Reservoir simulation from Python¶

PyJutulDarcy wraps JutulDarcy 0.4, a differentiable multiphase reservoir simulator. Run industry DATA files or build a model with Python and NumPy; Julia performs the simulation and adjoint calculations.

Install¶

Requires Python 3.10 or newer.

python -m pip install jutuldarcy

The first import jutuldarcy installs Julia 1.11 or newer and the registered Julia dependencies automatically. Allow several minutes and network access for this step. Later imports reuse the installation; the first simulation also compiles code for your model.

Run a DATA file¶

import jutuldarcy as jd

output = jd.simulate_data_file("MODEL.DATA", convert=True)
pressure = output["STATES"][-1]["Pressure"]  # NumPy array, Pa
days = output["DAYS"]

Start with the executed SPE9 example to try a public dataset, or build a model from scratch. The user guide explains indexing, results, backends and plotting; the API reference lists the Python entry points.

Learn more in the Julia documentation¶

The Python interface exposes a subset of the full simulator. Use the Julia manual for the physics and complete keyword options:

First model and high-level API.
Input formats and supported physical systems.
Examples, covering CO₂ storage, geothermal, compositional flow and history matching.

For scientific use, cite the JutulDarcy paper.