A Nonlinear Predictive Control Algorithm Based on Wiener Model

An improved nonlinear predictive control algorithm is proposed for a class of nonlinear systems which are described by Wiener model.In the algorithm,Laguerre functions are used to describe the control signals from the dynamic linear section of Wiener model,and the optimization solutions of the future control input sequences in predictive control are converted into the optimization of a set of immemorial Laguerre coefficients within prediction horizon in order to reduce the computation burden in optimization.A static fuzzy model is used to approximate the nonlinear section of Wiener model,and the optimization of nonlinear predictive control is converted into the optimization problem of linear predictive control.Consequently the difficulty in solving the nonlinear equations are overcome for obtaining control input.Analysis solution of predictive control input is further deduced.Simulation results of CSTR(Continuous Stirred Tank Reactor) process show that the proposed algorithm is valid and feasible.