中国电机工程学报
中國電機工程學報
중국전궤공정학보
ZHONGGUO DIANJI GONGCHENG XUEBAO
2013年
28期
130-136
,共7页
梁睿%靳征%王崇林%刘建华
樑睿%靳徵%王崇林%劉建華
량예%근정%왕숭림%류건화
故障测距%行波%时频域分析%单相接地故障%波头识别
故障測距%行波%時頻域分析%單相接地故障%波頭識彆
고장측거%행파%시빈역분석%단상접지고장%파두식별
fault location%traveling waves%time domain and frequency domain analysis%single-phase earth faults%wave front recognition
单端行波检测要实现单相接地故障精确测距,必须确定故障初始波头与其他反射波头的时间差,而准确识别故障反射波头的问题一直没有很好解决。为此,提出了一种精确识别故障点反射波头的方法。首先,在连续小波变换(continuous wavelet tranformation,CWT)自建母小波的单相接地故障信号频域分析的基础上,较为准确地得到故障点的特征频率,从而初步确定故障的距离;然后根据频域分析方法得到的故障初步位置结合 Daubechies(Db)小波对信号进行时域分析,准确判别故障线路对侧母线反射波头到达时刻,进而利用初始波头与对侧母线反射波头时间差实现故障精确测距。大量仿真实验表明,该方法能精确地辨别反映故障距离的对端母线反射波头,提高了行波频域分析的精度与时域分析的可行性,通过对故障暂态信号的时频域综合分析,较好地实现了单端辐射状配电网单端故障测距。
單耑行波檢測要實現單相接地故障精確測距,必鬚確定故障初始波頭與其他反射波頭的時間差,而準確識彆故障反射波頭的問題一直沒有很好解決。為此,提齣瞭一種精確識彆故障點反射波頭的方法。首先,在連續小波變換(continuous wavelet tranformation,CWT)自建母小波的單相接地故障信號頻域分析的基礎上,較為準確地得到故障點的特徵頻率,從而初步確定故障的距離;然後根據頻域分析方法得到的故障初步位置結閤 Daubechies(Db)小波對信號進行時域分析,準確判彆故障線路對側母線反射波頭到達時刻,進而利用初始波頭與對側母線反射波頭時間差實現故障精確測距。大量倣真實驗錶明,該方法能精確地辨彆反映故障距離的對耑母線反射波頭,提高瞭行波頻域分析的精度與時域分析的可行性,通過對故障暫態信號的時頻域綜閤分析,較好地實現瞭單耑輻射狀配電網單耑故障測距。
단단행파검측요실현단상접지고장정학측거,필수학정고장초시파두여기타반사파두적시간차,이준학식별고장반사파두적문제일직몰유흔호해결。위차,제출료일충정학식별고장점반사파두적방법。수선,재련속소파변환(continuous wavelet tranformation,CWT)자건모소파적단상접지고장신호빈역분석적기출상,교위준학지득도고장점적특정빈솔,종이초보학정고장적거리;연후근거빈역분석방법득도적고장초보위치결합 Daubechies(Db)소파대신호진행시역분석,준학판별고장선로대측모선반사파두도체시각,진이이용초시파두여대측모선반사파두시간차실현고장정학측거。대량방진실험표명,해방법능정학지변별반영고장거리적대단모선반사파두,제고료행파빈역분석적정도여시역분석적가행성,통과대고장잠태신호적시빈역종합분석,교호지실현료단단복사상배전망단단고장측거。
Calculation of the time difference between initial and reflected wave fronts originated from fault point is a key issue based on single terminal traveling wave method for single-phase ground fault location. But accurate identification of the reflected wave fronts from fault point have been not solved well yet. The paper proposed a new method for accurate identification of the reflected wave fronts. Firstly, transient signal originated from fault point is analyzed in frequency domain by fault waveform-inferred mother wavelets of continuous wavelet transformation (CWT), the precise characteristic frequency is obtained and fault location preliminarily determined. Secondly, the fault location determined from the previous step is used as a clue for Daubechies (Db) wavelet time domain analysis, and the arrival time of the reflected wave fronts from the far terminal bus is determined. Finally, more precise fault location is achieved based on the time difference between initial wave front from fault point and reflected one from far terminal bus. Many simulations show that reflected traveling waves associated with fault location have been accurately identified, precision in frequency domain analysis and feasibility in time domain analysis are both improved. The method integrates time and frequency domain analysis and gets a more accurate single terminal fault location procedure for radial power distribution network.