Predictive Modeling of Ground Vibration Induced by High-Speed Train Using Machine Learning Techniques

DOI

10.2991/978-94-6463-936-0_16How to use a DOI?

Keywords

Machine Learning; Ground Vibration; LS Boost; Decision Tree; Random Forest; Xg Boost; K-Nearest Neighbors (KNN); SHAP; Performance Parameters

Abstract

Vibrations induced by trains are a growing concern in urban and transport infrastructure planning. Understanding their impact is essential to protect nearby structures, ensure occupant comfort, and maintain the safety and durability of buildings, bridges, and tunnels. Traditional approaches to modeling ground vibration often rely on simplistic assumptions and empirical formulas, which may limit their accuracy and applicability in diverse real world scenarios. Nowadays, the machine learning (ML) techniques has offered a promising avenue for improving the precision and efficiency of ground vibration modeling. Developing a prediction model is highly valuable, where on site measurements are difficult, time consuming, or not feasible due to accessibility or safety constraints. Such models can efficiently handle complex datasets involving multiple variables and can be used to predict vibrations across various scenarios, including different train speeds, soil conditions, and distances from the track. This study investigates the ML techniques for modeling ground vibration induced by semi high-speed rail systems, focusing on the development and evaluation of five distinct models using a comprehensive dataset. Through rigorous analysis and comparison of various performance parameters, the Least Squares Boosting (LS Boost) model emerged as the most effective in accurately predicting ground vibration levels. Using SHAP (SHapley Additive exPlanations) analysis, the study identifies distance from the track center as a significant parameter influencing ground vibration. This study contributes to the advancement of predictive modeling techniques for assessing and mitigating ground vibration impacts in diverse applications such as civil engineering and urban planning.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - M. A. Meera
AU  - M. N. Sandeep
PY  - 2025
DA  - 2025/12/29
TI  - Predictive Modeling of Ground Vibration Induced by High-Speed Train Using Machine Learning Techniques
BT  - Proceedings of the International Conference on Civil Engineering and Architecture for Sustainable Infrastructure Development and Environment (CEASIDE 2025)
PB  - Atlantis Press
SP  - 215
EP  - 225
SN  - 3005-155X
UR  - https://doi.org/10.2991/978-94-6463-936-0_16
DO  - 10.2991/978-94-6463-936-0_16
ID  - Meera2025
ER  -