PI-FUZZY CONTROLLER DESIGN BASED ON AN OPTIMIZATION APPROACH

The paper proposes a new Iterative Feedback Tuning (IFT)-based approach to the optimisation of a class of Takagi-Sugeno PI-fuzzy controllers (PI-FCs). A combination of IFT and the Extended Symmetrical Optimum method is used for the design of PI-FCs for servo systems with controlled plants characterized by second-order integral type models. Since the gradient in the optimisation approach is estimation, the results are quasi-optimal. Experimental results for a PI-fuzzy controlled servo system validate the proposed IFT-based design.