Application of artificial neural network for predicting production flow rates of gaslift oil wells

In petroleum industry, the prediction of oil production flow rate plays an important role in tracking good performance as well maintaining rate. addition, a modelling with high accuracy will be useful optimizing properties to achieve expected rate, enhance recovery factor and ensure economic efficiency. However, is traditionally predicted by theoretical or empirical models. The model usually gives results wide variation error, this also requires lot input data that might time-consuming costly. models are often limited volume set used construct model, therefore values from applications these practical condition not highly accurate. research, authors propose use artificial neural network (ANN) establish better relationship between predict Using 5 wells which continuous gas lift method X field, Vietnam, ANN system was developed using back-propagation algorithm tansig function above set. This called (BPNN). comparison collected studied wells, constructed achieved very correlation coefficient (98%) low root mean square error (33.41 bbl/d). Therefore, can serve robust tool for oilfield