A Genetic Based Neuro-Fuzzy Controller for Thermal Processes

This paper presents a neuro-fuzzy network where all its parameters can be tuned simultaneously using Genetic Algorithms. The approach combines the merits of fuzzy logic theory, neural networks and genetic algorithms. The proposed neuro-fuzzy network does not require a priori knowledge about the system and eliminates the need for complicated design steps like manual tuning of input-output membership functions, and selection of fuzzy rule base. Although, only conventional genetic algorithms have been used, convergence results are very encouraging. A well known numerical example derived from literature is used to evaluate and compare the performance of the network with other modelling approaches. The network is further implemented as controller for two simulated thermal processes and their performances are compared with other existing controllers. Simulation results show that the proposed neuro-fuzzy controller whose all parameters have been tuned simultaneously using GAs, offers advantages over existing controllers and has improved performance.