Keywords: railway infrastructure, substructure condition, railway turnouts, track geometry, longitudinal level, root mean square, wavelength filtering, maintenance assessment.

This paper investigates the applicability of track geometry–based indicators for the assessment of substructure-related degradation in railway turnouts. While spectral analysis methods such as power spectral density have shown promising results for extended plain line sections, their application to turnouts is limited by the required analysis window length and the resulting influence of adjacent track sections. To address this limitation, a localized indicator based on the root mean square value of the longitudinal level signal is proposed. Prior to its computation, the signal is filtered to isolate long-wavelength components associated with substructure behaviour. The indicator is evaluated using repeated track geometry measurements from selected turnouts and is assessed through both cross-sectional comparisons and temporal analyses in relation to maintenance interventions. The results indicate that the filtered root mean square value is sensitive to substructure-related changes while showing limited response to ballast-related maintenance activities. This suggests that the proposed approach provides a suitable diagnostic support tool for turnout-scale substructure assessment based on existing track geometry data.

download the full-text of this paper (PDF, 13 pages, 575 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description