Related Experiment Video
Updated: Jun 30, 2025
08:27
Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
Published on: January 5, 2024
1.1K
Construction of environmental vibration prediction model for subway transportation based on machine learning
1Mechanical and Industrial Engineering Department, College of Engineering, Northeastern University, 360 Huntington Ave, Boston, MA, USA. z17738729171@163.com.
Scientific Reports
|March 16, 2024
Summary
This study introduces a new model using machine learning to predict subway vibrations, improving accuracy and operational stability. The developed model accurately forecasts vibration levels, ensuring smoother metro transport operations.
Area of Science:
- Civil Engineering
- Mechanical Engineering
- Data Science
Background:
- Metro transport vibrations, primarily from wheel-rail interaction, disrupt operations.
- Existing prediction models lack accuracy, speed, and broad applicability.
- Accurate vibration prediction is crucial for subway system efficiency and passenger comfort.
Purpose of the Study:
- To develop a novel vibration prediction model for the metro transport environment.
- To enhance the accuracy and efficiency of subway vibration forecasting.
- To address limitations of traditional prediction methods.
Main Methods:
- Utilized database theory and machine learning algorithms.
- Developed a predictive model for subway transportation environment vibrations.
- Validated the model against real-world operational data.
Main Results:
- The model demonstrated high accuracy with an average difference of 1.4 dB between predicted and real values.
- Maximum error differences were minimal (0.29% and 8.2%).
- Predictions showed good agreement with experimental data, especially within 100m ranges.
Conclusions:
- The developed model accurately predicts subway transport environment vibrations.
- This approach offers a viable solution for enhancing subway operational stability.
- The findings support the use of machine learning for infrastructure monitoring and prediction.
