计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (14): 231-234.DOI: 10.3778/j.issn.1002-8331.2010.14.069

• 工程与应用 • 上一篇    下一篇

神经网络信息融合用于电梯故障诊断的研究

李 端,艾永乐   

  1. 河南理工大学 电气工程与自动化学院,河南 焦作 454000
  • 收稿日期:2008-11-11 修回日期:2009-01-20 出版日期:2010-05-11 发布日期:2010-05-11
  • 通讯作者: 李 端

Study on fault diagnosis of elevator based on neural network information fusion technology

LI Duan,AI Yong-le
  

  1. School of Electrical Engineering and Automation,Henan Polytechnic University,Jiaozuo,Henan 454000,China
  • Received:2008-11-11 Revised:2009-01-20 Online:2010-05-11 Published:2010-05-11
  • Contact: LI Duan

摘要: 针对传统电梯故障诊断系统实时性有限、故障定位准确率低等问题,将多信息融合技术引入到电梯故障诊断中来,建立了基于模糊神经网络和D-S证据理论相结合的故障诊断模型。为了提高神经网络的训练速度和推广能力,采用了正则化算法对BP网络算法进行修改,并且利用D-S证据理论对神经网络的诊断结果进行决策融合,仿真结果表明了此方法有效地提高了故障诊断的准确率。

关键词: 故障诊断, 信息融合, 模糊神经网络, D-S证据理论

Abstract: Researchers have studied on elevator fault diagnosis,but timeliness is limited,and can’t discover reason of the faults accurately.To solve these problems,this paper introduces the multi-information fusion technology to the fault diagnosis of elevator control system.The fault diagnosis model:A combination of fuzzy neural network and D-S evidential theory is set up.The BP neural network’s algorithm has been revised in order to enhances its study speed and the ability for precise diagnosis from the D-S evidential theory is fully made use of.The simulation of the fault phenomena has the good effect of this fault diagnosis system.

Key words: fault diagnosis, information fusion, fuzzy neural network, D-S evidential theory

中图分类号: