An approach based on a single bogie-mounted sensor for estimating track vertical irregularity from in-service railway vehicles

The growing interest of the railway industry in utilising commercial trains as a cost-effective complement to conventional Track Recording Vehicles (TRVs) is widely recognised. Given that equipping commercial vehicles with sophisticated measurement systems comparable to those used in TRVs is impractical, measurement systems for commercial vehicles should ideally be based on a simplified set of sensors. This study introduces a methodology for estimating the longitudinal track level by employing the double integration of acceleration data collected from a single vertical accelerometer positioned centrally on the bogie frame. A strategy based on weighting functions has been developed to account for the filtering effect induced by the bogie wheelbase. For the assessment of track conditions, the standard deviation and the peak value of the longitudinal level, calculated within 100-metre segments, are employed as indicators. Linear regression models are established to correlate the bogie vertical displacement and track longitudinal level. The application of the methodology has been tested using experimental data obtaining satisfactory results that are summarised in the paper.