基于改进小波能熵和支持向量机的短时电能质量扰动识别-电工技术学报.pdfVIP

2009 年 4 月 电 工 技 术 学 报 Vol.24 No. 4 第 24 卷第 4 期 TRAN SACTION S OF CHINA ELECTROTECHNICAL SOCIETY Apr . 2009 基于改进小波能熵和支持向量机的短时电能质 量扰动识别 李庚银 王洪磊 周明 （华北电力大学电力系统保护与动态安全监控教育部重点实验室 北京 102206） 摘要 提出了一 基于改进小波能熵和支持向量机 （SVM）的短时电能质量扰动识别方法。 首先对采样信号进行小波多分辨分解与重构处理，然后引入滑动时间窗算法，从时-频域结合分 析的角度，选用高频带的小波系数进行特征提取 ；提出了改进小波能熵算法，并用此计算相应 的熵值作为扰动特征量，将这些特征量作为 SVM 的 入，实现短时电能质量扰动的辨识。通 过原始小波能熵与改进小波能熵的对比，仿真结果表明了改进算法的有效性。 关键词：电能质量 扰动识别 改进小波能熵 支持向量机 中 分类号：TM7 14 .3 Short-Time Power Quality Disturbances Identification Based on Improved Wavelet Energy Entropy and SVM L i Gengy in Wang H ong lei Zhou M ing （Key Laboratory of Power System Protection and Dynamic Security Monitoring and Control under Ministry of Education North China Electric Power University Beij ing 102206 China ） Abstract This paper proposes an approach to identify short-time power quality disturbances based on improved wavelet energy entropy and support vector machine (SVM). Firstly, the sampled signals are processed by using the multi-scale wavelet resolution and reconstruction . Then, the sliding time window is introduced into algorithm, combined the time domain analysis with frequency domain analysis. The wavelet coefficients of the high frequency regions are selected for feature extraction . The values of the entropy are then calculated according to the improved wavelet energy entropy proposed in this paper as the disturbances features. Furthermore, these features are used as the input vectors of SVM to classify the short-time power quality disturbances. The simulation results show that the proposed method has mer