基于多标签Rank-WSVM的复合电能质量扰动分类.pdfVIP

第33 卷 第28 期 中 国 电 机 工 程 学 报 Vol.33 No.28 Oct.5, 2013 114 2013 年10 月5 日 Proceedings of the CSEE ©2013 Chin.Soc.for Elec.Eng. 文章编号：0258-8013 (2013) 28-0114-07 中图分类号：TM 71 文献标志码：A 学科分类号：470·40 基于多标签 Rank-WSVM 的 复合电能质量扰动分类 张巧革，刘志刚，朱玲，张杨 (西南交通大学电气工程学院，四川省 成都市 610031) Recognition of Multiple Power Quality Disturbances Using Multi-label Wavelet Support Vector Machine ZHANG Qiaoge, LIU Zhigang, ZHU Ling, ZHANG Yang (School of Electrical Engineering, Southwest Jiaotong University, Chengdu 610031, Sichuan Province, China) ABSTRACT: Multi-label rank wavelet support vector machine 组合而成的复合扰动。 (Rank-WSVM) is proposed and applied to the recognition of 关键词：电能质量；复合扰动；多标签分类；排位小波支持 multiple power quality disturbances in this paper. By applying 向量机 the wavelet technique to Rank-SVM classifier the efficiency is improved. Firstly, all of the common power quality 0 引言 disturbances and their compound ones are decomposed by 电能质量扰动分类问题是 1 个十分复杂的问 discrete wavelet transform, and the Tsallis entropy of the wavelet coefficients are extracted as eigenvectors. And then, 题[1-2] 。目前，对单一扰动识别的研究较多，研究思 the type of multiple power quality disturbances is predicted 路和实验方法较为成熟，主要是先利用信号处理技 through the Rank-WSVM. The simulation results show that 术进行特征提取再设计合适的分类器进行分类[3-4] 。 Rank-WSVM proposed in this paper can effectively recognize 但由于实际系统的复杂性，采集到的电能质量扰动 the multiple power quality disturbances including voltage sag, 信号往往不是理想