执行依赖启发式动态规划PMSM控制策略研究

永磁同步电机算法中的PID参数固定不变，自适应能力较差。近似动态规划具有自学习能力，能动态调整参数，增强系统的自适应能力。本文将近似动态规划中的代表性算法——执行依赖启发式动态规划与类PI网络相结合，根据环境反馈对各个策略进行评价，选取评价最优的策略在线调整动作网络参数，对永磁同步电机进行自适应最优控制。仿真结果表明，执行依赖启发式动态规划加PI控制算法在系统速度变化和负载变化过程中，动态效果和稳态精度优于传统PID控制。

Control strategy research of Action-Dependent Heuristic Dynamic Programming in PMSM

The PID parameters are fixed in permanent magnet synchronous motors control algorithm and the adaptive ability is poor. Approximate dynamic programming with self-learning ability, can dynamically adjust the parameters and enhance the adaptive capacity of the system. In this paper, the representative algorithm of approximate dynamic programming - action-dependent heuristic dynamic programming combined with PI-Like network ,optimal control permanent magnet synchronous motors adaptively. The simulation results show that the algorithm of action-dependent heuristic dynamic programming combining with PI-Like network in the process of any speed and load changes, the dynamic effect and steady state accuracy of better than traditional PID control.