Original scientific paper
https://doi.org/10.17794/rgn.2026.5.4
GEOMECHANICAL IMPACTS OF PORE TYPES IDENTIFIED BY MACHINE LEARNING IN PERMIAN–TRIASSIC RESERVOIRS OF THE PERSIAN GULF
Saeed Karimkhani
; School of Geology, College of Science, University of Tehran, Tehran, Iran.
Vahid Tavakoli
; School of Geology, College of Science, University of Tehran, Tehran, Iran.
*
* Corresponding author.
Abstract
Understanding geomechanical behaviour is essential in the petroleum industry for optimizing wellbore stability, drilling operations, and hydrocarbon recovery. This study examines how different pore types influence the geomechanical properties of the reservoir. Petrographic analysis, wireline logs, routine core data, and geomechanical measurements were used. Geomechanical data include Young’s modulus (E), Poisson’s ratio (ϑ), shear modulus (G), bulk modulus (K), Schmidt hammer rebound values, unconfined compressive strength (UCS), cohesion (C), and internal friction angle (Ф). The DataFrame was divided based on porosity to evaluate its impact on geomechanical properties. Missing porosity values were predicted using the ExtraTreesRegressor model, which achieved an R² of 0.82. Since the datasets had different sampling depths, the radial basis function (RBF) method was used to interpolate wireline log geomechanical proper ties to core depths. Machine learning based barplots were then applied to assess how pore types affect geomechanics. The results show that rocks with non fabric selective pores have the highest geomechanical strength. In contrast, samples with fabric selective pores exhibit the weakest properties. Depositional pore types fall in between, having lower geomechanical properties than non fabric selective pores, but higher than fabric selective ones. By integrating petro graphic and petrophysical data with machine learning techniques, this study provides a practical workflow for evaluating geomechanical heterogeneity. This approach improves reservoir characterization and supports better development decisions in carbonate fields.
Keywords
pore type; geomechanical; properties; machine learning; ExtraTreesRegressor model
Hrčak ID:
352835
URI
Publication date:
27.10.2026.
Visits: 0 *