多类分类SVM在电能质量扰动识别中的应用  

Application of multi-class classification SVM in power quality disturbances classification

作　　者：陈春玲[1] 许童羽[1] 郑伟[1] 姜凤利[1] 郭丹[1] 

机构地区：[1]沈阳农业大学信息与电气工程学院,辽宁沈阳110161

出　　处：《电力系统保护与控制》2010年第13期74-78,共5页Power System Protection and Control

基　　金：辽宁省教育厅高校科研A类项目(2008635)

摘　　要：结合傅里叶变换良好的幅频特性、小波变换良好的时频特性和支持向量机优秀的统计学习能力,采用多类分类支持向量机进行电能质量扰动的分类识别。对电压骤升、电压骤降、电压中断、谐波、电压波动、暂态振荡、瞬时脉冲、频率偏差等八种常见电能质量扰动进行数学建模,利用傅里叶变换和小波变换对产生的样本波形进行特征提取,将特征量输入到osu_svm进行电能质量扰动多类分类。算例表明该方案具有识别正确率高,训练样本数少,训练时间短,实时性好,对噪声不敏感等优点,是电能质量扰动识别的有效方法。This paper uses the multi-class classification for support vector machine and combines the good amplitude-frequency characteristic of Fourier transform, the good time-frequency characteristics of wavelet transform and the excellent statistical learning ability of support vector machine to make the classification and recognition to the disturbances of power quality. Mathematical modeling is done for the 8 kinds of common power quality disturbances, namely voltage swell, voltage sag, voltage interruption, harmonic, voltage fluctuation, transient oscillation, transient pulse and frequency deviation, and then Fourier transform and wavelet transform are used to extract the characteristics ofthe waveform of the generated samples, and the characteristic value is input to the osu_svm and the quality disturbances multi-class classification are done. The example shows that this method has a high recognition accuracy, a few training samples and a short training time, a good real-time performance, and is not sensitive to noise, etc. It is an effective method for power quality disturbances classification.

关 键 词：电能质量 扰动识别 支持向量机 多类分类 

分 类 号：TM711[电气工程—电力系统及自动化]