Improved Performance of PMSM Control Based on Nonlinear Control Law and Computational Intelligence

Abstract

Based on the classic Field Oriented Control (FOC) structure of the Permanent Magnet Synchronous Motor (PMSM), a control structure based on the nonlinear currents control law is presented, and, for the optimization of its parameters, a set of computational intelligence (CI) algorithms were used: Particle Swarm optimization (PSO) type algorithm, Simulated Annealing (SA) type algorithm, Genetic Algorithm (GA) type, Gray-Wolf optimization (GWO) type algorithm, and Reinforcement Learning (RL) type of Twin Delayed Deep Deterministic Policy Gradient (TD3) agent. The corresponding control structures are presented, as well as the related stages: design, training, and performance evaluation. The numerical simulations realized in the Matlab/Simulink programming computing platform confirm the improvement of the performance of PMSM control system (PMSM-CS) using the nonlinear currents controller and CI.