Institute of Solid State Physics


 Physics-Informed Deep Learning for Reactive Gas Transport in Paper

Simulating the transport of gasses such as organic volatile molecules through porous media is challenging when the complex geometry of real systems has to be considered, especially when transport is accompanied by sorption and/or chemical reactions. This complexity in geometry challenges traditional numerical solvers in their ability to capture the intricate details of the system in an accurate yet feasible manner.

The transport through paper is influenced by a combination of material properties, including the polarity of the organic compound, its diffusivity in the porous and solid phases of the paper, and the paper’s micro-structural features such as porosity and tortuosity. Physics-informed neural networks (PINNs) can successfully predict such material parameters and the subsequent transport through paper (DOI: 10.1007/s11242-022-01864-7).

Based on experimental migration data, this PINN approach shall be extended to two- and three-dimensional systems by incorporating microstructural information obtained via micro-CT measurements of paper into the framework of PINNs domain.

Contact: Karin Zojer



Impressum, Datenschutzerklärung