Development and Application of Subgrid Upscaling

University Of Texas At Austin, Austin TX

Investigators

Abstract

The investigator considers a physical system representing a porous medium, such as a groundwater aquifer or a petroleum reservoir. The movement of, e.g., a contaminant or oil over several kilometers is strongly influenced by the properties of the intervening rock. What appears to us to be bulk fluid flow is actually a complex process of fluid flowing through high and low permeability regions. To simulate the flow, one generally uses a coarse computational grid. In subgrid upscaling, one incorporates fine-scale processes within the coarse grid cells through a local and easily computed subgrid technique. The projects objective is to obtain provably improved accuracy and computational efficiency. Moreover, since this technique provides a good and easily computed approximation to the fine-scale movement of the fluids, the technique will be adapted so that the full fine-scale problem itself can be solved in significantly less time. One post-doc, two graduate students, and one undergraduate student will be involved in the research, and educated in the interdisciplinary environment. The work will be disseminated at interdisciplinary meetings and in the mathematics, groundwater, and reservoir simulation literature to maximize its impact. The contamination of groundwater is one of the most serious environmental problems facing the nation. Improvements to the predictive capabilities of groundwater simulations in heterogeneous formations would have a wide-ranging impact. The engineering of petroleum and natural gas production uses heavily subsurface simulation technologies as well. The subgrid upscaling method has shown great promise, and this proposal will further its development while promoting important educational and societal objectives.