Metro Track Geometry Defect Identification Model Based on Car-Body Vibration Data and Differentiable Architecture Search
Source abstract
Abstract
Efficient and low-cost modes for detecting metro track geometry defects (TGDs) are essential for condition-prediction-based preventive maintenance, which can help improve the safety of metro operations and reduce the maintenance cost of metro tracks. Compared with the traditional TGD detection method that utilizes the track geometry car, the method that uses a portable detector to acquire the car-body vibration data (CVD) can be used on an ordinary in-service train without occupying the metro schedule line, thereby improving efficiency and reducing the cost. A convolutional neural network-based identification model for TGD, built on a differentiable architecture search, is proposed in this study to employ only the CVD acquired by a portable detector for integrated identification of the type and severity level of TGDs. Second, the random oversampling method is introduced, and a strategy for applying this method is proposed to improve the poor training effect of the model caused by the natural class-imbalance problem arising from the TGD dataset. Subsequently, a comprehensive performance-evaluation metric (track geometry defect F-score) is designed by considering the actual management needs of the metro infrastructure. Finally, a case study is conducted using actual field data collected from Beijing Subway to validate the proposed model.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Railway Engineering and Dynamics · Infrastructure Maintenance and Monitoring · Vehicle License Plate Recognition
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044, China
Key Laboratory of Transport Industry, Big Data Application Technologies for Comprehensive Transport, Beijing Jiaotong University, Beijing 100044, China
Funding
National Natural Science Foundation of China · Science and Technology Research and Development Plan Project of China State Railway Group Co., Ltd.
References (62)
Show source-provided references
1. Track geometry degradation and maintenance modelling: A review10.1177/0954409716657849 ↗
2. Classification-learning-based framework for predicting railway track irregularities10.1177/0954409714552818 ↗
3. (2017). Standard No. EN 13848-5:2017.
4. FRA (Federal Railroad Administration) (2022, March 30). Track and Rail and Infrastructure Integrity Compliance Manual: Volume II Track Safety Standards, Available online: https://railroads.dot.gov/elibrary/track-and-rail-and-infrastructure-integrity-compliance-manual-volume-ii-chapter-2-track-0.
5. (2020). Standard No. GB/T 39559.4-2020.
6. Hierarchical Bayesian modelling of rail track geometry degradation10.1177/0954409713486619 ↗
7. Optimizing the Alignment of Inspection Data from Track Geometry Cars10.1111/mice.12067 ↗
8. Zhang, Y.J., Rusk, K., and Clouse, A.L. (2012, January 16–19). Decades of automated track inspection success and strategy for tomorrow. Proceedings of the 2012 AREMA Annual Conference, Chicago, IL, USA.
9. Development of track condition assessment model based on visual inspection10.1080/15732470903194676 ↗
10. Development of a new track geometry assessment technique incorporating rail cant factor10.1243/09544097JRRT237 ↗
11. Jamieson, D., Bloom, J., and Kelshaw, R. (2001, January 9–12). T-2000: A railroad track geometry inspection vehicle for the 21st century. Proceedings of the 2001 AREMA Annual Conference, Chicago, IL, USA.
12. Perspectives on railway track geometry condition monitoring from in-service railway vehicles10.1080/00423114.2015.1034730 ↗
13. Development of railway track condition monitoring from multi-train in-service vehicles10.1080/00423114.2020.1755045 ↗
14. Development of metro track geometry fault diagnosis convolutional neural network model based on car-body vibration data10.1177/09544097221080366 ↗
15. Tsunashima, H. (2019). Condition Monitoring of Railway Tracks from Car-Body Vibration Using a Machine Learning Technique. Appl. Sci., 9.10.3390/app9132734 ↗
16. An overview: Modern techniques for railway vehicle on-board health monitoring systems10.1080/00423114.2017.1296963 ↗
17. Scott, G., Chillingworth, E., and Dick, M. (2010, January 27–29). Development of an Unattended Track Geometry Measurement System. Proceedings of the 2010 Joint Rail Conference, Urbana, IL, USA.10.1115/JRC2010-36282 ↗
18. Monitoring lateral track irregularity from in-service railway vehicles10.1243/0954409JRRT64 ↗
19. Monitoring vertical track irregularity from in-service railway vehicles10.1243/0954409JRRT65 ↗
20. Urban rail track condition monitoring based on in-service vehicle acceleration measurements10.1016/j.measurement.2015.11.033 ↗
21. Estimation of railway track longitudinal irregularity using vehicle response with information compression and Bayesian deep learning10.1111/mice.12802 ↗
22. Kawasaki, J., and Youcef-Toumi, K. (2002, January 8–10). Estimation of rail irregularities. Proceedings of the 2002 American Control Conference (IEEE Cat. No.CH37301), Anchorage, AK, USA.10.1109/ACC.2002.1024495 ↗
23. A Mixed Filtering Approach for Track Condition Monitoring Using Accelerometers on the Axle Box and Bogie10.1109/TIM.2011.2170377 ↗
24. Track geometry estimation from car-body vibration10.1080/00423114.2014.889836 ↗
25. Estimation of lateral and cross alignment in a railway track based on vehicle dynamics measurements10.1016/j.ymssp.2018.06.041 ↗
26. Deep Learning for Track Quality Evaluation of High-Speed Railway Based on Vehicle-Body Vibration Prediction10.1109/ACCESS.2019.2960537 ↗
27. Condition monitoring of railway track from car-body vibration using time-frequency analysis10.1080/00423114.2020.1850808 ↗
28. A novel model for determining the amplitude-wavelength limits of track irregularities accompanied by a reliability assessment in railway vehicle-track dynamics10.1016/j.ymssp.2016.10.010 ↗
29. Smartphone’s Sensing Capabilities for On-Board Railway Track Monitoring: Structural Performance and Geometrical Degradation Assessment10.1155/2019/1729153 ↗
30. Identification of Subway Track Irregularities Based on Detection Data of Portable Detector10.1177/03611981221097088 ↗
31. A reinforcement neural architecture search method for rolling bearing fault diagnosis10.1016/j.measurement.2019.107417 ↗
32. A comprehensive review on convolutional neural network in machine fault diagnosis10.1016/j.neucom.2020.07.088 ↗
33. One-shot neural architecture search for fault diagnosis using vibration signals10.1016/j.eswa.2021.116027 ↗
34. A Comprehensive Survey of Neural Architecture Search
35. Neural Architecture Search: A Survey
36. Zoph, B., and Le, Q.V. (2017). Neural architecture search with reinforcement learning. arXiv.
37. Baker, B., Gupta, O., Naik, N., and Raskar, R. (2017). Designing Neural Network Architectures using Reinforcement Learning. arXiv.
38. Liu, H., Simonyan, K., and Yang, Y. (2019, January 6–9). DARTS: Differentiable Architecture Search. Proceedings of the Seventh International Conference on Learning Representations (ICLR 2019), New Orleans, LA, USA.
39. Brock, A., Lim, T., Ritchie, J.M., and Weston, N. (2017). SMASH: One-Shot Model Architecture Search through HyperNetworks. arXiv.
40. A systematic study of the class imbalance problem in convolutional neural networks10.1016/j.neunet.2018.07.011 ↗
41. Bai, L., Liu, R., and Li, Q. (2020). Data-Driven Bias Correction and Defect Diagnosis Model for In-Service Vehicle Acceleration Measurements. Sensors, 20.10.3390/s20030872 ↗
42. Vinberg, E.M., Martin, M., Firdaus, A.H., Tang, Y., and Qazizadeh, A. (2018). Railway Applications of Condition Monitoring, KTH Royal Institute of Technology. Technical Report.
43. 1-D CNNs for structural damage detection: Verification on a structural health monitoring benchmark data10.1016/j.neucom.2017.09.069 ↗
44. Multiclass Damage Identification in a Full-Scale Bridge Using Optimally Tuned One-Dimensional Convolutional Neural Network10.1061/(ASCE)CP.1943-5487.0001003 ↗
45. Subway tunnel damage detection based on in-service train dynamic response, variational mode decomposition, convolutional neural networks and long short-term memory10.1016/j.autcon.2022.104293 ↗
46. Gradient-based learning applied to document recognition10.1109/5.726791 ↗
47. Zoph, B., Vasudevan, V., Shlens, J., and Le, Q.V. (2018, January 18–22). Learning Transferable Architectures for Scalable Image Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA.10.1109/CVPR.2018.00907 ↗
48. Chollet, F. (2016, January 21–26). Xception: Deep Learning with Depthwise Separable Convolutions. Proceedings of the 30TH IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), Honolulu, HI, USA.10.1109/CVPR.2017.195 ↗
49. Yu, F., and Koltun, V. (2016). Multi-Scale Context Aggregation by Dilated Convolutions. arXiv.
50. Ioffe, S., and Szegedy, C. (2015, January 6–11). Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. In Proceedings of Machine Learning Research, Lille, France.
51. Nair, V., and Hinton, G.E. (2010, January 21–24). Rectified Linear Units Improve Restricted Boltzmann Machines Vinod Nair. Proceedings of the International Conference on International Conference on Machine Learning, Haifa, Israel.
52. Liu, C., Zoph, B., Neumann, M., Shlens, J., Hua, W., Li, L.J., Fei-Fei, L., Yuille, A., Huang, J., and Murphy, K. (2018, January 8–14). Progressive Neural Architecture Search. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.10.1007/978-3-030-01246-5_2 ↗
53. Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep Learning, MIT Press.
54. The class imbalance problem: A systematic study10.3233/IDA-2002-6504 ↗
55. Bowyer, K.W., Chawla, N.V., Hall, L.O., and Kegelmeyer, W.P. (2011). SMOTE: Synthetic Minority Over-sampling Technique. arXiv.
56. Chinchor, N., and Sundheim, B.M. (1993, January 25–27). MUC-5 evaluation metrics. Proceedings of the Fifth Message Understanding Conference (MUC-5), Baltimore, MA, USA.10.3115/1072017.1072026 ↗
57. Ruder, S. (2016). An overview of gradient descent optimization algorithms. arXiv.
58. Loshchilov, I., and Hutter, F. (2017). SGDR: Stochastic Gradient Descent with Warm Restarts. arXiv.
59. Kingma, D.P., and Ba, J.L. (2014). Adam: A method for stochastic optimization. arXiv.
60. Devries, T., and Taylor, G.W. (2017). Improved Regularization of Convolutional Neural Networks with Cutout. arXiv.
61. Dropout: A Simple Way to Prevent Neural Networks from Overfitting
62. Wallach, H., Larochelle, H., Beygelzimer, A., Buc, F.D.T.A., Fox, E., and Garnett, R. (2019). Advances in Neural Information Processing Systems 32, Curran Associates, Inc.