基于改进pso算法的配电网无功补偿优化配置分析

《基于改进pso算法的配电网无功补偿优化配置分析》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、万方数据ABSTRCTUnreasonabledistributionofreactivepowerwilldecreasevoltagequality，increasenetworklossesandreducepowersystemstability．Thereactivepoweroptimizationcanintegrateexistingresources，maximumincreaseeconomicbenefitofthesystemandcustomer．Sothatthestudyofthepro

2、blemofreactivepoweroptimizationhasthegreatsignificanceinthetheoryandpracticalapplication．Afterthispaperanalyzestheadvantagesanddisadvantagesoftheexistingoptimizationalgorithm；animprovedParticleswarm0ptimizationalgorithmforthemathematicalmodelcorrespondstotheact

3、ualcalculationisoptimized．Theadvantagesofsimplicityandeasyimplementationofparticleswarmalgorithmhavebeenvalidatedinscienceandengineeringfields．However，theweaknessesofparticleswarmalgorithmarethesameasotherevolutionaryalgorithm’S，suchaseasytofallintolocalminimum

4、，prematureconvergence，etc．ThereasonsofdisadVantagesofParticleSwarmOptimization(PSO)Algorithmwereanalyzed，andaChaoticSequence-CosinePSO(CS—CPSO)algorithmhasbeenproposed．Itmakestheinitialparticlestotraversetheentiresearchspacetousethepopulation-initializedofchaot

5、icsequencesinthealgorithm．Andthatincreasesthediversityofinitialpopulation．ItmakestheparticlesintheearlytohavetheabilityofstrongerglobalsearchthattheinertiaweightoftheSPSOwaschangedbecauseofcosinefunctionsnonlinearity．Alongwiththefrequencyofiterationsincreasing，

6、theinertiaweightdecreases，andthenthelocalsearchcapabilityofparticlereinforced．Theaccuracyofthealgorithmimproves．TheparticleshavestrongabilityofsociallearningatanearlystagebecausethelearningfactorwaschangedbythecosinefunctionofnonlinearsymmetriC，0therparticleswi

7、lldrawc10setotheoptimalrapidly．Particleitselflearningabilityisenhancedlatterly．Itspeedsuptheconvergenceofthealgorithm；BacterialchemotaxishasbeenintroducedintheCS—CPSOalgorithm，whichmaintainsthediversityofpopulation，andpreventstheparticlesfallintolocaloptimumina

8、certainextent．TheCS—CPSOalgorithmissimulatedandanalyzedbyfivetestfunctions，comparedwiththeoriginalparticleswarmoptimizationalgorithmandstandardparticleswarmoptimizationalgor