This study delves into the heightened efficiency and accuracy of 11 deep learning models classifying 11 dynamograph classes in the oil production sector. Introducing a novel framework with the Grad–CAM method, we address the “black box” issue, providing transparency in the models’ decision-making processes. Our analysis includes a comparative study with human experts, revealing a comprehensive understanding of both machine and human interpretive strategies. Results highlight the notable speed and precision of machine learning models, marking a significant advancement in rapid, reliable dynamograph classification for oil production decision-making. Additionally, nuanced findings in the model’s diagnostic accuracy reveal limitations in situations featuring the simultaneous occurrence of multiple pump issues. This underscores the need for additional features and domain-specific logic to enhance discernment and diagnostic precision in complex scenarios. The exploration of qualitative aspects distinguishes interpretive approaches, highlighting strengths and limitations. Machines, driven by algorithmic patterns and data processing, excel in rapid identification, albeit with occasional misclassifications. In contrast, human experts leverage experience and domain-specific knowledge for nuanced interpretation, providing a comprehensive understanding of both quantitative metrics and qualitative nuances. In conclusion, this study not only demonstrates the accelerated and enhanced accuracy of dynamograph classification by machine learning models compared to junior and medior domain experts, but also provides valuable insights into specific features and patterns guiding the decision-making process. This understanding allows continuous refinement, combining machine speed with human understanding for improved results in oil production. The potential for further studies and improvements in this domain is substantial.

Gas wells, particularly those situated onshore, play a vital role in the global energy sector by supplying a significant portion of natural gas. However, operational challenges, notably gas hydrate formation, pose substantial issues, leading to complications such as flowline blockages and unexpected well shutdowns. Gas hydrates, crystalline structures resembling ice, form under specific conditions of low temperature and high pressure. This paper explores the complex process of hydrate formation in gas wells, emphasizing the challenges it presents and the need for specialized strategies to address these issues. The primary focus is a case study of an onshore gas well experiencing recurrent hydrate-related problems. Leveraging PipeSim software, a well model is developed, followed by a sensitivity analysis under various operational scenarios. The study investigates mitigation strategies, including choke position adjustments and methanol introduction, crucial for the safe production of oil and gas fields. The significance of this study lies in its aim to optimize well performance and mitigate risks associated with hydrate formation. Findings contribute to existing knowledge and offer practical solutions for industry practitioners and researchers dealing with onshore gas wells. The paper's structure includes a review of related work, details on the experimental setup and results, and concluding remarks. The perennial challenge of hydrate formation in gas wells necessitates a case-specific assessment and individualized approaches. Nodal analysis and well modeling software have become indispensable tools for engineers in developing preventative measures. This paper presents a methodological approach using a specific well as an example, evaluating the effectiveness of three methodologies: downhole choke installation, methanol dosing, and well transfer to a high-pressure separator.

The Integrated Production Model (IPM) acts as a digital twin of the actual field, representing reservoir behavior, all wells and their equipment, as well as the entire surface infrastructure. This model allows for changes in parameters within the virtual environment, enabling the simulation of various scenarios to determine how these changes affect the entire system. Consequently, it is possible to select the most optimal scenario and validate its impacts. The development of the surface infrastructure model represents the final step in integrated production modeling, and this study focuses on creating a surface infrastructure model of an oil field in the Pannonian Basin using two software tools: Aspen HYSYS and Petroleum Experts - GAP. The goal is to compare the results obtained and identify the advantages and limitations of these software solutions. This analysis offers valuable insights into the capabilities of each program in simulating and optimizing oil operations. Engineers use a range of computational strategies and mathematical models to design and operate processing facilities. Aspen HYSYS is a widely recognized tool in the oil and gas industry for process simulation, allowing engineers to model various operational scenarios in detail and assess their impact on system performance. Aspen HYSYS is used for modeling the entire production process, including oil and gas processing, refining, and chemical plants, while GAP specializes in integrated asset modeling (IAM), modeling wells, flowlines, risers, and surface facilities, optimizing the entire production system from the reservoir to the processing plant. This paper first provides the theoretical background relevant to the research topic, including the basic concepts of process simulation in Aspen HYSYS and GAP. Following that, the research methodology is presented, including the steps of simulation and analysis. Finally, the simulation results are discussed, and conclusions are drawn regarding the applicability of both software tools for modeling surface infrastructure in oil fields.

During analysis of work process after first chemical EOR project done in Serbia it was observed that our available resources were not used in an optimal way. Some of laboratory tests that were part of standard testing procedures for selection of chemicals gave us little or non-useful information but took a lot of time and resources. This drove us to analyze all our available resources and to develop a process algorithm that will give us best "value for money spent" in terms of time optimization, and developing testing methodology that will utilize equipment that is already available in our laboratory. In a way, entire process was adapted to local conditions - focusing on conditions in Serbian oilfields and chemical selection methods needed for those conditions. The process described here is applied after chemical EOR method selection and it covers all possible combinations: Surfactant, Polymer, SP or ASP EOR. In case that one component is excluded, workflow can be modified with ease.

Carbon dioxide (CO2) flooding is one of the most important and most used enhanced oil recovery (EOR) method because it does not only increase oil recovery efficiency but also is used as an underground CO2 storage. It is considered a very complex method as it involves knowing the fluid phase behavior with different CO2 concentrations. It should be noted that oil swelling (volume increase) with the dissolution of carbon dioxide has a significant effect on increase of oil recovery. When this occurs, a significant decrease in the viscosity of the oil is observed. In this study, a reservoir 3D simulation modeling approach was applied to evaluate immiscible and miscible CO2 flooding in a high WC reservoir. To reduce simulation time, the PVT composition was grouped into 5 fluid components. The 3-parameter, Peng-Robinson Equation of State (EOS) was used to match PVT experimental data by using the Schlumberger's ECLIPSE PVTi software. One-dimensional slim-tube model was defined using ECLIPSE 300 software to determine the minimum miscibility pressure (MMP) for injection of CO2. Beside this approach, an analytical MMP estimation was carried out using several correlations. Schlumberger Petrel software was used to set up a 3D simulation model of a static and dynamic model. Various scenarios of immiscible and CO2 injection have been simulated using ECLIPSE 300 software and these results have been compared.

Selection of adequate surfactant is one of the most important steps in preparation for ASP EOR. There are many parameters to be taken in considerations in this process but different authors are prioritizing different parameters. Shown here is comparative analysis of two surfactants chosen according difference set of priorities, in one low IFT and stability and type of created microemulsion was priority (Surfactant A) and in another mobility of created microemulsion (Surfactant B). Bottle test was done with both surfactants to assess the stability of microemulsion at formation temperature, and coreflood test to assess ability of surfactant to mobilize trapped oil. During first round of tests Surfactant A gave better results, very low IFT and stabile Windsor type III microemulsion while Surfactant B gave higher IFT and Windsor type I microemulsion. During coreflood test Surfactant B performed better in terms of oil recovery factor (ORF) and injection pressures. Apparently, stabile Windsor type III microemulsion that is considered desirable in ASP injection and widely prioritized in surfactant selection process can cause decrease in permeability and injectivity issues. Good results can be obtained with IFT in "moderately" low range and stability of microemulsion is not critical in terms of oil recovery factor.