基于MRSVD-RF的同杆双回线路故障识别  

Fault Identification of Double-circuit Line on the Same Tower Based on MRSVD-RF

作　　者：杨亮 吴浩 胡潇涛 杨杰 陈佳豪 刘益岑 YANG Liang;WU Hao;HU Xiaotao;YANG Jie;CHEN Jiahao;LIU Yicen(School of Automation and Information Engineering,Sichuan University of Science&Engineering,Zigong 643000,China;Artificial Intelligence Key Laboratory of Sichuan Province,Zigong 643000,China;Electric Power Research Institute,State Grid Sichuan Electric Power Company,Chengdu 610000,China)

机构地区：[1]四川轻化工大学自动化与信息工程学院,自贡643000 [2]人工智能四川省重点实验室,自贡643000 [3]国网四川电力公司电力科学研究院,成都610000

出　　处：《电力系统及其自动化学报》2022年第1期65-75,共11页Proceedings of the CSU-EPSA

基　　金：人工智能四川省重点实验室项目(2019RYY01);四川省科技厅项目(2018GZDZX0043,2019YJ0477);国家电网有限公司科技项目(合同号:521997180016)。

摘　　要：为提高行波保护在各种工况下的可靠性,分析区内、外故障时线路两端的初始行波电流变化规律,提出一种基于多分辨奇异值分解和随机森林的同杆双回线路区内外故障识别方法。该方法通过提取故障后线路两端的行波电流数据进行相模变换,选取变换后的同向模量电流进行6层多分辨奇异值分解,计算每层的电流积分作为特征向量输入随机森林分类器模型中进行训练和测试,来识别区内、外故障。仿真结果表明,该方法在各种工况下都能够准确地进行故障识别,并在抗噪声干扰和CT饱和方面也具有良好的表现。To improve the reliability of traveling wave protection under various working conditions,after the analysis of changes in initial traveling wave current at both terminals of the line under internal and external faults,a fault identification method for the double-circuit line on the same tower under internal and external faults is proposed,which is based on multi-resolution singular value decomposition and random forest(MRSVD-RF).This method performs phase-to-mode conversion by extracting the traveling wave current data at both terminals of the line after the fault,selects the converted current with modulus in the same direction to perform 6-layer MRSVD,and calculates the current integral of each layer as a feature vector,which enters into the RF classification and is used for training and testing.In this way,the internal and external faults can be identified.Experimental results show that the proposed method can accurately identify faults under various working conditions,and it also has satisfying performance in anti-noise interference and CT saturation.

关 键 词：同杆双回线路 电流积分 多分辨奇异值分解 随机森林 故障识别 

分 类 号：TM75[电气工程—电力系统及自动化]