Use of Deep Learning Algorithms To Prevent Pantograph-Catenary Malfunctions
Abstract
Today, the demand for rail transport is increasing. Studies in this area are increasing worldwide. While the railway infrastructure is increasing in the world, the suitability of the railroads and train sets built is of great importance in terms of road and passenger safety. The most important test to ensure road and passenger safety is on the electrification line. The energy required for the movement of the electric train is provided by the power line. Continuous contact between the power line and the pantograph is desired while in motion by providing continuous energy for the rail system to operate. Even short-term non-contact between the pantograph and the catenary adversely affects the rail system vehicle and the electronic systems inside. For this reason, the pantograph and catenary interaction should be controlled dynamically and statically in certain periods. In this study, dynamic and static control was provided by using deep learning. The data received from the system are recorded in CSV format. Using deep learning algorithms, failure points have been successfully detected up to 99.4%.
Keywords
Catenary, data acquisition, high speed train, fault detection, pantograph, Catenary, data acquisition, high speed train, fault detection, pantograph
Supporting Institution
TUBITAK
Project Number
EEEAG‑118E322
Thanks
This work has been supported by the Scientific and Technological Research Council of Turkey (TUBITAK) under Grant EEEAG‑118E322. (EEE‑AG: Electrical Electronics Engineering Research Group).
References
- Arı, D., Çıbuk M., & Ağgün, F. (2018). A New Proxy-Based Network Joining Method for Linear Wireless Sensor Networks. International Engineering and Natural Sciences Conference (IENSC 2018), 2018,715-723.
- Bucca, G., Collina, A., (2009). A Procedure for the Wear Prediction of Collector Strip and Contact Wire in Pantograph–Catenary System, Wear, 266, 46–59.
- Boguslavskii, A.A., & Sokolov. S.M., (2006). Detecting Objects in Images in Real-Time Computer Vision Systems Using Structured Geometric Models, Pleiades Publishing, Inc., 32, 177–18.
- Burrow, M., Ghataora, G., & Stirling, A. (2004). Rail Research UK, Department of Civil Engineering, University Of Birmingham, A Rational Approach To Railway Track Substructure Design, Birmingham, UK.
- Carbonell, J.G. (1983). Learning by analogy, Formulating and generalizing plans from past experience, In Machine learning, 137-161. Springer, Berlin, Heidelberg.
- Cengiz, M.S. (2013). Smart meter and cost experiment, Przeglad Elektrotechniczny 89(11), 206-209.
- Cengiz, Ç., Yapıcı, I., & Cengiz, M.S. (2018). Fourier Analysis in Rail Systems. International Conference on Multidisciplinary, Science, Engineering., & Technology (2018 Dubai, UAE).
- Cengiz, M.S., & Cengiz, Ç. (2018). Numerical analysis of tunnel lighting maintenance factor. International Islamic University Malaysia Journal, 19(2), 154-163.
- Cengiz, Ç., Eren, M., Kaynaklı, M., Yapıcı, I., Gencer, G., & Yurci, Y. (2017). Numerical Analysis of Maintenance Factor for Tunnel and Road In Solid State Lighting, International Conference on Multidisciplinary, Science, Engineering and Technology, October 27-29, 2017. Bitlis.
- Cengiz, Ç., Kaynaklı, M., Gencer, G., Eren, M., Yapici, İ., Yildirim, S., & Cengiz, MS. (2017). Selection Criteria and Economic Analysis, International Conference on Multidisciplinary, Science, Engineering, and Technology Bitlis Book of Abstracts,1 (1), 27-29.