Design of mixed static/dynamic output feedback controller using multi-objective distribute particle swarm optimization

Abstract

This paper discusses a new design method for a mixed static/dynamic output feedback controller that can simultaneously achieve conflicting goals such as the minimization of both H-2 and H-infinity norms. In a general procedure for designing the mixed controller, the structure of controller is first fixed, and then the controller is designed. However, if such a controller structure is arbitrarily chosen by the designer without careful consideration, no one can guarantee that the selected controller structure is the optimal one. From the above viewpoint, this research first proposes a novel multi-objective distribute particle swarm optimization algorithm, and then explains in detail how this algorithm can be applied to complex controller design problems of finding the optimal controller structure and its optimal parameters simultaneously. Finally, simulation results are provided to show the superiority of the proposed methodology.