Life Science Journal 2012;9(4) http://www.lifesciencesite.

com

Calculation of generation system reliability index: Loss of Load Probability

Hasan Fayazi Boroujeni 1, Meysam Eghtedari 2, Mostafa Abdollahi 3, Elahe Behzadipour 4

1,2,3,4
Department of Electrical Engineering, Boroujen Branch, Islamic Azad University, Boroujen, Iran
[email protected]

Abstract: Generation system reliability is an important factor in the long term planning for future system capacity
expansion to make sure that the total installed capacity is sufficient to support demand. The planning process utilizes
reliability indices as criteria to decide on new investments in new generation capacities. Generation system
reliability is evaluated by using different indexes. In this paper, loss load of probability (LOLP) is simulated to
evaluate the system reliability. Effects of the system parameters such as forced outage rate (FOR) are tested on the
LOLP index.
[Hasan Fayazi Boroujeni, Meysam Eghtedari, Mostafa Abdollahi, Elahe Behzadipour. Calculation of generation
system reliability index: Loss of Load Probability. Life Sci J 2012;9(4):4903-4908] (ISSN:1097-8135).
http://www.lifesciencesite.com. 736

Keywords: Generation System Reliability, Loss of Load Probability, Capacity Outage Probability Table,
Analytically Method.

1. Introduction and short-term evaluations to assist in day to day

Electricity has been the driving force for operating decisions. In short, these reliability indices
economies of the world and provides day-to-day (for long-term evaluations) are used by system
necessity for the population in the world. The planners and the authorities to decide on and advice for
generation, transmission and retailing of electricity new investments in building new generation capacities
have existed hundreds of years in providing the much [1].
needed electricity. Due to the nature of electricity Generation system reliability is an important
systems, the variable demand at every moment needs aspect in the planning for future system capacity
to be met by consistent electricity supply in making expansion. It provides a measurement of reliability or
sure the continuous availability of the resources. Not adequacy to make sure that the total generation system
meeting the demand in any case will lead to a huge capacity is sufficient to provide adequate electricity
loss of income to the generators as well as to the when needed [1].
consumers. The reliability of the generation, In this paper an important reliability index
transmission and distribution of electricity in this sense LOLP is evaluated for generation system. The
is crucial for the continuous supply of electricity to proposed index is simulated by using analytically
meet the demand. method. Effects of changing system parameters such as
A modern power system is complex, highly FOR are tested on the LOLP.
integrated and very large. Fortunately, the system can 2. Generation system reliability
be divided into appropriate subsystems or functional Reliability has been and always is one of the
areas that can be analyzed separately [1]. These major factors in the planning, design, operation, and
functional areas are generation, transmission and maintenance of electric power system. Generation
distribution. The function of the generation system is system reliability focuses on the reliability of
to make sure enough capacity is available to meet the generators in the whole electric power system where
load/demand at any time. Transmission and electric power is produced from the conversion process
distribution systems need to be reliable in making sure of primary energy (fuel) to electricity before
the electricity generator can be delivered to the transmission. The generation system is an important
consumers. System planners have been assigned the part of the electricity supply chain and it is crucial that
role of planning for forecasting the load into the future enough electricity is generated at every moment to
and plant capacity addition to meet the load and meet the demand. Generating units will occasionally
provide a level of reliability in case some of the plants fail to operate and the system operator has to make
are out on maintenance or breakdown. Probabilistic sure that enough reserve is available to be operated
method is often used to determine the system when this situation happens [2-31].
reliability and the system reliability can be summed up Reliability of the generation system is divided
into a single value, the reliability indices. Reliability into adequacy and security [32]. System adequacy
studies are conducted for two purposes. Long-term relates to the existence of sufficient generators within
evaluations are performed to assist in system planning

http://www.lifesciencesite.com 4903 [email protected]

 Life Science Journal 2012;9(4) http://www.lifesciencesite.com

the system to satisfy the consumer load demand or 3. Load model

system operational constraints. System adequacy is The load in a power system in any time period
associated with static conditions of the system and do is a stochastic process, which is difficult to describe
not include system disturbances. System security on with a simple mathematical formula. Different models
the other hand relates to the ability of the system to are created, starting from rimary load data and
respond to disturbances arising within the system. according to the need to calculate reliability. Primary
Therefore system security is associated with the load data will provide a minimum amount of data that
response of the system to whatever perturbation it is is needed to establish an hourly chronological load
subjected to. In this study, the reliability evaluations profile. Most primary load data consist of the
will be focused on the generation system adequacy and percentage of maximum monthly load or weekly load
will not take into account system security. in a year, the load in 24 hours in a typical day in each
The basic modeling approach for the season and the maximum load in each day in a week.
generating system adequacy assessment consists of With the percentages of these data available and the
three parts as shown in Figure 1. The generation and annual peak load known, the hourly chronological load
load models are convolved to form an appropriate risk profile can be established.
model where the element of interest is the risk of
generation capacity less than the load. In short, 4. Forced Outage Rate
adequacy evaluation of generation systems consists of There are many concepts in reliability
four general steps as Figure 1. evaluation such as: failure rate, repair time,
(i) Create a generation capacity model; (ii) unavailability, forced outage rate (FOR) and etc. Unit
create a load model; (iii) combined the generation unavailability is also known conventionally as “forced
capacity model with load model to obtain a risk model outage rate” (FOR), although the value is not a rate.
and (iv) calculating indexes. The FOR is defined as below.

Forced outage hours

FOR  (1)
In service hours  Forced outage hours

The FOR is calculated for a long period of

time (e.g. 365 days), is the same index as the
unavailability.
5. Generation system reliability indices
The quantification of reliability is an
important aspect of generation system reliability
assessment. The measurement used to quantify
Figure 1. Generation reliability evaluation process reliability of a generation system is given various
reliability indices. These reliability indices are used to
Analytical methods or Monte Carlo assess the reliability performance of a generation
simulation [33] can be used to calculate the reliability system against some predetermined minimum
indices. Analytical techniques represent the system by requirements or reliability standards, compare
analytical models and evaluate the indices from these alternative designs, identify weak spots and determine
models using mathematical solutions. Monte Carlo ways for correction in the generation system and to be
simulations, on the other hand estimate the indices by integrated with costs and performance considerations
simulating the actual process and random behavior of for decision making. These indices are better
the system, treating the problem as a series of understood as estimates of system-wide generation
experiments. The reliability indices obtained indicate adequacy and not as absolute measures of system
the ability of the generating facilities to meet the reliability [18].
system demand. Basically, system reliability evaluations can
In the analytical method, the generating be divided into deterministic and probabilistic. The
system model used for generation capacity adequacy most common deterministic indices are the Reserve
assessment is a Capacity Outage Probability Table Margin and the largest set in the system. An important
(COPT) which can be created using the recursive shortcoming of these methods is that they do not
technique. As for the load model, the daily peak load account for the stochastic nature of system behavior.
or hourly load for a period of one year is normally used Probabilistic methods can provide more
to form the Load Probability Table (LPT). meaningful information to be used in design and
resource in planning and allocation. There are two

http://www.lifesciencesite.com 4904 [email protected]

 Life Science Journal 2012;9(4) http://www.lifesciencesite.com

approaches that use probabilistic evaluation. The (LOEP), Loss of Energy Expectation (LOEE),
analytical methods and Monte Carlo simulation as can Expected Energy Not Served (EENS), and Loss of
be seen from Figure 2. The analytical methods Load Frequency (LOLF) and Loss of Load Duration
represent the system by mathematical models and use (LOLD). Most of these indices are basically expected
direct analytical solutions to evaluate reliability indices values of a random variable. Expectation indices
from the model. As for the Monte Carlo simulation, provide valid adequacy indicators which reflect
reliability indices are estimated by simulating the various factors such as system component availability
actual random behavior of the system. So of the and capacity, load characteristics and uncertainty,
commonly used probabilistic reliability indices are system configurations and operational conditions, etc
Loss of Load Probability (LOLP), Loss of Load [1]. Typical reliability indices used in power system
Expectation (LOLE), Loss of Energy Probability evaluations and their categorizing is shown in Figure 2.

Figure 2. Generation system reliability assessment indices category

6. Loss of load probability calculation is being done. The mathematical formula

Loss of load occurs when the system load for calculation of LOLP is shown as below [1].
exceeds the generating capacity available for use.
Loss of Load Probability (LOLP) is a projected value Pj.t j
of how much time, in the long run, the load on a    
LOLP j p CA  Cj .P L  Cj  j
100
(2)
power system is expected to be greater than the
where;
capacity of the available generating resources. It is P: the probability; L: expected load; CA:
defined as the probability of the system load available generation capacity; Cj: remaining
exceeding the available generating capacity under the generation capacity; pj: probability of capacity
assumption that the peak load of each day lasts all
outage; tj: percentage of time when the load exceeds
day [1].
Cj.
LOLP is based on combining the probability
Alternatively, a load duration curve consists
of generation capacity states with the daily peak
of daily peak loads arranged in descending order can
probability so as to assess the number of days during
be used to measure LOLP for long term generation
the year in which the generation system may be
capacity evaluation. The assumption used in this case
unable to meet the daily peak. LOLP can be
is that the peak load of the day would last all day.
calculated considering the daily peak loads for one
The LOLP calculation is illustrated with a daily peak
year duration or sometimes on each hour’s load for a load curve in Figure 3 [1].
24 hours day. Therefore, the same system may have
two or more values of LOLP depending on how the

http://www.lifesciencesite.com 4905 [email protected]

 Life Science Journal 2012;9(4) http://www.lifesciencesite.com

In this section LOLP index is calculated for

the proposed test system. The procedure presented in
section 6 is used to computing LOLP. In first the
Capacity Outage Probability Table (COPT) is
derived. Table 2 shows the COPT and the probability
of different outages is listed.

Table 2. COPT for the test system

Capacity Outage Probability
(MW)
0 0.88565791683600000
Figure 3. LOLP Calculation using Load Duration 25 0.05478296392800000
Curve 40 0.03614930272800000
50 0.00894603956400000
In the figure, the parameters are defined as 50 0.00084715923600000
following; Oj: is the magnitude of the jth outage in the 65 0.00223603934400000
system; pj: is the probability of a capacity outage of 75 0.00055336327200000
magnitude Oj; tj: is the number of days that an outage 80 0.00036887043600000
of magnitude Oj would cause a loss of load in the 90 0.00036514447200000
system. 90 3.45779280000000e-05
Capacity outage less than the amount of 100 0.00894603956400000
reserves will not contribute to a loss of load. When a 100 8.55716400000000e-06
particular capacity outage is greater than the reserve, 105 2.28167280000000e-05
115 2.25862560000000e-05
the risk in this case will be pj ⨯ tj.
125 0.00055336327200000
7. Case study
130 3.72596400000000e-06
In this section a numerical case study is 130 3.52836000000000e-07
carried out for reliability evaluation. Table 1 shows 140 0.00036514447200000
the proposed generation test system. This system 140 3.49272000000000e-07
contains four generation companies with six units. 150 9.03640360000000e-05
The system data and capacity of units are considered 150 8.55716400000000e-06
as typical. The load model is also considered as 155 2.30472000000000e-07
Figure 4. 165 2.25862560000000e-05
175 5.58952800000000e-06
Table 1. Generation system details 180 3.72596400000000e-06
Generation Number of Capacity of each
FOR
180 3.56400000000000e-09
Company units unit (MW) 190 3.68832800000000e-06
1 2 25 0.03 190 3.49272000000000e-07
2 2 40 0.02
200 8.64360000000000e-08
3 1 50 0.01
4 1 100 0.01
205 2.30472000000000e-07
215 2.28144000000000e-07
230 3.76360000000000e-08
230 3.56400000000000e-09
240 3.52800000000000e-09
255 2.32800000000000e-09
280 3.60000000000000e-11
Sum of probabilities=1

LOLP index is calculated as mentioned

above. In this regard, the LOLP is obtained as below.
LOLP 5.968138239671235e 004 (3)
In order to show the sensitivity of LOLP
index to the system parameters, an evaluation is
carried out and the results are listed in Table 3. It is
seen that changing FORs and load has a direct effect
Figure 4. Daily peak demand of year of the reliability of generation system.

http://www.lifesciencesite.com 4906 [email protected]

 Life Science Journal 2012;9(4) http://www.lifesciencesite.com

6. Conclusions [6] R. Billinton, T.K.P. Medicherla, Effect of system

In this paper a commonly used reliability peak load forecast uncertainty on annualized
index of generation system LOLP was successfully reliability indices of composite generation and
calculated and evaluated. Different conditions and transmission systems, Microelectronics
changing were considered. COPT was carried out and Reliability, 20 (1980) 91-100.
then the reliability calculated. Simulation results [7] R. Billinton, X. Tang, Selected considerations in
showed that changing components FOR and load utilizing Monte Carlo simulation in
level can directly affect of the system total reliability. quantitative reliability evaluation of composite
power systems, Electric Power Systems
Table 3. Effect of changing parameters on the LOLP Research, 69 (2004) 205-211.
index [8] C.L.T. Borges, An overview of reliability models
Parameter changing LOLP and methods for distribution systems with
FOR unit 25 renewable energy distributed generation,
5.043705938958903e-004
MW=0.01
FOR unit 25
Renewable and Sustainable Energy Reviews,
6.889275761643836e-004 16 (2012) 4008-4015.
MW=0.05
FOR unit 40
7.691760586986303e-004
[9] C.L.T. Borges, D.M. Falcão, Optimal distributed
MW=0.05 generation allocation for reliability, losses, and
FOR unit 50 voltage improvement, International Journal of
7.519234803835614e-004
MW=0.05
FOR unit 100 Electrical Power & Energy Systems, 28
0.0058208871141920000 (2006) 413-420.
MW=0.1
Increasing load by
8.500825094465753e-004
[10] H.-C. Chen, Optimum capacity determination of
10% in all levels stand-alone hybrid generation system
decreasing load by
10% in all levels
1.611630195397260e-004 considering cost and reliability, Applied
Energy.
Corresponding Author: [11] D.P. Clancy, G. Gross, F.F. Wu, Probabilitic
Hasan Fayazi Boroujeni flows for reliability evaluation of multiarea
Department of Electrical Engineering, power system interconnections, International
Boroujen Branch, Islamic Azad University, Journal of Electrical Power & Energy
Boroujen, Iran. Systems, 5 (1983) 101-114.
E-mail: [email protected] [12] S. Fockens, A.J.M. van Wijk, W.C. Turkenburg,
C. Singh, Reliability analysis of generating
References systems including intermittent sources,
[1] R. Billington, R.J. Ringlee, A.J. Wood, Power- International Journal of Electrical Power
system reliability calculations, 1973. & Energy Systems, 14 (1992) 2-8.
[2] R. Arya, S.C. Choube, L.D. Arya, Reliability [13] S. Fockens, A.J.M. van Wijk, W.C. Turkenburg,
evaluation and enhancement of distribution C. Singh, Application of energy based indices
systems in the presence of distributed in generating system reliability analysis,
generation based on standby mode, International Journal of Electrical Power
International Journal of Electrical Power & Energy Systems, 16 (1994) 311-319.
& Energy Systems, 43 (2012) 607-616. [14] H. Haroonabadi, M.R. Haghifam, Generation
[3] A.K. Azad, R.B. Misra, A new approach to reliability assessment in power markets using
evaluate reliability and production cost of a Monte Carlo simulation and soft computing,
power generating system, Computers & Applied Soft Computing, 11 (2011) 5292-
Electrical Engineering, 22 (1996) 343-351. 5298.
[4] B. Banerjee, S.M. Islam, Reliability based [15] Y.-Y. Hong, L.-H. Lee, Reliability assessment of
optimum location of distributed generation, generation and transmission systems using
International Journal of Electrical Power fault-tree analysis, Energy Conversion and
& Energy Systems, 33 (2011) 1470-1478. Management, 50 (2009) 2810-2817.
[5] J. Batut, J. Boisseau, E. Bourgade, M. Monti, [16] A. Jain, S.C. Tripathy, R. Balasubramanian,
Different approaches to reliability as applied to Reliability and economic analysis of a power
the generation — transmission system and its generation system including a photovoltaic
components, International Journal of Electrical system, Energy Conversion and Management,
Power & Energy Systems, 10 (1988) 168- 36 (1995) 183-189.
173. [17] D. Kottick, M. Blau, Y. Frank, Reliability of
supply to consumers as a function of the

http://www.lifesciencesite.com 4907 [email protected]

 Life Science Journal 2012;9(4) http://www.lifesciencesite.com

installed generation reserve margin, Electric Hourly Wind Speed Probability Model, Energy
Power Systems Research, 33 (1995) 63-67. Procedia, 12 (2011) 819-827.
[18] S.R. Lakshmi, S.C. Tripathy, K.S.P. Rao, R. [26] C. Singh, N.V. Gubbala, An alternative
Balasubramianian, Evaluation of generation approach to rounding off generation models in
system reliability indices by fast transform power system reliability evaluation, Electric
techniques, International Journal of Electrical Power Systems Research, 36 (1996) 37-44.
Power & Energy Systems, 17 (1995) 281- [27] C. Singh, N.V. Gubbala, Comparison of unit
287. addition and fast Fourier transform methods
[19] T. Lekane, J. Gheury, J. Van Hecke, Assessing for generation system reliability evaluation,
the impact of operating constraints and International Journal of Electrical Power
operating policy on the reliability level of a & Energy Systems, 18 (1996) 203-205.
generation system, International Transactions [28] R. Sundaramurthi, C. Smidts, Human reliability
in Operational Research, 3 (1996) 271-279. modeling for the Next Generation System
[20] Y.-F. Li, E. Zio, A multi-state model for the Code, Annals of Nuclear Energy.
reliability assessment of a distributed [29] A.K. Verma, A. Srividya, B.C. Deka, Impact of
generation system via universal generating a FACTS controller on reliability of composite
function, Reliability Engineering & power generation and transmission system,
System Safety, 106 (2012) 28-36. Electric Power Systems Research, 72 (2004)
[21] S. Narasimhan, S. Asgarpoor, A fuzzy-based 125-130.
approach for generation system reliability [30] Z.A. Yamayee, F.S. Ma, Effect of size and
evaluation, Electric Power Systems Research, location of conventional and intermittent
53 (2000) 133-138. generation on system reliability, International
[22] P. Pelacchi, D. Poli, The influence of wind Journal of Electrical Power & Energy
generation on power system reliability and the Systems, 5 (1983) 94-100.
possible use of hydrogen storages, Electric [31] X. Zhang, Z. Bie, G. Li, Reliability Assessment
Power Systems Research, 80 (2010) 249-255. of Distribution Networks with Distributed
[23] M.V.F. Pereira, S.H.F. Cunha, G.C. Oliveira, Generations using Monte Carlo Method,
Application of reliability evaluation methods Energy Procedia, 12 (2011) 278-286.
to the planning of large hydroelectric systems, [32] R. Billinton, E. Khan, A security based approach
International Journal of Electrical Power to composite power system reliability
& Energy Systems, 10 (1988) 142-150. evaluation, Power Systems, IEEE Transactions
[24] O. Salmela, The effect of introducing increased- on, 7 (1992) 65-72.
reliability-risk electronic components into 3rd [33] O. Alizadeh Mousavi, R. Cherkaoui, M. Bozorg,
generation telecommunications systems, Blackouts risk evaluation by Monte Carlo
Reliability Engineering & System Safety, Simulation regarding cascading outages and
89 (2005) 208-218. system frequency deviation, Electric Power
[25] W.-H. Shi, J. Chen, Reliability Assessment of Systems Research, 89 (2012) 157-164.
Interconnected Generation Systems Based on

09/04/2012

http://www.lifesciencesite.com 4908 [email protected]