Flow estimation solely from image data through persistent homology analysis

Suzuki, Anna and Miyazawa, Miyuki and Minto, James M. and Tsuji, Takeshi and Obayashi, Ippei and Hiraoka, Yasuaki and Ito, Takatoshi (2021) Flow estimation solely from image data through persistent homology analysis. Scientific Reports, 11 (1). 17948. ISSN 2045-2322 (https://doi.org/10.1038/s41598-021-97222-6)

[thumbnail of Suzuki-etal-SR-2021-Flow estimation-solely-from-image-data-through-persistent-homology-analysis]
Preview
 Text. Filename: Suzuki_etal_SR_2021_Flow_estimation_solely_from_image_data_through_persistent_homology_analysis.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo
Download (2MB)| Preview

Abstract

Abstract: Topological data analysis is an emerging concept of data analysis for characterizing shapes. A state-of-the-art tool in topological data analysis is persistent homology, which is expected to summarize quantified topological and geometric features. Although persistent homology is useful for revealing the topological and geometric information, it is difficult to interpret the parameters of persistent homology themselves and difficult to directly relate the parameters to physical properties. In this study, we focus on connectivity and apertures of flow channels detected from persistent homology analysis. We propose a method to estimate permeability in fracture networks from parameters of persistent homology. Synthetic 3D fracture network patterns and their direct flow simulations are used for the validation. The results suggest that the persistent homology can estimate fluid flow in fracture network based on the image data. This method can easily derive the flow phenomena based on the information of the structure.

ORCID iDs

Suzuki, Anna, Miyazawa, Miyuki, Minto, James M. ORCID logoORCID: https://orcid.org/0000-0002-9414-4157, Tsuji, Takeshi, Obayashi, Ippei, Hiraoka, Yasuaki and Ito, Takatoshi;
CORE (COnnecting REpositories)