Extracting connectivity paths in 3D reservoir property: A pseudo skeletonization approach. (February 2023)

• Record Type: Journal Article
• Authors: Thore, Pierre; Lucas, Antoine
• Is Part Of: Computers & geosciences. Volume 171(2023)
• Publication Date: 2023-02
• Subjects: Environmental policy -- Periodicals; 550.5
• DOI: 10.1016/j.cageo.2022.105262
• Languages: English
• ISSNs: 0098-3004

Abstract

Abstract: We propose a new algorithm to extract connectivity paths from a 3D reservoir property. Reservoir properties can either be derived from seismic inversion or simply be any property of a flow simulation model. Our algorithm results in connected paths starting at source point(s) and dithering towards reservoir limits in tree-like shape. The connectivity paths represent the internal structure of a reservoir property. Our algorithm extracts them with a density of paths as sparse or detailed as needed. As such the algorithm can be seen as a pseudo skeletonization of the reservoir property. Classic skeletonization algorithms are not directly fit for this type of data which have specific characteristics compared to classic images, such as: undefined volume limits and amplitude (grey level) representing the major information. In this article, we first provide a description of our algorithm and illustrate its application on synthetic and real data. In the second part we discuss some applications of the connectivity paths already used for reservoir characterization and consider new applications. As connectivity paths offer a unique manner of analyzing the interior of a reservoir property and extracting its structure, we believe that this new object will have a significant impact for improving reservoir models.