Estimating oil recovery factor using machine learning: Applications of XGBoost classification

In petroleum engineering, it is essential to determine the ultimate recovery factor, RF, particularly before exploitation and exploration. However, accurately estimating requires data that is not necessarily available or measured at early stages of reservoir development. We, therefore, applied machine learning (ML), using readily available features, to estimate oil RF for ten classes defined in this study. To construct the ML models, we applied the XGBoost classification algorithm. Classification was chosen because recovery factor is bounded from 0 to 1, much like probability. Three databases were merged, leaving us with four different combinations to first train and test the ML models and then further evaluate them using an independent database including unseen data. The cross-validation method with ten folds was applied on the training datasets to assess the effectiveness of the models. To evaluate the accuracy and reliability of the models, the accuracy, neighborhood accuracy, and macro averaged f1 score were determined. Overall, results showed that the XGBoost classification algorithm could estimate the RF class with reasonable accuracies as high as 0.49 in the training datasets, 0.34 in the testing datasets and 0.2 in the independent databases used. We found that the reliability of the XGBoost model depended on the data in the training dataset meaning that the ML models were database dependent. The feature importance analysis and the SHAP approach showed that the most important features were reserves and reservoir area and thickness.

Estimating oil and gas recovery factors via machine learning: Database-dependent accuracy and reliability

With recent advances in artificial intelligence, machine learning (ML) approaches have become an attractive tool in petroleum engineering, particularly for reservoir characterizations. A key reservoir property is hydrocarbon recovery factor (RF) whose accurate estimation would provide decisive insights to drilling and production strategies. Therefore, this study aims to estimate the hydrocarbon RF for exploration from various reservoir characteristics, such as porosity, permeability, pressure, and water saturation via the ML. We applied three regression-based models including the extreme gradient boosting (XGBoost), support vector machine (SVM), and stepwise multiple linear regression (MLR) and various combinations of three databases to construct ML models and estimate the oil and/or gas RF. Using two databases and the cross-validation method, we evaluated the performance of the ML models. In each iteration 90 and 10% of the data were respectively used to train and test the models. The third independent database was then used to further assess the constructed models. For both oil and gas RFs, we found that the XGBoost model estimated the RF for the train and test datasets more accurately than the SVM and MLR models. However, the performance of all the models were unsatisfactory for the independent databases. Results demonstrated that the ML algorithms were highly dependent and sensitive to the databases based on which they were trained. Statistical tests revealed that such unsatisfactory performances were because the distributions of input features and target variables in the train datasets were significantly different from those in the independent databases (p-value < 0.05).