Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Development of a wayside monitoring system for rail wheel defect detection using thermography and dynamic measurements
Thesis (MEng)--Stellenbosch University, 2025.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
Stellenbosch : Stellenbosch University
2025
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613947333443585 |
|---|---|
| access_status_str | Open Access |
| author | Lubbe, Enrico |
| author2 | Bekker, Annie |
| author_browse | Bekker, Annie Lubbe, Enrico |
| author_facet | Bekker, Annie Lubbe, Enrico |
| author_sort | Lubbe, Enrico |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (MEng)--Stellenbosch University, 2025. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/134682 |
| institution | Stellenbosch University (South Africa) |
| language | English |
| last_indexed | 2026-06-10T12:44:14.442Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/134682 Development of a wayside monitoring system for rail wheel defect detection using thermography and dynamic measurements Lubbe, Enrico Bekker, Annie Stellenbosch University. Faculty of Engineering. Dept. of Mechanical and Mechatronic Engineering. Railroad trains -- Wheels -- Maintenance and repair Railroad engineering Thermography Machinery -- Monitoring Infrared technology Thesis (MEng)--Stellenbosch University, 2025. Lubbe, E. 2025. Development of a Wayside Monitoring System for Rail Wheel Defect Detection Using Thermography and Dynamic Measurements. Unpublished masters thesis. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/87b5c51d-bf32-47d9-aefb-b672fbad57cb ENGLISH ABTRACT: This thesis develops and evaluates a wayside monitoring system for rail-wheel tread defect detection using thermography and conventional dynamic measurements (rail-mounted accelerometer). The work had two main objectives: (1) to design and validate a thermography wayside monitoring system capable of in-situ detection and geometric characterisation of wheel-tread defects under depot conditions, and (2) to benchmark thermography against an accelerometer-based reference to quantify detection performance, localisation, speed sensitivity, environmental robustness, processing time and cost. An experimental campaign was conducted on the X’Trapolis Mega train sets within a depot setting. Based on an initial thermography study, it was observed that wheel tread defects produce a thermal contrast that can be extracted. This observation informed the development of a defect length estimation method, which determines the defect length and compares it with a manual inspection measurement to assess accuracy. The thermography wayside system comprises a strategically positioned thermal camera that captures approximately one and a half revolutions of the wheel. Furthermore, camera calibration was conducted to correct lens distortions and generate a spatial map of the visible rail section. This map enables the system to associate defect locations with the closest calibration reference point, facilitating the application of the length estimation algorithm through back-projection of the points. A conventional wayside monitoring system was implemented utilising an accelerometer-based approach. The system consists of five rail-foot-mounted uniaxial sensors complemented by two-point lasers for axle counting and timing. The signals were analysed in the time domain through the use of statistical features, including Peak-to-Peak, root mean square, kurtosis, and crest factor, which were integrated into a joint metric for automated flagging. The key findings demonstrate complementary strengths. Both methods achieve high detection accuracy at depot speed (3 km/h). The accelerometer joint metric delivered F1-scores of 0.96 (3 km/h), 1.00 (10 km/h), and 0.96 (20 km/h). In comparison, thermography achieved perfect detection on the tested set at 3 km/h, revealing several surface defects missed in manual inspection. Thermographic length estimation produced typical errors of ±1.5 mm for defects less than 100 mm at speeds up to 10 km/h. Still, its performance degraded (mean error ≈ 8.7%) at 20 km/h due to motion blur and reprojection limitations. Processing and cost trade-offs are notable: accelerometers are rapid (≈ 75 ms per wheel), robust to ambient conditions, and cost-effective (≈ ZAR 264,790), whereas the thermography package is more ex-pensive (≈ ZAR 1.6 million), slower under the current semi-automated workflow (≈ 10 s per wheel), but uniquely provides direct visual severity measures and the ability to localise multiple defects per wheel. The thesis concludes that thermography is a feasible technique for wayside monitoring, as it enables the identification of wheel tread defects, which can be evaluated in accordance with regulatory standards. Furthermore, a hybrid approach is suggested, utilising accelerometers for continuous, automated screening and triggering thermographic inspections for quantitative severity evaluation and geometric analysis. To enhance readiness for field deployment, the study recommends automating thermal defect localisation, assessing more cost-effective cameras, and improving extrinsic calibration rigs to augment throughput and measurement capabilities. AFRIKAANSE OPSOMMING: Geen opsomming beskikbaar. Masters 2025-12-23T10:03:45Z 2025-12-23T10:03:45Z 2025-12 Thesis https://scholar.sun.ac.za/handle/10019.1/134682 en Stellenbosch University xvi, 122 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Railroad trains -- Wheels -- Maintenance and repair Railroad engineering Thermography Machinery -- Monitoring Infrared technology Lubbe, Enrico Development of a wayside monitoring system for rail wheel defect detection using thermography and dynamic measurements |
| title | Development of a wayside monitoring system for rail wheel defect detection using thermography and dynamic measurements |
| title_full | Development of a wayside monitoring system for rail wheel defect detection using thermography and dynamic measurements |
| title_fullStr | Development of a wayside monitoring system for rail wheel defect detection using thermography and dynamic measurements |
| title_full_unstemmed | Development of a wayside monitoring system for rail wheel defect detection using thermography and dynamic measurements |
| title_short | Development of a wayside monitoring system for rail wheel defect detection using thermography and dynamic measurements |
| title_sort | development of a wayside monitoring system for rail wheel defect detection using thermography and dynamic measurements |
| topic | Railroad trains -- Wheels -- Maintenance and repair Railroad engineering Thermography Machinery -- Monitoring Infrared technology |
| url | https://scholar.sun.ac.za/handle/10019.1/134682 |
| work_keys_str_mv | AT lubbeenrico developmentofawaysidemonitoringsystemforrailwheeldefectdetectionusingthermographyanddynamicmeasurements |