nmerovingian / PINN-CV

Using PINN to predict cyclic voltammetry with knowledge of only boundary condition and diffusion law

PINN-CV

This is a code repository in company with "Predicting Voltammetry using Physics-Informed Neural Networks" published at The Journal of Physical Chemistry Letters

Requirements

Python 3.7 and above is suggested to run the program. The neural networks was developed and tested with Tensorflow 2.3. To install required packages, run

$ pip install -r requirement.txt

• 1D Semi-Infinite: 1D simulation of cyclic voltammetry with semi-infinite boundary condition
• 1D Thin-Layer: 1D simulation at thin layer cyclic voltammetry
• 2D Microband: 2D simulation cyclic voltammetry at microband electrode
• 2D Square Electrode: 2D simulation of cyclic voltammetry at a square electrode

Animated Simulation

Please refer to SimulationDynamics.gif

Issue Reports

Please report any issues/bugs of the code in the discussion forum of the repository or contact the corresponding author of the paper

Cite

To cite, please refer to: J. Phys. Chem. Lett. 2022, 13, 2, 536–54

Since 2022, the authors have explored PINN's application in other fields of electrochemistry:

• Hydrodynamic voltammetry and channel electrode: Analyst, 2022,147, 1881-1891
• 3D cube electrode and best practices of PINN4Electrochemistry: J. Electroanal. Chem. 2022, 925, 116918
• Rotating disk electrode and edge effect: Anal. Chem. 2023, 95, 34, 12826–12834