Tuning of a robust PID controller-application to heating process in extruder

In this paper different methods are compared for tuning a robust PID controller for a heating process in an extruder. These tuning methods are internal model control, Clarke method, and Cohen & Coon method. Robustness is needed because the process is nonlinear. By using process model uncertainty the robust stability theorem is employed to assure that the feedback loop with the nonlinear process and the linear PID controller will be stable at all operating points. The model uncertainty is approximated with the help of a nonlinear physical model, which is based on heat balance equations. The controller parameters are selected to satisfy the requirement of the robust stability under model uncertainties