• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2016, Vol. 38 ›› Issue (06): 1294-1298.

• 论文 • 上一篇    

抑制非线性扰动的迭代学习控制系统研究

陆仲达1,2,王丽婧1,徐凤霞1   

  1. (1.齐齐哈尔大学计算机与控制工程学院,黑龙江 齐齐哈尔 161006;
    2.哈尔滨理工大学计算机科学与技术学院,黑龙江 哈尔滨 150001)
  • 收稿日期:2015-01-27 修回日期:2015-08-26 出版日期:2016-06-25 发布日期:2016-06-25
  • 基金资助:

    黑龙江省留学归国科学基金(LC2015024)

Inhibition of nonlinear perturbation of
iterative learning control system    

LU Zhongda1,2,WANG Lijing1,XU Fengxia1   

  1. (1.School of Computer and Control Engineering,Qiqihar University;Qiqihar 161000;
    2.College of Computer Science & Technology,Harbin University of Science & Technology,Harbin 150080,China)
  • Received:2015-01-27 Revised:2015-08-26 Online:2016-06-25 Published:2016-06-25

摘要:

多轴旋转机械体速率系统和位置系统中存在一类非线性扰动,这类非线性扰动具有多周期性的性质,并具有周期不变性。为抑制多周期非线性扰动对系统的影响,在系统满足连续里普希斯条件时,得出位置多周期非线性扰动转化为时间多周期非线性扰动的条件。提出一种迭代学习控制方法,通过系统误差收敛性分析来构造学习算子,利用系统的稳态误差信号构成前馈补偿,得出补偿多轴测旋转机械体周期非线性扰动的条件,并证明了算法稳定性。仿真表明,该方法能有效地补偿系统的多周期性非线性扰动,提高多轴旋转机械体系统的控制精度,具有较高的实用价值。

关键词: 多轴旋转机械体, 多周期, 非线性扰动, 迭代学习控制, 前馈补偿, 稳定性

Abstract:

There is a class of nonlinear perturbation in the rate system and location system of the multiplespindle rotating machinery, which features multiple and invariant cycles. Assuming that the system meets the continuous Lipschitz condition, the condition in which multicycle nonlinear perturbation of position can be converted to multicycle nonlinear perturbation of time can be obtained. And thus we propose an iterative learning control method, which constructs the learning operator and the feedforward compensation by analyzing the systematic error convergence and the steadystate error signals respectively, and the compensation terms for multicycle nonlinear perturbation of the multiplespindle rotating machinery can be obtained. The imitated model proves that this method can compensate for the systematic multicycle nonlinear perturbation and improve the control precision of the multispindle rotating machinery system.

Key words: multiple spindle rotating mechanical;multi-cycle;nonlinear perturbation;iterative learning control;feedforward compensation;stability