Wavefield character in high rise building under earthquake shake and CNN based damage detection

Wavefield character in high rise building under earthquake shake and CNN based damage detection

Aijia Zhang, Xin Wang, Ji Dang

Abstract. Vibration of buildings can be regarded as the wave propagation in the vertical direction. Stiffness deterioration of structures due to damages could be altered by the changes in velocity and attenuation of the traveling waves. Previous studies have proposed methods to construct a new wavefield from the original wavefield of the building, in which the propagation path of the waves can be more easily recognized. In this study, firstly, we construct the wavefield with the virtual source at the top of the building (deconvolved wave), which consist of one acausal up-going wave and one causal down-going wave. Then, the changes of deconvolved waves over time at the base and inter floors are visualized. Finally, the CNN is constructed to automatically recognize the change of the visualized wavefield. To generate training data of the CNN model, multivariate nonlinear vibration simulation, reconstruction of the wavefield and visualization of the wavefield based on the vibration data was performed. To validate the trained CNN, the data of a shake table test on a 1/3 scaled 18-story steel frame building is used. As the damage progresses, the changes in the wavefield are recognized.

Keywords
Wavefield Reconstruction, Deconvolved Wave, Visualization of Wavefield, CNN, Building Damage Recognition

Published online 3/30/2023, 8 pages
Copyright © 2023 by the author(s)
Published under license by Materials Research Forum LLC., Millersville PA, USA

Citation: Aijia Zhang, Xin Wang, Ji Dang, Wavefield character in high rise building under earthquake shake and CNN based damage detection, Materials Research Proceedings, Vol. 27, pp 215-222, 2023

DOI: https://doi.org/10.21741/9781644902455-27

The article was published as article 27 of the book Structural Health Monitoring

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

References
[1] Hatakeyama T, X. Wang, S. Ohno, and M. Motosaka. Change of natural frequency of low-rise buildings based on long-term vibration records. Journal of Japan Association for Earthquake Engineering, Vol. 18, No.3, 2018. https://doi.org/10.5610/jaee.18.3_15
[2] Todorovska M.D., S.S. Ivanovic, and M.I. Todorovska. Wave propagation in a seven-story reinforced concrete building: III. Damage detection via changes in wavenumbers, Soil Dynamics and Earthquake Engineering, Vol.23, pp.65-75, 2003. https://doi.org/10.1016/S0267-7261(02)00093-3
[3] Ministry of Education, Culture, Sports, Science and Technology (MEXT), Kyoto University. Research and development on quantification of collapse margin in high-rise buildings, May 2014.
[4] Maintenance and recovery of functionality in urban infrastructures, 2015. (in Japanese). https://www.toshikino.dpri.kyoto-u.ac.jp/report/2014/report_j_2014.html
[5] Xin Wang and Masayuki Naganob. Changes of Wave Propagation Characteristics with Damage Progression Using Shake-Table Test of an 18-Story Steel Building, The 13th International Conference on Structural Safety and Reliability, 2021.
[6] Xin Wang and Masayuki Naganob. Wave propagation in an 18-story steel building and the changes of wave filed with the progression of damages, SSA Annual Meeting, 2021.
[7] A. Krizhevsky, I. Sutskever, G. E. Hinton: ImageNet Classification with Deep Convolutional Neural Networks, In Proc. of NIPS, 2012.
[8] NIED. The NIED Strong-Motion Seismograph Networks, 2022.