基于概率与小波分析的故障检测方法及其应用  

A Fault Detection Method Based on Probability and Wavelet Analysis and Its Application

作　　者：赵佳佳[1] 马喜平[2] 贾嵘[1] 沈渭程[2] 党建[1] 

机构地区：[1]西安理工大学水利水电学院,陕西西安710048 [2]国网甘肃省电力公司电力科学研究院,甘肃兰州730050

出　　处：《电网与清洁能源》2016年第7期92-95,共4页Power System and Clean Energy

基　　金：国家自然科学基金项目(51279161)~~

摘　　要：针对小波阈值法无法准确从强噪声中检测到故障信号的问题,提出了一种基于概率与小波分析的故障检测方法。首先基于分位数的定义将各分解层的小波系数进行分割,利用各层的若干个分界值计算信号各个位置发生故障的概率,再结合小波阈值分析,得出最终的检测结果。通过仿真和实测分析,表明该方法可以准确、有效地检测到故障信号,具有良好的应用价值。As the wavelet threshold method can not detect the fault signal from strong noises accurately, a fault detection method based on probability and wavelet analysis is proposed in this paper. Firstly the wavelet coefficients of each decomposition layer are divided on the basis of quantile. And the probability which indicates the possibility of fault of each position in the signal is calculated based on boundary values and the final detection result is obtained based on the result of wavelet threshold analysis. Simulation and experimental results show that this method is able to detect the fault signal accurately and effectively, and it has good application value.

关 键 词：故障检测 概率 分位数 小波分析 

分 类 号：TP27[自动化与计算机技术—检测技术与自动化装置]