Detailseite
Projekt Druckansicht

Modeling of two-phase flow processes in strongly heterogeneous porous media using multi-rate mass transfer approaches

Fachliche Zuordnung Hydrogeologie, Hydrologie, Limnologie, Siedlungswasserwirtschaft, Wasserchemie, Integrierte Wasserressourcen-Bewirtschaftung
Förderung Förderung von 2011 bis 2015
Projektkennung Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 192984945
 
Erstellungsjahr 2016

Zusammenfassung der Projektergebnisse

Immiscible two phase flow processes in highly heterogeneous porous media, such as fractured rock, are important in many geotechnical applications, such as CO2 sequestration or oil recovery. In fractured rock classical modelling approaches are computationally intensive due to the strong contrast in the model parameters. In this project we derived upscaled two phase flow models on the macroscale, where the detailed fracture network is no longer described. In fractured rock the fractures are related to fast flow processes. Slow exchange of fluid takes place between the fractures and the rock matrix. For the upscaled model the fractured rock is divided into two zones. The fractures with the fast flow processes are the mobile zone and the rock matrix with the slow flow processes are the immobile zone. The upscaled flow model describes flow processes in the mobile zone only. The exchange processes between mobile and immobile zone are modelled with an additional sink-source term. This term is expanded in a way that the model becomes a multi-rate mass-transfer model for two-phase flow. With this modelling approach we derived two upscaled models on the macroscale. The first model is for oil recovery from fractured rock. This is an imbibition process, where oil as the nonwetting phase is displaced by water as the wetting phase. The flow in the fracture network is dominated by flow enforced by boundary conditions and the flow in the rock matrix is dominated by capillary counter-current flow. The second model is for CO2 storage in deep fractured rock. This is a drainage process, where brine as the wetting phase is displaced by supercritical CO2 as the nonwetting phase. The flow in the fracture network is dominated by viscous forces and by gravity. The flow in the rock matrix is dominated by gravity flow only. For both upscaled models, we developed strategies to predict the parameters for the upscaled model from the fluid and material parameters and information on structure. For this purpose we derived characteristic time scales for the different flow processes in the fractured rock. The characteristic time scale relates a flow velocity for each process to a distance. The time scales are needed to parameterize the upscaled models. Furthermore the time scales can be used to compare the different flow processes in the fractured rock in order to decide whether the upscaled model is applicable or needed for a given scenario. With this reasonableness check and the strategies to predict parameters the upscaled models can be used to predict two phase flow in fracture rock. We developed a numerical method for the upscaled models and implemented the resulting numerical models using Matlab. To analyze the applicability and the validity of the upscaled models, we simulated CO2 sequestration and oil recovery with the upscaled models as well as with a classical modelling approach numerically and compared the results. For the classical modelling approach on a two dimensional fracture network we used the simulator Dumux by Flemisch et al. (2011). We found a good match between our upscaled models and the classical modelling approach, when we applied the methods for parameterization that we developed during the project. The numerical simulations with the upscaled models are much faster than the detailed simulations. For the scenario of oil recovery from fracture rock the speedup is more than factor 1600.

Projektbezogene Publikationen (Auswahl)

 
 

Zusatzinformationen

Textvergrößerung und Kontrastanpassung