基于支持向量机的单端行波故障测距方法
Single Terminal Traveling Wave Fault Location Method Based on SVM
DOI: 10.12677/SG.2016.64027, PDF, HTML, XML,  被引量 下载: 1,537  浏览: 2,386 
作者: 沈兴来, 曹 琦:江苏省电力公司徐州供电公司,江苏 徐州;崔亚博:国网浙江省电力公司检修分公司,浙江 杭州;梁 睿:中国矿业大学信息与电气工程学院,江苏 徐州
关键词: 支持向量机单端行波法故障测距小波变换SVM Single Terminal Traveling Wave Method Fault Location Wavelet Transform
摘要: 单端辐射状配电网发生单相接地故障时,单端行波法故障测距的难点在于第二个到达测量点的故障行波波头的正确辨识。针对上述难点,本文提出了一种基于支持向量机的单端行波故障测距方法。首先,利用支持向量机正确地识别出故障段,然后根据行波传播的网格图确定第二个到达测量点的故障行波波头是来自于故障点还是来自于对端母线。通过相模变换消除电磁耦合的影响,通过对小波变换相关知识的研究,选择最优小波基进行小波变换,对故障行波波头进行识别,检测故障行波波头模极大值对应的时刻,进而利用相应波头到达测量点的时间差结合波速实现故障点的精确测距。
Abstract: When single-phase-to-earth fault occurred in radial distribution network supplied by single power, the key problem of the single-ended traveling wave fault location lies in the correct recognition of the second travelling wave head to reach the measuring point. In view of the above problem, this paper introduces single terminal traveling wave fault location method based on SVM. After identifying the fault section by the SVM, the second traveling wave head coming from the fault point or bus at opposite terminal is recognized based on lattice diagram of traveling waves. Elec-tromagnetic coupling effect is eliminated by phase mode transformation. Through the study of relevant knowledge of wavelet transform, the optimal wavelet base is selected for wavelet trans-form in order to identify the fault traveling wave head. The time corresponding to fault traveling wave modulus maximum is detected and time difference of arrival and velocity of the corre-sponding wave head is used to achieve fault location.
文章引用:沈兴来, 崔亚博, 曹琦, 梁睿. 基于支持向量机的单端行波故障测距方法[J]. 智能电网, 2016, 6(4): 242-251. http://dx.doi.org/10.12677/SG.2016.64027

参考文献

[1] 束洪春, 邬乾晋, 张广斌, 等. 基于神经网络的单端行波故障测距方法[J]. 中国电机工程学报, 2011, 31(4): 85-92.
[2] 徐丙垠, 李京, 陈平, 等. 现代行波测距技术及其应用[J]. 电力系统自动化[J], 2001, 25(23): 62-65.
[3] 刘健, 董新洲. 配电网故障定位与供电恢复[M]. 北京: 中国电力出版社, 2012.
[4] 季涛. 中性点非有效接地系统行波故障测距技术[M]. 北京: 北京理工大学出版社, 2008.
[5] 施慎行, 董新洲, 周双喜. 单相接地故障下第2个反向行波识别的新方法[J]. 电力系统自动化, 2006, 30(1): 41-45.
[6] Burges, C.J.C. (1998) A Tutorial on Support Vector Machines for Pattern Recognition. Kluwer Academic Publishers, Boston. Manufactured in The Netherlands.
[7] Deng, X.T., Yuan, R.X. and Xiao. Z.F. (2015) Fault Location in Loop Distribution Network Using SVM Technology. Electrical Power and Energy Systems, 65, 254-261.
[8] 卢占庆. 基于支持向量和小波变换的配电网单相接地故障分析及系统设计[D]. 沈阳: 东北大学, 2011.
[9] 郑蕊蕊, 赵继印, 赵婷婷, 等. 基于遗传支持向量机和灰色人工免疫算法的电力变压器故障诊断[J]. 中国电机工程学报, 2011, 31(7): 56-63.
[10] Parikh, U.B., Das, B. and Maheshwari, R.P. (2008) Combinedwavelet-SVM Technique for Fault Zone Detection in a Series Compensated Transmission Line. IEEE Transactions on Power Delivery, 23, 1789-1794.
[11] Livani, H. and Evrenosoglu, C.Y. (2012) A Hybrid Fault Location Method for Overhead Lines Combined with Underground Cables Using DWT and SVM. 2012 IEEE Power and Energy Society General Meeting, San Diego, 22-26 July 2012, 1-6.
http://dx.doi.org/10.1109/pesgm.2012.6345578
[12] 马丹丹, 王晓茹. 基于小波模极大值的单端行波故障测距[J]. 电力系统保护与控制, 2009, 37(3): 55-59.
[13] 和敬涵, 张飚, 范瑜, 等. 解耦变换在电力系统暂态保护中的应用研究[J]. 北京交通大学学报, 2006, 30(5): 101-104.
[14] 束洪春, 彭仕欣, 李斌, 赵兴兵. 配电网缆–线混合线路故障选线的可拓融合方法[J]. 电工技术学报, 2008, (11): 146-152.
[15] 梁睿, 孙式想. 单端行波故障测距的组合方法研究[J]. 电网技术, 2013, 37(3): 699-706.