Artificial intelligence in regional forecasting of oil and gas potential in the Southern Ustyurt region based on integrated geological and geophysical data
The article considers the technologies of regional prediction of oil and gas content using digital technologies, artificial intelligence and innovative methods of field exploration, processing, interpretation and modeling based on the example of the South Ustyurt region (SUR). The study utilizes data from airborne gamma-ray spectrometry survey (AGRSS), gravity and magnetometry, geothermal studies, structural models for three horizons, and machine learning methods. The application of neural networks, machine learning, inversion, and spatial analysis of geological and geophysical data has improved prediction accuracy and identified prospective areas for geological exploration. The use of IP Seismic modules (SOM_Clustering Seis plug-in, IP_Prediction 1D/2D) allowed for the discovery of hidden patterns in geophysical potential field anomalies. The inversion of gravity and magnetic data enabled the construction of geodensity and geomagnetic models at depths ranging from 100 to 20,000 meters. The hydrocarbon potential forecast was carried out in two stages. The first stage involved self-training of neural networks, which identified cluster centers. In the second stage, potential field attributes and distances from data points to cluster centers were calculated, allowing for the construction of standard deviation maps and the identification of the most promising areas. Comprehensive integration of geological data (structural, lithostratigraphic, geochemical, thermobaric, and other factors), statistical classification, and AI made it possible to assess the hydrocarbon potential of the SUR. The Central Ustyurt Dislocation System (CUDS) was identified as having minimal or no prospectivity. In the Assakeaudan Depression, an increase in class differentiation was observed, enhancing its gas and oil potential. On the Shakhpakhty Step, model calculations and classification results indicate active migration processes and hydrocarbon accumulation, making it the most promising area for further exploration using seismic surveys and deep drilling. The study provided a comprehensive picture of the structure of the promising formations within the SUR, revealed patterns in the distribution of radioactive elements, and anomalies in gravitational, geomagnetic, and geothermal fields. It also improved the accuracy of hydrocarbon prospecting forecasts and identified the most promising areas for further exploration.
Сведения
- Журнал
- Neft i gaz
- Опубликовано
- 2025 · № 2, С. 17-37
- Язык
- Английский
- ISSN
- 1562-2932 · 2708-0080
- Издатель
- Oil Research and Engineering Center of the National Engineering Academy of the Republic of Kazakhstan