Title :
An improved Fuzzy neural networks algorithm and its application in resistance furnace
Author :
Lu, Yong-Song ; Pan, Hai-Peng
Author_Institution :
Dept. of Autom., Zhejiang Sci-Tech Univ., Hangzhou, China
Abstract :
Resistance furnace has the characteristics of nonlinear, time-delay, time-varying and so on. In order to overcome the disadvantage of the traditional fuzzy neural network (FNN) controller with long setting time, an improved fuzzy neural network controller is presented in this paper. It consists of two parts. One is the traditional fuzzy neural network. The other is a compensator, which is used to accelerate the response of the plant. Based on the recursive least square (RLS) algorithm, the parameters of the model are identified on line. According to this model, the trend of the plant can be predicted and the overshoot can be restrained. And the simulation result indicates that this method has the characteristics of short setting time and obvious improvement of traditional fuzzy neural network.
Keywords :
delays; fuzzy control; fuzzy systems; industrial control; least squares approximations; neurocontrollers; nonlinear control systems; resistance furnaces; time-varying systems; FNN algorithm; RLS algorithm; fuzzy neural network controller; fuzzy system; industrial control; nonlinear system; recursive least square algorithm; resistance furnace; time-delay; time-varying system; Automatic control; Control systems; Error correction; Furnaces; Fuzzy control; Fuzzy neural networks; Fuzzy systems; Least squares methods; Mathematical model; Predictive models; FNN; compensator; recursive least square algorithm; resistance furnace;
Conference_Titel :
Computing, Communication, Control, and Management, 2009. CCCM 2009. ISECS International Colloquium on
Conference_Location :
Sanya
Print_ISBN :
978-1-4244-4247-8
DOI :
10.1109/CCCM.2009.5268009
