ZOU Shourui,WU Yinan,FANG Yongchun.Tracking control of piezoelectric actuator based on feedforward compensation of recurrent neural network[J].CAAI Transactions on Intelligent Systems,2021,16(3):567-574.[doi:10.11992/tis.202104010]





Tracking control of piezoelectric actuator based on feedforward compensation of recurrent neural network
邹守睿 武毅男 方勇纯
南开大学 人工智能学院,天津 300350
ZOU Shourui WU Yinan FANG Yongchun
College of Artificial Intelligence, Nankai University, Tianjing 300350, China
piezoelectric actuatorhysteresisnonlinearityrecurrent neural networkinverse modelfeedforward controlneuronadaptive control
However, PEAs’ inherent hysteresis, combined with other dynamic properties, negatively influences their tracking performance. Because recurrent neural networks can accurately fit nonlinear systems with memory storage capabilities, a recurrent neural network is designed to model the hysteresis of PEAs. Then, an accurate inverse model of the relationship between the output displacement and the input voltage is obtained, through which feedforward compensation is performed on PEAs. Furthermore, because modeling errors and other disturbances affect PEA tracking accuracy, a single neuron adaptive proportional-integral-derivative controller is designed to accurately track the desired signal by tracking the PEAs. Finally, experimental results verify the proposed model’s accuracy and the tracking performance of the designed controller.


[1] 张鑫, 孙小飞, 周文松, 等. 基于压电阻抗和主成分分析的斜拉索覆冰监测[J]. 哈尔滨工程大学学报, 2020, 41(12):1765-1771
ZHANG Xin, SUN Xiaofei, ZHOU Wensong, et al. Monitoring of stay-cable icing based on electro-mechanical impedance and principal component analysis[J]. Journal of Harbin Engineering University, 2020, 41(12):1765-1771
[2] 许延峰, 周天放, 蓝宇. 低频弯曲式水听器研究[J]. 应用科技, 2020, 47(1):99-104
XU Yanfeng, ZHOU Tianfang, LAN Yu. Research on the low frequency curved hydrophone[J]. Applied science and technology, 2020, 47(1):99-104
[3] WU Yinan, FAN Zhi, FANG Yongchun, et al. An effective correction method for AFM image distortion due to hysteresis and thermal drift[J]. IEEE transactions on instrumentation and measurement, 2021, 70:5004212.
[4] 黄心汉. 微装配机器人:关键技术、发展与应用[J]. 智能系统学报, 2020, 15(3):413-424
HUANG Xinhan. Microassembly robot:key technology, development, and applications[J]. CAAI transactions on intelligent systems, 2020, 15(3):413-424
[5] 莫喜平, 刘永平. 纵向换能器的频率方程[J]. 哈尔滨工程大学学报, 2019, 40(7):1245-1250
MO Xiping, LIU Yongping. Frequency equations for a longitudinal transducer[J]. Journal of Harbin Engineering University, 2019, 40(7):1245-1250
[6] GU Guoying, ZHU Limin, SU Chunyi, et al. Modeling of piezoelectric-actuated nanopositioning stages involving with the hysteresis[M]//RU Changhai, LIU Xinyu, SUN Yu. Nanopositioning Technologies:Fundamentals and Applications. Cham:Springer, 2016:183-212.
[7] ANG W T, GARMON F A, KHOSLA P K, et al. Modeling rate-dependent hysteresis in piezoelectric actuators[C]//Proceedings 2003 IEEE/RSJ International Conference on Intelligent Robots and Systems. Las Vegas, USA, 2003:1975-1980.
[8] STEPANENKO Y, SU Chunyi. Intelligent control of piezoelectric actuators[C]//Proceedings of the 37th IEEE Conference on Decision and Control. Tampa, USA, 1998:4234-4239.
[9] XIAO Shunli, LI Yangmin. Modeling and high dynamic compensating the rate-dependent hysteresis of piezoelectric actuators via a novel modified inverse preisach model[J]. IEEE transactions on control systems technology, 2013, 21(5):1549-1557.
[10] CHEN Hui, TAN Yonghong, ZHOU Xingpeng, et al. Identification of dynamic hysteresis based on duhem model[C]//2011 Fourth International Conference on Intelligent Computation Technology and Automation. Shenzhen, China, 2011:810-814.
[11] LIU Yanfang, SHAN Jinjun, GABBERT U, et al. Hysteresis and creep modeling and compensation for a piezoelectric actuator using a fractional-order Maxwell resistive capacitor approach[J]. Smart materials and structures, 2013, 22(11):115020.
[12] CHENG Long, LIU Weichuan, HOU Zengguang, et al. An adaptive Takagi-Sugeno fuzzy model-based predictive controller for piezoelectric actuators[J]. IEEE transactions on industrial electronics, 2017, 64(4):3048-3058.
[13] LING Jie, FENG Zhao, ZHENG Dongdong, et al. Robust adaptive motion tracking of piezoelectric actuated stages using online neural-network-based sliding mode control[J]. Mechanical systems and signal processing, 2021, 150:107235.
[14] ZHAO Xinlong, SHEN Shuai, SU Liangcai, et al. Elman neural network-based identification of rate-dependent hysteresis in piezoelectric actuators[J]. Journal of intelligent material systems and structures, 2020, 31(7):980-989.
[15] SERVAN-SCHREIBER D, CLEEREMANS A, MCCLELLAND J L. Learning sequential structure in simple recurrent networks[M]//TOURETZKY D S. Advances in Neural Information Processing Systems 1. San Francisco:Morgan Kaufmann Publishers Inc., 1989:643-652.
[16] XIE Shengwen, REN Juan. Predictive control of nano-positioning stage using recurrent-neural-network-based inversion model[C]//2019 IEEE 58th Conference on Decision and Control (CDC). Nice, France, 2019:7764-7769.
[17] WU Yinan, FANG Yongchun, LIU Cunhuan, et al. Gated recurrent unit based frequency-dependent hysteresis modeling and end-to-end compensation[J]. Mechanical systems and signal processing, 2020, 136:106501.
[18] CHOUZA A, BARAMBONES O, CALVO I, et al. Sliding mode-based robust control for piezoelectric actuators with inverse dynamics estimation[J]. Energies, 2019, 12(5):943.
[19] SOLEYMANZADEH D, GHAFARIRAD H, ZAREINEJAD M. Sensorless adaptive sliding mode position control for piezoelectric actuators with charge leakage[J]. Journal of intelligent material systems and structures, 2020, 31(1):40-52.
[20] CHENG Long, LIU Weichuan, HOU Zengguang, et al. Neural-network-based nonlinear model predictive control for piezoelectric actuators[J]. IEEE transactions on industrial electronics, 2015, 62(12):7717-7727.
[21] LIN F J, SHIEH H J, HUANG P K, et al. Adaptive control with hysteresis estimation and compensation using RFNN for piezo-actuator[J]. IEEE transactions on ultrasonics, ferroelectrics, and frequency control, 2006, 53(9):1649-1661.
[22] SCH?FER A M, ZIMMERMANN H G. Recurrent neural networks are universal approximators[C]//Proceedings of the 16th International Conference on Artificial Neural Networks. Athens, Greece, 2006:632-640.
[23] MORé J J. The Levenberg-Marquardt algorithm:implementation and theory[M]//WATSON G A. Numerical Analysis. Berlin, Heidelberg:Springer, 1997:105-116.
[24] XU Baochang, WU Jianzhang, CHEN Yongkun. An improved single neuron adaptive PID control algorithm[C]//2009 Fifth International Conference on Natural Computation. Tianjin, China, 2009:558-562.
[25] MAR?íK J, STREJC V. Application of identification-free algorithms for adaptive control[J]. Automatica, 1989, 25(2):273-277.


更新日期/Last Update: 2021-06-25