engleski

Production characteristics and reservoir quality at the Ivanić oil field (Croatia) predicted by machine learning system

At the Ivanić oil field, hydrocarbons are accumulated in fine to medium grained litharenites of the Ivanić-Grad Formation (Iva-sandstones member) of Upper Miocene age. Reservoir rocks are divided into eight depositional (production) units (iI-iVIII). Deposits of each unit are characterized by their own reservoir quality parameters (porosity, horizontal permeability, net pay...). Production characteristics of 30 wells have been studied by a simple statistical method. Two major production well categories ("good producers" and "bad producers") have been found. The contribution of each depositional unit to the total production of an individual well was studied by using a machine learning system which can serve as an expert system shell. The results have shown that: 1) the most important factors affecting the total hydrocarbon production of each well are horizontal permeability and net pay thickness of the depositional units iV, iIV and iIII., 2) after 20 wells had been drilled during the primary recovery period it was possible to predict the production category of a well to be drilled in the future with a reliability of 80%.

Sava depression; Ivanić oil field; reservoir quality; prediction of oil production; machine learning system; expert system

nije evidentirano