Identification and control of Wiener type process applied to real‐time heat exchanger

Abstract

AbstractA simple relay feedback method is proposed for the identification and control of Wiener‐type nonlinear processes. Wiener‐type processes comprise a linear dynamic subsystem followed by a nonlinear static element. A linear dynamic subsystem is identified from the shape of the output curve. If the dynamics are not captured then the order of the subsystem is improved. The nonlinear static function is identified by the simple least square estimation (LSE) method. This method avoids the iterative procedure. The inverse static nonlinearity is also obtained by the LSE method. The Wiener‐model–based nonlinear control strategy is used to compensate the nonlinear dynamics of the process, so that the controller parameters can be determined by the usual tuning rules developed for linear processes. Two important parameters are obtained through relay feedback test (RFT): amplitude of the periodic output (ku) and the ultimate period (pu). Proportional integral (PI) parameters are obtained directly from these two parameters. This proposed method is designed to adapt easily with the real‐time processes. Finally, simulated and experimental results are given to validate the proposed method. Copyright © 2008 Curtin University of Technology and John Wiley & Sons, Ltd.