GaPFlow

Gap-averaged flow simulations with Gaussian Process regression.

This code implements the solution of time-dependent lubrication problems as described in:

• Holey, H. et al., Tribology Letters 70 (2022)

The extension to atomistic-continuum multiscale simulations with Gaussian process (GP) surrogate models has been described in:

• Holey, H. et al., Science Advances 11, eadx4546 (2025)

The code uses µGrid for handling macroscale fields and tinygp as GP library. Molecular dynamics (MD) simulations run with LAMMPS through its Python interface. Elastic deformation is computed using ContactMechanics.

Installation

GaPFlow can be installed via

pip install GaPFlow

A serial build of LAMMPS is provided for most platforms which allows testing all of GaPFlow's functionality. For production simulations it is however recommended to build GaPFlow with parallel LAMMPS on your system. You need to have MPI installed (e.g. via apt install openmpi-bin libopenmpi-dev on Debian-based systems). To compile for your specific platform, run

pip install --no-binary GaPFlow GaPFlow[parallel]

You can check your installation by running gpf_info from the command line.

==========
GaPFlow
==========
Version:  ...

==========
LAMMPS
==========
Version: ...
Shared lib: <path-to-your-python-env>/lib/pythonX.Y/site-packages/GaPFlow/_vendor/lammps/liblammps_mpi[.so, .dylib]
MPI:  True
mpi4py:  True
packages:  ['EXTRA-FIX', 'MANYBODY', 'MOLECULE']

==========
muGrid
==========
Version:  ...
NetCDF4: True
MPI: False

We currently do not use parallel functionalities of µGrid, so MPI support is not required.

For installations from source, do not forget to initialize the LAMMPS submodule after cloning the repository

git submodule update --init

Minimal example

Simulation inputs are commonly provided in YAML files. A typical input file might look like this:

# examples/journal.yaml
options:
    output: data/journal
    write_freq: 10
grid:
    dx: 1.e-5
    dy: 1.
    Nx: 100
    Ny: 1
    xE: ['D', 'N', 'N']
    xW: ['D', 'N', 'N']
    yS: ['P', 'P', 'P']
    yN: ['P', 'P', 'P']
    xE_D: 877.7007
    xW_D: 877.7007
geometry:
    type: journal
    CR: 1.e-2
    eps: 0.7
    U: 0.1
    V: 0.
numerics:
    tol: 1e-9
    dt: 1e-10
    max_it: 200
properties:
    shear: 0.0794
    bulk: 0.
    EOS: DH
    P0: 101325
    rho0: 877.7007
    T0: 323.15
    C1: 3.5e10
    C2: 1.23

Note that this example uses fixed-form constitutive laws without GP surrogate models or MD data. More example input files can be found in the [examples] (examples/) directory.

The input files can be used to start a simulation from the command line

python -m GaPFlow -i my_input_file.yaml

or from a Python script

from GaPFlow.problem import Problem

myProblem = Problem.from_yaml('my_input_file.yaml')
myProblem.run()

Simulation output is stored under the location specified in the input file. After successful completion, you should find the following files.

• config.yml: A sanitized version of your simulation input.
• topo.nc: NetCDF file containing the gap height and gradients.
• sol.nc: NetCDF file containing the solution and stress fields.
• history.csv: Contains the time series of scalar quantities (step, Ekin, residual, ...)
• gp_[xz,yz,zz].csv (Optional): Contains the time series of GP hyperparameters, database size, etc.
• Xtrain.npy (Optional): Training data inputs
• Ytrain.npy (Optional): Training data observations
• Ytrain_err.npy (Optional): Training data observation error

The code comes with a few handy command line tools for visualizations like this one

which shows the transient solution of a 1D journal bearing with active learning of the constitutive behavior.

License

The GaPFlow project is distributed under the terms of the GNU General Public License version 2 (GPLv2).

The original GaPFlow source files under GaPFlow/ are licensed under the MIT License (see LICENSES/MIT.txt). This repository also contains third-party software that is not covered by the MIT License, in particular the LAMMPS distribution under lammps/ and Python bindings vendored under GaPFlow/_vendor/lammps/, which are licensed under GPLv2. The full text of the GPLv2 is included in the COPYING file at the repository root.

If you distribute a packaged artifact (for example an sdist or wheel) that includes the LAMMPS sources or the vendored LAMMPS Python bindings, the combined distribution is subject to the terms of the GPLv2. In that case, recipients must be granted the rights required by the GPLv2 and the COPYING file must be included in distributed artifacts. This does not change the MIT license that applies to the original GaPFlow sources; rather, when distributed together with GPLv2-covered files the resulting artifact must comply with GPLv2.

Funding

This work received funding from the German Research Foundation (DFG) through GRK 2450 and from the Alexander von Humboldt Foundation through a Feodor Lynen Fellowship.