Evolutionary Learning in Identification of Fuzzy Model of Air Flow Supply System

Evolutionary learning and especially genetic optimization algorithms have recently received a lot of research attention as tools for identifying fuzzy models of the systems. Most often fuzzy modelling employs the fuzzy IF-THEN rules. In this paper, besides AND-operator the OR-operator is also considered in constructing the premise rule base. A genetic algorithm is utilized to find the premise structure of the rules, also to optimize fuzzy set membership functions as well as the consequent part of model structure at the same time. The performance of the approach is demonstrated on the laboratory stand named "airflow stabilization system".