The deposition of solids in the wellbore is a widely known problem encountered during drilling, completion, and intervention operations. Innovative cleanout fluids alleviate problems associated with inefficient hole cleaning. Various methods are known to provide efficient cleanout; however, their effectiveness drops as the wellbore geometry becomes more complex. One of the innovative ways of resolving this issue is to add fibers to the drilling fluid and improve its lifting capacity by reducing the settling velocity of the solids. This study is aimed at evaluating the cleanout performance of fibrous fluids in horizontal wells using a large-scale flow loop. The flow loop has a 48-ft long annular test section that has a 5-inch wellbore and 2.375-inch drill pipe. In each experiment, the solids bed is initially formed in the annulus with natural sand that has similar particle size distribution as solids found in oil and gas wells. Low-viscosity and high-viscosity polymeric suspensions were used with and without fiber. During the investigation, flow rate and fluid rheology were varied while measuring the equilibrium bed height. The bed height was measured starting at the lowest flow rate. Then, the flow rate was increased step by step until the bed was completely cleaned. The measured test parameters include flow rate and pressure loss, which were monitored and recorded using the data acquisition system. The pipe section was used to monitor the rheology of the fluids before the completion of the test. The results demonstrated the effectiveness of fiber in improving hole cleanout in horizontal wellbores. When a small amount of fiber (0.04% wt.) was added, the cleanout performance of the high-viscosity fluid did not show a noticeable change while that of the low-viscosity improved significantly. Even though the addition of fiber had minimal impact on the rheological characteristics of the fluids, the fiber improved the solid carrying capacity of the low viscosity fluid. In the horizontal configuration, the cleaning process of solids requires resuspension of deposited solids and transportation of the suspended particles from one location to another. Fiber particles are more effective in improving the carrying capacity of the fluids than their resuspension capability. Hence, the low-viscosity fluid which has a better resuspension capacity as compared to the high-viscosity fluid exhibited better cleanout performance when the same amount of fiber was introduced. In the absence of pipe rotation during coiled tubing operations, wellbore cleanout after drilling and milling operations require efficient and optimized hole cleaning, especially in horizontal wells. The presented large-scale flow loop study provides a unique analysis of the dynamics of hole cleanout in horizontal wells using novel fibrous fluids.

The deposition of rock cuttings is a problem commonly faced during drilling, completion, and intervention operations. Using polymer-based fluids is a common technique to improve horizontal downhole cleaning. However, these fluids cannot always guarantee an efficient wellbore cleanout. One way to enhance cleanout efficiency is by rotating the drill pipe to mitigate the settling of solids and facilitate their removal. However, drill string rotation often increases equivalent circulating density (ECD). Therefore, this study explores how the impact of rotation on hole cleaning can be synergized by using fibrous fluids to perform cleanout at reduced rotational speeds with limited impact on ECD. The flow loop utilized for this study consists of a 48-ft long eccentric annular (5" × 2.375") test section. Each experiment began by forming a stationary bed of natural sand (average diameter of 1.2 mm) in the test section. High-viscosity and low-viscosity polymer-based suspensions with and without fibers were used. Drill pipe rotation speed was varied from 0 to 150 rpm. In each experiment, the flow rate was increased from 35 to 195 gpm stepwise. The equilibrium bed perimeter was measured at every test flowrate until a complete bed cleanout was achieved. As part of the test, flow rate, pressure gradient, and equilibrium bed height were measured. Rotational viscometers were used to measure fluid rheology before and after each test. Fiber particles improve the carrying capacity of the fluid by reducing the settling of solids and minimizing the re-deposition of particles. The results demonstrate the effectiveness of fiber in synergizing the effect of pipe rotation on hole cleanout performance in horizontal wellbores. The impact of fiber is more pronounced when used with low viscosity fluid. The cleanout performance of the low-viscosity fluid amplified significantly with rotation, almost entirely cleaning the bed at 75 gpm and a rotational speed of 50 RPM, compared to more than 195 gpm without rotation. Even more improvement could be achieved by adding a small amount of fiber (0.04% wt/wt). Furthermore, the fiber improved the cleanout performance of the high-viscosity fluid. The enhancement, however, was not as noticeable as with the low-viscosity fluid. In general, rotation combined with low-viscosity fibrous fluid exhibits the most effective sand cleaning performance. Rotating the pipe re-suspends the settled solids, which are then carried away by the fluids. Fiber particles enhance the fluid's ability to carry solids.

Summary 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, predicting the cleaning performance of fluids becomes more challenging. As a result, optimization is often difficult. Therefore, this research focuses on developing a data-driven model for predicting hole cleaning in deviated wells to optimize drilling performance. More than 500 flow loop measurements from eight 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. Experiments ranged in test-section length from 22 to 100 ft, in hole diameter from 4 to 8 in., and in pipe diameter from 2 to 4.5 in. The experiments provided measured equilibrium bed height at a specific flow rate for various fluids, including water-based and synthetic-based fluids and fluids containing fibers. Several relevant test parameters, including fluid and cutting properties, well inclination, and drillstring rotation speed (drillpipe rev/min), were also considered in the analysis. The collected data have been analyzed using the Cross-Industry Standard Process for Data Mining. This paper is unique because it systematically evaluates various ML models for their ability to describe hole cleanout processes. Six different ML techniques: boosted decision tree (BDT), random forest (RF), linear regression, multivariate adaptive regression spline (MARS), neural networks, and support vector machine (SVM) 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 drillstring rotation. Finally, using the ML model, a parametric study has been conducted to examine the impact of various parameters on the cleanout performance of selected fluids. The results show the relative influence of different variables on the prediction of cuttings bed. Accordingly, flow rate, drillpipe rev/min, 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 BDT algorithm has provided the most accurate prediction with an R2 of 92%, a root-mean-square error (RMSE) of 0.06, and a 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.

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.

Summary During drilling, completion, and intervention operations, solids can deposit in the wellbore. Innovative cleanout fluids reduce the problems associated with inadequate hole cleaning. Various methods have been developed to improve hole cleaning, but their effectiveness decreases as the wellbore inclination increases. One way to solve this issue is to add fibers to the drilling fluid and reduce the settling velocity of the solids to improve the fluid’s lifting capacity. The purpose of this study is to evaluate the cleanout performance of fibrous fluids in horizontal wells using a large-scale flow loop. Thus, flow loop experiments were conducted to assess the impact of fiber on equilibrium bed height. The experiment measures equilibrium bed height and pressure loss in an eccentric annular test section. During the investigation, the flow rate and apparent viscosity of the fluid and fiber length were varied. The results demonstrate the effectiveness of long fiber (length = 0.5 in.) in improving hole cleanout in horizontal wellbores. When a small amount (0.04% wt.) of long fiber was added, the cleanout performance of the high-viscosity fluid did not show a noticeable change. In contrast, the performance of the low-viscosity fluid improved. Even though adding fiber has minimal impact on the apparent viscosity of the fluids, the long fiber improved the cleaning performance of the low-viscosity fluid. Hole cleaning is challenging in operations such as coiled tubing (CT) in which rotating the drillstring is impossible. Hence, this study focuses on cleanout operations in horizontal wellbores without drillstring rotation. The novelty of this work lies in demonstrating how the adjustment of fluid viscosity can positively impact the hole cleaning performance of fibrous fluids in the absence of pipe rotation. The study also presents a new approach to modeling the effects of solids bed irregularity on wellbore pressure loss and equivalent circulating density (ECD).