Brushless DC Motor Control Based on Perturbation Observer in Model Predictive Control Method

At present, the FOC control method is often used in Brushless DC Motor (BLDCM), however, the accuracy of the FOC depends heavily on complex vector modulation and PI parameter regulation. Model predictive control (MPC) can deal with discrete switching states directly without vector modulation and PI parameter regulation, and has faster current response. However, no matter FOC strategy or MPC strategy, the noise of measured current is unavoidable. Moreover, the control accuracy of MPC mainly depends on the accuracy of motor model. It has been proposed to use Kalman observer to filter the measured current to reduce the noise in the quadrature current. In practical application, the parameters of the motor vary with different working conditions, and there are some unknown disturbances. These factors will greatly reduce the accuracy and stability of current predictive control. Therefore, this paper proposes a model predictive control method with disturbance observer. In this paper, the disturbance term is added to the motor model, disturbance observation is added to the Kalman current observer, and the current is used to estimate the disturbance value online, which reduces the accuracy requirement of the motor model by MPC strategy. The current robustness and disturbance rejection performance of the proposed strategy are demonstrated in a simulation based on MATLAB/Simulink.