Data Analytics for the of RCF Damages on the Dutch High Speed Line
AA Nunez (TU Delft - Railway Engineering)
Arjen Zoeteman (TU Delft - Integral Design & Management)
A. R M (Rogier) Wolfert (TU Delft - Integral Design & Management, TU Delft - Materials and Environment)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
During a typical measurement campaign, lots of temporal and spatial data can be gathered regarding the condition of the rail. This paper proposes two approaches that make use of data analytics techniques to find causes of rolling contact fatigue (RCF) damages. The first approach, named ‘bottom-up approach’, determines the influencing factors regarding RCF based on the worst
affected areas (hotspots). The second approach, called ‘top-down approach’, determines the influencing factors based on the condition of the whole track. The approaches use correlation analysis, clustering and similarity of parameters. To show the advantage of the approaches, they have been used for the study of the Dutch High Speed Line (HSL). The results indicates that severe RCF defects
occurred only under two very specific conditions. First, in specific curves where one type of train was driving under high tractive efforts and large cant excess through curves. Second, at the entry zones of the HSL where voltage locks are present, the same type of trains’ low driving speeds result in driving without cant excess/deficiency (theoretical cant). The conditions suggest that structurally driving below design speed on a high-speed track can be a cause of rail damages.