INTRODUCTION

The objective of this guidebook is to communicate how to successfully finance,

develop, construct and operate a utility scale Pv power plant. hroughout 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 35 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. The layout design
recommendations assume the plant is located in the northern hemisphere, although the
concepts may be extended to southern hemisphere locations.

SOLAR PV TECHNOLOGY
This section discusses Pv applications, module technologies, mounting systems,
inverters, monitoring and forecasting techniques. It provides an overview of current
commercially available technologies used in utility scale solar Pv projects. The
purpose is to provide a framework of understanding for developers and investors
before they commit to a specific technology.
Applications of Solar PV
There are four primary applications for Pv power systems:

Off-grid domestic - Providing electricity to households and villages that are not

connected to the utility electricity network (the grid).

Off-grid non-domestic - Providing electricity for a wide range of applications such
as
telecommunication, water pumping and navigational aids.

 Grid-connected distributed PV Providing electricity to a specific grid-connected

customer.

Grid-connected centralised PV Providing centralised power generation for the

supply of bulk power into the grid.

The focus of this is on grid-connected centralized PV power plants. However much of
the guidebook is also relevant to other applications.

Overview of Ground Mounted

PV Power Plant
It gives an overview of a megawatt scale grid-connected solar Pv power plant. The
main components include:
Solar PV modules - These convert solar radiation directly into electricity through
the photovoltaic effect in a silent and clean process that requires no moving parts. The
photovoltaic effect is a semiconductor effect whereby solar radiation falling onto the
semiconductor Pv cells generates electron movement. The output from a solar Pv cell
is direct current (dC) electricity. A Pv power plant contains many cells connected
together in modules and many modules connected together in strings to produce the
required dC power output.

Module mounting (or tracking) systems These allow Pv modules to be

securely attached to the ground at a fixed tilt angle, or on sun-tracking frames.

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
parallel strings are connected to the inverters.

Step-up transformers - The output from the inverters generally requires a

further step-up in voltage to reach the AC grid voltage level. The step- up transformer
takes the output from the inverters to the required grid voltage (for example 25 kv, 33
kv, 38 kv, 110 kv depending on the grid connection point and requirements).

The grid connection interface - This is where the electricity is exported into the

grid network. The substation will also have the required grid interface switchgear such
as circuit breakers and disconnects for protection and isolation of the Pv power plant
as well as generation and supply metering equipment. The substation and metering
point are often externalto the Pv power plant boundary and are typically located on the
network operators property.

SITE SELECTION
Selecting a suitable site is a crucial component of developing a viable solar Pv project.
There are no clear cut rules for site selection. viable projects have been developed in
locations that may seem unlikely on first look, such as on high gradient mountain
slopes, within wind farms and on waste disposal sites. In general, the process of site
selection must consider the constraints and the impact they will have on the cost of the
electricity generated. The main constraints that need to be assessed include:
Solar resource.
Local climate.
Available area.
Land use.
Topography.
Geotechnical.
Geopolitical.
Accessibility.
Grid connection.
Module soiling.
Water availability.
Financial incentives.
Showstoppers for developing a utility scale Pv power plant in a specific location
may include constraints due to a low solar resource, low grid capacity or insufficient
area. However, a low solar resource could be offset by high local financial incentives,
making a project viable. A similar balancing act applies to the other constraints. These
are discussed further below.

UTILITY SCALE SOLAR POWER PLANTS

Site Selection Constraints
Solar Resource
A high average annual GHI is the most basic consideration for developing a solar Pv
project. The higher the resource, the greater the energy yield per kWp installed. When
assessing the GHI at a site, care must be taken to minimise any shading that will
reduce the irradiation actually received by the modules. Shading could be due to
mountains or buildings on the far horizon, or mutual shading between rows of
modules, or shading near the location due to trees, buildings or overhead cabling.
Avoiding shading is critical as even small areas of shade may significantly impair the
output of a module or string of modules. The loss in output could be more than
predicted by simply assessing the proportion of the modules that are shaded. When
assessing shading, it must be remembered that the path the sun takes through the sky
changes with the seasons. An obstacle that provides significant shading at mid-day in
december may not provide any shading at all at mid-day in June. The shading should
be assessed using the full sun-path diagram for the location.

Area
The area required per kWp of installed power varies with the technology chosen. The
distance between rows of modules (the pitch) required to avoid significant inter-row
shading varies with the site latitude. Sites should be chosen with sufficient area to
allow the required power to be installed without having to reduce the pitch to levels
that cause unacceptable yield loss.depending on the site location (latitude) and the
type of Pv module selected (efficiency), a well-designed Pv power plant with a
capacity of 1 MWp developed in India is estimated to require between one and two
hectares (10,000 to 20,000 m2) of land. A plant using lower efficiency CdTe thin film
modules may require approximately 40 to 50% more space than a plant using polycrystalline modules. Figure 15 shows a large ground mounted plant.

Climate
In addition to a good solar resource, the local climate should not suffer from extremes
of weather that will increase the risk of damage or downtime. Weather events that may
need consideration include:

Flooding - May increase the risk of erosion of support structure and foundations,

depending on geo-technical conditions.

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 should be
avoided. Fixed systems do not shut down at high wind speeds, but tracking systems
must shut down in safe mode when speeds of 16-20 m/s are exceeded.

Snow - Snow settling on modules can significantly reduce annual energy yield if

mitigating measures are not taken. If the site is prone to snow, then one has to consider
factors such as extra burden on the mounting structures, the loss in energy production
and the additional cost of higher specification

Topography
modules or support structures. The cost of removing the snow needs to be weighed
against the loss in production and the likelihood of further snowfall. The effects of
snow can be mitigated by a design with a high tilt angle and frameless modules. The
design should also ensure that the bottom edge of the module is fixed higher than the
average snow level for the area. A site that has regular coverings of snow for a long
period of time may not be suitable for developing a solar Pv plant.

Temperature - The efficiency of a Pv power plant reduces with increasing

temperature. If a high temperature site is being considered, mitigating measures

should be included in the design and technology selection. For instance, it would be
better to choose modules with a low temperature coefficient for power.
Ideally, the site should be flat or on a slight south facing (in the northern hemisphere)
slope. Such topography makes installation simpler, and reduces the cost of technical

modifications required to adjust for undulations in the ground. With additional cost
and complexity of installation, mounting structures can be designed for most
locations. In general, the cost of land must be weighed against the cost of designing a
mounting structure and installation time.

Geotechnical
A geotechnical survey of the site is recommended prior to final selection. The purpose
is to assess the ground conditions in order to take the correct design approach, and to
ensure that the mounting structures will have adequate foundations. The level of the
geotechnical survey required will depend on the foundation design that is envisaged.

UTILITY SCALE SOLAR POWER PLANTS

Best practice dictates that either boreholes or trial pits are made at regular intervals
and at a depth appropriate for the foundation design. The boreholes or trial pits would
typically assess:
The groundwater level.
The resistivity of the soil.
The load-bearing properties of the soil.
The presence of rocks or other obstructions.

The soil pH and chemical constituents in order to assess the degree of corrosion

protection required and the properties of any cement to be used.

The geotechnical study may also be expected to include an assessment of the risk of
seismic activity and the susceptibility to frost, erosion and flooding.

Access
The site should allow access for trucks to deliver plant and construction materials.
This may require upgrading existing roads or building new ones. At a minimum,
access roads should be constructed with a gravel chip finish or similar.
The closer the site is to a main access road, the lower the cost of adding this
infrastructure. Safe packaging of the modules and their susceptibility to damage in
transport must also be carefully considered.

The site should be in a secure location where there is little risk of damage from either
people or wildlife. It should ideally be in a location where security and maintenance
personnel can respond quickly to any issue and this requirement should be stipulated
in the maintenance contract.

Grid Connection
A grid connection of sufficient capacity is required to enable the export of power. The
viability of grid connection will depend on three main factors: capacity, availability
and proximity. These factors should be considered thoroughly at an early stage of a
project; otherwise, the costs could become prohibitive if the site is later found to be in
an unfavourable area for grid connection.

Capacity - The capacity for the grid to accept exported power from a solar plant

will depend on the existing network infrastructure and current use of the system. The
rating of overhead lines, cables and transformers will be an important factor in
assessing the connection capacity available. Switchgear fault levels and protection
settings may also be affected by the connection of a generation plant. In cases where a
network does not have the existing capacity to allow connection, there are two options
available:
1) to reduce peak power export to the allowable limits of the network.
2) to upgrade the network toallow the desired export capacity. network upgrade
requirements will be advised by the network operator. But some aspects of that
upgrade can be carried out by contractors other than the network operator. Initial
investigation into network connection point capacitycan often be carried out by
reviewing published data.
However, discussion with the network operator will berequired to fully establish the
scope of work associated with any capacity upgrades.

Proximity - A major influence on the cost of connecting to the grid will be the

distance from the site to the grid connection point. Sites should be at locations where
the cost of grid connection does not adversely affect project economics. Besides, a

higher connection voltage will entail increased cost of electrical equipment such as
switchgear and transformers, as well as a higher conductor specification. A higher
voltage is also likely to increase the time taken to provide the connection resulting in a
longer development period.

Availability - The grid availability describes the percentage of time that the

network is able to export power from the solar Pv plant. The annual energy yield from
a plant may be significantly reduced if the grid has significant downtime. This may
have adverse effects on the economics of the project.
Availability statistics should be requested from the network operator to establish the
expected downtime of the network. In developed areas, the availability of the grid is
usually very high.

Land Use
Solar Pv power plants will ideally be built on low value land. If the land is not already
owned by the developer, then the cost of purchase or lease needs to be considered. The
developer must purchase the land or rights for the duration of the project. Besides
access to the site, provision of water, electricity supplies and the rights to upgrade
access roads must be considered along with relevant land taxes. Since government
permission will be required to build a solar plant, it is necessary to assess the site in
line with the local conditions imposed by the relevant regulatory bodies. If the land is
currently used for agricultural purposes, then it may need to be re-classified for
industrial use with cost and time implicationsand the possibility of outright
rejection. The future land use of the area must also be taken into account. It is likely
that the plant will be in operation for at least 25 years. As such, extraneous factors
need to be considered to assess the likelihood of their impact on energy yield. For
example, the dust associated with building projects could have significant impact on
the energy yield of the plant. Locating the plant in an environmentally sensitive area
should be avoided. Government stipulated environmental impact assessments or
plant/wildlife studies will slow down and potentially stop the development of a
project.

Any trees on the project site and surrounding land may need to be felled and removed,
with associated cost.
Clearances from the military may be required if the site is in or near a militarysensitive area. Glare from solar modules can affect some military activities.

Module Soiling
If the modules are soiled by particulates, then the efficiency of the solar plant could be
significantly reduced. It is, therefore, important to take into account local weather,
environmental, human and wildlife factors while determining the suitability of a site
for a solar Pv plant. The criteria should include:
dust particles from traffic, building activity, agricultural activity or dust storms.

Module soiling from bird excreta. Areas close to nature reserves, bird breeding

areas and lakes should be carefully assessed. Soiling of modules may require an
appropriate maintenance and cleaning plan at the site location.

Water Availability
Clean, low mineral content water is preferred for cleaning modules. A mains water
supply, ground water, stored water or access to a mobile water tank may be required;
the cost of the various options will have an impact on the project economics. The
degree to which water availability is an issue will depend upon the expected level of
module-soiling, the extent of natural cleaning due to rainfall and the required cleaning
frequency.

Financial Incentives
Financial incentives (such as feed-in tariffs or tax breaks) in different countries, or
regions within countries, have a strong bearing on the financial viability of a project.
Such incentives could outweigh the costs associated with one or more of the site
selection constraints.

UTILITY SCALE SOLAR POWER PLANTS

ENERGY YIELD PREDICTION
An important step in assessing project feasibility andattracting finance is to calculate
the electrical energy expected from the Pv power plant. The energy yield prediction
provides the basis for calculating project revenue. The aim of anenergy yield analysis
is to predict the average annual energy output for the lifetime of the proposed power
plant. Typically, a 25 to 30 year lifetime is assumed.The level of accuracy needed for
the energy yield prediction depends on the stage of project development. For example,
a preliminary indication of the energy yield can be carried out using solar resource
data and estimates of plant losses based on nominal values seen in existing projects.
For a more accurate energy yield prediction, software could be used to illustrate
detailed plant specifications and three-dimensional modelling of the layout. Modelling
will also help assess shading losses within time-step simulation.
To accurately estimate the energy produced from a Pv power plant, information is
needed on the solar resource and temperature conditions of the site in addition to the
layout and technical specifications of the plant components. Sophisticated software is
often used to model the complex interplay of temperature, irradiance, shading and
wind- induced cooling on the modules. While a number of software packages can
predict the energy yield of a Pv power plant at a basic level, financiers generally
require an energy yield prediction carried out by a suitable expert.
Typically, the procedure for predicting the energy yield of a Pv plant using time-step
(hourly or sub-hourly) simulation software will consist of the following steps:
1. Sourcing modeled or measured environmental data such as irradiance, wind speed
and temperature from land-based meteorological stations or satellite imagery (or a
combination of both). This results in a time series of typical irradiation on a
horizontal plane at the site location along with typical environmental conditions.
2. Calculating the irradiation incident on the tilted collector plane for a given time
step.

3. Modelling the performance of the plant with respect to varying irradiance and
temperature to calculate the energy yield prediction in each time step.
4.

Applying losses using detailed knowledge of the inverters, Pv module

characteristics, the site layout, DC and AC wiring, module degradation, downtime and
soiling characteristics.
5. Applying statistical analysis of resource data and assessing the uncertainty in input
values to derive appropriate levels of uncertainty in the final energy yield prediction.
These steps are described in more detail in the following sections.

Irradiation on Module Plane

In order to predict the solar resource over the lifetime of a project, it is necessary to
analyse historical data for the site. These data are typically given for a horizontal
plane. The assumption is that the future solar resource will follow the same patterns as
the historical values. Historical data may be obtained from land-based measurements
or from information obtained from satellite imagery as described in Section 3.2.1 data
in hourly or sub-hourly time steps are preferred. Statistical techniques can be used to
convert average monthly values into simulated hourly values if these are not
immediately available.
Horizontal plane irradiation may be divided into its diffuse and direct components.
Models are used to calculate the resource on the specific plane at which the modules
are tilted. Part f this calculation will take into account the irradiance reflected from
the surroundings towards the modules. The degree to which the ground is able to
diffusely reflect radiation is quantified by the albedo values, which vary according to
surface properties. A higher albedo factor translates into greater reflection and so
higher levels of diffuse irradiation.
For example, fresh grass has an albedo factor of 0.26, reducing down to a minimum of
approximately 0.15 when dry. Asphalt has a value between 0.09 and 0.15 or 0.18 if
wet.

Performance Modelling
Sophisticated simulation software is used to predict the performance of a Pv power
plant in time steps for a set of conditions encountered in a typical year. This allows a
detailed simulation of the efficiency with which the plant converts solar irradiance
into AC power and the losses associated with the conversion. Some of these losses
may be calculated within the simulation software, others are based on extrapolations
of data from similar Pv plants and analysis of the site conditions.depending on specific
site characteristics and plant design, losses may be caused by any of the factors
described. Energy yield prediction reports should consider and (ideally) quantify each
of these losses. In individual cases, some of the losses may be negated or considered
in logical groupings.

Energy Yield Prediction Results

The predicted annual energy yield may be expressed within a given confidence
interval. A P90 value is the annual energy yield prediction that will be exceeded with
90% probability; P75 is the yield that will be exceeded with 75% probability; while
P50, the expected value, is the annual energy yield prediction that will be exceeded
with 50% probability. Good quality energy yield reports used by investors will give
the P50 and P90 energy yield prediction values as a minimum.

Uncertainty in the Energy Yield Prediction

The uncertainty of energy yield simulation software depends on each modelling stage
and on the uncertainty in the input variables. Modelling software itself can introduce
uncertainty of 2% to 3%.
The typical relative accuracy of measurements at meteorological (MET) stations by a
well-maintained pyranometer is 3-5%. This represents the upper limit in accuracy of
resource data obtained through MET stations. However, in many cases, the presence
of a MET station at the project location (during preceding years) is unlikely. If this is
the case, solar resource data will likely have been obtained using satellites or
interpolation as described in Section 3.2. This will increase the uncertainty in the

resource data depending on the quality of the satellite or the distance from a well
maintained MET station. In general, resource data,.

PLANT DESIGN
Designing a megawatt-scale Pv power plant is a complex process that requires
considerable technical experience and knowledge. There are many compromises that
need to be made in order to achieve the optimum balance between performance and
cost. This section highlights some of the key issues that need to be considered when
designing a Pv plant.

Technology Selection
Modules
While certification of a module to IEC/CE/UL standards (as described in Section
2.3.6) is important, it says very little about the performance of the module under field
conditions of varying irradiance and temperature. It is also relatively difficult to find
comprehensive and independent module performance comparisons. In addition,
modules tested under a specific
set of conditions of irradiance, temperature and voltage, with a specific inverter, may
perform very differently under alternative conditions with a different inverter. This
makes choosing a module a more difficult process than it may first appear. Many
developers employ the services of an independent consultant for this reason. When
choosing modules, the following key aspects should be considered:

The aim is to keep the levelised cost of electricity (LCoE) at a minimum. When

choosing between high efficiency-high cost modules and low efficiency- low cost
modules, the cost and availability of land and plant components will have an impact.
High efficiency modules require significantly less land, cabling and support structures
per MWp installed than low efficiency modules.

When choosing between module technologies such as mono-crystalline silicon,

multi crystalline silicon and The cost per kWh of electricity generated that takes into
account the time value of money thin film amorphous silicon, it should be realised that
each technology has examples of high quality and low quality products from different
manufacturers.

different technologies have a differing spectral response and so will be better

suited for use in certain locations, depending on the local light conditions.

Amorphous silicon modules generally perform better under shaded conditions

than crystalline silicon modules. Many of them show a better response in low light
levels.

The nominal power of a module is given with a tolerance. Some modules may

be rated with a 5% tolerance while others are given with a 3% tolerance. Some
manufacturers routinely provide modules at the lower end of the tolerance, while
others provide modules that achieve their nominal power or above (positive
tolerance).

When ordering a large number of modules, it is recommended to have a sample

of modules independently tested to establish the tolerance.

The value of the temperature coefficient of power will be an important

consideration for modules installed in hot climates.

The degradation properties and long term stability of modules should be

understood. The results of independent testing of modules can sometimes be found in

scientific journals or papers from research institutes.

The manufacturers warranty period is useful for distinguishing between

modules but care should be taken with the power warranty. Some manufacturers offer
as guarantee of performance the percentage of the peak power for a given duration;
others give it as a percentage of the minimum peak power (that is, the peak power
minus the tolerance).

Frameless modules may be more suitable for locations that experience snow, as

snow tends to slide off these modules more easily.

other parameters important for selection of modules include: cost ($/Wp), lifetime,
and maximum system voltage.

Quality Benchmarks

Product guarantee - In the EU, manufacturers are legally bound to provide a

product guarantee ensuring that modules will be fully functional for a minimum of 2
years. Some companies guarantee a longer period, with 5-6 years being the usual
duration.

Power guarantee - In addition to the product guarantee, most manufacturers

grant nominal power guarantees. These vary between manufacturers but a typical
power guarantee stipulates that the modules will deliver 90% of the original nominal
power after 10 years and 80% after 25 years. So far no module manufacturer has
offered a power output guarantee beyond 25 years. The conditions listed in both the
power guarantee and product guarantee are
important, and vary between manufacturers.

Lifetime - Good quality modules with the appropriate IEC certification have a

design life in excess of 25 years. Beyond 30 years, increased levels of degradation

may be expected. The lifetime of crystalline modules has been proven in the field.
Thin film technology lifetimes are currently unproven and rely on accelerated lifetime
laboratory tests, but are expected to be in the order of 25-30 years also. The module
data sheet format and the information that should be included has been standardised
and is covered by En 50380, which is the data sheet and nameplate information for
photovoltaic modules. An example of the information expected in a data sheet is
provided.

Table: Comparison of Module Technical Specifications at STC

Manufacturer Xxxx

Module Model

Xxxx

nominal power (PMPP)

Power tolerance

210 Wp

3%

voltage at PMAX (vMPP) 26.5 v

Current at PMAX (IMPP) 7.93 A

open circuit voltage (voP) 33.2 v

Short circuit current (ISC) 8.54 A

Maximum system voltage 1000 vdC

Module efficiency 14.33%

operating temperature

-40C to +85C

Temperature coefficient of PMPP -0.41%/C

dimensions 148099050 mm

Module area 1.47 m2

Weight

18 kg

Maximum load

5400 Pa

Product warranty

5 years

Performance guarantee

90%: after 10 years; 80%: after 25 years

Inverters

No single inverter concept is best for all situations. In practice, the local conditions
and the system components have to be taken into account to tailor the system for the
specific application. different solar Pv module technologies and layouts may suit
different inverter types. So care needs to be taken in the integration of modules and
inverters to ensure optimum performance and lifetime. Among the major selection
criteria for inverters, the

financial incentive scheme and the dC-AC conversion

efficiency are major inverter selection criteria, directly affecting the annual revenue of
the solar Pv plant. It is also important to bear in mind that efficiency varies according
to a number of factors. of them, dC input voltage and percentage load are the two
dominant factors. Several other factors should inform inverter selection, including site
temperature, product reliability, maintainability, serviceability and total cost of
ownership. A thorough financial analysis is required to determine the most costeffective inverter option. Many of the inverter selection criteria listed in Table may
feed into this analysis.
Table Criterion Description
Incentive scheme
Banding of financial incentive mechanisms may have an
influence on the choice of inverter. For example, feed-in tariff
(FiT) schemes might be tiered for different plant sizes which
may, in turn, influence the inverter size.
Project size
Size influences the inverter connection concept. Central
inverters are commonly used in large solar PV plants.
Performance
High efficiency inverters should be sought. The additional
yield usually more than compensates for the higher
initial cost. The way the efficiency has been defined should be
carefully considered.
Consideration must also be given to the fact that efficiency
changes with DC input voltage, percentage of load, and several

other factors.
MPP range A wide MPP range allows flexibility and facilitates design.
3-phase or single phase output
National electrical regulations might set limits on the
maximum power difference between the phases in the case of
an asymmetrical load. For example, this limit is 4.6 kVA in the
German regulations.
Module technology
The compatibility of thin-film modules with transformer-less
inverters should be confirmed with manufacturers.

General Layout

Minimising cable runs and associated electrical losses may suggest positioning an
Lv/Mv station centrally within the plant. If this option is chosen, then adequate space
should be allocated to avoid the possibility of the station shading modules behind it.
The layout should allow adequate distance from the perimeter fence to prevent
shading. It should also incorporate access routes for maintenance staff and vehicles at
appropriate intervals.

Tilt Angle
Every location will have an optimal tilt angle that maximises the total annual
irradiation (averaged over the whole year) on the plane of the collector. For fixed tilt
grid connected power plants, the theoretical optimum tilt angle may be calculated
from the latitude of the site. However, adjustments may need to be made to account
for:
Soiling - Higher tilt angles have lower soiling losses.The natural flow of rainwater
cleans such modules more effectively and snow slides off more easily at higher
angles.

Shading - More highly tilted modules provide more shading on modules behind

them. As shading impacts energy yield much more than may be expected simply by
calculating the proportion of the module shaded, a good option (other than spacing the
rows more widely apart) is to reduce the tilt angle. It is usually better to use a lesser
tilt angle as a trade-off for loss in energy yield due to inter-row shading.

Seasonal irradiation distribution - If a particular season dominates the annual

distribution of solar resource (monsoon rains, for example), it may be beneficial to

adjust the tilt angle to compensate for the loss. Simulation software is able to assess
the benefit of this.

Inter-Row Spacing

The choice of row spacing is a compromise chosen to reduce inter-row shading while
keeping the area of the PV plant within reasonable limits, reducing cable runs and
keeping ohmic losses within acceptable limits. Inter-row shading can never be reduced
to zero: at the beginning and end of the day the shadow lengths are extremely long.
Figure 18 illustrates the angles that must be considered in the design process.
PV Array Design
The

design

of

Pv

array

will

depend

on

the

inverter

specifications and the chosen system architecture besides the specific context and
conditions of use. Using many modules in series in high voltage arrays minimises
ohmic losses. However, safety requirements, inverter voltage limits and national
regulations also need to be considered.
Maximum number of modules in a string -The maximum number of modules in a
string is defined by the maximum dC input voltage of theinverter to which the string
will be connected to (vMax

(Inv, dC)

). Under no circumstances should this voltage be

exceeded. Crossing the limit can decrease the inverters operational lifetime or render
the device inoperable. The highest module voltage that can occur in operation is the
open-circuit voltage in the coldest daytime temperatures at the site location. design
rules of thumb for Europe use - 10C as the minimum design temperature, but this
may vary according to location. The maximum number of modules in a string (n Max)
may therefore be calculated using the formula: VOC(Module)@Coldest Module Operating TemperatureNMax
<VMax(Inv, DC)

Minimum number of modules in a string The minimum number of modules is

governed by the requirement to keep the system voltage within the MPPT range of the
inverter. If the string voltage drops below the minimum MPP inverter voltage, then the
system will underperform. In the worst case, the inverter may shut down. The lowest
expected module voltage occurs during the highest operating temperature conditions.
design rules of thumb for Europe use 70C as the design benchmark, but this may
vary according to site conditions. The minimum number of modules in a string (n Min)
may therefore be calculated using the formula:

VMPP(Module) @Highest Module Operating TemperatureNMin>VMPP(Inv Min)

Voltage optimisation - As the inverter efficiency is dependent on the operating

voltage, it is preferable to optimise the design by matching the array perating voltage
and inverter optimum voltage as closely as possible. This will require voltage
dependency

graphs

in

19).

Figure

inverter

of
If

inverter
such

efficiency

graphs

manufacturers,

they

are

(see
not

may

example

provided

be

by

obtained

from independent sources. Substantial increases in the plant yield can be achieved by
successfully matching the operating voltages of the Pv array with the inverter.
Number of strings - The maximum number of strings permitted in a Pv array is a
function of the maximum allowable Pv array current and the maximum inverter
current. In general, this limit should not be exceeded as it leads to premature inverter
ageing and yield loss.

Inverter Sizing
It

is

strategy

not

possible

that

applies

to

formulate

in

all

an

cases.

optimal

Project

inverter

specifics

sizing

such

as

the

solar resource and module tilt angle play a very important role when choosing a
design.
been

to

this

is

option
to

While
use
not

always

might

lead

curtail

(based

an

on

compliance

power

the

rule

of

inverter-to-array
the
to

best
a

spikes

one

hour

data).

in

cases

where

power

design

situation
not

thumb
ratio

approach.
where

anticipated

or, it

could

reactive

fail

power

the
by
to

has

less
For

than

unity,

example,

this

inverter

manages

irradiance

profiles

achieve

injection

to

grid

code

the

grid

is required.
The optimal sizing is, therefore, dependent on the specifics of the plant design. Most
plants will have an inverter sizing range within the limits defined by:

0.8<Power ratio<1.2

Where:
PInverter DC rated
Guidance on inverter and PV array sizing can be obtained
from the inverter manufacturers, who offer system sizing software. Such tools also
provide an indication of the total number of inverters required.
A number of factors and guidelines must be assessed when sizing an inverter:

The maximum voC in the coldest daytime Temperature must be less than the

inverter maximum dC input voltage (vInv, dC Max).

The inverter must be able to safely withstand the maximum array current.

The minimum voC in the hottest daytime temperature must be greater than the

inverter dC turn-off voltage (vInv,

dC Turn-off

).

The maximum inverter dC current must be greater than the Pv array(s) current.

The inverter MPP range must include Pv array MPP points at different

temperatures.

When installed, some thin film modules produce a voltage greater than the

nominal voltage. This happens for a period of time until initial degradation has
occurred, and must be taken into account to prevent the inverter from being damaged.

Grid code requirements: for example, reactive power injection.

The operating voltage should be optimised for maximum inverter efficiency.

Power Ratio =PInverter DC rated =

P

PV Peak

Pinverter AC rated
n100%

Site conditions of temperature and irradiation profiles.

Economics and cost-effectiveness.

Inverters with reactive power control are recommended.

Inverters can control reactive power by controlling the phase angle of the current
injection. Moreover, aspects such as inverter ventilation, air-conditioning, lighting and
cabinet heating must be considered. When optimising the voltage, it should be borne
in
mind

that

the

inverter

efficiency

is

dependent

on

voltage.

Specification sheets and voltage dependency graphs are required for efficient voltagematching.

Cable Selection and Sizing

The

selection

should

and

take

sizing

for

Pv

plants

and

regulations

each

country.

Cables

specifically

designed

for

(solar

cables)

are

available

and

general,

three

installations
used.

In

codes

solar

national

solar

be

cables

account

to

should

dC

into

applicable
Pv

of

criteria

readily
must

be

observed

when sizing cables:

1. The cable voltage rating. The voltage limits of the cableto which the Pv string or
array cable will be connectedmust be taken into account. Calculations of the
maximum voC voltage of the modules, adjusted for the site minimum design
temperature, are used for this calculation.
2. The current carrying capacity of the cable. The cable must be sized in accordance
with the maximum current. It is important to remember to de-rate appropriately, taking
into account the location of cable, the method of laying, number of cores and
temperature. Care must be taken to size the cable for the worse case of reverse current
in an array.
3. The minimisation of cable losses. The cable voltage drop and the associated power
losses must be as low possible. normally, the voltage drop must be less than 3%, but
national regulations must be consulted for guidance. Cable losses of less than 1% are
achievable.

Sizing and Selecting Transformers

The purpose of transformers in a solar power plant is to provide suitable voltage levels
for transmission across the site and for export to the grid. In general, the inverters
supply
power at Lv. But for a commercial solar power plant, grid connection is typically
made at upwards of 11 kv (Hv levels).
It is therefore necessary to step up the voltage using a transformer between the
inverter and the grid connection point shows a high level single line diagram showing
typical voltages of operation for the AC system of a solar power plant. Selection and

specification - The selection of an appropriate transformer should consider several

basic issues. These include the required size of the transformer, its position within the
electrical system, and the physical location of installation. The size of the transformer,
which will depend on the projected maximum power exported from the solar array,
should be specified in MvA. Power would generally be expected to flow from the
solar arrays to the grid. To prepare for the reverse case (a need to supply power back
to the plant), this should be specified or an auxiliary transformer used. The position of
the transformer in the electrical system will define the required voltage levels on the
primary and secondary sides of the transformer. Tertiary supplies for substation
auxiliary

services

and/or

harmonic

mitigation should also be considered.

In addition, the physical location and the anticipated environmental conditions will
need to be specified.International and national standards should be specified as
required. The requirements for power transformers are defined in IEC 60076.

Infrastructures
A utility scale Pv power plant requires infrastructure appropriate to the specifics of the
design chosen. Locations should be selected in places where buildings will not cast
unnecessary shading on the Pv module. It may be possible to locate buildings on the
northern edge of the plant to reduce shading, or to locate them centrally if appropriate
buffer
zones are allowed for. depending on the size of the plant, infrastructure requirements
may include:

Office - A portable office and sanitary room with communication devices. This

must be watertight and prevent entry to insects. It should be located to allow easy
vehicular access.

LV/MV station - Inverters may either be placed amongst the module support

structures (if string inverters are chosen) in specially designed cabinets or in an

inverter house along with the medium voltage transformers, switchgear and metering
system. This Lv/Mv station may be equipped with an air conditioning system if it is
required to keep the electrical devices within their design temperature envelopes.
MV/HV station - An Mv/Hv station may be used to collect the AC power from the
medium voltage transformers and interface to the power grid. For string inverters, the
Lv/Mv station may be used to collect the AC power.

Communications - The plant monitoring system and the security system will

require a
communications medium with remote access for visibility and control of the plant.
There can also be a requirement from the grid network operator for specific telephone
landlines for the grid connection. often, an Internet broadband (dSL) or satellite
communications system is used for remote access. A GSM (Global System for Mobile
Communications) connection or standard telephone line with modems is an alternative
though it has a lower data transfer rate.

other issues to consider when designing and specifying a transformer may also
include:
Tap setting requirements.
Cooling medium.
Earthing.
Winding connections.
number of windings.
requirement for redundancy/spare transformer. Losses.
Bushings for connection of cabling and overhead lines.
Transformer trip and warning alarms.
Choosing a reputable manufacturer to carry out the detailed design and manufacture
should ensure that the transformer provided is of the required standard. There are, of
course,
many other parameters and design considerations that could be specified.
Losses

Transformers

can

lose

energy

through

magnetising current in the core, known as iron losses and copper losses in the
windings. Minimising the losses in a transformer is a key requirement as this will
increase the energy supplied to the grid and, as a result, enhance the revenue of a solar
power plant.
Test Requirements - Transformers should be subject to a number of routine and type
tests performed on each model manufactured; these tests are set out in IEC 60076. The
Manufacturer also can be requested to undertake special tests mentioned in IEC
60076.

Forecasting Technology
dispatchable

power

forecast

the

to

plants
network

typically
operator.

need
This

to

helps

provide
to

fix

a
plant

schedules and guarantee continuity of supply. often,

production forecasts (in half hourly time-steps) are required 24 hours in advance. This
entails weather forecasts coupled with power forecasting algorithmsmore so since
Pv power production is intermittent and random in nature. Such
forecasting algorithms can use physical models, statistical approaches or a
combination of both. At the least, the algorithms require the definition of:
Power plant capacity.
Module tilt and orientation.
Module specifications.
Latitude and longitude of the plant.
Meteorological agency data, gathered from ground
measurement stations and/or satellites.
The

algorithms

regional

typically

forecasts

and

take

break

three-hour

them

down

to

national
30

and/or

minute

local

forecasts (temporal interpolation) before using algorithms

to

forecast

production

power
and

production.

actual

weather

Comparison
can

also

of

historical

allow

learning

algorithms to be employed. Figure 23 shows the components of a forecasting system.

results of forecasting are typically posted on web portals. There are a variety of
commercial
forecasting

products

available

in

the

market

today.

But

availability may be limited to regions that have rapid access to meteorological

agencys weather data.

Conclusions
The performance of a Pv power plant may be optimised by reducing the system losses.
reducing the total loss increases the annual energy yield and hence the revenue,
though in some cases it may increase the cost of the plant. In addition, efforts to
reduce one type of loss may conflict with efforts to reduce losses of a different type. It
is the skill of the plant designer to make compromises that result in a plant with a high
performance at a reasonable cost. For plant design, there are some general rules of
thumb. But specifics of project locationssuch as irradiation conditions, temperature,
sun angles and shadingshould 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 scrutinized

PHOTOS