GNNPCSAFT Project

The project focuses on using Graph Neural Networks (GNN) to estimate the pure-component parameters of the Equation of State PC-SAFT.

Currently, the model takes into account the hard-chain, dispersive, and associative terms of PC-SAFT. Future work on polar and ionic terms is being studied.

Use cases of this package are demonstrated in Jupyter Notebooks:

• compare.ipynb (Open in Colab): comparison of the performance of trained models
• training.ipynb (Open in Colab): notebook for model training
• tuning.ipynb (Open in Colab): notebook for hyperparameter tuning

Model checkpoints can be found at Hugging Face.

Implementations with GNNPCSAFT:

• GNNPCSAFT CLI
• GNNPCSAFT APP
• GNNPCSAFT MCP
• GNNPCSAFT Webapp
• GNNPCSAFT Chat