Title | Applications of machine learning to geoscience: nanoporosity and fluid flow in tight formations |
Download | Downloads |
| |
Licence | Please note the adoption of the Open Government Licence - Canada
supersedes any previous licences. |
Author | Chen, Z ;
Stoyanov, S R ; Liu, X; Mane, J ; Little, E |
Source | Geological Survey of Canada, Scientific Presentation 96, 2019, 1 sheet, https://doi.org/10.4095/313629 Open Access |
Image |  |
Year | 2019 |
Publisher | Natural Resources Canada |
Document | serial |
Lang. | English |
Media | on-line; digital |
File format | pdf |
Subjects | fossil fuels; sedimentology; petroleum resources; hydrocarbons; hydrocarbon recovery; modelling; reservoir rocks; pore structure; pore size; porosity; fluid flow; scanning electron microscopy; resource
estimation; production; economic analyses; Methodology; Artificial intelligence; machine learning; Decision making |
Illustrations | photomicrographs; digital images; schematic diagrams; bar graphs; frequency distribution diagrams; spectra; flow diagrams |
Program | Geoscience for New Energy Supply (GNES) Shale-hosted petroleum resource assessment |
Released | 2019 03 19 |
Summary | (Plain Language Summary, not published) The Geological Survey of Canada and CanmetENERGY Devon are developing a collaborative engagement to enhance hydrocarbons recovery from tight formations.
This engagement couples artificial intelligence (AI)-based characterization with physico-chemical modeling of hydrocarbon distribution and flow through rock nanopores to develop an AI-enabled engineering model. This poster outlines our engagement
plan highlighting opportunities for collaboration with industry. It also presents preliminary results from applications of ML in feature extraction from SEM images for nanopore characterization and from physico-chemical modeling of the distribution
of hydrocarbons near mineral surfaces. The AI-enabled engineering modeling would be employed by government and industry to help enhance hydrocarbon recovery beyond the current several percent as well as to address environmental challenges by
developing additives and processes with lower impact on water pollution. |
GEOSCAN ID | 313629 |
|
|