Applications of Machine Learning Methods to Predict Hole Cleaning in Horizontal and Highly Deviated Wells

Abstract

Machine learning (ML) has become a robust method for modeling field operations based on measurements. For example, wellbore cleanout is a critical operation that needs to be optimized to enhance the removal of solids to reduce problems associated with poor hole cleaning. However, as wellbore geometry becomes more complicated, it gets more difficult to predict the cleaning performance of fluids. As a result, optimization is often challenging. Therefore, this study aims to develop a data-driven model for predicting hole cleaning in deviated wells to optimize drilling performance. More than 500 flow loop measurements from 8 studies are used to formulate a suitable ML model to forecast hole cleanout in directional wells. Measurements were obtained from hole-cleaning experiments that were conducted using different loop configurations. Test sections ranged in length from 22 to 100 feet, in hole diameter from 4 to 8 inches, and in pipe diameter from 2 to 4.5 inches. The experiments provided measured equilibrium bed height at a specific flow rate for various fluids, including water-based and oil-based fluids and fluids containing fibers. Several relevant test parameters, including fluid and cutting properties, well inclination, and drilling string rotation speed, were also considered in the analysis. The collected data has been analyzed using the Cross-Industry Standard Process for Data Mining (CRISP-DM). Six different machine learning techniques (Random Forest, Linear Regression, Neural Networks, Multivariate Adaptive Regression Spline, Support Vector Machine, and Boosted Decision Tree) have been evaluated to select the most appropriate method for predicting bed thickness in a wellbore. Also, we compared the predictions of the selected ML method with those of a mechanistic model for cases without drill string rotation. Finally, using the ML model, a parametric study has been conducted to investigate the impact of various parameters on the cleanout performance of selected fluids. Results show the relative influence of different variables on the prediction of cuttings bed. Accordingly, flow rate, drill string rotation, and fluid behavior index have a strong impact on dimensionless bed thickness, while other parameters such as fluid consistency index, solids density and diameter, fiber concentration, and well inclination angle have a moderate effect. The Boosted Decision Tree algorithm has provided the most accurate prediction with an R-square of approximately 90%, Root Mean Square Error (RMSE) of close to 0.07, and Mean Absolute Error (MAE) of roughly 0.05. A comparison between a mechanistic model and the selected ML technique shows that the ML model provided better predictions.