Scientists of Perm Polytechnic University together with foreign colleagues have developed a unique mathematical model for predicting oil production, studying the behavior of the flow of hydrocarbons through throttles, which are responsible for regulating the pressure of oil supply from the reservoir.
Forecasting oil production rates is the most important parameter for oil engineers. This process is very complicated and takes a lot of time because it is necessary to conduct daily continuous monitoring of well performance. The technology will help to evaluate the remaining raw materials present in productive formations, optimize production operations, manage formations and plan business.
The matmodels developed by scientists are trained and tested individually at oil wells located on the Norwegian continental shelf, in the central part of the North Sea.
Surface chokes are widely used equipment installed at the wellhead to regulate the supply of oil. They minimize various problems that may include jamming of surface equipment, preventing excessive sand extraction and cone formation. Therefore, the matmodel takes into account the flow behavior through the chokes. In addition, the polytechnics used and analyzed popular machine learning methods, as well as field data over the past few years to train and test the model.
Testing of the matmodel showed that the efficiency surpassed other existing, significant methods for predicting production.