The imperfect sinusoidal flux distribution, cogging torque, and current measurement errors can cause periodic torque ripple in the permanent magnet synchronous motor (PMSM). These ripples are reflected in the periodic oscillation of the motor speed and torque, causing vibration at low speeds and noise at high speeds. As a high-precision tracking application, ripple degrades the application performance of PMSM. In this paper, an adaptive dynamic programming (ADP) scheme is proposed to reduce the periodic torque ripples. An optimal controller is designed by iterative control algorithm using robust adaptive dynamic programming theory and strategic iterative technique. ADP is combined with the existing Proportional-Integral (PI) current controller and generates compensated reference current iteratively from cycle to cycle so as to minimize the mean square torque error. As a result, an optimization problem is constructed and an optimal controller is obtained. The simulation results show that the robust adaptive dynamic programming achieves lower torque ripple and shorter dynamic adjustment time during steady-state operation, thus meeting the requirements of steady speed state and the dynamic performance of the regulation system.

中文翻译：

---

永磁同步电机转矩脉动最小化的自适应动态规划方法

不正确的正弦波通量分布，齿槽转矩和电流测量误差会在永磁同步电动机（PMSM）中引起周期性的转矩波动。这些脉动反映在电动机速度和转矩的周期性振荡中，从而导致低速振动和高速噪声。作为高精度跟踪应用，纹波会降低PMSM的应用性能。本文提出了一种自适应动态规划（ADP）方案，以减少周期性转矩脉动。利用鲁棒自适应动态规划理论和策略迭代技术，通过迭代控制算法设计出最优控制器。ADP与现有的比例积分（PI）电流控制器结合使用，并在每个周期之间迭代生成补偿参考电流，以使均方转矩误差最小。结果，构造了优化问题并获得了最优控制器。仿真结果表明，鲁棒的自适应动态规划在稳态运行时实现了较低的转矩脉动和较短的动态调整时间，从而满足了稳定转速状态和调节系统动态性能的要求。