Ground improvement in transport geotechnics - from theory to practice
In order to meet the ever-increasing demand for public and freight mobility, concerted efforts are needed to improve transport efficiency, and ensure the maintenance cost of highways and railways to sustainable levels. Ballast and subgrade are major components of rail track infrastructure, and understanding their integrated behaviour as layered strata is of utmost importance. In order to investigate their performance, sophisticated numerical modeling techniques using the Discrete Element Method (DEM) and Finite Element Method (FEM) are commonly employed, but the micro-mechanics of particulate interactions are often taken for granted rather than studied in depth. The salient aspects of particle degradation and confining pressure are discussed through the use of advanced elasto-plastic constitutive models. Applications of DEM to study the behavior of coal-fouled ballast subjected to cyclic loading using a track process simulation apparatus are also presented. The DEM enabled the discrete nature of ballast aggregates to be modeled considering their realistic size and shape, as well as through the force chains developed at contacts.A radial consolidation model under cyclic loading is proposed to capture the behavior of soft clays subjected to cyclic loadingwhen radial drainage is allowed during the loading period. The effects of the cyclic stress history on the generation of excess pore pressure are considered in the proposed model. This model is verified against the results of large-scale triaxial tests. In addition, the study of the behavior of artificially compacted subgrade, is vital to avoid unwanted volume changes caused by moisture variation that can contribute to poor track performance. A model is proposed for assessing the performance of compacted subgrades by evaluating the small strain behavior, considering the soil matric suction, the water content and the compaction energy level. This paper discusses the stability of the transport infrastructure at small and large strains based on numerical methods supported through advanced constitutive algorithms and validated against large-scale laboratory data.