Understand grids Module

Author: Zakariya Abugrin | Date: May 2025

Introduction

In this tutorial, we will see how the grids module is an essential part of any model in reservoirflow which allows to create a grid model. Currently, only RegularCartesian grid models are supported. In the near future, Radial and IrregularCartesian models will be added.

import reservoirflow as rf

print(rf.__version__)

0.1.0b3

Create 1D Model

We can use RegularCartesian from grids module to build a gird, see the documentation.

Grid model dimensionality can be defined based on nx, ny, nz arguments as following:

# Create a 1D regular cartesian grid:
grid = rf.grids.RegularCartesian(
    nx=4,
    ny=1,
    nz=1,
    dx=200,
    dy=200,
    dz=40,
    phi=0.27,
    kx=270,
    ky=200,
    kz=10,
    comp=1e-6,
    unit="field",
)

# Show the grid model with cells id
a = grid.show(
    label="id",
    boundary=False,
    static=True,
    notebook=True,
    azimuth=25,
)

# Show the grid model with cells id
grid.show(
    label="coords",
    boundary=False,
    static=True,
    notebook=True,
    azimuth=25,
)

Cells id

Cells id are arranged in natural order starting from the bottom left corner and can be accessed in several ways:

• attribute cells_id provides cells_id without boundary cells as flat array:

grid.cells_id

array([1, 2, 3, 4])

• more options can be found under get_cells_id() method:

cells_id = grid.get_cells_id(boundary=False, fshape=True, fmt="array")
cells_id.shape  # -> (nz, ny, nx)

(1, 1, 4)

Note: fshape (flow shape) argument provides the cells_id array as flow shape (nz, ny, nx) and boundary argument can be used to include or exclude boundary cells.

Using nz=3, but actually nz=5 (i.e. 5 layers) in total including upper and lower boundaries. If you do not want to deal with boundary cells just set boundary=False.

Comments 💬

Feel free to make a comment, ask a question, or share your opinion about this specific content. Please keep in mind the Commenting Guidelines ⚖.