基于神经网络的超磁致伸缩传感执行器磁滞模型

中国机械工程 ›› 2011, Vol. 22 ›› Issue (5): 571-575.

基于神经网络的超磁致伸缩传感执行器磁滞模型

刘慧芳;贾振元;王福吉

大连理工大学精密与特种加工教育部重点实验室,大连,116024

出版日期:2011-03-10 发布日期:2011-03-24
基金资助:
国家自然科学基金资助项目(50775021)
National Natural Science Foundation of China（No. 50775021）

Study on Hysteresis Model of Giant Magnetostrictive Sensing Actuator Based on Neural Network

Liu Huifang;Jia Zhenyuan;Wang Fuji

Key Laboratory for Precision and Non-traditional Machining Technology of Ministry of Education, Dalian University of Technology, Dalian, Liaoning,116024

Online:2011-03-10 Published:2011-03-24
Supported by:
National Natural Science Foundation of China（No. 50775021）

摘要/Abstract

摘要：

依据超磁致伸缩材料Joule效应和Villari效应之间的耦合关系,提出一种超磁致伸缩传感执行器,该执行器利用Villari效应和Faraday效应产生的感应电动势驱动超磁致伸缩材料发生Joule效应而产生应变,给出了该传感执行器的结构和工作原理。为了解决材料的磁滞对超磁致伸缩传感执行器输出特性的影响,测量了在不同预紧力和最大工作电流作用下的磁滞回线,采用BP神经网络建立了磁化滞回模型。计算结果表明该模型能很好地描述在任意预紧力和最大工作电流等工作条件下的磁滞特性。

关键词:

超磁致伸缩, 传感执行器, 磁滞, 神经网络

Abstract:

Based on the coupling relationship between Joule effect and Villari effect of the giant magnetostrictive materials, it proposed a giant magnetostrictive sensing actuator which used induced electromotive force generated by Villari effect and Faraday effect to drive giant magnetostrictive materials occurring Joule effect and generating strain. Its structure and working principle were presented herein. Meanwhile, in order to solve the hysteresis characteristics of the material in giant magnetostrictive sensing actuator, hysteresis loop under different pre-tightening force and maximum working current were measured. Moreover, it put forward establishing the magnetization hysteresis model of the giant magnetostrictive sensing actuator by BP neural networks. The calculation results show that the model can describe hysteresis characteristics under arbitrary pre-tightening force and maximum working current well. It provides evidence for hysteresis compensation of giant magnetostrictive sensing actuator.

Key words: giant magnetostrictive, sensing actuator, hysteresis, neural network

中图分类号:

TP183
TB381

刘慧芳, 贾振元, 王福吉.

基于神经网络的超磁致伸缩传感执行器磁滞模型

[J]. 中国机械工程, 2011, 22(5): 571-575.

LIU Hui-Fang, GU Zhen-Yuan, WANG Fu-Ji.

Study on Hysteresis Model of Giant Magnetostrictive Sensing Actuator Based on Neural Network

[J]. China Mechanical Engineering, 2011, 22(5): 571-575.

参考文献

[1]李扩社[1],徐静[1],杨红川[1],袁永强[1],徐建林[1],于敦波[1],张深根[1].稀土超磁致伸缩材料发展概况[J].稀土,2004,25(4):51-56.
[2]郭东明,杨兴,贾振元,武丹.超磁致伸缩执行器在机电工程中的应用研究现状[J].中国机械工程,2001,12(6):724-727.
[3]卢全国[1,2],陈定方[1],钟毓宁[3],陈敏[1].超磁致伸缩致动器热变形影响及温控研究[J].中国机械工程,2007,18(1):16-19.
[4]Zhao Xinlong,Tan Yonghong. Modeling Hysteresis and Its Inverse Model Using Neural Networks Based on Expanded Input Space Method[J]. IEEE Transactions on Control Systems Technology, 2008, 16 (3) :484-490.
[5]黄平林[1,2],胡虔生[2].基于人工神经网络的Preisach磁滞模型与实现[J].电工电能新技术,2009,28(1):43-45.
[6]Dimitre M, Luc D, Marc D W, et al. Modeling of Quasistatic Magnetic Hysteresis with Feed-Forward Neural Networks[J]. Journal of Applied Physics, 2001,89 ( 11 ) : 6737-6739.
[7]郑军红[1],叶修梓[1],陈志杨[2].基于神经网络和遗传算法的智能夹具规划[J].中国机械工程,2008,19(19):2376-2381.
[8]Sixdenier F,Scorretti R,Marion R,et al. Quasistatic Hysteresis Modeling with Feed-Forward Neural Networks: Influence of the Last But One Extreme Values[J]. Journal of Magnetism and Magnetic Materials, 2008,320 (20) : 992-996.
[9]Kucuk I. Prediction of Hysteresis Loop in Magnetic Cores Using Neural Network and Genetic Algorithm [J]. Journal of Magnetism and Magnetic Materials, 2006,305(2) :423-427.
[10]Cincotti S, Marchesi M, Serri A. A Neural Network Model of Parametric Non-linear Hysteretic Inductors [ J]. IEEE Transactions on Magnetics, 1998,34(5) :3040-3043.
[11]Serpico C, Visone C. Magnetic Hysteresis Modeling via Feed-Forward Neural Networks[J]. IEEE Transactions on Magnetics, 1998,34 (3) : 623-628.
[12]贾振元[1],马建伟[1],刘巍[1],王福吉[1].多几何要素影响下液压阀件特性的混合神经网络预测模型[J].机械工程学报,2010(2):126-131.
[13]Wilamowski B M, Kaynak O, Iplikci S, et al. An Algorithm for Fast Convergence in Training Neu ral Networks [C]//Proceedings of International Joint Conference on Neural Networks. Washington,2001 : 1778-1782.