Performance Improvement of Permanent-Magnet Synchronous Motor Through a New Online Predictive Controller

In this article, a new online continuous control set-predictive controller (CCS-PC) is proposed with the objective of minimizing the stator-flux and torque ripples. In this controller, a constrained optimization problem is constructed to optimize the phase, amplitude, and time duration of the voltage vector applied to the motor. Fast Iterative Shrinkage-Thresholding Algorithm (FISTA) which is an advanced algorithm based on fast proximal gradient method is adopted to find the optimal solution in an online manner. In each control period, FISTA is executed 10 times where the voltage vector found in the last iteration is considered as the optimal voltage vector. Performance of the proposed controller is compared with the performance of an offline CCS-PC through various experiments. In these experiments, stator-flux ripple, torque ripple, stator-current total harmonic distortion, and computational complexity of the controllers are compared with each other. Based on the results, pros and cons of the proposed online CCS-PC are highlighted.