File(s) not publicly available

Rail short-wavelength irregularity identification based on wheel-rail impact response measurements and simulations

conference contribution
posted on 06.12.2017, 00:00 authored by Yan SunYan Sun, Mitchell McclanachanMitchell Mcclanachan, A Wilson, Sakdirat KaewunruenSakdirat Kaewunruen, M Kerr, Colin ColeColin Cole
Long and short-wavelength dip defects unavoidably exist on railway track top surface. Several manual measurement methods and AK track measurement car were used to measure and to identify these types of defects in a selected track section. The paper considers rail dip irregularities in detail when measured, the dip can be considered as two components. The first one is the large, long dip profile, which can be measured by the survey leveling. The other is a small short dip profile, which is superimposed on the large profile and can be measured using the dip gauge. The corresponding measurement and processed data from AK car were analysed and compared with the manual measurements. It has been found that the severities as indicated by the AK car for the short-wavelength dips in the bandwidth of 0.1~1.0m were in agreement with those measured by the dip gauge. Simulation analyses were also carried out in order to fully understand wheel-rail impact behaviours due to dip defects using the VAMPIRE package and the CRE-3D VTSD model. There was good agreement between the simulation results and the measurements from the AK car, and the simulations show that the impact accelerations are principally caused by the small dip profile.


Category 4 - CRC Research Income


Start Page


End Page


Number of Pages


Start Date





Shanghai, China


China Railway Publishing House

Place of Publication

Beijing, China

Peer Reviewed


Open Access


External Author Affiliations

Centre for Railway Engineering; Not affiliated to a Research Institute; Rail Corporation New South Wales; Rail Innovation Australia Pty Ltd;

Era Eligible


Name of Conference

International Heavy Haul Association. Conference

Usage metrics