引用本文: | 魏伟,夏鹏飞,左敏.压电驱动纳米定位台的线性自抗扰控制[J].控制理论与应用,2018,35(11):1577~1590.[点击复制] |
Wei Wei,XIA Peng-fei,ZUO Min.Linear active disturbance rejection control of piezoelectric nanopositioning stage[J].Control Theory and Technology,2018,35(11):1577~1590.[点击复制] |
|
压电驱动纳米定位台的线性自抗扰控制 |
Linear active disturbance rejection control of piezoelectric nanopositioning stage |
摘要点击 2695 全文点击 1106 投稿时间:2018-05-12 修订日期:2018-09-20 |
查看全文 查看/发表评论 下载PDF阅读器 |
DOI编号 10.7641/CTA.2018.80348 |
2018,35(11):1577-1590 |
中文关键词 压电驱动器 纳米定位 磁滞 自抗扰 |
英文关键词 piezoelectric actuator nanopositioning hysteresis active disturbance rejection control |
基金项目 国家自然科学基金项目(61403006, 61873005), 北京市自然科学基金暨北京市教委科技计划重点项目(KZ201810011012), 北京市属高等学校青年 拔尖人才培育计划项目(CIT& TCD201704044)资助. |
|
中文摘要 |
压电驱动的纳米定位系统, 磁滞严重降低其定位精度. 基于磁滞逆模型的定位控制方法, 可提高定位精度.
然而, 因其对模型依赖大, 无法在内、外扰动因素存在时, 仍保证良好的定位效果. 为此, 视磁滞为干扰, 设计不基于
磁滞及定位系统精确数学模型的自抗扰控制, 利用扩张状态观测器主动估计磁滞, 并实时补偿之, 以保证定位精度.
本文分析了扩张状态观测器的收敛性、闭环系统的稳定性; 讨论了自抗扰控制参数与纳米定位系统动态响应间的
关系; 对比了PI控制和自抗扰控制的动态响应、时间与误差绝对值乘积积分、平均绝对误差和均方根误差, 数值和
实验结果均表明自抗扰控制优于PI控制, 证实了自抗扰控制可主动估计磁滞、内部不确定性和外部扰动, 在其降低
定位精度之前即被补偿、保证良好的定位控制效果. |
英文摘要 |
In piezoelectric driven nanopositioning systems, hysteresis severely degrades its positioning accuracy. A
hysteresis inverse model based positioning approach is able to improve the accuracy. However, due to its large dependence
on model, it cannot guarantee satisfied performance when internal and external disturbances exist. To address such problem,
hysteresis is taken to be a kind of disturbance, a model independent control, i.e. active disturbance rejection control
(ADRC), is designed. An extended state observer (ESO) is used to estimate hysteresis actively and compensate it by
control signal in real time to ensure the positioning accuracy. The convergence of ESO and the stability of the closedloop
system are analyzed. Relationship between control parameters of ADRC and the response of the system has been
discussed. System response, integral of time-multiplied absolute-value of error (ITAE), mean absolute error (MAE) and
root mean square error (RMSE) of PI control and ADRC have been compared. Both numerical and experimental results
show that ADRC is superior to PI control. It confirms that ADRC can actively estimate hysteresis, internal uncertainties, and
external disturbances. Those undesired issues can be cancelled before they reducing the positioning accuracy. Therefore,
satisfied performance can be guaranteed. |
|
|
|
|
|