A Comparative Study of Fuzzy Systems and Neural Networks for System Modeling and Identification

In this paper, a comparative study is presented for modelling and identification of non-linear systems. In this paper, Fuzzy modeling is done using Gradient Descent method and Neural Network modeling is done using Back Propagation and their performance index (PI) and identifier output is used as a basis of comparison. In this paper, three case studies are presented to illustrate which method provides better approximation results and robustness. Simulation are done using MATLAB 2015a and simulation results are presented.