Dynamic Behaviour of Ballasted Railway Track with Special Reference to Transition Zones

Abstract

The poor performance of transition zones in railway tracks has long been a subject of concern for the rail infrastructure managers. These zones are the discontinuities along a railway line that are highly susceptible to differential settlement due to an abrupt variation in the support conditions over a short span. Consequently, these regions require frequent maintenance to ensure adequate levels of passenger safety and comfort. The rapid deterioration of track geometry in these zones is primarily ascribed to limited understanding of the underlying mechanism and scarcity of adequate tools to assess the severity of the potential issue. Therefore, a comprehensive evaluation of their behaviour is paramount to improve the design and ensure adequate service quality. With this objective, a novel methodology is developed which can predict the dynamic behaviour of the transition zones under train-induced repeated loading and assess the suitability of different countermeasures in improving the track performance.  

To this end, an integrated approach is first developed by combining track loading, resiliency, and settlement models to evaluate the transient and irrecoverable response of the substructure layers of a standard ballasted railway track. The track substructure layers (ballast, subballast, and subgrade) in this model are simulated as an array of lumped masses that are connected by elastic springs and viscous dampers. The irrecoverable response of the track is evaluated using the empirical settlement models for substructure layers. The accuracy of the method is validated by comparing the predicted results against the field investigation data reported in the literature. Subsequently, the practical applicability of the aforementioned method under different traffic loading and soil conditions is improved by replacing the empirical approach with a mechanistic approach, in which, plastic slider elements are employed to predict the inelastic deformation in the substructure layers. To validate the approach, the predicted results are compared with the in-situ measurements reported in the literature. A good agreement between the predicted results and the field data verified the accuracy of the novel geotechnical rheological track model. A parametric investigation is conducted which highlights the significant influence of axle load, train speed, and granular layer thickness, on the accumulated settlement in the track layers.

The novel geotechnical rheological track model is then applied to an open-track bridge transition by incorporating the inhomogeneous support conditions associated with the critical zone and the adequacy of different countermeasures to mitigate the differential track settlements is examined. The approach is successfully validated with published field data and predictions from the finite element (FE) analysis. The results revealed that an increase in axle load exacerbates the track geometry degradation problem. The results also show that the performance of transition zones with weak subgrade can be improved by increasing the granular layer thickness. Interpretation of the predicted differential settlement for different countermeasures exemplified the practical significance of the proposed methodology

Subsequently, the influence of principal stress rotation (PSR) experienced by the soil elements during a train passage is incorporated in the geotechnical rheological model. The results revealed that PSR causes significant cumulative deformation in the substructure layers, and disregarding it in the analysis leads to inaccurate predictions. Finally, the adequacy of using three-dimensional (3D) cellular geoinclusions to improve the performance of critical zones is investigated using the proposed methodology and FE analyses. A novel semi-empirical model is first developed to evaluate the magnitude of improvement provided by these inclusions under the 3D stress state. The proposed model is successfully validated against the experimental data. This model is then incorporated in the geotechnical rheological model and the effectiveness of 3D geoinclusions in improving the performance of an open track-bridge transition is investigated. The results show that the geoinclusions significantly reduce the magnitude of differential settlement and therefore, have a huge potential to be used in the transition zones to improve track performance.

The essential contribution of this thesis is that it provides reliable, practical, and adaptable techniques to assist the practising railway engineers in analysing the performance of various sections of ballasted railway tracks, identifying the most effective method to improve the track performance, planning the maintenance operations, and improving the design. The developed techniques are available in the form of MATLAB codes, which can readily be converted into an application that can be used by railway engineers. Nonetheless, the outcomes of this study have huge potential to influence the real-world design implications of track transition zones. The approaches developed in this study are original, simple yet elegant, and can enhance, if not fully replace, present complex track modelling procedures for anticipating the behaviour of critical zones and adopting appropriate mitigation strategies.