Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/1760
標題: 啟發式模糊學習控制系統
heuristic neural-fuzzy learning control system
作者: 陳盈達
Chen, Ying-Ta
關鍵字: neural-fuzzy;類神經模糊;on-line training;motor position control;線上學習;馬達定位控制
出版社: 機械工程學系
摘要: 
本論文中,我們結合TSK網路模式及Mamdain歸論方式,提出一套新而能夠真
實線上及時取樣學習之類神經模糊控制系統.我們先以線性方程式以取代
傳統邏輯推論過程,並運用TSK網路模式傳輸訊號,在以通用之規則庫與簡
易之解模糊化步驟,提昇網路推論能力;在網路訓練方面,經由啟發式學習
法則配合切換模式,在不需要繁雜的系統辨識工作下迅速取得有效之訓練
樣本,進而精確修改類神經模糊之前見部語言項隸屬函數之中心,寬度與後
件部歸論權值,而達到真正線上學習控制之目的.

In this thesis,a new neural-fuzzy control scheme that
combines the TSK type network structure,the Mamdain's
consequence mode,and the heuristic learning rule is proposed.
In this new neural-fuzzy structure,a set of linear equations is
usedto replace the convientional logical reasoning process and
the TSK typetransmission mode is executed for more efficient
data processing.In addition,the general fuzzy rule base and a
simple defuzzification method are alsoemployed to increase the
reasoning ability of the network .With those new functions,
parameter need to be learned in this new seheme are reduced
whilethe performance is improved.
URI: http://hdl.handle.net/11455/1760
Appears in Collections:機械工程學系所

Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.