MSE Based Drilling Optimizer Project for Large National Drilling Contractor

A large GCC National Drilling Contractor is planning to trial a new MSE (Mechanical Specific Energy) Drilling advisory system based in artificial intelligence (AI) on a conventional drilling rig. This system works by means of calculating the optimal auto driller input parameters to achieve higher drilling efficiency. The objective of the MSE based Drilling Optimization system is to drive drilling efficiency by way of advising surface drilling parameters (WOB and RPM). The system will be tuned to advise/automatically modify WOB and RPM for each specific run, section, or well. The parameters can be adjusted in one of the two following ways: Advisory Mode: A recommended WOB and RPM value is sent to the driller, who then manually applies the setpoint change Control Mode: The setpoints are sent to the automatic driller for instant and automated application The system shall enable reduction of NPT and rig days to drill wells by increasing efficiency with consequent cost reduction efficiency by utilizing advanced elements of Artificial Intelligence (AI).

Developing a new approach for the anticipation of subsurface pressure in three oil wells from various fields in Iraq

Subsurface pressure quantification is the main goal for all drilling engineers due to its vital importance to minimizing drilling costs and preventing various excavating dilemmas such as stuck pipe, pipe collapsing, lost circulation, and well kick. Formation pressure is measured directly in producing hydrocarbon zones using special equipment; however, it is difficult to obtain pore pressure measurements in other intervals; therefore, many approaches were suggested and developed to anticipate the geopressure due to its importance on the drilling operation. Traditionally, Eaton's method is the most widely used for geopressure anticipation from well logs. In the current study, a new proposed formula for geopressure determination was derived from the modified Eaton's equation which is based on the inception of the mechanical specific energy. The study was applied on three oil-producing wells from different fields in Iraq (Missan C, West Qurna 15, and Zubair 171). The estimated subsurface pressure from the new suggested approach was validated by comparison with the actual in situ subsurface pressure obtained from Drill Stem Test (DST) and Repeated Formation Test (RFT). Statistical analysis is used for the validation by applying Mean squared error (MSE) and Mean Absolute percentage Error (MAE). Encouraging results were obtained from the present research for all wells being studied. The main finding of the current work is involving a new parameter (mechanical specific energy) for subsurface pressure determination, where it was not included in the previous techniques of geopressure equations. It was found that the value of exponent (m) in the new proposed equation has a significant influence on the predicted geopressure, where (m) varies from 0.1626 to 0.6896 depending on the type of the excavated rock plus the materials that the drill bit is manufactured of, where the interaction between these components is vital. The present work could be a useful method when planning to excavate a new well adjacent to the area of the investigated wells, where special well logs are unavailable for pore pressure measurement and suitable usage of mud weight is highly needed for the overall drilling process.

An Experimental Investigation of Drilling Performance Improvement Using Reaming While Drilling

This work investigates the drilling performance by reaming while drilling (RWD) using a dual-body bit and compared it with conventional drilling by a standard drilling bit. The dual-body bit consisted of a 2.45-in pilot bit located at a short distance ahead of a 2.47*3.97-in reamer. Conducting a series of drilling experiments at a simulation drilling rig with full monitoring sensors, we further studied the drilling performance as a function of the distance between the pilot bit and the reamer which affect mud diffusion and the resultant change in pore pressure and stress. A method was devised to eliminate the drill-string vibration and its effect on the drilling performance and the energy consumed. The mechanical specific energy (MSE) calculated for each case was considered as a drilling performance indicator. Using two laboratory experiments as well as analytical thermo-poro-elastic calculations of the Mechanical Specific Energy (MSE), the MSE changes were monitored and recorded. Comparison of this drilling performance indicator was used in both the RWD and the conventional drilling assembly to analyze the effect of RWD. Based on the results, with increasing the distance between the pilot bit and reamer, there is an increase in improvement of drilling performance in terms of MSE reduction. The best drilling performance indicator (MSE reduction of 84%) was observed with the distance between the pilot bit and the reamer of 43.3 cm. This is considered a novel finding in reaming while drilling.

Estimating Formation Tops while Drilling Using Rate of Penetration ROP and Mechanical Specific Energy MSE

Accurate determination of formation tops while drilling is a critical part of exploration geology workflow. Operational decisions on coring, wireline logging, casing, and final well depth largely depend on it. One of the commonly used methods for picking formation tops while drilling is to correlate the rate of penetration (ROP) of the new well to wireline logs from offset wells where there is no logging while drilling (LWD) data. Picking formation tops based on only ROP from a new well can result in picking the wrong formation tops. To improve the workflow and outcome, this paper proposes the combination of ROP and Mechanical Specific Energy (MSE) for estimating formation tops while drilling. MSE is a measure of the energy required to crush or drill through a unit volume of rock. Because MSE is related to rock strength, it can be correlated to changes in lithofacies and formation tops. There are three key steps necessary for utilizing mechanical specific energy to estimate formation tops. First, select the input drilling data relevant to the applicable MSE equation. There are several empirical equations in the literature which can be used for estimating MSE. Input data are ROP, Weight on Bit (WOB), Bit Size (BS), Rotation Per Minute (RPM), and Torque (TORQ) from both the offset wells and the new well. Second, utilize a predetermined empirical equation to estimate MSE. Third, correlate MSE and ROP from the new well to both MSE, ROP, and wireline logs from offset wells (where available) to determine formation tops in the new well. Application of the proposed workflow to two wells show 1) distinct bed boundaries, which agree with formation tops picked using wireline logs; (2) that including MSE increases confidence and reliability of the data and makes it easy to identify the different formation boundaries based on the observed features of both MSE and ROP in the new well; and (3) that MSE variations are sensitive to formation strength, which may indicate rock mechanical changes and formation heterogeneity. This paper presents an alternative method of picking formation tops using MSE and ROP while drilling. The preliminary results based on the two test wells showed over 95% match with those picked using wireline logs of the same new well. As a result, this workflow enhances the ability of geoscientists to correlate subsurface geological features, reduces the uncertainty associated with picking formation tops, casing, and coring depths. Furthermore, it improves the confidence in the result, enhances the quality of operational decisions, and reduces the non-productive time (NPT) and well-cost.

Utilizing High-Frequency In-Bit Sensor Data Improves Drillbit Design and Modelling

Modern drill bits designs have become more efficient using static modelling, and in more advanced cases, time-based dynamic modelling. These methods have created improved cutting structures that fail rock more effectively, however, at-bit vibrations are difficult to estimate because of the high-frequency nature of the vibration and its proximity to typical vibration sensors. In conventional applications, vibration is not measured near the bit. A solution to capture this data on conventional assemblies and use the data in an actual bit design is presented in this paper with subsequent performance and vibration results. The relative efficiency, bit dull grading, and vibration performance are compared across these designs and explored in depth. This new generation of vibration tool fits inside the bit pin, enabling accurate at-bit vibration measurements by a suite of sensors. The tool includes a tri-axis accelerometer that measures lateral and axial acceleration, and gyro sensors to measure rpm and torsional acceleration. Together, these outputs combine with the rig surface data to have time- and depth-based vibration data in the context of the run. When used to quantify the dynamic model, this represents a modelling calibration that improves bit design performance. The lower-vibration environment created by the new bit design enables the operator to run increased parameters with a lower likelihood for measurement-while-drilling (MWD) failures, motor failures, and premature catastrophic bit failures leading to faster run times and less nonproductive time (NPT). These results also prove that meaningful bit design changes can take place more frequently than through traditional means, translating value to the operator in the form more successful BHA improvements and less drilling time. Using the new in-bit sensor in a baseline design to start the design cycle, a baseline mechanical specific energy (MSE) and vibration model was developed foot-by-foot. The worst areas of vibration were seen as the bit became dull in the lower section of the drilling interval. A new dull bit model was created in parallel to capture this section of data. A new design was proposed to Whiting Petroleum to improve both sharp and dull efficiency and vibration, and subsequently run with sensor in an offset well.