期刊刊名：修平學報 　 卷期：23期

篇名出版日期：2011年9月1日

作者：Chi-Huang Lu, Chi-Ming Lu, Yuan-Hai Charng,呂奇璜,呂奇明,常元海

語言：English

關鍵字：Elman neural network, hierarchical BP algorithm, model predictive control, PID controller,Elman類神經網路, Hierarchical演繹法,模型預估控制,PID控制器

被點閱次數：14次

閱讀時間：370sec

摘要：
This paper presents a predictive proportional-integral-derivative (PID) controller based on Elman neural network (ENN) for a class of nonlinear systems. The ENN with both online learning and well approximation capability is employed to estimate the nonlinear function of the controlled system. The weights of the ENN identifier are trained by the hierarchical backpropagation algorithm with the adaptive learning rate, the adaptive learning rate is suitable for the ENN identifier can be convergent. The predictive PID controller is derived via a predictive performance criterion and the adaptive optimal rate for guaranteeing the convergence of the proposed PID controller. The stability analysis of the closed-loop control system is presented by the discrete Lyapunov stability theorem. Numerical simulations reveal that the proposed control law gives satisfactory tracking and disturbance rejection performances.

　　　　　[全文下載]