[1]黄鹏升,丁 烨.基于高斯过程的压电驱动系统磁滞特性建模[J].机械与电子,2023,41(12):3-9.
　HUANG Pengsheng,DING Ye.Modeling of Hysteresis in Piezoelectric-actuated Systems Based on Gaussian Process[J].Machinery & Electronics,2023,41(12):3-9.
点击复制

基于高斯过程的压电驱动系统磁滞特性建模()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
41
期数:
2023年12期
页码:
3-9
栏目:
研究与设计
出版日期:
2023-12-30

文章信息/Info

Title:
Modeling of Hysteresis in Piezoelectric-actuated Systems Based on Gaussian Process
文章编号:
1001-2257 ( 2023 ) 12-0003-07
作者:
黄鹏升; 丁 烨
上海交通大学机械与动力工程学院,上海 200240
Author(s):
HUANG Pengsheng ;  DING Ye
( School of Mechanical Engineering , Shanghai Jiao Tong University , Shanghai 200240 , China )
关键词:
压电驱动器; 磁滞非线性; 系统建模; 高斯过程; 有源自回归模型
Keywords:
piezoelectric actuator ;  hysteresis nonlinearity ;  system modeling ;  Gaussian process ;  auto-regressive with extra inputs
分类号:
TP301
文献标志码:
A
摘要:
针对压电陶瓷驱动系统输入与输出间的磁滞非线性关系,现有建模方法存在无法兼顾建模过程复杂度与模型通用性的问题。为此,提出了一种新的系统输入 输出关系建模方法。首先,基于压电驱动器磁滞特性的率无关特点,建立了将磁滞特性与线性特性解耦的 Hammerstein 形式系统描述方法,以简化建模过程。在此基础上,基于高斯过程方法建立输入电压与输出位移及其一阶变化率间的统计描述,并导出磁滞特性模型;同时基于有源自回归模型对线性特性进行建模,并给出了相关的参数辨识流程。完整模型具有较高的通用性,可用于不同输入条件下。通过实际系统输出预测实验可以证明,所提方法可有效表征系统特性,输出预测误差控制在 2% 左右。与同类方法的对比实验则证明了所提方法在参数辨识效率与系统建模精度上均具有一定优势。
Abstract:
In current research on modeling of the hysteresis nonlinear relationship between output displacement and control voltage of a piezoelectric-actuated system , the complexity of modeling process and generality of model can not be guaranteed at the same time.To solve the problem , a new system modeling method is proposed.Firstly , based on the rate-independent characteristics of the hysteresis in piezoelectric actuators , a Hammerstein model which decouples the hysteresis and the linear dynamics is established to simplify the modeling process.On this basis , a statistical description about the control voltage , the output displacement and their first-order derivative is built with a Gaussian process ( GP ) .And the hysteresis characteristic model is derived accordingly.Meanwhile , the linear dynamics of the system is modeled with the auto-regressive with extra inputs ( ARX ) method and the parameter identification process is given. The complete model are universal so that it is applicable to input with different features.The system output prediction experiment shows that the proposed method can effectively characterize the system properties while the output prediction error is lower than about 2%.The comparison with existing method demonstrates the advantages of the proposed method in parameter identification efficiency and system modeling accuracy.

参考文献/References:

[ 1 ] 李春霞 . 压电驱动二维纳米定位平台的高速栅格扫描运动控制方法研究[ D ] . 上海:上海交通大学,2017.

[ 2 ] 杨梅菊 . 压电驱动纳米定位平台的高带宽控制理论和方法研究[ D ] . 上海:上海交通大学,2015.
[ 3 ] 黄维维 . 多功能快速刀具伺服系统的设计、控制及在光学微结构加工中的应用 [ D ] . 上海:上海交通大学,2022.
[ 4 ] ZHAO D P , ZHU Z H , HUANG P , et al.Development of a piezoelectrically actuated dual-stage fast tool servo [ J ] .Mechanical systems and signal processing , 2020 , 144 ( C ): 106873.
[ 5 ] 杜莹雪 . 基于 Play 模型的软磁材料磁滞特性模拟与实验研究[ D ] . 北京:华北电力大学(北京),2022.
[ 6 ] XIE S L , MEI J P , LIU H T , et al.Hysteresis modeling and trajectory tracking control of the pneumatic muscle actuator using modified prandtl – ishlinskii model [ J ] . Mechanism and machine theory , 2018 , 120 : 213-224.
[ 7 ] 黄雷,包广清,陈俊全 . 基于 Levenberg Marquardt 算法的改进 Preisach 模型磁特性模拟与验证[ J ] . 中国电机工程学报,2020 , 40 ( 18 ): 6006-6015.
[ 8 ] CHEN P , BAI X X , QIAN L J , et al.An approach for hysteresis modeling based on shape function and memory mechanism [ J ] .IEEE / ASME Transactions on mechatronics , 2018 , 23 ( 3 ): 1270-1278.
[ 9 ] WEN Z J , DING Y , LIU P K , et al.An efficient identification method for dynamic systems with coupled hysteresis and linear dynamics : application to piezoelectric actuated nanopositioning stages [ J ] .IEEE / ASME Transactions on mechatronics , 2019 , 24 ( 1 ): 326-337.
[ 10 ] 莫仕勋,杨皓 . 一种考虑 J A 磁滞的变压器电磁暂态模型[ J ] . 实验室研究与探索, 2023 , 42 ( 3 ): 119-124.
[ 11 ] FARROKH M. Hysteresis simulation using least squares support vector machine [ J ] .Journal of engineering mechanics , 2018 , 144 ( 9 ): 04018084.
[ 12 ] WU Z L , TANG H , HE S F , et al.A regularized on line sequential extreme learning machine with forgetting property for fast dynamic hysteresis modeling[ C ] ∥2017 IEEE / RSJ International Conference on Intelligent Robots and Systems( IROS ), 2017 : 383-388.
[ 13 ] HIROSE N , TAJIMA R.Modeling of rolling friction by recurrent neural network using LSTM [ C ] ∥2017 IEEE International Conference on Robotics and Automation ( ICRA ), 2017 : 6471-6478.
[ 14 ] TAO Y D , LI H X , ZHU L M.Rate-dependent hysteresis modeling and compensation of piezoelectric actuators using gaussian process [ J ] .Sensors and actuators a : physical , 2019 , 295 : 357-365.?
[ 15 ] LUO X , XIAO M B , DING Y , et al.Hysteresis modeling and compensation of a pneumatic end-effector based on gaussian process regression [ J ] .Sensors and actuators a : physical , 2020 , 315 : 112227.
[ 16 ] XU J H , XIAO M B , DING Y.Modeling and compensation of hysteresis for pneumatic artificial muscles based on gaussian mixture models [ J ] .Science china technological sciences , 2019 , 62 ( 7 ): 1094-1102.
[ 17 ] 黄涛,罗治洪,陶桂宝,等 . 压电定位平台 Hammerstein 建模与反馈线性化控制 [ J ] . 光学精密工程,2022 , 30 ( 14 ): 1716-1724.
[ 18 ] 吴与泽,张慧萍,林剑锋,等 . 压电驱动器 Wiener 模型的辨识与复合控 制 [ J ] . 机械设计与研 究,2022 , 38( 4 ): 65-70 , 77.
[ 19 ] RASMUSSEN C E , WILLIAMS C K I.Gaussian processes for machine learning [ M ] .Cambridge : MIT Press ,2005.
[ 20 ] 谷国迎 . 压电陶瓷驱动微位移平台的磁滞补偿控制理论和方法研究[ D ] . 上海:上海交通大学,2012.
[ 21 ] GU G Y , ZHU L M , SU C Y.Modeling and compensation of asymmetric hysteresis nonlinearity for piezoceramic actuators with a modified prandtl – ishlinskii model [ J ] .IEEE Transactions on industrial electronics , 2014 , 61 ( 3 ): 1583-1595.
[ 22 ] GU G Y , LI C X , ZHU L M , et al.Modeling and identification of piezoelectric actuated stages cascading hysteresis nonlinearity with linear dynamics [ J ] . IEEE / ASME Transactions on mechatronics , 2016 , 21( 3 ): 1792-1797.
[ 23 ] TAO Y D , LI H X , ZHU L M.Hysteresis modeling with frequency-separation-based gaussian process and its application to sinusoidal scanning for fast imaging of atomic force microscope [ J ] .Sensors and actuators a : physical , 2020 , 311 : 112070.

备注/Memo

备注/Memo:
收稿日期: 2023-06-27
基金项目:国家自然科学基金资助项目( 52275501 )
作者简介:黄鹏升 ( 1998- ),男,四川广元人,硕士研究生,研究方向为精密运动系统控制技术;丁 烨 ( 1982- ),男,江苏丹阳人,博士,教授,博士研究生导师,研究方向为机器人学与智能加工技术。
更新日期/Last Update: 2024-01-10