基于强化学习的刚性联接双电机系统无模型最优协调控制OA北大核心CSTPCD

Dual permanent magnet synchronous motor(PMSM)systems have the characteristics of uncertain model parameters,external load changes and coexistence of fast and slow dynamics,which brings challenges to their optimal coordinated control.This paper proposes a model-free optimal coordinated control method based on reinforcement learning(RL).First,the mathematical model of the dual-PMSM system under traditional master-slave control and PI controllers is formulated.Next,by output regulation and optimal control theories,an optimal coordinated controller is designed to solve the problem of external load change of the system.Then,a RL algorithm independent of model parameters is proposed for uncertainty of model parameters and the coexistence of fast and slow dynamic to learn the controller gain.The proposed control method can improve the tracking performance and synchroni-zation performance of the dual-PMSM system,suppress the interference of unknown time-varying loads,and avoid the influence of parameter uncertainty.Finally,the simulation and experimental results verify that the proposed control strategy can effectively improve the speed tracking performance and torque synchronization performance of the dual-PMSM system.