A Systematic Machine Learning Method for Reservoir Identification and Production Prediction

Wei Liu


This work developed an integrated ML system, formed by two interconnected predictive models. The results of reservoir identification revealed that ensemble techniques (RF, GBDT and XGB) perform better than single classifiers (LR, KNN, DT and ANN). The reservoir identification results of XGB were selected because of the outperformance of XGB in all evaluation metrics. Based on the prediction results of ⅠO and ⅡO obtained from the reservoir identification, the effective thickness (thickness of ⅠO and ⅡO) was an important input used in the production prediction process to predict the cumulative oil production of single wells. The very little difference (0.01 R2) between the prediction results of established ANN / XGB model (based on predicted effective thickness) and corresponding RM Ⅰ (based on real effective thickness) proved that the prediction of reservoir identification was sufficiently accurate and could reliably be used in production forecasting. ANN and XGB models with effective thickness provided higher prediction accuracy than corresponding RM Ⅱ (based on overall reservoir thickness) in training and testing data sets. In testing process, the R2 of XGB and ANN model using effective thickness was 10% and 13% higher than that of RM Ⅱ. XGB was better than ANN with a higher prediction accuracy and faster computing speed. In this study, the research data was mainly based on an oil production data set from CNPC. The integrated ML system has proven successful in the predictive test of CNPC’s subordinate blocks. In the future, introducing diverse data from different regions may improve the ML models and perhaps make the applicable on a global scale.