Multicriteria Fuzzy Control Using Evolutionary Programming

Abstract In this paper, we propose a multicriteria fuzzy control (MFC) using evolutionary programming (EP) by which its λ-fuzzy measure is optimized. Basically the MFC is an extended version of the existing fuzzy logic controller to obtain the proper weights of the fuzzy rules for the fine tuning control. The MFC is designed by three steps: first, the attributes of the system output response are selected, second, the partial evaluations from the viewpoints of the attributes are assigned to the fuzzy rules, and third, the fuzzy integral is employed to aggregate the partial evaluation and the user's fuzzy measure about the consideration degree of each attribute for the proper weight value. EP is used to optimize the λ-fuzzy measure for the linear tuning property of the MFC. Computer simulations indicate that the proposed scheme has excellent performance.