电力系统过电压支持向量机分层识别..doc

电力系统暂时过电压多级支持向量机分层识别 杜 林1，李 欣1, 2，王丽蓉3，司马文霞1 （1.重庆大学输配电装备及系统安全与新技术国家重点实验室，重庆 400044； 2.湖南省电力公司科学研究院，湖南 长沙 410007；3.湖南省电力公司超高压管理局，湖南 长沙 410004） 摘要：提出了一种电力系统暂时过电压多级支持向量机（M-SVM）分层识别的方法。根据暂时过电压分类，建立暂时过电压分层识别系统，并采用“二分树”法构建多级支持向量机分类器。在变电站实测过电压数据的基础上，提取了三相及零序电压的时域统计特征和小波时频特征，同时对特征量进行逐级选择，将这些特征量作为M-SVM的输入，实现暂时过电压类型辨识。现场数据测试表明，采用的M-SVM分层识别方法具有训练样本少、训练时间短、识别率高的优点，可较好地应用于电力系统暂时过电压类型识别。 关键词：暂时过电压；特征提取；过电压识别；零序电压；分层识别系统；多级支持向量机 Temporary overvoltage layered pattern identification based on multistage support vector machine in power system DU Lin1, LI Xin1, 2, WANG Li-rong3, SIMA Wen-xia1 (1. State Key Laboratory of Power Transmission Equipment System Security and New Technology, Chongqing University, Chongqing 400044, China; 2. Hunan Electric Power Corporation Research Institute, Changsha 410007, China; 3. EHV Administration of Hunan Electric Power Corporation, Changsha 410004, China) Abstract: This paper proposes an approach of temporary overvoltage layered recognition based on multistage support vector machine in power system. Firstly, on the basis of temporary overvoltage classification, a temporary overvoltage multi-level recognition system is built, and the classifiers are constructed based on binary trees of multistage support vector machine. Then, according to field overvoltage data in a substation, time-domain statistical features and wavelet time-frequency features of three-phase and zero sequence voltage are extracted and selected step by step, and finally these features are used as input vectors of M-SVM to classify temporary overvoltage. The field acquired overvoltage data indicates that the method in this paper, which has advantages of less training samples, shorter training time and high recognition rate, can be well applied to power system temporary overvoltage identification. This work is supported by the National Basic Research Program of China (973) (No. 2009CB724504). Key words: temporary overvoltage；feature extraction；overvoltage identification；zero