Optimal Placement of Distributed Generation Using Bacterial foraging Optimization in an Electrical Distribution System

Optimal Placement of Distributed Generation Using Bacterial foraging Optimization in an Electrical Distribution System

1J. A. Baskar, 3

1Research Scholar, Asso Prof.

Department of Electrical and Electronics Engineering, Narayana Engineering College, Gudur, Nellore Dt, A.P., India

2 Dr. R. Hariprakash (IITM),

2 Principal,

Jagans College of Engineering & Technology, Nellore, A.P., India

Dr. M. Vijayakumar

3Professor,

Dept of Electrical & Electronics Engineering and Director (Admissions) JNTUA, Anantapuramu, India

Abstract- Distributed generation (DG) has been placed in electrical distribution networks widely due to line loss reduction, environmental benefits, better voltage profile, postponement of system upgrading, and increasing reliability. Optimization techniques are tools which can be used for placement and sizing of the DG units in the distribution system. The impacts of DG placement issues, such as power loss minimization and voltage profile, should be analyzed effectively by proposing a method of locating and sizing DG units. This paper proposes Bacterial Foraging Optimization (BFO) algorithm technique for optimal placement of DG units in Distribution System (DS) to minimize the total real power loss, improve power factor and to regulate the voltage profile. BFO is a recently developed nature-inspired optimization technique, which is based on the foraging behavior of E. coli bacteria. To determine the validity of the BFO algorithm, a standard IEEE 69 bus radial distribution feeder system is examined with two test cases. The results show that the proposed BFO algorithm is more efficient, and capable of handling mixed integer nonlinear optimization problems effectively.

Keywords- Distributed Generation, Optimal DG size, Bacterial Foraging Optimization (BFO), E. coli,DS.

NOMENCLATURE

SLoad,k Apparent load power at bus k

Ssystem j,k System apparent power flows from bus j to bus k Ssystem k,,j System apparent power flows from bus k to bus j Srated j,k Apparent rated power flows from bus j to bus k

Srated k,,j Apparent rated power flows from bus k to bus j Ssystem,k Apparent load power at bus k

Pj Active power flows from bus j to bus k Qj Reactive power flows from bus j to bus k N Number of buses

Vj Bus voltage at bus j

Vk Bus voltage at bus k

A P k Active power injected bus k R P k reactive power injected bus k Lj Load demand at bus j

,

System apparent power flows from bus j to bus k

Maximum specified allowable voltage

PDGj Dispatchable DG rated active power at bus j QDGj Dispatchable DG rated reactive power at bus j rk Line resistance connecting buses j and k

xk Line reactance connecting buses j and k

µp Real power multiplier when there is no real power source set active power multiplier to 0 or when there is real power source set to 1

µq Reactive power multiplier when there is no real power source set active power multiplier to 0 or when there is real power source set to 1

Maximum DG unit size in KVA

Minimum DGunit size in KVA

. Maximum DG units working power factor

. Minimum DG units working power factor

1. INTRODUCTION

   With the increasing level of global warming due to emissions of CO2, renewable based distributed generation (DG) should play a major role in electrical power generation. DG may be defined as a small scale generating resource, placed near to load being served or at low voltage distribution side [1]. DG refers to small sources ranging between 1 kW and 50 MW, which are normally placed close to consumption centers. In general DS is designed for one- way power flow. The insertion of DG in the DS violates this basic assumption and can disrupt distribution operation if not carefully employed, potentially causing islanding, protection disturbances, upset voltage regulation, and other power quality problems. DG normally follows the utility voltage and injects a constant amount of real and reactive power [3].

   The optimal placement and sizing of DG units on the DS has been analyzed to achieve various solutions. The objective was the minimization of the active losses of the feeder or the minimization of the total network supply costs, which includes generators operation and losses compensation or even the best utilization of the available generation capacity. A stochastic multi-objective model for placement of DG in electrical distribution networks is proposed in [13] with a binary particle swarm optimization (PSO) algorithm. The DS expansion planning strategy with DG systems with intermittent power generation is presented in [14] to obtain optimal solutions for system planners. A DG placement planning study framework is brought in [15] that includes a coordinated reconfiguration feeder and voltage control to calculate the maximum allowable DG capacity at a given node in the electrical distribution system. Works have been dealing with the DG optimal placing and sizing problem by means of multi-objective optimization (MOO) tools, with a considerable number of them based on heuristic approach. Some basic concepts about MOO are introduced in order to support the methods comprehension. The MOO consists in minimize or maximize simultaneously a set of objectives subject to a number of constraints. The process of optimization in a multi-objective scenario occurs in two stages: the determination of the solution set, where all objective function values of each solution cannot be enhanced at the same time and the human decision making whose criteria can be applied before, after or even during the optimization process.

   based approach for optimal location of DG and determining the size of DG units in electrical distribution systems with different load models based on PSO is introduced in[17,18] and a combined genetic algorithm is presented in[19] for optimal location and sizing of DG on DS. A methodology for evaluating the impact of DG-units on power loss, reliability, and voltage profile of distribution networks was presented in reference [11].

   A heuristic approach for optimal location and size of DGs in distribution networks, with the objectives of minimizing the power loss and voltage profile improvement is proposed in [6].In [10] a Newton-Raphson algorithm based load flow program is used to solve the load flow problem. Moreover, some heuristic search requires exhaustive search for all possible locations which may not be applicable to more than one DG.

   As a contribution to the methodology for DG placement analysis, in this paper it is proposed a BFO algorithm for the allocation of distributed generators in distribution networks, in order to improve voltage profile and line loss reduction in DS. The organization of this paper is as follows. Section II addresses the problem formulation. Section III addresses the impact of the DG placement and size on DS. The BFO algorithm is represented in Section IV. Pseudo code for a BFO computation procedure for the problem is given in Section V. Simulation result on the test systems are illustrated in Section VI. Then, the conclusion is given in Section VII.

2. PROBLEM FORMULATION

Importance of placing a DG in distribution networks is to reduce the total system real power loss while satisfying certain operating constraints. In other words, the problem of DG application can be interpreted as determining the optimal placement and size of the DG to satisfy the desired objective function subject to equality and inequalityconstraints. Reliability, accuracy, and flexibility of the DG solution algorithm are influenced by the load flow analysis used. So the overall algorithm accuracy is highly depends on that analysis. It can be otherwise called that the load flow analysis is the heart of the DG-unit solution algorithm. Based on that the load flow algorithm used in [10] is applied in this paper. In Fig.1, a sample two bus system including DG-unit is considered. The mathematical formulations of the mixed integer nonlinear optimization problem for the DG-unit application are as follows: [9]

The objective function is to reduce the real power loss

Obj.Fun = min (

=0

2+2 2

) * rk ———- (1)

Fig: 1: Sample two-bus system with one DG

In [16] a distributed micro grid model has been introduced to optimize best location and the capacities within DG micro grid, in which wind power and photovoltaic

The equality constraints are the three nonlinear recursive power-flow equations describing the system [10]

(2+2)

2

Pj – – PL k +µpA P k – P k =0 ———— (2)

(2+2)

power are taken into consideration with both PSO and Elitism Genetic Algorithm (EGA).A multi-objective index-

Qj –

2

– QL k +µqR P k – Q k =0 ———- (3)

(2+2)(2+2) optimal placement of DGs [19]. Since the analytical methods

2=2-2(rkPj+xkQi )+ ———– (4)

2

Where i=0, 1, 2… n

The inequality constraints are the systems voltage limits, that is, +5% or – 5% of the nominal voltage value

< < ——————————- (5)

are generally poor to solve this type of function, almost many of the related papers are based on heuristic methods. Due to the discrete nature of the sizing and placement problem, the objective function has a number of local minima. The hybrid particle swarm optimization (HPSO) algorithm has applied as a useful tool for engineering optimization, to solve complex optimization problems [10-15].This paper presents a novel search approach with respect to the voltage profile

for the optimal placement of DGs using the BFO algorithm

In addition, the thermal capacity limits of the networks feeder lines are treated as inequality constraints

< < —————————— (6)

and compares it with the Bee colony optimization (BCO) algorithm and other methods. Optimal bus locations are determined to obtain the best objective. The multi objective optimization simultaneously covers the optimization of both

,

,

,

the voltage margin and active power loss. The problem is defined and the objective function is introduced to maximize

The discrete inequality constraints are the DG-units size

(KVA) and power factor

——————- (7)

the voltage profile index, and minimize losses. Hence, an algorithm is developed to assess the voltage profile based on the loadability limit. This method is executed on the IEEE 69

bus test systems, showing the robustness of this method in finding the optimal sizing and placement of DGs, efficiency

. . . ————– (8)

for improvement of voltage profile, power factor

improvement and reduction of system real power losses [14].

The power factors of DG are set to operate at practical values [27], that is, from unity to 0.85 towards the optimal result. The operating DG-units power factor whether lagging or leading must be dissimilar to the buss load at which t he DG-unit is placed [26]. Consequently, the net total of both active and reactive powers of that bus where the DG-unit is placed will also decrease.

III.IMPACT OF THE DG PLACEMENT AND SIZE ON DISTRIBUTION SYSTEM

The placement of DGs renders a group of advantages, such as economic, environmental, and technical. The economic advantages are the reduction of transmission and distribution costs, reduction of electricity price, and saving of fuel. Environmental advantages entail reductions of sound pollution and emissions of greenhouse gases. The technical benefits of DGs in existing networks are minimizing power losses, reducing the emission of pollutants, improving power quality, and relieving transmission and distribution congestion [4].It can also provide stand-alone remote applications with the required power. Therefore, the optimal placement and sizing of DGs attract active research interest.

Several researchers have worked in this area [7- 13].DGs are placed at optimal locations to reduce losses [7]. Some researchers have presented load flow algorithms to find the optimal size of DGs at each load bus [8, 9]. Wang and Nehrir have shown analytical approaches for optimal placement of DGs in terms of loss [10]. Chiradeja quantified the benefit of reduced line loss in a radial distribution feeder with a concentrated load [11]. Further, many researchers have used evolutionary computational methods for finding the optimal DG placement [14-19]. Mithulananthan used a genetic algorithm (GA) for placement of DGs to reduce the losses [15]. Celli and Ghiani used a multi objective Evolutionary algorithm for the sizing and placement of DGs [18]. Nara et al. used a tabu search algorithm to find the

1. BACTERIAL FORAGING OPTIMIZATION IMPLEMENTATION

   The BFO algorithm was first represented by Pasino in 2002. The idea in this method was adopted from biological and physical living behavior of E. coli bacteria existing in human intestine. This algorithm has three main processes namely Chemotaxis, Reproduction and Elimination fault Dispersal. When E. coli grows, it gets longer, and then divides in the middle into two daughters. Given sufficient food and held at the temperature of the human gut of 37 ° C, E. coli can synthesize and replicate everything it needs to make a copy of itself in about 20 min; hence growth of a population of bacteria is exponential with a relatively short time to double. The E. coli bacterium has a guidance system that enables it to search for food and try to avoid noxious substances. The behavior of the E. coli bacterium, will be explained as its actuator (the flagellum), decision making, sensors, and closed-loop behavior. This section is based on the work in [24, 25]. Fig 2 shows the simplified BFO optimization flowchart.

   Fig 2.Simplified BFO optimization flowchart

   also, the number of chemotaxis is represented by NC.

   1. Reproduction: After the number of NC Chemotaxis steps, reproduction step takes place. Nre represents the number of reproduction steps.

   2. Elimination-Dispersal: The swimming process prepares the conditions for local search and reproduction process speeds up the convergence. In a large space swimming and reproduction for searching global optimal point cannot be sufficient. In bacterial foraging, dispersion takes place after a definite number of reproduction processes. A bacterium is selected with regard to a prearranged probability of Ped to be dispersed in the environment and moved to another position. These events can effectively prevent trapping in local optimal point. Ned is the number of elimination and dispersal phenomenon and Ped is defined for every bacterium with the probability of elimination and dispersal.

      A flowchart for BFO algorithm by MATLAB Programming is shown in Figure 2.

2. PSEUDO CODE FOR BFO ALGORITHM: Step (1): Initialize the parameters,

S: Total number of bacteria

P: Number of parameters to be optimized Nc; Number of chemotactic steps

Ns: Number of swarming,

Nre: Number of reproduction steps Ned: elimination-dispersal steps dattract, wattract, drepellant, wrepellant :

Attractant and repellant values

1. Chemotaxis: An E. coli bacterium ca decide to move in two different ways depending on its environment. A

   Ped

   : Probability of elimination- dispersal

   bacterium is subject to change during its lifetime between the two ways of swimming (swim for a short time) and tumbling. In BFO, one moving unit length with random directions represents tumbling and one moving unit length with the same direction relative to the final stage represents swimming. The mathematical equation for

   Chemotaxis is expressed as follows:

   C(i) : step size

   step(2): Elimination-Dispersal loop l = l + 1 step(3): Reproduction loop k = k + 1 step(4): Chemotactics loop j = j + 1

   ( + 1, , ) = (, , ) + () ()

   ()()

   Where

   i : location of ith bacterium C(i) : movement length

   (i) : direction random vector

   j : is representing jth chemotaxis k : is representing jth reproduction l : is representing jth elimination

   and dispersal

   — (9)

   step(5): Every bacterium i = i + 1

   1. Compute fitness function J(i,j,k,l)

      Let, J (i, j, k, l) = J (i, j, k, l) + Jcc

   2. Let, J (last) = J(i,j,k,l) to save this value. we may find a better cost via a run.

   3. Tumble: Generate a random vector (i) such that 1 <= (i) <= -1.

   4. Move: Let, i (j+1,k,l) = i (j,k,l) + C(i)* ((i) /

      ( T(i)* (i)))

   5. Compute J(i,j+1,k,l) and let, J(i,j,k,l) = J(i,j,k,l)

      + Jcc

   6. Swim: Let, m = 0.(counter for swim length)

      (i) While m < Ns (ii)Let, m = m+1

      1. if J(i,j+1,k,l) < J(i,j,k,l) = J(i,j,k,l) + Jcc then Let, Jlast = J(i,j+1,k,l)

         i (j+1,k,l) = i (j,k,l) + C(i)* ((i) / ( T(i)* (i)))

      2. else, let m = Ns

   7. Go to next bacterium, if i S Step(6): if j < Nc, go to step(4) Step(7): Reproduction:

1. for the given k and l, and for each i = 1 to S,

   1. Let, Jhealth = i=1:Nc J(i,j,k,l)

   2. Sort the fitness in ascending order

2. The Sr bacteria with worse health value will die, the remaining Sr bacteria with best values will split into two.

Step(8): if k < Nre, go to step(3)

Step(9): Elimination-Dispersal

1. for i = 1 to S, with probability Ped, eliminate and disperse each bacterium.

2. if a bacterium eliminated, then add new one

Fig. 3.Single-line diagram of the 69-bus feeder system

Line loss reduction analysis is based on the simple case of an IEEE 69 bus radial distribution feeder [4] is considered with following cases:

1. System without DG

2. System with the inclusion of one DG to share full load

1. System with the inclusion of one DG & one Capacitor to share full load

Case (I): System without DG

This is a reference scenario, in which no DG unit is connected to the system (default case). The voltage profiles of the feeder system with out DG placement has shown in fig. 4.

Voltage magnitude in volts versus bus

to a random location on the search space.

Step (10): if l < Ned, go to step (2), Else Terminate. VI.RESULTS AND DISCUSSION

The proposed BFO algorithm is implemented in MATLAB programming, and was executed on an Intel dual core PC with 3.0-GHz speed and 4 GB RAM. To check the performance of the proposed BFO algorithm, the 69-bus radial distribution feeder system was considered in different test cases.. We studied two test cases with the loads are identical to the values given in [11], ie. The total demands of

1

0.98

Vm in Volts

0.96

0.94

0.92

0.9

0.88

0 10 20 30 40 50 60 70

Bus

the 69-bus system are 3802.19 kW and 2694.60 kVAR.. The substation voltage and load power factors in both scenarios were considered as 1.0 p.u. and lagging p.f., respectively.

Fig .4: Voltage magnitude in p.u volts versus bus numbers before DG placement.

It is observed that from bus number 59 to 64 the voltages in p.u are 0.919, 0.912, 0.9, 0.9, 0.899 and 0.897 respectively. These voltages are the lowest among 69 buses.

Table1: Optimized results before DG placement

Table 2: parameters after 1 DG placed to share full load

.

Case 2: System with one DG to share full load

In this, the voltages are improved in all the buses In particular from 59 to 64 shown in Appendix 1.

Fig .5: Fitness value versus chemotatic steps for one DG unit connected with actual load

Fig .6: Voltage magnitude in p.u volts versus bus No. after one DG placed to share full load

Case 3: System with 1DG and one capacitor to share full load

Fig.7: Fitness value versus chemotatic steps for one DG plus one capacitor unit connected with active power supply

Figure.8: Voltage magnitude in p.u versus bus number for one DG plus one capacitor unit connected to share full load.

Table 4: parameters after one DG unit plus one capacitor is connected with active power supply.

1. CONCLUSIONS

   In this paper, a new population-based BFO has been implemented to solve the mixed integer nonlinear optimization problem. The objective function was to reduce the total system real power loss subject to equality and inequality constraints. Simulations were conducted on the IEEE 69-bus radial distribution feeder systems. The proposed BFO algorithm successfully implemented the optimal solutions at various test cases. The BCO algorithm is simple, easy to implement, and capable of handling complex optimization problems.

   Bus No	Voltage (p.u)	Angle(degrees)	Power factor
   1	1	-0.001	0.9999995
   2	0.999963062		-0.001827696	0.99999833
   3	0.999926124	-0.002655453	0.999996474
   4	0.999855678	-0.004949589	0.999987751
   5	0.999432299	-0.01106437	0.99993879
   6	0.994842983	0.02785304	0.999612129
   7	0.990079005	0.068559551	0.997650714
   8	0.988978119	0.077936818	0.996964463
   9	0.988470617	0.082270981	0.996617651
   10	0.981698373	0.203068431	0.979452362
   11	0.980216291	0.229815898	0.97370835
   12	0.976211313	0.302187767	0.954687674
   13	0.973038305	0.359606179	0.936035484
   14	0.976556403	0.218740967	0.976171434
   15	0.972146218	0.418720309	0.91361
   16	0.971332098	0.109703624	0.99398859
   17	0.97	0.429546074	0.909154885
   18	0.969996457	0.089172913	0.99602673
   19	0.96999241	0.073656787	0.997288565
   20	0.969989812	0.291364934	0.957852677
   21	0.970973705	0.339078857	0.943061455
   22	0.970952302	0.339470934	0.94293097
   23	0.970718881	0.343725439	0.94150574

   Appendix 2: Optimized values after one DG unit is connected to share normal load in 69 bus system using BFO technique

   24	0.970210821	0.352992258	0.93834247
   25	0.969379503	0.368161557	0.932990579
   26	0.969036582	0.3744269	0.93071738
   27	0.968875267	0.377375947	0.929634752
   28	0.999915049	-0.003057645	0.999995325
   29	0.999800109	-0.007227664	0.999973881
   30	0.999605872	-0.00381987	0.999992704
   31	0.999571597	-0.003218224	0.999994822
   32	0.999400223	-0.000209374	0.999999978
   33	0.998989384	0.006888811	0.999976272
   34	0.998451355	0.016278797	0.999867503
   35	0.998343243	0.017987214	0.999838234
   36	0.999816573	-0.002874332	0.999995869
   37	0.998272995	-0.004258298	0.999990933
   38	0.99691289	0.021906947	0.999760052
   39	0.996520349	0.029476017	0.999565614
   40	0.996497915	0.029936174	0.999551946
   41	0.987737226	0.222239503	0.975406277
   42	0.984018194	0.305185779	0.953791148
   43	0.983527399	0.316222083	0.950417047
   44	0.983492904	0.320228582	0.949163489
   45	0.982120736	0.348132892	0.940011302
   46	0.982109566	0.348367352	0.939931292
   47	0.999779182	-0.00721671	0.99997396
   48	0.997877073	-0.063267456	0.997999282
   49	0.992037316	-0.232567382	0.973077881
   50	0.991257571	-0.248968181	0.969167182
   51	0.988921908	0.078400135	0.996928283
   52	0.988906784	0.078690196	0.996905524
   53	0.988338615	0.083477341	0.99651779
   54	0.988197196	0.084731136	0.996412464
   55	0.988102979	0.085561691	0.996341831
   56	0.988102979	0.085561691	0.996341831
   57	0.988102979	0.085561691	0.996341831
   58	0.988102979	0.085561691	0.996341831
   59	0.993826375	-0.166536859	0.986164758
   60	0.994800171	-0.165735206	0.986297329
   61	0.996149963	-0.150750104	0.988658706
   62	0.995803655	-0.147762429	0.989102981
   63	0.995351824	-0.143867794	0.989668867
   64	0.99313746	-0.124769784	0.992226343
   65	0.968048854	0.385011974	0.926793936
   66	0.980126182	0.231623866	0.973294907
   67	0.980125133	0.231645278	0.973289991
   68	0.975686721	0.311746083	0.951799461
   69	0.975685044	0.311775695	0.951790378

   Appendix 2: Optimized values after one DG and one capacitor connected to share full load in 69 bus system using BFO technique

   Bus No	Voltage (p.u)	Angle(degrees)	Power factor
   1	1	-0.001	0.9999995
   2	0.999964995	-0.001740283	0.99999848
   3	0.99992999	-0.002480619	0.99999692
   4	0.999865349	-0.004512547	0.99998981
   5	0.999518468	-0.014211835	0.99989901
   6	0.995425256	-0.028577277	0.99959169
   7	0.991177444	-0.043860235	0.99903829
   8	0.99020142	-0.048105159	0.99884317
   9	0.989760495	-0.051064597	0.99869648
   10	0.98371427	-0.052704446	0.99861144

   11	0.98239746	-0.054173249	0.99853298
   12	0.979015595	-0.0894759	0.99599970
   13	0.976736348	-0.188703683	0.98224823
   14	0.976557155	0.474460183	0.88953947
   15	0.972146476	0.365552354	0.93392645
   16	0.971332267	0.081474304	0.99668280
   17	0.97	0.481418947	0.88633879
   18	0.969996457	0.453698589	0.89883218
   19	0.96999241	0.283626498	0.96004691
   20	0.969989812	0.165374228	0.98635681
   21	0.97597783	-0.285118361	0.95962836
   22	0.975956537	-0.284730295	0.95973744
   23	0.975724318	-0.280519328	0.96091178
   24	0.975218876	-0.271347412	0.96341062
   25	0.974391844	-0.256333687	0.96732601
   26	0.974050692	-0.250132677	0.96887958
   27	0.973890208	-0.24721393	0.96959794
   28	0.999918916	-0.002882808	0.99999584
   29	0.999803976	-0.007052794	0.99997512
   30	0.99960974	-0.003645027	0.99999335
   31	0.999575465	-0.003043385	0.99995369
   32	0.999404092	-3.46E-005	0.99999999
   33	0.998993255	0.007063571	0.99997505
   34	0.998455228	0.016453484	0.99986464
   35	0.998347116	0.018161888	0.99983507
   36	0.999820425	-0.002699202	0.99999635
   37	0.998276633	-0.004078863	0.99999168
   38	0.996916356	0.022093903	0.99975594
   39	0.996523767	0.02966515	0.99956002
   40	0.99650133	0.030125435	0.99954626
   41	0.987739455	0.222481569	0.97535289
   42	0.984019919	0.305450745	0.9537115
   43	0.983529058	0.316490097	0.95033366
   44	0.983494561	0.320497306	0.94907886
   45	0.982122203	0.348409708	0.93991683
   46	0.982111032	0.348644235	0.93983673
   47	0.999780582	-0.005690748	0.99998380
   48	0.997674269	-0.034617581	0.99940087
   49	0.991134994	-0.110916198	0.99385510
   50	0.990156675	-0.100719798	0.99493204
   51	0.990145279	-0.047642986	0.99886528
   52	0.990130174	-0.047353642	0.99887902
   53	0.989628665	-0.04986138	0.99875717
   54	0.989487431	-0.048610852	0.99881872
   55	0.989393337	-0.047782462	0.99885863
   56	0.989393337	-0.047782462	0.99885863
   57	0.989393337	-0.047782462	0.99885863
   58	0.989393337	-0.047782462	0.99885863
   59	0.99433958	0.26457186	0.96520454
   60	0.995752254	0.312542175	0.95155498
   61	0.997510741	0.399094494	0.92141323
   62	0.997164908	0.40207402	0.92025135
   63	0.996713696	0.405958024	0.91872449
   64	0.994502371	0.425003764	0.91103718
   65	0.973068058	-0.239656408	0.97141959
   66	0.98230755	-0.052373302	0.99628832
   67	0.982306504	-0.052351985	0.99862994
   68	0.978492507	-0.079972292	0.99680392
   69	0.978490835	-0.07994284	0.996806272

2. REFERENCES

1. Pukar Mahat,Weerakorn Ongsakul and Nadarajah Mithulananthan, Optimal Placement of Wind Turbine DG in Primary Distribution Systems for Real Loss Reduction, 2005.

2. G. P. Harrison and A. R. Wallace. OPF evaluation of distribution network capacity for the connection of distributed generation .

3. Optimal Size and Location of Distributed Generations for Minimizing Power Losses in a Primary Distribution Network,

   R.M. Kamel and B. Kermanshahi, Computer Science & Engineering and Electrical Engineering,Vol. 16, No. 2, pp. 137-144.

4. N.Mithulanandan,Distributedgenerator placement in power distributed generator placement in power distribution system using genetic algorithm to reduce losses", TIJSAT, 9(3), pp. 55-62 (2004).

5. Borges, C. and Falcao, D.Impact of distributed Generation allocation and sizing on reliability, losses and voltage profile", in Proceedings of IEEE Bologna, Power Technology Conference (2003).

6. Row, N. and Wan, Y.Optimum location of resources in distributed planning", IEEE Trans PWRS, 9(4), pp.2014-2020 (1994).

7. Kim, J.O.Dispersed generation planning using improved Hereford ranch algorithm", Electric Power Syst. Res., 47(1), pp. 47-55 (1998).

8. Kim, K. et al.Dispersed generator placement using fuzzy-GA in distribution systems", in Proceeding of IEEE Power Engineering Society Summer Meeting, Chicago, pp. 1148- 1153 (July 2002).

9. Fahad S. Abu-Mouti, Student Member, IEEE, and M. E. El- Hawary, Fellow, IEEE, Optimal Distributed Generation Allocation and Sizing in Distribution Systems via Artificial Bee Colony Algorithm, IEEE Transactions on power delivery, Vol. 26, NO. 4, October 2011, pp. 2090-2101.

10. Hamed Piarehzadeh, Amir Khanjanzadeh and Reza Pejmanfer, Comparison of Harmony Search Algorithm and Particle Swarm Optimization for Distributed Generation Allocation to Improve Steady State Voltage Stability of Distribution Networks, Research Journal of Applied Sciences, Engineering and Technology 4(15): 2310-2315, Aug 2012.

11. M. E. Baran and F. F.Wu, Optimal capacitor placement on radial distribution systems, IEEE Trans. Power Del., vol. 4, no. 1, pp. 725735,Jan. 1989.

12. M.Afzalan and M.A.Taghikhani, DG Placement and Sizing in Radial Distribution Network Using PSO&HBMO Algorithms, Energy and Power, pp.61-66, 2012, 2(4).

13. A.Soroudi,M.Afrasiab;"BinaryPSO-Based Dynamic Mul-ti- Objective Model for Distributed Generation Planning un-der Uncertainty",IET Renewable Power Generation, 2012, Vol.6, No. 2, pp. 67 – 78.

14. Kai Zou, A.P.Agalgaonkar, K.M.Muttaqi, S.Perera; "Distri- bution System Planning with Incorporating DG Reactive Capability and System Uncertainties" ,IEEE Transactions on Sustainable Energy, 2012, Vol.3 , No.1, pp. 112 123.

15. Sheng-Yi Su, Chan-Nan Lu, Rung-Fang Chang, G. Gu-tierrez- Alcaraz;" Distributed Generation Interconnection Planning: A Wind Power Case Study ", IEEE Transactions on Smart Grid, 2011, Vol.2 , No.1, pp. 181 189.

16. Ruifeng Shi, Can Cui, Kai Su, Zaharn Zain; "Comparison Study of Two Meta-heuristic Algorithms with their Applica- tions to Distributed Generation Planning", Energy Procedia.

17. A.M. El-Zonkoly; "Optimal Placement of Multi-Distributed Generation Units Including Different Load Models using Particle Swarm Optimization", Swarm and Evolutionary Computation, 2011, Vol.1, No.1, pp. 50-59.

18. R. S. Maciel, A. Padilha-Feltrin, Universidade Estadual Paulista UNESP,Ilha Solteira, Brazil Distributed Generation Impact Evaluation Using a Multi-objective Tabu Search,2007.

19. Deependra Singh, Devender Singh, and K. S. Verma,GA based Optimal Sizing & Placement of Distributed Generation for Loss Minimization, proceedings of world academy of science, engineering and technology volume 26 december 2007, ISSN 1307-6884.pp.318-387.

20. M. Afzalan, M. A.Taghikhani, DG Placement and Sizing in Radial Distribution Network Using PSO&HBMO Algorithms,Energy and Power 2012, 2(4): 61-66 DOI: 10.5923/j.ep.20120204.04,pp-61-66.

21. M.R. AlRashidi, M.F. AlHajri; "Optimal Planning of Mul-tiple Distributed Generation Sources in Distribution Net-works: A New Approach", Energy Conversion and Man-agement, 2011, Vol. 52, No. 11, pp. 3301-3308.

22. M.H. Moradi, M. Abedini; "A Combination of Genetic Al- gorithm and Particle Swarm Optimization for Optimal DG Location and Sizing in Distribution Systems", International Journal of Electrical Power & Energy Systems, 2012, Vol.34, No.1, pp. 66-74.

23. Fahad S. Abu-Mouti, Student Member, IEEE, and

    M. E. El-Hawary, Fellow, IEEE, Optimal Distributed Generation Allocation and Sizing in Distribution Systems via Artificial Bee Colony Algorithm, IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 26, NO. 4, OCTOBER 2011.pp.2090-2101.

24. CaishengW and Hashem,M.Analytical approach for optimum placement of distributed generation sources in power systems", IEEE Trans. Power Syst., l19(4), pp. 2068-2075 (2004).

25. T. Audesirk and G. Audesirk, Biology: Life on Earth, 5th ed. Englewood Cliffs, NJ: Prentice Hall, 1999.

26. D. DeRosier, The turn of the screw: The bacterial flagellar motor, Cell,vol. 93, pp. 17-20, 1998.

27. P. P. Barker and R. W. De Mello, Determining the impact of distributed generation on power systems part 1: Radial distribution systems, in Proc. IEEE Power Eng. Soc. Summer Meeting, Jul. 1620, 2000, vol. 3, pp. 16451656.

Authors Biographies

J.A.Baskar received B.Tech degree in Electrical and Electronics Engineering from JNT University, Hyderabad India in1993, and M.Tech in Energy Management from S.V. University, Tirupathi, India in2005, India. He worked with Andhra Pradesh Dairy Dev Co-op

federation Ltd., Chittoor, for 17 years. At present he is pursuing PhD in Electrical Engineering at JNTU, Anantapur, India. He is working as Associate Professor in EEE Dept. at Narayana Engineering college,Gudur,A.P,India. He has published 6 research papers in inter-national conferences and journals His research areas include power systems, Electrical Machines, Distributed generation system and energy management.

Dr. R. Hari Prakash received B.Tech & M.Tech with distinction from Birla Institute of Technology Ranchi. He Obtained PhD from Indian Institute of Technology, Madras. He has 35 years of Industry/Research/Teaching at various levels. He worked as senior lecturer at

Nanyang University, Singapore. He worked in NBKR Institute of Science & Technology Vidyanagar as Assistant professor and became Professor and Head of the Department. He is a Member of Board of Studies at JNTU Anantapur, S.V.University, Anna University, and Vellore Institute of Technology. He has published seven papers in National & four papers in International Journals. He worked as Principal of Gokula Krishna College of Engineering, Sullurpet, and Sri Venkateswara College of Engineering Chittoor. Currently he is the Principal of Brahmaiah College of Engineering, North Rajupalem, Nellore-524 366. He has authored thebook Operations Research published by SCITECH publishers, Hyderabad. He is a Member of Indian Society for Technical Education, Indian Society for Non-Destructive Testing, Indian society for Automobile Engineering and American society of Mechanical Engineering, Fellow of Institute of Engineers. His areas of research specialization are Energy Management, Energy Distribution, Alternate source of Energy, Eco-fuels.

Dr. M. Vijaya Kumar graduated from

S.V. University, Tirupathi A.P India in 1988. He obtained M.Tech degree from Regional Engineering College, Warangal, India in 1990. He received Doctoral degree from Jawaharlal Nehru Technological University,

Hyderabad, India in 2000. Currently he is working as Professor in Electrical and Electronics Engineering Department & Director of Admissions, JNTUA College of Engineering, Anantapur, A.P, India. He is a member of Board of studies of few Universities in A.P., India. He has published 87 research papers in national and inter-national conferences and journals. Six Ph.Ds were awarded under his guidance. He received two research awards from the Institution of Engineers (India). He served as Director, AICTE, New Delhi for a short period. He was Head of the Department during 2006 to 2008. He also served as Founder Registrar of JNT University, Anantapur during 2008 to 2010. His areas of interests include Electrical Machines, Electrical Drives, Microprocessors and Power Electronics.