Trained Models on Synthetic Permeability Fields, 3+1 Data Points

Dataset

DOI

Models are trained with Heat Plume Prediction. Steps 1 and 3 of LGCNN (Local Global Convolutional Neural Network) are separate, step 2 is a numerical solver that does not require any trained model. The vanilla UNet can be applied directly end-to-end, just does not give very good results.

For inference follow the guidelines of Heat Plume Prediction and applied all 3 steps/models sequentially to your input data.

Based on raw data from https://doi.org/10.18419/darus-5063.

Please follow the instructions in README.md to prepare the data for training.

Identifier
DOI	https://doi.org/10.18419/DARUS-5080
Metadata Access	https://darus.uni-stuttgart.de/oai?verb=GetRecord&metadataPrefix=oai_datacite&identifier=doi:10.18419/DARUS-5080

Provenance
Creator	Pelzer, Julia
Publisher	DaRUS
Contributor	Pelzer, Julia; Schulte, Miriam
Publication Year	2025
Funding Reference	DFG EXC 2075 - 390740016
Rights	CC BY 4.0; info:eu-repo/semantics/openAccess; http://creativecommons.org/licenses/by/4.0
OpenAccess	true
Contact	Pelzer, Julia (University of Stuttgart); Schulte, Miriam (University of Stuttgart)

Representation
Resource Type	Dataset
Format	application/zip
Size	165252495; 141512697; 731643904
Version	1.0
Discipline	Chemistry; Construction Engineering and Architecture; Earth and Environmental Science; Engineering; Engineering Sciences; Environmental Research; Geosciences; Natural Sciences