Improvement Performances of Sensorless Control for PMSM Based on FOC Strategy Using Luenberger Observer, Sine Cosine Algorithm, and RL-TD3 Agent

Considering the need to increase the reliability of Permanent Magnet Synchronous Motor (PMSM) control systems by eliminating speed sensors and using observers to estimate PMSM rotor speed, this paper shows a sensorless control structure-type based on the Field Oriented Control (FOC) type strategy and a Luenberger observer. In order to achieve a better performance of the observer in terms of reducing the error given by the difference between the measured value of the PMSM rotor speed and the Luenberger observer output, a Sine Cosine Algorithm (SCA) is used to optimize the tuning parameters of the Luenberger observer. Also, in order to increase the Luenberger observer performance, a Reinforcement Learning (RL), namely Twin-Delayed Deep Deterministic Policy Gradient (TD3) agent is used, which after the training process supply a signal that is added to the Luenberger observer output, thus obtaining superior features of the estimation for PMSM speed. These methods to improve Luenberger observer performance are validated by numerical simulations realized in Matlab/Simulink.