We propose new models to find the vulnerable countries in terms of food security. These are based on network analysis under deep uncertainty. The conditions uncertainty affect supply and demand food, namely, carbohydrates countries. Under such networks carbohydrate between constructed. vulnerability security were studied by centrality indices taking into account consumptions possibility group influence a country. Also, our show direct indirect dependence import from other scenario one...

Abstract Recognizing vuggy zones in borehole imaging (BHI) presents significant challenges due to the intricate and heterogeneous nature of these geological features (fig.1). Traditional methods, which often rely heavily on subjective visual interpretation BHI data, are prone variability inconsistencies. These limitations contribute substantial uncertainties modeling subsurface characterization, impacting reliability reservoir assessments. The integration AI-driven techniques offers a...

Abstract Monitoring HSE violations while focusing on Operations is one of the main targets every industry. Unfortunately, Humans can be prone to put focus mainly operations rather than safety. Therefore, assisting them with an co-pilot in form AI solution alarming when they tend compromise safety, huge value. While many attempts have been made literature terms Personnel Protective Equipment (PPE), there no available that uses modern deep-learning based video analytics comprehensively detect...

Summary The objective of this research was examination machine learning algorithms in combination with a priori geological information applicability for automatical facies distribution from wireline logs problem. This study based on data Field M located Western Siberia which can be characterized by complex geology making results reliable. During the project different classification were evaluated to find most appropriate one interpretation task. Classifiers trained and tested M, produced...

Summary These days understanding of fault geometry distribution across a particular oil or gas reservoir becomes very important task. It arises from the fact that fluid flow present unconventional deposits is mostly driven by natural fractures instead sedimentary porosity and corresponding permeability. On other, complex compartmentalized separated into small discontinuous tectonic activity could lead to economic risks during field development. Most conventional tools for interpretation...

This paper describes an effective solution to the task of a remote monitoring super-extended objects (oil and gas pipeline, railways, national frontier). The suggested is based on principle simultaneously seismoacoustic optical/infrared physical fields. simultaneous those fields not new but in contrast known solutions approach allows control with very limited operational costs. So-called C-OTDR (Coherent Optical Time Domain Reflectometer) systems are used monitor field. Far-CCTV A data...

Summary Extra net thickness may bring a huge impact on projects NPV, especially in case of brownfields with vast production wells stock and maintained surface infrastructure. Reservoir beds sand be misinterpreted by petrophysicist within well miscorrelated spatially. We propose statistical learning methods to identify missed reservoir therefore extra predictions supervised model. Robustness analysis such identification is the main purpose our paper. Methodology tested 3 Western Siberia along...

Summary The main goal of this work is to develop a systematic approach with raw well log data. Toaccomplish goal, we propose fit simple unsupervised generative model the input data and au-tomate preprocessing step using model. This allows detect anomaliesin as regions that struggles explain (i.e., samples extremely low like-lihood), infer approximations missing features Bayes rule incorporate additional expert knowledge in design

Abstract The new approach for production prediction was developed and is described in the article which involves clustering analysis aimed to well logs such that reservoir non-reservoir rocks are obtained (presented by various clusters) subsequent linkage among clusters' types, their thicknesses characteristics found. It may be implemented of planned wells' ranking further. Such provide solution tasks. used geological features' estimation. For example, adjusted sedimentological...

The stratigraphic framework of the Lower Cretaceous in Abu Dhabi is significantly shaped by tectonic activity and eustatic sea-level fluctuations, which have governed sedimentation patterns diagenetic processes. A comprehensive understanding facies distribution, history, depositional settings these formations essential for effective hydrocarbon exploration production, as they represent key reservoirs Dhabi's prolific oil gas fields. Detailed thin section analysis microscale reservoir...

We construct a model to find the insecure countries in terms of food security using network analysis under conditions deep uncertainty. The uncertainty can be drought, flood, earthquakes, etc., which affect supply and demand countries. Under such are more vulnerable has been studied. In order study situations, for healthy consumption different products built, is constructed determine import export data basic (Rice, Wheat, Maize, Sorghum, Barley Rye) various Different measures have applied...

Abstract—The practical efficient approach is suggested for estimation of the seismoacoustic sources energy in C-OTDR monitoring systems. This represents sequential plan confidence both energy, as well absorption coefficient soil. The delivers non-asymptotic guaranteed accuracy obtained estimates form regions with prescribed sizes. These are valid a finite sample size when distributions observations unknown. Thus, and nonparametric, also these guarantee prior regions, value.

This paper presents new results concerning selection of an optimal information fusion formula for ensemble Lipschitz classifiers. The goal is to create integral classificatory which could provide better generalization ability the while achieving a practically acceptable level effectiveness. problem very relevant data processing in multi-channel C-OTDR-monitoring systems. In this case we have effectively classify targeted events appear vicinity monitored object. Solution based on usage...

Abstract At ADIPEC'2022, SmartVessel, a vision-based HSE monitoring system, showed remarkable results in addressing challenges Marine Vessels. Despite its innovative features, SmartVessel was limited analysis of operational situation on working site as it relies solely images, lacking temporal context. This deficiency leads to numerous false alarms and unnecessary operations interruptions. To address this issue, the new version AI solution called SmartVessel-v2 proposed. large number...

АНАЛИЗ И МОДЕЛИРОВАНИЕ

Summary These days machine learning models are used on a regular basis for various complex oil and gas tasks. One of the most popular business problems is well log data autointerpretation. However, these models, in cases, require large amount which can be obtained only from mature fields. It assumes high variability due to different tools, measurements record conditions. leads noisy datasets with number inconsistencies affecting accuracy model prediction resulting many misclassifications...