New technology from Perm Polytechnic scientists allows for real-time analysis of the state of oil reservoirs. The forecast accuracy reaches 99.7%, which eliminates the need to stop production for research and saves companies from multi-million dollar losses.
Every year, mining companies suffer colossal financial losses amounting to billions of rubles. The main reason is regular shutdowns of wells for hydrodynamic and geophysical research. Each such downtime means a direct shortage of significant volumes of oil.
A solution to this large-scale problem has been found by Perm Polytechnic scientists. They created a unique software package based on artificial intelligence. The key advantage of the technology is the ability to predict the permeability of oil reservoirs with a phenomenal accuracy of 99.7% without the slightest interruption of the production process.
The intelligent system was based on an extensive database, which included 3,500 hydrodynamic well studies. This body of information was carefully systematized and used to train machine learning algorithms. For each study, a whole range of critical parameters were taken into account: key reservoir and oil properties, operational indicators recorded before production was stopped, and detailed dynamics of pressure recovery.
This integrated approach allows the AI model to identify complex dependencies that are not obvious to human analysis and, based on them, build ultra-accurate forecasts. The complex continuously processes incoming data, providing specialists with up-to-date information about the state of the formation. This not only maintains production continuity, but also significantly improves management efficiency deposit.
The implementation of this development opens a new stage of development for the oil and gas industry, moving it to a qualitatively new level of technology. Companies receive a powerful tool to minimize financial losses and optimize the entire production process.
The material was prepared with the support of the Russian Ministry of Education and Science as part of the Decade of Science and Technology.
Source: @gis_proxima
