 Open Access
 Total Downloads : 400
 Authors : A Sudheer Kumar, Dr. A. V Giridhar
 Paper ID : IJERTV3IS100348
 Volume & Issue : Volume 03, Issue 10 (October 2014)
 Published (First Online): 15102014
 ISSN (Online) : 22780181
 Publisher Name : IJERT
 License: This work is licensed under a Creative Commons Attribution 4.0 International License
A New Meta Heuristic Algorithm Based Shunt Capacitive Compensation for Power Loss Reduction on Radial Distribution System
A. Sudheer Kumar Dr. A.V. `Giridhar
Asst. Professor
M.Tech Student Dept. of Electrical engineering
Dept. of Electrical engineering NIT Warangal ,India
NIT Warangal ,India
Abstract: This paper describes an efficient and novel approach for capacitor placement in radial distribution systems that determine the optimal locations and size of capacitor with an objective of enhancing the voltage profile and reduction of power loss. The solution method has got two parts: in first part the loss sensitivity factors are used to select the potential buses for the capacitor placement . These loss sensitivity factors are determined by single base case power flow study
.and in second part a new algorithm that employs Flower pollination Algorithm (FPA) is used to estimate the optimal size of capacitors at the optimal buses determined in part one.For the first time flower pollination algorithm is applied for capacitor placement and sizing. The main advantage of the proposed method is that it does require a very few control parameters. The proposed method is tested on 10, 15, 69 and 85bus radial distribution systems. The results obtained by the proposed method are compared with other methods. The proposed method has given quite promising results over the other methods in terms of the quality of solution.
Index Terms:Capacitor Placement, Radial Distribution Systems, Loss Sensitivity Factors and Flower pollination algorithm.

INTRODUCTION
Distribution Systems are growing large day by day in india and most of them are radial in nature being stretched too far ends give rise to increased system losses and poor voltage regulation. Studies have shown that as much as 13% of total power generated is wasted in the form of losses at the distribution level [5]. So the need for an efficient distribution system has therefore become an important issue. In this concern, Capacitor banks are installed on Radial Distribution system for Power Factor Correction, Reduction of Loss and Voltage profile enhancement. As the optimal capacitor placement is a complicated combinatorial optimization problem, This problem has been investigated over decades. In
the 1980s, more rigorous analysis were done as given by Grainger [3],[4] and Baran Wu [1],[6] proposed the Capacitor Placement as a mixed integer nonlinear program.In the 90s combinatorial algorithms wereintroduced as a means of solving the Capacitor Placement Problem and neural network technique based papers [7] and[8] were investigated. Ng and Salama [9] have proposed a solutionapproach to the capacitor placementproblembased on fuzzy sets theory. Sundharajan and Pahwa [13] proposed the genetic algorithm approach to determine the optimal placement of capacitors based on the mechanism of natural selection.
Flower pollination algorithm is developed by Xin SheYang in 2012, inspired by the flower pollination process of flowering plants. Based on the successfully characteristics of biological systems, many nature inspired algorithms have been developed over the last few decades .
From the biological evolution point of view , the objective of the flower pollination algorithm is the survival of fittest and the optimal reproduction of plants in terms of number as well as most fittest. In this paper, Capacitor Placement and Sizing of it is done by Loss Sensitivity Factors and Flower Pollination Algorithm (FPA) respectively.In this paper, simple and efficient Distribution Load Flow method [12] is used.

SENSITIVITYANALYSIS AND LOSS SENSITIVITY FACTORS
The potential buses for the placement of capacitors are determined using the loss sensitivity factors[14]. The determination of these potential buses or candidate nodes basically helps in reduction of the search space for the remaining optimization procedure.
Consider a radial distribution line connected between p and q buses.
P R +j Xq
k th line Peff+jQeff
R[k],X[k] resistance and reactance of kline respectively , p,q are sending ,receiving end nodes and Ik current through kth line
k
Active power loss in the kthline is given by [I 2
]*R[k],which can be expressed as,
(2 []+2 [])[]
case voltage magnitudes given by (norm[j]= V[j]/0.95).Here 0.95 is taken because it is the minimum voltage that should maintained at any bus in radial distribution system . Now for the buses whose norm[j] value is less than 1.01 are considered as the candidate buses requiring the Capacitor Placement. These candidate buses are stored in rank bus vector. The norm[j]decides whether the buses needs QCompensation or not. If the voltage at a bus in the sequence list is healthy (i.e.,norm[j]>1.01) such bus needs no compensation and that bus will not be listed in the rank bus vector. The rb_busvector (rank bus vector) gives the information regarding the possible candidate buses for capacitor placement. Now sizing of Capacitors at buses given in the rb_ bus vector is done by using Flower pollination
Plineloss=
[ ]2algorithm.

FLOWER POLLINATION ALGORITHM
Similarly the reactive power loss in the kth line is given by
(2 [ ]+2 [])[]
Characteristics of Flower Pollination:
The main purpose of a flower is ultimately
Qlineloss=
[]2reproduction via pollination [2]. Flower pollination typically associated with the transfer of pollen,and
Where, Peff[q]=Total effective active powersupplied beyond the node q.
Qeff [q] = Total effective reactive powersupplied beyond the node q.
Now, both the Loss Sensitivity Factors can be obtained asshown below:
such transfer is often linked with pollinators such as insects..etc.
Pollination can take in following forms:
Aboitic:This pollination does not require any pollinators.This pollination takes place only 10%.
Biotic:about 90% of flowering plants belong to this kind of pollination, pollen transferred by insects .
=
(21 [])[]
[]2
(21 [])[]
Selfpollination: It is fertilization of one flower ,from the pollen of the same flower.
Crosspollination:means pollination can occur from
=
[ ]2pollen of a flower of a different plant.
Flower constancy: It maximizes transfer of flower
Candidate Node Selection using Loss Sensitivity Factors:
The Loss Sensitivity Factors are
determinedfrom the base case power flows and the values are arranged in descending order for all the lines of the given system.A vector bus position
bus_pos [j] is used to store the respective end buses of the lines arranged in descending order of the Values. The descending order of elements of bus_pos[j] vector will decide the order in which the buses are to be choosen for compensation. At these buses of bus_pos[j] vector, normalized voltage magnitudes are calculated by considering the base
pollen to the same or conspecific plants, and thus maximizing the reproduction of the same flower species.
Flower Pollination algorithm:
Now we can idealize above characteristics of pollination process, flower constancy and pollinator behavior as following rules[2].

Biotic ad Crosspollination considered as global pollination.

Abiotic and self pollination considered as local pollination.
Pseudo code for flower pollination algorithm:
Objective :min or max f(x), x = (x1, x2, …, xd) Initialize a population of n flowers with random solutions. Find the best solution g in the initial population
Define a switch probability p [0, 1]
While(t < max generation) For i=1:n( for all flowers ) Ifrand < p
Draw a step vector L from Levy distribution Global pollination
X[i]t+1=X[i]t +L*(X[i]t – g * )
Else
Randomly choose j and k solutions. Local pollination
X[i]t+1=X[i]t +* (X[j]t X[k]t)
End if loop
Evaluate new solutions

Flower constancy can be regarded as reproduction probability.
4..Local pollination and Global pollination is controlled by a switch probability p [0,1].
From the above discussions in this algorithm two key steps are there. i.e.,global and local pollination.
In global pollination step, flower pollens are carried by pollinators and pollen can travel long distance because insects can often fly and move longer range.This ensures the pollination and reproduction of the most fittest , and thus we represent the most fittest as g *. The first rule plus flower constancy can be mathematically represented by
X[i]t+1=X[i]t +L(X[i]t – g * )
whereX[i]t is the pollen i or solution vector X[i]
at iteration t, and g *
is current best. The parameter
L is strength of pollination , which is essentially is a step size. That is , we draw L>0 from a Levy distribution.
L=step size drawn from levys distribution [] and its value is given by
L=
( ) sin ( /2)
1+ _
Here=1.5 andS0= 0.1 S>>S0 >0
The local pollination (Rule 2) and flower constancy can be represented by
X[i]t+1=X[i]t +rand*(X[j]tX[k]t)
Where X[j]tandX[k]tare pollens from thedifferent
flowers of the same plant species.
IV ALGORITHM FOR CAPACITOR PLACEMNT AND SIZING USING LOSS SENSITIVITY AND FLOWER POLLINATION ALGORITHM
Step1: Run the base case Distribution load flow and determine the active power loss.
Step2: Identify the Candidate buses for placement of capacitors using Loss Sensitivity Factors as foresaid. Step3: Generate randomly n number of flowers, where each flower is represented as X[i]={Qc 1,Qc2,.,Qcj} Where j represents number of candidate buses or potential buses and find best solution g* in the initial population by running load flow for each X[i].
Step4:Define switching probability p [0, 1]
Step5: Set the Iteration count, iter=1.
Step6: choose random number between [0,1]. If rand< p , go to step7 other wise go to step 8
Step7: draw a step vector L from Levys distribution and update flower solution by
X[i]t+1=X[i]t +L(X[i]t – g * )
where X[i] t = previous iteration value
L=step size drawn from levys distribution[] and its value is given by
L=
( ) sin ( /2)
1+ _
Here=1.5 andS0= 0.1 S>>S0>0
g*=current best flower (solution) Step8: Randomly choose j and k solutions from existing solutions and update solution by
X[i]t+1=X[i]t +* (X[j]t X[k]t)
where X[j]t = j th random solution(flower) X[k]t =k th random solution(flower)
= random number between [0,1]
Step9: Run the power flow with updated Xvalues. If power loss is less than previos iteration , update flower (Qc values) . if not keep old values as solutions
Step10: if all flowers not considered go to step 6. Other wise go to step 11.
Step11: Find current best solution g*
Step12: if iter<max_iter goto step5 other wise terminate.
TABLE I
COMPARISION OF PREVIOUS METHODS[10]&[14] AND PROPOSED FPA METHOD FOR OF 10BUS RADIAL DISTRIBUTION SYSTEM.
BASE CASE ACTIVE POWER LOSS=783.77 KW
Fuzzy method[10]
PSO metod[14]
Proposed FPA metod
Bus No.
Size (kvar)
Bus No.
Size (kvar)
Bus No.
Size (kvar)
4
1050
6
1174
6
1200
5
1050
5
1182
5
1200
6
1950
9
264
9
495
10
900
10
566
10
220
Total kvar
4950
Total kvar
3186
Total kvar
3115
Active power loss(kW)
704.88
Active power loss(kW)
696.21
Active power loss(kW)
692.72

TEST RESULTS
The proposed FPA method for loss reduction by capacitorplacement and sizing is tested on 10bus [10], 15bus [12], 69bus [1] and 85bus [12] radial distribution systems.The constant used in the proposed algorithm is only p (switching probability). In this algorithm p=0.8 is taken. The above process is implemented and coded in MATLAB.The test results are shown below in various tables. The 10 bus test system with the proposed FPA method is compared with the paper
[10] and with the paper [14] in which the total kvar placed is 4950 kvar with a loss reduction of 10.06% And total kvar placed is 3186 kvar with loss reduction of 11.17% respectively where as the proposed FPA method for the identified locations with the paper [14] the total kvar placed is only 3115 kvar that too with a loss reduction of 11.62% as shown in Table I.When the proposed FPA method is tested on 15 bus system and compared with the paper [11] and with the paper [14], in which the total kvar placed is 1193 kvar with a loss reduction of 47.24% And total kvar placed is 1192 kvar with loss reduction of almost same as before respectively where as the proposed FPA method for the identified locations with the paper [14] the total kvar placed is only 964 kvar with a loss reduction of 48.03% as shown in Table II. Similarly Table III shows the test results of the proposed FPA method on 69bus system with a loss reduction of 32.64% and compared with results given in paper[syd] and Table IV shows the test results of the 85bus radial distribution system with a loss reduction of 51.43% and compared with results given in paper[syd] in which loss reduction is of 48.26%.TABLE II
COMPARISION OF PREVIOUS METHODS[11]&[14] AND PROPOSED FPA METHOD FOR OF 15BUS RADIAL DISTRIBUTION SYSTEM.
BASE CASE ACTIVE POWER LOSS=61.79 KW
Method given in
[11]PSO method[14]
Proposed FPA
method
Bus No.
Size
(kvar)
Bus No.
Size
(kvar)
Bus No.
Size
(kvar)
3
805
3
871
6
356
6
388
6
321
3
608
Total kvar
1193
Total kvar
1192
Total kvar
964
Active power loss(kW)
32.6
Active power loss(kW)
32.7
Active power loss(kW)
32.11
TABLE III
COMPARISION OF PREVIOUS METHOD [14] AND PROPOSED METHOD FOR OF 69 BUS RADIAL DISTRIBUTION SYSTEM.
BASE CASE ACTIVE POWER LOSS=225 KW
PSO method[14]
Proposed FPA method
Bus No.
Size
(kvar)
Bus No.
Size
(kvar)
46
241
57
213
47
365
58
200
50
1015
61
1066
Total kvar
1621
Total kvar
1479
Active power loss(kW)
152.48
Active power loss(kW)
151.55
TABLE IV
COMPARISION OF PREVIOUS METHODS[15]&[14] AND PROPOSED METHOD FOR OF 85 BUS RADIAL DISTRIBUTION SYSTEM.
BASE CASE ACTIVE POWER LOSS=315.71 KW
PSO method[14]
Proposed FPA method
Bus No.
Size (kvar)
Bus No.
Size (kvar)
8
796
8
775
58
453
7
200
7
314
58
615
27
901
27
759
Total kvar
2464
Total kvar
2349
Active power loss(kW)
163.32
Active power loss(kW)
153.34

CONCLUSION
In this paper, an algorithm that employs Flower pollination algorithm, a meta heuristic optimization technique fordetermination of required level of shunt capacitive compensationto improve the voltage profile of the system and reduce activepower loss. This algorithm (FPA) for the first time presented in this paper for active power loss reduction in radial distribution systems. Loss Sensitivity Factors are used to determine theoptimum locations required for compensation. The mainadvantage of this proposed method is that it systematicallydecides the locations and size of capacitors to realize theoptimum sizable reduction in active power loss and significantimprovement in voltage profile. Test results on 10, 15, 69and 85 bus systems are presented and compared with other methods as loss reduction in proposed method is more.

REFERENCES


M. E Baran and F. F. Wu, Optimal Sizing of Capacitors Placed ona Radial Distribution System, IEEE Trans. Power Delivery, vol. no.1, pp.11051117, Jan. 1989.

XinShe Yang, Flower pollination algorithm for global optimiza tion, in: Unconventional Computation and Natural Computation 2012, Lecture Notes inComputer Science, Vol. 7445, pp. 240249 (2012).

J.J. Grainger and S. Civanlar, Volt/var control on Distributionsystems with lateral branches using shunt capacitors as Voltage regulatorspart I, IIand III, IEEE Trans. Power Apparatus and systems, vol. PAS104, no.11, pp. 32783297, Nov. 1985.

J.J .Grainger and S.H Lee, Capacitor release by shunt capacitor placement on Distribution Feeders: A new Voltage Dependent Model,IEEE Trans .PAS, pp 12361243 May 1982.

Y. H. Song, G. S. Wang, A. T. Johns and P.Y. Wang, Distribution network reconfiguration for loss reduction using Fuzzy controlled evolutionary programming, IEEE Trans. Gener., trans., Distri., Vol.144, No.4, July 1997.

M. E. Baran and F. F. Wu, Optimal Capacitor Placement onradial distribution system, IEEE Trans. Power Delivery, vol. 4, no.1, pp. 725734, Jan. 1989.

N. I. Santoso, O. T. Tan, Neural Net Based Real Time Control Capacitors Installed on Distribution Systems, IEEE Trans. Power Delivery, vol. PAS5, no.1, pp. 266272, Jan. 1990.

P. K. Dash, S. Saha, and P. K. Nanda, Artificial Neural Net Appro Ach forInternational forum on Applications of Neural Networks to Power Systems, pp. 247250, 1991.

H.N. Ng, M.M.A. Salama, Fuzzy Optimal Capacitor Sizing and Placement, Canadian Conference on Electrical and Computer Engineering, pp. 680683, 1995.

ChingTzong Su and ChihCheng Tsai, A New Fuzzy Reasoning Approach to Optimum Capacitor Allocation for Primary distributionSystems, proceedings of IEEE International Conference on IndustrialTechnology, 1996.

M. H. Haque, Capacitor placement in radial distribution systems For loss reduction, IEE ProcGener, Transm, Distrib, vol, 146, No.5, Sep. 1999.

D.Das, D. P. Kothari, and A. Kalam, Simple and efficient method forload flow solution of radial distribution networks Electrical Power &Energy Systems, vol. 17. N0.5,pp 335346, Elsevier Science Ltd 1995.

Sundharajan and A. Pahwa, Optimal selection of capacitors for Radial distribution systems using genetic algorithm, IEEE Trans. Power Systems, vol. 9, No.3, pp.14991507, Aug. 1994.

Prakash K. and Sydulu M, Particle swarm optimization based Capacitor placement on radial distribution systems, IEEE Power Engineering Society general meeting 2007. pp. 15.