摘要
针对某型飞机机载设备故障多,且具有模糊性、复杂性的特点,本文将模糊逻辑和神经网络相结合,采用模糊隶属函数来描述这些故障的程度,建立了模糊神经网络故障诊断模型。采用图形化编程技术,开发了一种故障诊断推理流程图,方便了用户的开发。该系统依据专家知识和测试数据,可将故障隔离到内场可更换单元(SRU)或某个功能电路。实践证明该诊断系统是有效的,具有推广应用价值。
In view of the complexity of fault of airborne equipment, combined fuzzy logic with artificial neural network, the degree of faults is depicted by applying fuzzy membership functions. And fault diagnosis model based on fuzzy neural network is built. A type of inference flow chart is developed using the technology of graphical programming, which is easy for user to develop further. According to expert knowledge and test data, diagnose faults can be isolated to Shop Replaceable Unit （SRU） or some functional circuit. The result shows that this fault diagnosis system has the characteristics of fast speed inference, strong fault-tolerance ability.
出处
《光电工程》
EI
CAS
CSCD
北大核心
2007年第11期131-134,共4页
Opto-Electronic Engineering
关键词
模糊神经网络
故障诊断
推理流程
fuzzy neural network
fault diagnosis
inference flow
