Quantification of Dynamic Track Stiffness Using Machine Learning
Railway track stiffness is an essential factor influencing the track conditions and long-term deterioration. However, the traditional ways to measure the track stiffness are based on inverse computations using multi-body simulations and/or finite element models, which are time-consuming and at low-s...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9830617/ |
| _version_ | 1818472518031245312 |
|---|---|
| author | Junhui Huang Xiaojie Yin Sakdirat Kaewunruen |
| author_facet | Junhui Huang Xiaojie Yin Sakdirat Kaewunruen |
| author_sort | Junhui Huang |
| collection | DOAJ |
| description | Railway track stiffness is an essential factor influencing the track conditions and long-term deterioration. However, the traditional ways to measure the track stiffness are based on inverse computations using multi-body simulations and/or finite element models, which are time-consuming and at low-speed operation. To overcome these challenges, we propose a convolutional neural network framework to predict the track dynamic stiffness using the accelerations captured by accelerometers mounted on the axle box in real-time. To provide a benefit of computational cost-friendly, a dilated convolutional layer has been added which allows the framework to be applied to a compact device. In our study, a nonlinear finite element model of train-track interactions has been calibrated and used to generate unbiased, full range of data sets of axle box accelerations under various track and operational factors. Subsequently, the simulated data is formatted to three different sample sizes: 250-timesteps, 500-timesteps, and 1,000-time steps. The fine-tuned CNN model is developed based on the three datasets and provides the optimal <inline-formula> <tex-math notation="LaTeX">$\textrm {R}^{2}$ </tex-math></inline-formula> of 0.94, 0.94, and 0.97. The insights gained from this study can assist the track stiffness measurement in the field with a novel measurement method providing continuous, cost-friendly, fast, and implementable benefits. The quantification of dynamic track stiffness will help track engineers to locate problematic and defective tracks promptly on the vast railway networks such as mud pumping, loss of support, pulverized ballast, and so on. |
| first_indexed | 2024-04-14T04:08:16Z |
| format | Article |
| id | doaj.art-4c68f9110c174464aa5fd8e04ca7fcd6 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-14T04:08:16Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-4c68f9110c174464aa5fd8e04ca7fcd62022-12-22T02:13:17ZengIEEEIEEE Access2169-35362022-01-0110787477875310.1109/ACCESS.2022.31912789830617Quantification of Dynamic Track Stiffness Using Machine LearningJunhui Huang0Xiaojie Yin1Sakdirat Kaewunruen2https://orcid.org/0000-0003-2153-3538Department of Civil Engineering, School of Engineering, University of Birmingham, Birmingham, U.K.Department of Civil Engineering, School of Engineering, University of Birmingham, Birmingham, U.K.Department of Civil Engineering, School of Engineering, University of Birmingham, Birmingham, U.K.Railway track stiffness is an essential factor influencing the track conditions and long-term deterioration. However, the traditional ways to measure the track stiffness are based on inverse computations using multi-body simulations and/or finite element models, which are time-consuming and at low-speed operation. To overcome these challenges, we propose a convolutional neural network framework to predict the track dynamic stiffness using the accelerations captured by accelerometers mounted on the axle box in real-time. To provide a benefit of computational cost-friendly, a dilated convolutional layer has been added which allows the framework to be applied to a compact device. In our study, a nonlinear finite element model of train-track interactions has been calibrated and used to generate unbiased, full range of data sets of axle box accelerations under various track and operational factors. Subsequently, the simulated data is formatted to three different sample sizes: 250-timesteps, 500-timesteps, and 1,000-time steps. The fine-tuned CNN model is developed based on the three datasets and provides the optimal <inline-formula> <tex-math notation="LaTeX">$\textrm {R}^{2}$ </tex-math></inline-formula> of 0.94, 0.94, and 0.97. The insights gained from this study can assist the track stiffness measurement in the field with a novel measurement method providing continuous, cost-friendly, fast, and implementable benefits. The quantification of dynamic track stiffness will help track engineers to locate problematic and defective tracks promptly on the vast railway networks such as mud pumping, loss of support, pulverized ballast, and so on.https://ieeexplore.ieee.org/document/9830617/Track stiffnessaxle box accelerationsdilated convolutionalmachine learningrailway infrastructure |
| spellingShingle | Junhui Huang Xiaojie Yin Sakdirat Kaewunruen Quantification of Dynamic Track Stiffness Using Machine Learning IEEE Access Track stiffness axle box accelerations dilated convolutional machine learning railway infrastructure |
| title | Quantification of Dynamic Track Stiffness Using Machine Learning |
| title_full | Quantification of Dynamic Track Stiffness Using Machine Learning |
| title_fullStr | Quantification of Dynamic Track Stiffness Using Machine Learning |
| title_full_unstemmed | Quantification of Dynamic Track Stiffness Using Machine Learning |
| title_short | Quantification of Dynamic Track Stiffness Using Machine Learning |
| title_sort | quantification of dynamic track stiffness using machine learning |
| topic | Track stiffness axle box accelerations dilated convolutional machine learning railway infrastructure |
| url | https://ieeexplore.ieee.org/document/9830617/ |
| work_keys_str_mv | AT junhuihuang quantificationofdynamictrackstiffnessusingmachinelearning AT xiaojieyin quantificationofdynamictrackstiffnessusingmachinelearning AT sakdiratkaewunruen quantificationofdynamictrackstiffnessusingmachinelearning |