picture of the institute with institute logo
homeicon uni sucheicon search siteicon sitemap kontakticon contact
unilogo Universität Stuttgart
Institute of Hydraulic Engineering

Research: Dept. of Hydromechanics and Modeling of Hydrosystems

german-icon print view
 
Mechanics of Partially Saturated Soils
Subproject III: Modelling of Non-Isothermal Gas-Water Flow and Transport Processes in Cohesive Soils
Project manager:Prof. Dr.-Ing. Rainer Helmig
Deputy:Dr.-Ing. Holger Class
Research assistants:Dipl.-Geoökol. Sandra Freiboth
Duration:1.12.2001 - 30.11.2007
Funding:externer Link German Research Foundation (DFG)
Project Partners:here: DFG Research Group Mechanics of Partially Saturated Soils
Publications: Link

Abstract:

The objective of this subproject is to develop methods for the numerical modelling of non-isothermal two-phase two-component flow and transport processes in cohesive soils, allowing for shrinkage and swelling processes. With the assumption of a rigid soil skeleton, only the influence of skeleton alterations on the hydraulic properties, such as porosity or permeability, is taken into account. The considered two-phase -- two-component model comprises the fluid phases water and gas and the components water (liquid and gaseous) and air (as gas and dissolved in water). Such models are, for example, required to describe the processes during the desiccation of a mineral liner under a waste-diposal site and are of great importance in investigating the permanent reliability of mineral liners. For this study, close cordination with experiments and laboratory tests is needed to obtain constitutive relations as well as soil and flow properties.

The first step is the further development of methods to numerically simulate the desiccation until a critical saturation is reached at which cracks occur. Therefore, a method is conceived which allows us to determine significant parameters and model-input variables by comparison with experiments (see subproject IV Meißner). The main point of interest here is the application of inverse modelling. Additionally, the influence of small-scale heterogeneities, load and shrinkage processes on the simulation results is investigated. Furthermore, the validity of Darcy's Law for such small permeabilities has to be verified. For the efficient simulation of coupled non-linear proceses, parallelisation techniques and multigrid methods have to be improved. An aim of the second project phase is to couple the developed model with a "Stoffmodell" (see subproject IV Meißner) so that skeleton alterations such as cracks are taken into account.