Model Predictive Control of PMSM Considering Online Parameter Identification

As a high performance control method, model predictive control has many advantages, such as fast dynamic response, simple structure and ability to deal with multivariables, but its control effect depends on the motor model. Aiming at the problem of the parameter sensitivity of model predictive control, in this paper, a model predictive control considering online parameter identification algorithm is proposed. By establishing model reference adaptive system, and designing the parameter adaptive law base on Popov's Superstability theory, the inductance and flux parameters online identified are updated to the model predictive control. Thus, it solves the problem of model predictive control in parameter mismatch, and realizes high performance control under the motor runs for a long time. The simulation results verify the effectiveness of the proposed method.