Currently, models based on the application of artificial intelligence methods are actively developed and applied in solving a variety of problems, including in the practice of petroleum engineering. Evaluation of the accuracy and reliability of the developed models is usually reduced to defining standard statistical criteria, while the developers do not always use a separate examination sample. This article presents the results of the study, which are reduced to multivariate testing of the neural network model previously developed by the authors to determine the dynamic reservoir pressure in the selection zones of oil wells. The model is characterized by a number of advantageous characteristics, including minimal requirements for the amount of initial data, which determines its relevance and practical demand. However, the closed nature of computational algorithms related to the "black box" category does not allow us to reasonably formulate the conditions and criteria for applying the model, the reliability of the retro- and prospective forecast of the reservoir pressure. Three oil deposits of one field with different geological and physical conditions were selected as the object of study. The availability of a large number of actual reservoir pressure determinations by means of hydrodynamic well testing at the field allowed testing the model under a variety of scenarios, for each of which the forecast error was estimated and analyzed. As a result, high estimates of the model for retro- and prospective reservoir pressure reproduction were confirmed. It was found that forecast errors are reduced to zero in the presence of a large number of actual reservoir pressure determinations. However, to perform the calculation for each well, a single measurement for the entire history is sufficient. It was found that a sharp change in the well flow rate should also be accompanied by an actual determination of reservoir pressure with the entry of the obtained value into the model. In the absence of even a single reservoir pressure measurement for the wells, the model reliably reproduces its value using the kriging procedure used in the algorithms.

Natural gas separation is an important process in wells equipped with electric submersible pumps (ESP) that affects the efficiency of the "wellbore-pump-tubing" system. Nowadays, the amount of knowledge about this process requires critical analysis and further improvement. The paper presents the results of studying the unsteady features of the process of separation of gas bubbles into the annular space in the near-intake domain of the well model with conditionally radial inlet. The results of the experimental bench tests, as well as the results of numerical simulation in dynamic multiphase flow simulator are analyzed. The experiments were carried out on a test rig with the inner diameter of the casing model 80 mm and the outer diameter of the intake module 64 mm, taking into account the possibility of measuring liquid and gas flow rates, as well as high-speed video recording of the processes occurring in the near-intake domain of the well model. Unsteady features of gas-liquid mixtures flow with the help of video frames in the near-intake domain for model mixtures “Water-Air” and “Water-Surfactant-Air” are shown. It is revealed that at small time intervals (<1 s) the regimes with slug-churn flow pattern are characterized by significant nonstationarity. The results of numerical simulation indicate that such unsteady behavior can lead to oscillatory operation of the well and ESP.

On the basis of critical analysis of the obtained research results the following promising directions are formulated: a study of theoretical basis of separation in the near-intake domain of a well; field and bench experiments; a numerical modeling of gas natural separation into the annular space of a well equipped with ESP.

Recent advances in machine learning have enabled the automatic analysis of microtomography (µCT) images, facilitating more efficient rock property identification. This study aims to predict the experimentally measured open porosity of reservoir rocks using µCT images of standard core plugs. A dataset of 136 core plugs was collected, including 49 sandstone and 87 carbonate samples. Open porosity was experimentally determined using gas volumetry. The core plugs (30 ± 1 mm in height and diameter) were scanned using µCT with a resolution of 34.6–38.0 µm, producing 16-bit image stacks. The dataset consisted of 100,232 images (64,119 carbonate and 36,113 sandstone). To label the images, we introduced a supervised method called Segmentation of Unresolved Pores via Experimental Reference (SUPER), which segments dark voxels to match the experimentally measured open porosity, adapting to each sample’s characteristics. Three shifted window (Swin) transformer models were trained: a universal model and specialized models for sandstone and carbonate. The models used transfer learning with ImageNet weights, followed by fine-tuning. Testing confirmed that specialized models outperformed the universal model. This highlights that training an ensemble of models adapted to specific rock types leads to better performance than a single general model for porosity prediction. A key challenge arose with sandstones, especially fine-grained types, where small pores merged due to resolution limitations. Future work should improve image resolution and feed detailed images into the model. The method has potential for full-scale core scans and early porosity assessment in raw core plugs, including fragile reservoirs with oil or bitumens.

This paper describes a novel approach for automatic depth shifting geophysical well logs data and whole core images. The proposed approach combines the use of machine learning algorithms for automatic lithotype description via whole core images and an amplitude modal inversion algorithm, enabling the integration of stratigraphic information during the shifting process. The automatic shifting with whole core images is performed without using the natural gamma radiation curve of drill core or core routine data. As part of this study, the proposed workflow was validated on clastic and carbonate datasets (including non-clayey carbonates rocks) for neutron and density logs. The automatic shifting results have high accuracy, with precision equal to that of the smoothing filter (approximately 60 cm for neutron logs and 80 cm for density logs.), according to the analysis of the processed data and comparison with core routine data. The proposed technique significantly reduces the time required for processing well logging data compared to traditional approaches, minimizes errors related to human factors, and can serve as a foundation for further research in this area.

Reliable data on the relative permeability of low-permeability organic-rich shale reservoirs are essential for optimizing the development and operation of these fields by enhancing hydrodynamic models. Key challenges and limitations in studying complex reservoir systems include limited experimental data due to a shortage of core samples and the inapplicability of conventional laboratory methods for characterizing gas flow due to properties of these reservoirs (low porosity and permeability, high amount of organic matter including kerogen).

Investigating the effects of ultra-low interfacial tension, sorption, and diffusion on fluid flow under simulated reservoir conditions is crucial. Despite the importance of these factors, mechanisms of sorption and diffusion of hydrocarbon gases in shale rocks are often overlooked in experimental and numerical research on recovery strategies.

This review aims to provide a comprehensive understanding of gas adsorption and diffusion related mechanisms in tight shale rocks. Main findings includes assessment how various rock properties affect fluid behavior in nanoscale pores and identify critical areas for experimental analysis. By generalizing the review results, this work also highlights emerging research trends and address limitations in integrating adsorption and diffusion data for recovery, improving the accuracy of recoverable reserve estimates and reducing economic risks in unconventional reservoir development.