Adaptive nonlinear model-based control scheme implemented on the nonlinear processes

Abstract

In this work, two adaptive nonlinear model-based control schemes have been proposed and implemented on the simulated model of the nonlinear benchmark processes. The servo performance of the proposed control schemes was found satisfactory. In order to improve the servo-regulatory performance of one of the proposed control schemes, the model state and parameters have been estimated simultaneously online with the help of a derivative-free Kalman filter and the predicted values of model states have been used in the proposed control law. The performances have been compared between two proposed control schemes with conventional adaptive PI control scheme. From the extensive simulation studies, it has been found that proposed control schemes implemented on nonlinear processes are having better performance over conventional adaptive PI control scheme. It was also found that proposed control schemes are able to eliminate measurement noise and having good robustness features.