模拟数字电路故障诊断新方法

New Method for Fault Diagnosis of Analog and Digital Circuits

利用小波变换与神经网络相结合的方法,采用能量分布特征提取方法和改进BP算法,给出了一种基于小波变换和BP神经网络相结合的模拟电路故障诊断方法。用正弦信号仿真模拟电路,应用小波变换对模拟电路的采样信号进行多尺度分解,再进行能量分布特征提取,然后利用神经网络对各种状态下的特征向量进行分类识别,实现模拟电路故障诊断。在用神经网络诊断模拟电路的基础上,进行了将神经网络用于数字电路单故障诊断的研究。对两者的实例电路仿真结果表明,神经网络可以有效、方便地实现电路的故障检测和定位,准确率高,为故障诊断的研究提供了一种新思路。

A method of fault diagnosis for analogue circuit based on the combination of BP neural network （BPNN） with wavelet transformation was presented, using the method of drawing energy feature and improved BP algorithm. A high-pass filter was stimulated with sinusoid input, and its output was sampled in time domain to collect training data for neural network. The collected data was prcprocessed by WT to generate fault features. Feature vectors under certain states could be classified using neural network. Based on diagnosing anlog circuit by neural network, researched the method diagnosing the single fault for digital circuit by neural network. The results shown that the neural network can detect and locate circuit fault effectively and expediently, which provides a new method for fault diagnose.