基于改进的SVM的电能质量复合扰动分类

本文利用支持向量机对电能质量复合扰动进行分类,解决其多重分类问题,为了提高其整体分类的准确率,对支持向量机中的核函数进行了改进。考虑到特征向量在核函数中心位置的聚集程度会影响支持向量的数目,本文在核函数中引进一个径向宽度因子和一个幅值调节因子,从而解决传统核函数存在的问题,减少支持向量数目,降低计算复杂度。将改进后的算法应用到电能质量复合扰动分类中,验证所提方法对于电能质量复合扰动分类不仅具有可行性,并且有较高的分类准确率。从仿真实验结果可以看出,改进的方法对常见的7种单一电能质量扰动信号和5种电能质量复合扰动信号能够进行分类,相对原算法提高了分类准确率。

Classification of multiple power quality disturbances based on improved SVM

In order to solve the problem of mixed disturbances classification, support vector machine is used to classify power quality disturbances in this paper, and an improved support vector machine algorithm is proposed to improve the accuracy of the whole classification. Considering that the aggregation degree of the eigenvector in the central position of the kernel affects the number of support vectors, a radial width factor and an amplitude adjust?ment factor are introduced to the kernel function to solve the problem of traditional kernel functions. The proposed method can also decrease the number of support vectors and reduces the computational complexity. The improved algorithm is applied to the classification of the multiple power quality disturbances to prove that the proposed method is feasible and has high classification accuracy. From the results of the simulation experiment, we can see that the proposed method can classify seven kinds of single power quality disturbance and five kinds of multiple power quali?ty disturbances. Compared with the original algorithm, the classification accuracy is improved.