基于多分辨奇异值分解包和随机森林的电能质量扰动分类研究

POWER QUALITY DISTURBANCE CLASSIFICATION BASED ON MULTI-RESOLUTIONSINGULAR VALUE DECOMPOSITION PACKAGE AND RANDOM FOREST

为解决电能质量扰动的分类问题,利用多分辨奇异值分解(Singular Value Decomposition, SVD)的信号逐层分解方式,提出基于多分辨SVD包与随机森林(Multi-Resolution SVD and Random Forest, MRSVD-RF)的电能质量扰动分类方法。通过实验证明了该算法对单一和复合电能质量信号的分类效果明显优于分解结构相似的基于的小波包的信号分解方式,比较了分类器模型的选择和特征提取数量对算法性能的影响。

In order to solve the problem of power quality disturbance classification,this paper proposes a power quality disturbance classification method based on multi-resolution singular value decomposition and random forest(MRSVD-RF)by using layer by layer decomposition of signals with multi-resolution SVD.The experiments prove that the classification performance of the proposed algorithm for single and composite power quality signals is significantly better than the signal decomposition method based on wavelet packets with similar decomposition structures.The effects of classifier model selection and feature extraction quantity on the algorithm performance were compared.