科研速览 · Science Skim继续刷下去 · Keep skimming →
◆ Energy Geoscience2026-06-18· Geology

Geosystematical insights into tectonic and sedimentary processes for analyzing hydrocarbon potential in the central Lurestan province, Zagros

Hashem Ahmadpour Gohort, Zahra Maleki, Ali Solgi, Mehran Arian, Pooria Kianoush

原始摘要(英文原文)· Original abstract
The Central Lurestan province of the Zagros Fold-Thrust Belt represents a major hydrocarbon play characterized by complex structural architecture, fracture-dominated reservoirs, and significant overpressure. Accurate hydrocarbon system characterization and drilling risk mitigation are often hindered by the isolated application of analytical methods, which fails to capture the integrated nature of tectonic history, stress evolution, and fluid dynamics. This study develops an integrated geosystematical workflow that synthesizes isopach mapping, fracture network and fractal analysis, geomechanical modeling, and deep learning to establish predictive, process-based relationships within the petroleum system. Detailed isopach mapping of 25 formations and members quantitatively reconstructs the basin's evolution, highlighting critical phases such as the Paleocene-Eocene foredeep infill. Quantitative analysis of over 1200 fractures reveals high-complexity networks (mean fractal dimension, D = 1.78 ± 0.06) whose intensity decays exponentially from major faults (P 32 = 2.8e -0 · 05d + 1.2). A hybrid Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) model, trained on 28 integrated geological and geomechanical features, predicted reservoir pressure with a mean absolute error of 112 psi—a 71% improvement over conventional Eaton methods—and achieved a coefficient of determination ( R 2 ) of 0.96. The model indicates that overpressure compartments are geomechanically controlled, showing a strong correlation ( R = 0.79) with zones of high fracture connectivity. The principal innovations of this research are the establishment of a quantitative, process-based chain linking sedimentary architecture to contemporary fluid systems and the demonstration of a geology-informed machine learning framework for superior reservoir property prediction. This integrated approach provides a transformative tool for de-risking exploration targets, optimizing well placement, and managing geohazards in the Central Lurestan province and analogous tectonically active settings.
读原文 · Read the paper ↗

AI 追问PRO

登录后使用 AI 追问

讨论区

登录后参与讨论

相关论文 · Related

Geosystematical insights into tectonic and sedimentary processes for analyzing hydrocarbon potential in the central Lurestan province, Zagros — 科研速览 Science Skim