摘要
提出了一种远程模糊神经网络故障诊断系统,该诊断系统采用多个子网络的结构实现对多种标准故障模式的记忆。对于未知样本输入,该系统针对每种标准故障模式采用双向联想记忆过程联想出最佳结果,通过计算联想结果和标准模式的子集函数,确定该样本的故障类别。利用船舶柴油主机系统实例,详细介绍了该系统的组件模型实现技术,最后给出相应的结果。
A trouble-shooting system based on fuzzy neural networks is presented in the paper. This system adopts a series of sub-networks to memorize the standard patterns of various faults. When an unknown sample is fed into the trained trouble-shooting system, it can produce the best result by means of the bi directional association. Through calculating the subset function between the associated result and the standard pattern, a fault type is determined. With the diesel engine system of a ship as an example, the module model and the implementation of the system are illustrated in detail. In the end, the result of the system is presented.
出处
《国防交通工程与技术》
2004年第4期26-29,共4页
Traffic Engineering and Technology for National Defence
关键词
模糊神经网络
故障诊断
双向联想
船舶柴油主机
fuzzy neural networks
fault diagnosis
bi-directional association
the diesel engine system of a ship
