Assaad, Wissam (Shell Global Solutions International B.V.) | Di Crescenzo, Daniele (Shell Exploration & Production Company) | Murphy, Darren (Shell Exploration & Production Company) | Boyd, John (Shell Exploration & Production Company)
In this paper, we present a method of modeling surge pressures and wave propagation that can occur during well execution. The surge pressures have an effect on formations [i.e., formation fracture resulting in mud losses and nonproductive time (NPT)]. Knowing the amplitude of surge pressure in advance can lead to operation redesign to avoid losses. Swab- and surge-pressure waves can occur at numerous events during well execution. For example, during liner operations, pressure waves can occur at dart landing or plug shearing, liner-hanger setting, or clearing a plugged shoe-track component. It is possible for surge-pressure waves to create fractures in shale and sand layers (i.e., when surge-pressure-wave amplitude exceeds formation fracturing resistance).
A transient-state physical model is built to compute pressure-wave propagation through drillstring, casing, and open hole to predict the amplitude of a surge-pressure wave and to warn when a fracture might occur in the formation, to avoid mud losses and NPT.
In the model, continuity and energy partial-differential equations (PDEs) are built for a cylindrical fluid element contained in an elastic hollow cylinder. The method of characteristics is applied to convert the PDEs to ordinary-differential equations (ODEs). The ODEs are solved numerically to compute pressure distribution along well depth and in time. The model is implemented as a graphical-user-interface (GUI) tool to be used by drilling engineers at the design phase of a well to avoid losses. The GUI tool is targeted to address different scenarios that take place during the cementation process. To date, the transient-state physical model has been applied successfully in various applications, such as monodiameter technology, running casing, and perforating operations. Two cases are studied, one for a well in the Gulf of Mexico (GOM) where mud losses have been reported, and the other for a well in Malaysia where no mud losses have occurred. Pressure-wave computations are performed with the GUI tool for the two cases. The results of both cases are presented in this paper and show that formation fracture can be predicted by the GUI tool and subsequent losses can be avoided.
Operators face the continuing challenge to improve drilling efficiency for cost containment, especially in deepwater drilling environments where drilling costs are significantly higher. Innovative drilling technologies have been developed and implemented continuously to support the initiative. In many areas of the world, including the Gulf of Mexico (GOM), hydrocarbon reservoirs exist below thick non-porous and impermeable sequences of salt that are considered a perfect cap rock. However, salt poses varied levels of drilling challenges due to its unique mechanical properties.
At ambient conditions, the unconfined compressive strength (UCS) of salt varies between 3,000 to 5,000 psi; however, the strain at failure for salt can be an order of magnitude higher when compared to other rocks. Consequently, during drilling salt's viscoelastic behavior requires that its must be broken with an inter-crystalline or trans-crystalline grain boundary breakage. When compared to other rock types, the unique isotropic nature of salt results in a level of strain that is much higher for the given elastic moduli. This strain level makes salt failure mechanics different from other rock types that are prevalent in the GOM.
Hybrid bits combine roller-cone and polycrystalline diamond compact (PDC) cutting elements to perform a simultaneous on-bottom crushing / gouging and shearing action. Two divergent cutting mechanics pre-stresses the rock and apply high strain for deformation and displacement, resulting in highly efficient cutting mechanics. To meet the drilling objectives, different hybrid designs have been implemented to combine stability and aggressiveness for improved drilling efficiency. An operator, while drilling salt sections at record penetration rates, has successfully used this innovative process of rock failure utilizing the dual-cutting mechanics of hybrid bits. This has resulted in significant value additions for the operator.
This paper analyzes field-drilling data from successful GOM wells and attempts to correlate salt failure mechanics and provide insight into dual-cutting mechanics and its correlation with salt failure. The paper also reviews the drilling mechanics of hybrid bits in salt and highlights importance of dual-cutting mechanics for achieving higher penetration rates in salt through improved drilling efficiency.
Human factors are identified as the major contributor to oil and gas drilling and other operations related accidents. Offshore oil and gas operations involve complex scenarios and decision-making with potentially catastrophic consequences. The current simulation-based training modules are often criticized for their lack of objective and validated measures for human factors and non-technical skills. There is also a need to include measures for enhanced situational awareness and decision-making for the offshore drilling crew. In this study, we present holistic human-centered training framework equipped with assessment techniques to analyses situational awareness of partcipants in customized well-control operations.
The training exercise used in this work included real-time well control operation customized for drilling break and kick detection scenarios. The assessment approach consisted of eye-tracking data analysis, questionnaire analysis, checklist score analysis, and communication log analysis. After individual analysis from each technique, a new framework was developed to triangulate results from each technique to provide a comprehensive assessment. The participants included seven group of novices and one group of experts. The preliminary results indicate significant differences between the situation awareness and performance of participants. Furthermore, there were observed notable differences between the perceptual, comprehensive, and projection ability of novices and experts in routine jobs on a drilling platform. The eye-tracking data features included fixation count and fixation duration, and it was inferred that eye-tracking results can be representative of cognitive abilities of the partcipants. Furthermore, the fixation count and duration results were highly correlated with the checklist scores.
Overall, the adopted methodology in this study have potential to open new avenues for human- centered training framework and improvement in traditional assessment approach. Furthermore, it can also be helpful in understanding of cognitive responses of the offshore professionals.
An operator and rig contractor have been implementing drilling operations automation (DOA) pursuing well design and drilling operational execution improvements in terms of safety, quality, delivery, and cost (SQDC). Today, drilling automation enables tighter process control of operations and well design stakeholders are beginning to fully understand and anticipate its value.
DOA requires applying a process control approach and defining well construction processes at a very detailed level. This process control approach is proposed as a method to study and improve work steps and integrate them into overall operational activities. Optimizing, much less controlling, a drilling system is a difficult task with a multitude of variables to manage. The process of automating operations may be one of the best tools to reduce the number of unknown variables and better deliver consistent SQDC results.
Automation case studies such as a downhole Weight on Bit (WOB) drilling system, a directional drilling advisory system, a sliding system for conventional steerable mud motors, and an integrated tubular running system are described to highlight the role of automation in assisting operators and contractors to efficiently manage and improve the well construction process. Process automation requires improvements in foundational systems, tools, and data quality to support operational performance. The most significant finding is how automated systems enable operations to be practically managed at a detailed level by drilling personnel, engineers, and other stakeholders. After practices and systems are proven and automated, they can be scaled and managed over an entire rig fleet. This will ultimately enable today's well construction and drilling system related risks to be mitigated and managed, leading to further SQDC rewards with more efficient well designs.
The operator and rig contractor will share perspectives for realizing value and opportunity through applying DOA. Experience shows DOA-influenced standardized operations can result in eliminating steps that are no longer needed. Automation enables changes to well design that are just beginning to be understood and anticipated by drilling teams. The challenge will be linking these opportunities to pursue new capabilities supporting well design improvement. This will be the true benefit from automating drilling operations.
Antipova, Ksenia (Skolkovo Institute of Science and Technology, Digital Petroleum) | Klyuchnikov, Nikita (Skolkovo Institute of Science and Technology, Digital Petroleum) | Zaytsev, Alexey (Skolkovo Institute of Science and Technology, Digital Petroleum) | Gurina, Ekaterina (Skolkovo Institute of Science and Technology, Digital Petroleum) | Romanenkova, Evgenia (Skolkovo Institute of Science and Technology, Digital Petroleum) | Koroteev, Dmitry (Skolkovo Institute of Science and Technology, Digital Petroleum)
Majority of the accidents while drilling have a number of premonitory symptoms notable during continuous drilling support. Experts can usually recognize such symptoms, however, we are not aware of any system that can do this job automatically. We have developed a Machine learning algorithm which allows detecting anomalies using the drilling support data (drilling telemetry). The algorithm automatically extracts patterns of premonitory symptoms and then recognizes them during drilling.
The machine learning model is based on Gradient Boosting decision trees. The model analyzes real time drilling parameters within a sliding 4-hour window. For each measurement, the model calculates the probability of an accident and warns about anomaly of particular type, if the probability exceeds the selected threshold.
Our training sample comes from 20+ oilfields and consists of sections related to 80+ accidents of the following types: stuck pipe, mud loss, gas-oil-water show, washout of pipe string, failure of drilling tool, packing formation, that occurred while drilling, trip-in, trip-out, reaming; the sample also includes more than 700 sections without accidents.
We have designed the prediction model to work during drilling new wells and to distinguish the normal drilling process from the faulty one. One can configure the anomaly threshold to balance amount of false alarms and the number of missed accidents.
To evaluate quality of the model we measure data science metrics and check an industry-driven criterion. The model can identify 40 accidents from about 80 with high confidence, whereas for the others there is still a room for improvement. Our findings suggest that including more accidents of underrepresented types will improve quality. Other data science metrics also support aptitude of the model. Finally, having data from multiple heterogeneous oilfields, we expect that the model will generalize well to new ones.
This paper presents a good practice of development and implementation of a data-driven model for automatic supervision of continuous drilling. In particular, the model described in the paper will assist specialists with drilling accidents prediction, optimize their work with data and reduce the nonproductive time associated with the accidents by up to 20%.
Directional drilling for hydrocarbon exploration has been challenged to become more cost-effective and consistent with fast-growing drilling operations for both offshore and onshore production areas. Autonomous directional drilling provides a solution to these challenges by providing repeatable drilling decisions for accurate well placement, improved borehole quality, and flexibility to adapt smoothly to new technologies for drilling tools and sensors. This work proposes a model predictive control (MPC)-based approach for trajectory tracking in autonomous drilling. Given a well plan, bottomhole assembly (BHA) configuration, and operational drilling parameters, the optimal control problem is formulated to determine steering commands (i.e., tool face and steering ratio) necessary to achieve drilling objectives while satisfying operational constraints. The proposed control method was recently tested and validated during multiple field trials in various drilling basins on two-and three-dimensional (2D and 3D) well plans for both rotary steerable systems (RSS) and mud motors. Multiple curve sections were drilled successfully with automated steering decisions, generating smooth wellbores and maintaining proximity with the given well plan.
Eustes III, Alfred W. (Colorado School of Mines) | McKenna, Kirtland I. (Colorado School of Mines) | Zody, Zach J. (Colorado School of Mines) | Healy, Carl (Colorado School of Mines) | Lang, Camden (XTO) | Joshi, Deep (Colorado School of Mines) | Yow, Stephen (Chevron) | McGowen, Kyle (Shell)
Drilling education must evolve continuously to keep up with the changes in the drilling industry. Part of that evolution includes the addition of data analytics in drilling operations. In addition, having a "hands on" experience of actual drilling operations is an important part of the drilling engineering educational process. At the Colorado School of Mines, both goals are achieved using a new coring rig equipped with a high-frequency data acquisition system.
A Sandvik DE 130 Diamond Coring Rig was acquired by the school through a grant from Apache Corporation that has proven to be an excellent analog to full-scale petroleum rigs. It has all drilling subsystems such as rotary, hoisting, power, and circulation. A data acquisition system has been added that tracks accelerations as well as various drilling operational parameters. During experiments, each student has an opportunity to operate the driller's controls and experience the complexities associated with drilling operations including the occasional error. The retrieved core helps the student correlate the formation with drilling data.
The inclusion of the drilling experience in the curriculum has benefited the students in several aspects. This experience has helped students visualize drilling operations and understand complexities and challenges associated with drilling. During the drilling operations, if any problems arise, the students have a chance to troubleshoot those problems in real-time and apply their theoretical knowledge. Operational safety and stop work authority are also a focus and demonstrated by students. This is likely to be the first experience most students have with high-frequency drilling data analysis. Monitoring, collecting, and handling real high-volume data gives a first glimpse into the complexities of data analytics. Noisy realtime data and errors are real and observed by the students. They also learn to handle and analyze high- frequency drilling data identifying normal trends and abnormalities. This coring rig has enhanced the drilling engineering education and data analysis skills of our students.
This work outlines a novel teaching methodology that combines the practical understanding of drilling and the application of data analytics. Getting out to the field and actually drilling rock has enhanced our drilling curriculum to align it with the latest industry practice and to educate future drilling engineers.
A mathematical model is developed to capture the dynamic features in the wellbore during drilling operations so that it could be used for real-time computations. The model comprises one-dimensional (1D) mud flow solvers, one for the drillpipe and the other for the wellbore annulus including the volume below the drill bit, integrated point models for the bell nipple, bottomhole assembly (BHA) nozzles, 1D shallow water model for the flowline, and point model for the bypass replicating the hydraulic circuit in the drilling rig. The model assumes compressibility of mud for the wellbore section along with its transient gel characteristics. The equations are solved using appropriate explicit numerical solvers and the results capturing the fast transients of the standpipe pressure, bottomhole equivalent circulating density (ECD), and the flow rates during dynamic drilling operations are presented to illustrate the performance of the model with field data.
The paper provides a technical overview of an operator's Real-Time Drilling (RTD) ecosystem currently developed and deployed to all US Onshore and Deepwater Gulf of Mexico rigs. It also shares best practices with the industry through the journey of building the RTD solution: first designing and building the initial analytics system, then addressing significant challenges the system faces (these challenges should be common in drilling industry, especially for operators), next enhancing the system from lessons learned, and lastly, finalizing a fully integrated and functional ecosystem to provide a one-stop solution to end users.
The RTD ecosystem consists of four subsystems as shown in architecture
RTD ecosystem architecture
RTD ecosystem architecture
All of these subsystems are fully integrated and interact with each other to function as one system, providing a one-stop solution for real-time drilling optimization and monitoring. This RTD ecosystem has become a powerful decision support tool for the drilling operations team. While it was a significant effort, the long term operational and engineering benefits to operators designing such a real-time drilling analytics ecosystem far outweighs the cost and provides a solid foundation to continue pushing the historical limitations of drilling workflow and operational efficiency during this period of rapid digital transformation in the industry.
Ryan, M. (Baker Hughes, a GE Company) | Gohari, K. (Baker Hughes, a GE Company) | Bilic, J. (Baker Hughes, a GE Company) | Livescu, S. (Baker Hughes, a GE Company) | Lindsey, B. J. (Baker Hughes, a GE Company) | Johnson, A. (Murphy Oil Company) | Baird, J. (Murphy Oil Company)
Development of unconventional reservoirs in North America has increased significantly over the past decade. The increased activity in this space has provided significant data with respect to through-tubing drillouts which had previously not been attainable. This paper is focused on using the field data from the Montney and Duvernay formations along with laboratory data and numerical modeling to understand the hole cleanout associated with through-tubing drillouts of frac plugs.
Initially, an extensive full-scale flow loop laboratory testing program was conducted to obtain data on debris transportation for hole cleanout during through-tubing applications. The testing was conducted on various coiled tubing (CT)-production tubing configurations using various solid particles. The laboratory data was used to develop empirical correlations needed for a transient debris transport model. This model was then used for frac plug drillouts to ensure successful hole cleaning in actual field applications. Computational fluid dynamics (CFD) modelling was also used to further understand and quantify the differences between the laboratory data, field data and transient debris transport model results.
The objective of the work conducted was to gain a better understanding of debris transport and validate the empirical modelling approach developed for hole cleaning. The validation process was conducted in several stages. The first stage was to validate the laboratory data against the Montney and Duvernay field data. The second stage was to verify the results obtained from the empirical model against the results obtained from a computational fluid dynamic model. The results from both modelling approaches were lastly compared to the field data. All these results challenge the current industry's understanding and best practices for through-tubing drillouts in the Montney and Duvernay formations. With the contentious increase of lateral lengths and higher stage counts, the process of drilling out frac plugs has become more complex. This study explicitly benefits all operators in their ever-increasing need to understand their frac plug drillout operations to ensure efficient, cost effective, and most importantly, consistent and repeatable results.
While efficient results for frac plug drillout operations have been accomplished to date, the on-going feedback from the field has been the requirement to produce repeatable drillouts. This paper is the first to show a holistic approach for obtaining a transient debris transport model used for through-tubing drillouts of frac plugs. The novelty also consists of the transient debris transport model validation through laboratory data and actual Montney and Duvernay field data.