This repo is developed from mechanoChemML library. It is specialized to reproduce results from our publication (Under-review, available on request):

M. Duschenes, S. Srivastava and K. Garikipati (2022), Numerical analysis of non-local calculus on finite weighted graphs, with application to reduced-order modelling of dynamical systems

The original readme from the mechanoChemML library is attached at the end of this file. Same license as the mechanoChemML library is used for this repo.

Installation

1. Download the repo

2. use conda to create a virtual env

3. pip3 install -r requirements.txt

Examples

Example folders:

1. diffusion Allen-Cahn dynamics example. Run with python main.py
2. microstructures Coupled Cahn-Hilliard, gradient elasticity example. Run with python main.py settings.prm
3. scaling Modified Taylor series error analysis scaling example. Run with python main.py

mechanoChemML library

Developed by the Computational Physics Group at the University of Michigan. This library was developed within the mechoChemML library framework.

List of contributors (alphabetical order):

• Arjun Sundararajan
• Elizabeth Livingston
• Greg Teichert
• Jamie Holber
• Matt Duschenes
• Mostafa Faghih Shojaei
• Siddhartha Srivastava
• Xiaoxuan Zhang
• Zhenlin Wang
• Krishna Garikipati

Overview

mechanoChemML is a machine learning software library for computational materials physics. It is designed to function as an interface between platforms that are widely used for scientific machine learning on one hand, and others for solution of partial differential equations-based models of physics. Of special interest here, and the focus of mechanoChemML, are applications to computational materials physics. These typically feature the coupled solution of material transport, reaction, phase transformation, mechanics, heat transport and electrochemistry. mechanoChemML is available on PyPi at https://pypi.org/project/mechanoChemML/.

License

BSD 4-Clause license. Please see the file LICENSE for details.

Installation

One can follow the installation instruction at https://mechanochemml.readthedocs.io/en/latest/installation.html to install the mechanoChemML library.

Documentation and usage

The documentation of this library is available at https://mechanochemml.readthedocs.io/en/latest/index.html, where one can find instructions of using the provided classes, functions, and workflows provided by the mechanoChemML library.

Acknowledgements

This code has been developed under the support of the following:

• Toyota Research Institute, Award #849910 "Computational framework for data-driven, predictive, multi-scale and multi-physics modeling of battery materials"

Referencing this code

If you write a paper using results obtained with the help of this code, please consider citing following papers:

• M. Duschenes, & K. Garikipati (2021), Reduced order models from computed states of physical systems using non-local calculus on finite weighted graphs. arXiv preprint arXiv:2105.01740.
• X. Zhang, G.H. Teichert, Z. Wang, M. Duschenes, S. Srivastava, A. Sunderarajan, E. Livingston, J. Holber, M. Shojaei, K. Garikipati (2021), mechanoChemML: A software library for machine learning in computational materials physics, arXiv preprint arXiv:2112.04960.