首页 | 本学科首页   官方微博 | 高级检索  
     

基于改进小波网络的TE过程故障诊断
引用本文:刘晓琴,申东日,苏成利. 基于改进小波网络的TE过程故障诊断[J]. 辽宁石油化工大学学报, 2007, 27(4): 64-67
作者姓名:刘晓琴  申东日  苏成利
作者单位:辽宁石油化工大学信息与控制工程学院,辽宁抚顺,113001;辽宁石油化工大学信息与控制工程学院,辽宁抚顺,113001;辽宁石油化工大学信息与控制工程学院,辽宁抚顺,113001
摘    要:针对BP 算法容易陷入局部极小值、收敛速度慢及容易振荡等缺点, 采用小波BP 网络且对小波网络采用基于梯度符号变化的局部学习率自适应算法和引入动量项的改进。将改进后的算法对多变量非线性的田纳西-伊斯曼过程进行了仿真研究, 结果表明改进算法提高了故障分类的辨识精度。

关 键 词:小波网络  故障诊断  TE过程
文章编号:1672-6952(2007)04-0064-04
收稿时间:2007-02-06
修稿时间:2007-09-24

Fault Diagnosis of Tennessee-Eastman Process Based on the Improved Wavelet Network
LIU Xiao-qin,SHEN Dong-ri,SU Cheng-li. Fault Diagnosis of Tennessee-Eastman Process Based on the Improved Wavelet Network[J]. Journal of Liaoning University of Petroleum & Chemical Technology, 2007, 27(4): 64-67
Authors:LIU Xiao-qin  SHEN Dong-ri  SU Cheng-li
Affiliation:School of Information and Control Engineering , Liaoning Univ ersity o f Petroleum &Chemical Technology , Fushun Liaoning 113001, P .R .China
Abstract:BP algorithm trends to fall into the local minimum value, slow convergence speed and frequent oscillation. The wavelet BP network was used, and self-adaptive learning rate algorithm based on the sign change of gradient and momentum item were added in it. The improved algorithm was applied in Tennessee-eastman process of a multiple-variable and nonlinear system. The results show that the algorithm can improve the recognition accuracy of fault classification.
Keywords:Wavelet network   Fault diagnosis  Tennessee - eastman process
本文献已被 维普 万方数据 等数据库收录!
点击此处可从《辽宁石油化工大学学报》浏览原始摘要信息
点击此处可从《辽宁石油化工大学学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号