模糊神经网络在自动变速器电控系统故障自诊断中的应用  

Application of fuzzy neural network in fault self-diagnosis for electronic control system of automatic transmission

作　　者：孔慧芳[1] 何吉柱[1] 

机构地区：[1]合肥工业大学电气与自动化工程学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2010年第4期501-505,518,共6页Journal of Hefei University of Technology：Natural Science

基　　金：国家科技部支撑计划重点资助项目(2009BAG12B03);安徽省自然科学基金资助项目(08040102001);合肥工业大学科学研究发展基金重点资助项目(105-036413)

摘　　要：文章介绍了故障自诊断的步骤,针对自动变速器电控系统的特点,建立了故障征兆与故障原因映射列表,并应用基于模糊神经网络的故障自诊断方法,实现了对自动变速器电控系统故障的模糊诊断;并介绍了模糊神经网络的结构和学习方法,以及如何对训练样本进行预处理;利用Matlab神经网络工具箱作为平台,进行测试样本仿真和试验,结果表明该诊断方法是行之有效的。This paper firstly introduces the approaches to the design of fault self-diagnosis. Then a list of fault symptom and fault cause is made in allusion to the characteristics of electronic control system of automatic transmission, and the approach of fault self-diagnosis based on fuzzy neural network is applied to realize the fuzzy diagnosis of faults in the mentioned system. The structure of the fuzzy neural network and corresponding learning methods are introduced as well as the ways to pretreat the training samples. Finally, this diagnosis approach is simulated and tested by Matlab＇s Neural Networks Toolbox to demonstrate its validity and practicability.

关 键 词：模糊逻辑 神经网络技术 自动变速器电控系统 故障自诊断 

分 类 号：TP206.3[自动化与计算机技术—检测技术与自动化装置]