Predictive Soft Computing Methods for Building Digital Rock Models Verified by Positron Emission Tomography Experiments. Issue 11 (8th November 2022)

• Record Type: Journal Article
• Authors: Ebadi, Mohammad; Armstrong, Ryan T.; Mostaghimi, Peyman; Wang, Ying Da; Alqahtani, Naif; Amirian, Tammy; James, Lesley Anne; Parmar, Arvind; Zahra, David; Hamze, Hasar; Koroteev, Dmitry
• Is Part Of: Water resources research. Volume 58:Issue 11(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-11-08
• Subjects: Digital rock physics -- Positron Emission Tomography -- multi‐scale modeling -- heterogeneity -- X‐ray computed tomography; Hydrology -- Periodicals; 333.91
• DOI: 10.1029/2021WR031814
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: The influence of core‐scale heterogeneity on continuum‐scale flow and laboratory measurements are not well understood. To address this issue, we propose a fully automated workflow based on soft computing to characterize the heterogeneous flow properties of cores for predictive continuum‐scale models. While the proposed AI‐based workflow inherently has no trained knowledge of rock petrophysical properties, our results demonstrate that image features and morphological properties provide sufficient measures for petrophysical classification. Micro X‐ray computed tomography ( μ xCT) image features were extracted from full core plug images by using a Convolutional Neural Network and Minkowski functional measurements. The features were then classified into specific classes using Principal Component Analysis followed by K ‐means clustering. Next, the petrophysical properties of each class were evaluated using pore‐scale simulations to substantiate that unique classes were identified. The μ xCT image was then up‐scaled to a continuum‐scale grid based on the defined classes. Last, simulation results were evaluated against real‐time flooding data monitored by Positron Emission Tomography. Both homogeneous sandstone and heterogeneous carbonate were tested. Simulation and experimental saturation profiles compared well, demonstrating that the workflow provided high‐fidelity characterization. Overall, we provided a novel workflow to build digital rock models in a fully automated way to better understand the impacts of heterogeneity on flow. Plain Language Summary: We demonstrate that soft computing methods can be used to classify rock images into specific classes that have unique flow properties. Without any prior knowledge of rock flow properties, a computer is able to interpret rock images in terms of image features that inherently delineate them into classes with unique physical properties. The method is unsupervised, and thus provides a fully automated approach for classification. The characterization approach provides a means to build predictive continuum‐scale digital rock models. The predictive nature of the models were tested by conducting flow experiments using Positron Emission Tomography, which provides full 3D real‐time images of fluid flow. The strong agreement between simulation and experimental results validated the proposed workflow. Key Points: The field of view of digital rock models can be split to various classes by a hybrid of soft computing methods Super‐resolution methods can help to compute more accurate k, p c, and k r models The presented workflow accounts for heterogeneities at the pore and core scale