基于时间序列压缩动态时间弯曲距离故障区段定位
|
摘要
针对基于暂态零模电流相关系数法的配电网故障区段定位方法存在数据处理量大且需要时间同步问题,提出一种基于时间序列压缩的动态时间弯曲(DTW)距离的小电流接地故障区段定位方法。先对终端控制单元(FTU)采集的故障录波数据预先处理,只提取故障发生后一个工频周波内的初始值、极值以及两极值点之间数据变化差值最大的两点,再将新的时间序列上传至主站,最后由主站采用动态时间弯曲距离算法求取两相邻FTU零模电流的相似性来进行故障区段定位。通过算法对比分析可得,在抗同步误差能力和信号幅值反应能力上,DTW算法较强于相关系数法;在直流分量表达能力上,两算法效果基本一致。此外,数据压缩后的新时间序列在DTW距离上的相似度表达能力基本与原时间序列相同,但数据处理量减少了至少一半以上。大量ATP/EMTP仿真说明文中所提方法能在不同故障角、接地电阻、故障位置下准确判出故障区段。
Since the fault section location method of transient zero mode current correlation coefficient in distribution network exists the problem with large amount data processing and needs time synchronization,this paper presents a dynamic time warping( DTW) distance method based on time series compression for fault section location. The proposed method only extracts data within a power frequency cycle after fault occurs which includes the initial value,extreme value and two points of maximum difference change between two extreme points. Then,a new time sequence is uploaded to the host,and finally,by the main dynamic time warping distance algorithm calculating the similarity of two adjacent FTU. Through the comparison of the algorithm,the DTW algorithm is better than the correlation coefficient method in the ability of antisynchronization error and signal amplitude. In addition,the similarity calculation of the new time series on DTW distance is the same as the original time series,but the data processing capacity is reduced by more than one half. A large amount of ATP/EMTP simulation shows that the proposed method can accurately find fault segments in different fault angles,grounding resistance and fault location.
Since the fault section location method of transient zero mode current correlation coefficient in distribution network exists the problem with large amount data processing and needs time synchronization,this paper presents a dynamic time warping( DTW) distance method based on time series compression for fault section location. The proposed method only extracts data within a power frequency cycle after fault occurs which includes the initial value,extreme value and two points of maximum difference change between two extreme points. Then,a new time sequence is uploaded to the host,and finally,by the main dynamic time warping distance algorithm calculating the similarity of two adjacent FTU. Through the comparison of the algorithm,the DTW algorithm is better than the correlation coefficient method in the ability of antisynchronization error and signal amplitude. In addition,the similarity calculation of the new time series on DTW distance is the same as the original time series,but the data processing capacity is reduced by more than one half. A large amount of ATP/EMTP simulation shows that the proposed method can accurately find fault segments in different fault angles,grounding resistance and fault location.
引文
[1]郭上华,肖武勇,陈勇,等.一种实用的馈线单相接地故障区段定位与隔离方法[J].电力系统自动化,2005,29(19):79-81.Guo Shanghua,Xiao Wuyong,Chen Yong,et al. Practical method of locating and isolation for single phase grounding fault section of feeders in distribution networks[J]. Automation of Electric Power Systems,2005,29(19):79-81.
[2]肖白,束洪春,高峰.小电流接地系统单相接地故障选线方法综述[J].继电器,2001,29(4):16-20.Xiao Bai,Shu Hongchun,Gao Feng. Survey of the methods of faults line section for single-phase-to-earth fault in networks with ungrounded neutral[J]. Power Systems Protection and Control,2001,29(4):16-20.
[3]齐郑,杨以涵.中性点非有效接地系统单相接地选线技术分析[J].电力系统自动化,2004,28(14):1-5.Qi Zheng,Yang Yihan. Analysis of technology of fault line selection for single-phase-to-earth faults in neutral point non-effectively grounded system[J]. Automation of Electric Power Systems,2004,28(14):1-5.
[4]马士聪,徐丙垠,高厚磊,等.检测暂态零模电流相关性的小电流接地故障定位方法[J].电力系统自动化,2008,32(7):48-52.Ma Shicong,Xu Bingyin,Xue Yongduan,et al. An earth fault locating method in feeder automation system by examining correlation of transient zero mode currents[J]. Automation of Electric Power Systems,2008,32(7):48-52.
[5]孙波,孙同景,薛永端,等.基于暂态信息的小电流接地故障区段定位[J].电力系统自动化,2008,32(3):52-55.Sun Bo,Sun Tongjing,Xue Yongduan,et al. Single phase to ground fault section locating based on transient signals in non-solidly earthed network[J]. Automation of Electric Power Systems,2008,32(3):52-55.
[6]倪广魁,鲍海,张利,等.基于零序电流突变量的配电网单相故障带定位判据[J].中国电机工程学报,2010,30(31):118-122.Ni Guangkui,Bao Hai,Zhang Li,et al. Criterion based on the fault component of zero sequence current for online fault location of singlephase fault in distribution network[J]. Proceedings of CSEE,2010,30(31):118-122.
[7]宋国兵,李广,于叶云,等.基于相电流突变量的配电网单相接地故障区段定位[J].电力系统自动化,2011,35(10):84-90.Song Guobing,Li Guang,Yu Yeyun,et al. Single-phase earth fault section based on phase current fault component in distribution network[J]. Automation of Electric Power Systems,2011,35(10):84-90.
[8]孙波,张承惠,孙同景,等.基于暂态相电流的小电流接地故障定位研究[J].电力系统保护与控制,2012,40(18):69-74.Sun Bo,Zhang Chenghui,Sun Tongjing,et al. Earth fault location based on transient phase current in non-solidly earthed network[J].Power Systems Protection and Control,2012,40(18):69-74.
[9]王铭,徐玉琴.基于小波变换和BP网络的非故障相暂态电流故障定位[J].电力自动化设备,2006,26(4):25-27.Wang Ming,Xu Yuqin. Locating faulting using transient component of non-fault-phase based on wavelet analysis and BP neural network[J].Electric Power Automation Equipment,2006,26(4):25-27.
[10]孙国强,卫志农,唐利锋,等.多目标配电网故障定位的Pareto进化算法[J].电力自动化设备,2012,32(5):57-61.Sun Guoqiang,Wei Zhinong,Tang Lifeng,et al. Pareto evolutionary algorithm for multi-objective fault location of distribution network[J].Electric Power Automation Equipment,2012,32(5):57-61.
[11]郭壮志,陈涛,洪俊杰,等.基于故障辅助因子的配电网高容错性故障区段定位方法[J].电力自动化设备,2017,37(7):93-100.Guo Zhuangzhi,Chen Tao,Hong Junjie,et al. High-tolerance faulty section locating based on fault accessory factors for distribution network based on genetic algorithm[J]. Electric Power Automation Equipment,2017,37(7):93-100.
[12]薛永端,徐丙垠,李天友,等.配网自动化系统小电流接地故障暂态定位技术[J].电力自动化设备,2013,23(12):27-32.Xue Yongduan,Xu Bingyin,Li Tianyou,et al. Small-current grounding fault location based on transient signals of distribution automation system[J]. Electric Power Automation Equipment,2013,23(12):27-32.
[13]张姝,杨健维,何正友,等.基于线路暂态重心频率的配电网故障区段定位[J].中国电机工程学报,2010,30(31):118-122.Zhang Shu,Yang Jianwei,He Zhengyou,et al. Fault section location of the distribution network based on transient center frequency[J]. Proceedings of CSEE,2015,35(10):2463-2470.
[14]肖辉,胡运发.基于分段时间弯曲距离的时间序列挖掘[J].计算机研究与发展,2005,42(1):72-78.Xiao Hui,Hu Yunfa. Data mining based on segmented time warping distance in time series database[J]. Journal of Computer Research and Development,2005,42(1):72-78.
[15]刘鹏辉,黄纯.基于动态时间弯曲距离的小电流接地故障区段定位方法[J].电网技术,2016,40(3):952-957.Liu Penghui,Huang Chun. A fault section location method of small-current grounding fault in distribution network based on dynamic time warping distance[J]. Power Systems Technology,2016,40(3):952-957.
[16]黄纯,刘鹏辉,江亚群,等.基于动态时间弯曲距离的主动配电网馈线差动保护[J].电工技术学报,2017,32(6):240-247.Huang Chun,Liu Penghui,Jiang Yaqun,et al. Feeder differential protection based on dynamic time warping distance in active distribution network[J]. Transactions of China Electrotechnical Society,2017,32(6):240-247.
[2]肖白,束洪春,高峰.小电流接地系统单相接地故障选线方法综述[J].继电器,2001,29(4):16-20.Xiao Bai,Shu Hongchun,Gao Feng. Survey of the methods of faults line section for single-phase-to-earth fault in networks with ungrounded neutral[J]. Power Systems Protection and Control,2001,29(4):16-20.
[3]齐郑,杨以涵.中性点非有效接地系统单相接地选线技术分析[J].电力系统自动化,2004,28(14):1-5.Qi Zheng,Yang Yihan. Analysis of technology of fault line selection for single-phase-to-earth faults in neutral point non-effectively grounded system[J]. Automation of Electric Power Systems,2004,28(14):1-5.
[4]马士聪,徐丙垠,高厚磊,等.检测暂态零模电流相关性的小电流接地故障定位方法[J].电力系统自动化,2008,32(7):48-52.Ma Shicong,Xu Bingyin,Xue Yongduan,et al. An earth fault locating method in feeder automation system by examining correlation of transient zero mode currents[J]. Automation of Electric Power Systems,2008,32(7):48-52.
[5]孙波,孙同景,薛永端,等.基于暂态信息的小电流接地故障区段定位[J].电力系统自动化,2008,32(3):52-55.Sun Bo,Sun Tongjing,Xue Yongduan,et al. Single phase to ground fault section locating based on transient signals in non-solidly earthed network[J]. Automation of Electric Power Systems,2008,32(3):52-55.
[6]倪广魁,鲍海,张利,等.基于零序电流突变量的配电网单相故障带定位判据[J].中国电机工程学报,2010,30(31):118-122.Ni Guangkui,Bao Hai,Zhang Li,et al. Criterion based on the fault component of zero sequence current for online fault location of singlephase fault in distribution network[J]. Proceedings of CSEE,2010,30(31):118-122.
[7]宋国兵,李广,于叶云,等.基于相电流突变量的配电网单相接地故障区段定位[J].电力系统自动化,2011,35(10):84-90.Song Guobing,Li Guang,Yu Yeyun,et al. Single-phase earth fault section based on phase current fault component in distribution network[J]. Automation of Electric Power Systems,2011,35(10):84-90.
[8]孙波,张承惠,孙同景,等.基于暂态相电流的小电流接地故障定位研究[J].电力系统保护与控制,2012,40(18):69-74.Sun Bo,Zhang Chenghui,Sun Tongjing,et al. Earth fault location based on transient phase current in non-solidly earthed network[J].Power Systems Protection and Control,2012,40(18):69-74.
[9]王铭,徐玉琴.基于小波变换和BP网络的非故障相暂态电流故障定位[J].电力自动化设备,2006,26(4):25-27.Wang Ming,Xu Yuqin. Locating faulting using transient component of non-fault-phase based on wavelet analysis and BP neural network[J].Electric Power Automation Equipment,2006,26(4):25-27.
[10]孙国强,卫志农,唐利锋,等.多目标配电网故障定位的Pareto进化算法[J].电力自动化设备,2012,32(5):57-61.Sun Guoqiang,Wei Zhinong,Tang Lifeng,et al. Pareto evolutionary algorithm for multi-objective fault location of distribution network[J].Electric Power Automation Equipment,2012,32(5):57-61.
[11]郭壮志,陈涛,洪俊杰,等.基于故障辅助因子的配电网高容错性故障区段定位方法[J].电力自动化设备,2017,37(7):93-100.Guo Zhuangzhi,Chen Tao,Hong Junjie,et al. High-tolerance faulty section locating based on fault accessory factors for distribution network based on genetic algorithm[J]. Electric Power Automation Equipment,2017,37(7):93-100.
[12]薛永端,徐丙垠,李天友,等.配网自动化系统小电流接地故障暂态定位技术[J].电力自动化设备,2013,23(12):27-32.Xue Yongduan,Xu Bingyin,Li Tianyou,et al. Small-current grounding fault location based on transient signals of distribution automation system[J]. Electric Power Automation Equipment,2013,23(12):27-32.
[13]张姝,杨健维,何正友,等.基于线路暂态重心频率的配电网故障区段定位[J].中国电机工程学报,2010,30(31):118-122.Zhang Shu,Yang Jianwei,He Zhengyou,et al. Fault section location of the distribution network based on transient center frequency[J]. Proceedings of CSEE,2015,35(10):2463-2470.
[14]肖辉,胡运发.基于分段时间弯曲距离的时间序列挖掘[J].计算机研究与发展,2005,42(1):72-78.Xiao Hui,Hu Yunfa. Data mining based on segmented time warping distance in time series database[J]. Journal of Computer Research and Development,2005,42(1):72-78.
[15]刘鹏辉,黄纯.基于动态时间弯曲距离的小电流接地故障区段定位方法[J].电网技术,2016,40(3):952-957.Liu Penghui,Huang Chun. A fault section location method of small-current grounding fault in distribution network based on dynamic time warping distance[J]. Power Systems Technology,2016,40(3):952-957.
[16]黄纯,刘鹏辉,江亚群,等.基于动态时间弯曲距离的主动配电网馈线差动保护[J].电工技术学报,2017,32(6):240-247.Huang Chun,Liu Penghui,Jiang Yaqun,et al. Feeder differential protection based on dynamic time warping distance in active distribution network[J]. Transactions of China Electrotechnical Society,2017,32(6):240-247.