Adaptive Neuro Fuzzy Controller for Process Control System

The generation of membership function for fuzzy system is a challenging problem. The response of fuzzy controller is highly dependent on the membership function which is one of the design parameter in fuzzy system. The optimization of membership function can be viewed as identification problem for a nonlinear dynamic system. This paper presents an Adaptive Neuro Fuzzy Inference System (ANFIS) based controller for water temperature control. The ANFIS based input - output model is used to tuned the membership functions in fuzzy system. Different membership functions are tested and the mean and variance has been calculated. Experimental results are compared with the conventional controller based on Zieglar - Nichols tuning method. Both the controllers are tested in various operating conditions and varying setpoint changes. This shows that better performance can be achieved with ANFIS tuning.