The differential settlement of subgrade may accelerate degradation of tracks, lower passenger comfort and increase derailment risk of trains, which is a troublesome to designers and infrastructure mangers. This paper aims at the numerical simulation of interaction between train, track, and subgrade and running properties of trains under the differential settlement of high-speed railway subgrade. A dynamic analysis model for train-ballasted track-subgrade-subsoil system is established by finite/infinite element method, and validated by experimental data and existing results. The comparison shows the method proposed in this paper is effective in simulating the train- ballasted track-subgrade interaction and riding properties of trains subjected to subgrade differential settlement.
The ballasted track, as a traditional track structure, has been widely used in high speed railways due to good elasticity, low cost, easy maintenance, and obvious absorption to noise. However, the differential settlement of subgrade is a troublesome to designers and infrastructure mangers (Kang, 2016), especially for the soft subsoil easy to generate the differential settlement. In 1997, the Swedish National Rail Administration opened a service with the X-2000 highspeed train along their West Coast Line between Göteborg and Malmö. Shortly after starting the service, excessive vibrations of the railway embankment and surrounding soil were detected at several soft soil areas when the train approaches around 200km/h (Madshus and Kaynia, 2000). Questions were raised about the running safety of the trains, degradation of the supporting track and subsoil, distortion of the embankment, fatigue failure in the rails. Actually, many factors can cause the differential settlement of subgrade and subsoil, such as soil condition, train cyclic loading, subgrade subsidence, and so on. However, once the settlement generates, it may accelerate degradation of tracks, lower passenger comfort and increase derailment risk of trains, especially in high-speed railways.
Issues related to track deformation due to the poor performance of the earthworks are difficult to detect and complex to solve because the subgrade and the subsoil are generally inaccessible. Taufan and Louis (2016) described some current empirical ballast settlement models, and evaluated them using experimental data generated using the Southampton Railway Testing Facility. Due to the limitation of empirical model, the model test and the numerical methods have been applied. Zou (2011) designed a 1:1 model test to study the influence of differential subgrade settlement on the ballasted track. Sol-Sanchez and Pirozzolo (2016) studied the mechanical performance of different configurations for the railway track section by a laboratory approach. Tatsuya (2014) proposed an analytical procedure with iterative calculation by linear finite element analysis to estimate the cyclic plastic deformation of railroad ballast under repeated moving-wheel loads. Paixão and Fortunato (2015) used a FEM method to simulate different scenarios in normal railway lines and gave a parametric study. Zhang (2016) proposed a two-dimensional ballasted railway tracks model utilizing the discrete element method with PFC, and analyzed the dynamic characteristics of the concrete sleeper, clustered ballast stones and the silty clay subgrade under irregular vibration levels caused by train passing the track.
