基于VMD-PCA-SVM的电能质量扰动辨识  

Power Quality Disturbance Identification Based on VMD-PCA-SVM

作　　者：胡汉铖 刘明萍[1] 张震[1] 张镇涛 汪庆年[1] HU Hancheng;LIU Mingping;ZHANG Zhen;ZHANG Zhentao;WANG Qingnian(Information Engineering School,Nanchang University,Nanchang 330031,China)

机构地区：[1]南昌大学信息工程学院,南昌330031

出　　处：《实验室研究与探索》2021年第10期97-102,共6页Research and Exploration In Laboratory

基　　金：国家自然科学基金项目(61665006,61865011)。

摘　　要：为提高电能质量扰动辨识的准确性,提出一种基于VMD-PCA-SVM的电能质量扰动辨识方法。用Matlab 2017a仿真得到电能质量扰动信号;用变分模态分解(VMD)分解得到本征模态函数(IMF)分量,计算其希尔伯特边际谱的能量值,构造特征向量;将特征向量用主成分分析法(PCA)降维;将降维特征向量输入支持向量机(SVM)中训练,实现对电能质量扰动信号的辨识。与现有文献对比,试验结果表明,该方法准确率高,鲁棒性强,在不同信噪比下能有效识别包括两种复合扰动在内的8种电能质量扰动信号,准确率高达99.94%。In order to improve the accuracy of power quality disturbance identification, a power quality disturbance identification method based on VMD-PCA-SVM is proposed in this paper. Firstly, the power quality disturbance signals are simulated by Matlab 2017 a;Secondly, the intrinsic mode function(IMF) component is obtained by variational mode decomposition(VMD), and the energy value of its Hilbert marginal spectrum is calculated, from which the eigenvector is constructed;Thirdly, the eigenvectors are reduced by principal component analysis(PCA);Finally, the reduced eigenvectors are input into the support vector machine(SVM) to identify the power quality disturbance signals. Compared with the existing literatures, the experimental results show that this method has high accuracy and strong robustness, and can effectively identify 8 kinds of power quality disturbance signals including two kinds of composite disturbances under different SNRs, with an accuracy of 99.94%.

关 键 词：变分模态分解 主成分分析 支持向量机 电能质量 扰动辨识 

分 类 号：TM933[电气工程—电力电子与电力传动]