Data-Driven Three-Phase Saturation Identification from X-ray CT Images with Critical Gas Hydrate Saturation

Abstract

This study proposes three-phase saturation identification using X-ray computerized tomography (CT) images of gas hydrate (GH) experiments considering critical GH saturation (SGH,C) based on the machine-learning method of random forest. Eight GH samples were categorized into three low and five high GH saturation (SGH) groups. Mean square error of test results in the low and the high groups showed decreases of 37% and 33%, respectively, compared to that of the total eight. Additionally, a universal test set was configured from the total eight and tested with two trained machines for the low and high GH groups. Results revealed a boundary at ~50% of SGH signifying different saturation identification performance and the ~50% was estimated as SGH,C in this study. The trained machines for the low and high SGH groups had less performance on the larger and smaller values, respectively, of SGH,C. These findings conclude that we can take advantage of suitable separation of obtained training data, such as GH CT images, under the criteria of SGH,C. Moreover, the proposed data-driven method not only serves as a saturation identification method for GH samples in real time, but also provides a guideline to make decisions for data acquirement priorities.

Evaluation of CO2 Hydrate Saturation in Porous Core Experiments Using Medical CT Images

“…(3) Differentiation of water–gas hydrate (due to the similar density of water and CO 2 hydrate) within each region of the core based on their distinct density and CT numbers compared to rock and gas. The water + gas hydrate saturations can then be calculated by comparing the CT values at each step of the experiment (CT exp ) to both fully water-saturated (CT wet ) and completely dry (CT dry ) conditions, using the following equation: S normalw + normalH = CT exp − CT dry CT wet − CT dry To minimize the noise in the measurements, the saturation values were then averaged by grouping the slices into sections of 1 cm in length. An average saturation for each cm along the core was then obtained.…”

“…In the context of porous media, the dynamic behavior of CO 2 hydrate formation, saturation, and distribution needs to be determined for CO 2 hydrate formation/saturation depending on the intrinsic permeability of the porous media, the water distribution, availability of nucleation sites, and the growth kinetics of hydrate formation. − To address these challenges, visualization methods serve as a promising approach to quantitatively characterize the water–hydrate saturation of porous media under various permeabilities and predict approximate trends. So far, extensive research has utilized nondestructive methods such as MRI , and micro-CT scanning to distinguish the phase states of water, gas, and gas hydrate, and estimate hydrate saturation.…”

Research Advances in Machine Learning Techniques in Gas Hydrate Applications

“…Most often poor machine learning predictions are a result of poor data representation, lack of enough data, and incorrect or an absence of data processing. The data used in gas-hydrate-related machine learning modeling are mostly gathered from experimental articles in literature, field flow assurance data, field reservoir rock properties data, and experimental procedures by others. , Data from literature is commonly used for the prediction of gas hydrate phase behavior conditions. This is because such data are easy to correlate from different data or experimental sources, thus allowing hydrate phase behavior prediction using machine learning techniques.…”