基于人工神经网络的电路故障诊断专家系统

An Expert System for Circuit Fault Diagnosis Based on Artificial Neural Networks

分析了电路故障的复杂性。针对传统电路专家诊断系统的不足 ,结合人工神经网络用于故障诊断的优点 ,构造了一种新型神经网络电路故障诊断专家系统。该系统采用数据表格形式表示测试诊断知识 ,使知识表达更为清晰 ,推理过程更具智能性 ,在实际运用中 ,诊断效率有很大提高。着重介绍了该系统的设计思想、组成结构和工作原理。

In this paper the complexity of circuit faults is analyzed. In view of the shortcomings in traditional expert systems for circuit fault diagnosis and the promising usage of artificial neural networks in this field, a new expert system for circuit fault diagnosis is constructed. Tables are used to present the knowledge of the circuits, which makes the expression of knowledge more clearly,and the process of rationalization more intelligent. The efficiency of the new system is verified in practice. Moreover, the design idea, architecture and working principle of the new system are described in detail.\;