基于GA与SVM的混合算法在电能质量扰动分类问题中的应用

Application of GA and SVM based hybrid algorithm for the classification of power-quality disturbances

本文针对电能质量扰动分类问题,提出了一种基于GA与SVM的混合算法。首先用小波变换技术对电信号进行特征提取,然后设计出一种遗传算法对提取出的特征进行筛选,最后把筛选后的特征提交给支撑向量机,并由支撑向量机进行分类。数值模拟实验验证了该方法的有效性。

For the classification of power quality disturbances, we present a GA and SVM based hybrid algorithm. Firstly, a wavelet transform technique is used for feature extraction. Then a genetic algorithm is proposed for feature selection. The support vector machine technique is used for classification of selected features. Numerical experiments validate the effectiveness of the proposed algorithm.