Skip to Main Content
A coordinated control strategy is often used to ensure a thermal power plant to have a higher rate of load change, but without violating the thermal constraints. Although model predictive control has been widely used for controlling power plant, handling input constraints is a major problem especially as these plants are nonlinear. Two alternative methods of exploiting the nonlinear predictive control are presented in this paper. One is the input-output feedback linearization technique based on a suitably chosen approximated linear model. The other is based on neuro-fuzzy networks to represent a nonlinear dynamic process using a set of local models. From the criteria based on the integral absolute errors and the relative optimization time for completing the simulation, it is shown that the performance of the coordinated control of a steam-boiler generation plant using these two nonlinear predictive methods are better than the conventional predictive method.