基于VMD和DBN的非线性结构模型参数识别

摘要
图/表
参考文献(20)
相关文章 (15)

全文: PDF (2839 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要为解决现有的非线性结构模型参数识别方法面临优化过程复杂的问题，提出一种基于变分模态分解（Variational Mode Decomposition, VMD）和深度置信网络（Deep Belief Network, DBN）的非线性结构模型参数识别方法。首先，利用VMD和希尔伯特变换（Hilbert transform, HT）识别振动响应的瞬时参数。将瞬时参数进行主成分分析后作为输入，非线性模型参数作为输出。然后，利用DBN拟合两者之间的非线性映射关系。最后，将实测振动响应的瞬时参数进行主成分分析后，输入训练好的DBN可直接识别修正后的非线性模型参数。通过对两个不同非线性类型的双自由度模型和一个复杂框架模型在地震作用下的数值模拟，与高压输电结构的振动台实验，验证了该方法的有效性。数值与实验结果表明，所提出的方法具有较高的计算效率和良好的抗噪性。

	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
	莫叶1
	王佐才1
	2
	丁雅杰1
	袁子青1

关键词 ：非线性结构模型, 参数识别, VMD, DBN, 振动响应, 瞬时参数

Abstract：A nonlinear structural model parameters identification approach based on variational mode decomposition (VMD) and deep belief network (DBN) was proposed to solve the complicated optimization process in existing methods. Firstly, the instantaneous parameters of the vibration responses were identified by the VMD method and the Hilbert transform (HT). The instantaneous parameters were regarded as independent variables after principal component analysis and the nonlinear model parameters as dependent variables. The DBN was utilized for approximating the nonlinear mapping relationship between them. Finally, the identified nonlinear model parameters could be identified directly by feeding the instantaneous parameters of the measured vibration responses after principal component analysis into the well-trained DBN. The proposed method was verified by numerical simulations of two different nonlinear 2 degrees of freedoms models and a complex frame model under earthquake excitation, and a shaking table experiment of a high voltage transmission structure. Numerical and experimental results show that the proposed method has relatively high calculation efficiency and strong anti-noise ability.

Key words： nonlinear structural model parameters identification mode decomposition (VMD) deep belief network (DBN) vibration responses instantaneous parameters

收稿日期: 2020-12-16 出版日期: 2022-05-15

引用本文:

莫叶1，王佐才1,2，丁雅杰1，袁子青1. 基于VMD和DBN的非线性结构模型参数识别[J]. 振动与冲击, 2022, 41(9): 136-143.
MO Ye1, WANG Zuocai1,2, DING Yajie1, YUAN Ziqing1. Parametric recognition of nonlinear structural model based on VMD and DBN. JOURNAL OF VIBRATION AND SHOCK, 2022, 41(9): 136-143.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2022/V41/I9/136

[1] Lei Y, Huang H F, Shen W A. Update the finite element model of Canton Tower based on direct matrix updating with incomplete modal data[J]. Smart Structures and Systems, 2012, 10(4-5): 471-483.
[2] Zhang J, Wan C, Sato T. Advanced Markov Chain Monte Carlo Approach for Finite Element Calibration under Uncertainty[J]. Computer- Aided Civil and Infrastructure Engineering, 2013, 28(7):522-530.
[3] 周林仁, 欧进萍. 斜拉桥结构模型修正的子结构方法[J]. 振动与冲击，2014, 33(19):52-58.
ZHOU Lin-ren, OU Jin-ping. A Substructure Method of Structural Model Updating for Long-Span Cable-Stayed Bridges[J]. Journal of Vibration and Shock, 2014, 33(19):52-58.
[4] 翁顺, 左越, 朱宏平，等. 基于子结构的有限元模型修正方法[J]. 振动与冲击，2017, 36(4): 99-104.
WENG Shun, ZUO Yue，ZHU Hong-ping, et al. Model updating based on a substructuring method[J]. Journal of Vibration and Shock, 2017, 36(4): 99-104.
[5] 张皓, 李东升, 李宏男. 有限元模型修正研究进展:从线性到非线性[J]. 力学进展, 2019, 49(00):542-575.
ZHANG Hao, LI Dong-sheng, LI Hong-nan. Recent progress on finite element model updating: From linearity to nonlinearity[J]. Advances in Mechanics, 2019, 49(00):542-575.
[6] Astroza R, Nguyen L T, Nestorovic T. Finite element model updating using simulated annealing hybridized with unscented Kalman filter[J]. Computers & Structures, 2016, 177:176-191.
[7] Song M M, Astroza R, Ebrahimian H, et al. Adaptive Kalman filters for nonlinear finite element model updating[J]. Mechanical Systems and Signal Processing, 2020, 143:106837.
[8] Canbaloğlu G, Özgüven H N. Model updating of nonlinear structures from measured FRFs[J]. Mechanical Systems and Signal Processing, 2016, 80:282-301.
[9] 宋正华, 姜东, 曹芝腑，等. 含铰可展桁架结构非线性模型修正方法研究[J]. 振动与冲击， 2018, 37(1):21-6.
SONG Zheng-hua, JIANG Dong, CAO Zhi-fu, et al. Nonlinear model updating method for deployable trusses with joints[J]. Journal of Vibration and Shock, 2018, 37(1):21-6.
[10] Asgarieh E, Moaveni B, Stavridis A. Nonlinear finite element model updating of an infilled frame based on identified time-varying modal parameters during an earthquake[J]. Journal of Sound and Vibration, 2014, 333(23):6057-6073.
[11] Wang Z C, Xin Y, Ren W X. Nonlinear structural model updating based on instantaneous frequencies and amplitudes of the decomposed dynamic responses[J]. Engineering Structures, 2015, 100:189-200.
[12] Xie S L, Zhang Y H, Chen C H, et al. Identification of nonlinear hysteretic systems by artificial neural network[J]. Mechanical Systems and Signal Processing, 2013, 34(1-2):76-87.
[13] Hasancebi O, Dumlupinar T. Linear and nonlinear model updating of reinforced concrete T-beam bridges using artificial neural networks [J]. Computers & Structures, 2013, 119:1-11.
[14] Dragomiretskiy K, Zosso D. Variational Mode Decomposition[J]. IEEE Transactions on Signal Processing, 2014, 62(3):531-544.
[15] Ni P H, Li J, Hao H, et al. Time-varying system identification using variational mode decomposition[J]. Structural Control and Health Monitoring, 2018, 25(6):20.
[16] Wang Z C, Ren W X, Chen G D. Time-Varying Linear and Nonlinear Structural Identification with Analytical Mode Decomposition and Hilbert Transform[J]. Journal of Structural Engineering, 2013, 139(12):06013001.
[17] Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. Science, 2006, 313(5786):504-507.
[18] Mckenna F. OpenSees: A Framework for Earthquake Engineering Simulation[J]. Computing in Science & Engineering, 2011, 13(4):58-66.
[19] Gupta A, Krawinkler H. Behavior of Ductile SMRFs at Various Seismic Hazard Levels[J]. Journal of Structural Engineering, 2000, 126(1):98-107.
[20] Wang Z C , Ren W X , Chen G D . Time-varying system identification of high voltage switches of a power substation with slide-window least-squares parameter estimations[J]. Smart Materials and Structures, 2013, 22(6):065023.