Intelligent Production Prediction of Deep Offshore Hydrocarbon Reservoir: A Case Study of Niger-Delta Region of Nigeria

Abstract

Current methods for predicting output, such as material balancing and numerical simulation, need years of production history, and the model parameters employed determine how accurate they are. The use of artificial neural network (ANN) technology in the production forecasting of a deep offshore field under water injection/water flooding in Nigeria’s Niger-Delta region is investigated in this study. Oil, water, and gas production rates were predicted using well models and engineering features. Real-world field data from producer and water injection wells in deep offshore is used to test the models’ performance. Ninety percent (90%) of the historical data were utilised for training and validating the model framework before being put to the test with the remaining information. The predictive model takes little data and computation and is capable of estimating fluid production rate with a coefficient of prediction of more than 90%, with simulated results that match real-world data. The discoveries of this work could assist oil and gas businesses in forecasting production rates, determining a well’s estimated ultimate recovery (EUR), and making informed financial and operational decisions.