A Second-Order Finite Volume Method for Field-Scale Reservoir Simulation

• DOI: 10.1007/s11242-023-01999-1
• Publication Date: 2023-08-18T15:58:48Z
• Authors: Anna Kvashchuk; Robert Klöfkorn; Tor Harald Sandve
• Keywords: Reservoir Simulation; Benchmark (surveying); Multiphase flow

Abstract Subsurface reservoirs are large complex systems. Reservoir flow models defined on grids that follow geology with relatively block sizes to make consistent simulations feasible. engineers rely established reservoir simulation software model fluid flow. Nevertheless, front position inaccuracies and smearing may cause significant errors it hard predict hydrocarbon production efficiency. We investigate higher-order methods reduce these undesired effects without refining the grid, thus making more accurate robust. For this paper, we implemented a second-order finite volume method linear programming (LP) reconstruction in open-source industry-grade simulator OPM Flow (part of open porous media initiative, OPM). benchmark against first-order full-scale cases standard coarse refined grids. prepared refined-grid synthetic an unstructured grid Norne field example. Our results confirm LP predicts positions as accurately for problems dominated by transport. These include water alternating gas scenario piston-type injection field. Moreover, study gains from CO $$_{2}$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msub> <mml:mrow /> <mml:mn>2</mml:mn> </mml:msub> </mml:math> full multi-phase complexity beyond observe relevant difference between first- cases. However, some configurations, overshadows methods.