Pore-throat radius is one of the key parameters that characterizes the microscopic pore structure of rock, which has an important impact on oil-gas seepage and the prediction of remaining oil’s microscopic distribution. Currently, the quantitative characterization of a pore-throat radius mainly relies on rock-core experiments, then uses capillary pressure functions, e.g., the J-function, to predict the pore-throat radius of rocks which have not undergone core experiments. However, the prediction accuracy of the J-function struggles to meet the requirements of oil field development during a high water-cut stage. To solve this issue, in this study, based on core experimental data, we established a deep neural network (DNN) model to predict the maximum pore-throat radius R_max, median pore-throat radius R₅₀, and minimum flow pore-throat radius R_min of rocks for the first time. To improve the prediction accuracy of the pore-throat radius, the key components of the DNN are preferably selected and the hyperparameters are adjusted, respectively. To illustrate the effectiveness of the DNN model, core samples from Q Oilfield were selected as the case study. The results show that the evaluation metrics of the DNN notably outperform when compared to other mature machine learning methods and conventional J-function method; the root-mean-square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) are decreased by 14–57.8%, 32.4–64.3% and 13.5–48.9%, respectively, and the predicted values are closer to the true values of the pore-throat radius. This method provides a new perspective on predicting the pore-throat radius of rocks, and it is of great significance for predicting the dominant waterflow pathway and in-depth profile control optimization.

Qualitative schematic models of the Mishrif Formation, which have previously dominated the research, are inadequate for predicting reservoir production performance due to their inability to quantify spatial heterogeneity. In contrast to these earlier approaches, this study integrates core analysis, wireline logs, and 3D seismic data to not only describe but also quantitatively characterize the depositional elements and their spatial distribution. A novel methodology was developed to define nine distinct depositional elements from cored wells and then continuously identify them in uncored wells using unique pseudo-wireline log responses, a step not achieved in prior work. Furthermore, moving beyond previous qualitative models, 3D quantitative versions were constructed using Sequential Indicator Simulation (SIS) explicitly constrained by depositional geometries derived from 3D seismic inversion volumes. For the first time, these models reveal the quantitative spatial extent and evolution of these elements. Updating the 3D petrophysical property model using this new depositional framework resulted in a 15% increase in successful production history matches, demonstrating the direct and superior predictive power of this integrated quantitative approach for forecasting oil reservoir production performance.