Morphology Decoder to Predict Heterogeneous Rock Permeability with Machine Learning Guided 3D Vision

Abstract

<p><a></a><a>Permeability has a dominant influence on the flow behavior of a natural fluid, and without proper quantification, biological fluids (Hydrocarbons) and water resources become waste. During the first decades of the 21<sup>st</sup> century, permeability quantification from nano-micro porous media images emerged, aided by 3D pore network flow simulation, primarily using the Lattice Boltzmann simulator. Earth scientists realized that the simulation process holds millions of flow dynamics calculations with accumulated errors and high computing power consumption. Therefore, accuracy and efficiency challenges obstruct planetary exploration. To effic­­­iently, consistently predict permeability with high quality, we propose the Morphology Decoder. It is a parallel and serial flow reconstruction of machine learning-driven semantically segmented heterogeneous rock texture images of 3D X-Ray Micro Computerized Tomography (μCT) and Nuclear Magnetic Resonance (MRI). For 3D vision, we introduce controllable-measurable-volume as new supervised semantic segmentation, in which a unique set of voxel intensity corresponds to grain and pore throat sizes. The morphology decoder demarks and aggregates the morphologies' boundaries in a novel way to quantify permeability. The morphology decoder method consists of five novel processes, which we describe in this paper, these novel processes are (1) Geometrical: 3D Permeability Governing Equation, (2) Machine Learning: Guided 3D Properties Recognition of Rock Morphology, (3) Analytical: 3D Image Properties Integration Model for Permeability, (4) Experimental: MRI Permeability Imager, and (5) Morphology Decoder (the process that integrates the other four novel processes).</a></p>

3D Adapted Random Forest Vision (3DARFV) for Untangling Heterogeneous-Fabric Exceeding Deep Learning Semantic Segmentation Efficiency at the Utmost Accuracy

“…7, we display the results of SAV-II semantic segmentation, which provides new insight into this heterogeneous fabric of carbonate rocks. The segmentation of PorThN determines the rock's physical and chemical properties [25,45]: lithology, porosity, permeability, capillary pressure, lithofacies, and relative permeability [46]. We display the comparison between the three finalist algorithms NB, DNN, and RF in Fig.…”

“…However, for the planetary exploration [16], some decisions are needed fast enough not to miss an opportunity or suffer costly challenges [17][18][19]. Micro-nano-porous media images determined heterogeneous fabric physical properties (e.g., porosity, pore throat network, permeability, capillary pressure, wettability, relative permeability) [20][21][22][23][24][25][26][27][28][29]. Computer vision also determined heterogeneous fabric chemical properties (e.g., lithology and mineral volume) [30,31], leading to digital rock typing (DRT), an artificial intelligence geoscience research agent.…”