Machine learning approach for wheel flat detection of railway train wheels
Nowadays, monitoring the condition of railway infrastructure has become necessary and led railway companies to take advantage of artificial intelligence (AI) technologies to improve safety and reduce operating costs. This paper aims to present an unsupervised methodology to detect railway wheel flats. The automatic damage detection algorithm is based on the acceleration evaluated on the rails for the passage of traffic loads. The results of this research study show that only one sensor is enough to detect wheel flat automatically, which enables the development of a low-cost monitoring system that can be easily implemented in actual operating conditions.
- Record URL:
- Record URL:
-
Availability:
- Find a library where document is available. Order URL: http://worldcat.org/issn/23521465
-
Supplemental Notes:
- © 2023 The Authors. Published by Elsevier B.V. Abstract reprinted with permission of Elsevier.
-
Authors:
- Mosleh, Araliya
- Meixedo, Andreia
- Ribeiro, Diogo
- Montenegro, Pedro
- Calçada, Rui
-
Conference:
- Transport Research Arena Conference (TRA Lisbon 2022)
- Location: Lisbon , Portugal
- Date: 2022-11-14 to 2022-11-17
- Publication Date: 2023
Language
- English
Media Info
- Media Type: Web
- Features: Figures; Illustrations; References; Tables;
- Pagination: pp 4199-4206
-
Serial:
- Transportation Research Procedia
- Volume: 72
- Issue Number: 0
- Publisher: Elsevier
- ISSN: 2352-1465
- Serial URL: http://www.sciencedirect.com/science/journal/23521465/
-
Publication flags:
Open Access (libre)
Subject/Index Terms
- TRT Terms: Flaw detection; Machine learning; Mechanical acceleration; Railroad wheelsets; Rolling contact; Sensors
- Subject Areas: Data and Information Technology; Maintenance and Preservation; Railroads; Vehicles and Equipment;
Filing Info
- Accession Number: 01904479
- Record Type: Publication
- Files: TRIS
- Created Date: Jan 12 2024 11:27AM