Application of Artificial Intelligence Technology in Single Well Production and Water Cut Prediction

Abstract

AbstractAfter the oilfield is fully developed, with the reduction of remaining underground recoverable reserves and the continuous consumption of formation energy, at a certain period of time, there will be a decline in production and an increase in water cut, and the prediction of future production and water cut of a single well in the oilfield is very important for achieving long-term stable production of crude oil and formulating reasonable countermeasures to slow down production decline. This paper comprehensively considers historical data such as single well static geological data, production data, and operation data, and adopts a new generation of artificial intelligence scientific technology such as long and short-term memory neural networks to establish forecast model for single well production decline and water cut in ultra-high water cut stage oil field. The results show that the established model has an average prediction accuracy rate of more than 90% for single well production decline and water cut, thus realizing multi-dimensional single well production decline and water cut prediction, which is accurate and reasonable optimization for the oil well production and provides a strong technical foundation for the effective oil field development.KeywordsSingle well productionWater cutPredictionArtificial intelligenceOilfield developmentLong and short-term memory neural network (LSTM)