دانلود رایگان مقاله مدل سازی مولتی فیزیکی جریان ترکیبی غیر همدما در شبکه تطبیقی

Abstract

A multi-physics framework for compositional and non-isothermal two-phase flow in a porous medium is presented which adapts locally the model complexity depending on the physical state. It is based on a sequential (IMPET) solution scheme. As a second adaptive strategy, the simulation grid can be refined locally. This may lead to a mesh with hanging nodes, which is treated by a multi-point flux approximation (MPFA) for improved flux representation. The two adaptivity concepts are employed to simulate a combined subsurface CO2 injection and geothermal application in an existing reservoir, the Tensleep formation.

5. Conclusions

This work combines two adaptive modeling strategies for compositional two-phase flow in porous media including non-isothermal effects: a multi-physics concept selects different numerical models locally depending on the physical processes at hand; an adaptive modification of the simulation grid is used to track features of interest locally with a refined grid while allowing a coarser mesh globally. For the latter, a multi-point flux approximation (MPFA) avoids errors in the flux representation around the hanging nodes that evolve through refinement. A large-scale example related to CO2 sequestration combined with geothermal application illustrates the potential and applicability of the presented framework.