WNA Report

Optimized Capacity:
Global Trends and Issues
A Report by the World Nuclear Association’s
Capacity Optimization Working Group
Contents

Executive Summary 2

1. Introduction 3

2. Data Model 4

3. Industry Trends
3.1 Global Overview 6
3.2 Unavailable Capacity 9
3.3 Available Capacity 14

4. Factors Affecting Capacity 16

5. Conclusions 17

6. Definitions
6.1 Values 18
6.2 Indicators 19

7. Contributors 20

Figures
1. Data Model 4
2. Performance Indicator Derivation 5
3. Global Capacity Factor over Time 6
4. Number of Reactors Not Operated for the Entire Year 6
5. Long and Short Term Capacity Factors by Region 7
6. Long and Short Term Capacity Factors by Country 7
7. Long and Short Term Capacity Factors by Reactor Type 8
8. Long and Short Term Capacity Factors by Reactor Age 9
9. Energy Loss Distribution 10
10. Median Performance Indicator by Reactor Type 10
11. Best Quartile Performance Indicator by Reactor Type 10
12. Planned Energy Loss Causes 11
13. Refuelling Outage Durations 11
14. Unplanned Energy Loss by Outage Type 12
15. Unplanned Energy Loss by System 12
16. Unplanned Energy Loss By Direct Cause 13
17. Availability of Reactor Units 14
18. Recent Capacity Factors 14
19. Average Number of Automatic Scrams for Capacity Factor Intervals 15

1
 Executive Summary
This paper reflects the research of the World Nuclear Association’s Capacity Optimization Working
Group; a group constituted to identify means by which nuclear operators worldwide can both determine
and attain their optimal capacity. In order to progress towards this goal, this report establishes a status
baseline and undertakes high-level analysis to understand at what point the global industry currently stands
and what the dominant issues in utilization of the installed power base of the current nuclear fleet are.

The electrical output of a nuclear power plant is dependent on a wide variety of factors. Issues as diverse
as engineering, organization, regulation and finances all impact on the ability of the plant to produce
electricity. Within the boundaries set by these ‘real world’ issues it is desirable that a nuclear reactor
should perform at its best achievable capacity, its optimized capacity. Additionally, the boundaries
themselves can also be questioned, understood and influenced.

In the 20-year period from 1980 there was a significant rise in the median global actual energy utilization
of reactors' maximum power capability1 from 68%, culminating in 2002 in a historical maximum of
86%2. However, since around the turn of the century this growth has levelled off and has remained
constant at around the 85% mark for the last 8 years, which suggests that renewed focus should be
placed on more effective utilization of the existing nuclear fleet.

In 2008 the global median capacity factor was 84.5%, but there was a very broad spread in this
performance indicator between individual plants. Generally, this variance is not explained by the reactor
type used, or by age of the reactor. Indeed, the best performing utilities represent a range of
technologies, vendors and regions, suggesting that performance is independent of these choices.

Globally speaking, in recent years 94% of a plant’s unavailable capacity is due to reasons under
management control, the dominant cause being maintenance combined with refuelling. Best performing
utilities have significantly shorter and better-controlled outages that are of at least as good quality. More
broadly than this, best performers maximize their availability and minimize their unplanned unavailability;
they plan for success and are able to stick with their plan.

It is seen that the major direct cause of unplanned loss is failure or problem with plant equipment, with
the turbine and auxiliary system having the greatest effect. Additionally we see that indicators of plant
safety and capacity are linked: a well-managed plant is generally both productive and safe.

With potential significant benefits in safety, economics, security and environmental performance available,
it is clear that further work on optimizing the current global nuclear fleet’s capacity has merit.
In particular, the Capacity Optimization Working Group has identified four key areas for development:
Root cause analysis applied to unplanned energy loss
Case studies in managing planned energy loss
More detailed investigation of ageing
Development of a mathematical model of optimal capacity

1
Under ambient conditions.
2
This performance indicator is known as the capacity factor.

The views expressed in this report do not necessarily represent those of any government or company with which individual
members of the WNA may be associated.

2
1 Introduction

The electrical output of a nuclear power plant is dependent on a wide variety of factors. Issues as
diverse as engineering, organization, regulation and finances all impact on the ability of the plant to
produce electricity. In 2008 these factors contributed to a spread of production between 0%3 and over
100%4 of the reactor’s maximum power capability5. Within the boundaries set by these ‘real world’
issues it is desirable that a nuclear reactor should perform at its best achievable capacity, its optimized
capacity. Additionally, the boundaries themselves can also be questioned, understood and influenced.

The benefits of moving towards the attainment of optimized capacity are numerous and include:
Safety – enhancing nuclear and industrial health and safety through minimizing unplanned outages and
reducing associated transients
Economic – maximizing the return on an asset-based business
Security – contributing to the security and diversity of energy supply
Environmental – increasing power generation from non-greenhouse gas emitting power and making
best use of available materials and resources.

In 2008 the median capacity factor6 for all the world’s operating nuclear reactors was 84.5%; if this
could be increased (in relative terms) by 10%, to a capacity factor of 93%, this would:
Result in the production of an extra 260TWh, enough to power over 54 million homes7
Be equivalent to connecting numerous8 new reactors to the grid
Avoid the emission of 260 million tonnes of carbon dioxide9.

The capacity of the nuclear fleet should therefore be of interest to a wide audience including operators,
financiers, policy makers and regulators, as well as the general public.

This paper reflects the research and collective experience of the WNA’s Capacity Optimization Working
Group, a group constituted to identify means by which nuclear operators worldwide can both determine
and attain their optimal capacity. This paper is intended as a broad overview of the global trends and
identifies the topics that will be covered in greater depth in subsequent reports.

3
Percentages quoted are the 2008 annual median capacity factor.
4
A reactor can achieve over 100% by performing above its reference energy generation, generally through experiencing ambient conditions
significantly different from its reference ambient conditions.
5
Under ambient conditions.
6
The capacity factor is a performance indicator which reflects the actual energy utilization of the unit for electricity and heat production.
7
Based on the average energy consumption of a UK home.
8
Approximately 35 1GW reactors.
9
Estimated if coal had been used as a direct replacement.

3
 2 Data Model

Data is required in order to evaluate the performance of a nuclear power plant, or groups of plants.
A model is used to determine what data is useful to collect and to categorize it in an agreed
standardized manner. Figure 1 shows the model that has been adopted for data values collected
from nuclear power plants.

The reference unit power10 is the maximum power capability of a unit under average ambient conditions.
As shown in Figure 1, it can be split into two components, ‘available capacity’ and ‘unavailable capacity’.
The balance between these two components is determined by eight factors: outage execution,
equipment reliability, regulatory environment, organizational factors, engineering, safety performance,
finances, supply chain processes. These factors are expanded on in Section 4 of this paper.

Further, ‘available capacity’ can be broken down into what is and what is not supplied to the grid. Similarly
‘unavailable capacity’ can be broken down into elements that are or are not under plant management
control. Finding and achieving the optimal balance between ‘generation supplied’ and the other three
components at this level is the essence of WNA’s work on capacity optimization.

Two additional important concepts can be defined using this model: ‘availability’ is the sum of the
‘generation supplied’ and ‘available but not supplied’; ‘capability’ is the sum of the ‘generation supplied’
and ‘available but not supplied’ and the element of unavailable capacity which is ‘not under plant
management control’.

Reference unit
power

Available Total unavailable

capacity capacity

Not under plant Under plant

Generation Available but
management management
supplied not supplied
control control

Availability
Capability

Figure 1: Data Model

10
The reference unit power is the maximum (electrical) power of the unit under reference ambient conditions. The reference power is
based on design values, adjusted for reference ambient conditions. The reference unit power is expected to remain constant unless
design changes that affect the capacity are made to the unit.

4
In order to determine performance measures, the concept of reference energy generation (REG) is
used. It is determined by multiplying the reference unit power by the reference period11. By dividing the
components at the lowest level of Figure 1 by REG, we derive a set of indicators that can be used across
the nuclear fleet. The relationship between values and indicators is shown in Figure 2.

Reference Unit Power (RUP)

Reference Energy Generation (REG)

Values

Available Capacity (P) Total Unavailable Capacity (EL)

Not under plant

Generation Available but
management Under Plant Management Control
Supplied (EG) not supplied
control (XEL)

Unplanned
Planned Energy
Energy Loss
Loss (PEL)
(UEL)

Energy Availability Factor (EAF) Energy Unavailability Factor (EUF)

Indicators

External Planned Unplanned

Capacity Factor
Unavailability Unavailability Unavailability
(CF)
Factor (XUF) Factor (PUF) Factor (UUF)

Unit Capability Factor (UCF)

Figure 2: Performance Indicator Derivation

Performance indicators allow for meaningful statistical analysis of current and historic data held on the
nuclear fleet. Of particular interest is the ‘capacity factor’ indicator that relates to the ‘generation
supplied’ as discussed above. This indicator reflects the actual energy utilization of the unit for electricity
and heat production.

More detailed definitions of these performance values and indicators can be found in Section 6 of
this document.

Note: The data model, performance indicators and data used in this document are drawn from figures held in the
International Atomic Energy Agency (IAEA) Power Reactor Information System (PRIS) which constitutes the most complete
and authoritative technical data bank on nuclear power reactors in the world. This is with the exception of the capacity
factor as defined here, which the IAEA refers to as ‘load factor’.

11
The reference period is the time (in hours) over which the indicator is calculated.

5
 3
3.1 GLOBAL OVERVIEW
Industry Trends

90
Median Capacity Factor (%)

85

80

75

70

65
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008

Year

Figure 3: Global Capacity Factor over Time

Figure 3 shows that in the 20-year period 1980-2000 there was almost a 20% rise in the global
capacity factor (CF) culminating in 2002 in a historical maximum of 86%. However, since around the
turn of the century this growth has levelled off and has remained constant at around the 85% mark
for the last 8 years. This trend suggests that renewed focus should be put on a more effective
utilization of installed capacity base.

The 20% rise in the period 1980-2000 was despite an 8 year ‘recovery’ in the CF following
the Chernobyl accident in 1986. There have been some specific cases that have affected progress
more recently:
TEPCO case in 2003 – long-term shutdown of 17 TEPCO units (2003 and 2004)
Earthquake at Kashiwazaki Kariwa in July 2007 – 7 reactors not operated from this date
Long-term shutdown in 2008 of Brunsbuettel and Kruemmel in Germany
Ageing of Nuclear Power Plants (NPPs) – extended reconstructions of several old reactor units
(for example, in 2008 eight reactors were not operated for this reason).

These cases are included in Figure 4. While the number of not-operated reactors is following an upward
trend, their numbers are still low compared to the total number of operating reactors. Therefore these
specific cases are perturbations in a general trend of levelling off, not the cause of the trend.

25

20
Number

15

10

0
2000 2001 2002 2003 2004 2005 2006 2007 2008

Year

Figure 4: Number of Reactors Not Operated for the Entire Year

6
While Figure 3 shows a global median average CF, the actual CF of individual plants varies and in some
cases varies very widely from the worldwide average as demonstrated in Figures 5 and 6. The fact there
are differences, the causes of which are not necessarily understood, is the basis for the WNA’s Capacity
Optimization Working Group’s work. The first step to improvement is understanding the differences.
100

90

80
Median CF (%)

70

60

50

40

30
Africa America America Asia Asia Europe Europe Global
- Latin - Northern - Far East - Middle - Central - Western
East and and
South Eastern

2008 1999-2008
Figure 5: Long and Short Term Capacity Factors by Region

Global
Argentina
Armenia
Belgium
Brazil
Bulgaria
Canada
China
Czech Republic
Finland
France
Germany
Hungary
India
Japan
Country

South Korea
Lithuania
Mexico
Netherlands
Pakistan
Romania
Russia
Slovakia
Slovenia
South Africa
Spain
Sweden
Switzerland
Taiwan, China
Ukraine
UK
USA
30 40 50 60 70 80 90 100

Median Capacity Factor (%) 2008 1999-2008

Figure 6: Long and Short Term Capacity Factors by Country

Figures 5 and 6 have been included to demonstrate the variation from the global average. There is a wide
spread of CFs between regions and between countries within the same region. Local conditions can be
seen to come into play more directly (for example fuel supply issues, seasonal demand variations, load
following). While regions and countries may have restraints imposed on them by their local conditions,
all can look to continuously improve performance within these boundaries.

7
 Countries and regions will always be an important common denominator due to national and regional
regulatory control. However, companies and workforces are becoming increasingly internationalized
over time. If this trend towards globalization continues in the nuclear industry it will reduce the
importance of reactor nationality.
90

80
Median Capacity Factor (%)

70

60

50

40

30

20
BWR FBR GCR LWGR PHWR PWR Global

Reactor Type 2008 1999-2008

BWR – Boiling Water Reactor (including ABWR)

FBR – Fast Breeder Reactor12
GCR – Gas Cooled Reactor (including AGR)
LWGR – Light Water Graphite Reactor (including RMBK)
PHWR – Pressurized Heavy Water Reactor (otherwise known as CANDU)
PWR – Pressurized Water Reactor (including VVER)

Figure 7: Long and Short Term Capacity Factors by Reactor Type

The PWR and BWR together account for over 80% of operating units. Figure 7 shows that there is
effectively no difference between their global CFs over time. This similarity exists between two
profoundly different reactor types. The ageing mechanisms, chemistry performance, and standard
equipment are very different between these technologies, as are the operations, especially with respect
to refuelling requirements. Yet, despite these differences, both technology types achieve similar
performance, suggesting that technology is not a limiting factor to sustainable and efficient operation. The
next most prevalent reactor is the PHWR, followed by the GCR and LWGR. Only two FBRs operate
today and therefore they have been excluded from the statistical analysis in this paper12.

The high availability of PWR, PHWR and BWR reactors is despite decreased performance of BWR
and PHWR reactors in the last few years. The availability of BWR units has been significantly affected by
the TEPCO case in 2003 and the earthquake in Japan in 2007 (all TEPCO units are of BWR or ABWR
type). Had these technologies been operating at, or near, maximum practical capacity factor, the recent
events would have resulted in a downturn in performance. This suggests that there is a strong reserve
margin of capacity to be realized through operational best practices.

LWGR reactors have increased their availability significantly over the last few years. Performance of GCRs
has decreased significantly mainly due to type-specific ageing plant issues. These decreases in
performance may be attributed to end-of-life performance issues related to these technologies, as
opposed to operational issues.

12
There are only two FBRs operating in the world and therefore they are not statistically significant but are retained in Figure 7
for completeness.

8
 100

90

80
Median Capacity Factor (%)

70

60

50

40

30

20

10

0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

Age of Reactor in Years 2008 1999-2008

Figure 8: Long and Short Term Capacity Factors by Reactor Age

In general, no significant global age-related trend in capacity factor can be detected from Figure 8. This
is good news for older plants, which can maintain historic output levels, and also for new plants, which
do not appear to require any ‘run-in’ time, suggesting that industrial good practice in operations is being
passed on.

Figure 4 shows an increase in the number of reactors off line for an entire year – it is believed that this
is caused by increasing numbers of ageing reactors coming off line for major items to be refurbished.
Therefore there are some ageing effects on the fleet which are being managed. But Figure 8 suggests that
ageing reactors that are on-line are operating as well as new reactors.

What cannot be seen here is the cost of keeping older plants performing at historic levels, and whether
this cost is comparable with the cost of operating younger plants. It is also important to remember that
capacity factor here is different from output – older plants tend to have significantly lower reference
unit power.

While Figure 8 gives an overview, it is suspected that there will be trends hidden within it. A further,
more detailed analysis of ageing requires investigation by reactor type and reactor model. It also
requires filtering to manage those cases when the capacity factor is affected by a non-ageing reason,
such as an earthquake.

3.2 UNAVAILABLE CAPACITY

As shown in Figures 1 and 2, the power of a unit can be split into two parts; the Available Capacity (P)
and the Total Unavailable Capacity (EL). EL can be further broken down into three components:

PEL – planned energy loss

UEL – unplanned energy loss
XEL – loss that is not under plant management control (external loss)

9
 Planned
68%
Unplanned Forced
17%

Unplanned
External Outage
6% Extension 9%

Unplanned 26%

Figure 9: Energy Loss Distribution (2004-2008)

Figure 9 shows that globally, 94% of unavailable capacity is within plant management control.

Planned losses are the most significant factor, followed by unplanned losses. External reasons, which are
not under plant management control, are the smallest cause. In Figure 9 unplanned losses (which are
under plant management control) have been split into two components, demonstrating the importance
of unplanned outage extensions. Clearly planned losses are important, but unplanned causes should also
be addressed.

Capacity Planned Unplanned External Forced

Factor Unavailability Unavailability Unavailability Loss Rate
Factor Factor Factor
TOTALS: 82.79 10.34 2.49 0.36 1.42
BWR 85.60 7.97 2.66 0.14 1.05
GCR 66.27 11.94 16.45 0.05 12.89
LWGR 71.43 20.21 1.14 1.67 1.05
PHWR 75.56 9.22 4.45 0.76 3.43
PWR 85.06 10.35 1.78 0.36 1.17

Figure 10: Median Performance Indicator by Reactor Type (2004-2008)

Capacity Planned Unplanned External Forced

Factor Unavailability Unavailability Unavailability Loss Rate
Factor Factor Factor
TOTALS: 90.30 7.08 0.79 0.00 0.43
BWR 91.87 5.17 0.79 0.00 0.36
GCR 77.37 9.92 10.44 0.00 8.03
LWGR 78.16 17.29 0.27 1.15 0.05
PHWR 88.91 5.38 2.30 0.15 1.61
PWR 91.06 7.71 0.65 0.00 0.38

Figure 11: Best Quartile Performance Indicator by Reactor Type (2004-2008)

Demonstrated again in Figures 10 and 11 is that planned losses are most significant for all reactor types,
except in the case of GCRs, where unplanned losses are most significant. Forced loss rate (FLR) is also
significantly higher for this reactor type.

10
 Maintenance
with refuelling

Maintenance
Cause of Planned Loss

without refuelling

Major back-fitting
without refuelling

Refuelling without
maintenance

Major back-fitting
with refuelling

0 10 20 30 40 50 60 70 80 90 100

Percentage Best 10% All reactors

Figure 12: Planned Energy Loss Causes (2004-2008)

Figure 12 looks at planned loss in more detail. It can be seen clearly that a combined maintenance and
refuelling outage is the dominant cause, accounting for approximately 72% of this loss category.

Looking at the best performing 10% of reactors in Figure 12 we can see this trend is exaggerated further.
In best performers a combined maintenance and refuelling outage accounts for close to 89% of planned
loss. Clearly both refuelling and maintenance are essential activities in nuclear power plant operations and
the best performers will parallel plan these to make best use of any outage. Additionally, this data suggests
that shorter outages do not result in increased losses in other categories, suggesting that the quality of
the shorter outages is as good, if not better, than that of longer outages.

Minimum

Average
Best 10%

Average
All Reactors

0 10 20 30 40 50

Outage Days

Figure 13: Refuelling Outage Durations (2008)

We can see in Figure 13 that not only do the best 10% of performers use outages to perform
refuelling and maintenance but they also have less than half the average outage duration. Clearly a
short refuelling duration is a key feature of best performance.

11
 Deferred
controlled
shutdown
Unplanned Loss Outage Type
Immediate
controlled
shutdown

Extension of
planned
outage

Automatic
scram

Manual
scram

0 5 10 15 20 25 30 35 40 45

Percentage Best 10% All

Figure 14: Unplanned Energy Loss by Outage Type (2006-2008)

Figure 14 looks at unplanned losses. It is immediately clear that for ‘All’ reactors, extension of planned
outages is the primary mechanism, followed by controlled shutdowns and then scrams. In stark contrast,
extension of planned outages is the weakest mechanism for best performers unplanned loss. Controlling
the duration of an outage is a key feature of best performance in addition to a short refuelling duration.

Turbine and
Auxiliaries
Main Generator
Systems
Reactor and
Accessories
Feedwater and Main
Steam System
Electrical Power
Unplanned Energy Loss by System

Supply Systems
Reactor I&C
Systems
Reactor Cooling
Systems
Steam Generation
Systems
Safety
Systems
Reactor Auxiliary
Systems
Fuel Handling and
Storage Facilities
Circulating Water
System
Miscellaneous
Systems
All other I&C
Systems
Safety I&C Systems
(excluding reactor I&C)

0 5 10 15 20 25

Percentage Best 10% All reactors

Figure 15: Unplanned Energy Loss by System (2004-2008)

12
Figure 15 shows the causes of unplanned energy loss by system.

The distribution of systems directly involved in unplanned energy losses for best performers is markedly
different from that of for all reactors. For both groups, turbine and auxiliary systems are the biggest
contributor to loss, followed by main generator systems. Best performers do significantly better in the
category of reactor and accessories and also better in feedwater and main steam system, but less well in
electrical power supply systems.

Not enough detailed data exists to perform an adequate analysis of what is driving the problems within
each of these systems – either for all reactors or the best performers. Good quality
equipment/component failure data to identify common causes and therefore prevent them would be of
benefit to the industry. The sharing of root cause analysis information on equipment and system failures
could result in global gains.

Plant
Unplanned Energy Loss by Direct Cause

equipment
problem/failure
Nuclear
regulatory
requirements
Human
factor
related

Fire

Others

0 10 20 30 40 50 60 70 80 90

Percentage

Figure 16: Unplanned Energy Loss by Direct Cause (2004-2008)

Figure 16 shows that the direct cause of unplanned energy loss is overwhelmingly attributable to equipment
problems and failure. As with the systems analysis in Figure 15, not enough detailed data is available to analyse
this further. Additionally, while direct cause is the immediate initiator for the unplanned loss and therefore
understanding this is important, it is of limited value compared to understanding root cause, which is the
initiating cause in the chain of events leading to the unplanned loss. It is suspected that root cause analysis for
unplanned energy loss events would reveal a very significantly higher proportion of human factor-related
causes. However, this cannot be substantiated, as the information required for root cause analysis is currently
unavailable. Again, the sharing of root cause analysis information could result in global gains.

13
 3.3 AVAILABLE CAPACITY

The other element of reference unit power is available capacity (P), the indicator of which is energy
availability (EAF). EAF is made up of both generation supplied and generation available but not supplied.
The indicator that relates solely to generation supplied to the grid is the capacity factor (CF).

Best 10% 94.7 4.6

All reactors 81.3 12.7 4.86

50 60 70 80 90 100

Percentage

Availability factor (%) Planned unavailability (%) Unplanned unavailability (%) External unavailability (%)

Figure 17: Availability of Reactor Units (2006-2008)

Figure 17 shows that best performers have significantly larger planned and actual availability than others.
In other words, they maximize their availability and minimize the amount of planned and unplanned
unavailability compared to other units. For best performers then, planning for success and being able to
stick to that plan is important.

Best 10%

All reactors

50 60 70 80 90 100

Percentage

Figure 18: Recent Capacity Factors (2006-2008)

In Figure 18 the difference in achieved capacity factors between the best performing 10% of units and
the global average is clear to see.

The best capacity factor performers in the years 2006-2008 represent a range of technologies13, vendors,
regions14 and countries suggesting that performance is independent of these choices. Best performers
in these years achieved a median CF of 95.68%.

14
 Automatic Scram Rate per 7,000 Hours Critical (UA7) (%)
4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.00
0 10 20 30 40 50 60 70 80 90 100

Capacity Factor Intervals (%)

Figure 19: Average Number of Automatic Scrams for Capacity Factor Intervals (2006-2008)
Trend line inserted through data points.

The indicator Automatic Scram Rate per 7,000 Hours Critical (UA7) relates to plant safety as it provides
a measure of undesirable and unplanned thermohydraulic and reactivity transients requiring reactor
scrams. It also therefore provides an indication of how well a plant is being operated and maintained and
indeed, it is seen in Figure 19 that there is a link between plant safety and performance. A higher CF is
linked to lower numbers of automatic scram rates. Therefore work done to improve automatic scram
rates will have a benefit to the plant’s CF. Scrams are caused by a wide range of issues including
equipment problems and human performance issues as well as nuclear safety practices. Good performers
manage across these operational issues to achieve both productivity and safety.

13
PWR, BWR and PHWR all appear in the list of best performers.
14
North America, Far East Asia and Europe all appear in the list of best performers.

15
 4 Factors Affecting Capacity

The Capacity Optimization Working Group has identified the following factors as affecting capacity:

1. Outage Execution 5. Engineering

Duration Design changes
Frequency Power uprates
Scope Design upgrades
Management Design change processes
Cost (life cycle management)
Fuel
2. Equipment Reliability Design
Human performance Reliability
Lifecycle management, asset management Front and back end (limiting factor)
Predictive maintenance Environment
System redundancy Water
Component reliability Heat sink
System diagnostics Hurricane
Culture of operations Earthquake
Digital controls Grid stability
Life extension
3. Regulatory Environment
Ageing – longer term management
Licences
Thermal performance
Working regulations
Market conditions 6. Safety Performance
Baseload vs load following Scram rate
Fuel cycles (12, 18, 24 months) Radiation exposure
Greenhouse gas emissions Industrial safety
Licensing process Fuel reliability
Public relations Safety system availability
Fuel cycle Safety culture
Surveillance extensions (component level)
Outage operations requirements 7. Finances
Cost benefit
4. Organizational Factors Investment analysis
Human resource availability Operating and Maintenance (O&M) cost
Training and education requirements Capital cost
Knowledge management
Governance (centralized/decentralized) 8. Supply Chain Processes
Financial decision making – financial steering model Contract management
Partnerships and alliances
Procurement

16
5 Conclusions

1. The industry’s steady progress in raising the capacity factor has halted in the last few years.
2. Age does not have a significant effect on the capacity factor.
3. Technology choice between the predominant reactor designs does not have a significant effect on
the capacity factor.
4. Best performers achieve both higher available capacity and lower unavailable capacity than
other reactors.
5. The vast majority of loss is within plant management control.
6. Planned losses have the biggest impact (except in the case of GCRs).
7. Combined maintenance and refuelling outages are the single biggest cause of loss – this is more
pronounced for the best performers who comparatively reduce other causes of loss.
8. The biggest cause of unplanned loss is an extension to a planned outage – suggesting not only
short outages but also well-planned and executed, predictable outages are beneficial.
9. Plant equipment problems and failure is the largest direct cause of unplanned loss, with the
turbine and auxiliary system responsible for the highest proportion of unplanned losses.
10. Plants with higher capacity factors have lower numbers of automatic scrams.

These conclusions have been drawn using indicators of performance for the global fleet and the best
performing 10% of the global fleet. To draw more locally applicable conclusions would require the
sorting and analysis of the data by factors such as region and reactor type.

Further, greater detail and understanding in each area would certainly be achieved by performing case
studies and root cause analysis. Root cause analysis applied to unplanned energy losses would result in
information which could be applied to achieve gains globally. However, adequate data collection and
sharing must be undertaken by the industry in order to enable this analysis to be performed. Case studies,
which are used to share operational experience, are more appropriate for examining best practice in
managing planned energy loss.

More than a third of the current global nuclear fleet is over 30 years of age; ageing is therefore a
significant concern for the industry. It is therefore warranted that more detailed investigation(s) should
be performed to analyse for currently hidden trends of if and how ageing affects capacity. Costs related
to maintaining capacity of ageing plants will be crucial to strategic business and investment decisions.
Equally the operations of ageing reactors will need to take into account how managing plant capacity now
may affect the plant’s ability to function at the same level in the future.

While this paper has given an overview of global trends and highlighted some main issues and findings, to
determine a nuclear power plant’s optimal capacity both the boundary conditions and balance between
the factors that affect capacity need to be understood and quantified. Therefore identifying a plant’s
optimal capacity would require advanced mathematical modelling due to its complex multifactoral
dependence. However, even without this modelling tool, lessons learnt from suitable root cause analysis
and case studies can still be implemented to achieve global improvement in fleet capacity.

With the potential benefits in safety, economics, security and environmental performance available, it is
clear that further work in optimizing the current global nuclear fleet’s capacity has merit.

17
 6 Definitions

For more detailed definitions and descriptions of accepted measurement techniques for the following
values and performance indicators please refer to either:
IAEA PRIS or
World Association of Nuclear Operators Performance Indicator Programme Reference Manual

6.1 VALUES

Reference Unit Power (RUP)

The maximum power capability of the unit under reference ambient conditions. Reference ambient
conditions are environmental conditions representative of the annual mean (or typical) ambient
conditions for the unit. The reference unit power remains constant unless permanent modification or
permanent change in authorization that affects the capacity is made to the unit. [MW(e)]

Reference Energy Generation (REG)

The energy that could be produced if the unit were operated continuously at full power under reference
ambient conditions. The reference energy generation is determined by multiplying the reference unit
power by the period hours. [MW(e).h]

Available Capacity (P)

The maximum net capacity at which the unit or station is able or is authorized to be operated at a
continuous rating under the prevailing condition assuming unlimited transmission facilities. [MW(e)]

Energy Loss (Total Unavailable Capacity) (EL)

The energy which could have been produced during the reference period by the unavailable capacity. It
is comprised of PEL, UEL and XEL. [MW(e).h]

Energy Generated (Generation Supplied) (EG)

The net electrical energy supplied during the reference period as measured at the unit outlet terminals,
i.e. after deducting the electrical energy taken by unit auxiliaries and the losses in transformers that are
considered integral parts of the unit. [MW(e).h]

External Energy Losses (XEL)

The energy that was not supplied due to constraints beyond plant management control that reduced
plant availability. [MW(e).h]

Planned Energy Loss (PEL)

The energy that was not supplied during the period because of planned shutdowns or load reductions
due to causes under plant management control. Energy losses are considered to be planned if they are
scheduled at least 4 weeks in advance. [MW(e).h]

Unplanned Energy Loss (UEL)

The energy that was not supplied during the period because of unplanned shutdowns, outage extensions
or load reductions due to causes under plant management control. Energy losses are considered to be
unplanned if they are not scheduled at least 4 weeks in advance. [MW(e).h]

18
6.2 INDICATORS

Capacity Factor (CF)

The ratio of the energy which the unit produced over the period, to the reference energy generation
over the same time period.
CF (%) = (EG/REG) x 100
This indicator reflects the actual energy utilization of the unit for electricity and heat production.
(Note: this is sometimes known as Load Factor (LF))

Energy Availability Factor (EAF)

The ratio of the energy that the available capacity could have produced during this period, to the
reference energy generation over the same time period.
EAF (%) = [(REG–PEL–UEL–XEL)/REG] x 100
This indicator reflects the unit’s ability to provide energy.

Energy Unavailability Factor (EUF)

The ratio of the energy losses during the period due to unavailable capacity to the reference energy
generation over the same time period.
EUF (%) = (EL/REG) x 100 = 100–EAF = PUF+UUF+XUF
This indicator reflects all the unit’s energy losses.

Unit Capability Factor (UCF)

The ratio of the energy that the unit was capable of generating over a given time period considering only
limitations under plant management control, to the reference energy generation over the same
time period.
UCF (%) = [(REG-PEL-UEL)/REG] x 100
This indicator reflects the unit’s energy production reliability.

Planned Capability Loss Factor (PCLF)/Planned Unavailability Factor (PUF)

The ratio of the planned energy losses during a given period of time, to the reference energy generation
over the same time period.
PCLF/PUF (%) = (PEL/REG) x 100
This indicator reflects planned activities that cause energy loss such as refuelling and maintenance.

Unplanned Capability Loss Factor (UCLF) /Unplanned Unavailability Factor (UUF)

The ratio of the unplanned energy losses during a given period of time, to the reference energy
generation over the same time period.
UCLF/UUF (%) = (UEL/REG) x 100
This indicator reflects outage time and power reductions that result from unplanned equipment failures
or other conditions.

External Unavailability Factor (XUF)

The ratio of the external energy losses during a given period of time, to the reference energy generation
over the same time period.
XUF (%) = (XEL/REG) x 100 = UCF-EAF
This indicator reflects energy loss caused by events beyond plant management control.

19
 Forced Loss Rate (FLR)
The ratio of all unplanned forced energy losses during a given period of time to the reference energy
generation reduced by energy generation losses corresponding to planned outages and unplanned outage
extensions of planned outages during the same period.
FLR (%) = FEL/ [REG-(PEL+OEL)] x 100
where FEL is unplanned forced energy losses and OEL is unplanned outage extension losses.
This indicator reflects the plant's ability to maintain systems for safe electrical generation when it is
expected to be at the grid dispatcher’s disposal.

Automatic Scram Rate per 7,000 Hours Critical (UA7)

The number of unplanned automatic scrams (reactor protection system logic actuations) that occur per
7000 hours of critical operation. This indicator reflects plant safety (the number of undesirable and
unplanned thermal-hydraulic and reactivity transients requiring reactor scrams).
UA7 = (total unplanned automatic scrams while critical) / (total number of hours critical) x 7000

7 Contributors

Paul Shoemaker, AREVA

Neil Caris, AREVA
Mark Ferri, CH2MHill
Scott Lumadue, ConverDyn
Mike Montecalvo, Constellation
Steven Lau, DNMC
Sylvain Hercberg, EdF
Valery Prunier, EdF
Robert Bergkvist, GE-Hitachi Nuclear Energy
Richard Rusin, GE-Hitachi Nuclear Energy
Jiri Mandula, International Atomic Energy Agency
Paul Adler, KorteQ
Akira Nagano, Mitsubishi Heavy Industries
Anatoly Kapitanov, Rosenergoatom
Fedor Aparkin, Rosenergoatom
David Jones, Southern Nuclear Operating Company
Bob Florian, Southern Nuclear Operating Company
Jun Matsumoto, TEPCO
Ruthanne Neely, The Ux Consulting Company
Martin Luthander, Vattenfall
Rene Bastien, Westinghouse
Bill Rinkacs, Westinghouse
Rebecca Holyhead, World Nuclear Association

The Capacity Optimization Working Group wishes to express its thanks to the International Atomic
Energy Agency for access to and generous assistance with using the Power Reactor Information System.

20
The World Nuclear Association is the international private-sector
organization supporting the people, technology, and enterprises
that comprise the global nuclear energy industry.

WNA members include

the full range of enterprises involved in producing nuclear
power – from uranium miners to equipment suppliers to generators of
electricity.

With a secretariat headquartered in London, the WNA serves

as a global forum for industry experts and as an authoritative information
resource on nuclear energy worldwide.

World Nuclear Association

Carlton House • 22a St. James’s Square • London SW1Y 4JH • UK
tel: +44(0)20 7451 1520 • fax: +44(0)20 7839 1501
www.world-nuclear.org • [email protected]