摘要
对模拟电路故障诊断提出了一种双神经网络的诊断方法,该方法通过将电路故障模式分类的预处理,再用双神经网络分别对不同的故障模式类进行诊断。仿真实验表明,该方法能够实现故障检测及定位,具有准确率高的特点。
A method of analog circuit fault diagnosis based on dual neural network was presented. By preprocessor, the proposed method classes circuit fault patterns, and uses dual neural networks as diagnosis tools working for each fault class respectively, The simulation experiments result shows that this scheme is feasible and has many powerful virtues, such as diagnosing and locating faults quickly and exactly.
出处
《低压电器》
北大核心
2008年第3期17-19,51,共4页
Low Voltage Apparatus
基金
广西壮族自治区自然科学基金项目资助(桂科自0640068)
关键词
故障诊断
模拟电路
双神经网络
fault diagnosis
analog circuit
dual neural networks
