基于WT和GA-SVM的电能质量扰动识别方法  

Power Quality Disturbance Recognition Method Based on WT and GA-SVM

作　　者：李波 曹敏[1,2] 李仕林 李春阳[3] LI Bo;CAO Min;LI Shi-lin;LI Chun-yang(Electric Power Research Institute of Yunnan Power Grid Co.,Ltd.,Kunming 650217,China;不详)

机构地区：[1]云南电网有限责任公司电力科学研究院,云南昆明650217 [2]中国南方电网公司电能计量重点实验室,云南昆明650217 [3]武汉大学,电子信息学院,湖北武汉430072

出　　处：《电力电子技术》2020年第3期52-55,共4页Power Electronics

基　　金：南方电网公司重点项目(YNKJQQ00000283);云南电网公司重点项目(YN2014-2-001)。

摘　　要：电能质量的监测是用电信息采集系统的主要任务,监测电能质量的关键在于对电能质量扰动的识别。针对电能质量扰动的识别和分类问题,提出了一种基于小波变换(WT)和遗传支持向量机(GA-SVM)的识别方法。首先对正常电压和6种常见的电能质量扰动信号进行小波分解,提取各层小波系数的能量熵作为特征向量,然后利用基于GA优化的SVM对扰动信号进行学习训练,得到电能质量扰动识别模型。算例实验中与单一的SVM和反向传播(BP)神经网络智能算法分别对比,结果表明该方法在各扰动信号的识别准确率和训练时间方面都有明显改善。The main task of electricity information collection system is power quality monitoring.Aiming at problems of power quality disturbance identification and classification,a new method classifying power quality disturbances based on wavelet transform(WT) and genetic algorithm-support vector machine(GA-SVM) is presented.Normal voltage and six kinds of power quality disturbances are decomposed by wavelet and the energy entropy of the wavelet packet coefficients of each end node are extracted as eigenvectors.GA is used to optimize parameters of SVM.The disturbances are classified by using optimization parameters of SVM.Compared with single SVM and back propagation(BP) intelligence algorithm,the results show that the method is improved at the aspect of identification accuracy of the distribution signal and the training time.

关 键 词：电能质量 小波变换 支持向量机 

分 类 号：R363.124[医药卫生—病理学]