System and load points reliability evaluation for electric power systems

In the solution evaluation of a composite electric power process, the optimal power flow analysis is system operating economically.. The firstperformedoverthe studiedsystem, then reliabi

2007 1st Annual IEEE Systems Conference

Waikiki Beach, Honolulu, Hawaii, USA April 9-12, 2007

SYSTEM AND LOAD POINTS RELIABILITY EVALUATION FOR

Ching-Tzong Su Senior Member IEEE Ji-Jen Wong Chi-Jen Fan Electrical Engineering Department Electrical Engineering Department

WuFeng Institute of Technology Chung Cheng University

117,Chian-Kuo Rd., Sec.2 168, University Rd.

Tel: +886-5-2792158

Abstract- This work deals with thereliability electric power systems In the solution evaluation of a composite electric power process, the optimal power flow analysis is system operating economically The firstperformedoverthe studiedsystem, then reliability evaluation and the economic the aforementioned reliability indices are

dispatch are conducted using the optimal computed based on the results from the power flow and the probabilistic method optimal power flow analysis Appropriate

Moreover, a linearprogramming method with load curtailment is performed if the load real power variables is employed for the demand is unable to be fully satisfied optimal power flow From power flow Consequently, a pertinent load shedding

solutions, the load demand satisfied and strategy is introduced Finally, the unsatisfiedcanbe calculated application and effectiveness of the

proposed method are demonstrated and The reliability evaluation in this study verifiedbyanapplication system

includes evaluating the reliability of system

and loadpoints. The basic reliabilityindices INTRODUCTION

are probability of failure and frequency of

failure, and those are used to derive many System reliability may bedefined astheability of other reliability indices for the system and the system tosupply its load taking into account load points The investigated reliability transmission constraints as well as scheduled indices are not only useful for design of and unscheduled outages of generators and flexible power supply reliability for various transmission facilities The degree of reliability customers but also beneficial to long-term asked by customers varies from one to one For system capacity expansion planning of example, the residential customer and industrial

customer ask different degree of reliability; the As mentioned above, in order to design the high-tech industrial customers require a higher flexible reliability for different types of electricity degree of reliability than the traditional industrial users, the capability of evaluating load points customers do Consequently, providing different reliability is essential

customers with different level of reliability is

necessary Moreover, in order to achieve the The solution steps of the proposed method flexible power supply reliability for different types include a dc load flow analysis for each of the

of customers, the evaluation of load points single and double contingencies, in order to reliability in the system becomes essential determine the effect of contingencies on the

system Ageneration rescheduling is conducted, The electric power system reliability evaluation after the occurrence of a contingency if the began in 60s, and the study was mainly on equipment operating constraints have been establishing a reliability model [1,2].Three violated If the generation rescheduling works, it different states- normal, fault, and switching could lead to the system normalstate; otherwise, states- for modeling components of the system further corrective action such asload curtailment

are presented [3] Amoreprecise model where a is performed to relieve the operating constraints fault is subdivided into a passive and an active Reliability indices to the system and load points state was then evaluated [4] Analytic are evaluated for all operating states of the approaches have a drawback considering powernetwork

complex situations such as restorative actions

performed recently to overcome the drawback

Monte Carlo simulation method is a pretty An acpowerflow analysis is the determination of popular method for reliability evaluation [5,6] An the bus voltage magnitude and phase angle, accurate probability model reflecting restorative generation and load at each bus in megawatts actions in reliability assessments is introduced in and megavars, flow of real and reactive powers [7] A fuzzy linear programming method, which on each transmission line, etc Power flow

can include uncertainties that exist in certain analysis is essential in planning the future variables and overcome the limitations of minor development of the system and satisfactorily

constraint violations in crisp linear programming operating the system

model, is introduced in [8] An innovative method

using genetic algorithms for the assessment of DC power flow isafurthersimplification of the ac

generation system reliability is presented in [9], powerflow It isa completely linear, non-iterative the genetic algorithm is used as a search tool to power flow algorithm It is assumed that each truncate the probability state space and to track bus voltage magnitude is 1.0 per unit The power the most probable failure states In our proposed flow on the transmission line connecting busi

work, not only the system reliability is evaluated, and bus]jis:

but also the load points reliability is investigated

where NGbNig NLb

powerflowing from bus ito busjonthe P1,i = E Li (5)

line connecting bus i and busj where

Oi phase angle at busi CRtot total generationcost

X/j : reactance of the linefrom bus i to busj pJj powergeneration ofgeneratorjat bus i

pmax pmi minimum and maximum

GENERATION generation limits for generator j at bus

i, respectively Generally, the power stations are not located at o amax

the same distance from the load centers and J i : initial line flow and transmission

they may use different fuels with different costs capacityof line i

Accordingly, there are lots of generation a (n,ij) generationshift factor ofgeneratorj

combinations for the power stations The issue at bus / on line n

of generation scheduling is an optimization k asystem state

problem, the objective is to search for a power dk time duration of state k

generation scheduling for each power station in

such a way as to minimize the operating cost Stot total numberofsystem states

subject to a few operating constraints, such as

generator capability limits, output limits of the

compensating devices, transmission lines

capability limits, power match, ,etc. The issue

of finding an optimal generation scheduling is g

called optimal dispatch of generation [10]

iJ : unit generation cost of generator] at The mathematical model for the optimal bus i

generation dispatch can beformulated as follows AT total numberof load buses

St0t NGbNig

Mi n =R Eo dki7 ij (2) L1 load demand at busi

s.t

RELIABILITY INDEX OF SYSTEM vv LdeF

NLb(9

eLi

In a practical electric power system, there are a i=l

lot of load points and each load point has a 5 SAPC/D (System Average Power distinct setof reliability indices The basic indices Curtailment / Disturbance)

arethe probability and frequency of failure at the

individual load point, however more indices can E E 60Ldef,ik Dik Fk

be defined The individual load point indices can SAPC/D= i kEs (10) also be aggregated to produce system indices YL

The indices used in this work are described as

follows [11] 6 SECI (System Energy Curtailment Index)

1 EUE (Expected Unserved Energy) L DkF

,/,/def,ikDikFk

EUE= ELdef ,ikDikFk (MWh) (6) SECI= i kes NLb()

Dik duration in hours of the load

THE SOLUTION METHOD curtailment arising due to the

outage k The selection of an appropriate outage level is

very important in the reliability evaluation of a

1> : frequency of occurrence of outage k

composite power system The computation time increases rapidly as the contingency level Lce k load curtailment at bus i due to

considered increases In order to limit the contingency k number of outage events, only the single and

s setof states with load curtailment double contingencies are taken into account

2 ELC (Expected Load Curtailed) Another important consideration in composite

electric power systems reliability evaluation is ELC= ZLdef,ikFk (7) the load curtailment in the event of a deficiency kes

in the system capacity. This curtailment of load

3 EDLC (Expected Duration of Load can be done in a lot of ways depending on the Curtailment) relative priority given to the individual load center

EDC ZDkF (8 Generally the load at each bus can be classified kes into two categories, namely, firm load and

4 SPII (System PowerlInterruption Index) curtailable load The proposed rule of load

curtailment is to interrupt the curtailable load first,

and then the firm load if necessary.

CONCLUSION The solution procedures of the proposed method

comprise 1 conducting a dc powerflow analysis A system and load points reliability evaluation

to each state of the system considering only the method for composite generation and single and doublecontingencies; 2 rescheduling transmission systems operating economically generation, i.e., conducting again a dispatch of and securely is presented In addition to system generation in case of a deficiency in system reliability, load points reliability is evaluated as

capacity; 3 conducting load curtailment if well Being aware of load points reliability helps necessary, considering the rule of load understand the reliability profile over the whole curtailment mentioned above; 4 computing the system and is essential to flexible reliability load points reliability indices- EUE, ELC, EDLC, design for different types of customers using equations (6) -(8) and the system reliability Reliability is evaluated considering only the indices- EUE, SPII, SAPC/D, SECI, using single and double contingencies in the system equations (9)-(11) In case of system capacity deficiency, unit

generation rescheduling is conducted Load

APPLICATION curtailment is performed according to a preset

rule for load curtailment if necessary The

A5-bus application system [11] shown in Fig 1 proposed method is applied to a modified

is employed for demonstrating the effectiveness five-bus application system to demonstrate its

of the proposed method The given data to this feasibility andeffectiveness Computation results application system for computing the system and of the system and load points are useful for load points reliability mainly include: expansion planning of composite generation and

1 generating unit data comprising: capacity, transmissionsystems

forced outage rate (FOR), expected failure rate

(A), expected repair rate (p), asshown in Table 1, REFERENCE

and operation costas shown in Table 2; 2 load

data comprising: MW peak load, as shown in [1] D.P Gaver, etal., "PowerSystem Reliability I-Table 3; 3 transmission line data comprising: Measures of Reliability and Methods of line resistance, reactance, susceptance, and Calculation," IEEE Trans on Power Apparatus reliability data, asshown in Table 4 and Systems, Vol 83, No 7, pp 723-737, 1964

[2] Z.G Todd, "A Probability Method for The computation results are shown in Tables 5 Transmission and Distribution Outage

Table 5 shows the outage probability and outage Calculation," IEEE Trans on Power Apparatus frequency for each load point, and the indices and Systems, Vol 83, No 7, pp 695-702, 1964 ELC, EUE, EDLC results for each load point, as [3] J Endrenyi, "Three-State Models in Power well as the indices EUE, SPII, SAPC/D, SECI for System Reliability Evaluation," IEEE Trans on the system Power Apparatus and Systems, Vol 90, No 4,

pp 1909-1916, 1971 Non-Exponential Distributions and Restorative [4] M.S Grover and R Billinton, "A Actions," IFAC Symposium on Power Plants & Computerized Approach and Switching Station Power Systems Control, Vol 1II, pp 1176-1181, Reliability Evaluation," IEEE Trans on Power 2003

Apparatus and Systems, Vol 93, No 5, pp [8] A.K Verma, et al., "Composite System 1488-1497, 1974 Reliability Assessment Using Fuzzy Linear [5] S Asgarpoor and M.J Mathine, "Reliability Programming," Electric Power Systems Evaluation of Distribution Systems with Research, Vol 73, Issue 2, pp 143-149, 2005 Non-Exponential Down Times," IEEE Trans on [9] A.A.S Nader, "Reliability Assessment of Power Systems, Vol.12, No 2, pp 579-584, Electric Power Systems Using Genetic

1997 Algorithms," Doctoral Dissertation, Texas A & M [6] G.Y Yongji, et al., "Composite System University, 2004

Reliability Evaluation Based on Monte Carlo [10] H Saadat, PowerSystem Analysis, 2ndEd., Simulation with Outages Screening," IEEE Trans McGRAW-Hill, 2004

on Power Systems, Vol 14, No 2, pp 785-790, [11 ]R Billinton and R.N Allan, Reliability

1999 Evaluation of Power Systems, Plenum Press, [7] G.W Kim and K.Y Lee, "Power System 1984

Stability Assessment Considering

Bus4

Fi.IAFiv-u ApIcainSse

61

Table 1 Capacity and Reliability for the Generating Units Plant Number of Generators Generator Capacity (MW) FOR A (flyr) p (r/yr)

Table 2 Power Production Cost for the Generating Units

Variable Cost ($IMWh)

Table 4 The Transmission Line Data

Li-3-1a |1.5 10 0.0342 0.18 0.0106 100 0.0017

Table 5 TheComputedResults

03 0.0019 2.7271 |41.5910 | 2.7271 271.9000 T16.7S30

System |EUE: 1977.0 MWh SPII: 1.335 SAPC/D: 12.19 MW/disturbance SECI: 0.00145