引用本文: | 许德智,黄泊珉,杨玮林.神经网络自适应的永磁直线同步电机超扭曲终端滑模控制[J].电力系统保护与控制,2021,49(13):64-71.[点击复制] |
XU Dezhi,HUANG Bomin,YANG Weilin.Neural network adaptive super twist terminal sliding mode control for a permanent magnet linear synchronous motor[J].Power System Protection and Control,2021,49(13):64-71[点击复制] |
|
摘要: |
为了提高永磁直线同步电机控制系统的鲁棒性和快速性,提出了一种基于超扭曲滑模控制的直线电机反推控制策略。首先,根据电机的机械动力学方程,建立了永磁直线同步电机的数学模型。其次,引入一种基于超扭曲控制的终端滑模控制器,削减系统的抖振,保证滑模面的快速收敛,从而提高系统的鲁棒性。针对直线电机易受到参数变化以及外界干扰影响的特点,为直线电机控制系统设计了一个神经网络干扰观测器。最后,通过直线电机实验平台验证控制方法的有效性。 |
关键词: 永磁直线同步电机 终端滑模控制 径向基函数神经网络 反推控制 |
DOI:DOI: 10.19783/j.cnki.pspc.201154 |
投稿时间:2020-09-19修订日期:2020-11-22 |
基金项目:国家自然科学基金项目资助(61973140, 61903158);中央高校基本科研业务费专项资金资助(JUSRP41911,JUSRP22030) |
|
Neural network adaptive super twist terminal sliding mode control for a permanent magnet linear synchronous motor |
XU Dezhi,HUANG Bomin,YANG Weilin |
(College of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China) |
Abstract: |
In order to improve the robustness and speed of a permanent magnet linear synchronous motor control system, a backstepping control strategy based on super twisting sliding mode control is proposed. First, given the mechanical dynamic equation of the motor, the mathematical model is established. Secondly, a terminal sliding surface based on a super twisting controller is introduced. This can reduce the chattering of the system and ensure the rapid convergence of the sliding surface, thus improving the robustness of the system. The linear motor control system is easy to be affected by parameter changes and external disturbances, thus a neural network disturbance observer is designed for it. Finally, it is verified by a motor experimental platform and the results show the effectiveness of the control method.
This work is supported by the National Natural Science Foundation of China (No. 61973140 and No. 61903158) and the Fundamental Research Funds for the Central Universities (No. JUSRP41911 and No. JUSRP22030). |
Key words: permanent magnet linear synchronous motor terminal sliding mode control radial basis function neural network backstepping control |