GPUMD-NEP NEMD Analysis

This repository contains GPUMD non-equilibrium molecular dynamics (NEMD) simulation outputs and a Jupyter notebook for analyzing interfacial thermal conductance (ITC) from the compute.out and model.xyz files.

The current dataset is organized as repeated simulation folders at 300 K. Each numbered folder contains one GPUMD run and the associated analysis inputs/outputs.

Repository Contents

.
├── 1/ ... 21/                  # Numbered GPUMD/NEMD simulation folders
│   ├── run.in                  # GPUMD input file
│   ├── nep.txt                 # NEP potential file
│   ├── model.xyz               # Atomic model and lattice vectors
│   ├── compute.out             # Temperature groups, Ein, and Eout
│   ├── thermo.out              # Thermodynamic output
│   ├── shc.out                 # Spectral heat current output
│   ├── movie.xyz               # Trajectory output
│   └── neighbor.out            # Neighbor information
├── ICT.txt                     # Existing DT/ITC summary
├── ICT_notebook.txt            # DT/ITC summary written by the notebook
└── project_introduction_itc.ipynb

Notebook

Open and run:

jupyter notebook project_introduction_itc.ipynb

or:

jupyter lab project_introduction_itc.ipynb

The notebook:

• scans numbered folders from 1 to 39;
• skips folders that are missing compute.out or model.xyz;
• reads temperature groups, Ein, and Eout from compute.out;
• reads the in-plane area from the Lattice field in model.xyz;
• computes DT and ITC for each valid folder;
• prints the DT/ITC table, mean, standard deviation, and SEM in one cell;
• plots the averaged temperature profile plus Ein/Eout for a selected folder.

To choose which folder to plot, edit this line in the plotting cell:

plot_folder = "2"

Change "2" to any folder listed in the notebook output, for example "1", "13", or "21".

Analysis Notes

The notebook assumes the same column layout as the included compute.out files:

• all columns except the final two are temperature-group columns;
• the final two columns are Ein and Eout;
• DT is computed from adjacent averaged temperature groups after averaging the second half of the trajectory;
• ITC is reported in MW/m^2/K.

The notebook writes its summary to ICT_notebook.txt so the original ICT.txt is not overwritten.

Python Requirements

The analysis notebook uses:

numpy
matplotlib
jupyter

Example setup:

python -m pip install numpy matplotlib jupyter

Uploading to GitHub

The target repository is:

https://github.com/wenlingren/GPUMD-NEP.git

Typical upload commands from this directory are:

git init
git add README.md project_introduction_itc.ipynb ICT.txt ICT_notebook.txt 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
git commit -m "Add GPUMD NEMD dataset and analysis notebook"
git branch -M main
git remote add origin https://github.com/wenlingren/GPUMD-NEP.git
git push -u origin main

Before pushing, check the repository size and file sizes. GitHub blocks files larger than 100 MB and recommends keeping repositories small. This project is several hundred MB, so if any files exceed GitHub limits, use Git LFS or exclude large trajectory/output files.

Recommended files to exclude from Git:

.ipynb_checkpoints/
.matplotlib/
__pycache__/

Citation

If this data or notebook is used in a publication or shared analysis, cite GPUMD and the relevant NEP potential/source data used to generate the simulations.