电工技术学报  2023, Vol. 38 Issue (6): 1519-1530    DOI: 10.19595/j.cnki.1000-6753.tces.221383
“高转矩性能电机及其系统”专题(特约主编:赵文祥 教授) |
基于自学习非线性PID的音圈电机精密定位系统
程苗苗1, 翟朋辉1, 张英杰2, 李健2, 冯凯2
1.湖南大学电气与信息工程学院 长沙 410082;
2.湖南大学机械与运载工程学院 长沙 410082
A Voice Coil Motor-Driven Precision Positioning System Based on Self-Learning Nonlinear PID
Cheng Miaomiao1, Zhai Penghui1, Zhang Yingjie2, Li Jian2, Feng Kai2
1. School of Electrical and Information Engineering Hunan University Changsha 410082 China;
2. School of Mechanical and Vehicle Engineering Hunan University Changsha 410082 China
全文: PDF (2831 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 基于音圈电机的精密宏微气浮运动系统,是一种能克服接触摩擦和行程限制的新型精密定位系统。针对系统中用于精密定位的音圈电机受到内外扰动从而影响系统最终定位精度的问题,在建立起音圈电机数学模型的基础上,设计了基于反正弦函数的自学习非线性PID控制器,利用自学习算法对非线性增益函数的增益系数进行实时调整。完成算法设计与仿真后,在搭建的系统平台分别进行了微动台短行程定位和宏微动台的长行程定位实验。仿真和实验结果表明,与传统PID控制器相比较,自学习非线性PID控制器的使用有效提高了系统的鲁棒性和定位精度,系统对位置指令响应迅速无超调,控制精度达到了亚微米级。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
程苗苗
翟朋辉
张英杰
李健
冯凯
关键词 精密宏微气浮运动系统音圈电机自学习非线性PID定位精度    
Abstract:As a typical precise-positioning and long-stroke motion system, the precision macro-micro air-floating motion table has been taking more and more research interest due to its free-of-contact friction characteristics. Voice coil motor (VCM) has been proposed as the main driving element for the micro motion table due to its fast response and non-contact feed drives. However, due to the mover suspending to the stator, VCM is susceptible to interference such as load disturbance, vibration, and noise. Some solutions have been proposed in the existing research to eliminate the disturbances and improve positioning accuracy. However, either limited effect or complex implementation problems exist.
Therefore, this paper proposes a novel self-learning nonlinear PID control method to improve the positioning accuracy and robustness of the VCM precision positioning system. The main idea of the proposed control method is to construct a nonlinear PID control law based on the arcsine function, and apply the neural network algorithm to adaptively adjust the weight parameters of the proposed nonlinear function with the time-varying system errors. Accordingly, the pre-designed PID control law assures the PID control parameters change reasonably by following the proposed arcsine function. The BP neural network can provide control flexibility, thus improving the model robustness and control accuracy. Therefore, the proposed self-learning nonlinear PID control is the potential for nonlinear, multivariable, and interference-susceptible systems, such as the precision macro-micro air-floating motion system.
The unit step response experiment is first performed on the micromotion table. The experimental results show that the overshoot of the traditional PID controller is about 8 %, and the steady-state positioning accuracy is 1 μm. The steady-state positioning accuracy of the self-learning nonlinear PID controller is 0.5 μm, and there is no overshoot. The proposed method provides better positioning accuracy and transient response. Some experiments are further carried out with varying load conditions. According to the experimental results, the traditional PID controller presents a greatly increased overshoot and settling time, while the proposed self-learning nonlinear PID presents the same control performance as the no-load case. The proposed method improves the robustness of the VCM-driven micromotion table.
Finally, the long-stroke positioning experiment is performed on the macro-micro motion stable. The micro-motion stable adopts the proposed self-learning nonlinear PID control, while the macro-motion stable adopts the traditional linear PID control. The results show that the macro-motion stable follows the micro motion stable with a fast response. Besides, the positioning accuracy is within 0.5 μm, consistent with the short-stroke positioning experimental results. That proves the decoupled control characteristics of the macro/micro-motion stable. The positioning accuracy of the system is determined by the control performance of the micro-motion stable.
Concluded above, a novel self-learning nonlinear PID controller is proposed for the VCM-driven micro motion stable. It provides with better positioning accuracy, fast transient response and enhanced robustness according to the experimental results. Both short-stroke and long-stroke positioning experiments are carried out. The results verify that the micro motion table is mechanically decoupled from the macro motion table. On the basis of this, the proposed self-learning nonlinear PID and the traditional PID are proposed to be used for the micro/macro motion table respectively. Finally, a reduced control complexity and improved control performance could be achieved for the precision macro-micro air-floating motion table.
Key wordsPrecision macro-micro air-floating motion system    voice coil motor    neural network nonlinear PID    positioning accuracy   
收稿日期: 2022-07-19     
PACS: TM359.4  
基金资助:国家重点研发计划资助项目(2020YFB2007604)
通讯作者: 冯 凯 男,1982年生,教授,博士生导师,研究方向为精密制造、气浮平台以及控制系统等。E-mail: kfeng@hnu.edu.cn   
作者简介: 程苗苗 女,1982年生,副教授,硕士生导师,研究方向为电机模型建立以及非线性控制算法等。E-mail: mmcheng_hnu@126.com
引用本文:   
程苗苗, 翟朋辉, 张英杰, 李健, 冯凯. 基于自学习非线性PID的音圈电机精密定位系统[J]. 电工技术学报, 2023, 38(6): 1519-1530. Cheng Miaomiao, Zhai Penghui, Zhang Yingjie, Li Jian, Feng Kai. A Voice Coil Motor-Driven Precision Positioning System Based on Self-Learning Nonlinear PID. Transactions of China Electrotechnical Society, 2023, 38(6): 1519-1530.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.221383          https://dgjsxb.ces-transaction.com/CN/Y2023/V38/I6/1519