Hsiuping University of Science and Technology Institutional Repository : Item 310993100/1603
English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 4334/7631
造訪人次 : 3199437      線上人數 : 366
RC Version 3.2 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
搜尋範圍 進階搜尋

請使用永久網址來引用或連結此文件: http://ir.hust.edu.tw/dspace/handle/310993100/1603

題名: Discrete-time neural predictive controller design
作者: Chi-Huang Lu
關鍵詞: generalized predictive control
recurrent neural network
nonlinear system
日期: 2009-03
上傳時間: 2009-05-24T04:06:26Z
摘要: This paper presents a design methodology for generalized predictive control (GPC) using recurrent neural network (RNN). A discrete-time mathematical model using RNN is constructed and a learning algorithm adopting an adaptive learning rate (ALR) approach is employed to identify the unknown parameters in the recurrent neural network model (RNNM). The neural predictive controller (NPC) is obtained via a generalized predictive performance criterion, and the convergence of the NPC including the adaptive optimal rate (AOR) by the Lyapunov stability theorem is presented. The illustrative process system is used to demonstrate the effectiveness of the proposed strategy. Results from numerical simulations show that the proposed method is capable of controlling nonlinear system with satisfactory performance under setpoint and load changes.
關聯: 修平學報 18, 27-38
顯示於類別:[電機工程系(含碩士班)] 期刊論文

文件中的檔案:

檔案 描述 大小格式瀏覽次數
18-03.pdf169KbAdobe PDF2637檢視/開啟

在HUSTIR中所有的資料項目都受到原著作權保護.

 

DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 回饋