Machine learning-based electrofacies characterization of the Asmari reservoir in the Ahvaz oil field (SW Iran) using multi-resolution graph-based clustering (MRGC) and hydraulic flow unit (HFU) analysis
- 1Department of Earth Sciences, College of Science, Shiraz University
Abstract
Electrofacies (EF) analysis has become an effective approach for reservoir characterization, particularly in heterogeneous reservoirs where core data are limited or unavailable. This study presents an integrated Artificial Intelligence (AI) and Machine Learning (ML)-based workflow combining electrofacies analysis, hydraulic flow unit (HFU) classification, and petrographic investigations for reservoir characterization to characterize the Asmari carbonate–clastic (Oligocene–Miocene) reservoir in the Ahvaz oil field, southwestern Iran. The novelty of this study lies in the cross-validation of electrofacies results using independent HFU and microscopic analyses to improve reservoir zonation and quality assessment. Petrophysical log data from the studied wells were analyzed using the Multi-Resolution Graph-Based Clustering (MRGC) algorithm, an unsupervised machine learning technique, resulting in the identification of five distinct electrofacies (EF1–EF5). The results indicate significant heterogeneity within the reservoir. EF1, mainly associated with the Ahvaz Sandstone Member, exhibits the highest reservoir quality, characterized by high effective porosity, low water saturation, and excellent permeability. In contrast, EF5 shows the poorest reservoir quality due to its shale-rich composition, negligible effective porosity (PHIE), and high-water saturation (SW). HFU analysis identified five hydraulic flow units that show strong correspondence with the electrofacies classification, with HFU-5 representing the best reservoir quality and HFU-1 the poorest. Petrographic investigations further confirmed these results by revealing abundant intergranular and secondary dissolution porosity in high-quality reservoir intervals and limited pore development in low-quality facies. The strong agreement among electrofacies classification, hydraulic flow unit analysis, and petrographic observations demonstrates the reliability of the proposed integrated reservoir characterization model. The developed electrofacies model was subsequently propagated to all wells with available log data, providing an effective framework for identifying reservoir-quality zones and supporting reservoir management and field development strategies in the Ahvaz oil field.
Related
- Spatiotemporal clustering and epidemiological characterization of suicides in Ecuador, 2015–2023
- Unsupervised phenotypic clustering, inferential characterization, and discriminant classification of elite flatwater canoeists: neuromuscular and anthropometric determinants
- High-Resolution Characterization of Cone Photoreceptor Mosaic Alterations in Retinopathy of Prematurity Using Adaptive Optics Scanning Laser Ophthalmoscopy
- Deep Learning Reconstruction for Ultra-High-Resolution Photon-Counting CT of the Lung: Image Quality and Texture Characterization
- Hydraulic Pressure‐Programmed Molecular Transport in Tough Hydrogels
BibTeX
@article{mohsenipour2026machine,
title = {{Machine learning-based electrofacies characterization of the Asmari reservoir in the Ahvaz oil field (SW Iran) using multi-resolution graph-based clustering (MRGC) and hydraulic flow unit (HFU) analysis}},
author = {Seyed Abouzar Mohsenipour},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62890-9},
url = {https://doi.org/10.1038/s41598-026-62890-9},
note = {CC BY},
}
RIS
TY - JOUR
TI - Machine learning-based electrofacies characterization of the Asmari reservoir in the Ahvaz oil field (SW Iran) using multi-resolution graph-based clustering (MRGC) and hydraulic flow unit (HFU) analysis
AU - Seyed Abouzar Mohsenipour
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62890-9
UR - https://doi.org/10.1038/s41598-026-62890-9
AB - Electrofacies (EF) analysis has become an effective approach for reservoir characterization, particularly in heterogeneous reservoirs where core data are limited or unavailable. This study presents an integrated Artificial Intelligence (AI) and Machine Learning (ML)-based workflow combining electrofacies analysis, hydraulic flow unit (HFU) classification, and petrographic investigations for reservoir characterization to characterize the Asmari carbonate–clastic (Oligocene–Miocene) reservoir in the Ahvaz oil field, southwestern Iran. The novelty of this study lies in the cross-validation of electrofacies results using independent HFU and microscopic analyses to improve reservoir zonation and quality assessment. Petrophysical log data from the studied wells were analyzed using the Multi-Resolution Graph-Based Clustering (MRGC) algorithm, an unsupervised machine learning technique, resulting in the identification of five distinct electrofacies (EF1–EF5). The results indicate significant heterogeneity within the reservoir. EF1, mainly associated with the Ahvaz Sandstone Member, exhibits the highest reservoir quality, characterized by high effective porosity, low water saturation, and excellent permeability. In contrast, EF5 shows the poorest reservoir quality due to its shale-rich composition, negligible effective porosity (PHIE), and high-water saturation (SW). HFU analysis identified five hydraulic flow units that show strong correspondence with the electrofacies classification, with HFU-5 representing the best reservoir quality and HFU-1 the poorest. Petrographic investigations further confirmed these results by revealing abundant intergranular and secondary dissolution porosity in high-quality reservoir intervals and limited pore development in low-quality facies. The strong agreement among electrofacies classification, hydraulic flow unit analysis, and petrographic observations demonstrates the reliability of the proposed integrated reservoir characterization model. The developed electrofacies model was subsequently propagated to all wells with available log data, providing an effective framework for identifying reservoir-quality zones and supporting reservoir management and field development strategies in the Ahvaz oil field.
SN - 2045-2322
ER -