Meysam Soleymani, Sadegh Soleymani, Mahdiyeh Eslami, Hadi .

2y ago
34 Views
4 Downloads
1.21 MB
5 Pages
Last View : 1m ago
Last Download : 3m ago
Upload by : Luis Waller
Transcription

1931INTERNATIONAL JOURNAL OF SCIENTIFIC & ENGINEERING RESEARCH, VOLUME 5, ISSUE 1, JANURAY 2014ISSN 2229-5518Application of Evolutional Algorithms for Productivity Improvement and Reductionof Active Power Losses in Radial Distribution SystemMeysam soleymani, Sadegh soleymani, Mahdiyeh eslami, Hadi zayandehroodiAbstract: Power losses in distribution system have become the most concerned issue in power losses analysis in anypower system. In the effort of reducing power losses within distribution system, reactive power compensation has becomeincreasingly important as it affects the operational, economical and quality of service for electric power systems. This paperpresents a new method for determining capacitor placement in radial distribution systems. The capacitor placementproblem consist of finding size & places to install capacitor bank in an electrical distribution network aiming to reducelosses and voltage profile Improvement due to the compensation of the reactive component of power flow.The capacitor placement is hard to solve in sense of global optimization due to the high non linear and mixed integerproblem. To solve the problem efficiently, this paper by Genetic Algorithm (GA) and Colonal Selection Algorithm (CSA)that is one of the efficient optimization methods.Key-word: Capacitor placement, Genetic Algorithm, Colonal Selection Algorithm, Back/Forward Sweep, Loss Reduction.—————————— ——————————1 IntroductionIJSERCapacitors are often installed in distribution systemfor reactive power compensation to carry out powerand energy loss reduction, voltage regulation, systemsecurity improvement and system capacity release.Economic benefits of the capacitor depends mainly onwhere and how many capacities of the capacitor areinstalled and proper control schemes of the capacitorsat different load levels in the distribution system [1].A variety of methods have been devoted to solvingthe capacitor placement problem. A capacitorallocation techniques can be found in [2]. Amongvarious approaches, the metaheuristics play a relevantrole, since exact optimization methods are not suitablefor tacking real world instances. Focusing only onmetaheuristic methods, [3-6] propose differentmethods for capacitor placement problem.This article presents a new approach base on aGenetic Algorithm (GA) & Colonal SelectionAlgorithm (CSA) for solving the capacitor �——Meysam soleymani, Department of Electrical Engineering Science andResearch Branch, Islamic Azad university, kerman, Iran. (E-mail:soleymani.m48@gmail.com).Sadegh soleymani, Department of Electrical Engineering Science andResearch Branch, Islamic Azad university, kerman, Iran. (E-mail:soleymani.sa@gmail.com).Mahdiyeh eslami, Department of Electrical Engineering Science andResearch Branch, Islamic Azad university, kerman, Iran. (E-mail:mahdiyeh eslami@yahoo.com).Hadi zayandehroodi, Department of Electrical Engineering Science andResearch Branch, Islamic Azad university, kerman, Iran. (E-mail:h.zayandehroodi@yahoo.com).problem. In this paper presents a very fast and simplepower flow problem for solving the capacitorplacement. The main contribution of this study iscombination of GA & CSA. To demonstrate theeffectiveness of proposed algorithm, this method isapplied to a real radial distribution feeder. Incomparison with conventional power flow method interms of solution accuracy and computational time.2 Problem Formulations2.1. Objective functionThe objective function of the problem can beexpressed as follows to minimize the capacitorinvestment cost and system energy loss and voltagesat different buses within the limits.:IMin f (i, c) C (i, Qc) Ploss (i, Qc)i 1Qc : discrete variables, for fixed typed capacitor:Inthis formulation, sizing vector whose components aremultiples of the standard size of one capacitor bank.C i (i, Qc)represent the investment cost associatedwith the capacitor installed at node i.Plossis thepower loss.2.2. Evaluation of fitness functionThe evaluation of fitness function is a procedure todetermine the fitness of each string in the population.Since the GA & CSA proceeds in the direction ofevolving best-fit strings and the fitness value is theIJSER 2014http://www.ijser.org

INTERNATIONAL JOURNAL OF SCIENTIFIC & ENGINEERING RESEARCH, VOLUME 5, ISSUE 1, JANURAY 2014ISSN 2229-5518only information available to the GA & CSA, theperformance of the algorithm is highly sensitive to thefitness values. The fitness function F, which has beenchosen in this problem, is1 (, )3 Proposed Algorithm3.1. Implementation of CSA3.3. Load flow methodIn order to evaluate the power distribution networkand examine the effectiveness of possible changes onsystem in network programing state, it is necessary toperform power flow analysis on the network which isprobably the most important of all networkcalculations.3.3.1 Backward Forward Sweep methodClonal Selection principle is a form of naturalselection [3] and it describe the essential featureswhich contain adequate diversity, discrimination ofself and non-self and long-lasting immunologicmemory. The main idea of clonal selection theory liesin that the antibodies can selectively react to theantigens, which are the native production and spreadon the cell surface in the form of peptides. Whenexposed to antigens, the immune cells that recognizeand eliminate the antigens will be selected and arousean effective response against them. The reaction leadsto cell proliferating clonally and the colony has thesame antibodies. Consequently, the process of clonalselection actually consists of three main steps:Clone:descend a group of identical cells from a singlecommon ancestor through asexual propagation.Mutation: gain higher affinity mainly throughhypermutation [4]. Selection: select some excellentindividuals from the sub-population generated byclonal proliferation.Assuming the objective function and restrainingconditions of optimization are the antigens invadingthe body and candidate solutions are the antibodiesrecognizing antigens, then the process of optimizationcan be considered as the reaction between antigensand antibodies, and the affinity between the antigensand the antibodies are the matching degree betweenobjective function and solutions.The methods proposed for solving distribution powerflow analysis are essentially classified into threecategories: direct methods, backward/forward sweepmethods and Newton-Raphson (NR) methods. In thispaper we utilize the back/forward sweep methodwhich is simple, flexible, reliable, and didn't needJacobian matrix and its inverse and have highconvergence speed.Study of power flow system isusually performed to achieve the following goals:a) Sufficient Active and reactive power flow innetwork branches.b) To avoid overloading in different sections.c) Effect of adding new parts to the system understudy.d) To analysethe loss in different section at thecritical situation.e) The optimal power flow analysis and assignment.f) Optimization of system losses.IJSER3.2. Implementation of GAIn artificial intelligence, an Evolutionary Algorithm(EA) is a subset of Evolutionary Computation thatinvolves combinatorial optimization problems. GA’sare generalized search algorithms based on themechanics of natural genetics[13]. GA maintains apopulation of individuals that represent the candidatesolutions to the given problem. Eachindividual in thepopulation is evaluated to give some measure to itsfitness to the problem from the objectivefunction.Genetic Algorithms combine solution evaluation withstochastic operators namely, selection, crossoverandmutation to obtain optimality.The Back/Forward Sweep algorithm is as below:First, the initial voltage of all buses is consider to be 10. With the known primary load of each line, thecurrent of last bus is calculated as below:s p JQ V I I vV In each iteration, new voltage and load flow iscalculated.New voltage of each bus is calculated usingkvl law and by starting from first bus.Having new voltages, new current of each line isobtained utilizing the last bus.This process willcontinue until the maximum total voltage differenceof all buses is greater than the pre-specified value[7,8][7, 10]. The flowchart of proposed Back/ForwardSweep method is shown in Fig (1).IJSER 2014http://www.ijser.org1932

INTERNATIONAL JOURNAL OF SCIENTIFIC & ENGINEERING RESEARCH, VOLUME 5, ISSUE 1, JANURAY 2014ISSN 2229-5518A number of tests on the performance of the proposedalgorithm have been carried to determine the mostsuitable GA & CSA parameters setting.Input the dataCalculation of currents ofsections when voltage ofTable 1: Specification of test networkbuses 1Calculation of currentsCalculation of voltage ofbased on new voltagebuses from one to endof busesAnalysisNoaccuracy ofcalculationsYesPrint magnitude of voltage, losses and current linesFig Fig.1 : Flowchart Back/Forward Sweep methodpowerKW .81.810.84.57.21.810.84.57.24.510.84.5Impedance of BusNumberSection 10.016 j0.03920.015 j0.0430.015 j0.03840.017 j0.04150.017 j0.04160.015 j0.03970.013 j0.03780.014 j0.0490.017 j0.042100.015 j0.039110.016 j0.041120.016 j0.041130.015 j0.038140.015 j0.038150.016 j0.041160.013 j0.044170.015 j0.039180.012 j0.035190.014 j0.039200.015 j0.041210.016 j0.041220.016 j0.041230.016 j0.041240.016 j0.041250.016 j0.041260.016 j0.041270.016 j0.041280.016 j0.041290.016 j0.04130IJSER4 Problem SolutionIn order to test the proposed algorithm, a real radialdistribution network has been considered, Figure 2and Table 1 show the single line diagram andspecifications of test network, respectively.Fig.3: Voltage profiles of system for twoMethodsFig.2 Test network single line diagramIJSER 2014http://www.ijser.org1933

INTERNATIONAL JOURNAL OF SCIENTIFIC & ENGINEERING RESEARCH, VOLUME 5, ISSUE 1, JANURAY 2014ISSN 2229-5518of investment cost of losses for 5 years was equal tocost of investments and the fitness will be 28550000Rials for 20 years period.Table 4: cost & size capacitors(a)methodGA3 4 15 26 27 22 30LocationCapacitor size (kvar)450 600 750 900 1050 12001350Fig.4: power loss profiles in each branchs of systemsfor two MethodsTable 2: Execution time and number of IterationGA MethodCSA MethodTimeTime(sec)Iteration(sec)IterationFeeder30 cost0.253 0.28 0.276 0.1890.238 0.1760.217(b)methodCSA3 8 10 16 22 25 28LocationIJSERCapacitor size (kvar)300 300 450 600 750 9001050 1200 12005 Cost AnalysisFor economic evaluating of the proposed algorithm,the following equation were considered for theeconomic gain:Annual Gain 8760cost0.36 0.36 0.253 0.28 0.2760.189 0.238 0.1760. 1760 kwh.cost PlossWhere:Annual Gain: the annual economic gain with using thecapacitors regard to losses reduction for one year.8760: the conversion factor of power losses to energylosses.Kwh.cost: the cost of energy. Ploss :Capacitor(Toman/kvar)power losses reduction regard to use ofcapacitors.Annual Cost: ( ( .CapCost ) /(1 (1 /(1 i ) ))Where:Annual Cost: the total cost of capacitors and theiraccessories for one year. : Investment period, :Interest rate. According to above relations, the fitnessfunction can be formulated as:Fitness Annual Gain – Annual CostIn this study for the example network and withconsidering of the cost in IRAN, for the planningstudy 20 years long and interest rate 20% , inflationrate25% , kwh. cost 200 Rials, the reduction costThese tables 4 show the CSA algorithm is powerfulfor allocation and sizing the capacitor bank in the testnetwork.6 ConclusionIn this paper, implementation of GA to the optimalplacement of capacitor bank has been illustrated. Theeffectiveness of the used power flow method to solvethe capacitor placement problem has beendemonstrated through the numerical example. Theresult showed the GA is a proper optimizationmethod for optimal placement of capacitors bank iNradial distribution network. Furthermore, It wasshowed the used power flow method in capacitorplacement problem is better than conventional powerflow method in terms of solution quality andconsumed time. The economic study showed theinvestments costs will be compensated in a few yearsby reduction costs of losses7 Reference[1] Turan Gonen, Electric power Distribution systemsEngineering, Mc.Graw-Hill International Edition,1986.[2] Mahdi Mozaffari Legha, Rouhollah AbdollahzadehSangrood, Ardalan Zargar Raeiszadeh, MohammadIJSER 2014http://www.ijser.org1934

INTERNATIONAL JOURNAL OF SCIENTIFIC & ENGINEERING RESEARCH, VOLUME 5, ISSUE 1, JANURAY 2014ISSN 2229-5518Mozaffari Legha,, “CONDUCTOR SIZE SELECTION INPLANNING OF RADIAL DISTRIBUTION OMPETITIVEALGORITHM”,International Journal on Technical and Physical Problemsof Engineering (IJTPE), Issue 15, Vol. 5, No. 2, pp. 65-69,June 2013[3] Chiang H.D., and Author, Optimal capacitorplacement in distribution systems, IEEE Trans on PowerDelivery, Vol.5, No.2, 1990,pp.643-649.[4] Huang Y.C., and Author, Solving the capacitorplacement problem in a radial distribution system usingtabu search approach, IEEE Trans on Power Systems,Vol.11, No.4, 1996,pp.1868-1873.[5] Gallego R.A., and Author, Optimal capacitorplacement in radial distribution networks, IEEE Trans onPower Systems, Vol.16, No.4, 2001,pp.630-637.[6] Sundhararajan S, and Author, Optimal selection ofcapacitors for radial distribution systems using a geneticalgorithm, IEEE Trans on Power Systems, Vol.9, No.3,1994,pp.1499-1507.[7] Jen-hao T, A Network- topology- based three phaseload flow for distribution systems, Proc. Natl. Sci.Counc. ROC(A), Vol.24, No.4, 2000,pp.259-264.[8] Goldberg.D.E, Genetic Algorithm in Search,Optimization and Machine Learning, MA, AddisonWesley, 1989.[9] Davis.L, Handbook of Genetic Algorithm, New York,Van Nostrand, 1991.[10] Awadh.B, and Author, A computer-aided processplanning model based on genetic algorithm, ComputerOperational Research, vol.22, no.8, 1995, pp. 841-856.[11] N. E. Chang, " Generalized Equation on LossReduction withShunt Capacitor ", IEEE Trans. on Power Apparatus andSystems, Vol. PAS-91, (S), 1972, pp. 2189-2195.[12] Y. G. Bae, " Analytical Method of CapacitorAllocation on Disbibution Primary Feeders 'I, IEEETrans. on Power Apparatus and Systems, Vol. PAS-97,No. 4, July/Aug.l978, pp. 1232-1237.[13] S. H. Lee, J. J. Grainger, " Optimum Placement ofFixed and Switched Capacitors on Primary DistributionFeeders ", lEEE Trans. on Power Apparatus and Systems,Vol. PAS-100, No.1, Jan. 198 1, pp. 345-352.[14] J. J. Grainger, S. H. Lee, '' Optimum Size andLocation of Shunt Capacitors for Reduction of Losses onDistribution Feeders ", IEEE Trans. on Power Apparatusand Systems, Vol. PAS-100, No.3, March1981, pp. 1105-1118.[15] M. Mozaffari Legha, (2011) Determination ofexhaustion and junction of in distribution network andits loss maximum, due to geographical condition, MS.cThesis. Islamic Azad University, Saveh Branch, MarkaziProvince, Iran.[16] J. J. Grainger, S. H. Lee, " Capacity Release by ShuntCapacitor Placement on Distribution Feeders: A NewVoltage-Dependent Model ", IEEE Trans. on PowerApparatus and Systems, Vol. PAS-101, No.5, May 1982,pp. 1236-1244.[17] Ng, H.N., Salama, M.M.A., and Author,Classification of capacitor allocation technique, IEEETrans on Power Delivery, Vol.15, No.1, 2000, pp.387-392IJSERIJSER 2014http://www.ijser.org1935

Meysam soleymani, Sadegh soleymani, Mahdiyeh eslami, Hadi zayandehroodi Abstract: Power losses in distribution system have become the most concerned issue in power losses analysis in any power system. In the effort of reducing power losses within distribution

Related Documents:

Ghaderzadeh, Sadegh; Ghorbani-Asl, Mahdi; Kretschmer, Silvan; Hlawacek, Gregor; Krasheninnikov, Arkady V. Channeling effects in gold nanoclusters under He ion irradiation Published in: Nanotechnology DOI: 10.1088/1361-6528/ab4847 Published: 01/01/2020 Document Version Publisher's PDF, also known as Version of record Please cite the original .

Functional and Structural Characterization of SARS-Cov-2 Spike Protein: An In Silico Study Hadi Sedigh Ebrahim-Saraie1, Behzad Dehghani2, Ali Mojtahedi3, Mohammad Shenagari3, Meysam Hasannejad-Bibalan3* O

846 Review EMHJ – Vol. 26 No. 7 – 2020 Prevalence of self-medication in university students: systematic review and meta-analysis Meysam Behzadifar,1 Masoud Behzadifar,2 Aidin Aryankhesal,3 Hamid Ravaghi,3 Hamid Reza Baradaran,4 Haniye Sadat Sajadi,5 Mojtaba Khaksarian6 and Nicola Luigi Bragazzi7 1Social Determinants of Health Res

This issue’s quote recalls the final words recited—for centu-ries—during the Passover seder. What’s your favorite quote? Send your suggestion to The Voice, c/o the Temple Israel office. Mazal Tov To. . . Louise and Dr. Yousef Soleymani on the birth of their grandso

Testing the Traditional CAPM and MCAPM on Tehran Stock Exchange Ebrahim Abbasi 1, Meysam Kaviani 2, Ebrahim Farbod 3 1 Faculty of Social Sciences & Economics, Alzahra University, Tehran, Iran 2 PhD student of Financial Management, Islamic Azad University, Iran 3Department of Management, Payame Noor University, Iran P A P E R I N F O A B S T R A C T .

RESEARCH ARTICLE Open Access Health impact assessment of air pollution in megacity of Tehran, Iran Kazem Naddafi1,2, Mohammad Sadegh Hassanvand1,2*, Masud Yunesian1,2, Fatemeh Momeniha1,3, Ramin Nabizadeh1,2, Sasan Faridi1 and Akbar Gholampour1 Abstract The aims of the present study were to provide quantitative data on the impact of air pollution on the health of

Mohsen Bahramipanah, Amirreza Mahigir, Mohammad Sadegh Abrishamian, Seyed Abdollah Mirtaheri 1210 230- "Extended Pret a Voterwith Discretely-Chaotic Family ofPermutations"

ARALING PANLIPUNAN - GRADE 5 Ikalawang Markahan Kahulugan at Layunin ng Kolonyalismo Ang Dahilan ng Espanya sa Pananakop ng Pilipinas Pananakop ng Espanya sa Pilipinas Ang Paglalayag ni Ferdinand Magellan Labanan sa Mactan Ang Ekspedisyon nina Loaisa, Cabot, Saavedra at Villalobos Ekspedisyon ng Legazpi Mga Unang Engkwentro sa Pagitan ng mga Espanyol at Pilipino Kristiyanismo sa Buhay ng mga .