Abstract:

Based on T-S fuzzy system, a model reference adaptive disturbance canceling control method is proposed for nonlinear systems. The proposed method combines the fuzzy identification theory with the model reference adaptive inverse control according to their respective characteristics. First, according to the fuzzy identification theory, the premise structure identification is achieved by using diagonal division and the consequent parameters are identified by the recursive least squares method. Then model and inverse model of the plant are obtained in offline manner. The inverse model is connected to the plant in series as the initial controller. The model is parallel to the plant so as to determine the disturbance response. The disturbance is feedback to the input of the plant through the inverse model. While the system is running, the parameters of the controller are adjusted by least mean square adaptive algorithm to minimize the system error. Finally, the proposed method is applied to tracking control of a hybrid nonlinear system and input/output nonlinear system. The simulation results verify the effectiveness of the proposed method.