Adaptive control of nonlinear systems using nonlinear and linear models

The authors are concerned with the evaluation of an adaptive linearizing output feedback control strategy developed for a class of nonlinear delayed discrete-time systems. These nonlinear models can be obtained by several methods and can describe plants whose dynamics depends on the operating point. For this class of nonlinear discrete-time systems, expressed in input-output form, a model reference nonlinear controller is considered. An extension to this control law by weighting the filtered incremental control in the cost function to be minimized is proposed. For adaptive control, algorithms for estimating the controller or model parameters are developed. The approach is evaluated through simulations performed using a simplified nonlinear model of a heat exchanger. Comparisons with fixed and adaptive controllers using linear models show the performance improvements obtained