Prediction Current Compensation Using Adaptive Torque Feedforward Method for Permanent Magnet Synchronous Motors

The disturbances or loads in motor drive systems can be classified into two categories, random unknowns and regular knowns such as cogging torque or sometimes regular load torque. In conventional predictive control method, all of them are treated as random unknown disturbances, which would increase the load of controller and lead to performance degradation. In this paper, the known disturbances are considered individually, and an improved predictive control strategy which compensates prediction current using adaptive torque feedforward method is proposed to reduce the effect of regular disturbances. Using the proposed predictive control combined with adaptive feedforward method, the fluctuations of output torque is reduced by 45% and the speed ripples is decreased to less than a quarter of the original values.