基于在线辨识补偿的永磁直线同步电机模型参考自适应神经网络速度控制

本文针对永磁直线同步电机（ＰＭＬＳＭ）的速度伺服控制，提出了一种模型参考自适应与神经网络技术相结合的控制策略。为提高系统的鲁棒性，采用了在线辨识技术，对参数变化实时补偿，及时修正神经网络教师值的计算，仿真实验表明，该系统具有良好的速度伺服性能，特别是在克服永磁直线同步电机所特有的端部效应方面效果显著。

Model Reference Adaptive Neural Networks Speed Control for Permanent Magnet Line Synchronous Motor Based on On-line Identification and Compensation

In this paper,we propose a control scheme for thespeedservocontrolofPMLSMwithmodelrefe rence adaptive and neural network techniques.In order to improve the robustness of the system,we use on line indentification technique to compensate the variation of the parameters and modify the calculation of the neural networks's teacher value.The simulation shows that the system has good speed servo performance,especially in overcoming the effect of the effect which PMLSM has.