摘要
介绍了模拟电路故障诊断的神经网络方法及小波神经网络结构和原理,以一带通滤波器为例,提出了一种基于输出灵敏度分析,利用多频测试生成故障特征向量训练小波神经网络进行故障诊断的方法,仿真结果表明小波神经网作为故障分类器具有收敛速度快,诊断准确等特点。
Introduces the method of analog circuit faults diagnosis based on neural network and the theory of wavelet neural network, offers the method which used the output sensitivity analysis and multi-frequency test to extract faults feature vectors, the simulation result shows that the wavelet neural network has the properties of fast convergence and accurate diagnosis as a faults classifier.
出处
《微计算机信息》
北大核心
2006年第06S期206-208,共3页
Control & Automation
关键词
小波神经网络
模拟电路
故障诊断
特征向量
wavelet neural network
analog circuit
fault diagnosis
feature vectors
