In this thesis, a series of numerical models have been developed based on the finite element method to investigate the subsoil dynamic responses for the concrete slab track. A two-dimensional finite element coupled train-track model is developed. In this model, the track is modeled in plain strain conditions and infinite elements are implemented to prevent artificial wave reflections at the domain boundaries of the track model. The vehicle model is modeled as a spring-mass-damper system which represents the bogie, primary suspension and wheels. The second suspension is assumed insignificant in this research. The rail is discretized as a finite Timoshenko beam element. The train and track are coupled at the wheels-rail contact points, employing a linear contact theory.

The dynamic simulation is performed in the time domain using the implicit integration scheme. The subsequent dynamic step is done by varying train speed from 100 to 700 km/h. The effect of the magnitude of subsoil stiffness was investigated by varying the Young’s modulus of the isotropic elastic material of subsoil from 10 MPa for very soft soil to 180 MPa for stiff soil. The subsoil responses, including vertical displacements, stress changes and cycles, strain cycles and the radiation of the elastic strain energy are analyzed. articularly, the production and radiation of the elastic strain energies in the subsoil was the key factor in the analysis.

In this research, we examine two definitions of critical velocity: the critical velocity in the ground, which is related to energy radiation in the track and ground, is referred as vcr1,and the critical velocity in the rail caused by vibrations in beams
on an elastic Winkler foundation, which is referred as v
cr2. The critical velocity in the ground, vcr1, is identical to the Rayleigh wave velocity of the subsoil. Based on the numerical results, the so called “real critical velocity” is numerically visualized, and is referred as vcr3. With increasing train speed, three different zones: zone1, zone 2 and zone 3. The response in zone 3 for train speeds beyond v
cr3 can be understood as a post-critical behavior that might reflect erratic dissipation of elastic strain energies generated by a superfast train. These 3 zones are practically guided to understand the dynamic response of the track and subsoil at various train speeds.

In order to assess the dependency of the critical velocity, vcr3, on the subsoil stiffness for ballast track, supplementary analyses are performed. The same traintrack coupled model is also employed to perform the dynamic analysis. It is found that the critical velocity, vcr3, is not only dependent on the subsoil stiffness but also on the upper stiffness. However, a notable difference between ballast track and slab track is observed at critical velocity, vcr3.

This difference elucidates the effects of the elastic train energies in each type of track. The influence of multiple vehicle cars on critical velocity, vcr3, was also studied for the concrete slab track. The study consists of extending one car to ten cars. It is found that the real critical velocity, vcr3, is not influenced by multiple wheelsets.

Analysis of the radiation of elastic strain energy elucidates all the dynamic responses of displacements and stresses in the track and subsoil for the concrete slab track. The delayed arrival of the peak vertical displacement, the expanding hysteretic loop of the stress paths, and inclined distribution of the elastic strain energy at train speeds larger than vcr1, confirm retardation of energy radiation in the subsoil. At the critical velocity of vcr3, the overlapping and accumulated elastic strain energy results in maximal dynamic amplification. The wide spread of accumulated elastic strain energies adversely reduces dynamic amplification of vertical displacement.

With ever-increasing of train-speed, heavier axle load and large number of train passages, the subsoil layer is experiencing complicated stress paths at relatively
high stress levels. Such stresses cycles after thousand millions of train passage; the subsoil accumulates considerably plastic strains which can latter result in track settlement. In this regard, a simplified accumulation model which can capture an envelope of the maximum plastic strain is developed based on the shakedown concept.This model is implemented in three-dimensional finite element slab track model. The model was validated by comparing the predicted by comparing
predicted response with experimental results. It is shown that the predicted plastic strains in the subsoil and roadbed layers are in a reasonable good agreement with
measured results.