OPM Flow is a fully-implicit, black-oil simulator capable of running industry-standard simulation models. The simulator is implemented using automatic differentiation to enable rapid development of new fluid models.
Model formulation
- black-oil with dissolved gas and vaporized oil
- rock-dependent capillary and relative-permeability curves
- end-point scaling and hysteresis
- oil vaporization controls (VAPPARS)
EOR options
- Todd-Longstaff type polymer model with adsorption, dead-pore space, permeability reduction, and shear effects (Flow-polymer)
- extra component equation(s), such as a solvent model (Flow-solvent)
Description of geology
- rectilinear and fully-unstructured grid
- corner-point grids from Eclipse input, including fault and region multipliers, minpv/pinch, etc
Wells and controls
- bottom-hole pressure and surface/reservoir rate
- group controls
- shut/stop/open individual completions
- history-matching wells
Input/output
- general reader/parser for Eclipse input decks
- XML-based or simple text-format input of additional parameters
- flexible output of summary and restart files in Eclipse format
- logging to terminal and print file
Simulation technology
- fully-implicit in time
- two-point flux approximation in space with upstream-mobility weighting
- flexible assembly through the use of automatic differentiation
- block-structured linear solver with ILU0 preconditioner
- CPR preconditioner and multigrid solver
- adaptive step-size controls
OPMFlow in the MSO4SC project
In the MSO4SC project, the idea is to use OPM Flow as a pilot case to demonstrate OPM on the HPC/cloud infrastructure and to improve OPM Flow in the following ways:
- Deployability. Ensuring a good user experience installing/deploying the OPM software through providing packages, containers and other means.
- Usability. Improving ease of understanding of the software as well as providing documentation.
- Scalability. Ensuring that the software performs well for typical problems on HPC-cloud type infrastructure.
For more information about OPM, the OPM software, datasets and community, please visit the OPM web site.