Estimation of lateral and cross alignment in a railway track based on vehicle dynamics measurements

Monitoring the condition of the infrastructure based on measurements coming from in-service trains is a recent trend in the policies of railway infrastructure managers and is seen as a major factor to reduce maintenance costs retaining the present levels of passengers’ ride comfort and availability of the lines. Aim of this paper is to develop and assess model-based methods for the identification of geometric track irregularities from acceleration measurements taken on-board vehicles travelling on the track. In particular, the paper focusses on the identification of lateral alignment irregularities, a yet unsolved problem which turns out to be much more demanding compared to the identification of longitudinal level irregularities due to the relative wheel-rail motion in lateral direction which occurs on account of the peculiar geometry of wheel-rail contact. The paper presents three different approaches to the estimation of the lateral and cross-level track irregularities, one defined in the frequency domain and two in the time domain. For each method, an assessment is performed based on numerical experiments in which virtual measurements from vehicle-mounted sensors are obtained by means of numerical simulations performed using a multi-body model of a railway vehicle. For the method defined in the frequency domain, a first application to line measurements is presented.