In this paper, for the purpose of simulating the mathematical model of adaptive neuro-fuzzy inference system (ANFIS), we use Matlab/Simulink environment with its powerful S-functions. The simulated model of ANFIS network can be then used to make the simulation of the identification and the control of linear or nonlinear systems. Created Simulink block of ANFIS give flexible exploitation of parameters of ANFIS network like learning rates and initial local parameters. We use the S-function of ANFIS to make the direct-inverse adaptive control of DC-motor. The obtained results of Direct-inverse control of DC-motor are compared with these of a classical controller which is the simplest type of controller, selected only to check the effectiveness of the proposed intelligent controller in terms of control performances and disturbance rejection.

Key words: Direct inverse control, adaptive neuro-fuzzy inference system (ANFIS)controller, S-function of Matlab.