Displacement Prediction by Acceleration of Bridge Pier Using Long Short-Term Memory

Yukihide Kajita; Ryo Morishige; Jeong Moon Kyeong; Taiji Mazda

doi:10.2749/tokyo.2025.2360

Displacement Prediction by Acceleration of Bridge Pier Using Long Short-Term Memory

Yukihide Kajita
, Ryo Morishige
, Jeong Moon Kyeong
, Taiji Mazda

Structural Analysis and Earthquake Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

After a large earthquake, prompt assessment of damage to seismic isolation bearings, especially in bridges, is critical. Although accelerometers are commonly used in structural health monitoring, displacement response calculations based on time integration of acceleration data are often inaccurate due to numerical errors. In this paper, using dynamic analysis results on a bridge model with seismic isolation bearings, a long short-term memory (LSTM) model was employed to predict displacement more accurately. Results showed that the LSTM model reduced the maximum response displacement error from 62.88% to 19.44% compared to a conventional neural network without LSTM. However, reproducing residual displacement at the pier top due to pier plasticization remained challenging.

Original language	English
Title of host publication	IABSE Symposium Tokyo 2025
Subtitle of host publication	Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report
Publisher	International Association for Bridge and Structural Engineering (IABSE)
Pages	2360-2366
Number of pages	7
ISBN (Electronic)	9783857482069
DOIs	https://doi.org/10.2749/tokyo.2025.2360
Publication status	Published - 2025
Event	IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Tokyo, Japan Duration: May 18 2025 → May 21 2025

Publication series

Name	IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report

Conference

Conference	IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches
Country/Territory	Japan
City	Tokyo
Period	5/18/25 → 5/21/25

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

All Science Journal Classification (ASJC) codes

Management of Technology and Innovation
Renewable Energy, Sustainability and the Environment
Safety, Risk, Reliability and Quality
Civil and Structural Engineering

Access to Document

10.2749/tokyo.2025.2360

Cite this

Kajita, Y., Morishige, R., Kyeong, J. M., & Mazda, T. (2025). Displacement Prediction by Acceleration of Bridge Pier Using Long Short-Term Memory. In IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report (pp. 2360-2366). (IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report). International Association for Bridge and Structural Engineering (IABSE). https://doi.org/10.2749/tokyo.2025.2360

Displacement Prediction by Acceleration of Bridge Pier Using Long Short-Term Memory. / Kajita, Yukihide; Morishige, Ryo; Kyeong, Jeong Moon et al.
IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report. International Association for Bridge and Structural Engineering (IABSE), 2025. p. 2360-2366 (IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Kajita, Y, Morishige, R, Kyeong, JM & Mazda, T 2025, Displacement Prediction by Acceleration of Bridge Pier Using Long Short-Term Memory. in IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report. IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report, International Association for Bridge and Structural Engineering (IABSE), pp. 2360-2366, IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches, Tokyo, Japan, 5/18/25. https://doi.org/10.2749/tokyo.2025.2360

Kajita Y, Morishige R, Kyeong JM, Mazda T. Displacement Prediction by Acceleration of Bridge Pier Using Long Short-Term Memory. In IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report. International Association for Bridge and Structural Engineering (IABSE). 2025. p. 2360-2366. (IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report). doi: 10.2749/tokyo.2025.2360

Kajita, Yukihide ; Morishige, Ryo ; Kyeong, Jeong Moon et al. / Displacement Prediction by Acceleration of Bridge Pier Using Long Short-Term Memory. IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report. International Association for Bridge and Structural Engineering (IABSE), 2025. pp. 2360-2366 (IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report).

@inproceedings{40c1e68c7b7b4d3ba3f7596f30cdbd56,

title = "Displacement Prediction by Acceleration of Bridge Pier Using Long Short-Term Memory",

abstract = "After a large earthquake, prompt assessment of damage to seismic isolation bearings, especially in bridges, is critical. Although accelerometers are commonly used in structural health monitoring, displacement response calculations based on time integration of acceleration data are often inaccurate due to numerical errors. In this paper, using dynamic analysis results on a bridge model with seismic isolation bearings, a long short-term memory (LSTM) model was employed to predict displacement more accurately. Results showed that the LSTM model reduced the maximum response displacement error from 62.88\% to 19.44\% compared to a conventional neural network without LSTM. However, reproducing residual displacement at the pier top due to pier plasticization remained challenging.",

keywords = "Damage estimation, Long Short-Term Memory, Nonlinear dynamic analysis, Seismic isolation bearing",

author = "Yukihide Kajita and Ryo Morishige and Kyeong, \{Jeong Moon\} and Taiji Mazda",

note = "Publisher Copyright: {\textcopyright} 2025 IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report All rights reserved.; IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches ; Conference date: 18-05-2025 Through 21-05-2025",

year = "2025",

doi = "10.2749/tokyo.2025.2360",

language = "English",

series = "IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report",

publisher = "International Association for Bridge and Structural Engineering (IABSE)",

pages = "2360--2366",

booktitle = "IABSE Symposium Tokyo 2025",

}

TY - GEN

T1 - Displacement Prediction by Acceleration of Bridge Pier Using Long Short-Term Memory

AU - Kajita, Yukihide

AU - Morishige, Ryo

AU - Kyeong, Jeong Moon

AU - Mazda, Taiji

PY - 2025

Y1 - 2025

N2 - After a large earthquake, prompt assessment of damage to seismic isolation bearings, especially in bridges, is critical. Although accelerometers are commonly used in structural health monitoring, displacement response calculations based on time integration of acceleration data are often inaccurate due to numerical errors. In this paper, using dynamic analysis results on a bridge model with seismic isolation bearings, a long short-term memory (LSTM) model was employed to predict displacement more accurately. Results showed that the LSTM model reduced the maximum response displacement error from 62.88% to 19.44% compared to a conventional neural network without LSTM. However, reproducing residual displacement at the pier top due to pier plasticization remained challenging.

AB - After a large earthquake, prompt assessment of damage to seismic isolation bearings, especially in bridges, is critical. Although accelerometers are commonly used in structural health monitoring, displacement response calculations based on time integration of acceleration data are often inaccurate due to numerical errors. In this paper, using dynamic analysis results on a bridge model with seismic isolation bearings, a long short-term memory (LSTM) model was employed to predict displacement more accurately. Results showed that the LSTM model reduced the maximum response displacement error from 62.88% to 19.44% compared to a conventional neural network without LSTM. However, reproducing residual displacement at the pier top due to pier plasticization remained challenging.

KW - Damage estimation

KW - Long Short-Term Memory

KW - Nonlinear dynamic analysis

KW - Seismic isolation bearing

UR - https://www.scopus.com/pages/publications/105008759038

UR - https://www.scopus.com/pages/publications/105008759038#tab=citedBy

U2 - 10.2749/tokyo.2025.2360

DO - 10.2749/tokyo.2025.2360

M3 - Conference contribution

AN - SCOPUS:105008759038

T3 - IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches - Report

SP - 2360

EP - 2366

BT - IABSE Symposium Tokyo 2025

PB - International Association for Bridge and Structural Engineering (IABSE)

T2 - IABSE Symposium Tokyo 2025: Environmentally Friendly Technologies and Structures: Focusing on Sustainable Approaches

Y2 - 18 May 2025 through 21 May 2025

ER -

Displacement Prediction by Acceleration of Bridge Pier Using Long Short-Term Memory

Abstract

Publication series

Conference

UN SDGs

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this