Rock Mineral Volume Inversion Using Statistical and Machine Learning Algorithms for Enhanced Geothermal Systems in Upper Rhine Graben, Eastern France

Accurately determining the mineralogical composition of rocks is essential for precise assessments of key petrophysical properties like effective porosity, water saturation, clay volume, and permeability. Mineral volume inversion is particularly critical in geological contexts characterized by heterogeneity, such as in the Upper Rhine Graben (URG), where both carbonate and siliciclastic formations are prevalent. The estimation of mineral volumes poses challenges that involve both linear and nonlinear relationships associated with geophysical data. To address this complexity, our methodology strategically integrates the robust insights from standard statistical approaches with three machine learning (ML) algorithms: multi‐layer perceptron, random forest regression, and gradient boosting regression. Furthermore, we propose a new hybrid ensemble model that incorporates a weighted average of multiple ML approaches to predict mineral composition within the Muschelkalk and Buntsandstein formations of the URG. ML techniques for mineral composition prediction in these formations exhibit robust predictive performance. The predicted mineral volumes align closely with quantitative estimates derived from X‐ray diffraction analysis. Additionally, they are in good qualitative agreement with mineral descriptions obtained from cores and cuttings of the Muschelkalk and Buntsandstein formations. Plain Language Summary We conducted an assessment of subsurface rock mineral compositions from their physical properties measured through logging tools, employing a combination of statistical and machine learning techniques. The outcomes derived from these methodologies demonstrate their complementary nature and robustness in elucidating the spatial distribution of minerals within Triassic rocks from the Upper Rhine Graben in France. This approach helps in deciphering complex mineralogical compositions and geological structures within subsurface geothermal reservoirs. Key Points We use well‐logs to estimate mineral volumes in Triassic formations of the Upper Rhine Graben using statistical and machine learning (ML) methods We show that the random forest and gradient‐boosting regressions provide better‐fitting predictions than the multi‐layer perceptron More realistic mineral volume estimates can be obtained by combining several ML algorithms rather than using a single one