The python code TrainPANACHE.py trains the physics-based neural network model for constituent steps in cyclic adsorption processes. The code follows the methodology proposed in Physics-based neural networks for simulation and synthesis of cyclic adsorption processes (https://doi.org/10.1021/acs.iecr.1c04731). As an example, relevant training data is provided here to train blowdown step neural network using this code.

1. TrainPANACHE.py: Trains physics-based neural networks for constituent steps.
2. trainfcn.m: Parser function loads train_data.mat and generates train_ads.mat file required for running TrainPANACHE.py.
3. train_data.mat: .mat data file containing blowdown step spatiotemporal solutions of all state variables.
4. train_ads.mat: .mat file containing training data for neural network training.

The following dependencies are required for the proper execution of this program.

1. MATLAB version 2019b onwards [required]
2. Python 3 [required]
3. Tensorflow v1.15 (GPU) [required]

1. Clone the full software package from the GitHub server into the preferred installation directory using:

git clone https://github.com/ArvindRajendran/PANACHE.git

1. Run trainfcn.m (with train_data.mat in the same directory) in MATLAB to generate train_ads.mat.
2. Run TrainPANACHE.ipyb (with train_ads.mat in the same directory) in Python notebook 3.
3. Save the weights and biases of the trained model for subsequent use in model predictions.

@article{Subraveti2022,
title = {Physics-based neural networks for simulation and synthesis of cyclic adsorption processes},
author = {Sai Gokul Subraveti and Zukui Li and Vinay Prasad and Arvind Rajendran},
journal = {Industrial & Engineering Chemistry Research},
year = {2022},
doi = {10.1021/acs.iecr.1c04731}
}

• Sai Gokul Subraveti ([email protected])

• Prof. Dr. Arvind Rajendran ([email protected])
• Prof. Dr. Vinay Prasad ([email protected])
• Prof. Dr. Zukui Li ([email protected])

Copyright (C) 2022 Arvind Rajendran

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see https://www.gnu.org/licenses/.