Multi-Objective Optimization of DG Sizing and Placement using BBO Technique

Multi-Objective Optimization of DG Sizing and Placement using BBO Technique

Mrs. R. Sai Lakshmi Surya,

PG Student, Dept. of Electrical Engineering, Annamalai University,

Dr. Ghadir Radman,

Professor,

Dr. R. Neela,

Professor,

Dept. of Electrical Engineering, Annamalai University,

Centre for Energy System Research, Tennessee Technological University.

Abstract Recent trends in power system have made it an open access environment to the customers. Distributed generations (DGs) have been preferred due to various merits such as a Green House Gas (GHG) emission reduction in generation, improvement in voltage profile as well as voltage stability at heavy load levels in distribution feeders, enhancement of reliability and power quality. Also the power market immunity has been enhanced using DG technologies. In spite of the enormous advantages of DG technologies, the DG management and power dispatch is a bigger challenge for distribution system operators. This project presents a novel application of Biogeography Based Optimization (BBO) with the aim of determining the optimal DGs placement & sizing. In the proposed multi-objective optimization, along with the operational constraints, such as improving voltage profile and stability, power-loss reduction, etc., we also take into account the economic aspects i.e., the electrical power distribution companys and DG owner's economic benefits. The proposed approach has been studied and tested on the IEEE 33-bus radial distribution test system and the consequences have been discussed.

Index Terms Distributed Generation (DG), DG placement and Sizing, Biogeography Based Optimization (BBO), Pareto Optimality, Distribution Company (DisCo), DG owner.

NOMENCLATURE

Csub Cost of the purchased active power from the substation by DisCo.

CDG Cost of the purchased active power from the DG Owner for DisCo.

Cinv Investment Cost of DG.

Cmain Maintenance Cost of DG.

Coper Operational Cost of DG.

INDG DG Owners Income.

CPDG Contract Price of DisCo with the DG.

CMWh,P Cost of active power in MWh bought from substation.

INF_R Inflation Rate.

INT_R Interest Rate.

Lb Length of the branch.

Vn Node Voltage.

Th Total number of hours in a year.

Td Total number of hours in a day.

CF Capacity Factor of the DG.

PDG,i, QDG,i Active Power &Reactive Power supplied by the DG.

Pn, Qn Active power &Reactive power line flow. TVPI Total Voltage Profile Index.

TVSI Total Voltage Stability Index. TPLI Total Power Loss Index.

Vrated Rated bus voltage.

k, k Immigration&Emigration rates.

E, I Maximum Emigration rate & Maximum Immigration rate of the habitat.

S No of Species.

1. INTRODUCTION

   For many years, electricity production and distribution has been carried out by centralized, regulated electric power authority that is responsible for both the operation of power generation facilities along with the transmission and distribution lines. Later the power system was opened to the

   other utilities apart from the main authority. This led to a competitive environment for electric power generation and the transmission system was made an open access. Utility deregulation is one of the reasons for the advancements in the Distributed Generation (DG) technology.

   DG plays a vital role in the enhancement of the performance of the distribution system. Distribution system generally indicates a radial power supply system and this challenges the integration of DGs with the power system. DG forms an important part in the distribution system planning because of its compact size and minimal investment cost along with environmental benefits. In addition, DG also causes considerable cost reduction being placed near the load and reduced power distribution losses. DG attracts peoples interest in case of low investment cost and uninterrupted power supply in the current scenario of unpredictable load growth.

   Some of the positive impacts of integrating DG to the distribution system has been listed below

   • Reduction in power distribution losses along the feeders.

   • Improvement in the power utility reliability.

   • Enhancement of Voltage profile and Power quality.

   • Prevention of overloading of transmission and distribution system equipment and their effective utilization.

   • Environment friendly.

   • Comparatively lower investment, operational and maintenance cost due to simpler construction and easier operation.

2. MOTIVATION FOR DG GROWTH

   The growth of DGs was motivated by various factors and they have been classified into two main categories, environmental and commercial. These factors are discussed briefly below.

   1. Environmental factor

      1. Reduction in Green House Gas (GHG) emissions

         Since the GHG emissions are a big concern nowadays, the need for renewable energy and Combined Heat & Power units (CHP) led to one of the major factors of DG technology. DGs can make use of renewable energy to produce electrical energy and thus DG integration to the grid can promote the concept of clean & green energy. But integration of these renewable energy DGs to the power system is not an easy job.

      2. Reduction in the necessity for new transmission circuits and large generating plants

         Another factor which has more public attraction for DG from the environmental perspective is the controversy towards the construction of generating stations and new transmission lines. Public awareness towards global warming and other environmental ill

         effects makes people go against the new transmission and generation facilities. This is where the DGs help in eliminating the need for new transmission lines as they being placed locally, near the loads. Also the distribution losses are reduced when they are properly sized and placed on the distribution network. Hence there is no necessity for a new generating unit and the power balance between the production and load can be achieved to an acceptable level.

   2. Commercial factor

   1. Unpredictability in power markets favours low capacity DGs

      In a restructed power system, the concept of competition and choice in electricity endangers all the players from the generating end through transmission and distribution businesses to retailers. The capital investment required to establish new power generation stations are very high. The fluctuating and competitive market environment leads to new generation setups with a low capacity rating where the risk in financial aspects is comparatively small.

   2. DGs are cost effective, efficient and reliable

      Since the DGs are placed locally near the loads, the quality and reliability of the power dispatched can be maintained. Voltage profile improvement and the duration of power interruption to the customers can be taken care automatically when the DGs are allowed and able to be alive on the network during outages i.e., islanding mode.

3. LITERATURE REVIEW

Multiple advantages of renewable energy leads greater increase of their penetration level into the power systems. Most noticeable merits among these are reduced fuel usage, power dispatch to remote places, the need for clean energy sources, the economic benefits for investors in the modern power market and the operational benefits for utilities.

Proper sizing and placement of DGs in the power distribution system play a sigificant role in minimizing power distribution losses. For the calculation of optimal size and an effective placement of DG, an analytical expression has been developed with the concern to minimize the total power loss in the primary distribution systems [1-3]. Loss allocation for customers and distributed generators in radial distribution networks have to be considered. Hence, an approach has been devised to evaluate the technical benefits of DG and then applied to assess the voltage improvement, line-loss reduction and reduction of environmental impact [4-6].

Additionally, the DG owners and the Distribution Companys (DisCo) economic aspects should be considered for the deregulated markets. The approach involves the interests of two different agents: the DG

owner who would like to claim his profit from the power sales to be maximum, and the DisCo, which seeks for the reduction in payments in meeting out the demand. A key to figure out the location and contract pricing of distributed generation units for their dispatchable power in distribution systems has been discussed [7-8].

Costbenefit analysis (CBA) is an organized methodology for evaluating the strengths and weaknesses of any functional requirements in the aspect of economy [9]. Main purpose of CBA are to justify the

A. Objective Function and Constraints

1. Objective Function

   As mentioned earlier, this optimization problem of allocation of DG is solved considering the profit for the DG owner and the reduction of power purchasing cost of DisCo with several constraints and limitations, along with certain operational indices.

   The objective function is given below

   F1 max( IN DG CINV CMAIN COPERATE )

   soundness of the investment or decision.

   F 2 min(C

   SUB

   • CDG )

   (4.1)

   The necessity for the enhancement of computation speed & accuracy led to the invention of Evolutionary Algorithms (EA). EAs are inspired by biological maturation, such as reproduction, evolution and selection. Candidate solutions considered in the optimization problem are important and the fitness function decides the accuracy of the solutions. Also, the computational complexity is due to fitness function assessment. Fitness approximation has to be done to deal with this difficulty [10-13].

   In the proposed approach, Biogeography Based Optimization (BBO) has been utilized. BBO is an EA, which stochastically optimizes a function and iteratively improves the accuracy of the candidate solutions with regard to the fitness function. Complex real-valued functions can be optimized through this method [11].

   1. Components of the Maximizing Function of DG Owners Profit

      This section gives the details of the various costs involved in the DG owner profit maximization function.

      • DG Investment Cost: This indicates the initial investment involved in the installation of the DG.

        =1

        = , (4.2)

      • DG Maintenance Cost: This cost includes the repair, restore and maintenance work involved in the long run.

        = ,

        =1 =1

        Since the problem is a multi-objective optimization

        (1+INF_R) (4.3)

        type, Pareto Optimality principle has been implemented.

        1+INT_R

        Pareto optimality is utilized to allocate the resources where there is a difficulty in selecting the optimal solution in a multi-objective problem[12, 13].

        IV. PROBLEM FORMULATION

        The problem wheels around DG placement & sizing in electrical distribution system with the objective of maximizing the profit of DG owner & minimizing the purchasing cost of DisCo. Several constraints have been considered while solving the multi-objective optimization problem, which will be discussed below.Following are the assumptions made for solving the problem:

        • There are no limitations over the terrestrial or fundamental resources to integrate various DG technologies with the distribution system.

        • It is considered as a negative PQ load for the connection from a DG unit through a bus in power-flow analysis.

        • The proposed approach of DG allocation considers the Discos and the DG owners perspectives in the power market.

      • DG Operational Cost: This includes the cost of generation, fuel and others. The equation is a cumulative of all such costs.

        =1 =1

        = ,

        (1+INF_R) (4.4)

        1+INT_R

      • DG Owners Income: This indicates the selling price of the generated power from the DG to the DisCo.

        =1 =1

        = , (1+INF_R) (4.5)

        1+INT_R

   2. Components of the Minimizing Function of DisCos Cost

      This section describes the cost spent by the DisCo on purchase of power from the substation and DG owner.

      • Cost of DG power purchased: This cost is the one which is evaluated earlier as the DG owners income.

      • Cost of Substation power purchased:This indicates the cost of power purchased by the DisCo from the substation.

      ,, = ,,, ,, ,

      =1

      =1

      (4.6)

      1. Algorithm Step 1

         Since the power purchase made every hour t of the

         day, the cost of power purchased from substation is given the equation below.

         Run the distribution power flow for a base case and determine the active and reactive power losses.

         Step 2

         Identify the candidate solutions of DGs size and

         = 24

         (1+INF_R)

         location with minimum power losses using BBO.

         =1

         =1

         ,,

         ,

         1+INT_R

         (4.7)

         Step 3

2. Constraints and Indices

While considering the economic aspects, we also must study the operational constraints of the power system taken into account. Below listed are the operational constraints and indices which are helpful in maintaining the system stability and power quality.

1. Constraints

   Basic constraints are laid upon the following factors:

   • Capital cost of DG owner

   • Capacity of DG

   • Contract Price

   • Active and Reactive Power flow

   • Bus Voltages and Branch Currents

2. Indices

   These indices help to evaluate whether the system is operating in a desired stable state.

   • Total Voltage Profile Index

   • Total Voltage Stability Index

   • Total Power Loss Index

1. MULTI-OBJECTIVE OPTIMIZATION Complex problems are solved with lesser

   computational efforts, time and high accuracy using various optimization techniques, which have gained greater importance nowadays. Various evolutionary algorithms that imitate natures behavior such as genetic algorithm, particle swarm optimization, etc., are widely used in varied fields. BBO is one of the EA found by Dan Simon in 2008. This optimization technique is analogous to the biogeography theory. The immigration rate and the emigration rate are functions of the number of species in the habitat.

   From the candidates, check for optimal set from multi-objective optimization using Pareto optimality principle, where DG owners profit is maximized and DisCos cost is minimized using equation (4.1).

   Step 4

   Run the power flow by assigning the appropriate size and location for DGs and calculate the active and reactive power losses.

   Step 5

   Repeat the steps 3 and 4 for all possible candidate solutions. Also calculate the operational indices TVPI, TVSI, TPLI for the particular candidate and check if the candidate is satisfy the limits. If not, ignore the candidate and go to step 2.

   Step 6

   Calculate and display the result for thebest possible candidate solution.

   1. Flowchart

   Fig 1: Flowchart of the proposed approach

2. RESULTS AND DISCUSSIONS

   In order to evaluate the effectiveness of the proposed approach, it has to be tested on a standard test system. Here, the algorithm is tested on IEEE 33-bus radial distribution system (Fig 2) and the results are discussed below. Table has the values used for the initialization of the parameters. The following cases not only show the potency of the proposed method, but also give a comparative analysis which proves the DGs in the distribution network has various merits as mentioned earlier.

   TABLE I INITIAL PARAMETERS

   Parameters	Values
   b (f/km.year) CF NN NBUS Nb INT_R(%) INT_F(%) Th Td Max(CPDG)($/MWh) Min(CPDG)($/MWh) Pmax (MW) Pmin (MW)	0.12
   	1
   	20
   	33
   	32
   	12.5
   	9
   	8760
   	365
   	50
   	45
   	1
   	0

   • Case I: This is the base case where the DGs were not placed and evaluating the power distribution losses & the cost of power bought by DisCo from the substation.

   • Case II: Firstly, the sizing and placement for one DG has been simulated and the relevant losses and costs are calculated. Then the DG count has been increased to three and the results were compared.

   Fig 2: IEEE 33-bus radial distribution test system TABLE II

   VARIOUS DG COSTS

   Parameters	Values
   DG investment cost ($/MW) DG maintenance cost ($/MWh) DG operation cost ($/MWh)	318000
   	7
   	29

   Fig 3: Voltage profile without & with DG

3. CONCLUSION

In this paper, a technique which involves Biogeography Based Optimization for optimum DG sizing and placement in the power distribution system has been given. Objectives such as voltage profile and voltage stability improvement and reduction of transmission losses have been taken into account, but the important objective is maximizing the profit of DG owner and minimizing cost of DisCo. This algorithm is convenient in such a way that it holds good for any system with suitable modifications. On testing with IEEE 33-bus distribution system, it has been detected that, there is a considerable voltage profile improvement in the system and there is a notable system power loss reduction. Also the computational time is lesser and accuracy is better when compared with other evolutionary algorithms.

Table Iii

Comprehensive Results Obtained From The Proposed Methodology

REFERENCES

1. T. Gozel and M. H. Hocaoglu, An analytical method for the sizing and sitting of distributed generators in radial systems, Elect. Power Syst. Res., vol. 79, no. 6, pp. 912918, 2009.

2. N. Khalesi, N. Rezaei, and M. R. Haghifam, DG allocation with application of dynamic programming for loss reduction and reliability improvement, Int. J. Elect. Power Energy Syst., vol. 33, no. 8, pp. 288295, 2011.

3. J. Lopez-Lezama, J. Contreras, and A. Padilha-Feltrin, Location and contract pricing of distributed generation using a genetic algorithm, Int. J. Elect. Power Energy Syst., vol. 36, no. 15, pp. 117126, 2012.

4. T. Ackermann, G. Anderson, and L. S. Soder, Distributed generation: Adefinition, Elect. Power Syst. Res., vol. 57, no. 3, pp. 195204, 2011.

5. W. Caisheng and M. H. Nehrir, Analytical approaches for optimal placement of distributed generation sources in power systems, IEEE Trans. Power Syst., vol. 19, no. 4, pp. 20682076, Nov. 2004.

6. A. Soroudi, M. Ehsan, R. Caire, and N. Hadjsaid, Hybrid immune- genetic algorithm method for benefit maximization of distribution network operators and distributed generation owners in a deregulated environment, IET Gen., Transm. Distrib., vol. 5, no. 9, pp. 961972, 2011.

7. F. S. Abu-Moutiand and M. E. El-Hawary, Optimal distributed generation allocation and sizing in distribution systems via artificial bee colony algorithm, IEEE Trans. Power Del., vol. 26, no. 4, pp. 2090 2101, Oct. 2011.

8. M. F. Akorede, H. Hizam, I. Aris, and M. Z. A. Ab Kadir, Effective method for optimal allocation of distributed generation units in meshed electric power systems, IET Gen., Transm. Distribut., vol. 5, no. 1, pp. 276287, 2011.

9. S. Aruul Vizhiy, R.K.Shanti, Biogeography Based Optimal Placement

   of Distributed Generation Units in Distribution Networks, ICEEOT,

   978-1-4673-9939-5/ 2016.

10. Dan Simon "Biogeography-Based Optimization" IEEE Trans. Comp.Vol.12.No.6, December 2008.

11. Sampath Kumar Bodapatla, Dr.H.P.Inamdar, "Loss Reduction by Optimal Placement of Distributed Generation on a Radial feeder", ACEEE Int. J. on Electrical and Power Engineering, Vol. 02, No. 01, Feb 2011.

R. Sai Lakshmi Surya was born in Ambattur, Chennai, in 1989. She completed her Bachelor of Technology (B.Tech) in Electrical & Electronics Engineering from SRM University, Kattankulathur in 2011. She completed her Master of Engineering (M.E) in Smart Energy Systems at Annamalai University, Chidambaram.

Dr. R. Neela, Professor, Dept of Electrical Engineering, Annamalai University was born on 6th May, 1969. She received the B.E. degree in Electrical Engineering, the M.E. degree in Power Systems and the Ph.D. degree in electrical engineering from Annamalai University, Annamalai Nagar, Chidambaram.

Her Area of Specialization is Field Theory, EHV AC & DC Transmission System, Electrical Measurements, Control Systems, Transmission & Distribution, Microprocessor and their applications, Digital Simulation of Power Systems, State Estimation & Security Control of Power Systems and Expert Systems. She has published many notable International publications in the area of Power Systems.

Dr. Ghadir Radman, Professor, Centre for Energy System Research, obtained his Ph.D. in electrical engineering from the Tennessee Technological University, USA. His research interests include power system dynamics, power flow control, FACTS, distributed generations and large-scale systems.