Performance Enhancement of Supervisory Control for Largely Mismatched Processes

Even through being an effective approach to handle robustness, the single H-infinity controller is challenging to meet strict performance requirement when the modeling mismatch is relatively large. This paper presents a supervisory method to enhance the robust stability and performance of H-infinity control using a multiple model switching configuration. A new scheduling logic which selects the most proper controller into loop is proposed to ensure bounded exponentially weighted H-infinity norm of the closed loop system. The candidate controllers are obtained by solving a set of linear matrix inequalities (LMIs). The effectiveness of this supervisory H-infinity control method is validated with a fist-order inertial process with pure time delay.