基于多类电量测试数据融合的模拟电路故障诊断

提出了一种利用多类电量测试数据、基于神经网络与模糊集理论的模拟电路故障诊断新方法.检测可及节点电压并测量电路在选定的各测试频率下输出对输入的增益,分别利用此两类测试数据,各用一个独立的BP网络对电路进行初步诊断,再运用所提出的模糊融合诊断算法及其故障定位规则进行故障定位.模拟实验结果表明:所提方法克服了基于单一信息诊断方法的不足,对硬故障与元件参数偏移较小的软故障均适用,故障定位准确率高.

Fault Diagnosis of Analog Circuits Based on Data Fusion for Multiform Circuit Responses

Based on BP network and fuzzy theory, a new fault diagnosis method for analog circuits is proposed by using data fusion for multiform circuit responses. By measuring accessible node voltages and circuit gains of output to input under different test frequencies, Preliminary diagnosis is performed separately by an independent improved BP network employing one kind of test data. Fault location is accomplished by using the proposed fussy fusion algorithm and fault location regulation according to the preliminary diagnosis results. Theoretical analysis and experimental results show that the proposed fusion diagnosis method avoids limitation of diagnosis based on single test information and has the capability to diagnose catastrophic and parametric faults of analog circuits with satisfactory accuracy.