基于小波变换的神经网络模拟电路故障诊断

The wavelet neural network method of the analogy circuit fault diagnoses

对模拟电路提出了一种基于小波变换与神经网络相结合的故障诊断方法,该方法用小波变换对模拟电路故障信号提取小波特征,并经小波变换压缩,再将故障特征量输入至神经网络处理。结果表明,该方法有效地减少神经网络输入层单元数,简化了神经网络结构,提高了故障诊断能力。

The methods based on the neural network technology combined with the wavelet for fault diagnose were developed, which used wavelet transforms as analogy circuit fault signal preprocessor to make the number of input neural network decreased effectively. The structure of neural network was predigested and the ability of fault diagnoses was prompted.