基于聚类改进S变换与直接支持向量机的电能质量扰动识别

Power quality disturbance identification based on clustering-modified S-transform and direct support vector machine

针对电能质量扰动信号的识别问题,提出基于聚类改进S变换与直接支持向量机(SVM)的电能质量扰动识别方法。提出聚类改进S变换方法,该方法结合电能质量扰动信号的特点,可同时对基频的时域分辨率及高频的频域分辨率进行最优化处理,保证特征提取的准确性;将直接支持向量机作为分类器,与最小二乘支持向量机相比,其求解简单,计算复杂度较低,训练与测试速度快,泛化能力较高,并且避免不能保证全局最优解的缺点;将聚类改进S变换与直接支持向量机相结合,应用于单一扰动及混合扰动的识别分类工作。仿真实验验证了所提方法的有效性。

A method based on CMST（Clustering-Modified S-Transform） and DSVM（Direct Support Vector Machine） is proposed to identify the power quality disturbance. Combined with the characteristics of power quality disturbance signal,the CMST method can optimally and simultaneously process the time-domain resolution of fundamental frequency signal and frequency-domain resolution of high-frequency signal to ensure the correctness of property extraction. Compared with the least squares support vector machine,DSVM,as a classifier,has simpler solving process,lower computation complexity,faster training and testing speed,higher generalization ability. Furthermore,it guarantees the global optimal solution. The CMST combined with DSVM is applied in the identification of single or mixed disturbance. Simulative experiment verifies the effectiveness of the proposed method.