摘要
由于模拟电路元件中存在容差性和边界模糊性等问题,提出一种基于云关联系数矩阵的模拟电路故障诊断方法。首先,利用模拟电路本身的特点,提出了云模型提取电路故障特征集的算法。利用综合云隶属度方法计算云关联系数矩阵,并将其作为被测模拟电路故障特征集。然后,该特征集被用作支持向量机(SVM)模型的输入以确定故障的判定类别。以Sallen＿Key带通滤波器电路和四运放双二次高通滤波器电路的单、双故障情景为例作为验证对象。结果表明,该方法提取的故障特征能够更好地反映电路信息,诊断精度均可达到95%以上,具有较高的诊断精确度,可用于模拟电路软故障诊断。
Due to the tolerance and ambiguity in the boundary of analog circuit components, an analog circuit fault diagnosis method is proposed and it is based on cloud correlation coefficient matrix. First, by using characteristic of analog circuit, the algorithm of extracting fault feature based on cloud model is proposed. Adopting the comprehensive cloud membership method that has a characteristics of fault feature set is fully extracted, and computing the coefficient matrix of cloud correlation coefficient as the feature set of the analog circuits under test. Then the feature is used as the input of the support vector machine(SVM) model to diagnose the fault category. Single and double fault circuit case of the Sallen_Key band-pass filter circuit and four op-amp double high-pass filter circuit is taken as examples to carry out experimental verification. The results show that the fault feature extracted by this method could better reflect the circuit information and the accuracy of diagnosis, it has a higher diagnostic accuracy with more than 95%, it can be used for analog circuit soft fault diagnosis.
出处
《电子测量与仪器学报》
CSCD
北大核心
2018年第10期65-72,共8页
Journal of Electronic Measurement and Instrumentation
基金
国家自然科学基金(61741403)资助项目
关键词
故障诊断
特征提取
云模型
云关联系数矩阵
支持向量机
fault diagnosis
feature extraction
cloud model
cloud correlation coefficient matrix
support vector machine
