Suche nach Personen

plus im Publikationsserver
plus bei BASE
plus bei Google Scholar

Daten exportieren


Estimating permeability of 3D micro-CT images by physics-informed CNNs based on DNS


Verfügbarkeit überprüfen

Gärttner, Stephan ; Alpak, Faruk O. ; Meier, Andreas ; Ray, Nadja ; Frank, Florian:
Estimating permeability of 3D micro-CT images by physics-informed CNNs based on DNS.
In: Computational geosciences : modeling, simulation and data analysis. 27 (2023). - S. 245-262.
ISSN 1420-0597 ; 1573-1499


Open Access
Text (PDF)
Verfügbar unter folgender Lizenz: Creative Commons: Attribution 4.0 International (CC BY 4.0) Creative Commons: Namensnennung (CC BY 4.0) .

Download (4MB) | Vorschau
Volltext Link zum Volltext (externe URL):


In recent years, convolutional neural networks (CNNs) have experienced an increasing interest in their ability to perform a fast approximation of effective hydrodynamic parameters in porous media research and applications. This paper presents a novel methodology for permeability prediction from micro-CT scans of geological rock samples. The training data set for CNNs dedicated to permeability prediction consists of permeability labels that are typically generated by classical lattice Boltzmann methods (LBM) that simulate the flow through the pore space of the segmented image data. We instead perform direct numerical simulation (DNS) by solving the stationary Stokes equation in an efficient and distributed-parallel manner. As such, we circumvent the convergence issues of LBM that frequently are observed on complex pore geometries, and therefore, improve the generality and accuracy of our training data set. Using the DNS-computed permeabilities, a physics-informed CNN (PhyCNN) is trained by additionally providing a tailored characteristic quantity of the pore space. More precisely, by exploiting the connection to flow problems on a graph representation of the pore space, additional information about confined structures is provided to the network in terms of the maximum flow value, which is the key innovative component of our workflow. The robustness of this approach is reflected by very high prediction accuracy, which is observed for a variety of sandstone samples from archetypal rock formations.

Weitere Angaben

Sprache des Eintrags:Englisch
Institutionen der Universität:Mathematisch-Geographische Fakultät > Mathematik > Mathematisches Institut für Maschinelles Lernen und Data Science (MIDS)
Mathematisch-Geographische Fakultät > Mathematik > Lehrstuhl für Geomatik und Geomathematik
DOI / URN / ID:10.1007/s10596-022-10184-0
Open Access: Freie Zugänglichkeit des Volltexts?:Ja
Verlag:Springer Nature
Die Zeitschrift ist nachgewiesen in:
Titel an der KU entstanden:Ja
Eingestellt am: 07. Mai 2024 13:53
Letzte Änderung: 10. Mai 2024 10:58
URL zu dieser Anzeige:
AnalyticsGoogle Scholar