基于优选小波基和模糊SOM网络的模拟电路故障诊断  

Analog Circuit Fault Diagnosis Method Based on Preferred Wavelet Basis and Fuzzy SOM

作　　者：徐志成[1] 王宏[2] 徐长英[2] 邓芳明[1,3] 何怡刚[3] 

机构地区：[1]华东交通大学现代教育技术中心,江西南昌330013 [2]南昌航空大学工程训练中心,江西南昌330063 [3]合肥工业大学电气与自动化工程学院,安徽合肥230009

出　　处：《测控技术》2016年第11期5-8,13,共5页Measurement & Control Technology

基　　金：国家自然科学基金(51577046);国防科技计划项目(C1120110004;9140A27020211DZ5102);教育部科学技术研究重大项目(313018);江西省自然科学基金(20142BAB217008)

摘　　要：为了解决模拟电路故障诊断中的特征提取困难并实现对模拟电路故障模式准确的分类,提出一种优选小波基、模糊理论和自组织特征映射网络(SOM,self-organizing feature map)相结合的模拟电路故障诊断方法。该方法首先对模拟电路故障响应信号进行小波分解、提取能量值、均值和方差组成输入特征向量,同时采用余弦分离度评价小波变换在不同小波基函数下获取故障特征的有效性,据此选择余弦分离度最小的小波基分解的特征向量输入到自组织特征映射网络进行故障分类。仿真实验表明,利用余弦分离度选择的最优小波基能有效提高模拟电路故障特征提取,模糊神经网络能对故障模式进行精确分类。In order to solve the difficulty of feature extraction in analog circuits fault diagnosis and to accurately classify fault pattern,a new analog circuit diagnosis method based on the combination of optimal wavelet basis,fuzzy theory and self-organizing feature map is proposed.The response signals of the analog circuit are preprocessed by wavelet transform to extract feature consisting of energy,mean and variance,and the optimal wavelet coefficients with better classification capacity are obtained by using the minimum cosine separating degree.Then the feature set that is composed of optimal wavelet energy,mean and variance is inputted into a self-organizing feature map（SOM） to identify different fault cases.Simulation and experiment results show that the optimal wavelet transform based on cosine separating degree can extract analog circuit feature effectively,and the fuzzy SOM neural network can classify fault cases accurately.

关 键 词：模拟电路 故障诊断 小波基 余弦分离度 自组织特征映射网络 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] TM930[自动化与计算机技术—控制科学与工程]