基于短期负荷预测的配电网动态无功优化策略分析-analysis of distribution network dynamic reactive power optimization strategy based on short-term load forecasting.docx

下载文档 降价啦

9
0
约4.97万字
约 61页
2018-06-05 发布于上海
举报
版权申诉
保障服务

基于短期负荷预测的配电网动态无功优化策略分析-analysis of distribution network dynamic reactive power optimization strategy based on short-term load forecasting.docx

1、本文档共61页，可阅读全部内容。
2、有哪些信誉好的足球投注网站（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。

基于短期负荷预测的配电网动态无功优化策略分析-analysis of distribution network dynamic reactive power optimization strategy based on short-term load forecasting

ABSTRACTAlongwiththedevelopmentofthenationaleconomyandtherapidgrowthoftheload,thegridquality,economicefficiencybecomemoreandmoreimportant,usersalsoputforwardhigherrequesttopowerquality.Asthemostimportantpartforusers，themanagementofreactivepowerandvoltagedistributionsystemisparticularlyimportant，whichisdirectlyrelatedtoeconomy,securityandstabilityofthepowergrid.Reactivepoweroptimizationofdistributionnetworkisoneoftheimportantmeanswhichguaranteethepowerquality，systemsecurityandstability.Butthetraditionalreactivepoweroptimizationforacertaintimeisnolongersuitableforrequirementsofonlineanalysisandactualapplication.Becauseloadisconstantlychanginginadispatchingcycle,andinfactcontrolequipmentalsohasalimitoperationtimesperadispatchingcycle,therefore,itisnecessarytostudydynamicreactivepoweroptimizationconsideringtheloadchangeanddeviceoperationconstraints.Higherprecisionofshort-termloadforecasting,amoreeffectiveglobaloptimizationmethod,thereasonablecontrolstrategyforcontrolequipmentareurgentquestionneedtobesolved.Focusonthedynamicreactivepoweroptimizationofregionaldistributionnetwork,thepapercarriedonathoroughresearch.Dynamicshort-termloadforecastingmodelisestablished.Thequantumparticleswarmoptimizationalgorithm(QPSO)isintroducedintothedynamicreactivepoweroptimization,andactionprivilegedynamicadjustingisproposedtosolvethespace-timecouplingproblemsindynamicreactivepoweroptimization.Short-termloadforecastingisthepreconditionofdynamicreactivepoweroptimization.Aimedatthehighlynonlinearcharacteristicsoftheloadforecast,thepaperselecttheRBFneuralnetwork,however,conventionalneuralnetworkpredictionmodelisstaticmodel,thenetworkparametersaredeterminedafterthetraining,theneuralnetworkstructureandparametersarenolongerchanging,therefore,astimegoeson,andaccidentalfactors,thecurrentstructureandparametersoftheneuralnetworkismoreandmorenotmeettherequirementsofthecurrentloadforecastingaccuracy,andthepredictionerrorwillbemoreandmorebig.Therefore,thearticleputsforwardadynamicerrordiscriminantfunct