| A MIMO subsystem that regulated the steam temperatures by adjusting the burner tilt positions was extracted from the integrated control system of Unit 9 at Kingston Steam Power Plant to study the application of advanced control techniques on power plant. A mathematical model was developed for this physical system with the MATLAB SITB based on system identification theory. The model's uncertainty was analyzed.;A MIMO controller was then designed for this model using D-K iteration method. The nominal performance, robust stability, robust performance, and worst-case performance were analyzed. The purpose is to compare the LQG design method with D-K iteration method and improve the estimate of the uncertainty weighting function. The influences of different weighting functions Wp on closed-loop systems were analyzed for the D-K iteration design.;The MIMO controller was tested in the real plant through the interface of the DCS in Unit 9. Test results showed that the new MIMO controller behaved quite well. |