Moscow, April 15 (SANA) Russian researchers have developed a machine learning model designed to improve oil extraction efficiency by identifying key reservoir properties with up to 90 percent accuracy.
According to a report by RT, scientists from the Moscow Institute of Physics and Technology, in cooperation with the Russian Academy of Sciences and the Tyumen Scientific Oil Centre, said the model can accelerate the identification of optimal water salinity and gas composition in oil reservoirs.
These processes have traditionally required time-consuming and costly laboratory testing.
The researchers said the system reduces uncertainty associated with conventional forecasting methods, which can have error margins of up to 40 percent, while shortening testing times through the use of advanced algorithms.
They added that the model shows potential for use in enhanced oil recovery techniques, including carbon dioxide injection into geological formations.
Initial experiments have produced encouraging results, and further applications are being explored, including use in heavy crude oil extraction and the analysis of fluid behavior at the nanoscale.
The development reflects growing use of artificial intelligence in the energy sector to optimize production processes and improve resource management.
R.H
