Modify the grid#

Change the grid type, model dimensions, and spatial refinement, then inspect the generated deck before running an expensive simulation.

Configure a Cartesian grid#

Copy the SPE11B configuration and modify these variables:

grid = "cartesian"
dims = [8400.0, 1.0, 1200.0]
x_n = [420]
y_n = [1]
z_n = [60]
cartesian

Uses one cell count in each direction.

tensor

Uses refinement arrays to divide each direction into regions.

corner-point

Uses lateral refinement arrays and an 11-entry or 18-entry z_n array for the geological levels.

Generate and inspect the deck#

pyopmspe11 -i modified_grid.toml -o modified_grid -m deck

Inspect the generated grid before running Flow. When ready, reuse the existing deck and generate benchmark data:

pyopmspe11 -i modified_grid.toml -o modified_grid -m flow_data

Result#

Facies in a generated SPE11 corner-point grid

Facies generated for an SPE11 corner-point grid.#

Next#

Continue with Compare results. See Corner-point grids with 11 and 18 levels for a comparison of 11-level and 18-level corner-point grids, and use Configuration reference for exact variable definitions.