Best Choosing of Rule Set for Adaptive Neuro-Fuzzy Inference System (ANFIS)

Abstract

The neuro-fuzzy controller incorporates fuzzy logic algorithm with an artificial neural network (ANN) structure. The conventional PI controller is replaced by Adaptive Neurofuzzy Inference System (ANFIS), which tunes the fuzzy inference system with hybrid learning algorithm, This makes fuzzy system training with performance of the neuro-fuzzy based vector controlled of the system under controlled. This paper concentrates on choosing the best rule base and gives some useful results. The simulation of the design is achieved by using Matlab Ver.2010a. Chopper-Fed DC Motor Drive (Continuous / Discrete) are consider as case study. Satisfactory results are obtained, that is explaining the ability of ANFIS controller to control with the dynamic high nonlinear system and can be get very good results by tunes the fuzzy controller.