Abstract
This paper presents an innovative approach to oil field management using digital twin and machine learning technology. A detailed experimental setup utilizing oil displacement techniques was developed, equipped with sensors, actuators, flow meters and solenoid valves. Experiments focused on oil displace- ment with water, polymer and oil, resulting in valuable data. This data played a key role in creating a digital twin model of the oil field. Using the digital twin, ML algorithms were trained to predict oil produc- tion levels, detect potential equipment malfunctions, and prevent operational problems. Our results show a marked improvement of 10-15% in oil production efficiency, emphasizing the transformative potential of combining DT and ML in the oil industry.