Ivan Deshenenkov
Geological Consultant - Digital Transformation Lead
Saudi Aramco
Dr. Ivan Deshenenkov is Lead of Digital Transformation in Geological Operations of Saudi Aramco. He holds PhD in Petroleum Engineering from Russian Academy of Sciences and has over 17 years of industrial experience. His expertise is focused on petrophysics, digital rocks and special core analysis. He has several patents and more than 40 technical papers.
Participates in
TECHNICAL PROGRAMME | Primary Energy Supply
Advances in Geoscience
Forum 05 | Technical Programme Hall 1
29
April
14:30
16:00
UTC+3
The oil and gas industry is undergoing a significant transformation fueled by the rapid advancements of digital technologies. The integration of advanced digitalization tools and artificial intelligence (AI) methodologies is revolutionizing the field of geoscience, supporting optimization of complex geological workflows, unlocking new insights from diverse data sources, and reactivating legacy assets. This digital transformation deepens the understanding of subsurface reservoirs and empowers more informed, efficient, and effective decision-making. By leveraging multi-scale, multimodal digital imaging and AI methods, we can analyze vast datasets and extract valuable insights from millions of geological rock samples and analytical reports. This paper highlights the potential of digital transformation to revolutionize oil and gas geoscience, with the goal of transforming exploration workflows and reservoir characterization outcomes.
The discovery of oil in Saudi Arabia in the 1930s marked the beginning of a rich legacy of geological knowledge, resulting in a vast and diverse repository of operational data and geological sample sets. However, the unstructured and variable nature of this legacy data has hindered its full potential, limiting its accessibility and integration. To overcome this challenge, we employed Generative AI (GenAI) technologies to transform the original data into a structured database, utilizing Optical Character Recognition (OCR) models and Retrieval-Augmented Generation (RAG) pipelines. This approach enabled efficient access and utilization of the information, including the application of AI-driven analytics. Concurrently, we digitized and analyzed a vast collection of subsurface rock samples to create a 'Digital Rock,' generating high-fidelity digital twins of the rock samples and predicting rock properties via simulations of physical and chemical processes at the pore scale.
Our digital transformation initiatives have significantly enhanced the efficiency of exploration and petroleum system analysis, saving thousands of man-hours in the search and retrieval of geological information. The Digital Rock has unlocked novel insights into complex petroleum system elements, including microporous, bioturbated, thin-bedded, tight sandstones, hot shales, and other challenging formations. Two exemplary case studies demonstrate the potential of digital transformation: (1) the regularization of over 90 years of biostratigraphic analysis datasets using GenAI, and (2) the assessment of flow heterogeneity in mega-porous reservoirs utilizing a multi-scale digital imaging workflow. These applications underscore the transformative potential of digital technologies in geoscience, enabling more accurate and efficient reservoir characterization.
The integration of rock sample images with corresponding interpretation reports has enabled the creation of a comprehensive and robust database, providing a solid foundation for training and validating AI models. This multidisciplinary approach has significantly enhanced our understanding of the subsurface environment. The synergistic relationship between geological data, interpretation, and AI-modeling has the potential to revolutionize the field of geoscience, enabling more informed decision-making for effective exploration and field development strategies.
Co-author/s:
Mohammed Sadah, Lead Geologist, Saudi Aramco.
Abid Bhullar, Senior Geologist Consultant, Saudi Aramco.
The discovery of oil in Saudi Arabia in the 1930s marked the beginning of a rich legacy of geological knowledge, resulting in a vast and diverse repository of operational data and geological sample sets. However, the unstructured and variable nature of this legacy data has hindered its full potential, limiting its accessibility and integration. To overcome this challenge, we employed Generative AI (GenAI) technologies to transform the original data into a structured database, utilizing Optical Character Recognition (OCR) models and Retrieval-Augmented Generation (RAG) pipelines. This approach enabled efficient access and utilization of the information, including the application of AI-driven analytics. Concurrently, we digitized and analyzed a vast collection of subsurface rock samples to create a 'Digital Rock,' generating high-fidelity digital twins of the rock samples and predicting rock properties via simulations of physical and chemical processes at the pore scale.
Our digital transformation initiatives have significantly enhanced the efficiency of exploration and petroleum system analysis, saving thousands of man-hours in the search and retrieval of geological information. The Digital Rock has unlocked novel insights into complex petroleum system elements, including microporous, bioturbated, thin-bedded, tight sandstones, hot shales, and other challenging formations. Two exemplary case studies demonstrate the potential of digital transformation: (1) the regularization of over 90 years of biostratigraphic analysis datasets using GenAI, and (2) the assessment of flow heterogeneity in mega-porous reservoirs utilizing a multi-scale digital imaging workflow. These applications underscore the transformative potential of digital technologies in geoscience, enabling more accurate and efficient reservoir characterization.
The integration of rock sample images with corresponding interpretation reports has enabled the creation of a comprehensive and robust database, providing a solid foundation for training and validating AI models. This multidisciplinary approach has significantly enhanced our understanding of the subsurface environment. The synergistic relationship between geological data, interpretation, and AI-modeling has the potential to revolutionize the field of geoscience, enabling more informed decision-making for effective exploration and field development strategies.
Co-author/s:
Mohammed Sadah, Lead Geologist, Saudi Aramco.
Abid Bhullar, Senior Geologist Consultant, Saudi Aramco.
