Академический Документы
Профессиональный Документы
Культура Документы
Nomenclature
1 Introduction
IDMT
T
IF
TDS
PS
CT
CTpr-rating
Irelay
OF
TDSmin
TDSmax
Tmin
Tmax
CTI
Tpri-cl-in
Tpri-far-bus
Tprimary
Tbackup
E-ISSN: 2224-350X
240
2 Optimal
Problem
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
Optimization Algorithm
Evolutionary Algorithm (EA)
Differential Evolution Algorithm (DEA)
Modified Differential Evolution Algorithm
Self-Adaptive Differential Evolutionary
Particle Swarm Optimization (PSO)
Modified Particle Swarm Optimization
Evolutionary Particle Swarm Optimization
Box-Muller Harmony Search
Zero-one Integer Programming
Covariance Matrix Adaptation Evolution
Strategy
Seeker Algorithm
Teaching Learning Based Optimization
Chaotic Differential Evolution Algorithm
Informative Differential Evolution Algorithm
Firefly Optimization Algorithm
Krill Herd Algorithm
Non-dominated Sorting Genetic Algorithm
Biogeography Based Optimization
E-ISSN: 2224-350X
Coordination
TABLE I
Literature for optimisation algorithm
Ref.
[2]
[3]
[4]
[5]
[6]
[7-8]
[9]
[10]
[11]
[12]
Relay
T=
TDS
IF
PS
CT
pr rating
(1)
241
I relay =
IF
(2)
CTpr rating
Ti min Ti Ti max
N cl
(3)
where,
i
Tpri
cl in =
0.14 TDS i
Ii
i F i
PS CTpr rating
(4)
0.02
0.14 TDS
Ij
j F j
PS CTpr rating
i
=
Tbackup
(5)
0.02
E-ISSN: 2224-350X
I
x
i
PS CTpr rating
i
F
(10)
0.02
0.14 TDS y
I
y
i
PS CTpr rating
i
F
(11)
0.02
(6)
0.14 TDS x
i
Tprimary
=
2.2 Constraints
i
i
TDS min
TDS i TDS max
(9)
N far
i
pri cl in
=i 1 =j 1
(8)
(7)
242
1
M
E
j =1
(13)
( Ab )ij =
1
1 + E (2)
(14)
( Ag )i =
1
1 + OPTi
(15)
E-ISSN: 2224-350X
(12)
i =1
243
3.2 Overview
of
Particle
Optimization (PSO) Algorithm [26]
Swarm
Particle Swarm Optimization (PSO) is a populationbased evolutionary technique which has a number of
key advantages over other optimization techniques.
PSO finds the optimal solution using a population of
particles. Each particle represents a candidate
solution to the problem. PSO is basically developed
through simulation of bird flocking in two
dimensional spaces [27]. Attractive features of the
PSO include ease of implementation, the fact that no
gradient information is required and its application
can be extended in neural network training and
minimizing function. The PSO is presented by [2632] as follows:
Step 1: Each individual particle has the following
properties: a current position in search space xi, a
current velocity Vi, and a personal best position
Xpbest,
Step 2: The personal best position Xpbest,
corresponds to the position in search space, where
particle i presents the smallest error as determined
by the objective function f, assuming a minimization
task,
Step 3: The global best position denoted by Xgbest,
represents the position yielding the lowest error
among all the Xpbests.
Consider a swarm of P particles; with each
particles position representing possible solution
point in the design problem space. For each particle,
the authors in [29-31] proposed that its position xi is
updated in the following manner:
Vi ( t + 1) = W + Vi ( t ) + c1 r1.[ Pbest (t ) xi (t ) ]
+c2 r2 .[Gbest (t ) xi (t ) ]
(16)
and,
xi ( t + 1=
) xi ( t ) + Vi ( t + 1)
(17)
E-ISSN: 2224-350X
244
information picked-up from a persons preknowledge about the problem to be solved, are
regarded as bacteria used to change a certain
integrant of the particle, aimed at guiding the search
process. However, the post-vaccinal particle must be
checked by immune selection, which is capable of
suppressing the degradation phenomena. If the
fitness of the post-vaccinal particle is smaller than
the original one, the original one will be preserved;
otherwise, the post-vaccinal particle will be
regarded as the new particle and replace the original
particle. Therefore, the optimized particle swarm is
undergoes immunization and hence, a new particle
swarm is generated,
Step 6: The newly generated particle at Step 5
above is returned to Step 2 and calculations are
repeated until the optimal solution is found or the
maximum iterative number is reached.
(b)
Fig. 3. Case study systems:
(a) IEEE 4-bus, (b) IEEE 6-bus.
4 Case Study
The optimization algorithms IA, PSO and IA-PSO
are validated and tested on two systems, namely
IEEE 4-bus and IEEE 6-bus models as shown in
Figures 3.a and 3.b, respectively.
The first case study consists of two power
generators, four lines and eight IDMT directional
overcurrent relays. The objective of the optimization
problem in this case is to coordinate the settings of
eight relays. Accordingly, there are 16 decision
variables which are TDS1 to TDS8 and PS1 to PS8.
The second case study consists of three power
generators, seven lines and fourteen IDMT
directional overcurrent relays. The objective of the
optimization problem in this case is to coordinate
the settings of fourteen relays. Accordingly, there
are 28 decision variables which are TDS1 to TDS14
and PS1 to PS14.
CTI is selected to take the value of 0.30 sec in
each of the studied cases.
TABLE II
IF and CTpr-rating for T ipri_cl_in and Tjpri_far_bus in case study:
(a) IEEE 4-bus, (b) IEEE 6-bus.
(a)
T ipri_cl_in
IiF
TDS1
TDS2
TDS3
TDS4
TDS5
TDS6
TDS7
TDS8
20.32
88.85
13.60
116.81
116.70
16.67
71.70
19.27
CTipr-rating TDSj
TDS2
TDS1
TDS4
TDS3
TDS6
TDS5
TDS8
TDS7
0.4800
0.4800
1.1789
1.1789
1.5259
1.5259
1.2018
1.2018
IjF
CTjpr-rating
23.75
12.48
31.92
10.38
12.07
31.92
11.00
18.91
0.4800
0.4800
1.1789
1.1789
1.5259
1.5259
1.2018
1.2018
(b)
T ipri_cl_in
(a)
E-ISSN: 2224-350X
TDSi
T jpri_far_bus
245
TDSi
IiF
TDS1
TDS2
TDS3
TDS4
TDS5
TDS6
TDS7
TDS8
TDS9
TDS10
TDS11
TDS12
TDS13
TDS14
2.5311
2.7376
2.9723
4.1477
1.9545
2.7678
3.8423
5.6180
4.6538
3.5261
2.5840
3.8006
2.4143
5.3541
T jpri_far_bus
CTipr-rating TDSj
0.2585
0.2585
0.4863
0.4863
0.7138
0.7138
1.7460
1.7460
1.0424
1.0424
0.7729
0.7729
0.5879
0.5879
TDS2
TDS1
TDS4
TDS3
TDS6
TDS5
TDS1
TDS2
TDS3
TDS4
TDS5
TDS6
TDS1
TDS2
IjF
CTjpr-rating
5.9495
5.3752
6.6641
4.5897
6.2345
4.2573
6.3694
4.1783
3.8700
5.2696
6.1144
3.9005
2.9011
4.3350
0.2585
0.2585
0.4863
0.4863
0.7138
0.7138
1.7460
1.7460
1.0424
1.0424
0.7729
0.7729
0.5879
0.5879
TABLE III
IF and CTpr-rating for Txbackup and T yprimary in case study:
(a) IEEE 4-bus, (b) IEEE 6-bus.
(a)
T xbackup
Relay
No.
5
5
7
7
1
2
2
4
4
T ypimary
IFi
CTipr-rating
20.32
12.48
13.61
10.38
116.81
12.07
16.67
11.00
19.27
1.5259
1.5259
1.2018
1.2018
0.4800
0.4800
0.4800
1.1789
1.1789
Relay
No.
1
1
3
3
4
6
6
8
8
IFj
CTjpr-rating
20.32
12.48
13.61
10.38
116.81
12.07
16.67
11.00
19.27
0.4800
0.4800
1.1789
1.1789
1.1789
1.5259
1.5259
1.2018
1.2018
(b)
T xbackup
Relay
No.
8
11
8
3
3
10
10
13
1
1
12
12
14
3
3
11
2
11
2
13
4
13
4
14
6
14
6
8
2
8
2
12
6
12
10
4
10
4
Typrimary
IFi
CTipr-rating
4.0909
1.2886
2.9323
0.6213
1.6658
0.0923
2.5610
1.4995
0.8869
1.5243
2.5444
1.4549
1.7142
1.4658
1.1231
2.1436
2.0355
1.9712
1.8718
1.8321
3.4386
1.6180
3.0368
2.0871
1.8138
1.4744
1.1099
3.3286
0.4734
4.5736
1.5432
2.7269
1.6085
1.8360
2.0260
0.8757
2.7784
2.5823
1.7460
0.7729
1.7460
0.4863
0.4863
1.0424
1.0424
0.5879
0.2585
0.2585
0.7729
0.7729
0.5879
0.4863
0.4863
0.7729
0.2585
0.7729
0.2585
0.5879
0.4863
0.5879
0.4863
0.5879
0.7138
0.5879
0.7138
1.7460
0.2585
1.7460
0.2585
0.7729
0.7138
0.7729
1.0424
0.4863
1.0424
0.4863
E-ISSN: 2224-350X
Relay
No.
1
1
1
2
2
3
3
3
4
4
5
5
5
6
6
7
7
7
7
9
9
9
9
11
11
11
11
12
12
12
12
13
13
13
14
14
14
14
IFj
CTjpr-rating
5.3752
5.3752
2.5311
2.7376
5.9495
4.5897
2.9723
4.5897
4.1477
6.6641
4.2573
1.9545
4.2573
6.2345
6.2345
4.1783
4.1783
3.8423
3.8423
5.2696
5.2696
4.6538
4.6538
3.9005
3.9005
2.5840
2.5840
3.8006
3.8006
6.1144
6.1144
4.3350
4.3350
2.4143
2.9011
2.9011
5.3541
5.3541
0.2585
0.2585
0.2585
0.2585
0.2585
0.4863
0.4863
0.4863
0.4863
0.4863
0.7138
0.7138
0.7138
0.7138
0.7138
1.7460
1.7460
1.7460
1.7460
1.0424
1.0424
1.0424
1.0424
0.7729
0.7729
0.7729
0.7729
0.7729
0.7729
0.7729
0.7729
0.5879
0.5879
0.5879
0.5879
0.5879
0.5879
0.5879
(a)
(b)
Fig. 4. Convergence characteristics of IA, PSO and
IA-PSO in case study: (a) IEEE 4-bus, (b) IEEE 6-bus.
246
TABLE IV
Optimal relays settings: (a) IEEE 4-bus, (b) IEEE 6-bus.
(a)
Relay
No.
TDS
1
PS
TDS
2
PS
TDS
3
PS
TDS
4
PS
TDS
5
PS
TDS
6
PS
TDS
7
PS
TDS
8
PS
IA
PSO
IA-PSO
0.0511
1.2922
0.2148
1.4243
0.0504
1.2645
0.1536
1.5267
0.1279
1.5893
0.0503
1.6961
0.1346
1.5765
0.0506
1.3569
0.0528
1.3533
0.2235
1.5999
0.0529
1.3263
0.1602
1.5913
0.1339
1.5914
0.0526
1.3276
0.1411
1.4922
0.0543
1.3276
0.0560
1.4583
0.2415
1.7156
0.0561
1.4309
0.1732
1.6174
0.1441
1.6172
0.0563
1.4309
0.1543
1.7178
0.0566
1.4305
TABLE V
Optimal CTI value: (a) IEEE 4-bus, (b) IEEE 6-bus.
(a)
Relay
No.
1 4
2 6
2 6
4 8
4 8
5 1
5 1
7 3
7 3
(b)
Relay
No.
TDS
1
PS
TDS
2
PS
TDS
3
PS
TDS
4
PS
TDS
5
PS
TDS
6
PS
TDS
7
PS
TDS
8
PS
TDS
9
PS
TDS
10
PS
TDS
11
PS
TDS
12
PS
TDS
13
PS
TDS
14
PS
IA
PSO
IA-PSO
0.1534
0.9833
0.2836
0.9776
0.1469
0.8219
0.1522
0.9844
0.0754
0.8215
0.0754
0.9065
0.0754
0.8212
0.0744
0.8208
0.0758
0.8215
0.0854
0.9868
0.0984
0.9819
0.0781
0.9853
0.0772
0.9624
0.1078
0.9842
0.2602
0.5864
0.4739
0.5764
0.2406
0.5102
0.2711
0.5349
0.1268
0.4891
0.1264
0.5411
0.1265
0.4892
0.1265
0.4886
0.1268
0.4854
0.1424
0.5877
0.1647
0.5872
0.1401
0.5459
0.1265
0.5715
0.1779
0.5864
0.4064
0.4722
0.7506
0.4709
0.3872
0.4119
0.4031
0.4726
0.2005
0.4118
0.2011
0.4437
0.2003
0.4109
0.2133
0.4108
0.2006
0.4124
0.2265
0.4724
0.2610
0.4618
0.2039
0.4739
0.2002
0.4642
0.2837
0.3731
TLBO
[14]
0.539
0.649
0.600
0.510
0.432
0.300
0.356
0.355
0.382
IA
PSO IA-PSO
0.438
0.538
0.523
0.436
0.348
0.312
0.311
0.503
0.322
0.384
0.372
0.376
0.309
0.408
0.338
0.492
0.357
0.313
0.304
0.312
0.318
0.313
0.324
0.329
0.322
0.309
0.317
IA
PSO
IA-PSO
0.9532
0.9242
1.2657
1.1544
0.3369
0.3187
0.8239
1.0543
0.5564
0.3653
0.6249
1.0658
0.7862
0.3675
0.3045
0.9156
0.6550
0.8814
0.6795
0.6123
0.4267
0.9043
0.6539
0.32567
0.3156
0.3447
0.3257
0.3355
0.3387
0.3012
0.4259
0.3294
0.3139
0.3034
0.3754
0.3134
0.4223
0.3142
(b)
Relay
No.
8 1
11 1
8 1
3 2
3 2
10 3
10 3
13 3
1 4
1 4
12 5
12 5
14 5
3 6
3 6
11 7
2 7
11 7
2 7
MDEA
[4]
0.288107
4.029328
0.80684
1669.695
0.199929
-0.18123
0.378005
0.300372
0.458382
0.199845
0.225775
0.839275
0.519282
0.578187
0.347919
0.200146
0.238046
0.237149
0.200045
TLBO
[14]
2.18859
1.534819
3.249765
2.201281
0.438233
0.418234
1.236218
0.30791
0.843753
0.517069
0.937599
1.525362
1.180526
0.551088
0.30015
1.373804
0.982896
1.472532
1.019525
0.3439
0.2362
0.3002
0.3531
0.3004
0.3057
0.3054
0.3003
0.3009
0.3129
0.3455
0.3324
0.3753
0.3435
0.3325
0.3212
0.3564
0.3247
0.3038
E-ISSN: 2224-350X
MDEA
[4]
0.300
0.348
0.299
0.397
0.299
0.299
0.400
0.299
0.349
247
TABLE VI
OF Comparison for case study:
(a) IEEE 4-bus, (b) IEEE 6-bus.
(a)
Algorithm
TLBO [14]
MDEA [4]
CDEA [15]
IA
PSO
IA-PSO
OF (sec)
5.5890
3.6674
3.6774
3.6758
3.6524
3.1239
(b)
Algorithm OF (sec)
TLBO [14] 23.7878
CDEA [15] 10.6272
MDEA [4] 10.3514
IA
9.3468
PSO
8.1245
IA-PSO
7.6722
7 Appendix
7.1 IA Algorithm
Replacement rate = 0.15,
Cloning rate = 0.20,
Mutation rate = 0.15,
Suppression threshold = 10-6,
Percentile amount of clones to be re-selected = 0.80,
Pruning threshold = 1.10,
Population size = 100,
Maximum number of generation = 150.
6 Conclusions
In this paper, three optimization algorithms,
namely IA, PSO, and IA-PSO, were presented to
solve the coordination problem of IDMT directional
overcurrent relays. The proposed optimization
algorithms were validated and tested on IEEE 4-bus
and IEEE 6-bus meshed power system models.
Though the three algorithms showed better results
than those obtained in literature for other
optimization algorithms, such as TLBO, CDEA and
MDEA, robustness and feasibility of IA-PSO
algorithm were clearly observed in the obtained
results.
Based on the obtained simulation results, IA-PSO
in particular proved its superiority in providing the
minimum operating time T of relays at a fast
convergence rate as well as securing minimum CTI
between primary and backup relays. This was
achieved through finding the optimum TDS and PS
values of each relay. The advantages encountered
when using IA-PSO are attributed to its hybrid
nature which combines the immune information
processing mechanism and the particle swarm
optimization algorithm to achieve better and fast
E-ISSN: 2224-350X
References:
[1] D. Birla, R. P. Maheshwari, and H. O. Gupta, An
Approach to Tackle the Threat of Sympathy Trips in
Directional Overcurrent Relay Coordination, IEEE
Transactions on Power Delivery, Vol. 22, No. 2,
2007, pp. 851-858.
[2] J. A. Sueiro, E. Diaz-Dorado, E. Mguez, and J.
Cidrs, Coordination of Directional Overcurrent
Relay using Evolutionary Algorithm and Linear
Programming, International Journal of Electrical
Power and Energy Systems, Vol. 42, 2012,
pp. 299-305.
248
E-ISSN: 2224-350X
249
E-ISSN: 2224-350X
250