Over the past 20 years, the structure of reserves of Russian oil fields has undergone significant changes: with a general increase in reserves, there has been a significant increase in the share of hard to recover reserves (HTR). In the HTR structure of the Russian Federation, low-permeability and heterogeneous reservoirs are identified as the main subgroup (64 %), in 2022, approximately 30 % of all Russian oil was produced from HTR (166 from 535 million tons). In such reservoirs, to maintain and ensure high levels of oil recovery, a workovers such as hydraulic fracturing (HF) is used.Petroleum production companies have various technical and infrastructal capabilities for the selection of candidate wells for workover. Some oil companies use the support of design institutes and service companies that mining andhave specialized geological software products for monitoring and selection of workovers, others make do with the knowledge and skills of selecting their own specialists. However, due to the development of residual recoverable reserves and the unprofitability of the workover every year to select candidate wells, it is becoming increasingly difficult to select candidate wells. Working out the entire well stock and selection candidate wells using standard techniques takes a huge amount of time for a specialist. The article proposes a solution to the topical issue of optimizing the selection of candidate wells for hydraulic fracturing by predicting the dynamics of production wells based on historical data for various geological conditions. The developed method for forecasting the flow rate of liquid and oil on the basis of the rate of change in reservoir pressure and water cut of the well allows you to predict the flow rate for two years ahead, monthly estimates the possible increase from hydraulic fracturing and allows you to determine the optimal time for repeated or primary hydraulic fracturing at the well. This approach sharply narrows the number of potential candidates from a coupe of hundred to two to three dozen potential candidates.
Read full abstract