H∞ controller design for a distillation column using genetic algorithms

In the conventional H∞ optimal output feedback control design, the order of the controller is higher than that of the plant. Thus, it is not easy to implement H∞ optimal control in real world engineering problems. This paper tries to bridge the gap between the theoretical H∞ control and practical applications. It proposes a simple genetic algorithm (GA), blended with multi-objective characteristics, for the design of a structure specified H∞ sub-optimal controller. The proposed procedure is applied to control the distillation benchmark problem formulated for the IEEE Conference on Decision and Control (CDC 1991).