A new method for identification of fuzzy models based on evolutionary algorithms and its application to the modeling of a wind turbine

This paper presents a novel fuzzy model identification method, which is based on Genetic Algorithms and Particle Swarm Optimization. The proposed method is compared to other existing strategies for identification of fuzzy systems and equivalent linear models. A wind turbine system is used to verify and validate the proposed strategy. For purposes of this work, it is assumed that the simulator of the plant represents the actual system that needs to be identified. Simulations are carried out in continuous time and data are acquired with fixed sample time to generate a black box model of the system, using different techniques of identification.