Optimal power flow solution using League Championship

Algorithm for Indian utility 62 bus system

by
Yeshitela Shiferaw Maru

Andhra University, College of Engineering (A)

Visakhapatnam-530003, A.P, India
2/5/2020 January 2020 1
Outlines of presentation:

1. Introduction

2. Optimal power flow formulation

3. Proposed algorithm steps

4. Result and discussion

5. Conclusions

2/5/2020 2
 1.Introduction
 The optimization can be defined as the process of finding the
conditions that provide the minimum or maximum value of a
chosen objective function.
 The optimal power flow was at first developed by Carpentier in
1962 to minimize generator cost.

 With the ever-increasing demand for electricity and the complexity

of the power system, optimal power system operation issue becomes
important.

 Meta-heuristic model is a way to solve complex optimization

problems by using simple mathematic rules and powerful
computers.

2/5/2020 3
2/5/2020 4
 1.Introduction

 The main objective of this study is to reduce fuel cost of

generation, reduce active and reactive power loss, improve
voltage deviation and enhance voltage stability index within the
given control variable constraints using LCA algorithm.

2/5/2020 5
 2.Formulation of optimal power flow

 The optimal power flow problem is mainly concerned with the

minimizations of the fuel cost of active power generations and power
losses in the power system.
 In this paper, the proposed LCA algorithm goes to minimize the
objective function shown in equation below.

𝑁𝐺
 Cost Minimize 𝐹 = ෍ 𝑎𝑖 𝑃2 𝑔𝑖 + 𝑏𝑖 𝑃𝑔𝑖 + 𝑐𝑖 $Τℎ + 𝑝𝑒𝑛𝑎𝑙𝑡𝑦
𝑖=1

𝑛𝑔
 The voltage 𝑉𝑖
𝐿𝑗 = |1 − ෎ 𝐹𝑗𝑖 ∠𝜃𝑖𝑗 + 𝛿𝑖 − 𝛿𝑗 | + 𝑝𝑒𝑛𝑎𝑙𝑡𝑦
stability index (L) 𝑉𝑗
at the load bus 𝑖=1
stated as
2/5/2020 6
 2.Formulation of optimal power flow

Real power loss minimization using equation

𝑛𝑙
 𝑃𝑙𝑜𝑠𝑠(𝑖ሻ = ෍𝑗=1 𝑔𝑖,𝑗 (𝑉𝑖2 + 𝑉𝑗2 − 2𝑉𝑖 . 𝑉𝑗 cos(𝛿𝑖 − 𝛿𝑗 ሻሻ +
𝑖≠𝑗
𝑝𝑒𝑛𝑎𝑙𝑡𝑦

𝑁𝐿
 Voltage Deviation (VD) 𝐹 = ෍ |𝑉𝑖 − 1.0| + 𝑝𝑒𝑛𝑎𝑙𝑡𝑦
𝑖=1

2/5/2020 7
 2.Formulation of optimal power flow

Equality constraints
𝑛
 𝑃𝐺𝑖 − 𝑃𝐷𝑖 − ෍ |𝑉𝑖 ||𝑉𝑗 | |𝑌𝑖𝑗 |cos 𝜃𝑖𝑗 − 𝛿𝑖 + 𝛿𝑗 = 0
𝑗=1

𝑛
 𝑄𝐺𝑖 − 𝑄𝐷𝑖 + ෍ |𝑉𝑖 ||𝑉𝑗 | |𝑌𝑖𝑗 |sin൫𝜃𝑖𝑗 − 𝛿𝑖 +
𝑗=1

Inequality constraint
min max
 Real power generation limits 𝑃𝐺𝑖 ≤ 𝑃𝐺𝑖 ≤ 𝑃𝐺𝑖

min max
 Voltage limits for generator bus 𝑉𝐺𝑖 ≤ 𝑉𝐺𝑖 ≤ 𝑉𝐺𝑖

min max
 Reactive power generation limits 𝑄𝐺𝑖 ≤ 𝑄𝐺𝑖 ≤ 𝑄𝐺𝑖
2/5/2020 8
 2.Formulation of optimal power flow

 Transmission line loading limits 𝑆𝑖 ≤ 𝑆𝑖max

 Voltage stability indices 𝐿𝑗𝑖 ≤ 𝐿𝑗𝑖max

 Transformer constraints 𝑇𝑖min ≤ 𝑇𝑖 ≤ 𝑇𝑖max

 Stability indexes are state variables which are limited

through the penalty determination method.

2/5/2020 9
 3, PROPOSED LEAGUE CHAMPIONSHIP ALGORITHM

 League Championship Algorithm (LCA) is stochastic population-

based algorithm for continuous global optimization which tries to
mimic a championship environment.
 A sports league is an organization that exists to provide a
regulated competition for a number of teams to compete in a
specific sport
The main aim of match analysis is:

 To identify strengths (S) which can then be further built upon

 To identify weaknesses (W) which suggest areas for

improvement,
 To use data to try to counter opposing strengths (threats (T)) and
exploit weaknesses (opportunities (O))
2/5/2020 10
 3, PROPOSED LEAGUE CHAMPIONSHIP ALGORITHM

League champion algorithm similarities of parameters with other

evolutionary terminology

Sporting terminology Evolutionary terminology

(LCA)
League Population
week iteration
Team i ith member in the
population
formation solution
playing strength fitness value
Number of seasons Maximum iterations
2/5/2020 11
 3, PROPOSED LEAGUE CHAMPIONSHIP ALGORITHM

Steps to implement LCA algorithm for proposed system

Step 1: Reading of input parameter (transmission line, generator and
load data)

Step 2: . Initializing league size (L), the number of seasons (S) and a
maximum number of seasons, and generate league schedule compete in
pairs for S*(L-1), where S is a number of seasons and week is denoted
by t. Also initializing team formations.
Step 3: determine the winner/loser at the first match.

Step 4:Devise a new formation for the team i for the forthcoming
match ,and evaluate the playing strength of the resulting arrangement
ln(1 − 1 − 𝑝 𝑁 ሻ𝑟
𝑐
𝑞𝑖𝑡 =
2/5/2020 ln(1 − 𝑝𝑐 ሻ 12
 3, PROPOSED LEAGUE CHAMPIONSHIP ALGORITHM

Step 5: If the new formation is the fittest one (that is, the new
solution is the best solution achieved so far for the ith member of
the population),

Step 6: Generating the initial values of the control variables

randomly to be optimized, then an objective function evaluated.
The penalty factors are added to the real equation if the constraints
are violated

Step 7: Search for the optimal set of the control variables step
2,3,4,5 and 6

2/5/2020 13
 3, PROPOSED LEAGUE CHAMPIONSHIP ALGORITHM

Step 9: Compare the objective function values for the previous and
updated solution

Step 10. Stop and report the optimal value if the maximum season
is touched, then display the results otherwise return to step 2.

2/5/2020 14
 4. Result and discussion

The proposed Indian utility 62 bus test system has the following
characteristics: 19 generators, 11 transformers with an off-nominal
tap, 2 injected VAR source, 89 number of the transmission line, and
33 PQ buses. The system base is 100MVA, 50Hz and the total load
demand is 2908MW and 1270MVAr.

Another test system for comparison of proposed LCA algorithm is

IEEE-30 bus test system, has been the following characteristics: 6
generators, 4 transformer, 9 shunt VAR compensation, 41 branches,
and 24 PQ bus. The system has 100MVA, 60Hz, and

2/5/2020 15
 4. Result and discussion

The proposed technique has been applied to solve the OPF

problem for the following cases:

Case 1: Minimization of generation fuel cost

Case 2: Voltage profile improvement with fuel cost minimization

Case 3: Voltage stability enhancement with fuel cost minimization

Case 4: Minimization of real power loss

Case 5: Minimization of reactive power loss

2/5/2020 16
 Optimal setting of control variable IEEE-30 bus using proposed
LCA
Control Mini Maxim Case 1 Case 2 Case 3 Case 4 Case 5
variable mum um
value value
Fuel cost - -
($/hr.) 799.1087 803.2752 800.0185 964.1988 965.3353
Real power - - 8.61 9.8557 8.6701 2.9104 3.1163
loss(MW)

Reactive power - - 4.3327 6.9673 - -

loss(MVar) 10.8515 17.1466 24.0322
∑Voltage - - 1.7318 0.1062 1.9654 1.9691 0.9557
deviation(pu)

Lmax (pu) - - 0.1183 0.1366 0.1143 0.1168 0.1297

2/5/2020 17
 Optimal setting of control variable Indian utility 62 bus using
proposed LCA
Control Mini Maximu Case 1 Case 2 Case 3 Case 4 Case 5
variable mum m value
value
Fuel cost - - 9106.53 9881.88 9017.588 9976.189 10148.28
($/hr.)

Real power - - 84.8563 165.964 105.507 68.6251 71.0714

loss(MW)

Reactive - - -484.232 -95.889 -427.648 -570.693 -566.566

power
loss(MVar)
∑Voltage - - 1.4424 0.9922 1.3341 1.48381 1.5987
deviation(p
u)
Lmax (pu) - - 0.1318 0.0930 0.0852 0.09475 0.108786

2/5/2020 18
  Fuel cost variation for case 1
2/5/2020 19
2/5/2020
 Variation of fuel cost, and voltage 20
deviation for case 2
  Variation of fuel cost , and stability index for case
3
2/5/2020 21
 Variation of real power for case 4
2/5/2020 22
 Variation of reactive power for case 5

2/5/2020 23
Comparison of LCA and other algorithm reported in the literature
for OPF of IEEE-30 bus system

Algorith Fuel cost VD minimization L-index minimization Power

m minimizati along with fuel cost) along with fuel cost loss
on ($/h) Fuel cost VD(pu) Fuel cost Lmax(pu minimizati
($/h) ($/h) ) on
799.116 804.99 0.1020 - - 8.63
[19](BBO)
[25] (DE) 799.289 805.26 0.1357 - 0.1219 8.615
[20] 800.41 806.38 0.0891 - 0.1246 -
(PSO)
[16] 800.413 803.7198 0.09454 800.4 0.13745 9.01
(DE,CH)
800.5099 803.3125 0.10842 801.2248 0.13713 9.0345
[18](MSA
)
Proposed 799.108 803.2752 0.1062 800.0185 0.1143 8.61
(LCA)
2/5/2020 24
 . For Indian utility 62 bus test system, there is no much work
done in the literature.

only in [26], and [27] explain the total fuel cost of minimization

ALGORITHM FUEL COST OF

GENERATOR
($/h)
[26] PSO 10328.9
[27] EP 11406
PROPOSED LCA 9106.51

2/5/2020 25
 5. Conclusions

The following conclusions are drawn from the work carried

out:-.

 The optimal power flow is mathematically formulated in the

power system network with equality and inequality constraints.

 Voltage stability index is formulated.

 Initialization of League Championshiop Algorithm was done

and incorporated with Newton Raphson load flow equations.

2/5/2020 26
 5. Conclusions
 The proposed algorithm result was compared with those
reported in the literature in terms of the objective function such
as minimization of fuel cost, fuel cost along with voltage
deviation improvement, fuel cost along with voltage stability
enhancement, and minimization of real & reactive power

 Hence, LCA algorithm is a powerful method to solve optimal

power flow on standard IEEE-30 bus test system, and Indian
utility 62 bus system because it could not have more internal
tuning parameter as compared to the other algorithm mentioned
in the literature which confirms its effectiveness.

2/5/2020 27
 Thank You !

2/5/2020 28