Refinable-precision in mesh compression for upscaling and upgridding in reservoir simulation with HexaShrink

@inproceedings{BouardRM2019,
 abstract = { A reservoir mesh in Corner Point Grid format may contain up to millions of cells. It accurately models geological fields, in order to obtain realistic (yet computationally expensive) flow simulations. To gain in computing time, upscaling methods have been proposed to decrease the number of cells. Such approaches generally end up with several models of different precision, built iteratively, in a trade-off between computing complexity and model simplification. However storage and transmission of the generated data remain a major challenge. In Peyrot et al. (2019), a multiresolution scheme for structured hexahedral volume meshes was developed, allowing efficient mesh visualization and storage. Our objective is now to evaluate whether this multiresolution framework is also suitable for flow simulations at different scales, obtained by partial decoding with a refinable precision of the compressed mesh structure. In this paper, promising results on a two-phase flow model are presented, suggesting significant speed improvements with consistent approximate simulation results. },
 author = { Bouard, Lauriane AND Duval, Laurent AND Preux, Christophe AND Payan, F{\'e}d{\'e}ric AND Antonini, Marc },
 booktitle = { 2019 Ring Meeting },
 publisher = { ASGA },
 title = { Refinable-precision in mesh compression for upscaling and upgridding in reservoir simulation with HexaShrink },
 year = { 2019 }
}