Utility Scale Solar Power Plants
Utility Scale Solar Power Plants
Utility Scale Solar Power Plants
February 2012
India is well-placed to benefit from the successful development of the solar energy industry. It is
hoped that this guidebook will encourage project developers and financiers in India and elsewhere
to adopt industry best practices in the development, construction, operation and financing of
solar projects.
This guidebook is a best practice manual for utility-scale solar power plants in India. It
focuses primarily on ground mounted, fixed tilt PV projects and also covers solar tracking
system technology. Intended to be a practical toolkit, the guidebook includes an annex that
covers Concentrated Solar Power (CSP) technology and highlights aspects of the CSP project
development process that differ from the PV equivalent. It also has annexes on construction,
operation and maintenance contract terms.
While the guidebook is focused on utility-scale, grid-connected solar projects, much of the
technical content is equally relevant to off-grid solar applications and is likely to be helpful to
readers keen to deepen their understanding of this exciting sector.
Solar PV Technology
The applications of solar PV power systems can be PV modules must be mounted on a structure. This helps to
split into four main categories: off-grid domestic; off-grid keep them oriented in the correct direction and provides them
non-domestic; grid-connected distributed; and grid-connected with structural support and protection.
centralised. This guidebook is focussed on grid-connected
centralised applications. Mounting structures may be either fixed or tracking. Since
fixed tilt mounting systems are simpler, cheaper and have
The main components of a PV power plant are PV modules, lower maintenance requirements than tracking systems, they
mounting (or tracking) systems, inverters, transformers and are the preferred option for countries with a nascent solar
the grid connection. market and with limited indigenous manufacturers of tracking
technology (such as India). Although tracking systems are
Solar PV modules are made up of PV cells, which are most more expensive and more complex, they can be cost-effective
commonly manufactured from silicon but other materials are in locations with a high proportion of direct irradiation.
available. Cells can be based on either wafers (manufactured
by cutting wafers from a solid ingot block of material) or PV modules are generally connected together in series to
“thin film” deposition of material over low cost substrates. In produce strings of modules of a higher voltage. These strings
general, silicon-based crystalline wafers provide high efficiency may then be connected together in parallel to produce a higher
solar cells but are relatively costly to manufacture, whereas thin current DC input to the inverters.
film cells provide a cheaper alternative but are less efficient.
Inverters are solid state electronic devices that convert DC
Since different types of PV modules have different electricity generated by the PV modules into AC electricity,
characteristics (in terms of efficiency, cost, performance in low suitable for supply to the grid. In addition, inverters can
irradiation levels, degradation rate), no single type is preferable also perform a range of functions to maximise the output
for all projects. In general, good quality PV modules are of a PV plant.
expected to have a useful life of 25 to 30 years, although their
performance will steadily degrade over this period. In general, there are two main classes of inverters: central
inverters and string inverters. Central inverters are connected
The PV module market is dominated by a few large to a number of parallel strings of modules. String inverters
manufacturers based predominantly in Europe, North America are connected to one or more series strings. While numerous
and China. Selecting the correct module is of fundamental string inverters are required for a large plant, individual
importance to a PV project, keeping in mind the numerous inverters are smaller and more easily maintained than a
internationally accepted standards. When assessing the quality central inverter.
of a module for any specific project, it is important to assess
its specifications, certifications and performance record besides While central inverters remain the configuration of choice
the track record of the manufacturer. for most utility-scale PV projects, both configurations have
their pros and cons. Central inverters offer high reliability
and ease of installation. String inverters, on the other hand,
are cheaper to manufacture, simpler to maintain and can give
enhanced power plant performance on some sites.
A Guide For Developers and Investors 5
Solar Resource
The efficiency of proposed inverters should be carefully Reliable solar resource data are essential for the development
considered during the development process. While there is no of a solar PV project. While these data at a site can be defined
universally accepted method for quantifying inverter efficiency, in different ways, the Global Horizontal Irradiation (the total
there are a number of established methods that can help in solar energy received on a unit area of horizontal surface) is
making an informed decision. Almost half of the inverter generally of most interest to developers. In particular, a high
market is dominated by SMA Solar Technology AG, which long term average annual GHI is desired.
has a higher market share than the combined share of the next
four largest vendors. Following a global shortage of inverters There are two main sources of solar resource data: satellite
in 2010, some big name players are starting to enter the solar derived data and land-based measurement. Since both sources
inverter market. A key parameter is the Performance Ratio have particular merits, the choice will depend on the specific
(PR) of a PV power plant, which quantifies the overall effect site. Land based site measurement can be used to calibrate
of losses on the rated output. The PR, usually expressed as a resource data from other sources (satellites or meteorological
percentage, can be used to compare PV systems independent stations) in order to improve accuracy and certainty.
of size and solar resource. A PR varying from approximately
77% in summer to 82% in winter (with an annual average PR As solar resource is inherently intermittent, an
of 80%) would not be unusual for a well-designed solar PV understanding of inter-annual variability is important. At least
installation or plant, depending on the ambient conditions. 10 years of data are usually required to give the variation to a
reasonable degree of confidence.
It is also important to consider the capacity factor of a PV
power plant. This factor (usually expressed as a percentage) In India, solar resource data are available from various
is the ratio of the actual output over a period of a year to sources. These include the Indian Meteorological Department,
theoretical output if the plant had operated at nominal power NASA’s Surface Meteorology and Solar Energy data set,
for the entire year. The capacity factor of a fixed tilt PV plant METEONORM’s global climatological database, and satellite-
in southern Spain will typically be in the region of 16%. Plants derived geospatial solar data products from the United States
in India operating within a reliable grid network are expected National Renewable Energy Laboratory. These sources are of
to have a similar capacity factor. varying quality and resolution. Appropriate expertise is needed
to interpret the data.
This apart, the “specific yield” (the total annual energy
generated per kWp installed) is often used to help determine
the financial value of a plant and compare operating results
from different technologies and systems.
6 Utility Scale Solar Power Plants
Project Development
The development of a PV project can be broken down The development phase takes the project from the feasibility
into the following phases: conceptual, pre-feasibility study, stage through to financial close and is likely to consist of:
feasibility study, development and design. In general, each
succeeding phase entails an increased level of expenditure but
• Preparation and submission of the permit
applications for the proposed solar PV project.
reduces the risk and uncertainty in the project. In practice,
the progression through these phases is not strictly linear. The • Preparation and submission of a grid
amount of time and money committed in each phase will vary, connection application.
depending on the priorities and risk appetite of the developer. • Revision of the design and planning permissions.
A typical scope for a feasibility study would include • Decision on contracting strategy (turnkey EPC
the items below (again, these are covered in more contract or multi- contract).
detail in the book): • Decision on the financing approach.
• Assessment of shading (horizon and nearby buildings • Preparation of construction tender documentation.
and objects).
• Contractor selection and ranking.
• Outline layout of areas suitable for PV development.
• Contract negotiations.
• Assessment of technology options providing cost/
benefit for the project location: • Completion of a bankable energy yield.
• Predicted energy yields. The design phase will prepare the necessary detail and
documentation to enable the tendering and construction of
• Financial modelling. the solar PV plant.
A Guide For Developers and Investors 7
• Local climate – flooding, high winds, snow and To estimate accurately the energy produced from a PV
extreme temperatures. power plant, information is needed on the solar resource and
temperature conditions of the site. Also required are the layout
• Available area – area required for different module
and technical specifications of the plant components.
technologies, access requirements, pitch angle and
minimising inter-row shading.
To make life easy for project developers, a number of solar
• Land use – this will impact land cost and energy yield prediction software packages are available in the
environmental sensitivity. The impact of other land market. These packages use time step simulation to model the
users on the site should also be considered.
performance of a project over the course of a year. To ensure
• Topography – flat or slightly south facing slopes are more accurate results that would satisfy a financial institution’s
preferable for projects in the northern hemisphere. due diligence and make the project bankable, the analysis
should be carried out by a qualified expert. Realistic allowance
• Geotechnical – including consideration of
groundwater, resistivity, load bearing properties, soil should be made for undermining factors such as air pollution,
pH levels and seismic risk. grid downtime and electrical losses.
• Geopolitical – sensitive military zones should Annual energy yields may be expressed within a given
be avoided.
confidence interval (for example, the P90 annual energy
• Accessibility – proximity to existing roads, extent of yield prediction means the energy yield value with a 90%
new roads required. probability of exceedance). Since the energy yield simulation
software is heavily dependent on the input variables,
• Grid connection – cost, timescales, capacity,
proximity and availability. any uncertainty in the resource data gets translated into
uncertainty in the yield prediction results. As the energy yield
• Module soiling – including local weather, depends linearly, to a first approximation, on the plane of
environmental, human and wildlife factors.
array irradiance it is the uncertainty in the resource data that
• Water availability – a reliable supply is required for dominates the uncertainty in the yield prediction.
module cleaning.
Plant Design
The design of a PV plant involves a series of compromises The electrical design of a PV project can be split into the
aimed at achieving the lowest possible levelised cost[1] of DC and AC systems. The DC system comprises the following:
electricity. Choosing the correct technology (especially
modules and inverters) is of central importance. Selecting a
• Array(s) of PV modules.
module requires assessment of a complex range of variables. • Inverters.
At the very least, this assessment would include cost, power
output, benefits / drawbacks of technology type, quality,
• DC cabling (module, string and main cable).
spectral response, performance in low light, nominal power • DC connectors (plugs and sockets).
tolerance levels, degradation rate and warranty terms.
• Junction boxes/combiners.
The factors to consider when selecting inverters include • Disconnects/switches.
compatibility with module technology, compliance with grid
code and other applicable regulations, inverter-based layout,
• Protection devices.
reliability, system availability, serviceability, modularity, • Earthing.
telemetry requirements, inverter locations, quality and cost.
The AC system includes:
In designing the site layout, the following aspects
are important:
• AC cabling.
• Switchgear.
• Choosing row spacing to reduce inter-row shading
and associated shading losses. • Transformers.
• Orientating the modules to face a direction that In order to achieve a high performance PV plant, the
yields the maximum annual revenue from power incorporation of automatic data acquisition and monitoring
production. In the northern hemisphere, this will
technology is essential. This allows the yield of the plant to be
usually be true south.
monitored and compared with calculations made from solar
irradiation data to raise warnings on a daily basis if there is a
shortfall. Faults can then be detected and rectified before they
have an appreciable effect on production.
[1] Levelized cost is the net cost to install and operate a renewable energy system
divided by its expected life-time energy output.
A Guide For Developers and Investors 9
• Military authorities.
Selection of suitable technology and optimisation of
the main electrical systems is clearly vital. Alongside, Early engagement with all relevant authorities is highly
detailed consideration should be given to the surrounding advisable to minimise risk and maximise the chances of
infrastructure, including the mounting structures, control successful and timely implementation of the project.
building, access roads and site security systems. While these
systems should be relatively straightforward to design and Construction
construct, errors in these systems can have a disproportionate
impact on the project. The management of the construction phase of a solar
PV project should be in accordance with construction
Permits and Licensing management best practice. The aim should be to construct
the project to the required level of quality within the time and
Permit and licensing requirements vary, depending on cost deadlines.
the location of the project but the key permits, licences
and agreements typically required for renewable energy During construction, the environmental impact of the
projects include: project as well as the health and safety issues of the workforce
(and other affected people) should also be carefully managed.
• Land lease contract.
The IFC Performance Standards give detailed guidance on
• Environmental impact assessment. these issues. Compliance with these standards can facilitate the
financing of a project.
• Building permit/planning consent.
• Grid connection contract. Typical issues that arise during the construction of a PV
project include:
• Power purchase agreement.
The authorities, statutory bodies and stakeholders that • Foundations not being suited to ground conditions.
should be consulted also vary from country to country but will • Discovery of hazardous / contaminated substances
usually include the following organisation types: during excavation.
• Civil aviation authorities (if located near an airport). • Damaging cables during construction / installation.
10 Utility Scale Solar Power Plants
Commissioning should prove three main criteria: Scheduled maintenance typically includes:
• The power plant operates as designed and performs • Thermographic detection of faults.
as expected.
• Inverter servicing.
Commissioning tests are normally split into three groups: • Inspecting mechanical integrity of
mounting structures.
• Visual acceptance tests. These tests take place before
any systems are energised and consist of a detailed • Vegetation control.
visual inspection of all significant aspects of the plant.
• Routine balance of plant servicing / inspection.
• Pre-connection acceptance tests. These include an
open circuit voltage test and short circuit current test. Common unscheduled maintenance requirements include:
These tests must take place before grid connection.
• Tightening cable connections that have loosened.
• Post-connection acceptance test. Once the plant is
connected to the grid, a DC current test should be • Replacing blown fuses.
carried out. Thereafter, the performance ratio of the
plant is measured and compared with the value stated • Repairing lightning damage.
in the contract. An availability test, usually over a
period of 5 days, should also be carried out.
• Repairing equipment damaged by intruders or
during module cleaning.
• Repairing mounting structure faults. MWp for solar PV power projects commissioned during fiscal
years 2010/11 and 2011/12. The CERC benchmark also gives
• Rectifying tracking system faults.
a breakdown of the various cost elements that can be used by
Careful consideration should be given to selecting an developers for planning or comparison purposes.
operation and maintenance (O&M) contractor and drafting
the O&M contract to ensure that the performance of the plant The financial benefits and drawbacks to the developer
is optimised. After the project is commissioned, it is normal should be explored in detail through the development
for the EPC contractor to guarantee the performance ratio and of a full financial model. This model should include the
the O&M contractor to confirm the availability and, ideally, following inputs:
the performance ratio.
• Capital costs – these should be broken down
as far as possible. Initially, the CERC assumption
Economics and Financial can be used but quoted prices should be included
when possible.
Modeling
• Operations and maintenance costs – in addition
The development of solar PV projects can bring a range of to the predicted O&M contract price, operational
economic costs and benefits at the local and national levels. expenditure will include comprehensive insurance,
administration costs, salaries and labour wages.
Economic benefits can include:
• Annual energy yield – as accurate an estimate as is
• Job creation. available at the time.
• Use of barren land. • Energy price – this can be fixed or variable and will
depend on the location of the project as well as the
• Avoidance of carbon dioxide emissions.
tariff under which it has been developed.
• Increased energy security.
• Certified Emission Reductions – under the Clean
• Reduction of dependence on imports. Development Mechanism, qualifying Indian solar
projects may generate Certified Emission Reductions,
• Increased tax revenue. which can then be sold. However, this revenue is
difficult to predict.
An awareness of the possible economic benefits and costs
will aid developers and investors in making the case for • Financing assumptions – including proportion of
debt and equity, interest rates and debt terms.
developing solar energy projects to local communities and
government bodies. • Sensitivity analysis – sensitivity of the energy
price to changes in the various input parameters
India’s Central Electricity Regulatory Commission (CERC) should be assessed.
has produced a benchmark capital cost of INR 169 million/
12 Utility Scale Solar Power Plants
Financing PV Projects
Solar PV projects are generally financed on a project finance In Europe, it is quite normal to see the equity partners and
basis. As such, the equity partners and project finance partners developers form a special purpose vehicle (SPV) to develop the
typically conduct an evaluation of the project covering the project. The equity component is typically around 15-20% of
legal aspects, permits, contracts (EPC and O&M), and the total project investment cost. The debt portion—usually
specific technical issues prior to achieving financial closure. provided by an investment bank offering project finance
The project evaluations (due diligence) identify the risks and or leasing finance—is typically 80-85% of the total project
methods of mitigating them prior to investment. investment cost.
There are typically three main due diligence evaluations: Despite the recent turmoil in the international credit
markets, many financial institutions are willing to provide
• Legal due diligence – assessing the permits and long term finance for the solar energy sector. Individual
contracts (EPC and O&M). projects from smaller developers may receive financing with a
loan to value (LTV) ratio of 80%, whereas portfolios of solar
• Insurance due diligence – assessing the adequacy of
the insurance policies and gaps in cover. power projects from experienced developers may be financed
with a LTV ratio of 85%. Typical terms of the project finance
• Technical due diligence – assessing technical aspects loan are approximately 18 years. In India, a debt-equity split
of the permits and contracts. These include:
of 75:25 is taken as standard and the debt term is typically
• Sizing of the PV plant. around 14 years.
• Layout of the PV modules, mounting and/or At present, the insurance industry has not standardised the
trackers, and inverters.
insurance products for PV projects or components. However,
• Electrical design layout and sizing. an increasing demand for PV insurance is expected to usher
in standardisation. In general, while large PV systems require
• Technology review of major components
liability and property insurance, many developers may also opt
(modules/inverters/mounting or trackers).
to add policies such as environmental risk insurance.
• Energy yield assessments.
Conclusion
Solar power is becoming a widely accepted technology.
India is well-placed to benefit from the successful
development of a solar energy industry. It is hoped
that this guidebook will encourage Indian project
developers and financiers to adopt industry best
practices in the development, construction, operation
and financing of solar projects.
14 Utility Scale Solar Power Plants
Contents
1 INTRODUCTION���������������������������������������������������������������������������������������������������������������������� 23
2 SOLAR PV TECHNOLOGY�������������������������������������������������������������������������������������������������������� 24
2.1 Applications of Solar PV����������������������������������������������������������������������������������������������������������������� 24
2.2 Overview of Ground Mounted PV Power Plant���������������������������������������������������������������������������� 24
2.3 Solar PV Modules����������������������������������������������������������������������������������������������������������������������������� 24
2.4 Mounting and Tracking Systems���������������������������������������������������������������������������������������������������� 31
2.5 Inverters������������������������������������������������������������������������������������������������������������������������������������������� 34
2.6 Quantifying Plant Performance����������������������������������������������������������������������������������������������������� 39
2.7 Solar PV Technology Conclusions��������������������������������������������������������������������������������������������������� 41
3 THE SOLAR RESOURCE� ����������������������������������������������������������������������������������������������������������� 42
3.1 Quantifying the Resource��������������������������������������������������������������������������������������������������������������� 42
3.2 Solar Resource Assessment������������������������������������������������������������������������������������������������������������� 42
3.3 Variability in Solar Irradiation�������������������������������������������������������������������������������������������������������� 44
3.4 Indian Solar Resource���������������������������������������������������������������������������������������������������������������������� 44
4 PROJECT DEVELOPMENT�������������������������������������������������������������������������������������������������������� 50
4.1 Overview of Project Phases������������������������������������������������������������������������������������������������������������� 50
4.2 Concept�������������������������������������������������������������������������������������������������������������������������������������������� 50
4.3 Pre-Feasibility Study������������������������������������������������������������������������������������������������������������������������ 50
4.4 Feasibility Study������������������������������������������������������������������������������������������������������������������������������� 51
4.5 Development����������������������������������������������������������������������������������������������������������������������������������� 52
4.6 Detailed Design������������������������������������������������������������������������������������������������������������������������������� 55
5 SITE SELECTION���������������������������������������������������������������������������������������������������������������������� 55
5.1 Introduction������������������������������������������������������������������������������������������������������������������������������������� 55
5.2 Site Selection Constraints��������������������������������������������������������������������������������������������������������������� 56
6 ENERGY YIELD PREDICTION� ��������������������������������������������������������������������������������������������������� 60
6.1 Irradiation on Module Plane����������������������������������������������������������������������������������������������������������� 60
6.2 Performance Modelling������������������������������������������������������������������������������������������������������������������ 61
6.3 Energy Yield Prediction Results������������������������������������������������������������������������������������������������������ 61
6.4 Uncertainty in the Energy Yield Prediction����������������������������������������������������������������������������������� 61
A Guide For Developers and Investors 15
Contents
7 PLANT DESIGN������������������������������������������������������������������������������������������������������������������������ 68
7.1 Technology Selection����������������������������������������������������������������������������������������������������������������������� 68
7.2 Layout and Shading������������������������������������������������������������������������������������������������������������������������� 74
7.3 Electrical Design������������������������������������������������������������������������������������������������������������������������������� 77
7.4 Infrastructure����������������������������������������������������������������������������������������������������������������������������������� 92
7.5 Site Security�������������������������������������������������������������������������������������������������������������������������������������� 93
7.6 Monitoring and Forecasting����������������������������������������������������������������������������������������������������������� 94
7.7 Optimising System Design��������������������������������������������������������������������������������������������������������������� 98
7.8 Design Documentation Requirements����������������������������������������������������������������������������������������� 100
7.9 Plant Design Conclusions�������������������������������������������������������������������������������������������������������������� 102
8 PERMITS AND LICENSING����������������������������������������������������������������������������������������������������� 105
8.1 Permitting, Licensing and Regulatory Requirements – General������������������������������������������������ 105
8.2 IFC Performance Standards On Social And Environmental Sustainability�������������������������������� 105
8.3 Permitting, Licensing and Regulatory Requirements – India����������������������������������������������������� 106
9 CONSTRUCTION�������������������������������������������������������������������������������������������������������������������� 110
9.1 Introduction������������������������������������������������������������������������������������������������������������������������������������110
9.2 Interface Management������������������������������������������������������������������������������������������������������������������110
9.3 Programme and Scheduling����������������������������������������������������������������������������������������������������������110
9.4 Cost Management��������������������������������������������������������������������������������������������������������������������������113
9.5 Contractor Warranties�������������������������������������������������������������������������������������������������������������������115
9.6 Quality Management���������������������������������������������������������������������������������������������������������������������116
9.7 Environmental Management���������������������������������������������������������������������������������������������������������116
9.8 Health and Safety Management���������������������������������������������������������������������������������������������������117
9.9 Specific Solar PV Construction Issues��������������������������������������������������������������������������������������������117
9.10 Construction Supervision�������������������������������������������������������������������������������������������������������������119
10 COMMISSIONING���������������������������������������������������������������������������������������������������������������� 122
10.1 General Recommendations��������������������������������������������������������������������������������������������������������� 123
10.2 Pre-Connection Acceptance Testing������������������������������������������������������������������������������������������ 123
10.3 Grid Connection�������������������������������������������������������������������������������������������������������������������������� 123
10.4 Post Connection Acceptance Testing����������������������������������������������������������������������������������������� 124
10.5 Provisional Acceptance��������������������������������������������������������������������������������������������������������������� 124
10.6 Handover Documentation���������������������������������������������������������������������������������������������������������� 125
16 Utility Scale Solar Power Plants
Contents
11 OPERATION AND MAINTENANCE��������������������������������������������������������������������������������������� 126
11.1 Scheduled/Preventative Maintenance���������������������������������������������������������������������������������������� 126
11.2 Unscheduled Maintenance��������������������������������������������������������������������������������������������������������� 128
11.3 Spares�������������������������������������������������������������������������������������������������������������������������������������������� 129
11.4 Performance Monitoring, Evaluation and Optimisation���������������������������������������������������������� 129
11.5 Contracts�������������������������������������������������������������������������������������������������������������������������������������� 129
11.6 Operations and Maintenance Conclusions�������������������������������������������������������������������������������� 133
12 ECONOMICS AND FINANCIAL MODEL�������������������������������������������������������������������������������� 134
12.1 Economic Benefits and Costs������������������������������������������������������������������������������������������������������ 134
12.2 Central Electricity Regulatory Commission (CERC) Cost Benchmarks�������������������������������������� 136
12.3 Financial Model���������������������������������������������������������������������������������������������������������������������������� 137
13 FINANCING PV PROJECTS� ��������������������������������������������������������������������������������������������������� 144
13.1 Introduction��������������������������������������������������������������������������������������������������������������������������������� 144
13.2 Project Financing������������������������������������������������������������������������������������������������������������������������� 144
13.3 Risks���������������������������������������������������������������������������������������������������������������������������������������������� 146
13.4 Insurance�������������������������������������������������������������������������������������������������������������������������������������� 149
14 CONCLUSION���������������������������������������������������������������������������������������������������������������������� 150
Contents
4.2.5 Examples from Industry������������������������������������������������������������������������������������������������������������� 162
4.2.6 Losses������������������������������������������������������������������������������������������������������������������������������������������� 163
4.2.7 Costs�������������������������������������������������������������������������������������������������������������������������������������������� 164
4.2.8 Conclusions��������������������������������������������������������������������������������������������������������������������������������� 165
4.3 Power Tower���������������������������������������������������������������������������������������������������������������������������������� 166
4.3.1 Heliostat and the Tracking and Control Mechanisms�������������������������������������������������������������� 166
4.3.2 Receiver, Heat Transfer Medium and Tower���������������������������������������������������������������������������� 167
4.3.3 Examples in Industry������������������������������������������������������������������������������������������������������������������ 167
4.3.4 Conclusions��������������������������������������������������������������������������������������������������������������������������������� 168
4.4 Parabolic Dish�������������������������������������������������������������������������������������������������������������������������������� 169
4.4.1 Stirling Engines��������������������������������������������������������������������������������������������������������������������������� 169
4.4.2 Conclusions��������������������������������������������������������������������������������������������������������������������������������� 170
4.5 Power Block����������������������������������������������������������������������������������������������������������������������������������� 170
4.6 Energy Storage and Supplementary Heating������������������������������������������������������������������������������ 171
4.6.1 Overview������������������������������������������������������������������������������������������������������������������������������������� 171
4.6.2 Storage Medium (Including Molten Salts)�������������������������������������������������������������������������������� 172
4.6.3 Supplementary Heating (Use of Natural Gas or LPG)�������������������������������������������������������������� 172
4.6.4 Costs�������������������������������������������������������������������������������������������������������������������������������������������� 172
4.6.5 Conclusions��������������������������������������������������������������������������������������������������������������������������������� 173
4.7 Cooling and Water Consumption������������������������������������������������������������������������������������������������� 173
4.7.1 Cooling Options�������������������������������������������������������������������������������������������������������������������������� 173
4.7.2 Water Consumption��������������������������������������������������������������������������������������������������������������������174
4.8 Integrated Solar Combined Cycle������������������������������������������������������������������������������������������������� 175
4.9 Concentrated Photovoltaic (CPV)������������������������������������������������������������������������������������������������� 176
4.9.1 Manufacturers and Examples from Industry���������������������������������������������������������������������������� 176
4.9.2 CPV Advantages and Disadvantages���������������������������������������������������������������������������������������� 177
4.10 Linear Fresnel Reflector��������������������������������������������������������������������������������������������������������������� 177
4.10.1 Applications and Examples������������������������������������������������������������������������������������������������������ 177
4.10.2 Reflector and Structure������������������������������������������������������������������������������������������������������������ 178
4.10.3 Receiver and Heat Transfer������������������������������������������������������������������������������������������������������ 179
4.10.4 Conclusions�������������������������������������������������������������������������������������������������������������������������������� 179
18 Utility Scale Solar Power Plants
Contents
5 Site Selection� ����������������������������������������������������������������������������������������������������������������������� 180
6 Energy Yield Prediction� ������������������������������������������������������������������������������������������������������� 180
6.1 Site Conditions and Data Measurements������������������������������������������������������������������������������������ 180
6.2 Technology Characteristics����������������������������������������������������������������������������������������������������������� 181
6.3 Energy Yield Modelling���������������������������������������������������������������������������������������������������������������� 181
7 Project Implementation� ������������������������������������������������������������������������������������������������������� 182
7.1 Overview����������������������������������������������������������������������������������������������������������������������������������������� 182
7.2 Design��������������������������������������������������������������������������������������������������������������������������������������������� 182
7.2.1 Project Size and Land Area�������������������������������������������������������������������������������������������������������� 182
7.2.2 Load Matching Generation�������������������������������������������������������������������������������������������������������� 182
7.2.3 Solar Multiple������������������������������������������������������������������������������������������������������������������������������ 182
7.2.4 Capacity Factor�������������������������������������������������������������������������������������������������������������������������� 182
7.2.5 Grade of Heat����������������������������������������������������������������������������������������������������������������������������� 183
7.3 Development���������������������������������������������������������������������������������������������������������������������������������� 183
7.4 Engineering, Procurement and Construction������������������������������������������������������������������������������ 184
7.5 Uncertainties and Risks����������������������������������������������������������������������������������������������������������������� 184
7.5.1 Achieving Performance Improvements������������������������������������������������������������������������������������� 184
7.5.2 Realising Learning Rate Effects������������������������������������������������������������������������������������������������� 184
7.5.3 Supply Chain Competition��������������������������������������������������������������������������������������������������������� 185
7.5.4 Short Term Cost Uncertainties��������������������������������������������������������������������������������������������������� 185
8 Conclusion���������������������������������������������������������������������������������������������������������������������������� 185
Contents
LIST OF FIGURES AND TABLES
FIGURES
Figure 1: Overview of Solar PV Power Plant���������������������������������������������������������������������������������������� 25
Figure 2 (a): PV Technology Classes������������������������������������������������������������������������������������������������������ 26
Figure 2 (b): Development of Research Cell Efficiencies�������������������������������������������������������������������� 30
Figure 3: Effect of Tilt on Solar Energy Capture��������������������������������������������������������������������������������� 31
Figure 4: Benefit of Dual Axis Tracking System����������������������������������������������������������������������������������� 32
Figure 5: An Example of a Tracking PV Plant�������������������������������������������������������������������������������������� 33
Figure 6: PV System Configurations����������������������������������������������������������������������������������������������������� 35
Figure 7: Transformer and Transformerless Inverter Schematic��������������������������������������������������������� 36
Figure 8: Efficiency Curves of Low, Medium and High Efficiency Inverters as
Functions of the Input Power to Inverter Rated Capacity Ratios������������������������������������ 37
Figure 9: Inverter Manufacturer Market Share 2009�������������������������������������������������������������������������� 39
Figure 10: Annual Average Global Horizontal Irradiation����������������������������������������������������������������� 43
Figure 11: Example Pyranometer Measuring GHI (Image courtesy: Kipp & Zonen)������������������������� 43
Figure 12: Annual Average Daily Global Horizontal Irradiation for India at 40-km Resolution.���� 45
Figure 13: Annual Mean Horizontal Solar Irradiation for Three Cities in India�������������������������������� 46
Figure 14: Monthly Diffuse and Direct Solar Irradiation in Chennai, India�������������������������������������� 47
Figure 15: Large Scale PV Plant������������������������������������������������������������������������������������������������������������� 57
Figure 16: Uncertainty in Energy Yield Prediction������������������������������������������������������������������������������ 65
Figure 17: Sun-path Diagram for Chennai, India��������������������������������������������������������������������������������� 75
Figure 18: Shading Angle Diagram (Image courtesy of Schletter GmbH)����������������������������������������� 76
Figure 19: Voltage and Power Dependency Graphs of Inverter Efficiency��������������������������������������� 79
Figure 20: PV Array Showing String Cables����������������������������������������������������������������������������������������� 82
Figure 21: Typical Transformer Locations and Voltage Levels in a Solar Plant where
Export to Grid is at HV�������������������������������������������������������������������������������������������������������� 89
Figure 22: PV System Monitoring Schematic��������������������������������������������������������������������������������������� 95
Figure 23: Components of a Forecasting System�������������������������������������������������������������������������������� 97
Figure 24: Spacing Between Module Rows����������������������������������������������������������������������������������������119
Figure 25: Solar Panel Covered with Dust������������������������������������������������������������������������������������������ 127
Figure 26: Benchmark Solar PV Plant Cost Breakdown according to CERC������������������������������������� 136
20 Utility Scale Solar Power Plants
Contents
FIGURES
Contents
TABLES
Appendix B – AC Benchmarks
Table 1: Cable Specification���������������������������������������������������������������������������������������������������������������� 186
Table 2: Switchgear Specification������������������������������������������������������������������������������������������������������� 186
Table 3: Transformer Specification����������������������������������������������������������������������������������������������������� 187
List of abbreviations
°C Degrees Centigrade CPV Concentrating HV High Voltage Tracking
Photovoltaic
A Amp IFC International Finance MTBF Mean Times Between
CSP Concentrated Solar Corporation Failures
AC Alternating Current Power
IEC International NAPCC National Action Plan
AOD Aerosol Optical Depth DC Direct Current Electrotechnical for Climate Change
Commission
a-Si Amorphous Silicon DIN Deutsches Institut für NREL National Renewable
Normung IP International Energy Laboratory
CB Circuit Breaker
Protection Rating or
DNI Direct Normal Internet Protocol NVVN NTPC Vidhyut Vyapar
CDM Clean Development
Irradiation Nigam (NVVN)
Mechanism
IRR Internal Rate of Return
DSCR Debt Service Coverage O&M Operations and
c-Si Crystalline Silicon
Ratio ISC Short-Circuit Current Maintenance
CCGT Combined Cycle Gas
EIA Environmental Impact JNNSM Jawaharlal Nehru PPA Power Purchase
Turbine
Assessment National Solar Mission Agreement
CdTe Cadmium Telluride
EN European Norm kWh Kilowatt Hour PR Performance Ratio
CE Conformité
EPC Engineer Procure LTV Loan to Value PV Photovoltaic
Européenne
Construct
LV Low Voltage REC Renewable Energy
CER Certified Emission
FiT Feed-in Tariff Certificate
Reduction MET Meteorological
GHI Global Horizontal RPO Renewable Purchase
CERC Central Electricity MNRE Ministry of Renewable
Irradiation Obligation
Regulatory Energy
Commission GSM Global System SCADA Supervisory Control
for Mobile MPP Maximum Power Point and Data Acquisition
CIGS Copper Indium
Communications
(Gallium) Di-Selenide MPPT Maximum Power Point
A Guide For Developers and Investors 23
1. INTRODUCTION
The objective of this guidebook is to communicate how to successfully finance, develop, construct and operate a utility scale PV
power plant.
Throughout the guide, a number of case studies have been included to illustrate specific aspects of the development and
construction of solar PV projects in India. These case studies are based on a real solar PV project of 5 MWp capacity located in
India. While the studies are based on this one specific project, many of the issues addressed are relevant to other locations and
many of the challenges faced by this project will be common across solar power plant projects in India. This plant was the first of
its kind on this scale in India and served as a demonstration plant. It is in this context that the case studies should be read.
The layout design recommendations assume the plant is located in the northern hemisphere, although the concepts may be
extended to southern hemisphere locations.
Annex A gives an overview of CSP technology and highlights key differences between the PV and CSP project development
processes. Other appendices provide technical benchmarks for AC equipment and heads of terms for EPC and O&M contracts.
The guidebook may also be read in conjunction with the Indian market analysis report[2].
[2] IFC Market Analysis on Solar Power Generation in India (SgurrEnergy Limited).
Case Studies
The case studies will highlight a wide range of issues and lessons learned from the
development and construction of the 5 MW plant in Tamil Nadu, India. However, it should
be noted that many of these issues (e.g. the losses to be applied in an energy yield prediction
or the importance of a degree of adjustability in a supporting structure) come down to the
same fundamental point: it is essential to get suitable expertise in the project team. This does
not only apply to technical expertise but also to financial, legal and other relevant fields. It
can also be achieved in a variety of ways: e.g. hiring staff, using consultants or partnering with
other organisations.
Issues and lessons described in these case studies will inform the actions of other developers
and help promote good practice in the industry to ensure that best practices can be followed
to support project financing in this sector.
24 Utility Scale Solar Power Plants
2. SOLAR PV TECHNOLOGY
This section of the guidebook discusses PV applications, onto the semiconductor PV cells generates electron
movement. The output from a solar PV cell is direct
module technologies, mounting systems, inverters, monitoring
current (DC) electricity. A PV power plant contains
and forecasting techniques. It provides an overview of current many cells connected together in modules and many
commercially available technologies used in utility scale modules connected together in strings[3] to produce
solar PV projects. The purpose is to provide a framework of the required DC power output.
understanding for developers and investors before they commit
• Module mounting (or tracking) systems – These
to a specific technology.
allow PV modules to be securely attached to the
ground at a fixed tilt angle, or on sun-tracking
2.1 Applications of Solar PV frames.
There are four primary applications for PV power systems: • Inverters – These are required to convert the DC
electricity to alternating current (AC) for connection
to the utility grid. Many modules in series strings and
• Off-grid domestic – Providing electricity to
parallel strings are connected to the inverters.
households and villages that are not connected to the
utility electricity network (the “grid”). • Step-up transformers – The output from the
inverters generally requires a further step-up in
• Off-grid non-domestic – Providing electricity
voltage to reach the AC grid voltage level. The step-
for a wide range of applications such as
up transformer takes the output from the inverters to
telecommunication, water pumping and
the required grid voltage (for example 25 kV, 33 kV,
navigational aids.
38 kV, 110 kV depending on the grid connection
• Grid-connected distributed PV – Providing point and requirements).
electricity to a specific grid-connected customer.
• The grid connection interface – This is where the
• Grid-connected centralised PV – Providing electricity is exported into the grid network. The
centralised power generation for the supply of bulk substation will also have the required grid interface
power into the grid. switchgear such as circuit breakers and disconnects
for protection and isolation of the PV power plant as
The focus of this guidebook is on grid-connected centralised well as generation and supply metering equipment.
PV power plants. However much of the guidebook is also The substation and metering point are often external
to the PV power plant boundary and are typically
relevant to other applications.
located on the network operator’s property[4].
• Solar PV modules – These convert solar radiation [3] Modules may be connected together in series to produce a string of modules.
directly into electricity through the photovoltaic When connected in series the voltage increases. Strings of modules connected
effect in a silent and clean process that requires in parallel increase the current output.
no moving parts. The photovoltaic effect is a [4] Responsibility for this is defined in the grid connection contract. Normally,
it is the grid operator‘s onus to maintain the equipment in the grid operator‘s
semiconductor effect whereby solar radiation falling boundary—and there will be a cost to be paid by the PV plant owner.
A Guide For Developers and Investors 25
Utility Grid
Sunlight
Solar Modules
LV/MV
Voltage
Mounting Racks Step Up
AC Utility
Net Meter
Inverter
&
Transfers DC
DC/AC Disconnects
Electricity to
Inverter
AC
Transfers the Service
Converted AC Panel
Electricity
The unusual electrical properties required for PV cells limit Figure 2(a) summarises the technology classes:
the raw materials from which they may be manufactured.
Silicon is the most common material while cells using • Crystalline Silicon (c-Si) – Modules are made from
cells of either mono-crystalline or multi-crystalline
cadmium telluride and copper indium (gallium) di-selenide
silicon. Mono-crystalline silicon cells are generally
are also available. Each material has unique characteristics that the most efficient, but are also more costly than
impact the cell performance, manufacturing method and cost. multi-crystalline.
As of January 2010, approximately 78% of the global Crystalline wafers provide high efficiency solar cells but are
installed capacity of solar PV power plants use wafer-based relatively costly to manufacture. In comparison, thin film cells
crystalline silicon modules. Amorphous silicon and cadmium are typically cheaper due to both the materials used and the
telluride thin film modules make up the remaining 22%. simpler manufacturing process. However, thin film cells are
less efficient.
Mono
Crystalline
Crystalline
silicon cells
Poly/Multi
Crystalline
Amorphous
Microcrystalline
CdTe
Crystalline silicon modules consist of PV cells (typically The most well-developed thin film technology uses silicon
between 12.5 square cm and 20 square cm) connected in its less ordered, non-crystalline (amorphous) form. Other
together and encapsulated between a transparent front (usually technologies use cadmium telluride and copper indium (gallium)
glass), and a backing material (usually plastic or glass). di-selenide with active layers less than a few microns thick. In
general, thin film technologies have a less established track record
Mono-crystalline wafers are sliced from a large single crystal than many crystalline technologies. The main characteristics of
ingot in a relatively expensive process. thin film technologies are described in the following sections.
Amorphous silicon (a-Si) can be deposited on a wide range process, the quality of assembly and packaging of the cells
of both rigid and flexible low cost substrates. The low cost of into the module, as well as maintenance levels employed
a-Si makes it suitable for many applications where low cost is at the site. Regular maintenance and cleaning regimes may
more important than high efficiency. reduce degradation rates but the main impact is specific to
the characteristics of the module being used. It is, therefore,
2.3.3.2 Cadmium Telluride important that reputable module manufacturers are chosen
and power warranties are carefully reviewed.
Cadmium telluride (CdTe) is a compound of cadmium
and tellurium. The cell consists of a semiconductor film stack The extent and nature of degradation varies among module
deposited on transparent conducting oxide-coated glass. A technologies. For crystalline modules, the cells may suffer from
continuous manufacturing process using large area substrates irreversible light-induced degradation. This can be caused by
can be used. Modules based on CdTe produce a high energy the presence of boron, oxygen or other chemicals left behind
output across a wide range of climatic conditions with good by the screen printing or etching process of cell production.
low light response and temperature response coefficients. The initial degradation occurs due to defects that are activated
on initial exposure to light.
2.3.3.3 Copper Indium (Gallium) Di-Selenide
(CIGS/CIS) Amorphous silicon cells degrade through a process called
the Staebler-Wronski Effect[5]. This degradation can cause
CIGS is a semiconductor consisting of a compound of reductions of 10-30% in the power output of the module
copper, indium, gallium and selenium. in the first six months of exposure to light. Thereafter, the
degradation stabilises and continues at a much slower rate.
CIGS absorbs light more efficiently than crystalline silicon,
but modules based on this semiconductor require somewhat Amorphous silicon modules are generally marketed at
thicker films than a-Si PV modules. Indium is a relatively their stabilised performance levels. Interestingly, degradation
expensive semiconductor material, but the quantities required in amorphous silicon modules is partially reversible with
are extremely small compared to wafer based technologies. temperature. In other words, the performance of the modules
may tend to recover during the summer months, and drop
Commercial production of CIGS modules is in the early again in the colder winter months.
stages of development. However, it has the potential to offer
the highest conversion efficiency of all the thin film PV Additional degradation for both amorphous and crystalline
module technologies. technologies occurs at the module level and may be caused by:
• Mechanical stress and humidity on the contacts. efficiency technologies are more costly to manufacture, less
efficient modules require a larger area to produce the same
• Cell contact breakdown.
nominal power. As a result, the cost advantages gained at the
• Wiring degradation. module level may get offset by the cost incurred in providing
additionally required power system infrastructure (cables and
PV modules may have a long term power output mounting frames) for a larger module area. So using the lowest
degradation rate of between 0.3% and 1% per annum. For cost module does not necessarily lead to the lowest cost per
crystalline modules, a generic degradation rate of 0.5% per Wp for the complete plant. The relationship between plant
annum is often considered applicable (if no specific testing has area and module efficiency is discussed in Section 5.2.2.
been conducted on the modules being used[6]). Banks often
assume a flat rate of degradation rate of 0.5% per annum. At the time of writing, crystalline silicon technology
comprises almost 80% of global installed solar capacity and is
In general, good quality PV modules may be expected to likely to remain dominant in the short term. But the presence
have a useful life of 25-30 years. The possibility of increased of thin film technologies is growing. As of 2010, Cadmium
rates of degradation becomes higher thereafter. Telluride accounted for the large majority of installed
thin film capacity but CIGS is thought to have promising cost
2.3.5 Module Cost and Efficiency reduction potential
[6] US Department of Energy, National Renewable Energy Laboratory. The International Electrotechnical Commission (IEC)
[7] ASIF: Informe Annual 2009. Maximum efficiency values taken from “Best
Production-Line PV Module Efficiency Values From Manufacturers‘ Websites”
issues internationally accepted standards for PV modules. PV
compiled by Bolko von Roedern, NREL 9-2009 modules will typically be tested for durability and reliability
[8] “Global Market Outlook for Photovoltaics until 2014” www.epia.org
(Concentrating PV and other minor technologies not shown).
[9] The temperature coefficient for power describes the dependence on power
output with increasing temperature. Module power generally decreases as the
module temperature increases.
A Guide For Developers and Investors 29
according to these standards. Standards IEC 61215 (for Table 2 summarises major PV quality standards. These are
crystalline silicon modules) and IEC 61646 (for thin film an accepted quality mark and indicate that the modules can
modules) include tests for thermal cycling, humidity and safely withstand extended use. However, they say very little
freezing, mechanical stress and twist, hail resistance and about the performance of the module under field conditions
performance under some fixed test conditions, including of varying irradiance and temperature experienced at a specific
standard testing conditions (STC )[10]. site location.
IEC 60364-4-41 Protection against electric shock. 2). High dielectric strength.
There are efforts to create a new standard, IEC 61853, There are currently a few large module manufacturers
which will test the performance and energy rating of modules dominating the market. Financial institutions often keep lists
under a variety of irradiance and temperature conditions. of module manufactures they consider bankable. However,
This standard should facilitate comparison and selection of these lists can quickly become dated as manufacturers
modules, based on performance. introduce new products and quality procedures. Often, the
larger manufacturers (such as Suntech, Sunpower or First
2.3.7 Module Manufacturers Solar) are considered bankable, but there is no definitive and
accepted list.
Manufacturers of PV modules are based predominantly
in Europe, China and North America[11]. A 2009 survey by 2.3.8 Module Technology Developments
Photon International (2-2009) indicated that there were over
220 suppliers of PV modules and over 2,700 products. When Solar PV module technology is developing rapidly. While
assessing the quality of a module for any specific project, it a wide variety of different technical approaches are being
is important to assess its specifications, certifications, track explored, the effects of these approaches are focused on either
record, and the track record of the manufacturer. improving module efficiency or reducing manufacturing costs.
[11] There are several module manufacturers in India but the current installed Incremental improvements are being made to conventional
production capacity is still under 1000 MWp per annum. India exports a
major portion of the production. The dominant technology is crystalline silicon c-Si cells. One of these improvements is the embedding of
with amorphous silicon gaining ground.
[12] Image courtesy from United States National Renewable Energy Laboratory the front contacts in laser-cut microscopic grooves in order to
www.nrel.com, it can be accessed April 2011.
A Guide For Developers and Investors 31
reduce the surface area of the contacts and so increase the area Figure 2(b) illustrates the development of the efficiencies
of the cell that is exposed to solar radiation. Similarly, another of research cells from 1975 to the present day. It should be
approach involves running the front contacts along the back of noted that commercially available cells lag significantly behind
the cell and then directly through the cell to the front surface at research cells in terms of efficiency.
certain points.
2.4 Mounting and Tracking Systems
Different types of solar cell inherently perform better at
different parts of the solar spectrum. As such, one area of PV modules must be mounted on a structure, to keep them
interest is the stacking of cells of different types. If the right oriented in the correct direction and to provide them with
combination of solar cells is stacked (and the modules are structural support and protection. Mounting structures may
sufficiently transparent) then a stacked or “multi-junction” cell be fixed or tracking.
can be produced which performs better across a wider range of
the solar spectrum. This approach is taken to the extreme in 2.4.1 Fixed Mounting Systems
III-V cells (named after the respective groups of elements in the
Periodic Table) in which the optimum materials are used for each Fixed mounting systems keep the rows of modules at a
part of the solar spectrum. III-V cells are very expensive but have fixed tilt angle[13] while facing a fixed angle of orientation[14].
achieved efficiencies in excess of 40%. Less expensive approaches Figure 3 illustrates why the tilt angle is important for
based on the same basic concept include hybrid cells (consisting maximising the energy incident on the collector plane.
of stacked c-Si and thin film cells) and multi-junction a-Si cells.
Time of Day
The tilt angle and orientation is generally optimised for each In locations with a high proportion of direct irradiation
PV power plant according to location. This helps to maximise including some regions of India, single or dual-axis
the total annual incident irradiation[15] and total annual energy tracking systems can be used to increase the average total
yield. For Indian sites, the optimum tilt angle is generally annual irradiation. Tracking systems follow the sun as it
between 10º and 35º, facing true south. There are several moves across the sky. They are generally the only moving
off-the-shelf software packages that may be used to optimise parts employed in a PV power plant.
the tilt angle and orientation according to specifics of the site
location and solar resource. Single-axis trackers alter either the orientation or tilt angle
only, while dual-axis tracking systems alter both orientation
Fixed tilt mounting systems are simpler, cheaper and have and tilt angle. Dual-axis tracking systems are able to track the
lower maintenance requirements than tracking systems. sun more precisely than single-axis systems.
They are the preferred option for countries with a nascent
solar market and limited indigenous manufacturing of Depending on the site and precise characteristics of the
tracking technology. solar irradiation, trackers may increase the annual energy
yield by up to 27% for single-axis and 37% for dual-axis
trackers. Tracking also produces a smoother power output
plateau as shown in Figure 4. This helps meet peak demand in
[15] Irradiation is the solar energy received on a unit area of surface. It is defined afternoons, which is common in hot climates due to the use of
more fully in section 3.1.
air conditioning units.
A Guide For Developers and Investors 33
Almost all tracking system plants use crystalline silicon An example of a tracking PV plant is shown in Figure 5.
modules. This is because their higher efficiency reduces Aspects to take into account when considering the use of
additional capital and operating costs required for the tracking tracking systems include:
system (per kWp installed). However, relatively inexpensive
single-axis tracking systems have recently been used with some Financial:
thin film modules.
• Additional capital costs for the procurement
There are many manufacturers and products of solar PV and installation of the tracking systems
(typically $140-700/kWp).
tracking systems. A 2009 survey of manufacturers by Photon
International magazine (11-2009) listed 170 different devices • Additional land area required to avoid shading
with a wide range of tracking capabilities. Most fall into one compared to a free field fixed tilt system of the same
of six basic design classes (classic dual-axis, dual-axis mounted nominal capacity.
on a frame, dual-axis on a rotating assembly, single-axis • Large tracking systems may require cranes to install,
tracking on a tilted axis, tracking on a horizontal axis and increasing the installation cost.
single-axis tracking on a vertical axis). In general, the simpler
the construction, the lower the extra yield compared to a fixed
• There is a higher maintenance cost for tracking
systems due to the moving parts and actuation
system, and the lower the maintenance requirement. systems. Typical additional maintenance costs range
from $2.8-21/kWp per annum.
34 Utility Scale Solar Power Plants
• Tracking angles: all trackers have angular limits that There are two broad classes of inverters: central inverters
vary among products. Depending on the angular and string inverters. The central inverter configuration shown
limits, performance may be reduced.
in Figure 6 remains the first choice for many medium and
• High wind capability and storm mode: dual-axis large scale solar PV plants. A large number of modules are
tracking systems especially need to go into a storm connected in series to form a high voltage string. Strings are
mode when the wind speed is over 16-20 m/s. This then connected in parallel to the inverter.
could reduce the energy yield and revenues at high
wind speed sites.
Central inverters offer high reliability and simplicity of
• Direct/diffuse irradiation ratio: tracking systems will installation. However, they have disadvantages: increased
give greater benefits in locations that have a higher mismatch losses[16] and absence of maximum power point
direct irradiation component. This is discussed tracking[17] (MPPT) for each string. This may cause problems
further in Section 3.
for arrays that have multiple tilt and orientation angles, suffer
2.4.3 Certification from shading, or use different module types.
Support structures should adhere to country specific Central inverters are usually three-phase and can include
standards and regulations, and manufacturers should conform grid frequency transformers. These transformers increase the
to ISO 9001:2000. This specifies requirements for a quality weight and volume of the inverters although they provide
management system where an organisation needs to: galvanic isolation from the grid. In other words, there is
no electrical connection between the input and output
• Demonstrate its ability to consistently provide voltages—a condition that is sometimes required by national
products that meet customer and applicable electrical safety regulations.
regulatory requirements.
• Aim to enhance customer satisfaction through the Central inverters are sometimes used in a “master slave”
effective application of the system. These include configuration. This means that some inverters shut down
processes for continual improvement as well as the when the irradiance is low, allowing the other inverters to run
assurance of conformity to customer and applicable more closely to optimal loading. When the irradiance is high,
regulatory requirements. the load is shared by all inverters. In effect, only the required
number of inverters is in operation at any one time. As the
2.5 Inverters
operating time is distributed uniformly among the inverters,
design life can be extended.
Inverters are solid state electronic devices. They convert DC
electricity generated by the PV modules into AC electricity,
ideally conforming to the local grid requirements. Inverters
can also perform a variety of functions to maximise the output
of the plant. These range from optimising the voltage across
the strings and monitoring string performance to logging data,
and providing protection and isolation in case of irregularities
in the grid or with the PV modules. [16] Mismatch refers to losses due to PV modules with varying current/voltage
profiles being used in the same array.
[17] Maximum Power Point Tracking is the capability of the inverter to adjust its
impedance so that the string is at an operating voltage that maximises the
power output.
A Guide For Developers and Investors 35
In contrast, the string inverter concept uses multiple String inverters, which are usually in single phase, also have
inverters for multiple strings of modules. String inverters are other advantages. For one, they can be serviced and replaced
increasingly being used as they can cover a very wide power by non-specialist personnel. For another, it is practical to keep
range and can be manufactured more cheaply in a production spare string inverters on site. This makes it easy to handle
line than central inverters. Additionally, they provide MPPT unforeseen circumstances, as in the case of an inverter failure.
on a string level with all strings being independent of each In comparison, the failure of a large central inverter—with
other. This is useful in cases where modules cannot be a long lead time for repair—can lead to significant yield loss
installed with the same orientation, where modules of different before it can be replaced.
specifications are being used, or when there are shading issues.
Inverters may be transformerless or include a transformer to
step up the voltage. Transformerless inverters generally have a
higher efficiency, as they do not include transformer losses.
36 Utility Scale Solar Power Plants
Power quality and grid code requirements are country- A number of different measures of efficiency have been
dependent. It is not possible to provide universally applicable defined for inverters. These describe and quantify the
guidelines. The national regulations and standards should be efficiency of different aspects of an inverter’s operation.
consulted when selecting an inverter and designing a solar PV The search for an objective way of quantifying inverter
power plant. performance is still ongoing. New ways of measuring efficiency
are frequently suggested in industry literature. The most
In general, one of the quantities used to describe the quality commonly used methods are discussed below.
of a grid-connected inverter is total harmonic distortion
(THD). It is a measure of the harmonic content of the inverter The conversion efficiency is a measure of the losses
output and must be limited in most grid codes. For high experienced during the conversion from DC to AC.
quality devices, THD is normally less than 5%[21] . These losses are due to multiple factors: the presence of
a transformer and the associated magnetic and copper
100
90
80
70
Inverter Efficiency (%)
60
High
50 Efficiency Medium
Efficiency
40 Low
Efficiency
30
20
10
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Ratio of Input Power to the Inverter’s rated Capacity
[22] Jayanta Deb Mondol, Yigzaw G. Yohanis, Brian Norton. Optimal sizing of
[21] Handbook of Photovoltaic Science and Engineering (Wiley, 2003) array and inverter for grid-connected photovoltaic systems, 2006.
38 Utility Scale Solar Power Plants
2.5.4 Certification
In order to ensure a high level of quality and performance, The latest global market survey for inverters conducted by
and to minimise risk, inverters must be compliant with Photon International magazine (04-2010) lists 1,068 inverter
a number of standards. The requirements, in terms of types, 532 of which are rated at 10 kW or less. Over a quarter
compliance with standards, depend on the location of the of the inverter types belong to the 10 to 100 kW range.
project and the type of inverter.
Over the past year, a number of major industry players have
Important standards bodies for inverters are DIN VDE, started to enter the inverter market. These include GE, ABB,
IEC and EN. Inverters must be CE compliant in order to and Schneider Electric (through the acquisition of Xantrex).
be installed in Europe. Table 3 is a non-exhaustive list of In 2010, the growth in the solar PV market and delays in
standards to which inverters should conform according to production (due to scarcity of key electronic components) led
European practice. to a global shortage of inverters.
The inverter market is dominated by SMA Solar The performance of a PV power plant is expected to fall
Technology AG, which has a higher market share than the during its lifetime, especially in the second and third decade of
combined share of the next four largest vendors (Kaco, its life as modules continue to degrade and plant components
Fronius, Power-One and Siemens) (as illustrated in Figure 9). age. In addition to the quality of the initial installation, a high
Other inverter manufacturers hold the remaining 18% share of degree of responsibility for the performance of a PV plant lies
the global market[24]. with the O&M contractor. This section discusses how the
operational performance of a PV plant may be quantified.
Figure 9: Inverter Manufacturer Market Share 2009[24] [24] Sun & Wind Energy February 2010
40 Utility Scale Solar Power Plants
irradiation into useful energy. The PR of a plant may be CF=(Energy generated per annum (kWh))/(8760
predicted using simulations, or alternatively may be calculated (hours / annum)×Installed Capacity (kWp))
for an operational plant by measuring irradiation, ambient
temperature, wind velocity, module temperature, voltage and The capacity factor of a fixed tilt PV plant in southern
current over a given time period. Spain will typically be in the region of 16%. This means
that a 5 MWp plant will generate the equivalent energy of
As PV plant losses vary according to environmental a continuously operating 0.8 MW plant. Plants in India
conditions through the year, the PR also varies. For example, operating within a reliable grid network are expected to have a
the more significant negative temperature coefficient similar capacity factor.
of power for crystalline modules may lead to increased
losses at high ambient temperatures. A PR varying from 2.6.3 Specific Yield
approximately 77% in summer to 82% in winter (with an
annual average PR of 80%) would not be unusual for a well- The “specific yield” (kWh/kWp) is the total annual energy
designed solar PV power plant that is not operating in high generated per kWp installed. It is often used to help determine
ambient temperature conditions. the financial value of an array[25] and compare operating results
from different technologies and systems. The specific yield of a
Some plants using amorphous silicon modules show the plant depends on:
opposite effect: in summer months, the PR increases, dropping
again in the colder winter months. This is due to the fact that • The total annual irradiation falling on the collector
plane. This can be increased by optimally tilting the
Staebler-Wronski degradation is partially reversible at high
modules or employing tracking technology.
temperatures. It is common to observe seasonal oscillations
in the PR of amorphous silicon plants due to this thermal • The performance of the module, including sensitivity
annealing process. to high temperatures and low light levels.
2.7
Thin film cells provide a cheaper alternative but are less efficient. There are three main types of thin film cells:
• Amorphous Silicon – The low cost of a-Si makes it suitable for many applications where low cost is more
important than high efficiency.
• Cadmium Telluride – Modules based on CdTe produce a high energy output across a wide range of climatic
conditions with good low light response and temperature response coefficients.
• Copper Indium (Gallium) Di-Selenide (CIGS/CIS) – Commercial production of CIGS modules is in the
early stages of development. However, it has the potential to offer the highest conversion efficiency of all the
thin film PV module technologies.
The performance of a PV module will decrease over time due to a process known as degradation. Typically,
the degradation rate is highest in the first year of operation and then it stabilises. PV modules may have a long
term degradation rate of between 0.3% and 1% per annum. Banks often assume a flat rate of degradation rate
of 0.5% per annum.
Modules are either mounted on fixed angle frames or on sun-tracking frames. Fixed frames are simpler to install,
cheaper and require less maintenance. However, tracking systems can increase yield by up to 34%. Tracking,
particularly for areas with a high direct/diffuse irradiation ratio, also enables a smoother power output.
Inverters convert DC electricity generated by the PV modules into AC electricity, ideally conforming to the local
grid requirements. They are arranged either in string or central configurations. Central configuration inverters are
considered to be more suitable for multi-megawatt plants. String inverters enable individual string MPPT and
require less specialised maintenance skills. String configurations are becoming increasingly popular as they offer more
design flexibility.
PV modules and inverters are all subject to certification, predominantly by the IEC. However, one major absence in
the standards is performance and energy rating testing other than at standard testing conditions (STC). A standard is
being prepared for this, which should enable easier comparison of manufacturers.
The performance ratio (PR) of a well-designed PV power plant will typically be in the region of 75% to 85%,
degrading over the lifetime of the plant. The capacity factor should typically be in the region of 16%. In general, good
quality PV modules may be expected to have a useful life of 25 to 30 years.
42 Utility Scale Solar Power Plants
3.1 Quantifying the Resource Long term annual average values of GHI and DNI can be
obtained for a site by interpolating measurements taken from
Site selection and planning of PV power plants requires ground based sensors or indirectly from the analysis of satellite
reliable solar resource data. Power production depends imagery. Ideally, historical values of daily or hourly irradiation
linearly on the plane of array irradiance[26], at least to a first with a spatial resolution of 10 km or less are required to
approximation. The solar resource of a location is usually generate regional solar atlases.
defined by the values of the global horizontal irradiation[27],
direct normal irradiation and diffuse horizontal irradiation as As the distance between a solar resource and a ground-based
defined below. sensor increases, the uncertainty of interpolated irradiation
values increases. Under such circumstances, satellite derived
• Global Horizontal Irradiation (GHI) – GHI is the data may be preferred. The uncertainty in satellite-derived data
total solar energy received on a unit area of horizontal is reducing as new models develop. The precise point at which
surface. It includes energy from the sun that is satellite data become preferable over data interpolated from
received in a direct beam and from all directions of
ground sensors depends on the individual case. The relative
the sky when radiation is scattered off the atmosphere
(diffuse irradiation). The yearly sum of the GHI is merits of these alternative data sources are discussed below.
of particular relevance for PV power plants, which
are able to make use of both the diffuse and beam 3.2.1 Satellite Derived Data
components of solar irradiance.
Satellite-derived data can offer a wide geographical coverage
• Direct Normal Irradiation (DNI) – DNI is the
total solar energy received on a unit area of surface and can often be obtained retrospectively for historical
directly facing the sun at all times. The DNI is of periods in which no ground-based measurements were taken.
particular interest for solar installations that track This is especially useful for assessing long term averages. A
the sun and for concentrating solar technologies combination of analytical, numerical and empirical methods
(as concentrating technologies can only make use of can offer half-hourly data with a nominal spatial resolution
the direct component of irradiation).
down to 2.5 km, depending on the location and field of view
• Diffuse Horizontal Irradiation (DHI) – DHI is the of the satellite.
energy received on a unit area of horizontal surface
from all directions when radiation is scattered off the One advantage of satellite resource assessment is that data is
atmosphere or surrounding area. not susceptible to maintenance and calibration discontinuities.
The same sensor is used to assess locations over a wide area.
Irradiation is measured in kWh/m2, and values are often
This can be particularly useful in comparing and ranking sites
given for a period of a day, a month or a year. A high long
as bias errors are consistent.
term average annual GHI is typically of most interest to PV
project developers. Average monthly values are important
A comparison of the GHI values shows that statistics
when assessing the proportion of energy generated in each
obtained from satellite readings correspond well with
month. Figure 10 shows the annual average of GHI for India.
ground-measured data. But it is not so in the case of DNI
[26] Irradiance is the power incident on a surface per unit area. (Watts per square values. Currently, it is not so clear if this dissonance is due
meter or W/m 2). to the satellite methodology or the poor maintenance of
[27] Irradiation is a measure of the energy incident on a unit area of a surface in a
given time period. This is obtained by integrating the irradiance over defined ground-based measurement stations, but is likely to be a
time limits. (energy per square meter or kWh/m 2)
combination of both.
A Guide For Developers and Investors 43
The traditional approach to solar resource measurement is For locations that have a low density of MET stations and
to use land-based sensors. A variety of sensor technologies is rely on satellite data, on-site resource monitoring may be
available from a number of manufacturers with differing accuracyconsidered during the feasibility stage of the project. On-site
and cost implications. The two main technology classes are: resource monitoring may be used to calibrate satellite-derived
estimates, thereby reducing bias and improving accuracy.
• Thermal Pyranometers – These are also known as In general, up to four months of measured data can reduce
solarimeters and typically consist of a black metal plate existing bias, and improve the estimation of the long term
absorber surface below two hemispherical glass domes
mean. A further, four to eight months of measured data will
in a white metal housing. Solar irradiance warms up
the metal plate in proportion to its intensity. The improve the capability to capture seasonal variations. But
degree of warming, compared to the metal housing, the best results are obtained by monitoring for a full twelve
can be measured with a thermocouple. High precision months or longer.
can be achieved with regular cleaning and recalibration.
Since thermal pyranometers have a slow response time, 3.3 Variability in Solar Irradiation
they might not be able to capture rapidly varying
irradiance levels due to clouds. Also, diffuse irradiance
can be measured if a sun – tracking shading disc is used In terms of irradiation, the solar resource is inherently
to block out irradiance travelling directly from the sun. intermittent. In any given year, the total annual global
An example of a pyranometer is shown in Figure 11. irradiation on a horizontal plane varies from the long term
average due to climatic fluctuations. This means that though
• Silicon Sensors – These are cheaper than pyranometers
the plant owner may not know the energy yield to expect in
and consist of a PV cell, often using crystalline silicon.
The current delivered is proportional to the irradiance. any given year, he can have a good idea of the expected yield
Temperature compensation can be used to increase averaged over the long term.
accuracy but its scope is limited by the spectral
sensitivity of the cell. Some wavelengths (for example To help lenders understand the risks and perform a
long wavelength IR) may not be accurately measured, sensitivity analysis, it is important to quantify the limits of
resulting in a lower irradiance measurement of up to
the inter-annual variation. This can be achieved by assessing
5% compared to thermal pyranometers.
the long-term irradiation data (in the vicinity of the site)
Well maintained land-based sensors can measure the solar sourced from nearby MET stations or satellites. At least 10
resource with a relative accuracy of 3-5%. Long term data years of data are usually required to give a reasonably confident
from such stations may be used to calibrate satellite – derived assessment of the variation. Research papers[29] show that for
irradiation maps. However, maintenance is very important since southern Europe (including Spain), the coefficient of variation
soiled or ill-calibrated sensors can easily yield unreliable data. (standard deviation divided by the mean[30]) is below 4%.
Table 4 shows the coefficient of variation for four locations in
In Europe, it is not common for the solar resource to be India as derived from data provided by NASA.
measured at the site of a PV plant for any significant length
of time, prior to construction. Energy yield predictions
typically rely on historical irradiation data taken from nearby
meteorological (MET) stations or derived from satellite imagery. [29] Uncertainties in photovoltaic electricity yield prediction from fluctuation of
solar radiation. Proceedings of the 22nd European Photovoltaic Solar Energy
Conference, Milano, Italy 3-7.9.2007.
[30] The coefficient of variation is a dimensionless, normalised measure of the
dispersion of a probability distribution. It enables the comparison of different
data streams with varying mean values.
A Guide For Developers and Investors 45
Figure 12: Annual Average Daily Global Horizontal Irradiation for India at 40 km Resolution[31]
[31] Data from United States National Renewable Energy Laboratory www.nrel. • NASA’s Surface Meteorology and Solar Energy data
com, accessed May 2010. National boundaries shown may not necessarily set. This holds satellite – derived monthly data for a
accurately reflect the Indian national boundaries.
46 Utility Scale Solar Power Plants
2500
Horizontal Solar Irradiation (kWh/m²)
2000
1500
1221
1027 1320
Direct
1000 Diffuse
500
820 800 656
0
Mumbai Chennai New Delhi
Figure 13: Annual Mean Horizontal Solar Irradiation for Three Cities in India
Figure 13: Annual Mean Horizontal Solar Irradiation for Three Cities in India
grid of 1°x1° covering the globe for a 22 year period METEONORM generates climatological averages
(1984-2005). The data are considered accurate estimated by using interpolation algorithms and
for preliminary feasibility studies of solar energy satellite data.
projects in India. They are also particularly useful
for estimating the inter-annual variability of the • Satellite-derived geospatial solar data products from
solar resource. the United States-based National Renewable Energy
Laboratory (NREL). Annual average DNI and GHI,
• The METEONORM global climatological database latitude tilt, and diffuse data are available at 40 km
and synthetic weather generator. This contains resolution for South and East Asia and at 10 km
a database of ground station measurements of resolution for India, as shown in the examples of
irradiation and temperature. In cases where a site is Figure 10 and Figure 12.
over 20 km from the nearest measurement station,
A Guide For Developers and Investors 47
250
Horizontal Solar Irradiation (kWh/m2)
200
150
Direct
100 Diffuse
50
0
rch
ril
ber
y
ber
y
ber
st
ary
ber
Ma
uar
Jul
Jun
gu
Ap
Ma
to
vem
cem
u
em
Au
r
Jan
Oc
Feb
No
De
Sep
Figure 14: Monthly Diffuse and Direct Solar Irradiation in Chennai, India
Figure 13 shows the proportion of direct and diffuse Figure 14 shows how the average monthly solar irradiation
radiation for three cities in India obtained from the varies over a year at Chennai. The energy yield and revenue
METEONORM database using land-based sensors supervised of a PV power plant may be expected to vary approximately
by the Word Meteorological Organisation (WMO). Due to in proportion. The annual mean GHI in Chennai is
the higher proportion of direct irradiation, it may be expected 2,021 kWh/m2. By optimally-orientating a fixed tilt plant,
that tracking technologies will offer a greater advantage in the yearly sum of global irradiation may be increased to
New Delhi than in Mumbai. 2,048 kWh/m2. Based on this resource, a 1 MWp plant with a
PR of 80% will give an AC yield of 1,638 MWh.
48 Utility Scale Solar Power Plants
Case Study 1
Resource Assessment
In order to support financing, the developer of the 5 MW plant in Tamil Nadu had a basic solar resource assessment
carried out. However, only one data source was used and there was no assessment made of the inter-annual variability
of the resource. Nor was any analysis provided of the historical period on which the data was based. However, as has
been seen globally, financing institutions are becoming more sophisticated in their analysis of solar plants and their
requirements are moving towards including analysis from additional data sets.
In a competitive market, financial institutions will tend to give better terms of financing to those projects that have the
lowest risk financial return. An important component of the risk assessment is the confidence that can be placed in the
solar resource at the site location. Developers can reduce the perceived long term solar resource risk by:
• Comparing different data sources, assessing their uncertainty and judiciously selecting the most appropriate
data for the site location.
• Assessing the inter-annual variation in the solar resource in order to quantify the uncertainty in the revenue in
any given year.
There are a variety of possible solar irradiation data sources that may be assessed for the purpose of estimating the
irradiation at potential solar PV sites in India. The datasets either make use of ground-based measurements at well-
controlled meteorological stations or use processed satellite imagery.
The location of the 5 MW plant in Tamil Nadu was more than 200 km from the nearest MET station. It was, therefore,
necessary to rely on data interpolation between distant MET stations, and on data from satellites.
The image below compares the data obtained for the site location from three such data sources. There is a significant
discrepancy between them. A robust solar resource assessment would compare the data sources, discuss their uncertainty
and select the data most likely to represent the long term resource at the site location.
Where there is significant uncertainty in the data sources (or in the case of large capacity plants), a short term data
monitoring campaign may be considered. Short term monitoring (ideally up to one year in duration) may be used to
calibrate long term satellite-derived data and increase the confidence in the long term energy yield prediction.
A Guide For Developers and Investors 49
7.0
6.0
5.0
MET station interpolation
4.0 (20 years)
0.0
h
y
y
er
ber
ber
ary
st
Oc r
e
ri
Ma
rc
uar
be
Jul
Jun
gu
tob
Ap
Ma
vem
cem
u
em
Au
r
Jan
Feb
No
De
Sep
50 Utility Scale Solar Power Plants
4. PROJECT DEVELOPMENT
To move from concept to construction, a project must • Development – The development phase takes
pass through a number of development stages. The key the project from the feasibility study to financial
consideration during project development is the balance of closure. This involves moving the project forward
on a number of fronts including outline design and
expenditure and risk. There is no definitive detailed “road
selection of contractors.
map” for developing a solar PV project. The approach
taken will depend on the developer’s priorities and • Detailed Design – The key systems and structures
requirements such as risk profiles or deadlines, as well as site will be designed in detail. This will generally be
dependent parameters. completed by a contractor.
[32] The Sponsor(s) is (are) the party that owns the specific Project Company which
is set up to develop and own a specific project or portfolio. Contracts and other
agreements are entered into in the name of the Project Company. The Sponsor
provides financial (and other) resources, thereby allowing the Sponsor to retain
control of the project in a way that can allow it to minimise its exposure to risk.
A Guide For Developers and Investors 51
• The project site and boundary area. The feasibility phase will focus on the possible site or sites
outlined in the pre-feasibility study. It will take into account
• A conceptual design of the project, including each of the constraints in more detail and, if multiple sites are
estimation of installed capacity.
being assessed, should highlight the preferred site.
• The approximate costs for development, construction
and operation of the project and predicted revenue. A typical scope for a feasibility study would include:
developed as part of the pre-feasibility study, it may be • Information required for submission.
worthwhile assessing various design configurations at this stage
• Method of submission (online or via the planning
in order to ensure that an optimised design is selected. department office).
Specific tasks include: • Standard restrictions for the area of the development
(for example zoning regulations).
• Calculation of shade and initial PV plant layout. This
• Process for making amendments at a later date.
process of optimisation typically takes into account:
• Mounting frame or tracking system selection, • Preparation and submission of the permit
including consideration of site specific conditions. applications for the proposed solar PV project.
• Electrical cabling design and single line diagrams. • Preparation and submission of a grid
connection application.
• Electrical connections and monitoring equipment.
• Revision of the design and planning permissions.
• Grid connection design.
• Decision on contracting strategy (i.e. single EPC
• Full energy yield analysis using screened solar data contract or multi-contract).
and the optimised layout.
• Decision on the financing approach.
4.4.2 Planning Applications • Preparation of solar PV module
tender documentation.
Advice on planning documentation requirements in
the project area can be obtained from the local planning • Supplier selection and ranking.
department or from an experienced consultant. The type of • Preparation of construction tender documentation.
information that needs to be considered includes:
• Contractor selection and ranking.
• Permits or licences required.
• Contract negotiations.
• Timescales for submission and response.
A Guide For Developers and Investors 53
• Completion of a bankable energy yield. The EIA should consider the likely environmental effects of
the proposed development based upon current knowledge of
• Preparation of a financial model covering the full
the site and the surrounding environment. This information
lifecycle of the plant.
will determine what specific studies are required. The EIA
• Completion of a project risk analysis. should then assess ways of avoiding, reducing or offsetting
any potentially significant adverse effects. The studies will
• Environmental impact assessment.
also provide a baseline case that can be used in the future to
• Production of a detailed project report. determine the impact of the project.
• Securing financing for the project.
Guidance on the significance of impacts is mainly
of a generic nature. However, it is broadly accepted that
4.5.1 Bankable Energy Yield Prediction
this significance reflects the relationship between a number
In the development stage, a bank grade energy yield will be of factors:
required to secure finance. It is advised that this energy yield
• The magnitude or severity of an impact (that is, the
is either carried out or reviewed by an independent specialist. actual change taking place to the environment).
This will ensure that confidence can be placed in the results
and will help attract investment. • The importance or value of the affected resource
or receptor.
The energy yield should include: • The duration involved.
• An assessment of the inter-annual variation and yield • The reversibility of the effect.
confidence levels.
• The number and sensitivity of receptors.
• Consideration of site-specific factors, including
soiling or snow, and the cleaning regime specified in The significance, importance or value of a resource is
the O&M contract. generally judged on the following criteria:
• Full shading review of the PV generator including • The land’s designated status within the land use
near and far shading. planning system.
• Detailed losses. • The number of individual receptors.
• A review of the proposed design to ensure that • An empirical assessment on the basis of
parameters are within design tolerances. characteristics such as rarity or condition.
The main output of the development phase will be a • Spare parts inventory cost.
detailed project report. This will be used to secure finance
from banks or investors (more information on financing is
• Connection cost for electricity and services.
in Section 13). The information should be project-specific • Details of the permitting and planning status.
including all relevant information in a professional and
clear format. The items detailed below give examples of the
• Environmental impact assessment, restrictions and
mitigation plans.
information that should be included:
4.5.4 Contract Strategy
• Site layout (showing the location of modules,
inverters and buildings). Indicative plans showing:
There are two main contracting strategies that a developer
• Mounting frame and module layout. may consider: multi-contract and single EPC contract.
• Losses assumed with regard to the energy There is no single preferred contracting approach. The
yield forecast.
approach taken will depend on the experience, capabilities and
• Financial model inputs including: cost sensitivity of the developer.
• Availability assumptions.
• Geopolitical.
• Overall single line diagrams.
• Accessibility.
• MV & LV switchgear line diagrams.
• Grid connection.
• Protection systems.
• Module soiling.
• Interconnection systems and design.
• Water availability.
• Auxiliary power requirements.
• Financial incentives.
• Control systems.
“Showstoppers” for developing a utility scale PV power plant
The civil engineering items should be developed to a level
in a specific location may include constraints due to a low solar
suitable for construction. These will include designs of array
resource, low grid capacity or insufficient area. However, a low
foundations and buildings, as well as roads and infrastructure
solar resource could be offset by high local financial incentives,
required for implementation and operation. The design basis
making a project viable. A similar balancing act applies to the
criteria should be determined in accordance with national
other constraints. These are discussed further below.
standards. The wind loadings should be calculated to ensure
that the design will be suitable for the project location.
56 Utility Scale Solar Power Plants
A high average annual GHI is the most basic consideration Depending on the site location (latitude) and the type
for developing a solar PV project. The higher the resource, the of PV module selected (efficiency), a well-designed PV
greater the energy yield per kWp installed. When assessing the power plant with a capacity of 1 MWp developed in India
GHI at a site, care must be taken to minimise any shading that is estimated to require between one and two hectares
will reduce the irradiation actually received by the modules. (10,000 to 20,000 m2) of land. A plant using lower efficiency
Shading could be due to mountains or buildings on the far CdTe thin film modules may require approximately 40 to 50%
horizon, or mutual shading between rows of modules, or more space than a plant using poly-crystalline modules. Figure
shading near the location due to trees, buildings or overhead 15 shows a large ground mounted plant.
cabling. Avoiding shading is critical as even small areas of
shade may significantly impair the output of a module or 5.2.3 Climate
string of modules. The loss in output could be more than
predicted by simply assessing the proportion of the modules In addition to a good solar resource, the local climate should
that are shaded. not suffer from extremes of weather that will increase the
risk of damage or downtime. Weather events that may need
When assessing shading, it must be remembered that the consideration include:
path the sun takes through the sky changes with the seasons.
An obstacle that provides significant shading at mid-day in • Flooding – May increase the risk of erosion of
support structure and foundations, depending on
December may not provide any shading at all at mid-day in
geo-technical conditions.
June. The shading should be assessed using the full sun-path
diagram for the location. • High wind speeds – The risk of a high wind event
exceeding the plant specifications should be assessed.
Locations with a high risk of damaging wind speeds
5.2.2 Area
should be avoided. Fixed systems do not shut down
at high wind speeds, but tracking systems must shut
The area required per kWp of installed power varies with the down in safe mode when speeds of 16-20 m/s are
technology chosen. The distance between rows of modules (the exceeded.
pitch) required to avoid significant inter-row shading varies
with the site latitude. Sites should be chosen with sufficient
• Snow – Snow settling on modules can significantly
reduce annual energy yield if mitigating measures
area to allow the required power to be installed without having are not taken. If the site is prone to snow, then one
to reduce the pitch to levels that cause unacceptable yield loss. has to consider factors such as extra burden on the
mounting structures, the loss in energy production
and the additional cost of higher specification
A Guide For Developers and Investors 57
5.2.4 Topography
modules or support structures. The cost of removing Ideally, the site should be flat or on a slight south facing
the snow needs to be weighed against the loss in (in the northern hemisphere) slope. Such topography
production and the likelihood of further snowfall. makes installation simpler, and reduces the cost of technical
The effects of snow can be mitigated by a design modifications required to adjust for undulations in the
with a high tilt angle and frameless modules. The
ground. With additional cost and complexity of installation,
design should also ensure that the bottom edge of the
module is fixed higher than the average snow level mounting structures can be designed for most locations. In
for the area. A site that has regular coverings of snow general, the cost of land must be weighed against the cost of
for a long period of time may not be suitable for designing a mounting structure and installation time.
developing a solar PV plant.
Best practice dictates that either boreholes or trial pits A grid connection of sufficient capacity is required to
are made at regular intervals and at a depth appropriate for enable the export of power. The viability of grid connection
the foundation design. The boreholes or trial pits would will depend on three main factors: capacity, availability and
typically assess: proximity. These factors should be considered thoroughly at
an early stage of a project; otherwise, the costs could become
• The groundwater level. prohibitive if the site is later found to be in an unfavourable
• The resistivity of the soil. area for grid connection.
• The load-bearing properties of the soil. • Capacity – The capacity for the grid to accept
exported power from a solar plant will depend on the
• The presence of rocks or other obstructions. existing network infrastructure and current use of
the system. The rating of overhead lines, cables and
• The soil pH and chemical constituents in order to
transformers will be an important factor in assessing
assess the degree of corrosion protection required and
the connection capacity available. Switchgear fault
the properties of any cement to be used.
levels and protection settings may also be affected
The geotechnical study may also be expected to include an by the connection of a generation plant. In cases
where a network does not have the existing capacity
assessment of the risk of seismic activity and the susceptibility
to allow connection, there are two options available:
to frost, erosion and flooding. 1) to reduce peak power export to the allowable
limits of the network or 2) to upgrade the network to
5.2.6 Access allow the desired export capacity. Network upgrade
requirements will be advised by the network operator.
The site should allow access for trucks to deliver plant and But some aspects of that upgrade can be carried out
by contractors other than the network operator. Initial
construction materials. This may require upgrading existing
investigation into network connection point capacity
roads or building new ones. At a minimum, access roads can often be carried out by reviewing published data.
should be constructed with a gravel chip finish or similar. However, discussion with the network operator will be
The closer the site is to a main access road, the lower the cost required to fully establish the scope of work associated
of adding this infrastructure. Safe packaging of the modules with any capacity upgrades.
and their susceptibility to damage in transport must also be
• Proximity – A major influence on the cost of
carefully considered. connecting to the grid will be the distance from the
site to the grid connection point. Sites should be
The site should be in a secure location where there is little at locations where the cost of grid connection does
risk of damage from either people or wildlife. It should ideally not adversely affect project economics. Besides,
be in a location where security and maintenance personnel can a higher connection voltage will entail increased
cost of electrical equipment such as switchgear
respond quickly to any issue and this requirement should be
and transformers, as well as a higher conductor
stipulated in the maintenance contract. specification. A higher voltage is also likely to increase
the time taken to provide the connection resulting in a
longer development period.
yield from a plant may be significantly reduced if Clearances from the military may be required if the site is in
the grid has significant downtime. This may have or near a military-sensitive area. Glare from solar modules can
adverse effects on the economics of the project. affect some military activities.
Availability statistics should be requested from
the network operator to establish the expected
downtime of the network. In developed areas, the 5.2.9 Module Soiling
availability of the grid is usually very high.
If the modules are soiled by particulates, then the
5.2.8 Land Use efficiency of the solar plant could be significantly reduced.
It is, therefore, important to take into account local
Solar PV power plants will ideally be built on low value weather, environmental, human and wildlife factors while
land. If the land is not already owned by the developer, then determining the suitability of a site for a solar PV plant.
the cost of purchase or lease needs to be considered. The The criteria should include:
developer must purchase the land or rights for the duration
of the project. Besides access to the site, provision of water, • Dust particles from traffic, building activity,
electricity supplies and the rights to upgrade access roads agricultural activity or dust storms.
must be considered along with relevant land taxes. • Module soiling from bird excreta. Areas close to
nature reserves, bird breeding areas and lakes should
Since government permission will be required to build a be carefully assessed.
solar plant, it is necessary to assess the site in line with the
local conditions imposed by the relevant regulatory bodies. Soiling of modules may require an appropriate maintenance
If the land is currently used for agricultural purposes, then it and cleaning plan at the site location.
may need to be re-classified for “industrial use” with cost and
time implications—and the possibility of outright rejection. 5.2.10 Water Availability
The future land use of the area must also be taken into Clean, low mineral content water is preferred for cleaning
account. It is likely that the plant will be in operation for modules. A mains water supply, ground water, stored water
at least 25 years. As such, extraneous factors need to be or access to a mobile water tank may be required; the cost
considered to assess the likelihood of their impact on energy of the various options will have an impact on the project
yield. For example, the dust associated with building projects economics. The degree to which water availability is an issue
could have significant impact on the energy yield of the plant. will depend upon the expected level of module-soiling, the
extent of natural cleaning due to rainfall and the required
Locating the plant in an environmentally sensitive area cleaning frequency.
should be avoided. Government stipulated environmental
impact assessments or plant/wildlife studies will slow down 5.2.11 Financial Incentives
and potentially stop the development of a project.
Financial incentives (such as feed-in tariffs or tax breaks) in
Any trees on the project site and surrounding land may different countries, or regions within countries, have a strong
need to be felled and removed, with associated cost. bearing on the financial viability of a project. Such incentives
could outweigh the costs associated with one or more of the
site selection constraints.
60 Utility Scale Solar Power Plants
greater reflection and so higher levels of diffuse irradiation. The uncertainty of energy yield simulation software depends
For example, fresh grass has an albedo factor of 0.26, reducing on each modelling stage and on the uncertainty in the input
down to a minimum of approximately 0.15 when dry. Asphalt variables. Modelling software itself can introduce uncertainty
has a value between 0.09 and 0.15 or 0.18 if wet. of 2% to 3%.
The predicted annual energy yield may be expressed within The energy yield depends linearly, to a first approximation,
a given confidence interval. A P90 value is the annual energy on plane of array irradiance. Therefore uncertainty in the
yield prediction that will be exceeded with 90% probability; resource data has a stronger bearing on the uncertainty in
P75 is the yield that will be exceeded with 75% probability; the yield prediction than does the accuracy of PV modelling.
while P50, the expected value, is the annual energy yield Total uncertainty figures of up to 10% may be expected. A
prediction that will be exceeded with 50% probability. Good good energy yield report will quantify the uncertainty for the
quality energy yield reports used by investors will give the P50 specific site location.
and P90 energy yield prediction values as a minimum.
[33] “Quality of Meteonorm Version 6.0”, Jan Remund, World Renewable Energy
Congress 2008.
62 Utility Scale Solar Power Plants
Losses due to soiling (dust and bird droppings) depend on the environmental conditions,
rainfall frequency and on the cleaning strategy as defined in the O&M contract. This loss
can be relatively large compared to other loss factors but is usually less than 4%, unless there
Soiling
is unusually high soiling or problems from snow settling on the modules for long periods of
time. The soiling loss may be expected to be lower for modules at a high tilt angle as inclined
modules will benefit more from the natural cleaning effect of rainwater.
A Guide For Developers and Investors 63
Most PV modules do not match exactly the manufacturer’s nominal specifications. Modules
are sold with a nominal peak power and a guarantee of actual power within a given tolerance
Module quality
range. The module quality loss quantifies the impact on the energy yield due to divergences
in actual module characteristics from the specifications.
Losses due to “mismatch” are related to the fact that the modules in a string do not all present
Module mismatch exactly the same current/voltage profiles; there is a statistical variation between them which
gives rise to a power loss.
Electrical resistance in the cable between the modules and the input terminals of the inverter
DC cable resistance give rise to ohmic losses (I2R). This loss increases with temperature. If the cable is correctly
sized, this loss should be less than 3% annually.
Inverter performance Inverters convert from DC into AC with an efficiency that varies with inverter load.
This includes transformer performance and ohmic losses in the cable leading
AC losses
to the substation.
Downtime is a period when the plant does not generate due to failure. The downtime periods
Downtime will depend on the quality of the plant components, design, environmental conditions,
diagnostic response time and the repair response time.
64 Utility Scale Solar Power Plants
Degradation The performance of a PV module decreases with time. If no independent testing has been
conducted on the modules being used, then a generic degradation rate depending on the
module technology may be assumed. Alternatively, a maximum degradation rate that
conforms to the module performance warranty may be considered.
MPP tracking The inverters are constantly seeking the maximum power point (MPP) of the array
by shifting inverter voltage to the MPP voltage. Different inverters do this with
varying efficiency.
Curtailment of
Yield loss due to high winds enforcing the stow mode of tracking systems.
tracking
Auxiliary power Power may be required for electrical equipment within the plant. This may include security
systems, tracking motors, monitoring equipment and lighting. It is usually recommended to
meter this auxiliary power requirement separately.
Grid Compliance Loss This parameter has been included to draw attention to the risk of a PV power plant losing
energy through complying with grid code requirements. These requirements vary on a country
to country basis.
A Guide For Developers and Investors 65
Figure 16 represents the typical combined uncertainties in uncertainty in energy yield when inter-annual variability is
the yield prediction for a PV power plant. The dashed blue combined with the uncertainty in the yield prediction. The
line shows the predicted P50 yield. The green lines represent total uncertainty decreases over the lifetime of the PV plant.
uncertainty in energy yield due to inter-annual variability The lower limit on the graph corresponds to the P90 and the
in solar resource. The solid red lines represent the total upper limit corresponds to the P10.
66 Utility Scale Solar Power Plants
Case Study 2
Energy Yield Prediction
The developer of a the 5MW plant in Tamil Nadu required a solar energy yield prediction to confirm project feasibility
and assess likely revenues. The developer did not consider the range of available input data or conduct a long term
yield prediction over the life of the project; both of these would have been useful to derive a more accurate yield
figure, particularly for potential project financiers.
The developer sourced global horizontal irradiation data for the site location (see case study 1). Commercially available
software was used to simulate the complex interactions of temperature and irradiance impacting the energy yield. This
software took the plant specifications as input and modelled the output in hourly time steps for a typical year. Losses
and gains were calculated within the software. These included:
The software gave an annual sum of electrical energy expected at the inverter output in the first year of operation.
Although this is a useful indicative figure, an improved energy yield prediction would also consider:
• Near shading from nearby obstructions including poles, control rooms and switch yard equipment.
• AC losses.
• Degradation of the modules and plant components over the lifetime of the plant.
The results will ideally show the expected output during the design life of the plant and assess the confidence in the
energy yield predictions given by analysing:
The energy yield prediction for the 5MW plant was provided as a first year P50 value (the value expected with
50% probability in the first year) excluding degradation. The confidence that can be placed in the prediction would
typically be expressed by the P90 value, the annual energy yield prediction that will be exceeded with 90% probability.
Projects typically, have a financing structure that requires them to service debt once or twice a year. The year on year
uncertainty in the resource is therefore taken into account by expressing a “one year P90”. A “ten year P90” includes
the uncertainty in the resource as it varies over a ten-year period. The exact requirement will depend on the financial
structure of the project and the requirements of the financing institution.
68 Utility Scale Solar Power Plants
7. PLANT DESIGN
Designing a megawatt-scale PV power plant is a complex thin film amorphous silicon, it should be realised that
each technology has examples of high quality and low
process that requires considerable technical experience and
quality products from different manufacturers.
knowledge. There are many compromises that need to be
made in order to achieve the optimum balance between • Different technologies have a differing spectral
performance and cost. This section highlights some of the key response and so will be better suited for use in certain
issues that need to be considered when designing a PV plant. locations, depending on the local light conditions.
• Other parameters important for selection of modules module manufacturer has offered a power output
include: cost ($/Wp), lifetime, and maximum guarantee beyond 25 years. The conditions listed in
system voltage. both the power guarantee and product guarantee are
important, and vary between manufacturers.
7.1.2 Quality Benchmarks
• Lifetime – Good quality modules with the
• Product guarantee – In the EU, manufacturers appropriate IEC certification have a design life in
are legally bound to provide a product guarantee excess of 25 years. Beyond 30 years, increased levels
ensuring that modules will be fully functional for of degradation may be expected. The lifetime of
a minimum of 2 years. Some companies guarantee crystalline modules has been proven in the field. Thin
a longer period, with 5-6 years being the usual film technology lifetimes are currently unproven and
duration. rely on accelerated lifetime laboratory tests, but are
expected to be in the order of 25-30 years also.
• Power guarantee – In addition to the product
guarantee, most manufacturers grant nominal power The module data sheet format and the information that
guarantees. These vary between manufacturers but a should be included has been standardised and is covered
typical power guarantee stipulates that the modules by EN 50380, which is the “data sheet and nameplate
will deliver 90% of the original nominal power information for photovoltaic modules”. An example of the
after 10 years and 80% after 25 years. So far no
information expected in a data sheet is provided in Table 6.
7.1.2 Inverters
No single inverter concept is best for all situations. In efficiency are major inverter selection criteria, directly affecting
practice, the local conditions and the system components have the annual revenue of the solar PV plant. It is also important
to be taken into account to tailor the system for the specific to bear in mind that efficiency varies according to a number
application. Different solar PV module technologies and of factors. Of them, DC input voltage and percentage load are
layouts may suit different inverter types. So care needs to be the two dominant factors. Several other factors should inform
taken in the integration of modules and inverters to ensure inverter selection, including site temperature, product reliability,
optimum performance and lifetime. maintainability, serviceability and total cost of ownership. A
thorough financial analysis is required to determine the most
Among the major selection criteria for inverters, the cost-effective inverter option. Many of the inverter selection
financial incentive scheme and the DC-AC conversion criteria listed in Table 7 may feed into this analysis.
The guarantee offered for inverters varies among Mounting structures will typically be fabricated from steel
manufacturers. A minimum guarantee of two years is typical, or aluminium, although there are examples of systems based
with optional extensions of up to twenty years or more. A on wooden beams. A good quality mounting structure may be
2009 survey of inverters[35] showed that only one inverter expected to:
manufacturer in the 100-500 kW range offers a guarantee
longer than 20 years. • Have undergone extensive testing to ensure
the designs meet or exceed the load conditions
experienced at the site.
While many manufacturers quote MTBF of 20 years or
more, real world experience shows that inverters generally need • Allow the desired tilt angle to be achieved within a
to be replaced every five to ten years. Based on a 2006 study[36], few degrees.
investment in a new inverter is required three to five times over
• Allow field adjustments that may reduce installation
the life of a PV system. time and compensate for inaccuracies in placement
of foundations.
Inverter protection should include:
• Minimise tools and expertise required for installation.
• Incorrect polarity protection for the DC cable. • Adhere to the conditions described in the module
• Over-voltage and overload protection. manufacturer’s installation manual.
• Islanding detection for grid connected systems • Allow for thermal expansion, using expansion joints
(depends on grid code requirements). where necessary in long sections, so that modules do
not become unduly stressed.
• Insulation monitoring.
Purchasing good quality structures from reputable
Inverters should be accompanied by the appropriate manufacturers is generally a low-cost, low-risk option. Some
type test certificates, which are defined by the national and manufacturers provide soil testing and qualification in order to
international standards applicable for each project and country. certify designs for a specific project location.
The inverter datasheet format and the information that Alternatively, custom-designed structures may be used to
should be included has been standardised and is covered solve specific engineering challenges or to reduce costs. If this
by EN 50524:2009 – “Data sheet and name plate for route is chosen, it is important to consider the additional
photovoltaic inverters”. An example of the information liabilities and cost for validating structural integrity. This apart,
expected in a datasheet is provided in Table 8. systems should be designed to ease installation. In general,
installation efficiencies can be achieved by using commercially
available products.
4. Choosing a tilt angle that optimises the annual describe the celestial motion of the sun throughout the year for
energy yield according to the latitude of the site and any location on earth, plotting its altitude[38] and azimuth[39]
the annual distribution of solar resource. angle on a sunpath diagram as shown in Figure 17. This, along
5. Orientating the modules to face a direction that with information on the module row spacing, may be used to
yields the maximum annual revenue from power
production. In the northern hemisphere, this will 1. Calculate the degree of shading and
usually be true south[37].
2. Simulate the annual energy losses associated with
Computer simulation software could be used to help design
various configurations of tilt angle, orientation
the plant layout. Such software includes algorithms which
and row spacing.
75
60
Sun Height [˚]
45
30
15 Behind Behind
The Plane The Plane
0
-180 -150 -120 -90 -60 -30 0 30 60 90 120 150 180
Azimuth [˚]
Figure 17: Sun-path Diagram for Chennai, India[40]
[37] True south differs from magnetic south, and an adjustment should be made [39] The azimuth is the location of the sun in terms of north, south, east and west.
from compass readings. Definitions may vary but 0° represents true south, -90° represents east, 180°
[38] The elevation of the sun above the horizon (the plane tangent to the Earth‘s represents north, and 90° represents west.
surface at the point of measurement) is known as the angle of altitude. [40] PVSYST V5.06
76 Utility Scale Solar Power Plants
Minimising cable runs and associated electrical losses may • Shading – More highly tilted modules provide
suggest positioning an LV/MV station centrally within the more shading on modules behind them. As
plant. If this option is chosen, then adequate space should shading impacts energy yield much more than may
be expected simply by calculating the proportion
be allocated to avoid the possibility of the station shading
of the module shaded, a good option (other than
modules behind it. spacing the rows more widely apart) is to reduce
the tilt angle. It is usually better to use a lesser tilt
The layout should allow adequate distance from angle as a trade-off for loss in energy yield due to
the perimeter fence to prevent shading. It should also inter-row shading.
incorporate access routes for maintenance staff and vehicles at
• Seasonal irradiation distribution – If a particular
appropriate intervals. season dominates the annual distribution of solar
resource (monsoon rains, for example), it may be
7.2.2 Tilt Angle beneficial to adjust the tilt angle to compensate for
the loss. Simulation software is able to assess the
Every location will have an optimal tilt angle that maximises benefit of this.
the total annual irradiation (averaged over the whole year) on
7.2.3 Inter-Row Spacing
the plane of the collector. For fixed tilt grid connected power
plants, the theoretical optimum tilt angle may be calculated The choice of row spacing is a compromise chosen to reduce
from the latitude of the site. However, adjustments may need inter-row shading while keeping the area of the PV plant
to be made to account for: within reasonable limits, reducing cable runs and keeping
ohmic losses within acceptable limits. Inter-row shading can
• Soiling – Higher tilt angles have lower soiling losses.
The natural flow of rainwater cleans such modules never be reduced to zero: at the beginning and end of the day
more effectively and snow slides off more easily at the shadow lengths are extremely long. Figure 18 illustrates the
higher angles. angles that must be considered in the design process.
h
α
β
a b
d
Figure 18: Shading Angle Diagram (Image courtesy of Schletter GmbH)
A Guide For Developers and Investors 77
The shading limit angle[41] α is the solar elevation angle south. For example, an array facing in a westerly direction will
beyond which there is no inter-row shading on the modules. be optimised to generate power in the afternoon. The effect of
If the elevation of the sun is lower than α then a proportion tilt angle and orientation on energy yield production can be
of the module will be shaded. Alongside, there will be an effectively modelled using simulation software.
associated loss in energy yield.
7.3 Electrical Design
The shading limit angle may be reduced either by reducing
the tilt angle β or increasing the row pitch d. Reducing the tilt For most large solar PV plants, reducing the levelised cost
angle below the optimal is sometimes a choice as this may give of electricity is the most important design criteria. Every
only a minimal reduction in annual yield. The ground cover aspect of the electrical system (and of the project as a whole)
ratio (GCR), given by l/d, is a measure of the PV module area should be scrutinised and optimised. The potential economic
compared to the area of land required. gains from such an analysis are much larger than the cost of
carrying it out.
For many locations a design rule of thumb is to space
the modules in such a way that there is no shading at It is important to strike a balance between cost savings
solar noon on the winter solstice (December 21st in the and quality. Engineering decisions should be ‘careful’ and
northern hemisphere). In general, if there is less than a 1% ‘informed’ decisions. Otherwise, design made with a view to
annual loss due to shading, then the row spacing may be reduce costs in the present could lead to increased future costs
deemed acceptable. and lost revenue due to high maintenance requirements and
low performance.
Detailed energy yield simulations can be carried out to assess
losses due to shading, and to obtain an economic optimisation The design of each project should be judged on a case-
that also takes into account the cost of land if required. by-case basis, as each site poses unique challenges and
constraints. While general guidelines and best practices can be
7.2.4 Orientation formulated, there are no “one-size-fits all” solutions. While the
recommendations in the following sections are based on solar
In the northern hemisphere, the orientation that PV plants with centralised inverter architectures, many of the
optimises the total annual energy yield is true south. In the concepts discussed also apply to plants with string inverters.
tropics, the effect of deviating from true south may not be
especially significant. The following sections are based mainly on European
practices. Practices will differ elsewhere. It is, therefore,
Some tariff structures encourage the production of energy crucial to bear in mind that in all cases the relevant national
during hours of peak demand. In such “time of day” rate and applicable international codes and regulations are
structures, there may be financial (rather than energy yield) consulted and followed, to ensure that the installation is safe
benefits of orientating an array that deviates significantly from and compliant.
7.3.1 DC System
The DC system comprises the following plant: higher than that given by standard multiplication factors. So,
this effect should also be taken into consideration. If in doubt,
• Array(s) of PV modules. the manufacturer should be consulted for advice.
• DC cabling (module, string and main cable).
7.3.1.1 PV Array Design
• DC connectors (plugs and sockets).
The design of a PV array will depend on the inverter
• Junction boxes/combiners.
specifications and the chosen system architecture besides the
• Disconnects/switches. specific context and conditions of use. Using many modules in
series in high voltage arrays minimises ohmic losses. However,
• Protection devices.
safety requirements, inverter voltage limits and national
• Earthing. regulations also need to be considered.
When sizing the DC component of the plant, the maximum • Maximum number of modules in a string –
voltage and current of the individual strings and PV array(s) The maximum number of modules in a string is
should be calculated using the maximum output of the defined by the maximum DC input voltage of the
individual modules. Simulation programs can be used to help inverter to which the string will be connected to
with sizing but their results should be cross checked manually. (VMax (Inv, DC)). Under no circumstances should this
voltage be exceeded. Crossing the limit can decrease
the inverter’s operational lifetime or render the device
For mono-crystalline and multi-crystalline silicon modules, inoperable. The highest module voltage that can
all DC components should be rated as follows, to allow for occur in operation is the open-circuit voltage in the
thermal and voltage limits[42]: coldest daytime temperatures at the site location.
Design rules of thumb for Europe use – 10°C as the
Minimum Voltage Rating: VOC(STC) ×1.15 minimum design temperature, but this may vary
according to location. The maximum number of
modules in a string (nMax) may therefore be calculated
Minimum Current Rating: ISC(STC) ×1.25
using the formula:
The multiplication factors used above (1.15 and 1.25) are VOC(Module)@Coldest Module Operating Temperature×NMax <VMax(Inv, DC)
location-specific and cover the maximum voltage and current
• Minimum number of modules in a string – The
values that can be expected under UK conditions of irradiance. minimum number of modules is governed by the
While different multiplication factors may apply for other requirement to keep the system voltage within the
locations, national standards should be consulted. MPPT range of the inverter. If the string voltage
drops below the minimum MPP inverter voltage,
For non-crystalline silicon modules, DC component then the system will underperform. In the worst case,
the inverter may shut down. The lowest expected
ratings should be calculated from manufacturer’s data, taking
module voltage occurs during the highest operating
into account the temperature and irradiance coefficients. In temperature conditions. Design rules of thumb for
addition, certain modules have an initial settling-in period, Europe use 70°C as the design benchmark, but this
during which the VOC and ISC output they produce is much may vary according to site conditions. The minimum
number of modules in a string (nMin) may therefore
be calculated using the formula:
[42] Guide to the installation of PV systems – DTI (2006), 2nd edition.
A Guide For Developers and Investors 79
in the plant yield can be achieved by successfully Guidance on inverter and PV array sizing can be obtained
matching the operating voltages of the PV array from the inverter manufacturers, who offer system sizing
with the inverter. software. Such tools also provide an indication of the total
• Number of strings – The maximum number of number of inverters required.
strings permitted in a PV array is a function of
the maximum allowable PV array current and the A number of factors and guidelines must be assessed when
maximum inverter current. In general, this limit sizing an inverter:
should not be exceeded as it leads to premature
inverter ageing and yield loss. • The maximum VOC in the coldest daytime
temperature must be less than the inverter maximum
7.3.1.2 Inverter Sizing DC input voltage (VInv, DC Max).
It is not possible to formulate an optimal inverter sizing • The inverter must be able to safely withstand the
strategy that applies in all cases. Project specifics such as the maximum array current.
solar resource and module tilt angle play a very important • The minimum VOC in the hottest daytime
role when choosing a design. While the rule of thumb has temperature must be greater than the inverter DC
been to use an inverter-to-array power ratio less than unity, turn-off voltage (VInv, DC Turn-Off ).
this is not always the best design approach. For example, this
• The maximum inverter DC current must be greater
option might lead to a situation where the inverter manages than the PV array(s) current.
to curtail power spikes not anticipated by irradiance profiles
(based on one hour data). Or, it could fail to achieve grid code • The inverter MPP range must include PV array MPP
compliance in cases where reactive power injection to the grid points at different temperatures.
is required. • When installed, some thin film modules produce
a voltage greater than the nominal voltage. This
The optimal sizing is, therefore, dependent on the specifics happens for a period of time until initial degradation
of the plant design. Most plants will have an inverter sizing has occurred, and must be taken into account to
prevent the inverter from being damaged.
range within the limits defined by:
• Grid code requirements: for example, reactive
0.8<Power Ratio<1.2 power injection.
Inverters with reactive power control are recommended. 3. The minimisation of cable losses. The cable
Inverters can control reactive power by controlling the phase voltage drop and the associated power losses must
angle of the current injection. Moreover, aspects such as be as low possible. Normally, the voltage drop must
be less than 3%, but national regulations must be
inverter ventilation, air-conditioning, lighting and cabinet
consulted for guidance. Cable losses of less than 1%
heating must be considered. are achievable.
When optimising the voltage, it should be borne in In practice, the minimisation of voltage drop and associated
mind that the inverter efficiency is dependent on voltage. losses will be the limiting factor in most cases.
Specification sheets and voltage dependency graphs are
required for efficient voltage-matching. 7.3.1.4 Cable Management
7.3.1.3 Cable Selection and Sizing DC cabling consists of module, string and main cables.
Issues such as routing the main cables, and proper laying
The selection and sizing of DC cables for solar PV plants and trenching are to be considered in the detailed design of
should take into account national codes and regulations a solar PV plant. Additionally, management of overground
applicable to each country. Cables specifically designed for cables (for example, module cables) also needs attention.
solar PV installations (“solar” cables) are readily available and Importantly, these cables need to be properly routed and
should be used. In general, three criteria must be observed secured to facilitate commissioning and troubleshooting.
when sizing cables: Proper management ensures that cables are properly protected
from inclement weather and extraneous factors (for example,
1. The cable voltage rating. The voltage limits of abrasion on the sharp edges of the support structures).
the cable—to which the PV string or array cable
will be connected—must be taken into account.
A number of cable connection systems are available:
Calculations of the maximum VOC voltage of the
modules, adjusted for the site minimum design
temperature, are used for this calculation. • Screw terminals.
For module cables the following should apply[44]: In an array comprising of N strings connected in parallel
and M modules in each string, as shown in Figure 21, sizing of
Minimum Voltage Rating = VOC(STC) ×1.15 cables should be based on the following:
Minimum Current Rating = ISC(STC) ×1.25 • Array with no string fuses (applies to arrays of three or
fewer strings only)[45].
The cables should be rated to the highest temperature
Voltage: VOC(STC) ×M×1.15
they may experience ( for instance, 80°C). Appropriate de-
rating factors for temperature, installation method and cable
Current: I(SC(STC)) ×(N - 1)×1.25
configuration should also be applied.
• Array with string fuses[45].
Single conductor, double insulation cables are preferable
for module connections. Using such cables helps protect Voltage Rating = VOC(STC) ×M×1.15
against short-circuits. When sizing string cables, the number
of modules and the number of strings per array need to be Current Rating = ISC(STC) ×1.25
considered. The number of modules defines the voltage at
which the cable should be rated. The number of strings is Usually, single conductor, halogen-free cables are preferred[46].
used to calculate the maximum reverse current that can flow If there is a high risk of lightning, cables should be screened.
through a string—especially, in case of a fault when there are Again, opting for “solar” cables is advisable, as they are designed
no string fuses. to meet the relevant requirements.
N number of strings
(connected in parallel) Strings Cables
Main DC Cable
Figure 20: PV Array Showing String Cables
[44] Guide to the installation of PV systems – DTI (2006), 2nd edition. Assumes [46] Halogen cables release corrosive and toxic gases if ignited.
mono- and multi-crystalline silicon modules, and UK conditions for the
multiplication factors.
A Guide For Developers and Investors 83
The formulae that guide the sizing of main DC cables incentive schemes such as FiTs). The increased costs for higher
running from the PV array to the inverter, for a system as cross section cables are thus amortised much faster.
shown in Figure 20, are given below[45]:
7.3.1.7 Junction Boxes
Minimum Voltage Rating = VOC(STC) ×M×1.15
Junction boxes or combiners are needed at the point where the
Minimum Current Rating = ISC(STC) ×N×1.25
individual strings forming an array are marshalled and connected
together in parallel before leaving for the inverter through the main
Standard de-rating factors must be also applied, as described DC cable. Junctions are usually made with screw terminals and must
in Section 7.3.1.3. be of high quality to ensure lower losses and to prevent overheating.
In order to reduce losses, the overall voltage drop between Junction boxes have protective and isolation equipment
the PV array and the inverter (at STC) should be minimised. like string fuses and disconnects (also known as load break
A benchmark voltage drop of less than 3% is suitable, and switches)[47], and must be rated for outdoor placement using, for
cables should be sized to reflect this benchmark. In most example, ingress protection (IP) 65. An explanation of the IP
cases, over sizing cables to achieve lower losses is a worthwhile (International Protection Rating) bands is provided in Table 9.
investment, since the allocated price for solar PV energy is Depending on the solar PV plant architecture and size, multiple
usually much higher than the normal market price (due to levels of junction boxes can be used.
[45] Assuming mono- and multi-crystalline silicon modules, and UK conditions for [47] Disconnects should be not confused with disconnectors/isolator that are dead circuit
the multiplication factors. Relevant national standards should be consulted. devices (or, devices that operate when there is no current flowing through the circuit).
84 Utility Scale Solar Power Plants
7.3.1.8 Connectors
It is important to remember that the DC side of a PV Specialised plug and socket connections for PV applications
system cannot be switched off and that the terminals remain have been developed, and are normally pre-installed to module
live during the day. Therefore clear and visible warning cables to facilitate assembly. These plug connectors, which
signs should be provided to inform anyone operating on provide secure and touch-proof connections, are currently the
the junction box. Furthermore, all junction boxes should be preferred choice.
properly labelled as per national regulations for safety.
Connectors should be correctly rated and used for DC
As a precaution, disconnects and string fuses should be applications. As a rule, the connector current and voltage
provided. Disconnects permit the isolation of individual ratings should be at least equal to those of the circuit they are
strings, while string fuses protect against faults, as discussed in installed on.
Section 7.3.1.10. Disconnects should be capable of breaking
normal load and should be segregated on both the positive and Connectors should carry appropriate safety signs that warn
negative string cables. against disconnection under load. Such an event can lead to
arcing (producing a luminous discharge across a gap in an
To ensure protection against short-circuits, it is electrical circuit), and put personnel and equipment in danger.
recommended that: Any disconnection should take place only after the circuit has
been properly isolated.
• The junction box enclosure is fabricated from non-
conductive material. In order to avoid errors during installation, the design of
• The positive and negative busbars are adequately DC connectors should be incompatible with AC connectors.
separated and segregated.
7.3.1.9 String Fuses
• The enclosure layout should be such that short-
circuits during installation and maintenance are
extremely unlikely. The main function of string fuses is to protect strings from
over-currents. They must be designed for DC operation.
The system should be designed so that the main DC cable Miniature fuses are normally used in PV applications. National
exiting the junction box can easily transition to a trench. codes and regulations must be consulted when selecting and
sizing fuses.
7.3.1.10 DC Switching
designer must also consult the module manufacturer DC switching is installed in the DC section of a solar
to ensure the module can withstand these reverse PV plant to provide protection and isolation capabilities.
currents. Importantly, while it might not be necessary DC switches/disconnects and DC circuit breakers are
to fit fuses in systems consisting of two or three
discussed below.
strings, it might still be beneficial to do so as this can
facilitate testing and tracing of faults.
DC Switches/Disconnects – Judicious design practice calls
• Since faults can occur on both the positive and for the installation of switching devices in PV array junction
negative sides, fuses must be installed on all boxes. DC switches provide a means of manually electrically
unearthed cables.
isolating entire PV arrays, which is required during installation
• To avoid nuisance tripping, the nominal current of and maintenance. DC switches must be:
the fuse should be at least 1.25 times greater than
the nominal string current. Overheating of fuses can • Double pole to isolate both the positive and negative
also cause nuisance tripping. For this reason, junction PV array cables.
boxes should be kept in the shade.
• Rated for DC operation.
• The string fuse must trip at less than twice the string
short-circuit current at STC or at less than the string • Capable of breaking under full load.
cable current carrying capability, whichever is the
lower value. • Rated for the system voltage and maximum
current expected.
• The trigger current of string fuses should be taken
into account when sizing string cables. It should • Equipped with proper safety signs.
not be larger than the current at which the string
DC Circuit Breaker – String fuses cannot be relied upon
cable is rated.
for disconnection of supply in case of fault conditions. This is
• The string fuse must be rated for operation at the due to the fact that PV modules are current-limiting devices,
string voltage using the formula[48]: with a short-circuit current only a little higher than the
nominal current. In other words, the fuse would not blow
String Fuse Voltage Rating= VOC(STC) ×M×1.15
since the fault current would be less than the trigger current.
For this reason, most PV codes and regulations recommend
Miniature circuit breakers (MCBs) can also be used for over-
that main DC circuit breakers (CB) should be installed
current protection, but they are less common than fuses due to
between the PV array fields and the grid connected inverters.
their higher cost.
[48] Assuming mono- and multi-crystalline silicon modules, and UK conditions for
the multiplication factors. Relevant national standards should be consulted.
86 Utility Scale Solar Power Plants
• Have a wide temperature range (-55 to 125°C). Cabling for AC systems should be designed to provide
a safe, economic means of transmitting power from the
• Be resistant to ultraviolet (UV) radiation and
weather if laid outdoors without protection. inverters to the transformers and beyond. Cables should
be rated for the operating voltage. They should also have
• Be single core and double insulated. conductors and screens sized for operating currents and
• Have mechanical resistance (animal proof, short circuit currents.
compression, tension and bending).
It is important that the cable chosen is specified correctly.
Apart from these, the risk of earth faults and short The following should be considered when designing
circuits should be minimised. Consequently, cable the cabling:
runs should be kept as short as possible. The following
cable options are preferable because they offer • The cable must be rated for the maximum
increased protection: expected voltage.
• A suitable number of cores should be chosen (either Switchgear type will largely be dependent on the voltage
single or multi-core). of operation. Switchgear up to 33 kV is likely to be an
internal metal clad cubicle type. Also, it will have gas – or
• Earthing and bonding should be suitably designed
for the project application. air-insulated busbars, and vacuum or SF6 breakers. For higher
voltages, the preferred choice will most likely be air-insulated
• The installation method and mechanical outdoor switchgear or, if space is an issue, gas-insulated
protection of the cable should be suitably designed
indoor switchgear.
for the project.
Cables should comply with relevant IEC standards or All switchgear should:
national standards. Examples of these include:
• Be in accordance with relevant IEC and
• IEC 60502 for cables between 1 kV and 36 kV. national standards.
• IEC 60364 for LV cabling (BS 7671 in UK). • Have the option to be secured by locks in off/
earth positions.
• IEC 60840 for cables rated for voltages above 30 kV
and up to 150 kV. • Clearly show the ON and OFF positions with
appropriate labels.
7.3.2.2 AC Switchgear
• Be rated for operational and short circuit currents.
Appropriately rated switchgear and protection systems • Be rated for the correct operational voltage.
should be provided throughout the electrical system to provide
• In the case of HV switchgear, have remote switching
disconnection, isolation, earthing and protection for the capability.
various components of the plant. On the output side of the
inverters, provision of a switch disconnector is recommended • Be provided with suitable earthing.
as a means to isolate the PV array. Other switchgear may be
It should be noted that HV switching is a hazardous
required, depending on the size and set up of the electrical
procedure, and safety measures to minimise risk should be
infrastructure in the solar plant.
adopted as good practice.
88 Utility Scale Solar Power Plants
The purpose of transformers in a solar power plant is to Other issues to consider when designing and specifying a
provide suitable voltage levels for transmission across the site transformer may also include:
and for export to the grid. In general, the inverters supply
power at LV. But for a commercial solar power plant, grid • Tap setting requirements.
connection is typically made at upwards of 11 kV (HV levels). • Cooling medium.
PV Array
PV Array
PV Array
PV Array
LV <1000
Inverters outputing at LV
MV 33000-1000v
LV/MV Transformers
MV collection switchgear
MV/HV Transformers
HV >33000v
GRID
Figure 21: Typical Transformer Locations and Voltage Levels in a Solar Plant where Export to Grid is at HV
Delivery and Commission – Consideration should be transportation will still be significant and road delivery may
given to the period of time required for manufacture and require special measures such as police escort.
delivery of transformers. Most large transformers will be
designed and built on order, and will therefore have a lengthy The positioning of the transformer in the power plant
lead time, which can stretch to several years. should also be decided at the planning stage. By doing this,
a transformer can be easily and safely installed, maintained
The delivery of large transformers to the site can also and—in the event of a failure—replaced. Liquid-filled
be a problem. Large transformers can be broken down to transformers should be provided with a bund to catch any
some extent, but the tank containing the core and winding leakage. Oil-filled transformers, if sited indoors, are generally
will always need to be moved in one piece. In the case of considered a special fire risk. As such, measures to reduce the
transformers around the 100 MVA size, the burden of risk to property and life should be considered.
90 Utility Scale Solar Power Plants
7.3.2.4 Substation
The substation houses the primary and secondary electrical Metering – Tariff metering will be required to measure the
equipment for the central operation of the solar plant and export of power. This may be provided at the substation or
connection to the local electricity grid. The substation can also at the point of connection to the grid. Current transformers
provide an operational base for staff required for operation and voltage transformers provided in the switchgear will be
and maintenance as well as stores or other auxiliary functions connected to metering points by screened cable.
associated with the solar plant. Equipment such as the LV/MV
transformers, MV switchgear, SCADA (Supervisory Control Data Monitoring/SCADA – SCADA systems provide
and Data Acquisition) systems, protection and metering control and status indication for the items included in the
systems can be placed within the substation. substation and across the solar plant. The key equipment
may be situated in the substation in control and protection
The layout of the substation should optimise the use of rooms. Air conditioning should be considered due to the heat
space while still complying with all relevant building codes and generated by the electronic equipment in the modules.
standards. A safe working space should be provided around the
plant for the operation and maintenance staff. Auxiliary equipment – The design of the substation should
take into account the need for auxiliary systems required for a
The substation may be wholly internal or may consist of functioning substation/control room. All auxiliary equipment
internal and external components such as transformers, HV should be designed to relevant standards and may include:
switchgear and backup generators. Separation between MV
switch rooms, converter rooms, control rooms, store rooms • LV power supplies.
and offices is a key requirement, besides providing safe access, • Back-up power supplies.
lighting and welfare facilities. In plants where the substation is
to be manned, care should be taken to provide facilities like a • Uninterruptible power supply (UPS) batteries.
canteen and washrooms. • Diesel generators.
Where HV systems are present, an earth mat may need • Auxiliary transformers and grid connections.
to be provided to obtain safe step/touch potentials and • Telephone and internet connections.
earth system faults. Earth mats should be installed prior
to setting the foundation. Lightning protection should • Lighting.
be considered to alleviate the effect of lightning strikes on • Heating Ventilation and Air Conditioning (HVAC).
equipment and buildings.
• Water supplies.
A trench is often required as a means for easing the routing
• Drainage.
of power and data cables to the substation.
• Fire and intruder alarms.
Where necessary, the substation may also need to
accommodate the grid company’s equipment (which
might be in a separate area of the building). Additional
equipment may include:
A Guide For Developers and Investors 91
The earthing of a solar PV plant influences a number of risk The earthing arrangements on each site will vary, depending
parameters, namely: on a number of factors:
• The risk of fire during a fault. • Installation guidelines for module manufacturers.
• Array frame earthing. While the system designer must decide the most appropriate
earthing arrangement for the solar PV plant, one can follow
• System earthing (DC conductor earthing). the general guidelines given below:
• Inverter earthing.
• Ground rods should be placed close to junction
• Lightning and surge protection. boxes. Ground electrodes should be connected
between the ground rod and the ground lug in the
Earthing should be provided as a means to protect against junction box.
electric shock, fire hazard and lightning. By connecting to the
• A continuous earth path is to be maintained
earth, charge accumulation in the system during an electrical throughout the PV array.
storm is prevented.
• Cable runs should be kept as short as possible.
The entire PV plant and the electrical room should be • Surge suppression devices can be installed at the
protected from lightning. Protection systems are usually inverter end of the DC cable and at the array
based on early streamer emission, lightning conductor air junction boxes.
terminals. The air terminal will be capable of handling
• Both sides of an inverter should be properly isolated
multiple strikes of lightning current and should be
before carrying out any work, and appropriate safety
maintenance-free after installation. signs should be installed as a reminder.
These air terminals will be connected to respective earthing • Many inverter models include internal surge
arrestors. Besides, separate additional surge protection
stations. Subsequently an earthing grid will be formed,
devices may be required. Importantly, national codes
connecting all the earthing stations through the required and regulations, and the specific characteristics of
galvanised iron tapes. each project must be taken into account.
92 Utility Scale Solar Power Plants
• Certification to current IEC and appropriate national • Office – A portable office and sanitary room with
standards such as IEC 62271 for HV switchgear and communication devices. This must be watertight and
IEC 61439 for LV switchgear. prevent entry to insects. It should be located to allow
easy vehicular access.
• Type testing to appropriate standards.
• LV/MV station – Inverters may either be placed
• A minimum warranty period of two years. amongst the module support structures (if string
inverters are chosen) in specially designed cabinets
• An expected lifetime at least equivalent to the design
or in an inverter house along with the medium
life of the project.
voltage transformers, switchgear and metering
system[49]. This “LV/MV station” may be equipped
Transformers should be supplied by reputable manufacturers
with an air conditioning system if it is required
accredited to ISO 9001. They should have: to keep the electrical devices within their design
temperature envelopes.
• Certification to IEC and appropriate national
standards such as IEC 60076 for the power • MV/HV station – An MV/HV station may be used
transformer, IEC 60085 for electrical insulation and to collect the AC power from the medium voltage
IEC 60214 for tap changers. transformers and interface to the power grid.
[49] For string inverters, the LV/MV station may be used to collect the AC power.
A Guide For Developers and Investors 93
• Communications – The plant monitoring PV power plants represent a large financial investment. The
system and the security system will require a modules are not only valuable but also portable. Efforts should
communications medium with remote access for be made to reduce the risk of theft and tampering. Such efforts
visibility and control of the plant. There can also be
may include:
a requirement from the grid network operator for
specific telephone landlines for the grid connection.
Often, an Internet broadband (DSL) or satellite • Reducing the visibility of the power plant by planting
communications system is used for remote access. A shrubs or trees at appropriate locations. Care should
GSM (Global System for Mobile Communications) be taken that these do not shade the plant.
connection or standard telephone line with • Installing a wire mesh fence with anti-climb
modems is an alternative though it has a lower protection. A fence is also recommended for
data transfer rate. safety reasons and may be part of the grid code
requirements for public safety. Measures should be
7.4.1 Quality Benchmarks taken to allow small animals to pass underneath the
fence at regular intervals.
Some benchmark features of PV plant infrastructure
include: • Security cameras, lights and microwave sensors with
GSM and TCP/IP transmission of alarms and faults
• Water-tight reinforced concrete stations or pre- to a security company as an option.
fabricated steel containers. • Anti-theft module mounting bolts may be used
• Sufficient space to house the equipment and facilitate and synthetic resin can be applied once tightened.
its operation and maintenance. The bolts can then only be released after heating the
resin up to 300°C.
• Inclusion of:
• Anti-theft module fibre systems may be used. These
• Ventilation grilles, secure doors and concrete systems work by looping a plastic fibre through all
foundations that allow cable access. the modules in a string. If a module is removed, the
plastic fibre is broken. This triggers an alarm.
• Interior lighting and electrical sockets.
• A permanent guarding station with security guard
• Either adequate forced ventilation or air- providing the level of security required in the
conditioning with control thermostats, insurance policy.
depending on environmental conditions.
• An alarm system fitted to the power plant gate and
the medium voltage station, metering station and to
any portable cabins.
94 Utility Scale Solar Power Plants
Some benchmark security features include: three main methods for obtaining the solar irradiance and
environmental conditions:
• Fence at least two meters high.
• On-site weather stations – To measure the plane of
• Metallic posts installed every 6m. array irradiance, module temperature and preferably
• Galvanised and plastic coated fencing. horizontal global irradiance, humidity and wind
speed. This is the option of preference for many
• Video surveillance: current utility scale PV power plants. It allows data
to be collected and compared remotely with yield
• Multiple night and day cameras at a set figures on a daily basis for immediate fault detection.
distance apart.
• Meteorological data gathered from weather
• Illumination systems (infrared) for cameras satellites – Simulation and calculation algorithms
along the perimeter of the site. measure the projected power plant output. This
figure becomes the benchmark for comparing values
• A minimum of 12 months recording time. received from the PV plant on a daily basis, and helps
detect faults immediately. This method removes the
7.6 Monitoring and Forecasting need for an onsite weather station. A number of good
commercial providers of packages use this technique
7.6.1 Monitoring Technology in Europe. Rapid fault detection depends on data
being made available from satellites and being
If high performance, low downtime and rapid fault analysed quickly.
detection is required, automatic data acquisition and
• Local weather stations – This is the least desirable
monitoring technology is essential. This allows the yield of the of the three options as data may not be available for
plant to be monitored and compared with calculations made several months. During that period, the plant may
from solar irradiation data. Monitoring and comparison also lose considerable revenue if faults in the plant go
help raise warnings on a daily basis if there is a shortfall. Faults undetected. It is also possible that the local weather
can be detected and rectified before they have an appreciable station does not accurately track the conditions at the
site (especially if it is some kilometres distant).
effect on production.
In case there are other PV power plants in the vicinity of the
Relying solely on manual checks of performance is not site—or one large plant is split into a number of components-
advisable. A high level of technical expertise is needed to detect it is possible to compare production data and identify a fault
certain partial faults at the string level. In fact, it can take with one plant. Internet-based solutions are available that
many months for reduced yield figures to be identified. The function in this manner.
lower yield may lead to appreciable revenue loss for a utility
scale PV power plant. The on-site weather station solution is currently the most
common option. Data-loggers can be used to collect data from
The key to a reliable monitoring and fault detection the weather station, inverters, meters and transformers. This
methodology is to have good knowledge of the solar information is transferred once a day to a server which carries
irradiance, environmental conditions and plant power output out three key functions:
simultaneously. This allows faults to be distinguished from,
for example, passing clouds or low resource days. There are
A Guide For Developers and Investors 95
• Operations management – The performance service team via fax, email or text message.
management (either onsite or remote) of
the PV power plant enables the tracking of • Reporting – The generation of yield reports
inverters or strings. detailing individual component performance, and
benchmarking the reports against those of other
• Alarm management – Flagging any element of components or locations.
the power plant that falls outside pre-determined
performance bands. Failure or error messages can be Figure 22 illustrates the architecture of an internet portal
automatically generated and sent to the power plant based monitoring system.
Main Servers
Temperature
Wind Speed
Irradiance
/ Direction
Metroligical
Sensors
Internet DSL/GPRS/GSM/TL
Inverters(s)
SCADA
Meter(s) • Real time data
Web Interface
• Historical data
• Reporting
• Alarms (SMS, email, FAX)
• Monitoring Power • Events
• Visualisation Transformer
• Forecasting
Grid
Operator
Dispatchable power plants typically need to provide a Monitoring systems should be based on commercially
forecast to the network operator. This helps to fix plant available software/hardware which is supplied with user
schedules and guarantee continuity of supply. Often, manuals and appropriate technical support.
production forecasts (in half hourly time-steps) are required
24 hours in advance. This entails weather forecasts coupled Depending on the size and type of the plant, minimum
with power forecasting algorithms—more so since PV power parameters to be measured include:
production is intermittent and random in nature. Such
forecasting algorithms can use physical models, statistical • Plane of array irradiance measured to accuracy
within 5% and stability within 0.5% per year. The
approaches or a combination of both. At the least, the
irradiation sensor will be of the same technology
algorithms require the definition of: as the modules being measured, or technology
independent. Silicon sensor reference cells are not
• Power plant capacity. advisable for use in Performance Ratio calculations.
• Module tilt and orientation. • Ambient temperature measured in a location
representative of site conditions with accuracy better
• Module specifications.
than ±1°C.
• Latitude and longitude of the plant.
• Module temperature measured with accuracy better
• Meteorological agency data, gathered from ground than ±1°C. This is done using a sensor thermally
measurement stations and/or satellites. bonded to the back of the module in a location
positioned at the centre of a cell.
The algorithms typically take three-hour national and/or
• Array DC voltage measured to an accuracy of
regional forecasts and break them down to 30 minute local
within 1%.
forecasts (temporal interpolation) before using algorithms
to forecast power production. Comparison of historical • Array DC current measured to an accuracy of
production and actual weather can also allow learning within 1%.
algorithms to be employed. Figure 23 shows the components • Inverter AC power measured as close as possible to
of a forecasting system. Results of forecasting are typically the inverter output terminals with an accuracy of
posted on web portals. There are a variety of commercial within 1%.
forecasting products available in the market today. But
• Power to the utility grid.
availability may be limited to regions that have rapid access to
meteorological agency’s weather data. • Power from the utility grid.
Local Monitoring
Systems
National/Regional
Local weather
weather forecast
Satellites
Temporal
interpolation
Learning system
Database
Solar PV plant
simulation
The performance of a PV power plant may be optimised The aim is to minimise losses. Measures to achieve this are
by a combination of several enabling factors: premium described in Table 10. Reducing the total loss increases the
quality modules and inverters; a good system design with annual energy yield and hence the revenue, though in some
high quality and correctly installed components; and a good cases it may increase the cost of the plant. Interestingly, efforts
preventative maintenance and monitoring regime leading to to reduce one type of loss may be antagonistic to efforts to
low operational faults. reduce losses of a different type. It is the skill of the plant
• Have a robust O&M strategy that removes the risk of shading due to
vegetation growth.
Incident angle • Use anti-reflection coatings, textured glass, or tracking.
Low irradiance • Use modules with good performance at low light levels.
Module temperature
• Choose modules with a lower negative temperature coefficient for power at
high ambient temperature locations.
• Choose modules less sensitive to shading (for example amorphous silicon).
Soiling • Ensure a suitable O&M contract that includes an appropriate cleaning
regime for the site conditions.
• Choose modules with a low tolerance. A tolerance of ±3% is typical but
Module quality
tolerances of between ±1.5% to ±10% are common.
• Sort modules with similar characteristics into series strings where possible.
Curtailment of tracking
• Ensure that tracking systems are suitable for the wind loads to which they
will be subjected.
100 Utility Scale Solar Power Plants
There are a number of minimum requirements that should • Mounting structure drawings with structural
be included within design documentation. These include: calculations reviewed and certified by a
licensed engineer.
• Datasheets of modules, inverters, array mounting
• A detailed resource assessment and energy
system and other system components.
yield prediction.
• Wiring diagrams including, as a minimum, the
• A design report. It will include information on the
information laid out in Table 11.
site location, site characteristics, solar resource, a
• Layout drawings showing the row spacing and summary of the results of the geotechnical survey,
location of site infrastructure. design work and the energy yield prediction.
A Guide For Developers and Investors 101
PV String Information
• String over-current protective device specifications (where fitted) – type and voltage/
current ratings.
• Array over-current protective devices (where applicable) – type, location and rating
(voltage/current).
• Details of all earth/bonding conductors – size and connection points. This includes
details of array frame equipotential bonding cable where fitted.
Earthing and protection
• Details of any connections to an existing Lightning Protection System (LPS).
devices
• Details of any surge protection device installed (both on AC and DC lines) to include
location, type and rating.
• AC isolator location, type and rating.
7.9
For plant design, there are some general rules of thumb. But specifics of project locations—such as irradiation
conditions, temperature, sun angles and shading—should be taken into account in order to achieve the optimum balance
between annual energy yield and economic return.
It may be beneficial to use simulation software to compare the impact of different module or inverter technologies and
different plant layouts on the predicted energy yield and plant revenue.
The solar PV modules are typically the most valuable and portable components of a PV power plant. Safety
precautions may include anti-theft bolts, anti-theft synthetic resins, CCTV cameras with alarms and security fencing.
The risk of technical performance issues may be mitigated by carrying out a thorough technical due diligence exercise
in which the final design documentation from the EPC contractor is scrutinised (as described in Section 13.2.2).
A Guide For Developers and Investors 103
Case Study 3
Design
It is vital to ensure that suitable technical expertise is brought to bear on every aspect of the plant design through in-house or
acquired technical expertise.
In case of the 5 MW project, the most significant of the design flaws were:
Foundations:
• The foundations for the supporting structures consisted of concrete pillars, cast in situ, with steel reinforcing bars and
threaded steel rod for fixing the support structure base plates. This type of foundation is not recommended due to the
inherent difficulty in accurately aligning numerous small foundations.
• Mild steel was specified for the fixing rods. As mild steel is prone to corrosion, stainless steel rods would
have been preferable.
Supporting structures:
• The supporting structures were under-engineered for the loads they were intended to carry; in particular, the purlins
sagged significantly under the load of the modules. As supporting structures should be designed to withstand wind
loading and other dynamic loads over the life of the project, this was a major problem. Extensive remedial work was
required to retrofit additional supporting struts.
• The supporting structure was not adjustable (i.e. no mechanism was included to allow adjustment in the positioning of
modules). This is a basic mistake which compounded the flaw in the choice of foundation type; the combination of these
two mistakes led to extensive problems when it came to attempting to align the solar modules.
104 Utility Scale Solar Power Plants
Electrical:
• String diodes were used for circuit protection instead of string fuses. Current best practice is to use string fuses,
as diodes cause a voltage drop and power loss, as well as a higher failure rate.
• No protection was provided at the submain or main junction boxes. This means that for any fault occurring
between the array junction boxes and the DC distribution boards (DBs), the DBs will trip and take far more of
the plant offline than is necessary.
• No load break switches were included on junction box before the DBs. This means it is not possible to isolate
the plant at the array, submain or main junction box levels for installation or maintenance.
• The junction boxes did not allow for string monitoring. This reduces fault diagnosis capability.
The design flaws listed above cover a wide range of issues. However, the underlying lesson is that is it vital to ensure
that suitable technical expertise is brought to bear on every aspect of the plant design. Should the developer not have all
the required expertise in-house, then a suitably experienced technical advisor should be engaged. It is also recommended
that, regardless of the level of expertise in-house, a full independent technical due diligence is carried out on the design
before construction commences.
It should be borne in mind that it is far cheaper and quicker to rectify flaws at the design stage than during
or after construction.
A Guide For Developers and Investors 105
• Local and/or regional planning authority. Compliance with the IFC performance standards will not
• Environmental agencies/departments. only ensure a socially and environmentally sustainable project
but will also facilitate the sourcing of finance for the project.
• Archaeological agencies/departments.
For further detail on the IFC’s performance standards
see www.ifc.org.
106 Utility Scale Solar Power Plants
• Planning Department – The project will normally • Local Governing Bodies – In some areas, a project
require prior approval from the relevant planning may fall under the jurisdiction of governing bodies
department at town and district levels. for small villages. Consultation with these local
bodies is key to getting consent for the project from
• Archaeological Department – Consultation and the local population. Their approval can facilitate
approval from the relevant archaeological department work in the construction and operation phases.
will confirm that the land acquired for the project is
not of historical significance. In addition, the following requirements should be noted.
• Fire Safety Authority – Consultation and approval
from the relevant authority may be required with
• Construction power requirements – This
specific licence is normally obtained from the state
respect to relevant fire safety requirements during
distribution utility for obtaining power required
construction and operation of the project.
during construction of the plant. Otherwise, stand-
• Forest Authority – Consultation and approval from alone diesel generators can be utilised with prior
the relevant forest authority may be required if trees permission from the pollution control board.
are to be felled to prevent any shading of PV plant.
It may also be prudent to confirm that the land
• Environmental Impact Assessment – An assessment
of the potential environmental impacts of the
to be developed has not been reserved for future
development should be undertaken. If required,
forestry operations.
appropriate mitigation measures should be identified
• Pollution Control Board – Consent from the local in consultation with relevant stakeholders. As
pollution control board may be required with respect a guideline, projects should adhere to the IFC
to wastewater management and noise emission performance standards (see Section 8.2).
control, particularly during the construction phase of
the project. 8.3.3 Power Export
Case Study 4
Permits and Licensing
There are many permits required for a multi-megawatt PV power plant in India. An indicative, non-exhaustive list
of the permits obtained for the 5MW plant built in Tamil Nadu in 2010 is shown in the table below. These apply
specifically to this project and permitting will differ in other states. In addition to the permits that are suggested below,
there will naturally be permits and licenses required as a result of simply operating a business in India (eg. human
resource requirement) which have not been included below in the interest of focus. However, these need to be given
equal attention in the development phase.
A lesson learnt in the case of the Tamil Nadu plant was that comprehensive legal advice on the permits is required as
well as a stringent management and follow up of the application processes.
It must be noted that some permit requirements were not relevant to the Tamil Nadu plant. Permission from the
Ministry of Defence, for example, was not required as the site was not in a militarily sensitive zone.
The majority of the permits were applied for and in place prior to the start of construction. This is deemed best
practice and sets a good example for other developers. One permission, involving land access rights, was overlooked.
The main access route to the plant was through land owned by another party. Until rights are obtained, the project
remains vulnerable to the risk of goodwill being withdrawn.
Some permits were issued on the condition that the plant was to be completed before a certain date. This caused
problems when the project was delayed. As a result, a re-application or extension was required. This illustrates the
importance of effective planning of projects and scheduling of construction.
A Guide For Developers and Investors 109
9. CONSTRUCTION
9.1 Introduction project interfaces, describe which organisations are involved,
allocate responsibility for each interface to a particular
The management of the construction phase of a solar PV individual, and explicitly state when the interface will be
project should be in accordance with general construction reviewed. In general, design and construction programmes
project management best practice. Therefore, the aim is should be developed to minimise interfaces wherever possible.
to construct the project to the required level of quality,
and within the time and cost limits. During construction, Opting for a turnkey EPC contract strategy will, in effect,
issues like environmental impact, and health and safety of pass the onus for interface management from the developer
the workforce (and other affected people) should also be to the EPC contractor. But interface management will remain
carefully managed. an important issue and one that requires ongoing supervision.
To some extent interfaces between the project and its
The approach to construction project management for a surroundings (for example grid connection) will remain the
solar PV plant will depend on many factors. Of them, one of responsibility of the developer.
the most important is the project contract strategy.
If a turnkey EPC strategy is chosen, then a contractor with a
From a developer’s perspective, construction project suitable track record in the delivery of complex projects should
management for a full turnkey EPC contract will be be selected to minimise this risk. Information should also be
significantly less onerous than that required for a multi- sought from potential contractors on their understanding
contract approach. However, a multi-contract approach gives of the project interfaces and their proposed approach to
the developer greater control over the final plant configuration. managing them.
Regardless of the contract strategy selected, there are a number
of key activities that will need to be carried out, either by the 9.3 Programme and Scheduling
developer or a contractor. These activities are described in the
following sections. A realistic and comprehensive construction programme is
a vital tool for the construction planning and management
Typical EPC contract terms may be found in Appendix C – of a solar PV project. The programme should be sufficiently
EPC Contract Model Heads of Terms. detailed to show:
For a multi-contract strategy, the developer should develop a • Project critical path.
robust plan for interface management. This plan should list all • Actual progress against plan.
A Guide For Developers and Investors 111
• Security contractor
• Civil contractor
• Mounting system supplier (for string Foundations for larger central inverters,
inverters) or suitability for the mounting system.
• Module supplier Suitability of the module string design
5 Inverter
for the inverter. Interface with the
• Inverter supplier communications for remote monitoring
• Electrical contractor and input into the SCADA system.
• Communications contractor
112 Utility Scale Solar Power Plants
All tasks and the expected timescale for completion should • Electrical site works.
be detailed along with any restrictions to a particular task. For
• Grid interconnection works.
example, if permits or weather constraints stop construction
during particular months. • Commissioning and testing.
For a solar PV project, it is likely that the programme will A high level programme should be produced to outline
have different levels, incorporating different levels of detail the timescales of each task, the ordering of the tasks and
around each of the following main work areas: any key deadlines. This should be completed as part of the
detailed design.
• Site access.
The programme will then be built up to detail all the
• Security. associated tasks and sub tasks, ensuring that they will
• Foundation construction. be completed within the critical timescale. A thorough
programme will keep aside time and resources for any
• Module assembly. contingency. It will also allocate allowance for weather risk or
• Mounting frame construction. permit restrictions for each task.
• Substation construction.
A Guide For Developers and Investors 113
Interdependencies between tasks will allow the programme Appropriate sequencing of tasks is a vital part of the
to clearly define the ordering of tasks. A project scheduling planning process. The tasks must be sequenced logically and
package will then indicate the start date of dependent tasks as efficiently. The overall sequence of works is generally: site
well highlighting the critical path. access, site clearance, security, foundation construction, cable
trenches and ducts, substation construction, mounting frame
Critical path analysis is important to ensure that tasks construction, electrical site works, communications, onsite
that can affect the overall delivery date of the project are grid works and then testing and commissioning. Each of these
highlighted and prioritised. A comprehensive programme work areas should be broken down into a series of sub-tasks.
should also take into account resource availability. This will Alongside, an assessment of the inputs required for each task
ensure that tasks are scheduled for when required staff or plant (especially when interfaces are involved) will help develop a
are available. logical and efficient sequence.
Incorporating a procurement schedule that focuses on items Consideration should also be given to any factors that
with a long manufacturing lead time (such as transformers, could prevent or limit possible overlap of tasks. These factors
central inverters and modules) will ensure that they are ordered could include:
and delivered to schedule. It will also highlight any issues with
the timings between delivery and construction, and the need • Access requirements.
for storage onsite.
• Resource availability (plant and manpower).
To share this information and to save time and effort, • Planning (or other regulatory) restrictions.
it is strongly recommended that an “off-the-shelf ” project
• Safety considerations.
scheduling package is used.
9.3.3 Risk Management
9.3.1 Milestones
The risks associated with the project should be identified,
Milestones are goals that are tied in with contractual assessed and managed throughout the construction process.
obligations, incentives or penalties. Incorporating milestones The hazards need to be incorporated in the planning and
in the programme helps the project team to focus on achieving scheduling of the project. Each aspect of the project should be
these goals. In effect, construction must be planned around assessed for likelihood and impact of potential risks. The next
certain milestones or fixed dates (for example, the grid step would be to develop a suitable action plan to mitigate
connection date). identified risks. If a particular risk could affect the delivery of
the whole project, alternatives for contingency (in terms of
If the contracted milestones are included in the programme, time and budget) should be included.
the impact of slippage on these dates will be apparent.
Appropriate budgetary and resourcing decisions can then be 9.4 Cost Management
made for those delays. The milestones can also indicate when
payments are due to a contractor. The viability of a solar project will be affected by the
duration of the construction period. During construction,
the project will be in debt owing to interest and finance
114 Utility Scale Solar Power Plants
charges, and lack of income to make payments. Therefore, a Earned value management – This is an approach based
shorter construction period is generally preferable. The period on monitoring the project plan, actual work completed and
of construction also requires prudent cost management, work-completed value to assess if a project is on track. Earned
which is tied in with the project schedule and the contracted value management indicates how much of the budget (and
payment structure. available time) should have been spent, with regards to the
amount of work done to date. This method necessitates
The payment structure will depend upon the type advance calculation of both the baseline cost for a task and
of contract opted for, but is likely to involve milestone the resources required. If used correctly, this is a powerful tool
payments. The typical range of EPC payment schedules is for estimating and controlling project overspends as early as
detailed in Table 13. If a multi-contract strategy is chosen, possible in the construction period.
then a similar structure for phased-out payments for each
contract is advisable. Completion certificates – Completion certificates are
issued once the entire (or a specific) part of the plant is
This schedule shows that a high percentage of the payments physically complete. These certificates are issued prior to any
are made once the goods have been delivered to site. It also tests taking place and confirm that the contractor has installed
allows enough money to be held back to ensure that the the equipment correctly. They may only be for a specific part
contractor completes the works. of the project (for example a string of modules). Payment for a
particular item of work will not be made until the appropriate
The tools used for construction cost management in a completion certificate has been issued.
solar PV venture are the same as for any major engineering
project. These can include:
Snagging lists – Compiling “Snag Lists” as an ongoing The owner of a PV power plant will typically use the
exercise is recommended. A prerequisite to the hand-over services of an EPC contractor to design and build a project.
phase, this list is a process of monitoring and tracking any He will also require an O&M contractor to operate and
defects. These should be addressed and rectified to the maintain the plant during its operational phase.
satisfaction of the developer. In some cases, the take-over
will occur with some minor defects still outstanding. In The EPC contractor offers project planning services, and
such a scenario, the snagging list will detail these minor will provide the necessary engineering for project design. The
defects, which will then have to be addressed within a contractor will typically be responsible for material selection
stipulated period. and procurement of modules, inverters, and balance of plant
components. Construction may be carried out by the EPC
Take-Over certificates – Take-over certificates will be issued contractor or through partnerships with local installation
by the contractor for acceptance by the developer. These will and project development companies (in this case, the EPC
be issued once all tests have been completed and defects contractor will provide on-site inspection). Warranties within
addressed. It is normal for the power purchaser to request a the EPC contract may include a Defect Warranty, Module
copy of the take-over certificate. To expedite the launch of the Capacity Warranty, Performance Ratio Warranty and Structure
project, the developer may choose to take over a project with Warranty as described in Table 14.
minor snags, subject to the contractor taking responsibility
to complete them. While this conditionality is acceptable,
progress in addressing defects should be monitored.
Warranties for EPC contracts are typically in the two to five • Acceptance criteria.
year range. If the O&M contractor is also the EPC contractor,
• Completion date.
it can be easier to enforce these warranties. However, if they
are separate companies, the exclusions to the warranties should • Details of any records to be kept (for example,
be checked carefully. Also, the maintenance carried out by photographs or test results).
the O&M contractor should be in compliance with the EPC • Signature or confirmation of contractor completing
contractor’s requirements. tasks or accepting delivery.
The original manufacturer’s warranties on components such • Signature of person who is confirming tasks or tests
on behalf of the developer.
as inverters, support structures and modules should be passed
from the EPC contractor to the plant owner at handover. Quality audits should be completed regularly. These will
Acceptance testing is often completed before the plant is help developers verify if contractors are completing their works
handed over to the owner. in line with their quality plans. Audits also highlight quality
issues that need to be addressed at an early stage. Suitably
Even in a multiple contract structure, the component experienced personnel should undertake these audits.
warranties described in Table 14 above would be applicable
and should be incorporated. However, these warranties will 9.7 Environmental Management
be contract-specific, and need to be decided upon before
tendering for the work. This is a judicious way to ensure that The IFC Environmental, Health, and Safety (EHS)
the guarantees provided are on a par with those expected by Guidelines are technical reference documents with general
the developers. and industry-specific examples of good international industry
practice. With respect to environmental management, the
9.6 Quality Management guidelines cover the following areas:
• Energy Conservation.
A quality plan is an overview document (generally in a
tabular form), which details all works, deliveries and tests to be
• Wastewater and Ambient Water Quality.
completed within the project. This allows work to be signed • Water Conservation.
off by the contractor and enables the developer to confirm if
the required quality procedures are being met. A quality plan
• Hazardous Materials Management.
will generally include the following information: • Waste Management.
It is recommended that these guidelines are followed as Community Health and Safety:
a benchmark, along with any specific local guidelines and
information from the EIA. • Water Quality and Availability.
The health and safety of the project work force and other • Emergency Preparedness and Response.
affected people should be carefully overseen by the project
The IFC guidelines give guidance on how each of these
developer. Apart from ethical considerations, the costs of not
aspects of H&S should be approached, outline minimum
complying with health and safety legislation can represent
requirements for each aspect and list appropriate control
a major risk to the project. Furthermore, a project with a
measures that can be put in place to reduce risks.
sensitive approach to health and safety issues is more likely to
obtain international financing.
As a minimum standard, compliance with local H&S
legislation should be rigorously enforced. Where local legal
The IFC EHS guidelines cover two main areas of health
requirements are not as demanding as the IFC guidelines, it is
and safety: occupational health and safety and community
recommended that the IFC guidelines be followed.
health and safety. The issues covered under these areas are
listed below.
9.9 Specific Solar PV Construction Issues
Occupational Health and Safety:
The following sections describe common pitfalls or mistakes
• General Facility Design and Operation. that can occur during the construction phase of a solar PV
project. Most of these pitfalls can be avoided by appropriate
• Communication and Training. design, monitoring, quality control and testing on site.
• Physical Hazards.
9.9.1 Civil
• Chemical Hazards.
The civil works relating to the construction of a solar PV
• Biological Hazards.
plant are relatively straightforward. However, there can be
• Personal Protective Equipment (PPE). serious and expensive consequences if the foundations and
road networks are not adequately designed for the site.. The
• Special Hazard Environments.
main risks lie with the ground conditions. Importantly, ground
• Monitoring. surveys lacking in meticulous detailing or proper interpretation
could lead to risks such as unsuitable foundations.
118 Utility Scale Solar Power Plants
Used land also poses a risk during the civil engineering Underground cables should be buried at a suitable depth
works. Due to the nature of digging or pile driving for (generally between 500mm and 1,000mm) with warning tape
foundations, it is important to be aware of hazardous obstacles or tiles placed above and marking posts at suitable intervals on
or substances below the surface. This is especially important in the surface. Cables may either be buried directly or in ducts. If
former industrial sites or military bases. cables are buried directly, they should be enveloped in a layer
of sand or sifted soil should be included to avoid damage by
9.9.2 Mechanical backfill material.
The mechanical construction phase usually involves the Comprehensive tests should be undertaken prior to
installation and assembly of mounting structures on the site. energisation to verify that there has been no damage
Some simple mistakes can turn out to be costly, especially if to the cables.
these include:
9.9.4 Grid Connection
• Incorrect use of torque wrenches.
The grid connection will generally be carried out by a
• Cross bracing not applied.
third party over whom the project developer will have limited
• Incorrect orientation. control. Close communication with the grid connection
contractor is essential to ensure that the grid requirements
If a tracking system is being used for the mounting
are met. Delay in the completion of the grid connection
structure, other risks include:
will affect the energisation date, which will delay the start of
• Lack of clearance for rotation of modules. commercial operation.
It is recommended that the owner and lenders of Design reviews will generally be carried out on:
the project are kept informed of developments during
construction. Construction supervision may be carried out • Design basis statements.
by in-house resources. Alternatively, a “technical advisor” or • Studies/investigations.
“owner’s engineer” may be commissioned to carry out the
work on their behalf. • Design specifications.
• Design of structures.
The role of the technical advisor during the construction
phase is to ensure contractor compliance with the relevant • Drawings (all revisions).
contracts, as well as to report on progress and budget. The • Calculations.
construction supervision team would generally comprise
of a site engineer supported by technical experts in an office. • Execution plans.
The main parts of the technical advisor’s role are: review • Risk assessments and method statements.
of proposed designs, construction monitoring and witnessing
of key tests. • Quality plans.
• Safety plans/reports.
• O&M manuals.
• Test reports.
120 Utility Scale Solar Power Plants
The objective of the design review is to ensure that the • Inspection of cable tracks.
contractor has designed the works in accordance with the
• Witnessing of delivery/off-load of solar modules,
contract agreements and relevant industry standards. It also transformers, inverters and switchgear.
aims to ascertain that the works will be suitably resourced
and sequenced to deliver the project as specified. The design • Inspection of module, switchgear and inverter
review could also cover specific areas such as grid compliance installation.
or geotechnical issues, depending upon the specific project • Witnessing of site acceptance tests.
requirements and experience of the developers.
• Witnessing of completion tests.
Key stages and tests for witnessing will include: • Monitoring and expediting defects.
• Inspection of road construction. Besides the owner’s engineer, the lender’s engineer has the
additional role of signing off and issuing certificates that state
• Inspection of foundations.
the percentage of the project completed. These certificates are
• Verification of cable routes. required by the lenders prior to releasing funds in accordance
with the project payment milestones.
Case Study 5
Construction and Project Management
The 5MW PV power plant in Tamil Nadu was constructed during 2010. At commencement, construction was
projected to be completed within 38 weeks; due to various factors, many of which are covered in the case studies in
this book, construction ended up taking approximately 52 weeks – a significant and costly delay. In addition, the
constructed plant suffered from serious quality issues. The construction schedule should be carefully thought through
by suitably experienced personnel. It is also recommended that project management software tools are used as these
enable the developer to track the progress of a project, identify resource constraints and understand the impact of
uncompleted tasks.
• Design flaws. Poor design of components, such as the support structures, lead to costly and time-consuming
remedial measures.
• Poorly planned construction schedule. The illogical sequencing of construction tasks caused a number of delays:
A Guide For Developers and Investors 121
• Monsoon rains restricted access to the site as the access road had not been sealed. The access route should
have been sealed well before the arrival of the monsoon.
• Modules were damaged (and were at risk of theft) as they were stored unprotected on site for long periods of
time. Modules and other valuable components should not be delivered to site until shortly before they are
required. If they must be delivered earlier then they should be stored in a controlled and secure environment.
The constructed plant suffered from a range of serious quality issues. These included:
While a wide range of factors undoubtedly contributed to each of these issues, the following factors are considered
to have been particularly significant:
• Design quality. Certain basic aspects of the design led directly to construction quality issues. The clearest
example of this is the design of the foundations and substructures leading to misalignment of the modules.
These problems could have been avoided if suitable expertise had been used in the design stage.
• Design documentation and document control. It is preferable to have a full set of “for construction” design
drawings before construction commences. Throughout the construction process, it is vital that document
management is thoroughly carried out; in particular, design changes and revision of drawings should be
rigorously controlled. The failure to do this led to basic mistakes such as foundations being constructed in the
wrong locations.
• Contractor capability. It is fundamentally important to select a suitably experienced and capable construction
contractor. Ideally, a contractor with demonstrable experience of similar projects should be selected. In any
case, potential contractors’ proposed approach to quality management should be thoroughly scrutinised
during the contractor selection process.
• Project Management – supervision / monitoring. Regardless of the capability of the selected contractor, the
developer must monitor construction progress closely. Suitably experienced personnel should regularly inspect
the progress and quality of the works (and the completeness of the quality records) as they progress. If the
developer does not have suitable resources in-house to carry out construction supervision then they should
engage a competent party to do it on their behalf.
122 Utility Scale Solar Power Plants
10. COMMISSIONING
The commissioning process certifies that the project owner’s Inverter commissioning should follow the protocol
requirements have been met, the power plant installation is described in the inverter installation manual. Tasks
complete and the power plant complies with grid and safety may include:
requirements. Successful completion of the commissioning
process is often considered to be part of the provisional or final
• Checking inverter cabling for conformity to
schematic diagrams.
acceptance of the PV plant.
• Checking that cable connections are firm.
Commissioning should follow the procedure described in
• Checking DC voltages for polarity and verifying
IEC 62445 and prove three main criteria: that they are approximately the same for each string.
Voltages must not exceed the maximum voltage of
1. The power plant is structurally and electrically safe. the inverter.
2. The power plant is sufficiently robust (structurally • Checking AC grid voltage. AC voltage
and electrically) to operate for the specified lifetime measurements between the external conductors
of a project. should be approximately the same as the nominal
voltage of the inverter.
3. The power plant operates as designed and its
performance is as expected. • Checking the internal AC power supply.
Critical elements of a PV power plant that require • Mounting the inverter panelling.
commissioning include:
• Inserting fuses or insulation blades (if applicable).
• Module strings. • Switching on the voltage supply by turning on the
grid monitoring circuit breaker (if applicable) and
• Inverters.
the external voltage supply circuit breaker. Status
• Transformers. lights should not be showing a fault.
• Earthing protection systems. With the exception of the module strings, the
• Electrical protection systems. commissioning of the remaining plant components follows
standard procedures for power plant (the guide does not
• Grid connection compliance protection and dwell further on the details). The following sections provide
disconnection systems. an overview of the pre-connection and post-connection
• Monitoring systems (including irradiation sensors). acceptance testing and documentation requirements for the
module strings.
• Support structure and tracking systems
(where employed).
A Guide For Developers and Investors 123
Commissioning should start immediately after installation inverters. These tests according to IEC 62446 should include:
has been completed or, where appropriate, sequentially as
strings are connected. For power plants employing modules • Open Circuit Voltage Test.
which require a settling-in period, for example, thin film • Short Circuit Current Test.
amorphous silicon modules, performance testing should
begin once the settling in period has been completed and the 10.2.1 Open Circuit Voltage Test
modules have degraded.
This test checks whether all strings are properly connected
Since irradiance has an impact on performance, tests should (module and string polarity) and whether all modules are
be carried out under stable sky conditions. The temperature of producing the voltage level as per the module data sheet. The
the cells within the modules should be recorded in addition to test should be conducted for all strings.
the irradiance and time.
The open circuit voltage, Voc, should be recorded and
Ideally, commissioning should be carried out by an compared with temperature adjusted theoretical values.
independent specialist third party selected by the owner. It
should include both visual and electrical testing. In particular, 10.2.2 Short Circuit Current test
visual testing should be carried out before any system is
energised. The testing outlined in this section does not This test verifies whether all strings are properly connected
preclude local norms which will vary from country to country. and the modules are producing the expected current. The test
should be conducted for all strings.
Test results should be recorded as part of a signed-off
commissioning record. While an independent specialist The short circuit current, Isc, should be recorded and
would be expected to carry out these tests, it is important that compared with the temperature adjusted theoretical values.
the developer and owner are aware of them and make sure
that the required documentation is completed, submitted 10.3 Grid Connection
and recorded.
Grid connection should only be performed once all
10.2 Pre-Connection Acceptance Testing DC string testing has been completed. It is likely that the
distribution or transmission system operator will wish to
Prior to connecting the power plant to the grid, electrical witness the connection of the grid and/or the protection relay.
continuity and conductivity should be checked by the Such a preference should be agreed in advance as part of the
electrical contractor. Once completed, pre-connection connection agreement.
acceptance testing should be carried out on the DC side of the
The grid connection agreement often stipulates the level of
parameters—such as electrical protection, disconnection and
fault—to which the PV power plant is required to adhere.
Usually, these conditions need to be met before commissioning
the grid connection.
124 Utility Scale Solar Power Plants
Once the power plant is connected to the grid, the for the temperatures observed during the test. This is known
inverters will be powered up according to the manufacturer’s as the adjusted performance ratio. The electrical energy
start-up sequence. Inverter internal meters and displays generated is typically considered to be acceptable if it is within
should be verified prior to use. ± 3-5% of the value given by the agreed temperature adjusted
performance ratio.
Post grid connection should include:
If there are significant differences between the contractual
• DC current test. and actual adjusted performance ratio, the EPC contractor
• Performance ratio test. should identify and rectify the discrepancy before repeating the
performance ratio test.
10.4.1 DC Current Test
10.4.3 Availability Test
This test verifies whether all strings are producing
adequate and consistent operating current as per the module An availability test should be performed in parallel with the
data sheets. The test should be conducted for all strings. performance ratio test as described above. This will confirm
that an acceptable availability is being achieved according to
The string current values per inverter will be checked guaranteed values. The test will typically be performed for a
against the average values of all strings connected to the minimum of ten consecutive days under stable sky conditions.
same inverter and checked against acceptance criteria.
The availability of the system can be defined using
10.4.2 Performance Ratio Test the formula:
• Warranty documentation.
• Performance guarantees.
[50] The commissioning record is a checked-off list of tasks that that have been
completed as part of the commissioning process.
126 Utility Scale Solar Power Plants
11. OPERATION AND
MAINTENANCE
Specific scheduled maintenance tasks are covered in the
Compared to most other power generating technologies,
following sections.
PV plants have low maintenance and servicing requirements.
However, proper maintenance of a PV plant is essential to
11.1.1 Module Cleaning
optimise energy yield and maximise the life of the system.
Module cleaning is a simple but important task. It can
Maintenance can be broken down as follows:
produce significant and immediate benefits in terms of
energy yield.
• Scheduled or preventative maintenance – Planned
in advance and aimed at preventing faults from
occurring, as well as keeping the plant operating at its The frequency of module cleaning will depend on local
optimum level. site conditions (for example, prevalence of dust or rain) and
the time of year. As the soiling of modules is site – specific,
• Unscheduled maintenance – Carried out in
the duration between clean-ups is likely to vary between sites.
response to failures.
However, it is generally recommended to clean the modules at
Suitably thorough scheduled maintenance should least twice annually. Figure 25 shows the solar panel covered
minimise the requirement for unscheduled maintenance with dust. When scheduling module cleaning, consideration
although, inevitably, some failures still occur. A robust and should be given to the following:
well-planned approach to both scheduled and unscheduled
maintenance is important. • Environmental and human factors (for instance,
autumn fall debris and soiling from local
agricultural activities).
11.1 Scheduled/Preventative Maintenance
• Weather patterns: cleaning during rainy periods is
The scheduling and frequency of preventative maintenance less likely to be required.
is dictated by a number of factors. These include the
• Site accessibility based upon weather predictions.
technology selected, environmental conditions of the site,
warranty terms and seasonal variances. The scheduled • Availability of water and cleaning materials.
maintenance is generally carried out at intervals planned in
If the system efficiency is found to be below the expected
accordance with the manufacturers’ recommendations, and as
efficiency, then module cleaning should be scheduled
required by the equipment warranties. Scheduled maintenance
as necessary.
should be conducted during non-peak production periods
and, where possible, at night.
The optimum frequency of module cleaning can be
determined by assessing the costs and benefits of conducting
Although scheduled maintenance will both maximise
the procedure. The benefit of cleaning should be seen in an
production and prolong the life of the plant, it does represent
improved performance ratio due to the lower soiling loss—and
a cost to the project. Therefore, the aim should be to seek the
resultant increase in revenue). A cost estimate to clean all the
optimum balance between cost of scheduled maintenance and
modules at the PV plant should be obtained from the cleaning
increased yield through the life of the system.
contractor. If the cost to clean is less than the increased
revenue then it is beneficial to clean the modules.
A Guide For Developers and Investors 127
Checking module connection integrity is important for Potential faults across the PV plant can often be detected
systems that do not have string level monitoring. This is through thermography. This technique helps identify weak and
more likely for central inverter systems for which no string loose connections in junction boxes and inverter connections.
monitoring at the junction/combiner boxes has been designed. It can also detect hot spots within inverter components and
In such cases, faults within each string of modules may along strings of modules that are not performing as expected.
be difficult to detect. Therefore, the connections between
modules within each string should be checked periodically Thermography should be conducted by a trained specialist
(this may include measuring the string current). using a thermographic camera.
All junction boxes or string combiner boxes should Generally, inverter faults are the most common cause
be checked periodically for water ingress, dirt or dust of system downtime in PV power plants. Therefore, the
accumulation and integrity of the connections within the scheduled maintenance of inverters should be treated as a
boxes. Loose connections could affect the overall performance centrally important part of the O&M strategy.
of the PV plant. Any accumulation of water, dirt or dust could
cause corrosion or short circuit within the junction box. The maintenance requirements of inverters vary with size,
type and manufacturer. The specific requirements of any
Where string level monitoring is not used, periodic checks particular inverter should be confirmed by the manufacturer
on the integrity of the fuses in the junction boxes, combiner and used as the basis for planning the maintenance schedule.
boxes and, in some cases, the module connection box should
be conducted.
128 Utility Scale Solar Power Plants
The annual preventative maintenance for an inverter should, The remaining systems within a PV power plant, including
as a minimum, include: the monitoring and security systems, auxiliary power supplies,
and communication systems should be checked and serviced
• Visual inspections. regularly. Communications systems within the PV power plant
• Cleaning/replacing cooling fan filters. and to the power plant should be checked for signal strength
and connection.
• Removal of dust from electronic components.
• Any additional analysis and diagnostics Vegetation control and ground keeping are important
recommended by the manufacturer. scheduled tasks for solar PV power plants since there is a
strong likelihood for vegetation (for example, long grass, trees
11.1.6 Structural Integrity
or shrubs) to shade the modules. The ground keeping can
The module mounting assembly, cable conduits and any also reduce the risk of soiling (from leaves, pollen or dust)
other structures built for the PV plant should be checked on the modules.
periodically for mechanical integrity and signs of corrosion.
This will include an inspection of support structure
11.2 Unscheduled Maintenance
foundations for evidence of erosion from water run-off.
Unscheduled maintenance is carried out in response
to failures. As such, the key parameter when considering
11.1.7 Tracker Servicing
unscheduled maintenance is diagnosis, speed of response
Similarly, tracking systems also require maintenance and repair time. Although the shortest possible response
checks. These checks will be outlined in the manufacturers’ is preferable for increasing energy yield, this should be
documentation and defined within the warranty conditions. balanced against the likely increased contractual costs of
In general, the checks will include inspection for wear shorter response times.
and tear on the moving parts, servicing of the motors or
The agreed response times should be clearly stated within
actuators, checks on the integrity of the control and power
the O&M contract and will depend on the site location—
cables, servicing of the gearboxes and ensuring that the
and whether it is manned. Depending on the type of fault,
levels of lubricating fluids are suitable.
an indicative response time may be within 48 hours, with
The alignment and positioning of the tracking system liquidated damages if this limit is exceeded.
should also be checked to ensure that it is functioning
The majority of unscheduled maintenance issues are
optimally. Sensors and controllers should be checked
related to the inverters. This can be attributed to their
periodically for calibration and alignment.
complex internal electronics, which are under constant
operation. Depending on the nature of the fault, it may be
possible to rectify the failure remotely – this option is clearly
preferable if possible.
A Guide For Developers and Investors 129
Other common unscheduled maintenance It is important that spares stock levels are maintained.
requirements include: Therefore, when the O&M contractor uses some spares
he should replenish the stocks as soon as possible. This
• Tightening cable connections that have loosened. arrangement will reduce the time gap between the
• Replacing blown fuses. identification of the fault and replacement of the non-
operational component. This can be of particular relevance for
• Repairing lightning damage. remote locations with poor accessibility and adverse weather
• Repairing equipment damaged by intruders or conditions. Consultation with manufacturers to detail the
during module cleaning. spare parts inventory, based upon estimated component
lifetimes and failure rates, is recommended.
• Rectifying SCADA faults.
In order to facilitate a rapid response, a suitably stocked Section 7.6.1 describes some of the information required
spares inventory is essential. The numbers of spares required for an effective monitoring system. A SCADA system is able
will depend on the size of the plant and site-specific to monitor the real-time efficiency and continuously compare
parameters. Adequate supplies of the following components it with the theoretical efficiency to assess if the system is
should be held: operating optimally. This information can be used by the
O&M contractor to establish the general condition of the
• Mounting structure pieces. system and schedule urgent repair or maintenance activities
such as cleaning.
• Junction/combiner boxes.
It is common for the O&M of PV plants to be carried • Warranties and operational targets.
out by specialist O&M contractors. The contractor will be
• Terms and conditions.
responsible for the operation and maintenance of the whole
plant. This is likely to include: • Legal aspects.
• Modules and mounting frames or tracking system. • Insurance requirements and responsibilities.
• Standards, legislation and guidelines to which the • Ensure that the required maintenance will be
contractor must comply. conducted on all components of the system. As a
minimum, these activities should be in line with
• Payment structure. manufacturer recommendations and the conditions
of the equipment warranties.
A Guide For Developers and Investors 131
• Provide appropriate cleaning of the modules and the • Local engineering practices (unless the
removal of snow (site specific). documents and conditions listed above require a
higher standard).
• Make sure that the natural environment of the
system is maintained to avoid shading and aid 11.5.5 Payment
maintenance activities.
• Replace defective system components and system The cost and remuneration of the O&M contract is
components whose failure is deemed imminent. generally broken down in to:
• Provide daily remote monitoring of the performance • Fixed remuneration and payment dates.
of the PV plant to identify when performance drops
below set trigger levels. • Other services remuneration and
expenditure reimbursement.
11.5.3 Obligations on the Owner
Fixed remuneration outlines the payment for the basic
In an O&M contract for a PV plant, the obligations on the services that are to be provided by the maintenance
owner/developer are generally limited to: contractor under the O&M contract. This section should
include the following:
• Granting the O&M contractor access to the system
and all the associated land and access points. • Cost – this is usually a fixed price per kWp installed.
• Obtaining all approvals, licences and permits • Payment structure (that is, monthly, quarterly
necessary for the legal operation of the plant. or annually).
• Providing the O&M contractor with all documents • Payment indexation over the duration
and information available to them and necessary for of the contract.
the operational management of the plant.
Remuneration for other services includes payment for any
11.5.4 Standards, Legislation and Guidelines services above the basic requirement. This should include:
This section of the contract outlines the various conditions • Method for determining level of other services
to which the O&M contractor must comply while carrying carried out.
out the O&M of the plant. These conditions are contained
• Agreed rates for conducting these services.
within the following documentation:
• Agreed method for approving additional expenses or
• Building or construction permits. services with the owner.
• Planning consents and licences. • Any required spare parts and other components that
are not covered by individual warranties.
• Grid connection statement, the grid connection
agreement and PPA (or similar). 11.5.6 Warranties
• System components installation handbooks.
An availability warranty can be agreed between the
• Applicable legislation. maintenance contractor and the owner of the system. It then
becomes the responsibility of the maintenance contractor
132 Utility Scale Solar Power Plants
to make sure that the system operates at a level greater It is also recommended that this section includes the
than the agreed value. If the system operates below the circumstances in which either the maintenance contractor or
warranted level, then the maintenance contractor may be the developer would be entitled to terminate the contract.
liable to pay a penalty.
11.5.10 Response Time
11.5.7 Legal
The guaranteed response time of a maintenance contractor
The contract will have a section outlining the governing is an important component of the O&M contract. As soon
law and jurisdiction of the O&M contract. The governing as notification of a fault occurs, it is the responsibility of the
law is normally the law of the country in which the project contractor to go to the site within a set period of time. The
is located. A legal succession or a transfer of rights condition faster the response time, the swifter the issues can be diagnosed
is required for the developer to reserve the right to assign the and resolved towards the aim of returning the system to full
O&M contract to a third party. production. The distance between the PV plant and the
contractor’s premises has a direct correlation to the duration of
It is also recommended that every contract has a non- the guaranteed response time.
disclosure agreement. This agreement between the O&M
contractor and the developer will outline the information The time of year coupled with the accessibility to the
that is to be treated as confidential and that which could be site can have a bearing on the actual response time for any
disclosed to third parties. unscheduled maintenance event. Restrictions in access roads,
at certain times of the year, can delay response. Adverse
11.5.8 Insurance conditions can also reduce the size of the payload that can
be transported to the site, thus extending the duration of the
The contract should have a section outlining the insurance maintenance work.
responsibilities of the contractor for the operations and
maintenance activities. This insurance should cover damage to 11.5.11 Selecting a Contractor
the plant, as well as provide cover for employees conducting
the maintenance. When choosing an O&M contractor, his experience should
be thoroughly examined. In particular, the following aspects
It is normal for the O&M contractor to also arrange and should be considered:
pay for the full site insurance.
• Familiarity of the contractor with the site and
11.5.9 Term of Agreement equipment.
11.6
It is normal for an O&M contractor to provide a warranty guaranteeing the availability of the PV plant. In some
cases when the O&M contractor is also the EPC contractor, it is possible for the warranty to include targets for the
PR or energy yield. The agreed availability limits are often based on the independently verified energy yield report, but
with some leeway.
In general, the O&M activities for a solar PV power plant are less demanding than those related to other forms of
electricity generation. This is mainly due to the fact that there are no moving parts in a solar PV system (unless it is a
tracking system). However, maintenance is still an important factor in maximising both the performance and lifetime
of the plant components.
134 Utility Scale Solar Power Plants
There are a number of national or macroeconomic benefits significant levels of solar power are installed in areas with a
which are likely to accrue from the development of solar power weak transmission network.
generation within India. An awareness of these benefits will
aid developers and investors when pitching the case for solar Budget diversions may also be significant as higher
development to policy makers. government budget allocations for solar projects may divert
resources from low income groups. Engagement of the
National economic benefits may typically include developer with the local community (by supporting local
the following: employment, for example) would be one way to work out a
mutually agreeable solution.
• Increased energy security arising from diversification
from coal-fired generation. 12.1.3 Benefits to Developers
• Long term energy price pressures mitigated due to
diversification of generation mix and technology Investment in solar projects offers a number of economic
development. The cost of installing solar power benefits to potential developers, the most important of which
generation is currently more expensive than coal-fired are outlined below:
generation in India. This gap would be expected to
reduce as the solar market matures and coal prices • Preferential tariff and guaranteed returns – Solar
rise. Energy price stability can lead to a wide range projects in India receive a FiT for 25 years.
of social and economic benefits. These include
improved global competitiveness of domestically • Concessional duties and tax breaks – The
produced goods and services, inflation reduction and Government of India has announced a concessional
social cohesion. customs duty of 5% on imports, with an exemption
on excise duty for some project components.
• Reduced dependence on imports resulting from long
term solar project development targets and mandates • Meeting the renewable energy obligation –
set by the National Solar Mission for consumption of Utilities and independent conventional power
domestically produced project components. producers have been mandated by the State
Electricity Regulatory Commissions (SERC) to
• Technology development, which leads to a purchase renewable energy under the Renewable
redeployment of human resources from primary Purchase Obligation (RPO). At present, the
industrial activities to higher value-creating proportion of renewable energy to be purchased
secondary industries. varies from 3% to 5% of the total generation across
various SERCs. This is likely to increase to 15% by
• Climate change mitigation, in line with the NAPCC.
2020.
• Reduction in pollution externalities such as health
and environmental consequences.
• Renewable Energy Certificate (REC) – RECs are
market-based instruments which give the developer
• Increased tax revenue. the option to either sell power produced at the state
specific average power pooled cost, or alternatively to
These benefits must be weighed against the cost of upgrades trade RECs separately.
to major transmission lines. Grid upgrades are likely if
• Certified Emission Reduction (CER) revenue[53].
[53] CER credits can be sold under the Clean Development Mechanism.
See Section 12.3.6
136 Utility Scale Solar Power Plants
• Improvement in corporate image – Investment in produced a benchmark capital cost of INR 169 million/MWp
solar power projects allows developers to demonstrate for solar PV power projects commissioned during fiscal years
their commitment to environmental concerns. 2010-11 and 2011-12. This capital cost is considered to be a
• Business diversification – Development of expertise reference cost in India as no large utility scale projects have yet
and technical skills within the developer organisation, been commissioned.
allowing diversification of income generation streams
and access to a large emerging market. Figure 26 gives the percentage breakdown of cost for a typical
1MWp size project. These costs are discussed in more detail
12.2 Central Electricity Regulatory Commission
in Table 15. It should be noted that the various elements of
(CERC) Cost Benchmarks
the capital cost will vary depending on the technology selected
and other project specific parameters; as an example, while
12.2.1 Capital Cost
the CERC benchmark costs show modules accounting for
approximately 60% of the overall capital cost, it is not unusual to
In order to determine the level of the feed-in tariff, the
see module costs ranging from 50% to 60% of the overall cost.
Central Electricity Regulatory Commission (CERC) has
Land
Preliminary and Pre- [0.89%] Civil and
Operative Expenses General Works
[10.71%] [5.33%]
Evacuation to Inter-
connection [5.03%]
Power
Conditioning
Unit [11.83%]
PV Modules
[60.3%]
For projects commissioned in financial years 2010-11 and been assumed to be INR 0.951 million/MWp for projects
2011-12, the tariff has been structured (assuming a useful life commissioned in fiscal year 2010–11. There shall be an annual
of 25 years) at a levelised rate of INR 17.91/kWh. This tariff escalation of 5.72% over the tariff period.
takes into account a reasonable return of equity, interest on
loan capital, depreciation factor, interest on working capital A more detailed discussion of the O&M costs associated
and O&M costs. with solar PV is provided in Section 12.3.
A more detailed discussion of the tariff structure is provided 12.3 Financial Model
in Section 12.3.
12.3.1 Introduction
12.2.3 Operations and Maintenance
It is clear from the discussion in the previous section
O&M expenses comprising extended warranties, repairs, that many of the economic benefits and costs of solar PV
routine maintenance, employee and administrative costs have project development do not accrue directly to the developer.
Instead, these act as “externalities”, which stem from environment. For example, a site located in a dusty
investment choices made largely on the basis of financial environment is likely to require frequent cleaning of modules.
benefits and drawbacks.
It is difficult to predict the O&M cost over the latter part
The financial benefits and drawbacks to the developer are of the 25 year design life as there are very few large scale
explored in detail through the construction of a full financial solar projects that have been generating for sufficient time
model. This facilitates the identification of key variables to have reached the end of their design life. The modules,
affecting the project value and enables financing decisions. which typically comprise over 60% of the total project cost,
are generally supplied with performance guarantees for 25
The following sections describe the key items and years. However, other project components require routine
assumptions that would be included in the financial modelling maintenance and component replacement. Aside from O&M,
of a typical Indian solar project, and discuss the conclusions operational expenditure will include comprehensive insurance,
that can be drawn from the results of the modelling process. administration costs, salaries and labour wages.
According to a CERC report, capital cost per MWp for There are a number of factors which affect the annual
solar PV plant in India is expected to vary between INR 150 energy yield of a solar PV project as discussed in Section 6.
million to INR 170 million. This total capital cost includes The confidence level of the yield forecast is important, as the
the cost of land, PV modules, mounting structure, inverters, annual energy yield directly affects the annual revenue.
balance of plant and support infrastructure, and start-up costs.
The cost variation largely depends on the project location, the 12.3.5 Energy Price
project design (such as the voltage level of power cables), the
technology utilised and the grid connection cost. Besides the power generated, the solar PV project revenue is
dependent upon the power price. This may be fixed or variable
In addition to overall project cost, there can be significant according to the time of day or year, and must be clearly
variation in component costs depending on the type of PV stipulated in the power purchase agreement.
technology used.
Economic return has historically been the key limiting
A project with crystalline PV technology requires less surface factor for development of large scale grid-connected solar
area per kWp installed compared to thin film modules. As a PV projects. PV has a high initial capital cost. High energy
result, the mounting structure and DC cabling costs are lower. prices are required for projects to be economic. Currently,
However, there is not significant variation in the other cost grid-connected solar projects are highly dependent on policy
components. support initiatives such as grants, feed-in tariffs, concessional
project funding and mandatory purchase obligations.
12.3.3 Operations And Maintenance
(O&M) Cost In India, the power tariffs for solar PV projects are
determined by the Ministry of New and Renewable Energy
O&M costs for solar PV are significantly lower than other (MNRE). Incentive policies include the generation-based
renewable energy technologies. O&M costs depend on many incentives (GBI) and the recently created Jawaharlal Nehru
factors, including the project location and the surrounding National Solar Mission (JNNSM).
A Guide For Developers and Investors 139
Under these regulatory regimes and incentive schemes, there fixed rate tariffs. However, it has potential for better revenue
are five main tariff options for the sale of the renewable power than some of the other options.
that is generated:
All projects should carefully assess the current tariffs
• Demonstration scheme GBI – tariffs aimed at available to them to capitalise on the best rate. It is
supporting pilot projects. advisable to reassess the rate at any stage when the tariffs
• JNNSM scheme – tariffs to encourage both large and vary or new options (for which the plant would be eligible)
small scale projects. become available.
• State Government Incentives – localised tariffs. As India is a non-Annex 1 party under the UN Clean
Development Mechanism (CDM), qualifying Indian solar
• Selling electricity and trading RECs separately.
projects could generate Certified Emission Reductions (CERs).
Under the GBI scheme, the project developer signs a PPA These CERs can then be sold to Annex 1 parties and help
with the relevant state utility grid operator for a period of 10 them comply with their emission reduction targets. This
years, whereas under the JNNSM scheme, PPAs will be signed effectively causes transference of wealth from Annex 1 parties
for 25 years. such as the UK and Germany to Indian developers.
CERC has ruled that projects commissioned in financial Each CER is equivalent to the prevention of one tonne of
year 2010-2011 and 2011-2012 shall have a tariff term of 25 carbon dioxide emissions. The income from CERs can be
years. This term has been fixed on the basis of a reasonable substantial. However, this revenue source cannot be predicted
deemed internal rate of return (IRR). Equity is assumed to as it is uncertain whether the project will be accredited.
comprise 20% of project cost, with a rate of return of 19% Moreover, CER values fluctuate considerably. Therefore,
for the first 10 years of operation, and 24% for the rest of a sensitivity analysis around the CER price (and the period of
plant’s useful life. time for which the project is accredited) is important.
Under the JNNSM scheme, large scale solar projects with an The National CDM Authority under the Ministry of
installed capacity of 5MWp and above—connected to the grid Environment and Forests (MoEF) is the designated authority
at 33kV and above—will sign a PPA with NVVN. This, in in India for approving CDM projects.
turn, shall bundle the power with conventional power and sell
it to various utilities through the RPO. For projects with an 12.3.7 Financing Assumptions
installed capacity of less than 5MWp, connecting to the grid at
less than 33kV, the project developer will sign a PPA with the The project financing structure generally comprises of debt
state utilities. and equity as described in Section 13.
Trading of RECs must be conducted through power The general financial assumptions for a project in India
exchanges within the price-range set by CERC. This range is are as follows:
subject to variation. Given the variability of the price of RECs,
this policy involves a higher level of risk for developers than
• Financing structure – equity 20% and debt 80% as
assumed in CERC tariff order.
A project financial model will calculate a range of project A Minimum DSCR value of less than one indicates the
value indicators in order to allow developers, lenders, project is unable to service the debt in at least one year.
investors and relevant government bodies to assess the project Lenders will conduct sensitivity analysis around the key
economics from several perspectives. variables in order to determine whether the project will be
able to service the debt in a bad year, for example if energy
From an investor’s point of view, a project is generally yield is lower than expected, or operational expenditure is
considered to be a reasonable investment only if the internal higher than expected.
rate of return (IRR) is higher than the weighted average cost of
capital (WACC). Investors will have access to capital at a range 12.3.9 Sensitivity Analysis
of costs; the return arising from investment of that capital
must be sufficient to meet the costs of that capital. Moreover, Sensitivity analysis involves changing the inputs in the
the investment should generate a premium associated with the financial model (such as power tariff, capital cost, and
perceived risk levels of the project. energy yield) to analyse how the value of the project changes
(measured using Net Present Value, Internal Rate of Return, or
Solar projects are usually financed with equity and debt the Debt Service Cover Ratio).
components. As a result, the IRR for the equity component
can be calculated separately from the IRR for the project as a Sensitivity analysis gives lenders and investors a greater
whole. The developer’s decision to implement the project or understanding of the effects of changes in inputs such as power
not, will be based on the equity IRR. tariffs on the project’s profitability and bankability. It helps
them understand the key risks associated with the project.
As returns generated in the future are worth less than
returns generated today, a discount can be applied to future Typical results that are monitored during sensitivity
cash flows to present them at their present value. The sum of analysis include:
discounted future cash flows is termed the net present value
(NPV). Investors will seek a positive NPV, assessed using a • Post tax Project IRR.
discount rate that reflects the WACC and perceived risk levels • Post tax Equity IRR.
of the project.
• Average DSCR.
Lenders will be primarily concerned with the ability of • Minimum DSCR.
the project to meet debt service requirements. This can be
measured by means of the debt service coverage ratio (DSCR), Typical variables investigated during sensitivity analysis are:
which is the cash flow available to service debt divided by the
debt service requirements. The Average DSCR represents the • Capital costs.
average debt serviceability of the project over the debt term. • Operational costs.
A higher DSCR results in a higher capacity of the project to
service the debt. Minimum DSCR represents the minimum • Annual energy production.
repayment ability of the project over the debt term. • Interest rate.
A Guide For Developers and Investors 141
Case Study 6
Economics and Finances
The 5MW plant’s costs increased during the project development and implementation phase, but the escalation was
only 2% of the original predicted cost. The risk of escalating cost was reduced by employing EPC contracts for the major
site works. Projects developed using EPC contracts are able to reduce the risk of cost overruns. Detailed financial models,
including sensitivity analysis should be carried out.
The plant was financed by a combination of equity and loan, including 18% of the final expected loan cost from the
IFC. Cost estimates of the project were provided by the developer at different stages of development and construction.
The first cost estimate was made in 2008/09; a final expected cost was calculated in late 2010. A comparison of the costs
against CERC benchmark values has been made in the table below.
The comparison shows that the CERC benchmark provides reasonable indicative estimates but these must be adjusted
according to project specific details. The main variation in the 5MW Tamil Nadu project compared to the CERC
benchmark can be attributed to the cost of PV modules and electrical works. The price of PV modules was found to be
13% higher than the CERC benchmark and the cost of electrical works was 113% higher than the benchmark. The high
cost of electrical works was partly due to the requirement for an electrical transmission line extension.
142 Utility Scale Solar Power Plants
The benchmark land costs (0.3 INR per acre) provided by CERC were in line with land costs incurred during the
project. However, the project used 13.2 acres of land per MW installed compared to the CERC benchmark of 5 acres. As
a result, the cost of land per MW installed was more than double that of the CERC estimate.
In total, the Tamil Nadu project cost 20% more than the CERC benchmark. Although the costs increased during the
project development and implementation phase, the escalation was only 2% of the original predicted cost. The risk of
escalating cost was reduced by employing EPC contracts for the major site works.
The table below shows how the operating expenditure (estimated for the 5MW Tamil Nadu plant) compares to CERC
benchmarked costs, including the expected yearly escalation rates. As the operating costs are only estimates at the time of
writing, no lessons can be learnt at this stage. However, the indication is that the Tamil Nadu plant O&M cost estimates
are in line with those of CERC.
Project costs and energy yield predictions were incorporated into a financial model to assess the viability of the project
with a tariff rate of INR 17.91/kWh. A sensitivity analysis was also conducted to determine the viability of the scheme
under different stress tests.
A Guide For Developers and Investors 143
Conclusions
• CERC benchmarks can be used to make reasonable estimates of project costs at the feasibility stage.
• Costs need to be adjusted according to the specifics of the project, such as the distance to the grid
connection point.
• Predicted costs for the Tamil Nadu project have been adhered to as they fall within the 2% escalation range.
Discrepancies with the CERC benchmarks were due to module cost, electrical connection costs and the total area
of land required for the project.
• Projects developed using EPC contracts are able to reduce the risk of cost overruns.
The project evaluations (due diligence) identify the risks and In some cases, the equity partners will not commit equity to
methods of mitigating any risks prior to investment. Where projects unless they have received firm commitments of debt
the project has inherent risks, the exposure to these risks project finance or leasing finance.
will be negotiated between the parties and reduced wherever
possible with insurance. The debt portion is typically provided by an investment
bank providing project finance or leasing finance. The debt
The following sections cover each of these steps portion is the larger investment, which is typically 80-85% of
and processes. the total project cost.
13.2 Project Financing Despite the recent turmoil in the international credit
markets, many financial institutions are willing to provide
13.2.1 Financing Alternatives long term finance for the solar energy market. Individual
projects from smaller developers may receive financing with a
The equity portion can be provided by the developer or loan to value (LTV) ratio of 80%, whereas portfolios of solar
from equity partners that sign agreements or letters of intent projects from experienced developers may be financed with a
to purchase the projects from the developers. Equity partners LTV ratio of 85%. The usual term of a project finance loan is
may be individual firms, developers or equity funds managed approximately 18 years.
by management firms, bank equity fund managers or pension
fund managers. Large corporations and utilities may develop solar plants
without the need for project finance; these projects are
The equity funds can be used as the seed capital to start the financed from the corporate balance sheet. But the corporation
construction of the project, following completion of the design or utility will still conduct a similar due diligence review before
and environmental studies, legal analysis, permit applications committing the project funds.
and grid connection applications. The equity is typically
around 15-20% of the total project investment cost.
A Guide For Developers and Investors 145
Both equity and debt finance investors typically evaluate the The due diligence phase of evaluating a project takes three
legal, permitting, consent and technical due diligence areas of main forms:
the project. The due diligence conducted at the equity stage
may be based on preliminary technical information. On the • Legal due diligence – assessing the permits and
other hand, the due diligence for project finance is conducted contracts (EPC and O&M).
at a later stage and often supported with detailed technical • Insurance due diligence – assessing the adequacy of
information and designs. the insurance policies and gaps in cover.
• Applies for and receives planning permits and • Layout in the land area available.
consents.
• Appropriate buffer zone around the plant to
• Applies for and receives grid connection offer or account for shading/other activities.
letters of intent.
• Overall size appropriate for the grid connection.
• Applies for feed-in tariff (FiT) and/or PPA.
• Layout of the PV modules, mounting and/or
The equity investor typically evaluates the work in the list trackers, and inverters:
above. The next key step is to validate and confirm all the
• Assessment of level of inter-row shading.
permits, consents and power purchase agreements. Along with
the equity partner, the project SPV can look towards securing • Access to plant components for maintenance
the technical solution, detailed design and equipment supply. In and installation activities.
some cases, these are carried out by the developer, depending on • Electrical design layout and sizing:
the stage in which the equity partner enters the project.
• Assessment of cable losses in the
The project finance partner can often influence the choice of DC/AC cabling.
the equipment technology, based on what they perceive to be • Assessment of appropriateness of the cable
“bankable”. This often affects the selection of modules, inverters placement and connectors.
or mounting structures. One way to avoid such issues is to have
• Appropriateness of the earthing and
discussions early in the design phase with the project finance
protection systems.
partner. This can help assess the equipment selection and satisfy
the requirements of all partners. • Compliance to safety standards.
146 Utility Scale Solar Power Plants
• Technology review of major components (modules/ The process of due diligence can require considerable
inverters/mounting or trackers): effort from the developer to satisfy the requirements
of the lenders. It is important that the developers have
• Suitable for environment.
realistic financial models with contingencies clearly shown.
• Integration of components. Alongside, it is also imperative to have a sensible construction
programme, which takes contingencies into account. Such
• Track record of suppliers and models.
a programme will clearly show that the target deadlines are
• Quality and compliance certificates. realistic and achievable.
• Compliance to safety standards.
The due diligence process is likely to identify risks, and help
• Warranties. develop solutions to mitigate the issues found. It may result
in changes in the design or use of components in the plant to
• Design life.
make the project “bankable” for the lenders.
• Degradation assumptions.
13.3 Risks
• Energy yield assessments:
• Appropriateness of any assumptions made. This section describes the key risks considered to be
applicable to an investment in solar PV projects. The list
• Source of solar irradiation data.
of risks identified below is not an exhaustive list. Investors
• Assessment of shade. and developers should satisfy themselves that the level of
risk attached with any development is appropriate to their
• Degradation assumptions.
investment criteria. Developers and investors should make
• Uncertainty analysis. every effort to mitigate the risks where possible.
The interest rate is payable at specified dates before maturity. Besides considering quality and warranties, the overall
This can be the best form of natural hedging to match long configuration of the PV power plant must be designed
term income with long term sources of finance. correctly. This will ensure that the maximum power reaches
the grid based on the gross irradiance reaching the modules.
13.3.1.2 Leverage
The technology and manufacturer choice for the inverters
In cases where projects are to be financed through a is also important for ensuring trouble-free operation suited to
mixture of equity and non-recourse debt finance, leverage the environment and design of the PV plant. Warranties and
may potentially increase the total return of the equity maintenance activities for the inverters need to be carefully
investors. But it may also lead to increasing losses in adverse assessed to ensure that the risk of inverter failure is minimised.
market conditions.
13.3.2.3 Technology Failure
13.3.1.3 New Business Start-up
Generation of electricity involves mechanical and electronic
As the solar PV industry is fairly new, very few companies processes. These may fail under certain conditions, leading
have a long history in operating in this renewable energy to loss of revenue and repair or replacement costs. Selection
sector. A possible way to tackle this problem is to have key of modules, tracking systems (if used) and inverters should
contractor and strategic partners that have experience in be based on the track record of manufacturers—and the
constructing and operating solar power plants. warranties they offer. These warranties help reduce the risk
of technology failure in the initial years of the PV plant’s
13.3.2 Technical Risks operational life.
It is important to consider the quality of modules (for One of the key factors in determining the energy yield of
example, checking whether degradation will occur faster than a solar plant is the solar irradiation at the site. Changes in
expected) and the strength of the module manufacturer’s weather patterns such as cloud cover, rainfall and heat waves
warranties. Any problems in the installation are usually could reduce the energy output and, consequently, investor
identified within the first year and corrected under EPC returns. However, meteorological assessments and long term
(construction) warranties. Later, problems can be rectified averaging show that inter-annual variation over the lifetime of
under manufacturer’s warranties. But as far as possible, it is a PV plant is generally quite low, generally in the order of 5%,
preferable to avoid any interruptions to production. depending on location.
Given the long term nature of the project, choosing 13.3.2.5 Solar Module Degradation
the right technology is essential in achieving consistent
results and maximising power output over the life of the The efficiency of solar modules as well as their degradation
project. A productive and viable PV power plant will (loss of performance) has a direct effect on the yield of a solar
automatically become an attractive proposition to potential plant. The degradation is indicated by the supplier (usually less
buyers in the future. than 1% per year). Any unexpected loss of performance could
have an adverse effect on the business.
148 Utility Scale Solar Power Plants
Module manufacturer’s power warranties generally cover The connection to the third party distribution or
larger losses of power due to degradation. However, the transmission network is often non-contestable. Therefore,
warranties need to be reviewed carefully for exclusions. The the final grid connection is reliant on the works of the third
financial strength and backing of the module manufacturers party network operator or their contractor. Grid connection
should be assessed to verify that the manufacturer can support contracts and deadlines should be finalised to mitigate
any claims against their warranties. In some cases, insurance this risk.
policies may be taken out by the manufacturers to cover
warranty claims. 13.3.3.5 Delay in Obtaining an Operating
Permit
13.3.3 Pre Completion Risks
In some jurisdictions, the relevant authorities must
13.3.3.1 Cost Overrun determine whether the construction of the plant and
connection facilities has been carried out in conformity with
Exposure to changes in the prices of components can the approved design, and whether they comply with the legal
account for a cost overrun. A change in prices for certain key requirements. Delays or difficulties in obtaining the operating
components, in particular modules and inverters, may have an permit may affect the income and profitability of the solar PV
adverse effect on the bottom line. plant.
Every operational solar power station engages an O&M It is important to note that insurance is no substitute for quality
Contractor to carry out the day-to-day maintenance of (design and components), and should not be seen as a ‘magic wand’
the solar power station. Inefficiencies in the operation and to mitigate issues related to design, equipment or contract.
management of the project could reduce the energy output.
This can be reduced by adding performance clauses within the 13.4.2 General Liability Insurance
O&M contract, based on the availability of the PV plant and
targets for energy yield or performance ratio. General liability insurance covers policyholders for death or injury
to persons or damage to property owned by third parties. General
13.4 Insurance liability coverage is especially important for solar system installers, as
the risk to personnel or property is at its greatest during installation.
13.4.1 Introduction
13.4.3 Property Risk Insurance
At present, the insurance industry has not standardised
the insurance products for PV projects or components. A The PV system owner usually purchases property insurance to
number of insurers are providing PV insurance policies, but protect against risks not covered by the warranty or to extend the
underwriters’ risk models have not been standardised. The coverage period. The property risk insurance often includes theft
data required for the development of fair and comprehensive and catastrophic risks.
insurance policies are lacking as insurance companies often
have little or no experience with solar projects. Property insurance typically covers PV system components
beyond the terms of the manufacturer’s warranty. For example,
However, demand for PV insurance is increasing. In general, if a PV module fails due to factors covered by the warranty, the
large PV systems require liability and property insurance, manufacturer is responsible for replacing it, not the insurer.
and many developers may also opt to add policies such as However, if the module fails for a reason not accounted for in
environmental risk insurance. the warranty, or if the failure occurs after the warranty period,
the insurer must provide compensation for the replacement of
Though PV insurance costs can be quite high, it is likely the PV module.
that rates will drop as insurers become familiar with PV
plants and as installed capacity increases. A recent study 13.4.4 Environmental Risk Insurance
by NREL stated:
Environmental damage coverage indemnifies PV system owners
“Insurance premiums make up approximately 25% of a against the risk of either environmental damage inflicted by their
PV system’s annual operating expense. Annual insurance development or pre-existing damage on the development site.
premiums typically range from 0.25% to 0.5% of the total
installed cost of a project, depending on the geographic 13.4.5 Business Interruption Insurance
location of the installation. PV developers report that
insurance costs comprise 5% to 10% of the total cost of energy Insurance against the risk of business interruption is often
from their installations, a significant sum for a capital-intensive required to protect the cash flow of the solar project. This insurance
technology with no moving parts.” policy can often be a requirement of the financing process.
150 Utility Scale Solar Power Plants
14. CONCLUSION
It is widely being accepted that solar energy has a major part to play in promoting ecologically sustainable growth and
tackling climate change. In addition, tapping the power of the sun can improve the energy security of those countries,
including India, which are currently dependent on fossil fuel imports.
India has excellent reserves of solar resource and is well-placed to benefit from the development of a solar energy
industry. However, there are numerous and varied challenges to be overcome at various levels in order to establish a
successful solar power industry.
It is vitally important that developers and financiers of solar energy projects follow best practices in developing,
constructing, operating and financing projects. It is hoped that this guidebook will go some way towards promoting such
best practices in the sector. However, it should be borne in mind that there is no substitute for experience and expertise.
A Guide For Developers and Investors 151
Appendix A
Concentrated Solar Power: A Guide For Developers and Investors
1. INTRODUCTION
The objective of this report is to provide an overview of Concentrated Solar Power (CSP) technology. The report is
intended to be read in conjunction with “Utility Scale Solar Power Plants: A Guide for Developers and Investors”, and
focuses on those areas where CSP differs from PV. The report describes the following technologies and associated aspects
of CSP projects:
The renewed market interest in CSP since 2007 is due Financial viability of projects will depend upon the resource,
to a combination of rising fossil fuel costs, firm renewable technology and project costs, and the extent of government-
energy targets, and substantial governmental subsidies or driven financial support. Current costs of the technology and
other financial support mechanisms. These factors have constraints on financial support indicate that only projects
helped CSP technology become commercially attractive, that are located in the areas with the highest direct normal
resulting in increased investment in CSP projects. There is irradiation are likely to be viable in the near future with
now 900 MW of total capacity commissioned and a further annual average direct normal irradiation values of greater than
1,900 MW under construction. 2.2 MWh/m2/year or 6.0 kWh/m2/day[4].
[1] SgurrEnergy CSP project database collated from information in Photon [2] IEA; Technology Roadmap, Concentrating Solar Power; 2010
International, Sun & Wind Energy [particularly 6/2010 issue], NREL CSP [3] National Renewable Energy Laboratory http://nrel.com
projects database [http://www.nrel.gov/csp/solarpaces/by_project.cfm], and [4] SgurrEnergy for Asian Development Bank; Central-West Asian Countries Solar
developer or project websites. Potential, Concentrated Solar Power; Draft report, November 2010.
A Guide For Developers and Investors 153
Concentrated
Solar Radiation
CONCENTRATOR
Heat
Heat
Rejected
Receiver
Concentration Losses
Losses
Work/Electricity
Heat
Rejected
Receiver
Losses
Concentration
Losses
Receiver
Parabolic
Power trough
tower
Receiver
Concentrator
Steam at 350-550°C
80-120 bar
Molten Salt
Air or Helium
at 600-1200°C Absorber tube
Parabolic 1-20 bar
dish
Sunlight
Fresnel
Receiver reflector
Secondary
Reflector
Fresnel
Reflector
Concentrator
Steam Turbine
Solar Field
Solar
G
Reheater
Condenser
Hot Salt Tank
Solar
Superheater
Steam
Cooling Tower
Generator
Deareator
Water Supply
Cold Salt Tank
Solar
Preheater
Expansion Vessel
Figure 4: Typical CSP Power Plant Schematic (Parabolic Trough with Storage)[6]
each dish unit. Utility scale parabolic dish power plants would Geographically, south-west USA dominates in terms of
therefore comprise many dish units. Energy storage is not operating experience and future projects, with Spain dominant
included in such plants since the heat engine operates directly in projects recently completed and under construction.
from the solar concentrator heat source.
4.1.2 CSP Cost Trends
The arrangement of linear Fresnel reflectors and power
plant could be similar to that for parabolic troughs. However, Due to variations in the configuration of the CSP plant (as
since current linear Fresnel reflector installations yield lower described later in this guide), generating cost or levelised cost
temperatures than parabolic troughs, and hence lower of electricity (US$/MWh) is a much better indicator of true
efficiency conversion to electrical power, linear Fresnel costs than installed cost (US$/MW installed capacity). The
technology is generally used for direct heat applications rather capacity factor (or load factor) varies over a wide range for
than utility power generation. CSP plants. Plants with similar areas of solar field, and annual
energy yields, may have differently sized generators depending
4.1.1 Uptake and Track Record on whether or not they have energy storage.
The uptake and track record of concentrating solar thermal Generating costs have been calculated from available
technologies is shown in Table 1 and Figure 5[7]. cost and performance information for recently completed
projects and projects under construction. This information
Parabolic trough technology is by far the most established is shown in Figure 6[9] (as a comparison the levelised cost of
technology in terms of operating experience. It also shows the generation for PV plants in 2010 was estimated to range from
highest anticipated build rate, taking into account projects
which have reached the advanced planning stage[8].
[7] Project data in the SgurrEnergy CSP database is collated from information in development permits are in place or subject to final approval, power purchase
Photon International, Sun & Wind Energy [particularly 6/2010 issue], NREL agreements are in place and construction contracts and financing are at an
CSP projects database http://www.nrel.gov/csp/solarpaces/by_project.cfm, and advanced stage of negotiation.
developer or project websites. [9] Calculated from selected project data in the SgurrEnergy CSP database
[8] Projects in the advanced planning stage are defined as projects where assuming 8% discount rate.
12, 000
10, 000
Installed Capacity (MW)
8, 000
6, 000
Advance planning
4, 000 Construction
Operation
2, 000
0
Parabolic Parabolic Power Parabolic Linear
Trough Trough Tower Dish Fresnel
hybrid Reflector
or ISCC
Figure 5: Implementation of CSP Technologies
1000
900
700
400
Power Tower without Storage
300
200
Power Tower with Storage
100
0
1.50 1.70 1.90 2.10 2.30 2.50 2.70 2.90
Direct solar insolation (MWh/m /year)
2
Figure 6: Generating Cost for Recently Completed and Under Construction CSP Projects
158 Utility Scale Solar Power Plants APPENDIX A
170 US$/MWh (Middle East) to 400 US$/MWh (Northern to energy consumers. However, the pool price option enables
Europe). While there is no clear trend, or differentiation generators to sell electricity at the market price and receive a
between the various technologies, it is clear that the sites with lower premium tariff (€254/MWh) linked to the average or
the greatest solar resource achieved the lowest energy costs. reference electricity price. Once the additional incentive for
participation in this option and the continuity supplement
Consistent CSP cost trends have yet to emerge but the are considered, the total tariff can typically be of the order
following general comments can be made: of €330/MWh (US$440/MWh) dependent on pool power
price. The terms for the feed-in tariff limit the capacity of CSP
• Reported costs vary widely: plants in Spain to 50 MW each, and encourage incorporation
• Projects which have recently been completed, of energy storage.
or are currently under construction, show
increasing cost trends; and Recent costs have been two to three (or more) times
previous current estimated costs – these factors should
• Previously projected cost reductions are currently
arguably be applied to the future projected costs. For all CSP
not being realised.
technologies in the foreseeable future, substantial economic
• Generating costs will depend on: support will be required for project economic viability,
through a support mechanism which is specifically designed to
• Direct normal irradiation (DNI) at the
project location; support CSP projects.
• Increased competition in the supply chain; and A parabolic trough system is composed of a solar field, a
• Learning rate effects. power block and an optional thermal storage system. The solar
field consists of parallel rows of parabolic, trough-shaped solar
Recent CSP development in Spain has been driven by collectors that focus direct normal solar radiation onto tubular
the feed-in tariff which was €269/MWh (approximately receivers located at the focal point of the collectors. The
US$365/MWh at the time of writing) for the regulated tariff collectors are installed on single-axis tracking structures that
option, for years 2 to 25, linked to the average or reference can be aligned on a north-south or east-west horizontal axis
electricity price, with a lower tariff thereafter. Elimination depending on the electricity demand profile.
of feed-in tariff for year one has recently been introduced to
reduce a government budget deficit or pass-through of costs
A Guide For Developers and Investors 159
Each receiver consists of a metal tube with a solar radiation used to generate steam which is fed into a conventional steam
absorbing surface in a vacuum inside a coated glass tube. turbine to produce electricity.
A heat transfer fluid is circulated through the receivers and
transports the heat generated at the receivers to a series of heat Alternatively, parabolic troughs may be used for steam
exchangers in the power block of the plant. The temperatures augmentation or to supply process heat.
achievable are generally in the region of 400°C. The heat is
Examples of a parabolic trough collector and a parabolic array. Mirrors need to be highly reflective to avoid losses and
trough collector solar field are shown in Figure 7 and Figure 8. durable to resist the harsh environments encountered in the
desert. The majority of mirrors currently used are of glass type
The main elements of the collector system are: with a reflective backing. Mirrors can be manufactured from
thick or thin glass; however, thick glass mirrors are currently
• Reflector the most commonly used.
• Receiver tube
Thick glass mirrors are typically constructed of 3-5mm
• Heat transfer fluid thick tempered glass or float glass (glass made by floating
• Base frame molten glass on a bed of molten metal), which due to the high
curvature required is normally pre-curved during construction.
• Tracking system The mirrors are typically fixed directly on to the parabolic
• Connecting elements trough supporting structure.
Research[12], including commissioning of a pilot plant, is collectors besides helping rectify minor misalignment issues.
currently underway investigating the use of molten salts (a Some problems were observed during operations and the flexible
mixture of sodium and potassium nitrate) as a heat transfer hosing was replaced with a ball joint. This eliminated the
fluid. Molten salts are also used for energy storage as described original problem. However, some concerns have been raised with
in Section 4.5. However, as the freezing point of these salts is regard to the ball joint at high temperatures. Further research is
typically above 200°C, a mechanism needs to be in place so that being performed in this area.
the molten salt does not freeze in any component of the system,
including the pipe runs in the solar field from the receivers to The orientation of the solar field can be along either a north-
the power island. south axis or an east-west axis. A north-south axis is the norm to
allow collectors to track the sun’s azimuth over each day, hence
Other heat transfer fluids undergoing research include maximising annual output. An east-west axis, in contrast, allows
pressurised gases and Diphenyl. for seasonal adjustments for the sun’s elevation and latitude, thus
maximising mid-day output. Since adjustment is seasonal rather
4.2.4 Base Frame, Tracking System and than daily, east-west axis systems could dispense with motorised
Connecting Elements and automated tracking systems. Usually, utility scale projects in
or near the tropics employ north-south axis systems.
A structure is required to support the concentrators and allow
the mechanism to track the daily path of the sun as it moves 4.2.5 Examples from Industry
across the sky. The plants commissioned in the 1980s used a
supporting structure designed by Luz International. Since then, To date, parabolic trough technology is the most widely
there have been several evolutions from the original design. One deployed of the concentrating solar power technologies.
such structure is the Euro Trough torque box design. Between 1984 and 1990, nine parabolic trough CSP plants,
comprising 354 MW of generation capacity, were installed in
The torque box is designed so that collector elements can the Mojave Desert of California in the south-west of the United
be connected together on one drive resulting in the reduction States. These pioneering plants were the SEGS plants of Luz
of the total number of drives and interconnecting pipes. The International; they are still in operation today having produced
overall reduction in components results in a reduction of cost more than 11,000 GWh of electricity.
and of the thermal losses in the system.
The technology ceased to develop for a period of time
Throughout the day, the parabolic troughs are normally set and from 1990 until 2007 no further commercial scale
on tracking mode by electric motors driving through gearboxes, developments were commissioned. From 2007 to date a
or directly by hydraulic drives. Hydraulic drives, which provide further 464 MW of parabolic trough power plants have been
mechanical energy to move the collector, are currently the most commissioned, 400 MW of which was installed in Spain.
common tracking mechanism used.
An example of a recently completed project is the 50
Generally, the solar field comprises many collectors. Earlier, MW Solnova I project which is reported to have achieved
collector panels were connected to one another by flexible significant construction and alignment (tracking) technology
hosing, which facilitated movement between neighbouring improvements. In addition 1,310 MW of capacity is currently
under construction in Spain and the USA, with 7,960 MW at
[12] Sandia National Laboratories - National Solar Thermal Test Facility
the advanced planning stage, mainly in south-west USA.
A Guide For Developers and Investors 163
4.2.6 Losses
Losses within a parabolic trough plant arise from geometrical, power block efficiently, depending upon the heat transfer
optical and thermal factors. These losses, which are described fluid used. Thermal losses mainly consist of convection and
below, have to be given serious consideration during the design conduction losses to air and surrounding elements.
and siting stages of the plant development process.
The maximum theoretical efficiency of a steam turbine plant
Spacing between collectors must be adequate to minimise (or any other heat engine) is limited by the Carnot Efficiency,
shading losses, taking into account daily and seasonal variations defined as:
in the sun’s path. As well as lateral spacing, losses in the effective
tinlet steam - tcondenser outlet
collector length will also have to be taken into account in sizing ηCarnot=
tinlet steam
the solar field.
Where:
Optical and thermal losses will be present within the receiver ηCarnot = Carnot Efficiency, (%)
and will need to be accounted for when determining a plant’s
tinlet steam = Temperature of steam at inlet to turbine (K)
overall energy production. Optical losses are associated with the
reflectivity, transitivity, absorptivity (the fraction of incident tcondenser outlet = Temperature of water at outlet of condenser (K)
radiation that is absorbed) and beam irradiance inception factor
properties of the receivers, including thermal incidence angle. Actual efficiency will always be lower than the theoretical
maximum. As the steam temperature in a concentrating solar
Thermal losses and flow losses occur in the solar field plant is likely to be lower than in a conventional steam power
between the receivers and the power island, which places an plant, and condenser temperatures may also be higher, the
upper limit on the size of solar field which can supply each achievable efficiency is correspondingly lower.
13% 14%
Others
6% Land Preparation
Foundations
Thermal Oils
11%
27% Tubes
Mirrors
4%
Structures
11% Installation
14%
Figure 9: Component Cost as a Percentage of Overall Plant Cost[13]
[[13] http://www.leonardo-energy.org/csp-training-course-5-lessons
164 Utility Scale Solar Power Plants APPENDIX A
12,000
10,000
Installed Cost (1000 US$/MW)
8,000 Other
Site Infrastructure
6,000 Power block / Steam plant
Storage
2,000
4.2.7 Costs
The following section explores the installed and generating on these inputs, the levelised cost of energy was US$377/
cost of a reference Parabolic Trough plant. MWh, with the largest component of this cost being capital
repayment on the cost of the project.
Figure 9 provides a high level overview of the cost of each
component in the solar field as a percentage of the overall cost The main opportunities for cost reduction have been
of the parabolic trough concentrator system. reported as:
[14] NREL SAM model reference plant information from: Technical Report
NREL/TP-550-47605, July 2010; Parabolic Trough Reference Plant for Cost
Modelling with the Solar Advisor Model (SAM); C. Turchi.
A Guide For Developers and Investors 165
350
300
Generating Cost (US$/MWh)
100
50
Conclusions
Parabolic trough technology is currently the most established concentrated solar power technology. It can therefore be
considered to be of relatively low technology risk. The main application is utility scale power generation, although it is also used
for smaller scale power generation or process steam applications.
The relative maturity of this technology is an advantage, as this has allowed refinement of many elements of the system. A large
number of projects are currently in development, as seen in Section 4.2.5, reinforcing the technical viability of the technology.
Drawbacks include:
• shortage of manufacturers for key components (such as the power trough, receivers and mirrors) limits available
manufacturing capacity and increases supply lead times;
• limited maximum temperature, due to the limited concentration ratio achievable, which limits the maximum efficiency;
• requires to be installed on flat ground, typically with a slope of less than 3%.
Despite substantial financial support, the economic viability of this technology will depend greatly on the actual costs of
completing projects currently under construction, as well as the effects of expected future reductions in costs from this baseline.
166 Utility Scale Solar Power Plants APPENDIX A
4.3 Power Tower 4.3.1 Heliostat and the Tracking and Control
Mechanisms
Power tower systems (see figure 12), also known as central
receiver systems (CRS), consist of: A heliostat is an instrument consisting of a mirror mounted
on a structure which allows the mirror to rotate. This allows
• a heliostat field; direct solar radiation to be steadily reflected in one direction,
• a tower and receiver; despite the movement of the sun. The heliostat should be
positioned so that the reflected ray is consistently orientated
• a power block; and towards the receiver.
• an optional thermal storage system.
Each heliostat is composed of a flat reflective surface,
The field of heliostats (flat, dual-axis tracking mirrors) a supporting structure and a solar tracking mechanism.
focuses direct normal solar radiation onto a receiver located Currently, the most commonly used reflective surface is the
at the top of a tower at the centre of the heliostat field. The glass mirror. Membrane technology is under development
receiver absorbs the concentrated radiation and transforms it consisting of a thin film reflective membrane stretched across
into thermal energy in a working fluid, which is then pumped a mounting structure. This technology is still in its infancy
to the power block. The power block generates steam (from and is not yet commercially available. Problems observed
the heated fluid) to drive a conventional steam turbine and with stretched membrane heliostats include the durability
generator to produce electricity. of the reflecting membrane and possible shape change of the
heliostats surface due to wind effects. Heliostat sizes vary
The temperatures achievable with power tower systems widely and aperture areas of up to 150 m2 have been assessed
are greater than those achievable through parabolic trough experimentally.
technology, and are in the region of 400-550°C. Temperatures
of up to 1000°C are being mooted for future plants that In order to function properly, the heliostats must be cleaned
will have demonstrable improvements in beam focussing at regular intervals as dirty heliostats can greatly reduce the
on the receiver. This would enable much higher efficiency efficiency of the entire system.
of conversion from heat to mechanical energy in the steam
turbine, and thence to electrical energy in the generator. One difficulty encountered with the Abengoa Solar PS10
pilot plant (in Spain) was related to the wind conditions
Although power tower technology is commercially less under which the heliostats could be utilised. In wind speeds
mature than parabolic trough technology, a number of greater than 10 m/s the heliostats must be stowed (secured in a
components and experimental systems were field tested as horizontal position) in order to avoid structural damage of the
early as the 1980s and early 1990s. components. Very high wind speeds could cause damage.
The principal components of power tower systems are The heliostat field is normally arranged to surround the
described in further detail in the following sections. power tower. The most common layouts utilise a full circular
field or a surrounding field in a north/south direction.
tracking, each heliostat has its own individual control system. Since the effectiveness of focussing irradiation on the receiver
The tracking algorithm takes into account various factors such diminishes when the heliostats are at too great a distance from
as the distance from the heliostat to the receiver. the receiver, large power projects may comprise of more than one
power tower, each with its own heliostat field.
4.3.2 Receiver, Heat Transfer Medium and Tower
Experimental projects, such as the 2 MW Eureka tower
The receiver transfers the concentrated solar energy reflected constructed by Abengoa Solar, are testing higher temperature
from the heliostats to the transfer medium. Dependent on the technologies to achieve increased efficiency.
technology, the receiver can be a boiler or steam drum. This
directly produces superheated steam at around 550°C and a 4.3.3 Examples in Industry
pressure of 160 bar for supply to the steam turbine or steam
storage tank (as in the case of the BrightSource technology[15]). Currently there is 44 MW of power tower capacity in
Alternatively, molten salt can be used as the heat transfer fluid operation, 416 MW under construction, and 1,291 MW at an
and heat storage medium – see Section 4.5 for further details advanced planning stage.
of heat storage and molten salts.
The first power towers to be built (of pre-commercial scale)
The tower supports the receiver, which needs to be located were the 10 MW Solar One and Solar Two plants in southern
at a certain height above the heliostats to avoid, or at least California. Both installations were developed as demonstration
reduce, shading and blocking of the heliostats. Tower heights plants. Solar One was operated from 1982 to 1988 and, after the
can vary from 50 metres to up to 165 metres depending on the initial test and evaluation phase, operated reliably.
distance of the heliostats from the tower.
In Solar One, the water was converted to steam in the receiver
and used directly to power a conventional Rankine cycle steam
[15] BrightSource website, technology section turbine. The heliostat field consisted of 1818 heliostats with an
168 Utility Scale Solar Power Plants APPENDIX A
aperture area of 39.3 m2. The plant generated power for eight In June 2010 the World Bank approved a $200m loan to
hours per day at summer solstice and four hours per day close co-finance a 100 MW power tower plant near Upington,
to winter solstice. Although Solar One demonstrated that South Africa.
power tower technology could be successful, it also revealed
the disadvantages of a water/steam system, such as intermittent In Rajasthan, India, ACME are commissioning a 10 MW
operation during cloud cover and the lack of thermal storage. plant, based on technology from eSolar (USA), and have plans
to scale up to production of 50 MW units and implementation
Solar Two was built to replace the original Solar One power of utility-scale projects.
tower. The aim of the re-designed Solar Two was to test
and validate the use of nitrate salt technology, to reduce the 4.3.4
technical and economic risk of power towers, and to stimulate
the commercialisation of power tower technology. The plant Conclusions
was built with sufficient thermal storage capacity to allow it
to operate at full capacity for up to 3 hours after the sun had Power tower technology is the second-most proven technology
set. The conversion of Solar One to Solar Two required a new for utility scale power generation after parabolic troughs.
molten salt heat transfer system including the receiver, thermal
storage, piping, steam generator and a new control system. Higher operating temperatures allow for higher performance
and hence potentially lower costs than parabolic troughs.
The first European plant and the first commercially However, there is very limited information available on costs
operational power tower plant to be installed was PS10, near which have been achieved, and on what is likely to be achieved
the Spanish city of Seville, which has a capacity of 11 MW. in the future; consequently, cost estimates remain uncertain.
This plant was installed by Abengoa Solar and was completed
in the first quarter of 2006. The project has a heliostat field of Many developers are now adopting receivers which produce
624 movable mirrors with a surface area of 120 m2. This field steam directly in the receiver, as this type of steam receiver is
concentrates the sun’s rays on the top of a 115 m tower where currently of lower cost and risk than molten salt receivers.
the solar receiver and steam turbine are located.
A major factor in the scalability, performance and cost of
A second plant, the 20 MW PS20, was installed in energy using power towers is the effectiveness with which
2009 next to PS10; the design of PS20 was based on the the heliostats focus the beam on the central receiver. This is
original PS10 plant and included a number of significant an area suitable for significant technical development and
improvements. These included a higher efficiency receiver, potential cost reduction.
various improvements in the control and operational systems
and a better thermal energy storage system. PS20 consists The terrain requirements for power tower plants are not as
of a solar field made up of 1,255 heliostats with an aperture restrictive as those for parabolic trough technology, and plants
area of 120 m2. This field concentrates solar irradiation on can be installed in areas with terrain gradients of up to 5%.
the top of a 162 m tower which produces steam to drive the
electrical generator. The actual costs of completing projects currently under
construction, combined with future reductions in costs from
Plants under construction and at an advanced planning this baseline will be key to economic viability of power tower
stage are dominated by large scale (29 to 200 MW) plants technology, even where substantial economic support is in place.
in the USA. In Europe, there is a 17 MW plant due for
commissioning in Spain in 2011.
A Guide For Developers and Investors 169
• Performance of production units, including efficiency The power system boiler is largely designed on a project-
and reliability specific basis and requirements depend on the characteristics of
• Ramp up of serial production of the units, including the CSP technology used.
quality and costs
Turbines are supplied by established manufacturers, such as
• Longevity of units.
Siemens, GE, Alstom or MAN Turbo. The Siemens SST-700
• Overall costs for construction of the site steam turbine, with a capacity of up to 175 MW is commonly
infrastructure, including the foundations, access, and used for CSP applications. However, a specific CSP system
electrical connections for the numerous dish units. steam turbine has not been released at this time.
• Robustness of implementation plans, costing and
commercial case. Due to the nature of CSP, the heat exchangers/boilers and
turbines are subject to significantly more thermal cycling
(heating and cooling) than conventional steam plants, which
run for long periods at a time without outages. This cycling
reduces the lifetime of the heat exchangers / boilers and can
cause wear and damage to the turbine if the steam quality is
not maintained. Low temperature steam turbines which are
being proposed for larger solar fields or direct steam systems
may not be well tested technology solutions.
A Guide For Developers and Investors 171
3500
3000
(MW)(kWh/m²)
2500
Irradiation
2000
capacity
Advance planning
Solar
1500
Construction
Installed
Horizontal
Operation
1000
500
0
Parabolic Parabolic Power Tower Power Tower
Trough without Trough with without with Storage
Storage Storage Storage
Figure 14: Implementation of Energy Storage in CSP Plants
4.6.1 Overview
The choice of whether or not to include energy storage
Energy storage is an established option for parabolic trough depends on local market conditions.
and power tower plants. Figure 14 compares the uptake of
CSP plants with and without energy storage. To date, in the south-west USA, parabolic trough plants
have typically been installed without energy storage as the solar
CSP with thermal energy storage, hydro with resource is coincident with times of peak load (largely due to
impoundment, and biomass are the only established air conditioning). However, some parabolic trough projects
renewable power generation technologies which offer load under construction, such as the 280 MW Solana project, and
matching capability[17]. It will become increasingly important other large projects being planned for the south-west USA,
for plants to offer this capability as both the need to reduce will incorporate energy storage.
reliance on combined cycle gas turbine (CCGT) plants and
the increasing proliferation of wind and solar power will In contrast, in Spain the feed-in tariff mechanism has
reduce overall load matching capacity in power grids. This will encouraged the use of energy storage and the technology has
ensure further financial premium to support power plants with become established in parabolic trough plants.
energy storage.
Power tower plants tend to include energy storage although
direct steam plants such as the Ivanpah plants in the USA
[17] Excepting coincident resource with load demand, curtailment of generation,
and use of separate energy storage plants such as pumped hydro or hydrogen. (370 MW) may not have energy storage.
172 Utility Scale Solar Power Plants APPENDIX A
Energy storage in CSP plants was first demonstrated transferring the storage medium from the cold tank for heating
by Solar One and Solar Two 10 MW power towers; it has from the solar field.
also been demonstrated in a number of 50 MW parabolic
trough plants, the first of which was the Andasol 1 plant Molten salt is the current storage medium of choice,
commissioned in 2008. Current parabolic trough plants with although it has drawbacks, primarily logistical, as handling
storage typically have between 6 to 8 hours energy storage is complicated. The molten salts used for energy storage,
at full plant capacity. Some power tower plants have up to and sometimes for heat transfer in CSP plants typically
15 hours storage. Such storage is sufficient to accommodate comprise nitrate salt mixtures of 60% sodium nitrate and
daily variation in solar irradiance and load demands, and 40% potassium nitrate. The use of calcium nitrate is a future
enables the capacity factor of parabolic troughs to be increased possibility. The use of molten salts as a heat transport fluid is
from 23-28% without storage to 36-41% with storage. well established in the chemicals and metals industries.
Capacity factors for power tower plants range from 24% to
67%, the lower figures relating to plants without storage, while Innovation in storage media continues, including use of
the higher figures are from the Solar Tres / Gemasolar plant, pressurised steam, specialised salts such as liquid fluoride
which has storage. salts, phase change materials, thermo-chemical storage using
ammonia, solids such as concrete or graphite, and hydrogen.
It should be noted that overall energy output is likely to be Short-term energy storage may be provided using pressurised
reduced as a consequence of incorporating a storage system, steam where the heat transfer fluid is direct steam, such as the
but the system and economic benefits of the output matching 0.5 hours storage in the 64 MW Nevada Solar 1 parabolic
demand in real time may outweigh these losses. The higher trough plant and the 1.0 hour storage provided in the 11 MW
capacity factors from storage-equipped plants are likely to PS10 and 20 MW PS20 power towers.
reflect a smaller turbine running at a higher percentage output
for more hours. 4.6.3 Supplementary Heating (Use of Natural
Gas or LPG)
4.6.2 Storage Medium (Including Molten Salts)
The freezing of salts in storage tanks and circuits must be
Heat storage using a liquid storage medium typically prevented, both overnight and during maintenance outages
comprises two tanks: a hot tank and a cold tank. During high when heat is not supplied from the solar collectors and there
solar irradiance, excess heat from the solar field is transferred is insufficient residual heat in the fluid, as the freezing points
to the storage medium which is then transferred to the hot of the salt solutions are typically above 200°C, and hence
tank. During low solar irradiance and high load demand, heat well above ambient temperatures. This is generally achieved
from the hot tank is transferred to drive the steam turbine. through the use of natural gas or liquefied petroleum gas
The storage medium is transferred to the cold tank once the (LPG) for direct heating or electrical heating.
heat has been extracted from it. The cycle is completed by
A Guide For Developers and Investors 173
4.6.4 Costs
Partly due to the necessity for heating to prevent freezing of Thermal storage has been reported to add 5% or US$50/
molten salts and to provide essential back-up power, and also MWh to the generating costs of CSP. However, given
to improve project economics, the use of natural gas or LPG variations in plant configuration, technical developments,
is generally permitted by feed-in tariff mechanisms to generate changes in the costs of CSP technologies (including the solar
up to a specified proportion of the total power generated. For field, storage, and steam power plant), and value of peak
example the tariff system in Spain allows up to 15% of power electricity, the economic case for inclusion of energy storage
to be generated from fossil fuels, and the SEGS plants in the should be assessed on a project by project basis, and in relation
USA are permitted to use natural gas to supply up to 25% of to local electricity demand patterns.
the heat energy into the turbines. Since continuity of power
generation and ability to meet transient loads significantly 4.6.5
increases the value of electricity sales, operators generally use
natural gas or LPG to generate power up to these limits and Conclusions
export the excess output that is not required for essential self-
loads. For example, generation from fossil fuels may be used Molten salts storage is technically proven but may
in the evening or in the early hours once storage has depleted. be considered to be at an early stage of commercial
This is much more cost-effective than increasing the size of demonstration.
solar field or storage, and hence significantly improves the
overall project economics. As thermal storage experience grows, it is likely that
the operational flexibility of CSP will be broadened with
In hybrid CSP plants, generation from natural gas is subsequent positive impacts on the economics.
combined with solar energy generation. Natural gas may be
used to increase the steam temperature above that provided 4.7 Cooling and Water Consumption
from the solar field, to increase the efficiency of the steam
power plant. In the case of integrated solar combined 4.7.1 Cooling Options
cycle plants, generation from natural gas is dominant and
a combined cycle gas turbine and steam turbine is used to In common with conventional steam turbine power
maximise generation efficiency from the natural gas – this is generation plants, all CSP plants used for power generation
discussed in Section 4.8. condense the steam exhausting from the steam turbine. This is
necessary to achieve an acceptable efficiency when converting
heat energy to mechanical energy, and to enable most of the
purified water to be re-circulated back to the feed system and
steam generators. As explained in Section 4.2.6, achieving a
lower condenser temperature directly increases the theoretical
efficiency limit of a turbine system.
174 Utility Scale Solar Power Plants APPENDIX A
The main cooling options for the condenser are summarised Conventional CSP plants such as parabolic troughs and
in Table 4. Although dry cooling uses less water, initial power towers typically consume large volumes of water for
capital costs are higher. Wet cooling is the most cost effective cooling to condense steam, provide make-up water for the
approach in water rich areas. Dry cooling should be considered steam cycle, and for mirror washing. Water requirements for
for water-stressed areas. different CSP technologies are given in Table 5.
G
G
High Pressure
Steam
Solar Steam
Expansion Generator
Vessel Deaerator
Feedwater Low Pressure
Preheater
As the locations with the best solar resource are typically existing thermal power plants or as part of a new hybrid
in arid areas, the supply of water can be a costly and complex installation. This is known as Integrated Solar Combined-
exercise. Alternatives to water cooling should therefore be Cycle (ISCC). All combined cycle plants falling under this
considered. Such methods are discussed in Section 4.7.1. The category use gas turbines in conjunction with steam turbines
US Department of Energy (DoE) reports that dry cooling as the established conventional thermal power plant known
requires approximately 350 litres per MWh for either type of as Combined Cycled Gas Turbine (CCGT). A typical ISCC
plant. Hybrid wet/dry systems are also being developed which plant schematic is provided in Figure 15.
balance water usage against cost and yield loss.
Two different thermodynamic cycles, a gas-turbine Brayton
All systems require water for mirror washing; this is typically cycle and a steam-turbine Rankine cycle, are combined in
in the region of 60 to 90 litres/MWh. a single system through a Heat Recovery Steam Generator
(HRSG). Fuel is combusted in the gas turbine in the
4.8 Integrated Solar Combined Cycle conventional way, and the hot exhaust gas goes through the
HRSG. Here the energy from the gas generates and superheats
It is possible to combine a solar thermal plant with a fossil steam to be used in the steam turbine cycle. In ISCC plants
fuelled thermal generation plant, either by integrating with solar heat from CSP technology is integrated either at high
[18] CSP Today article – Cost efficiency Vs water usage, and US DoE Power Electricity Generation; http://www1.eere.energy.gov/solar/pdfs/csp_
[19] US Department of Energy (2001); Concentrating Solar Power Commercial water_study.pdf (accessed 26/07/2010)
Application Study: Reducing Water Consumption of Concentrating Solar [20] Graphic courtesy Abengoa Solar.
176 Utility Scale Solar Power Plants APPENDIX A
pressure in the HRSG or directly in the low pressure casing of Due to these advantages much of the CSP capacity outside
the steam turbine. The general concept is an oversized steam the USA and Spain is made up of ISCC plants; a selection
turbine, using solar heat for steam generation and gas turbine of projects that are currently under construction or recently
waste heat for preheating and superheating steam[21]. completed is listed in Table 6.
[21] A. Fernandez-Garcia, E. Zarza, L. Valenzuela, M. Perez (December 2009): Installed capacity of CPV remains low. Only four
Parabolic-trough solar collectors and their applications;
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VMY- companies, Amonix, ENTECH, Guascor Foton and SolFocus,
4YNB5Y6-1&_user=8452154&_coverDate=09%2F30%2F2010&_
rdoc=1&_fmt=high&_orig=search&_sort=d&_docanchor=&view=c&_ have installed plants of more than 1 MW. Amonix-Guascor
searchStrId=1413265686&_rerunOrigin=google&_acct=C000050221&_
version=1&_urlVersion=0&_userid=8452154&md5=dae73bcebffd373c8a Foton, a joint venture, built the largest CPV plant at Parques
628c8a942157eb (accessed 26/07/2010); Renewable and Sustainable Energy Solares de Navarra, 7.8 MW, in 2008.
Reviews, Volume 14, Issue 7, September 2010, Pages 1695-1721
A Guide For Developers and Investors 177
Secondary Mirror
Primary Mirror
High-Efficiency
Optical Rod Multi-Junction Solar Cell
Figure 16: Illustration of Typical CPV Concentrating Mechanism Figure 17: Example of SolFocus CPV Installation
Suncore Photovoltaics, a joint venture between Emcore and Linear Fresnel reflectors differ from parabolic troughs in that
San’an Optoelectronics, is pursuing multiple projects as part the absorber is fixed in space above the slightly curved or flat
of the 280 MW solar energy plan recently announced by the Fresnel reflector. Several mirrors are fitted into the system, all
Chinese Government[22]. of which focus their energy on the central line-receiver. In some
cases a small parabolic mirror is added to the top of the receiver
4.9.2 CPV Advantages and Disadvantages to further focus the sunlight.
The principal advantages of CPV are that it requires water The options for siting and orientation of linear Fresnel plant
for cleaning purposes only, is modular and is more flexible are similar to those for parabolic trough plant. A flat land
than thermal CSP in terms of site requirements. Like parabolic area is required, and it is usual to orientate the reflectors in a
dishes, CPV systems can be installed at sites with undulating north-south direction, in order to maximise sunlight captured
terrain. throughout the day.
Commercialisation of CPV is held back by the availability Linear Fresnel reflector technology has historically operated at
of concentrator cells. A number of new companies with the lowest temperature of the available CSP technologies.
the capability for epitaxial (single-crystal) growth of multi-
junction cells have started competing with the established 4.10.1 Applications and Examples
manufacturers Emcore[23] and Spectrolab. The availability of
cells is expected to increase rapidly following the entry of these Implementation of linear Fresnel plants has been led by Areva
new suppliers. Solar (previously Ausra, USA), with the 0.36 MW Liddell plant
in Australia commissioned in 2007 and the 5 MW Kimberlina
Although some standards are available, many key areas are plant commissioned in 2008. Areva Solar (Ausra) technology
not covered in comparison with conventional PV. The most has been designed for application in utility-scale solar and solar
critical of the required standards is ‘IEC 61853, Photovoltaic hybrid plants, steam augmentation and industrial processes such
(PV) module performance testing and energy rating’, which as desalination and food processing. The overall concept was to
has been in draft for over two years. In the absence of this design the technology to be utilised at the MW scale, although
standard, nameplate energy ratings are debatable and hence
some investors have lower confidence in CPV technology than
[22] Emcore is the only vertically integrated CPV product provider at present, see:
in conventional PV. In certain situations financing risk may be http://www.euroinvestor.co.uk/news/story.aspx?id=11207570
reduced if CPV is installed alongside a mature technology such [23] yourrenewablenews.com; EMCORE enters into multi-year agreement to
supply Solar systems for utility scale power projects in US; http://www.
as flat plate crystalline PV in a hybrid installation. yourrenewablenews.com/news_item.php?newsID=4085
178 Utility Scale Solar Power Plants APPENDIX A
Areva Solar may be withdrawing from the construction of buildings for hospitals, factories and schools. These companies
large utility power plants to focus on heat plants for industrial may develop and scale up their technologies for applications
processes or smaller scale generation projects (up to 50 MW) such as process heat, desalination and power generation
where the permitting process is more straightforward. in the future.
The second major player in utility scale linear Fresnel 4.10.2 Reflector and Structure
plants is Novatec Biosol which is currently constructing the
30 MW Puerto Errado 2 (PE2) project. This follows the The mirrors utilised in linear Fresnel technology are
commissioning of their 1.4 MW Puerto Errado 1 (PE1) generally manufactured from float glass and have a thickness of
project in Murcia, Spain in 2009. around 1-2 mm. This allows the mirror to be sourced from a
number of manufacturers worldwide, unlike those of parabolic
Other companies such as UK based Heliodynamics troughs where precision bent mirrors are required. Fresnel
have developed and implemented facility scale systems for mirrors are relatively cheap to procure at around $9.8 and
applications where the solar heat is used directly, without 3 kg per square metre, which corresponds to approximately
conversion to electricity, for air conditioning and cooling in one third of the weight of a parabolic trough mirror.
Figure 18: Kimberlina solar thermal energy plant, installed by AREVA Solar (USA)
A Guide For Developers and Investors 179
4.10.4
The structures for mounting the mirror systems in linear
Fresnel systems are simpler than other CSP technologies Conclusions
allowing a higher volume of automated manufacture. As the
In the near term, it appears that linear Fresnel
mirrors do not need to support the weight of the receiver, the
technology is most likely to be implemented
overall weight of the structure can be reduced in comparison
in heat, rather than electricity generation
to parabolic trough technology. As linear Fresnel systems use
applications, where its lower cost can more than
mirrors located close together within a few metres of ground
outweigh its lower performance compared with the
level, the wind has a reduced effect on the structure, allowing
other CSP technologies.
for a lighter structure to be used. When not in use, the mirrors
can turn upside down for further protection from the wind,
Its use for utility scale power generation is likely to
sand storms or hail.
depend on:
4.10.3 Receiver and Heat Transfer
• The commercial success of the Puerto Errado
2 (PE2) project and a small number of other
Linear Fresnel reflector plants use steam as the heat proposed projects.
transfer fluid.
• Commercial success of small scale applications,
economies of serial production, and transfer of
The receiver of the Areva Solar / Ausra system is made from
technology and production from small scale to
steel tubes with heat absorption coating. Water is vaporised larger scale units.
within the receiver tube. The steam is piped directly to the
required application, whether it is for electricity generation, • Technology improvements and cost reductions
steam augmentation or industrial processes. This allows for achieved for other CSP technologies.
the elimination of expensive receivers, performance reducing • Continuing substantial economic support for
heat exchangers, and the costly transfer fluids that are CSP power generation.
required by parabolic trough technology. The simplification
of components required by the technology means they can be
sourced from a variety of manufacturers.
5. SITE SELECTION
As discussed in Section 3, CSP technology can only capture Fresnel reflectors. An average slope of 3% or less is preferable
the direct normal irradiance (DNI) component of the solar for parabolic troughs and linear Fresnel reflectors, and 5%
resource. This is a primary driver in site selection. Prime CSP for power towers . Parabolic dishes and CPV, due to their
locations typically require DNI exceeding 2,200 kWh/m2 modularity, can be installed on steeper slopes. CSP may
per year. The site screening threshold should be determined require less land area per MWh than PV, depending on the
taking into account the envisaged technology, plant design, technologies employed.
and region-specific factors such as the cost of grid connection,
power prices, and the additional economic support which will A far greater quantity of water is required for parabolic
be available. When considering a technology for a specific site, troughs, power towers and linear Fresnel reflectors than for
the following factors should be considered: solar PV plant, due to the need for turbine condenser cooling.
Depending on whether dry or wet cooling is employed, a
• Nature and scale of energy generation plant, for large quantity of water is required and therefore a local water
example utility scale power generation or process source at an environmentally acceptable and economic price is
steam augmentation. essential.
• Land availability providing sufficient direct
irradiation and area which is within slope limits 6. ENERGY YIELD PREDICTION
for the candidate technologies. If land availability
is limited this may indicate a preference for a high 6.1 Site Conditions and Data Measurements
efficiency technology to capture, convert and
generate the required amount of energy. As for PV plants, predicting a plant’s energy yield begins
• Economic support available and affordability. with quantifying the available solar resource. Accuracy is
Minimum generation cost ($/MWh) will generally crucial in this area, due to the CSP requirement for clear skies
determine the final choice of technology, irrespective and high DNI.
of energy capture and conversion efficiencies.
• Load matching requirements, and value of peak Depending on the location, DNI constitutes between 50%
energy, will determine the choice of energy storage and 90% of Global Horizontal Irradiance (GHI), and varies
system and its capacity. considerably in time and location. Annual DNI can vary
by up to 30% from one year to the next, so irradiance data
• Water availability is likely to determine the choice of
should be measured over as long a period as possible in order
condensing system.
to increase the confidence in the long-term prediction. DNI
Regions with high levels of airborne dust, haze or smog data availability varies greatly by location, and local data may
are not suitable. This often rules out sites near large cities, not be available. The availability of local data will determine
particularly in arid developing nations. Higher altitudes lead how much site data is required, but one year is considered a
to clearer skies and higher DNI, so accessible high altitude minimum. Measure-correlate-predict (MCP) methods can
locations can be considered. then be used to compare satellite and local meteorological data
against site measurements. Use of satellite data on its own can
CSP also generally requires land with minimal gradient, result in over-estimation of energy yield due to the effects of
although parabolic dishes and power tower technology are near-ground haze which may not be measured by satellite.
not as sensitive to slope as are parabolic troughs or linear
A Guide For Developers and Investors 181
Ambient temperature and relative humidity should be Several computer based energy yield and optimisation
measured. The site wind speed is also important, as the software tools have been developed to model the performance
plant may need to shut down when wind speeds exceed a of a variety of photovoltaic systems (including CPV) and
certain limit. concentrating solar power (CSP) systems including parabolic
trough, power tower, and dish-Stirling systems.
The energy yield prediction must therefore be based on
measurement and analysis of the solar resource and other Two categories of models can be found: physical models
conditions at the site combined with a detailed analysis of the based on heat transfer and thermodynamic principles, and
proposed technology, including thermodynamic modelling of empirical models based on data obtained from performance
the plant design. analysis of installed systems. Although physical models are
generally more flexible than empirical models, for which
6.2 Technology Characteristics only a limited range of system components can be included,
they also add more uncertainty to performance predictions
Each CSP technology has its own energy yield than empirical models. Models used for each technology are
characteristics. A difference when conducting energy yield described below.
analysis for CSP, as opposed to PV, is the added dimension
of the thermal conversion technology (with the exception of • CPV – models typically use the plant layout,
parabolic dishes). CSP requires that thermal inertia and the technical specifications for the PV modules and
inverters, tracking parameters and weather data for
efficiency of the Rankine cycle steam turbine is accounted
the site.
for. This, in turn, depends upon the form of cooling system
employed. However, unlike PV, CSP does not lose efficiency • Parabolic Trough – models typically use the plant
due to degradation over time, or losses in inverters. Higher design and layout, optical and technical parameters
ambient temperatures generally increase efficiency, providing for the collectors, receivers, power block, heat transfer
fluid, thermal storage system (if any) and cooling
that the condenser performance is not adversely affected,
system, as well as control parameters and weather
data for the site.
Modelling of candidate technology systems for a particular
project therefore needs careful consideration. Whereas • Power Tower – models typically use the heliostat
empirical guides based on previous project experience can be configuration, technical parameters for the tower,
receivers, power block, thermal storage system
used for preliminary scoping, proper definition of the project
(if any) and cooling system, as well as control
and performance modelling will require detailed information parameters and weather data for the site.
and analysis of:
• Dish-Stirling – models typically use the solar field
• Site conditions and resource data. layout, technical parameters for the collectors,
receivers and Stirling engine, and weather data for
• Site layout for each technology. the site.
7.1 Overview Load matching options for CSP plants (excluding CPV) are
considered in Table 7, indicating the relative size of the steam
For commercial projects completed to date, few major turbine and heat storage for a given size of solar field.
project problems have been reported in the public domain.
It can be considered that the base technology, at least for At present the cost of the solar field and storage are more
parabolic trough plants is proven. significant than variations in size of the steam power plant,
and other plant such as natural gas fired plant provides peak
Engineer Procure Construct (EPC), Design Build Operate power. Hence most current plants follow an intermediate load
(DBO), or Build Own Operate (BOO) contracting strategies matching regime.
are all established in the CSP sector. However, long term
operating experience is confined to Florida Power Limited’s 7.2.3 Solar Multiple
SEGS plants in the USA.
The ratio of installed capacity of the steam turbine to the
There are very large cost uncertainties and some planned size of solar field varies with the project design, particularly
projects are on hold or being redeveloped as PV solar depending on the integration of storage.
generation projects due to relative costs, economic support,
and uncertainties. The solar multiple is defined as:
As with any industry that depends on government policy to Actual area of Solar Field
set subsidies, there is potential for a boom and bust cycle in Area required to operate turbine at design output at time of
the CSP industry, unless consistent support is available with maximum solar irradiance
appropriate long-term tapered reductions.
Plants without storage have an optimal solar multiple
7.2 Design of roughly 1.1 to about 1.5 (up to 2.0 for linear Fresnel
Reflector plants), depending primarily on the amount of
Various technology-specific design characteristics have been irradiation the plant receives and its variation through the day.
considered in Section 4. Some of the main design parameters Plants with large storage capacities may have solar multiples
for CSP projects in general are considered here. of up to 3 to 5.
Whilst many current CSP projects are of 50 MW capacities, Currently the capacity factors of parabolic trough plants
many reports indicate a capacity of 200 MW would be more generally fall in the range of 23-28% without storage and
cost-effective. A rule of thumb indicator for the ratio of 36-41% with storage. Capacity factors for power tower
installed capacity to land area is around 0.3 MW/ha. plants range from 24% to 67% with the lower end of the
range applying to plants without storage and the higher
figure applying to the Solar Tres / Gemasolar plant which
includes storage.
A Guide For Developers and Investors 183
Improvements in effectiveness of storage should enable • High grade heat enables more efficient conversion
capacity factors to be increased in the future. However, since of heat energy to mechanical energy (and thence to
the capacity of the plant is not directly determined by the size electrical energy) in the steam turbine.
of the solar field, an increase in capacity factor does not usually 7.3 Development
give a proportionate reduction in generating costs.
Development requirements will be highly specific to the
7.2.5 Grade of Heat region and regulatory regime for planning and permitting in
the region, including specific regulations applying to land use,
The grade of heat refers to the temperature and pressure of
water use, and power / renewable energy projects.
the heat transfer fluid delivered from the receiver to the heat
storage or steam power plant after transfer losses. The grade of The size of the project may be a major factor in determining
heat is important since: the regulatory regime which applies and hence the time and
likelihood of progressing projects through the planning and
• High grade heat may enable more effective and lower
cost heat storage; and permitting process.
The CSP sector is dominated by several solar energy • Parabolic trough mirrors – Flabeg, Rioglass;
technology companies and affiliates who are experienced in:
• Parabolic trough receiver tubes – Solel (now owned
by Siemens), Schott;
• CSP technologies;
• Steam turbines for CSP projects – the market
• CSP project development; and
currently appears to be dominated by Siemens.
• Undertaking or managing CSP projects on an
EPC basis. 7.5 Uncertainties and Risks
Examples of solar energy technology / EPC companies Large uncertainties remain for the future of CSP both in
for parabolic trough developments active in Spain and USA the short and long term. Some of the major uncertainties are
include: summarised in this section.
• Sufficient proportion of the value of the plant which market volume will be required to attract new supply
can be expected to be subject to learning rate effects; companies into the market.
and
• The percentage learning rate effect which is predicted 7.5.4 Short Term Cost Uncertainties
must be realistic.
Other cost uncertainties include:
7.5.3 Supply Chain Competition
• Commodity price variations; and
Cost projections for CSP often assume increasing
• Project development and construction cost
competition in the supply chain. However, there is currently
uncertainties (typically greater than ±30%) remaining
low diversity of supply of specialist components, and sustained
once preliminary project design is defined.
8. Conclusions
For all CSP technologies in the foreseeable future, substantial economic support will be required for project economic viability, through
a support mechanism specifically designed to support CSP projects. Hence the first requirement for bankability is the availability of such
support, including commitment to provide revenue support at a defined level and for the period necessary to achieve appropriate project
returns.
For utility scale power production, parabolic trough is considered to be the most bankable CSP thermal technology, due to its
operational track record, which gives it a moderate technology risk (low relative to other CSP technologies). However, actual project
construction costs are currently very high and show a very large range. The actual completion costs of projects currently under
construction will be key to providing the basis for acceptable financial risk.
Power tower technology is considered to be the second most bankable CSP thermal technology based on operational experience of 44
MW projects over the last 1 to 3 years, and current project construction experience, which gives it a medium technology risk. As with
parabolic trough, actual project construction costs are currently very high and show a very large range. The actual completion costs of
projects currently under construction will be key to providing the basis for future financing.
Parabolic dish technology has been demonstrated at unit level for dishes of 25 kW capacity. However, the construction of a large array of
dishes to form a commercial scale project has yet to be completed. Bankability of this technology in the short term will be reliant on the
success of one technology and project developer partnership, namely Tessera / Stirling Energy Systems (SES), in taking its plans forward.
There is insufficient actual construction experience to confirm costs at this stage.
Linear Fresnel technology and project development is currently being aimed at smaller-scale building-integrated, process steam, or
desalination applications, although there are some technical developments currently being undertaken which could make it more
competitive with parabolic trough for utility-scale generation. Until this is demonstrated pre-commercially, linear Fresnel technology is
not considered bankable for utility-scale power generation. There is insufficient actual construction experience to confirm costs of utility-
scale projects at this stage.
Concentrating PV is currently perceived as a relatively high risk investment compared to other solar technologies. Reasons for this
include a lack of standardisation and certification, lack of volume production, and lack of an established supply chain with demonstrated
capability. Financing risk may be reduced if CPV is installed alongside a mature technology such as flat plate crystalline PV in a hybrid
installation.
186 Utility Scale Solar Power Plants APPENDIX B
Appendix B
AC Benchmarks
Appendix C
of goods, materials, equipment, salaries and other
items or any changes in general economic conditions.
EPC Contract Model Heads
• Ground Studies – means the ground studies
of Terms commissioned by the Contractor at its own expense
prior to the date of this Agreement in respect of the
Definitions and Interpretation Site.
• The Contract is for the Contractor to carry out the authorisations for connection to the electricity
turnkey construction, commissioning and delivery of distribution network. These permissions and
the Plant to the Client. authorisations will entitle the Contractor to carry out the
construction and installation of the plant on the Site.
• The Contractor shall design and construct the
Plant in accordance with the EPC contract, relevant Inspections
international and national standards, industry best
practice, OEM instructions, technical conditions set
• The Client shall have the right to have the construction
out in the Grid Connection Agreement and Power
works monitored by a representative of their own choice
Purchase Agreement (or similar), and applicable
and on their own terms. The representative shall be
legislation.
independent and not affiliated to the Contractor in any
• The subject of this Agreement involves engineering, way.
design, procurement, construction, assembly, testing,
start-up, commissioning of the Project by the Alterations to Client Scope of Work
Contractor. The Project includes the construction
and operation of the transformer station, connecting • All required alterations / deviations to the agreed scope
the power plant to the feeding point, to the of work by the Contractor costing over €/£/$[xxx]
transformer station, and connecting the power plant shall require the approval of the Client or the Client’s
at the transformer station to the mean voltage system Representative.
operated by [grid operator]. This includes switchgear
(interface) as specified in Schedule of Scope of Tasks and Obligations of the Contractor
Work, on a turnkey basis with fixed term fixed price
including fixed timeline and fixed performance. General Scope of Performance
Tasks and Obligations of the Client • The Client shall be provided with all appropriate site
drawings/documentation before construction works
General begin on site.
Permits
• The Contractor shall also be responsible for the safe Prior to the date on which the Contractor shall commence
and proper operation of the Works in conjunction Works on the Site, the Client shall obtain all necessary
with the works and services to be provided by the planning permissions, building permits and authorisations for
Client for the purpose of completing the Plant.
connection to the electricity distribution network. This will
• The Contractor shall provide a guarantee for a be done in order to entitle the Contractor to carry out the
duration of [x] years that the plant will contain all construction and installation of the plant on the Site.
characteristics detailed within the EPC contract.
• For the duration of the site works, the Contractor The Contractor is responsible for designing,
shall be responsible for the provision of security/ constructing and the installation of the plant in accordance
surveillance of the Site and all goods, materials, with all necessary planning permissions, building permits
equipment and other items located on the Site. and authorisations for connection to the electricity
distribution network that are in place and have been
• The contractor shall conduct studies to satisfy
themselves of the ground conditions on site. obtained by the Client.
Delay Documentation
The Contractor shall be required to complete the The Contractor shall provide the Client with all appropriate
installation and construction of the plant in accordance site drawings/documentation before construction works begin
with the timelines set out within the Project Schedule. If the on the site.
Contractor is unable to complete the works within the agreed
timescales, then the Contractor will be required to pay the The Contractor shall be required to provide, at a minimum,
Client compensation. the following documentation after the final acceptance tests
have been completed:
The contractor shall be required to pay the client €/£/$xxxx
per day of delay past the agreed completion date within the • Technical data sheets for all components / materials.
Project Schedule.
• ‘Flash Lists’ for the modules.
The amount of compensation that the Contractor will be • Operation and maintenance manuals.
obliged to pay to the Client as a result of failure to meet the
• Procedures for managing faults/malfunctions/issues.
timelines set out within the Project Schedule shall be limited
to a maximum of [xx]% of the total value of the Contract. • Characteristics of the components.
The Contractor shall only be allowed to appoint sub- • Acceptance testing date
contractors only after approval from the Client or the
• Completion/Take-Over date
Client’s Representative.
Payment Schedule
The appointment of sub-contractors does not relieve the
Contractor of the responsibility for the complete, accurate and The Contract Price shall be a fixed cost for the complete delivery
timely execution of the Contract. of the plant, and shall be calculated using the following formula:
• electrical substation
100% mechanical and electrical installation of the PV modules,
M3 40 70
and inverters.
Payable after commissioning and availability of at least 90%
M4 10 80
of the modules
Acceptance of the Provisional Acceptance Tests and issuance of the
M5 10 90
Provisional Acceptance Certificate. Completed scope of work and snag list.
M6 First receipt of income from generation (of power) on the site. 10 100
192 Utility Scale Solar Power Plants APPENDIX C
The Contractor shall not be entitled to payment for any The minimum documentation to be provided with the test
milestone until all requirements of the milestone are met. The report is as follows:
Client shall retain 10% of the contract until first receipt of
income from generation (of power) on the site. • Technical data sheets for components / materials.
Pre-final testing at a minimum shall consist of the following: • Both general and detailed final drawings.
• Voc and Isc for each string greater than [93]% of The Final Acceptance Certificate shall have a guaranteed
nominal – with I>600W/m2.
level of performance of:
• Submission of:
• [xx]% PR (to grid export meter) with temperature
• Construction Progress Report. and irradiation correction.
• [xx]% efficiency (temperature and irradiation • [xx]% Performance Ratio (to grid export meter) with
corrected) for the DC generation. temperature and irradiation correction
• [xx]% conversion efficiency.
Make Good Periods
Or
If the Plant does not meet the guaranteed levels of
• [xx]% Performance Ratio (to grid export meter) with performance set out in the Pre-Final, Provisional Acceptance
temperature and irradiation correction or Final Acceptance tests, the Contractor has [x] days to make
good the deficiencies. Retesting of the Plant will be at the
A Guide For Developers and Investors 193
Insurance
Contractor’s expense, including the costs for any external/ The Contractor shall have, at a minimum, the following
third party testing experts. If the Plant does not achieve the insurance cover for the duration of the site works:
guaranteed levels of performance even after repair and retests,
then the Contractor shall either make compensation to the • Erection Insurance
Client or the Client will be entitled to withdraw from the • Public Liability Insurance
agreement. [Note: This clause needs to be carefully worded
and agreed between the Client and Contractor. Even small • All Risks Insurance
deficiencies in performance can add up to large reductions in
The contractor shall also, at their own expense, be required
revenue over the project lifetime.]
to hold any other legally required insurance for the location of
the Plant.
Warranty
Duration and Termination of Agreement
The Contractor warrants that their work will remain defect free
for a period of [x] years following final acceptance of the Plant.
Sections required for covering:
The Contractor will transfer all equipment warranties and • Commencement of the agreement
guarantees to the Client following Final Acceptance.
• Standard duration of the agreement
Legal, Governing Law and Jurisdiction • Extension of the agreement (at the option of the
Principal)
The contract shall have sections covering:
• Termination of the agreement due to:
• Governing law and court of jurisdiction of the
agreement. The governing law is normally the law
• Financial arrears.
of the country in which the Plant is located. • Insolvency.
• A legal succession or a transfer of rights condition • “Good cause” relating to statutory law, failure
is required so that the Principal reserves the right to to comply with the requirements contained in
assign the EPC contract to a third party. this agreement, or environmental, planning or
legal breaches.
• Non-disclosure agreement. This agreement between
the Contractor and the developer will outline what • Consequences of the termination.
information is to be considered confidential and what
may be disclosed to third parties.
Arbitration
• Location of the arbitration hearing and under 6. Construction Schedule [this should include all the
what law. timelines for the construction tasks for the plant]
Sections required for detailing how communications shall 8. Energy Yield Study (Yield Assessment Report)
be conducted between the Client and Contractor, including
9. Performance Ratio
language and acceptable channels of communication.
10. Performance Bond
Contact details for both the Client and the Contractor’s
11. Warranty Bond
representatives shall be included within this section.
12. Parts of the Project to which the Structure
Applicable Law Warranty applies
1. Site Plans [this should include all layout plans for 20. Form of the Site Hand-over Protocol
the proposed plant]
2. Scope of Work
A Guide For Developers and Investors 195
Appendix D
O&M Contract Model Heads • Plant – the power plant to which this contract refers.
Details of the power plant should be contained in an
of Terms annex to the contract, including the total nominal
power output (MWp), details of the equipment and
Definitions and Interpretation systems that form the plant.
• Guaranteed Response Time – is the time taken • The Contractor shall maintain the Plant in
to resolve faults, and is dependent on the nature accordance with the O&M contract, O&M
of the fault and its impact on the total output Handbook, OEM instructions, technical conditions
of the project. set out in the Grid Connection Agreement and
Power Purchase Agreement (or similar), and
• Maintenance – the execution of all operations, applicable legislation.
required to maintain the functioning of the Plant in
accordance with the agreed maintenance schedule. • The Contractor will maintain the plant as defined in
the definition and annex [x]. This will include (but
• Measured Average Availability – the actual average not necessarily be limited to):
generation availability of the Project. This is
measured and calculated as defined in the contract. • DC Generation components (modules, cables,
mounting structures)
• Normal Performance – defined as performance
within [5%] of the Performance Ratio as defined in • AC Generation components (inverters, cables,
Appendix [x]. transformers, MV switchgear)
• The Contractor shall maximise the output of the include the assessment of the condition of the Plant
Plant in both the short and long term by monitoring including (but not limited to) the following:
and rectifying disruptions, and enable the Plant to
adhere to the Minimum Availability parameter. • Condition of the generation equipment and
supporting structures.
• The Contractor is to provide periodic reports and
advise on technical issues during operation. • Functional check and condition of the
security systems.
• The Contractor shall maintain the Plant to ensure
that the degradation of the Performance Ratio is • Visible damage and defects including functional
not more than [1%] per year [as defined in the check of safety equipment.
Acceptance Test]
• Inspection of the medium voltage components
• The Contractor shall operate and maintain the Plant (up to the grid metering point).
such that warranties under the EPC contract are not
• The Contractor shall, as a minimum, maintain Plant
restricted and remain enforceable.
components in order to comply with their warranty
conditions.
Tasks and Obligations of the Contractor
• The Contractor shall remove snow and other
Operation of the Project accretions (dust and dirt) from the modules as
required.
• The Contractor shall:
• The Contractor shall keep the Site free from
• Operate the Plant on behalf of the Principal, undesirable growth of plants and shrubs, to achieve
and ensure uninterrupted operation (wherever the best possible energy yields. Green areas shall be
possible) and optimal usage of the Plant, subject mowed and waste materials removed. Dust shall be
to the weather conditions. kept to a minimum to reduce soiling. Vehicles and
machines will be used such that they do not damage
• Advise the Principal on all significant issues plant components.
relating to the operation of the Project.
• The Principal shall pay the cost of electricity and
• Make sure that the Principal has unlimited right water required for maintenance purposes (within the
to inspect the Plant and perform works on it. boundaries of the Plant)—and not expressly covered
by another clause within this agreement.
Maintenance
• The Contractor shall maintain the Plant so that the
• The Contractor shall maintain the Plant in degradation of the Performance Ratio is not more
accordance with the Maintenance Schedule than [1%] per year, and shall inform the Principal if
(contained as an annex to this agreement). The this limit is likely to be exceeded.
Maintenance Schedule will show the minimum
maintenance required and minimum frequency for • The Contractor shall perform specialised cleaning
each item per annum. as required, on a site-specific basis to avoid seasonal
dust and dirt accumulation.
• The Contractor shall carry out a minimum of
[two] maintenance inspections per year. This shall • All maintenance work that may affect the energy
generation of the Plant, will be carried out, as far as
reasonably possible, during low-irradiation periods.
A Guide For Developers and Investors 197
Monitoring of the operation / analysis of the Failure Messages and Reaction Time
data / switching off of the Plant
• The Contractor shall monitor the Plant in accordance
• The Contractor shall: with this agreement. In the event of a malfunction
occurring in one of the functions monitored, a failure
• Monitor the operation of the plant and feed- message will be sent to the Contractor.
in capacity, without interruption, 365 days a
year, 24 hours a day. This will be done using • The Contractor is obliged to take all reasonable
remote data monitoring systems which will be measures necessary to rectify any malfunction
maintained and updated by the Contractor. notified by a failure message, or detected through
an inspection, as soon as possible to restore the
• Set right faults that can be rectified remotely. functionality of the Plant.
This shall be done as soon as possible and within
[1] day. All other faults shall be rectified within • The Contractor is obliged to acknowledge and react
the Guaranteed Response Time. to failure messages, depending on the severity of the
failure or malfunction, as shown below:
• Check and assess the data collected on a daily
basis, and compare the readings with the • The Contractor is obliged to acknowledge the
assumed targets, by referring to the following failure message within [1] calendar day.
functions and operations:
• If the failure has a low impact on the yield
• DC Component Availability. loss of the Plant [<5%] per calendar day, then
the Contractor will commence and conclude
• AC Component Availability. measures to resolve the problem within [14]
calendar days of having received the failure
• Irradiation. message.
• Generation (kWh). • For other failures in the Plant, the Contractor
will commence measures to remedy the
• Network availability.
malfunction within [24] hours after having
• Function of the Security System. received the failure message.
Repair
• The Contractor is responsible for repairs in order to • The Contractor gives an assurance to keep parts
bring the Plant to fault-free operation. available to PV plant experts or for inspection by
insurance advisors/estimators whenever an insurance
• The Contractor is to correct defects that become claim is made.
known through technical monitoring, and visual and
functional checks. Documentation and Reporting Obligations
• The Contractor is to replace damaged components,
or parts that are causing disruption to the operation. The Contractor is to provide the following reports:
• In cases where material and spare parts are required • Monthly reports describing the availability, day
by Contractor, the amount will be paid by Principal by day production, performance ratio, irradiation
without the need for prior approval, provided the measurements, faults and maintenance or repair
value does not exceed [USD 1500] within a [six] activities that are conducted on the Plant.
month period.
• Quarterly reports describing material events
• Approval in writing is to be sought from the Principal (that occurred or are expected to occur) such as
for repairs (including material and spare parts) that maintenance tasks and repairs, a schedule for
are expected to exceed [USD 1500] within a [six] repairs and maintenance, a stock list of spare parts
month period. and consumables, and a description of availability,
irradiation, Performance Ratio and downtime.
• The Principal shall approve the cost of repairs within
[3] business days. • Semi-annual inspection reports documenting the
visual and function checks with photographic
• The Contractor is to keep a spares stock of
evidence of observed issues.
particularly susceptible components.
• Annual reports for each year of the contract
• Spare parts, tools, measurement and test equipment
period giving:
are the responsibility of the Contractor.
• A summary of repairs and maintenance
Materials and Lubricants tasks completed.
• The Contractor will provide under this agreement • Parts and consumables used for repairs and
all consumables, small items, and lubricants (with maintenance, and total cost thereof.
a value less than [x]) required for the maintenance,
inspection and repairs without charge. • A summary of performance and operation:
monthly availability, Performance Ratio
• The Contractor guarantees that all parts to be and production.
delivered and installed within the scope of this
agreement will correspond to the parts being • A summary of plane of array irradiation
replaced, with respect to functionality and durability. measured on Site.
• The Contractor is responsible for the removal • The annual Performance Ratio and a
and disposal of old parts, lubricants, packaging as comparison of production with irradiation-
required by the relevant laws and regulations. corrected target values given by the energy
yield prediction.
• Reports on significant disruptions, damage documents (for instance, O&M handbook, relevant
or defects. permits, consents and licences).
• The Contractor shall check invoices that have been [Depending on the requirements of the Principal, the
received by the Principal from third parties—in the Contractor may be made responsible for the security
course of the operation, maintenance and repair of
arrangements for the Plant. If so, the requirements for
the Plant—for plausibility and accuracy.
monitoring and maintaining the security system—and
Defect and Insurance Claims the costs of providing these services—should be contained
in the agreement]
• The Contractor is required to inform the Principal
about possible and actual defects in the Plant • Security to be provided and monitored 365 days/year
for which the Principal may have Warranty for 24 hours/day.
claims under the EPC Contract or under other
agreements that the Principal may enter into in Warranty and Liability
respect of the Plant as soon as it gains knowledge of
such defects. Warranty Period
• The Contractor’s own works, equipment, spare
parts and materials provided should be covered by • The Contractor warrants all respective maintenance
warranty for a period of [2] years. work undertaken in accordance with this agreement.
Guarantee is also given for operations to remedy any
• The Contractor will provide full support to the agreed failures for a period of cover of [three] years
Principal in communication with insurance from the time of the work.
companies for matters related to the Plant.
Replacement of Parts
In the event that the Contractor cannot remedy the
defects or fails to redress or make good any defect as soon • The Contractor provides a three-year warranty for
as possible, the O&M contractor shall pay the Principal any replacement of parts and components provided
and installed under this agreement.
damages to compensate the loss incurred (including
loss of profit). Guaranteed Availability
Carrying out the operational management in • The Contractor is to ensure that the Measured
compliance with the applicable laws and using Average Availability in each generation period equals
professional personnel or exceeds Guaranteed Average Availability figures in
the following manner:
• The Contractor should be required to ensure that
• Guaranteed Average Availability for each
the Plant is operated and maintained in compliance
[twelve] month Generation Period of this
with the applicable laws and contract supporting
agreement will be in excess of: 97%
200 Utility Scale Solar Power Plants APPENDIX D
• Measured Average Availability = Grid-connected Clauses are required to specify the entitlement of the
available hours / possible available hours * 100%. Contractor to use competent third parties to fulfil obligations
arising from this O&M Contract. The Contractor should be
• Grid-connected available hours are the number fully responsible for the action of sub-contractors.
of hours that the inverters are connected to the
grid and available for export of power. In effect, it
also means the number of daylight hours (which Often, the Principal has the right to refuse the use of
are to be defined) during which the Plant is able third parties. However, this should be expressly covered in
to convert irradiation into energy. Importantly, this agreement.
availability outside daylight generation hours
is not to count towards the grid connected Obligations of the Principal
availability hours.
parts that are subject to warranty claims, or The contract should have sections covering:
minor repairs of value less than [x].
• Governing law and court of jurisdiction of the
• Services and ancillary costs that are approved in agreement. The governing law is normally the law of
advance by the Principal and are required for the country in which the Plant is located.
maintenance and repair. This will not include
services that are due to parts that are subject to • A legal succession or a transfer of rights condition is
warranty claims, or minor repairs of value less required to reserve the Principal’s right to assign the
than [x] that are to be corrected during semi- O&M contract to a third party.
annual inspections.
• Non-disclosure agreement. This agreement between
• Agreement is to be made on the invoicing dates/ the Contractor and the Developer will outline what
frequency and payment terms for the Principal. is to be considered confidential and what information
may be disclosed to third parties.
• Agreement is to be made on the indexation of the
remuneration over the term of the agreement. Often • Contractual language. Defining the language that
this is linked to a consumer price index, power the official legal contract and agreement would be
purchase or tariff increase or inflation index. drafted in.
Exclusions and work outside the agreement Agreement on the language for delivery of reports,
documentation and accounts, and the version for the
• Exclusions to the availability guarantee and agreed legal copy.
remuneration, including maintenance or repair work
that are caused by:
Insurance
• “Acts of God”
The contract should also have a section outlining the
• Extreme weather effects
insurance responsibilities of the Contractor for the operations
• Improper influence of the Principal and maintenance activities. This insurance should cover
damage to the plant as well as provide cover for employees
• Third parties not attributable to the Contractor
on the Plant conducting the maintenance.
• Work covering these exclusions will be decided upon It is also normal for the Contractor to arrange and pay
between the Principal and Contractor at the rates for insurance for the full site. The agreement should contain full
additional work agreed in this agreement.
details and remuneration requirements, while the annex to the
• Work outside the agreement including repair work contract should include the insurance documentation.
undertaken within the scope of the insurance policy.
In this case, the Contractor will carry out work as
agreed with the insurance company and Principal.
Costs will be agreed within the insurance terms and
additional costs agreed with the Principal.
202 Utility Scale Solar Power Plants APPENDIX D
Acknowledgements: Alexios Pantelias, Hemant Mandal, Patrick Avato, Matt Willis, Anjali Garg, Naomi Bruck
Contact Information
2011