Parameter Iterative Optimization for Repetitive Motion Control of Permanent Magnet Synchronous Motor

This paper studies the iterative optimization of controller parameters in repetitive motion control of permanent magnet synchronous motors. In order to achieve fast and accurate motion tracking, a position-velocity double closed-loop control structure with added feedforward is used to ensure that it has the structural potential to achieve accurate tracking. Since the model characteristics can never be completely accurate, the performance of rule-based parameters often falls short of expectations, while further manual optimization is complicated with relatively large number of parameters. Since iterative learning is very suitable for the parameter optimization problem of the repetitive trajectories, an improved PSO algorithm with boundary strategies is proposed to iteratively optimize the controller parameters. The stability and convergence speed of the PSO algorithm are enhanced through inertia weight reduction, early termination conditions, and constraining particle positions within tuning boundaries. Simulation results show that the motion tracking performance of the optimized controller is greatly improved compared with the manually tuned parameters. This paper provides an effective method for automatic iterative optimization of controller parameters for high precision and fast repetitive motion control of permanent magnet synchronous motors.