Rail switches are critical infrastructure components of a railroad network, that must maintain high-levels of reliable operation. Given the vast number and variety of switches that can exist across a rail network, there is an immediate need for robust automated methods of detecting switch degradations and failures without expensive add-on equipment. In this work, we explore two recent machine learning frameworks for classifying various switch degradation indicators: (1) a featureless recurrent neural network called a Long Short-Term Memory (LSTM) architecture, and (2), the Deep Wavelet Scattering Transform (DWST), which produces features that are locally time invariant and stable to time-warping deformations. We describe both methods as they apply to rail switch monitoring and demonstrate their feasibility on a dataset captured under the service conditions by Alstom Corporation. For multiple categories of degradation types, the baseline models consistently achieve near-perfect accuracies and are competitive with the manual analysis conducted by human switch-maintenance experts.
Skip Nav Destination
Close
Sign In or Register for Account
2018 Joint Rail Conference
April 18–20, 2018
Pittsburgh, Pennsylvania, USA
Conference Sponsors:
- Rail Transportation Division
ISBN:
978-0-7918-5097-8
PROCEEDINGS PAPER
Classification of Rail Switch Data Using Machine Learning Techniques
Kaylen J. Bryan
,
Kaylen J. Bryan
Florida Institute of Technology, Melbourne, FL
Search for other works by this author on:
Mitchell Solomon
,
Mitchell Solomon
Florida Institute of Technology, Melbourne, FL
Search for other works by this author on:
Emily Jensen
,
Emily Jensen
Case Western Reserve University, Cleveland, OH
Search for other works by this author on:
Christina Coley
,
Christina Coley
East Carolina University, Greenville, NC
Search for other works by this author on:
Kailas Rajan
,
Kailas Rajan
University of Southern California, Los Angeles, CA
Search for other works by this author on:
Charlie Tian
,
Charlie Tian
University of California, Berkeley, Berkeley, CA
Search for other works by this author on:
Nenad Mijatovic
,
Nenad Mijatovic
Alstom Signaling Operations LLC, Melbourne, FL
Search for other works by this author on:
James M. Kiss
,
James M. Kiss
Alstom Signaling Operations LLC, Melbourne, FL
Search for other works by this author on:
Benjamin Lamoureux
,
Benjamin Lamoureux
Alstom Signaling Operations LLC, Paris, France
Search for other works by this author on:
Pierre Dersin
,
Pierre Dersin
Alstom Signaling Operations LLC, Paris, France
Search for other works by this author on:
Anthony O. Smith
,
Anthony O. Smith
Florida Institute of Technology, Melbourne, FL
Search for other works by this author on:
Adrian M. Peter
Adrian M. Peter
Florida Institute of Technology, Melbourne, FL
Search for other works by this author on:
Kaylen J. Bryan
Florida Institute of Technology, Melbourne, FL
Mitchell Solomon
Florida Institute of Technology, Melbourne, FL
Emily Jensen
Case Western Reserve University, Cleveland, OH
Christina Coley
East Carolina University, Greenville, NC
Kailas Rajan
University of Southern California, Los Angeles, CA
Charlie Tian
University of California, Berkeley, Berkeley, CA
Nenad Mijatovic
Alstom Signaling Operations LLC, Melbourne, FL
James M. Kiss
Alstom Signaling Operations LLC, Melbourne, FL
Benjamin Lamoureux
Alstom Signaling Operations LLC, Paris, France
Pierre Dersin
Alstom Signaling Operations LLC, Paris, France
Anthony O. Smith
Florida Institute of Technology, Melbourne, FL
Adrian M. Peter
Florida Institute of Technology, Melbourne, FL
Paper No:
JRC2018-6175, V001T04A005; 10 pages
Published Online:
June 14, 2018
Citation
Bryan, KJ, Solomon, M, Jensen, E, Coley, C, Rajan, K, Tian, C, Mijatovic, N, Kiss, JM, Lamoureux, B, Dersin, P, Smith, AO, & Peter, AM. "Classification of Rail Switch Data Using Machine Learning Techniques." Proceedings of the 2018 Joint Rail Conference. 2018 Joint Rail Conference. Pittsburgh, Pennsylvania, USA. April 18–20, 2018. V001T04A005. ASME. https://doi.org/10.1115/JRC2018-6175
Download citation file:
- Ris (Zotero)
- Reference Manager
- EasyBib
- Bookends
- Mendeley
- Papers
- EndNote
- RefWorks
- BibTex
- ProCite
- Medlars
Close
Sign In
30
Views
0
Citations
Related Proceedings Papers
Related Articles
Mitigating Scattering Effects in Light-Based Three-Dimensional Printing Using Machine Learning
J. Manuf. Sci. Eng (August,2020)
Correlation of Warpage Distribution With the Material Property Scattering for Warpage Range Prediction of PBGA Components
J. Electron. Packag (December,2018)
Use of Wavelets for Analyzing Transient Radiative Heat Transfer in an Inhomogeneous Medium
J. Heat Transfer (August,2004)
Related Chapters
Model and Simulation of Low Elevation Ground-to-Air Fading Channel
International Conference on Instrumentation, Measurement, Circuits and Systems (ICIMCS 2011)
Study on Load Position Switching of Radial Scattering Dispenser
International Conference on Mechanical Engineering and Technology (ICMET-London 2011)
Scattering of Out-Plane Line Source Load by a Shallow-Embedded Circular Lining Structure and the Ground Motion
Geological Engineering: Proceedings of the 1 st International Conference (ICGE 2007)