T-S FUZZY MODEL IDENTIFICATION USING A DIFFERENTIAL EVOLUTION ALGORITHM WITH DOUBLE OPTIONAL EXTERNAL ARCHIVES

摘要

This paper presents a new differential evolution (DE) algorithm for automatically extracting the locally linear Takagi-Sugeno (T-S) fuzzy model from a set of data. Two optional external archives and a new process of replacement are designed to enhance the global search capability of DE and prevent the algorithm from becoming trapped around local optima. Validation results with other swarm intelligence algorithms used in T-S model identification are presented to demonstrate the effectiveness and efficiency of the proposed algorithm.