Nonlinear system identification based on ANFIS

System identification is the basis of designing control system, and it is very difficulty to identify the nonlinear system today. A method of fuzzy identification had been provided in reference 1, and another method of using neural networks to identify system had been provided in reference 2. In this paper, Author points out the disadvantages of those methods and a new identification method based adaptive neural-fuzzy inference system (ANFIS) is provided, as assembles the advantages of fuzzy theory and neural networks. The structure and algorithms of ANFIS is designed firstly, then a nonlinear function is tested using the method, and the simulation results show that ANFIS is very effective to identify the nonlinear system.