论文部分内容阅读
针对现有BP网络在模拟电路故障诊断中存在的问题,提出了一种基于BP小波神经网络的故障诊断方法。该法将小波函数与BP网络结合构成BP小波网络,这种网络具有小波变换的时频局域化性质和BP网络的自学习能力。分别用BP小波网络和传统BP网络对实例电路进行故障诊断,结果表明本方法是有效的,而且比传统BP网络方法的学习收敛速度快得多。
Aiming at the existing problems in analog circuit fault diagnosis of existing BP network, a fault diagnosis method based on BP wavelet neural network is proposed. This method combines the wavelet function and the BP network to form the BP wavelet network, which has the time-frequency localization features of wavelet transform and the self-learning ability of the BP network. The BP neural network and the traditional BP network are respectively used to diagnose the fault of the example circuit. The results show that this method is effective and much faster than the traditional BP neural network in learning convergence.