基于串联LSTM网络的振动台子结构试验模拟

王玉梅,纪金豹,王东岳

振动与冲击 ›› 2023, Vol. 42 ›› Issue (23) : 80-86.

PDF(1850 KB)
PDF(1850 KB)
振动与冲击 ›› 2023, Vol. 42 ›› Issue (23) : 80-86.
论文

基于串联LSTM网络的振动台子结构试验模拟

  • 王玉梅,纪金豹,王东岳
作者信息 +

Simulation of shaking table substructure tests based on series LSTM network

  • WANG Yumei,JI Jinbao,WANG Dongyue
Author information +
文章历史 +

摘要

振动台子结构试验具有试验测试与数值模拟的双重优势,可实现大尺度甚至足尺模型结构的地震作用复现。振动台子结构试验中数值子结构的实时求解是影响试验精度和系统稳定性、决定试验能否顺利实现的关键因素之一。为提升数值子结构的求解性能,将长短时记忆网络(Long Short-Term Memory Network, LSTM)引入到振动台子结构试验中,分别构建了用于模拟试验子结构和数值子结构的神经网络模型,并且在训练数据中引入时滞以模拟系统延迟带来的影响。选择5层钢框架模型结构对神经网络模型进行验证,结果显示,所构建的神经网络模型具有良好的精度、稳定性和时滞补偿能力,计算效率可满足实时控制要求。所提出的神经网络模型可用于振动台子结构试验的数值子结构实时求解。

Abstract

Substructural shaking table test has the advantages of real-time testing and numerical analysis, and can realize the recurrence of earthquake response of large-scale or even full-scale model. The real-time numerical solution of the numerical substructure in the substructural shaking table test is one of the key factors that affects the test accuracy and system stability, and determines whether the test can be successfully implementation. To improve the performance of numerical substructure analysis, the long short-term memory network (LSTM) was introduced into the substructure shaking table test, and the neural network models to simulate the physical substructure and the numerical substructure were modeled respectively. The time-delay was introduced into the training data to simulate the time-delay effect caused by the system delay. A 5-story steel frame structure model was selected to verify the neural network model. The results show that the network models have good accuracy, stability and time delay compensation ability, and the calculation efficiency can meet the real-time control requirements. The proposed neural network model can be used to solve the numerical substructure of the substructural shaking table test.

关键词

神经网络 / LSTM / 振动台子结构试验 / 实时混合试验 / 结构实验技术

Key words

neural network / LSTM;substructural shaking table test / real time hybrid test / structure experiment technology

 

引用本文

导出引用
王玉梅,纪金豹,王东岳. 基于串联LSTM网络的振动台子结构试验模拟[J]. 振动与冲击, 2023, 42(23): 80-86
WANG Yumei,JI Jinbao,WANG Dongyue. Simulation of shaking table substructure tests based on series LSTM network[J]. Journal of Vibration and Shock, 2023, 42(23): 80-86

参考文献

[1]  王向英,田石柱. 子结构地震模拟振动台混合试验原理与实现[J]. 地震工程与工程振动. 2009, 29(04): 46-52.
WANG Xiang-ying, TIAN Shi-zhu. Principle and implementation of the hybrid testing method based on substructure techniques by using shaking table[J].Earthquake Engineering and Engineering Dynamics. 2019, 39(4):7.
[2] Wu B, Xu G, Wang Q, et al. Operator-splitting method for real-time substructure testing[J]. Earthquake Engineering & Structural Dynamics. 2006, 35(3): 293-314.
[3]  吴斌,保海娥.实时子结构实验Chang算法的稳定性和精度[J]. 地震工程与工程振动. 2006(02): 41-48.
WU Bin,BAO Hai-e. Stability and accuracy of Chang algorithm for real-tmi e substructure testing[J].Earthquake Engineering and Engineering Dynamics. . 2006(02): 41-48.
[4] Nakashima M, Ishida M, Ando K. Integration techniques for subatructure pesudo dynamic test: Pseudo dynamic test using substructuring techniques[J]. Journal of Structural and Construction Engineering (Transactions of AIJ), 1990, 417:107-117.
[5] Chang S Y. Explicit Pseudodynamic Algorithm with Unconditional Stability[J]. Journal of Engineering Mechanics, 2002, 128(9):935-947.
[6] Wang T, Zhou H, Zhang X, et al. Stability of an explicit time-integration algorithm for hybrid tests, considering stiffness hardening behavior[J]. Earthquake Engineering and Engineering Dynamics. 2018, 17(3): 595-606.
[7] 陈再现,韩光,王焕定,等. 传统界面处理的子结构拟动力试验误差分析[J]. 地震工程与工程振动. 2014, 34(S1): 657-662.
CHEN Zai-xian, HAN Guang, WANG Huan-ding, et al. Error study on conventional interface scheme of substructure pseudo-dynamic testing method[J]. Earthquake Engineering and Engineering Dynamics. 2014, 34(S1): 657-662.
[8] 李振宝,李晓亮,唐贞云,等. 基于振动台的动力子结构试验界面反力获取方法[J]. 地震工程与工程振动. 2011, 31(03): 65-70.
LI Zhen-bao, LI Xiao-liang, TANG Zhen-yun, et al. Research on the methods of evaluating reaction force for dynamic substructure experiments using shaking table[J]. Earthquake Engineering and Engineering Dynamics. 2011, 31(03): 65-70.                             
[9] 王向英,田石柱. 子结构地震模拟振动台混合试验原理与实现[J]. 地震工程与工程振动. 2009, 29(04): 46-52.
WANG Xiangying,TIAN Shizhu. Principle and implementation of hybrid testing method based on substructure techniques by using shaking table[J]. Earthquake Engineering and Engineering Dynamics. 2009, 29(04): 46-52.
[10] 纪金豹,丛鹏里,王晨光. 基于多参量控制AMD的振动台子结构试验数值仿真[J]. 北京工业大学学报. 2019, 45(03): 213-220.
JI Jinbao,CONG Pengli,WANG Chenguang. Simulation of Shaking Table Substructure Test Based on Multi-variable Controlled AMD Device[J]. Journal of Beijing University of Technology,2019, 45(03): 213-220.
[11] Wu B,  Wang Z,  Bursi O S. Actuator dynamics compensation based on upper bound delay for real‐time hybrid simulation[J]. Earthquake Engineering & Structural Dynamics, 2013, 42(12).
[12] 王贞,李强,吴斌. 实时混合试验的自适应时滞补偿方法[J]. 工程力学. 2018, 35(09): 37-43.
WANG Zhen, LI  Qiang, WU Bin. Adaptive delay compensation method for real-time[J]. Engineering mechanics. 2018, 35(09): 37-43.
[13] 李宁,周陈,周子豪,等. 二阶段在线迭代时滞补偿方法及试验验证[J]. 振动与冲击. 2020, 39(17): 31-38.
LI Ning, ZHOU Chen, ZHOU Zi-hao,et al. A two-stage online iteration time-delay compensation method for real time hybrid testing: simulation and test verification[J]. Journal of Vibration and Shock. 2020, 39(17): 31-38.
[14] Bao Y, Tang Z, Li H, et al. Computer vision and deep learning-based data anomaly detection method for structural health monitoring[J]. Structural Health Monitoring-an International Journal, 2019, 18(2): 401-421.
[15] Voulodimos A, Doulamis N, Doulamis A, et al. Deep Learning for Computer Vision: A Brief Review[J]. Computational Intelligence and Neuroscience, 2018(7068349).
[16] Russakovsky O, Deng J, Su H, et al. ImageNet Large Scale Visual Recognition Challenge[J]. International Journal of Computer Vision. 2015, 115(3): 211-252.
[17] Dahl G E ,  Yu D ,  Deng L , et al. Context-Dependent Pre-Trained Deep Neural Networks for Large-Vocabulary Speech Recognition[J]. IEEE Transactions on Audio Speech & Language Processing, 2011, 20(1):30-42.
[18] Zhang R, Liu Y, Sun H. Physics-guided convolutional neural network(PhyCNN) for data-driven seismic response modeling[J]. Engineering Structures. 2020, 215: 110704.
[19] 许泽坤,陈隽. 非线性结构地震响应的神经网络算法[J]. 工程力学. 2021, 38(09): 133-145.
XU Ze-kun, CHEN Jun. Neural network algorithm for nonlinear structural seismic response[J]. Engineering mechanics. 2021, 38(09): 133-145.
[20] Xu Z, Chen J, Shen J, et al. Recursive long short-term memory network for predicting nonlinear structural seismic response[J].  Engineering Structures. 2022, 250: 113406.
[21] 涂建维,戴葵,瞿伟廉.磁流变阻尼器的磁滞效应与神经网络预测调整[J].华中科技大学学报( 自然科学版) ,2007,35(3):110-112.
TU Jian wei,DAI Kui,QU Wei-lian. Magnetic hysteresis of magnetorheological dampers and its compensation by neural network prediction[J]. Journal of Huazhong University of Science & Technology ( Natural Science Edition) ,2007,35(3) : 110 - 112.
[22] 王涛,翟绪恒,孟丽岩,等. 基于在线神经网络算法的混合试验方法[J]. 振动与冲击. 2017, 36(14): 1-8.
WANG Tao, ZHAI Xu-heng, MENG Li-yan, et al. Hybrid
testing method based on an online neural network algorithm
[J]. Journal of Vibration and Shock. 2017, 36(14): 1-8.
[23] 王燕华,吕静,吴京. 基于遗忘因子和LMBP神经网络的混合试验在线模型更新方法[J]. 振动与冲击. 2020, 39(09): 42-48.
WANG Yan-hua, LV Jing, Wu Jing. On-line model updating method for hybrid testings based on the forgetting factor and LMBP neural network[J]. Journal of Vibration and Shock. 2017, 36(14): 1-8.
[24] 王涛,翟绪恒,孟丽岩. 在线自适应神经网络算法及参数鲁棒性分析[J]. 振动与冲击. 2019, 38(08): 210-217.
WANG Tao, ZHAI Xu-heng, MENG Li-yan. On online adaptive neural network algorithm and its parameters robustness analysis[J]. Journal of Vibration and Shock. 2019, 38(08): 210-217.
[25] Bas E E, Moustafa M A. Real-Time Hybrid Simulation with Deep Learning Computational Substructures: System Validation Using Linear Specimens[J]. Machine Learning and Knowledge Extraction. 2020, 2(4): 469-489.
[26]杨丽, 吴雨茜, 王俊丽,等. 循环神经网络研究综述[J]. 计算机应用, 2018, 38(A02):7.
TANG Li,WU Yuxi , WANG Junli, et al. Research on recurrent neural network[J]. Journal of Computer Applications. 2018, 38(A02):7.
[27] 纪金豹, 王晨光, 闫维明. 基于TMD加载的振动台子结构试验技术研究[J]. 地震工程与工程振动, 2019, 39(4):7. 2018, 38(A02):7.
JI Jin-bao, WANG Chen-guang,YAN Wei-ming. Study of shaking table substructure test loading by TMD device[J].Earthquake Engineering and Engineering Dynamics. 2019, 39(4):7.
[28] 王倩颖,吴斌,欧进萍,等. 应用MTS控制系统的实时子结构实验[J]. 哈尔滨工业大学学报. 2008, 40(12): 1895-1899.
WANG Qian,WU Bin,OU Jin-ping, et al. Real-time substructure test with MTS control system[J]. Journal of Harbin Institute of Technology. 2008, 40(12): 1895-1899.
[29] Horiuchi T, Inoue M, Konno T, et al. Real-time hybrid experimental system with actuator delay compensation and its application to a piping system with energy absorber[J]. Earthquake engineering & structural dynamics. 1999, 28(10): 1121-1141.
[30] Agrawal A K, Yang J N. Compensation of time-delay for control of civil engineering structures[J]. Earthquake engineering & structural dynamics. 2000, 29(1): 37-62.
[31] Chae Y, Kazemibidokhti K, Ricles J M. Adaptive time series compensator for delay compensation of servo-hydraulic actuator systems for real-time hybrid simulation[J]. Earthquake Engineering & Structural Dynamics. 2013, 42(11): 1697-1715.

PDF(1850 KB)

Accesses

Citation

Detail

段落导航
相关文章

/