摘要
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。
A new neural network which is BP wavelet neural network is proposed to used for the fault diagnosis of analog circuit instead of BP network, in order to improve the performance of fault diagnosis systems based on BP network. This network makes full use of the advantages of wavelet transform time-frequency localization and artificial neural network self-learning. An application on fault diagnosis of analog circuit illustrates that the new method is feasible and effective. The simulation results show that compared to BP neural network, BP wavelet neural network has faster convergence.
出处
《测控技术》
CSCD
2007年第7期64-66,69,共4页
Measurement & Control Technology
基金
国家自然科学基金资助项目(60372001)
