DocumentCode :
387609
Title :
A prediction algorithm based on self-organizing fuzzy neural networks
Author :
Liu, Min ; Gu, Yun-dong ; Chai, Ying-chun
Author_Institution :
Dept. of Autom., Tsinghua Univ., Beijing, China
Volume :
3
fYear :
2002
fDate :
2002
Firstpage :
1688
Abstract :
The dyeing process in the cloth-weaving works is a complex chemical reaction. It is a complex nonlinear multivariable problem to predict the success rate of dyeing since it is affected by many factors. A new prediction algorithm based on structure-adaptive self-organizing fuzzy network is proposed in this paper, which combines supervised competitive learning algorithm with node generation method. The algorithm can adjust both structure and the weights according to the change of environment. Besides, Fuzzy synthesis prediction mechanism is introduced into the algorithm to improve the precision and stability of prediction. Digital simulations show that the algorithm is effective and suitable for larger scale prediction problem.
Keywords :
digital simulation; dyeing; fuzzy neural nets; large-scale systems; multivariable systems; nonlinear systems; prediction theory; process control; self-organising feature maps; textile industry; cloth-weaving works; complex chemical reaction; complex nonlinear multivariable problem; digital simulations; dyeing process; fuzzy synthesis prediction mechanism; prediction algorithm; self-organizing fuzzy neural networks; structure-adaptive self-organizing fuzzy network; success rate prediction; supervised competitive learning algorithm; Automation; Chemical processes; Couplings; Digital simulation; Fuzzy neural networks; Network synthesis; Neural networks; Prediction algorithms; Production; Stability;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Machine Learning and Cybernetics, 2002. Proceedings. 2002 International Conference on
Print_ISBN :
0-7803-7508-4
Type :
conf
DOI :
10.1109/ICMLC.2002.1167502
Filename :
1167502
Link To Document :
بازگشت