A GENETIC-BASED NEURO-FUZZY APPROACH FOR MODELING AND CONTROL OF DYNAMICAL-SYSTEMS

Linguistic modeling of complex irregular systems constitutes the heart of many control and decision making systems, and fuzzy logic represen ts one of the most effective algorithms to build such linguistic model s. In this paper, a linguistic (qualitative) modeling approach is prop osed. The approach combines the merits of the fuzzy logic theory, neur al networks, and genetic algorithms (GA's). The proposed model is pres ented in a fuzzy-neural network (FNN) form which can handle both quant itative (numerical) and qualitative (linguistic) knowledge. The learni ng algorithm of an FNN is composed of three phases. The first phase is used to find the initial membership functions of the fuzzy model. In the second phase, a new algorithm is developed and used to extract the linguistic-fuzzy rules. In the third phase, a multiresolutional dynam ic genetic algorithm (MRD-GA) is proposed and used for optimized tunin g of membership functions of the proposed model. Two well-known benchm arks are used to evaluate the performance of the proposed modeling app roach, and compare it with other modeling approaches.