高级智能故障信息处理系统中关键技术的开发研究——子站部分

《高级智能故障信息处理系统中关键技术的开发研究——子站部分》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、回线故障测距算法。通过对线路两回线电流的六序分量变换，解决了两回线之间的互感问题。由此得到的六序故障分量网络中，线路两端电流故障分量有效值之比仍然只是线路阻抗、两端系统阻抗和故障距离的函数。在分析各序故障分量之间关系的基础上，通过简单的测距方程求出故障距离，避免了迭代运算带来的不收敛或者伪根问题。大量仿真结果表明，该方法可以准确地实现同杆双回线单回线故障测距，且不受过渡电阻影响。另外，该算法不受故障类型的约束，从而避免故障类型判别错误带来的误差。由于单端测距算法无法获取对侧准确系统阻抗，所以容易受系统运行

2、方式变化的影响。如何准确获取对侧系统阻抗有待进一步的研究。关键字：三次B样条二进小波，故障选线，故障选相，小波能量，故障测距RESEARCHOFKEYTECHNOLOGYOFSUPERIORINTELLIGENTFAULTDIAGNOSISSYSTEM(PARTSUBSTATION)ABSTRACTThemodernpowernetworkdependsoneffectiveanalysisandmanagementofvariousinformation.It’sdifficultformanageme

3、ntpersonaltojudgecorrectlyjustaccordingtovariableinformationofSCADAsystemwhenlarge-areacomplexfaulthappens.Informationgotfromfaultrecorderandprotectionbecomesmoreandmoreimportantinfaultanalysisandsystemresumption.Faultdiagnosissysteminsubstationisnecessar

4、ytofaultanalysisaccordingtofaultrecorddata.Iffault,faultline、faultphaseandfaultpointcanbegotbyanalyzingonrecorddataofsubstation,thenthediagnosisresultcanbetransmittedtodispatchcenterinformoffaultbriefreportandtimelyreferencecanbeprovidedfordutypersonal.Fa

5、ultsignaldetectioninpowersystembasedonwavelettheoryhasimportantmeaning.Howtofindtimewhenfaulthappensisapivotalprobleminfaultanalysisanddetectionofpowersystem.Inthispaper,therankthreeB-splinedyadicwaveletisproposedtofindoutthesingularpointsoffaultcurrentan

6、dthetheorybasisisdiscussedindetail.SimulationfromATPandrecorddataofpowernetworkofEastChinashowsthaterrordetectedbythiswaveletisshorterthan0.6mswhichisshorterthanthatgotfromrecorder.Determinationoffaulttimefromtwo-enddatacanalsobeachievedbywavelet.Onthebas

7、isoftheabovework,theabruptchangecharacteristicofthefaultlinecurrentisanalyzedindetailinvirtueoftheconfiguredtimeofmainprotectandthereclosingintervalwhensimplefaultsoccuronlinesinlocalsubstationandnearbysubstationofEHVmulti-loopselectricnetworkunderseveral

8、circumstancessuchasnormalrelayandbreaker,action-resistantrelayorbreaker,faultyrelayorbreaker.Anewalgorithmbasedonthefaultanalogmessageisproposedatlast.Thealgorithmconquersthelimitationoftraditionalfaultlineselection