Prediction performance advantages of deep machine learning algorithms for two-phase flow rates through wellhead chokes

Abstract

Two-phase flow rate estimation of liquid and gas flow through wellhead chokes is essential for determining and monitoring production performance from oil and gas reservoirs at specific well locations. Liquid flow rate (Q _L) tends to be nonlinearly related to these influencing variables, making empirical correlations unreliable for predictions applied to different reservoir conditions and favoring machine learning (ML) algorithms for that purpose. Recent advances in deep learning (DL) algorithms make them useful for predicting wellhead choke flow rates for large field datasets and suitable for wider application once trained. DL has not previously been applied to predict Q _L from a large oil field. In this study, 7245 multi-well data records from Sorush oil field are used to compare the Q _L prediction performance of traditional empirical, ML and DL algorithms based on four influencing variables: choke size (D ₆₄), wellhead pressure (P _wh), oil specific gravity (γ _o) and gas–liquid ratio (GLR). The prevailing flow regime for the wells evaluated is critical flow. The DL algorithm substantially outperforms the other algorithms considered in terms of Q _L prediction accuracy. The DL algorithm predicts Q _L for the testing subset with a root-mean-squared error (RMSE) of 196 STB/day and coefficient of determination (R ²) of 0.9969 for Sorush dataset. The Q _L prediction accuracy of the models evaluated for this dataset can be arranged in the descending order: DL > DT > RF > ANN > SVR > Pilehvari > Baxendell > Ros > Glbert > Achong. Analysis reveals that input variable GLR has the greatest, whereas input variable D ₆₄ has the least relative influence on dependent variable Q _L.