Fuzzy supervisory control using fuzzy c-means clustering combined with fuzzy gain scheduling

In this paper, a technique using fuzzy c-means clustering combined with fuzzy gain scheduling to derive fuzzy structure, namely membership junctions and fuzzy IF-THEN rules was proposed. This technique was applied to perform fuzzy supervision for adjusting the parameters of a PI controller. The proposed fuzzy supervisory PI controller is applied to a binary distillation column. The task of the controller is to maintain the product composition when the disturbance enters the column in the form of the changes in feed flow rate. The simulation results show that better control performance were obtained compared with a PI controller tuning with fixed parameters.