HGA fuzzy supervisory control system of a binary distillation column

This paper presents the design of a hierarchical genetic algorithm (HGA) fuzzy supervisory control system. The reliable PI controller is still used in the control loop, but its parameters are adapted by a HGA fuzzy system. The effectiveness of this genetic algorithm enables the fuzzy membership functions and rules to be optimally reduced while consuming less computing time. The proposed scheme is applied to a binary distillation column to maintain product´s composition as close as possible to the set point despite the presence of perturbations in the feed flow rate. The HGA fuzzy supervisory control system has revealed a good compromise between performance and computational efforts