中压电力设备局部放电信号模式识别研究

Research on Partial Discharge Patter n Recog nition of Power Equipment in MV

本文利用多分辨率分析法和BP神经网络相结合的方法对局部放电信号进行模式识别,通过相关的理论分析,采用多分辨率分析法提取各样本信号特征量,并通过BP神经网络作为识别器对各特征量进行模式识别,最终依托MATLAB进行识别验证,识别结果证明了该方法的正确性和可行性。

In this paper,the method of combi ning multi-resoluti on an alysis and BP n eural network for patter n recog nition of partial discharge signals is studied.Based on related theoretical analysis,the multi-resolution analysis method is used to extract the signal feature quantities of each sample,and the BP neural network is used as a classifier.The pattern recog nition was performed on each feature,and the recog nition was finally verified by MATLAB.The recog nition results proved the correctness and feasibility of the method.