基于改进经验模态分解的直线电机伺服系统迭代学习控制  

Iterative learning control of linear motor servo system based on modified empirical mode decomposition

作　　者：刘思诺 武志涛[1] LIU Sinuo;WU Zhitao(School of Electronics and Information Engineering,University of Science and Technology Liaoning,Anshan,Liaoning 114051)

机构地区：[1]辽宁科技大学电子与信息工程学院,辽宁鞍山114051

出　　处：《电气技术》2024年第4期32-37,共6页Electrical Engineering

基　　金：国家自然科学基金项目(51677122)。

摘　　要：针对直线电机伺服系统迭代学习过程中因误差积累效应引起的收敛速度慢和跟踪精度差的问题,提出一种基于改进经验模态分解的控制策略。首先,设计一种具有自适应性调节特点的迭代学习位置控制器。然后,提出一种基于三角极值波延拓与互补集合经验模态分解的改进算法,该算法可将各次迭代的跟踪误差进行分解,筛选并剔除影响误差收敛的分量。通过仿真分析并与传统迭代学习控制进行比较,证明了本文方法具有更快的收敛速度,能够以较少的迭代次数实现直线电机的高精度跟踪控制。In order to address the issue of low convergence speed and poor tracking performance caused by error accumulation effects in iterative learning control of linear motor servo systems,a method based on a modified empirical mode decomposition algorithm is proposed.Firstly,a self-adaptive iterative learning position controller is designed.Subsequently,an improved algorithm based on the extension of triangular extreme wave and complementary set empirical mode decomposition is proposed.This algorithm can decompose the tracking errors of each iteration,screen and eliminate the components that affect error convergence.Through simulation analysis and a comparison with traditional iterative learning control,the paper demonstrates that the proposed method exhibits faster convergence speed and can achieve high-precision tracking control of linear motors with fewer iterations.

关 键 词：永磁直线同步电机(PMLSM) 迭代学习 改进经验模态分解 收敛速度 

分 类 号：TM359.4[电气工程—电机] TP273[自动化与计算机技术—检测技术与自动化装置]