CIESC Journal

• RESEARCH PAPERS • 上一篇    下一篇

基于小波网络的非线性动态过程建模和在环氧氯丙烷生产过程中的应用

黄德先a; 金以慧a; 张杰b; A. J. Morrisb   

  1. a Department of Automation, Tsinghua University, Beijing 100084, China
    b CPACT, Merz Court, University of Newcastle, Newcastle upon Tyne, NE1 7RU, U.K.
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2002-08-28 发布日期:2002-08-28
  • 通讯作者: 黄德先

Non-linear Chemical Process Modelling and Application in Epichlorhydrine Production Plant
Using Wavelet Networks

HUANG Dexiana; JN Yihuia; ZHANG Jieb; A. J. Morrisb   

  1. a Department of Automation, Tsinghua University, Beijing 100084, China
    b CPACT, Merz Court, University of Newcastle, Newcastle upon Tyne, NE1 7RU, U.K.
  • Received:1900-01-01 Revised:1900-01-01 Online:2002-08-28 Published:2002-08-28
  • Contact: HUANG Dexian

摘要: A type of wavelet neural network, in which the scale function is adopted only, is proposed
in this paper for non-linear dynamic process modelling. Its network size is decreased
significantly and the weight coefficients can be estimated by a linear algorithm. The
wavelet neural network holds some advantages superior to other types of neural networks.
First, its network structure is easy to specify based on its theoretical analysis and
intuition. Secondly,network training does not rely on stochastic gradient type techniques
and avoids the problem of poor convergence or undesirable local minima. The excellent
statistic properties of the weight parameter estimations can be proven here. Both
theoretical analysis and simulation study show that the identification method is robust and
reliable.Furthermore, a hybrid network structure incorporating first-principle knowledge
and wavelet network is developed to solve a commonly existing problem in chemical
production processes. Applications of the hybrid network to a practical production process
demonstrates that model generalisation capability is significantly improved.

关键词: wavelet;neural network;non-linear system identification;hybrid neural network

Abstract: A type of wavelet neural network, in which the scale function is adopted only, is proposed
in this paper for non-linear dynamic process modelling. Its network size is decreased
significantly and the weight coefficients can be estimated by a linear algorithm. The
wavelet neural network holds some advantages superior to other types of neural networks.
First, its network structure is easy to specify based on its theoretical analysis and
intuition. Secondly,network training does not rely on stochastic gradient type techniques
and avoids the problem of poor convergence or undesirable local minima. The excellent
statistic properties of the weight parameter estimations can be proven here. Both
theoretical analysis and simulation study show that the identification method is robust and
reliable.Furthermore, a hybrid network structure incorporating first-principle knowledge
and wavelet network is developed to solve a commonly existing problem in chemical
production processes. Applications of the hybrid network to a practical production process
demonstrates that model generalisation capability is significantly improved.

Key words: wavelet, neural network, non-linear system identification, hybrid neural network