基于GRNN的特高压直流输电线路故障识别方法

打开文本图片集

关键词：广义回归神经网络；特高压直流输电线路；继电保护；广义S变换；故障识别DOI：10.15938/j. jhust.2025.02.014中图分类号：TM723 文献标志码：A 文章编号：1007-2683（2025）02-0131-09

Abstract：Aprotection methodforultra-highvoltage directcurenttransmisson linesbasedongeneralizedregressonneural network（GRNN）isproposedtoaddresstheisuesofeasyrejectionandlongfault detectiontimeinultra-highvoltagedirectcurrent protection.FirstlyasedonthegneralizedStransfo，thefaultcharacterisiciformationintefrequencydomainistaindto constructhe input data for GRNN.Secondly，thechaosquantum particleswarm optimization（CQPSO）algorithm isused tooptimize theparametersofthegeneralizedregressonneuralnetwork，foranidealnetworkmodelbasedontheprincipleofthelowestfinss function，andbeterlathultcharacterissofthulra-hghvageDCtrasmssonlne.TeSofaxlasifierisuiledto clasifydep-levelfeatures，dentifingfaultsasexteal，us，orlinefults，andpolarizingtemintopositive，negative，orbipolar faults，thenoutpuigrecogitioesultsinally，heultra-higagediecturnt tasmissnodelbuiltineCA/C simulatioenvronmentisvalidated，andtevadationresultsshowedthattheproposedmethodhasgoodpeformanceinfaultetection andfaultpoleselectionofultra-highvoltagedirectcurenttransmissionlinerelaprotectionComparedtotradionalonolutional neuralnetwoks，generalizedregressionneuralnetworks，upportvectormacines，andotermethods，thefultrecognitionacyof the proposed method in this paper has been improved by 6.6% ， 0.65% ，and 7.69% ，respectively，meeting the requirements of protection speed and reliability.

Keywords：generalizedregressonneuralnetwork；UHVDCtransmissionline；relayprotection；generalizedS-transform；faulti dentification

0 引言

随着新型电力系统建设的推进，特高压直流输电由于其传输容量大、传输距离远、线路损耗少等显著优势，可以很好地接入分布式电源和储能装置，进而推进我国“双碳”目标，故研究特高压直流输电技术已经成为热点。（剩余13451字）