基于FCM-LLE的模拟电路故障诊断  

A FCM-LLE-based Fault Diagnosis Algorithm for Analog Circuits

作　　者：王月海[1] 吕晓雯 程冉[1] 张齐昌 WANG Yuehai;LYU Xiaowen;CHENG Ran;ZHANG Qichang(Col.of Information,North China Univ.of Tech.,100144,Beijing,China)

机构地区：[1]北方工业大学信息学院,北京100144

出　　处：《北方工业大学学报》2020年第2期85-89,共5页Journal of North China University of Technology

基　　金：国家自然科学基金面上项目“基于概率光流计算的仿生偏振/光流辅助INS组合导航方法研究”(61573019).

摘　　要：针对模拟电路的故障数据具有多源异构、维度高、数据量大且传递关系复杂等特点,本文提出了一种基于模糊C均值聚类-局部线性嵌入(FCM-LLE)的模拟电路故障诊断算法,使用局部线性嵌入法对特征数据进行降维,利用模糊C均值聚类方法进行故障分类诊断,并根据诊断情况进行多轮诊断以提升诊断率.仿真结果表明该方法对较复杂的模拟电路进行故障诊断,且具有较高的诊断率.Fault data of analog circuit are characterized by multi-source heterogeneity, high dimension, large volume of data and complex data correlation. An algorithm based on fuzzy C-means clustering combined with locally linear embedding(FCM-LLE) is proposed to reduce the dimension of diagnostic data by LLE and classify diagnosis category by FCM clustering. Meanwhile, multiple rounds of diagnosis are also performed to improve the diagnostic rate based on the outcome. The simulation results verified the effectiveness of the proposed algorithm as it provides a higher diagnostic accuracy.

关 键 词：故障诊断 模糊C均值聚类 聚类中心 隶属度 局部线性嵌入 

分 类 号：TP206[自动化与计算机技术—检测技术与自动化装置]