Результаты исследований: Научные публикации в периодических изданиях › статья › Рецензирование
A machine learning framework for mineralogical composition assessment in unconventional formations. / Gainitdinov, B.; Meshalkin, Y.; Orlov, D.; Chekhonin, E.; Zagranovskaya, J.; Koroteev, D.; Popov, Y.
в: Advances in Geo-Energy Research, Том 19, № 1, 25.01.2026, стр. 14-29.Результаты исследований: Научные публикации в периодических изданиях › статья › Рецензирование
}
TY - JOUR
T1 - A machine learning framework for mineralogical composition assessment in unconventional formations
AU - Gainitdinov, B.
AU - Meshalkin, Y.
AU - Orlov, D.
AU - Chekhonin, E.
AU - Zagranovskaya, J.
AU - Koroteev, D.
AU - Popov, Y.
N1 - Export Date: 29 March 2026; Cited By: 0
PY - 2026/1/25
Y1 - 2026/1/25
N2 - Quantitative determination of mineralogy, through laboratory core studies and high-definition spectroscopic logging, is effective but underutilized due to cost and complexity. Unconventional formations present additional challenges, such as kerogen presence, hetero-geneity, and anisotropy. This problem can be addressed by utilizing well logs and thermal profiling with specialized wrappers, such as multioutput regressor and regressor chain. Several machine learning models and strategies for combining well logs on multiscale data from an unconventional formation in West Siberia were tested to predict the mass and volumetric fractions of minerals obtained from the Litho Scanner. The gradient boosting regressor, wrapped in a regressor chain and combined with conventional well logs, demonstrated superior performance in predicting both mineral weight and volume fractions, effectively capturing the heterogeneity of the rock structure. A comparison between the machine learning-based model and the Litho Scanner showed an average discrepancy, measured by the root mean squared error for weight fraction, of 0.026 in the Bazhenov Formation. The relationship between certain minerals and the thermal properties of the rock was validated by assessing the importance of thermal core logging data for quartz and pyrite. Moreover, the volume fraction of the rock matrix, composed of total organic carbon and other minerals, was predicted more accurately by incorporating thermal core logging data. The mineral densities, required for obtaining mineral volumes, were determined by solving an optimization problem. Subsequently, a theoretical model was used to calculate thermal conductivity from the mineral volume fractions, revealing a significant similarity between the predicted and experimental values. © The Author(s) 2025.
AB - Quantitative determination of mineralogy, through laboratory core studies and high-definition spectroscopic logging, is effective but underutilized due to cost and complexity. Unconventional formations present additional challenges, such as kerogen presence, hetero-geneity, and anisotropy. This problem can be addressed by utilizing well logs and thermal profiling with specialized wrappers, such as multioutput regressor and regressor chain. Several machine learning models and strategies for combining well logs on multiscale data from an unconventional formation in West Siberia were tested to predict the mass and volumetric fractions of minerals obtained from the Litho Scanner. The gradient boosting regressor, wrapped in a regressor chain and combined with conventional well logs, demonstrated superior performance in predicting both mineral weight and volume fractions, effectively capturing the heterogeneity of the rock structure. A comparison between the machine learning-based model and the Litho Scanner showed an average discrepancy, measured by the root mean squared error for weight fraction, of 0.026 in the Bazhenov Formation. The relationship between certain minerals and the thermal properties of the rock was validated by assessing the importance of thermal core logging data for quartz and pyrite. Moreover, the volume fraction of the rock matrix, composed of total organic carbon and other minerals, was predicted more accurately by incorporating thermal core logging data. The mineral densities, required for obtaining mineral volumes, were determined by solving an optimization problem. Subsequently, a theoretical model was used to calculate thermal conductivity from the mineral volume fractions, revealing a significant similarity between the predicted and experimental values. © The Author(s) 2025.
KW - gradient boosting
KW - Machine learning
KW - mineral composition prediction
KW - subsurface characterization
KW - thermal profiling
KW - unconventional reservoir
KW - well logs
KW - Adaptive boosting
KW - Chains
KW - Experimental mineralogy
KW - Learning systems
KW - Organic carbon
KW - Organic minerals
KW - Rocks
KW - Thermal conductivity
KW - Thermal logging
KW - Gradient boosting
KW - Machine-learning
KW - Mineral composition
KW - Mineral composition prediction
KW - Subsurface characterizations
KW - Thermal
KW - Thermal core
KW - Thermal profiling
KW - Unconventional reservoirs
KW - Well logs
KW - chemical composition
KW - core logging
KW - machine learning
KW - optimization
KW - prediction
KW - pyrite
KW - quartz
KW - thermal conductivity
KW - total organic carbon
KW - volume
KW - weight
KW - well logging
KW - Mean square error
KW - Volume fraction
UR - https://www.mendeley.com/catalogue/a50d5b08-a284-307c-9e98-4a8ca69daf33/
U2 - 10.46690/ager.2026.01.02
DO - 10.46690/ager.2026.01.02
M3 - статья
VL - 19
SP - 14
EP - 29
JO - Advances in Geo-Energy Research
JF - Advances in Geo-Energy Research
SN - 2207-9963
IS - 1
ER -
ID: 151309001