Speed control of sensorless PMSM drive based on EKF optimized by variable scale chaotic particle swarm optimization

To investigate the parameter characteristics of permanent magnet synchronous motor (PMSM) speed sensorless vector control system and capture the noise matrices quickly and accurately in the speed estimation process of the extended Kalman filter for PMSM, The recursive least square method with forgetting factor is proposed to determine the actual parameters of the system, and then a new variable-scale chaotic particle swarm optimization (VCPSO) algorithm is put forward to accurately obtain the system noise matrix and the measurement noise matrix. The simulation results show that noise matrix optimization of extended Kalman filter by employing VCPSO algorithm under actual motor parameters is better than those employing standard PSO or chaotic PSO algorithms with faster speed and higher accuracy.