Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 104
PROCEEDINGS OF THE SECOND INTERNATIONAL CONFERENCE ON RAILWAY TECHNOLOGY: RESEARCH, DEVELOPMENT AND MAINTENANCE
Edited by: J. Pombo
Paper 89

Predicting Track Geometry Exceedances by Using Digital Signal Processing

P. Joksimovic and G. van der Werf

Department for Mobility, ARCADIS, Amersfoort, The Netherlands

Full Bibliographic Reference for this paper
P. Joksimovic, G. van der Werf, "Predicting Track Geometry Exceedances by Using Digital Signal Processing", in J. Pombo, (Editor), "Proceedings of the Second International Conference on Railway Technology: Research, Development and Maintenance", Civil-Comp Press, Stirlingshire, UK, Paper 89, 2014. doi:10.4203/ccp.104.89
Keywords: rail track geometry, digital signal processing.

Summary
To run rail infrastructure effectively and at a high management level, reliable information about the tracks must be available. Providing data for this process requires systematic data collection by means of various monitoring, detection and measuring systems. With the development of technology, the quality and quantity of data has changed resulting in large amounts of unused data. The measurement of rail tracks are not an exception. By increasing measurement density and the introduction of new data types, contemporary measuring devices have improved insights into track conditions and behaviour patterns far beyond the limits that could have been imagined just a few years ago. This has however introduced new problems with data filtering, over-fitted data, and imbalanced datasets. This paper demonstrates how methods of digital signal processing could be applied to track geometry measurements in order to isolate potential critical track locations. This method also contributes to an automated process developed by ARCADIS, an artificial intelligence based decision support system for optimal track management database.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £65 +P&P)