Deep neural networks for improving physical accuracy of 2D and 3D multi-mineral segmentation of rock micro-CT images

Wang, Ying Da; Shabaninejad, Mehdi; Armstrong, Ryan T.; Mostaghimi, Peyman

doi:10.1016/j.asoc.2021.107185

Cited by 78 publications

(41 citation statements)

References 73 publications

(96 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After mineral compositions were quantified by Energy-Dispersive X-ray Spectroscopy, the 2-D BSEM images were processed with NanoMin software and then registered into the 3-D μCT tomogram to align the BSEM image with the corresponding region in the registered μCT 3-D image volume (Latham et al, 2008). Through mapping the gray levels of μCT images to the mineral phases of BSEM images (Golab et al, 2010), the X-ray intensity range of each mineral is distinguished, and minerals of the μCT 3-D images are segmented, as presented by Wang et al (2020). An example of mineral segmentation along with the dry μCT tomogram and registered 2-D mineral map image acquired from QEMSCAN is illustrated in Figure S4.…”

Section: Core Flooding Experiments and Qemscan Imagingmentioning

confidence: 99%

Influence of Clay Wettability Alteration on Relative Permeability

Fan

McClure

Armstrong

et al. 2020

Geophysical Research Letters

Self Cite

View full text Add to dashboard Cite

Understanding the wettability of porous materials is important to model fluid flow in the subsurface. One of the critical factors that influences wetting in real reservoirs is the composition of geologic materials. The wetting properties for clay minerals can have a particularly strong impact on flow and transport. In this work, we analyze the chemical composition of a Mt. Simon sandstone core to resolve the microscopic structure of clay regions and assess how alterations to the local wetting properties influence multiphase transport based on core flooding experiments and relative permeability simulations. We show that whichever fluid has greater affinity toward clay minerals will tend to accumulate within these high surface area regions, leading to dramatic shifts in the relative permeability. This work establishes the essential importance of the mineral composition and associated wetting properties in the modeling of flow and transport in reservoir-scale systems. Plain Language Summary Surface interactions between fluid and solid materials have important consequences for underground fluid flow. Wettability refers to the relative preference for one fluid to coat the solid surface when multiple fluids are present. Exposure to different fluids can dramatically alter the wettability for mineral surfaces in geological systems, which can have a significant effect on multiphase flow in geological reservoirs. We show that alteration of the wetting properties for clay regions can dramatically alter the behavior for water and CO 2 to flow through reservoir rocks. The underlying causes for these effects are understood by examining microscopic fluid configurations obtained based on the composition of geologic materials. A deeper comprehension of CO 2 flow and transport behavior in a reservoir rock with high mineralogical heterogeneities will aid the CO 2 storage capacity prediction and reduce CO 2 leakage potential.

show abstract

Section: Core Flooding Experiments and Qemscan Imagingmentioning

confidence: 99%

Influence of Clay Wettability Alteration on Relative Permeability

Fan

McClure

Armstrong

et al. 2020

Geophysical Research Letters

Self Cite

View full text Add to dashboard Cite

show abstract

“…The data size is 1000×1000×750 voxels with a resolution of 1.8 micrometers. Further details on the micro-CT data are provided by Da Wang et al (2020). For the simulations, a mixed-wet system was generated by using a morphological approach to achieve irreducible water saturation (McClure et al 2021).…”

Section: Geometrical and Topological Characterization Of Wettingmentioning

confidence: 99%

“…Pore-scale micro-CT images of sedimentary rocks provide complex 3D images on the spatial distributions of minerals (Wildenschild and Sheppard 2013) yet lack the resolution needed to define thin brine films. In addition, the accurate quantification of micro-CT data beyond distinguishing solid from void remains a challenge (Da Wang et al 2020). Sedimentary rocks are commonly composed of many different solid materials, such as quartz, K-feldspar, muscovite, clay minerals, and other precipitated salts, as displayed in Fig.…”

Section: Introductionmentioning

confidence: 99%

Multiscale Characterization of Wettability in Porous Media

et al. 2021

Self Cite

View full text Add to dashboard Cite

Wettability is one of the key controlling parameters for multiphase flow in porous media, and paramount for various geoscience applications. While a general awareness of the importance of wettability was established decades ago, our fundamental understanding of how wettability influences transport and of how to characterize wettability has improved tremendously in recent years through breakthroughs in imaging technology and modeling techniques. Numerical modeling studies clearly show not only that macroscopic two-phase flow is influenced by the average wettability, but also that the spatial distribution of wetting significantly impacts the macroscopic parameters. Herein, we explore the thermodynamics for porous multiphase systems, and recent breakthroughs in wettability characterization. Our view is that bridging the multiscale characterization of wetting must consider two fundamental perspectives: geometry and energy. Advancing the overall description requires an improved understanding of the operative mechanisms that dominate at various scales, and the development of quantitative approaches to capture these effects. We take a multistage approach, looking at these fundamental perspectives from the sub-pore-to-pore length scales, followed by the pore-to-core length scales using various analytical techniques and numerical simulations. Within this context, there remain many open-ended questions, and we therefore highlight these issues to provide guidance on future research directions. Our overall aim is to provide comprehensive guidance on the multiscale characterization of wettability in porous media, in order to facilitate novel research.

show abstract

“…One of the most well-attended subjects in multi-phase segmentation of porous material images is mineralogy analysis of sedimentary deposits (Andrew, 2018;Da Wang et al, 2020;Karimpouli & Tahmasebi, 2019). Sedimentary rocks are naturally formed by different minerals carried and deposited under high pressure and temperature conditions (Adams et al, 2017) and knowing their internal mineralogy in non-destructively has many applications in mining, petroleum geology and geoscience (Garfi et al, 2020;Massara et al, 2019).…”

Section: Multi-phase Segmentationmentioning

confidence: 99%