电网技术
電網技術
전망기술
POWER SYSTEM TECHNOLOGY
2015年
1期
143-149
,共7页
李雅洁%孟晓丽%宋晓辉%史常凯
李雅潔%孟曉麗%宋曉輝%史常凱
리아길%맹효려%송효휘%사상개
小电流接地系统%单相接地故障%故障选线%最优FIR滤波器%层次聚类
小電流接地繫統%單相接地故障%故障選線%最優FIR濾波器%層次聚類
소전류접지계통%단상접지고장%고장선선%최우FIR려파기%층차취류
small current grounding system%single-phase earth fault%fault line detection%optimal FIR filter%hierarchical clustering
由于故障特征不明显、噪声干扰、配电网不平衡运行状况影响等问题,现有小电流接地系统单相接地故障选线方法在可靠性与普适性方面仍存在不足。提出一种小电流接地系统单相接地故障选线方法,该方法依据故障暂态量的频率特性,通过最优 FIR 滤波器滤除干扰因素,提取有效故障特征,并利用层次聚类法,基于滤波后波形,检出变化趋势不同于其他线路的单根馈线,即为故障线路。仿真实验表明,较之于常用的小波算法,该方法避免了对数据的大量逐层分解,在各种故障状况下均可可靠检出故障线路,且降低了采样频率需求,节约系统资源,易于工程实现。
由于故障特徵不明顯、譟聲榦擾、配電網不平衡運行狀況影響等問題,現有小電流接地繫統單相接地故障選線方法在可靠性與普適性方麵仍存在不足。提齣一種小電流接地繫統單相接地故障選線方法,該方法依據故障暫態量的頻率特性,通過最優 FIR 濾波器濾除榦擾因素,提取有效故障特徵,併利用層次聚類法,基于濾波後波形,檢齣變化趨勢不同于其他線路的單根饋線,即為故障線路。倣真實驗錶明,較之于常用的小波算法,該方法避免瞭對數據的大量逐層分解,在各種故障狀況下均可可靠檢齣故障線路,且降低瞭採樣頻率需求,節約繫統資源,易于工程實現。
유우고장특정불명현、조성간우、배전망불평형운행상황영향등문제,현유소전류접지계통단상접지고장선선방법재가고성여보괄성방면잉존재불족。제출일충소전류접지계통단상접지고장선선방법,해방법의거고장잠태량적빈솔특성,통과최우 FIR 려파기려제간우인소,제취유효고장특정,병이용층차취류법,기우려파후파형,검출변화추세불동우기타선로적단근궤선,즉위고장선로。방진실험표명,교지우상용적소파산법,해방법피면료대수거적대량축층분해,재각충고장상황하균가가고검출고장선로,차강저료채양빈솔수구,절약계통자원,역우공정실현。
Due to the impacts of unconspicuous fault characteristics, noise interference, imbalanced operation condition of distribution network and so on, there are insufficiencies in reliability and universality in existing fault line detection methods for single-phase earth fault occurred in small current grounding system. A new fault line detection method for single-phase earth fault occurred in small current grounding system is proposed. Based on frequency characteristics of fault transient quantity the proposed method filters interference factors by finite impulse response (FIR) filter and extracts available fault characteristics; utilizing hierarchical clustering and based on the filtered waveform, the variation trend of transient zero-sequence capacitive current component at the beginning end of the faulty feeder is different from variation trends other healthy feeders, so the faulty feeder can be detected. Simulation results show that using the proposed method the data decomposition step by step can be avoided, and the faulty feeder under various fault conditions can be reliably detected, besides, the demand on the sampling frequency is decreased, thus the system resource can be saved and the proposed method is easy to implement.