优化核参数的SVM在电能质量扰动分类中的应用

Recognition of Power Quality Disturbances Based on Support Vector Machine with Optimal Kernel-parameter

针对短时电能质量变化和暂态扰动现象的不同特点,建立常见电能质量扰动的数学模型.运用小波变换对暂态电能质量扰动现象的内在特征进行提取,将扰动电压变化率绝对值、扰动能量变化量作为暂态电能质量扰动的特征向量.根据支持向量机的基本原理,给出一种推广误差上界估计判据,利用此判据进行最优核参数的自动选取,利用支持向量机进行训练和测试.结果表明,优化核参数的支持向量机分类器准确率高,实时性好.

According to the different features of short-term power quality variation and transient disturbance,the mathematical models of frequent power quality disturbances（PSD） were established.In this method the time characteristic of the disturbance is extracted by wavelet transform;the duration,amplitude and frequency of the disturbance and the absolute value of voltage regulation are taken as the inputs of classifiers to test classification accuracies.The fundamental of support vector machine（SVM） based on structure risk minimization was introduced.An estimation formula of upper bound of generalization error was given,and the optimal kernel-parameter of the SVM was selected automatically by the formula.The classification accuracies indicates that the optimal kernel-parameter method can get optimal results.