A Type-3 Fuzzy Parameter Adjustment in Harmony Search for the Parameterization of Fuzzy Controllers

The use of metaheuristics is currently on the rise for solving real problems due to their complexity and uncertainty management. Most of the current existing metaheuristic algorithms have the problem of local minima and fixed parameters. Fuzzy logic has contributed to solving this problem. It has been shown that the use of type-1 and type-2 fuzzy theory applied in parameter adaptation has contributed to effectively solve this problem. The advantage of utilizing fuzzy theory in parameter adaptation is the uncertainty management that offers a significant enhancement in finding solutions. The main goal is to utilize type-3 fuzzy theory in parameter adaptation of harmony search. Type-3 membership functions can use vertical slices for their construction. A new type-3 fuzzy harmony search approach is utilized to find the antecedent and consequent parameters of a fuzzy controller problem. Experiments with different lower scale parameters were carried out to verify the benefits of modeling the uncertainty domain with the type-3 fuzzy approach. A level of disturbance was applied to the control process to validate the performance of the method with respect to those existing in the literature.