作　　者： ; ; ;

机构地区： 湖南大学电气与信息工程学院

出　　处： 《微电子学》 2005年第2期149-152,共4页

摘　　要： 　论述了基于多类电量测试信息模糊融合的模拟电路故障诊断方法的基本原理,提出了分别基于K故障节点诊断法和最小标准差法的元件故障隶属函数构造方法,以及基于可测点电压与不同测试频率下电路增益的模糊信息融合诊断算法。分别利用此两类测试信息及K故障诊断法和最小标准差法,对电路进行初步诊断,再运用模糊变换及故障定位规则,得到融合的故障诊断结果。模拟实验结果表明,所提方法大大提高了故障定位的准确率。 Fundamental principles of diagnosis methods based on fuzzy fusion of multiform circuit responses are discussed. By using fuzzy theory, branch fault attaching functions are proposed based on K-fault node diagnosis method and least dispersion method, respectively. A fuzzy data fusion algorithm based on accessible node voltages and circuit gains of output to input at different frequencies is developed. The preliminary diagnosis results are presented by fault attaching values of components. And fusion diagnosis is performed based on fuzzy transformation and fault location rules. Experimental results show that the proposed fusion diagnosis method is capable of diagnosing catastrophic and parametric faults of tolerance circuits with satisfactory accuracy.

关 键 词： 故障定位 模糊集理论 决策融合 模拟电路

领　　域： [电气工程]