摘要
大型电路故障诊断过程复杂,故障征兆和故障原因之间存在许多不确定因素,建立精确的故障诊断系统存在一定困难。单纯运用专家系统存在许多不足之处。为此,文中基于BP(反向传播)神经网络,利用神经网络与专家系统相结合建立故障诊断系统,阐述了这种故障诊断系统的构建方法及其推理机的训练进程,并以某电路为对象对该诊断系统进行了试验验证。结果表明,该方法是可行和有效的。
The process of the electro circuit fault diagnosis is complex. There are many uncertain factors between the fault symptoms and the fault causes , so it＇s difficult to construct the fault diagnosis system with high precision. The diagnosis expert systems exist a lot of shortages. This paper constructs the fault diagnosis system based on the fuzzy neural network （FNN） and expert system and also illustrates the construct method of this diagnosis system and the training process of the inference mechanism. This method is been tested on a type of circuit . The result shows that this method is feasible and effective.
出处
《电子工程师》
2005年第8期71-73,79,共4页
Electronic Engineer
关键词
神经网络
专家系统
故障诊断
BP算法
fuzzy neural network, expert system, fault diagnosis, BP arithmetic
