带神经网络转矩观测器的PMSM自适应前馈PID控制器设计

Design for adaptive feedforward PID controller of PMSM with neural network torque observer

推导了永磁同步电机(PMSM)的ARMA模型,由递归神经网络构成综合转矩观测器,从而把综合负载转矩视为可测干扰;结合极点配置自校正控制、PID控制和前馈控制思想,设计了PMSM的自适应前馈PID控制器。有效地解决了PMSM系统中参数变化和负载扰动等不确定性问题。仿真实验结果表明,该方法具有较强的鲁棒性。

Based on deriving for ARMA model of PMSM, a neural-network-based synthetical load torque observer is designed, and the synthetical load torque is considered as measurable disturbance. Combined the idea of a pole placement self-tuning control, PID control and feedforward control, an adaptive feedforward PID controller for permanent magnet synchronous motor (PMSM) is designed, such that the uncertainties of parameter variations and load disturbances for PMSM drive system are overcome. Experiment simulation results show that the proposed control scheme has stronger robustness.