GEOSCAN Search Results: Fastlink

GEOSCAN Menu


TitleThree-dimensional structural geological modeling using graph neural networks
 
AuthorHillier, MORCID logo; Wellmann, FORCID logo; Brodaric, BORCID logo; de Kemp, EORCID logo; Schetselaar, EORCID logo
SourceMathematical Geosciences 2021 p. 1-25, https://doi.org/10.1007/s11004-021-09945-x Open Access logo Open Access
Image
Year2021
Alt SeriesNatural Resources Canada, Contribution Series 20200743
PublisherSpringer
Documentserial
Lang.English
Mediapaper; digital; on-line
File formatpdf; html
Subjectsstructural geology; Science and Technology; Nature and Environment; bedrock geology; structural features; models; modelling; geometric analyses; Methodology; Classification
Illustrationsschematic representations; 3-D diagrams; 3-D models; tables; models
ProgramOpen Geoscience
Released2021 06 30
AbstractThree-dimensional structural geomodels are increasingly being used for a wide variety of scientific and societal purposes. Most advanced methods for generating these models are implicit approaches, but they suffer limitations in the types of interpolation constraints permitted, which can lead to poor modeling in structurally complex settings. A geometric deep learning approach, using graph neural networks, is presented in this paper as an alternative to classical implicit interpolation that is driven by a learning through training paradigm. The graph neural network approach consists of a developed architecture utilizing unstructured meshes as graphs on which coupled implicit and discrete geological unit modeling is performed, with the latter treated as a classification problem. The architecture generates three-dimensional structural models constrained by scattered point data, sampling geological units and interfaces as well as planar and linear orientations. The modeling capacity of the architecture for representing geological structures is demonstrated from its application on two diverse case studies. The benefits of the approach are (1) its ability to provide an expressive framework for incorporating interpolation constraints using loss functions and (2) its capacity to deal with both continuous and discrete properties simultaneously. Furthermore, a framework is established for future research for which additional geological constraints can be integrated into the modeling process.
Summary(Plain Language Summary, not published)
A new 3D geological modelling framework using a geometric deep learning approach has been developed. Compared with existing geological modelling approaches the presented approach permits new forms of data and knowledge to be incorporated into generating 3D geological models, does not impose modelling assumptions that can compromise modelling, and produces reasonable models from noisy datasets. A Graph Neural Network (GNN) is developed and applied to two diverse geological datasets to demonstrate its modelling performance.
GEOSCAN ID328099

 
Date modified: