2001-2002 OTRC PROJECT: DEVELOPMENT OF A BLOWOUT INTERVENTION METHOD AND DYNAMIC KILL SIMULATOR FOR BLOWOUTS OCCURRING IN ULTRA-DEEPWATER
OBJECTIVE:
Ultra-Deepwater drilling activity has increased dramatically in the last two years. Operations that were once exceptional and characterized by several man-years of well and operations planning, equipment qualification and contingency planning are now being done routinely several times each rig year.
DEA – 63, Floating Vessel Blowout Control, completed in the early 90’s did not contemplate operations in water as deep as we commonly operate in now. While the project did contain a good deal of information, it was not widely available or read within the industry. One reason for this was massive restructuring that continues to take place within the oil business and lack of a publication mechanism to make it available to a wide audience
We propose a project to expand DEA – 63 for application into ultra-deepwater, develop a Visual Basic / Spreadsheet based dynamic kill program for ultra-deepwater and make the document available through the Texas A&M University Press, the International Association of Drilling Contractors, or other means of publication that would best reach the intended audience as either a technical report or handbooks for end users or both.
We propose to expand on DEA – 63 in the following areas:
1) Mechanical intervention – We would update the deepwater intervention methods proposed in DEA – 63 taking into account advancements made in deepwater construction since the late 80’s. We would also evaluate the hydraulic requirements for methods that have been proposed in the past now taking into account the very long sections of pipe necessary to reach the sea bed.
Additional new work would be done in the following areas:
1) Bridging tendencies in ultra-deep water blowouts – Gulf of Mexico and other ultra-deepwater sediments are generally poorly consolidated. Many believe that a high rate ultra-deep water blowout will bridge and self kill. We will investigate the likelihood of this and define the parameters for evaluation of bridging including conditions with open hole drilling and cased hole completions.
2) Dynamic kill investigation of ultra-deepwater blowouts – we would develop a dynamic kill model for deepwater blowouts and investigate methods and pump rates necessary to kill the blowout from the existing well bore or from one or more relief wells.
3) Development of Dual Density blowout control methods – In the event that a deepwater blowout results in loss of the riser or a disconnect it may not be possible or safe to reconnect the riser and divert flow to the surface. If that is the case, dynamic kill could only be accomplished from a relief well using Dual Density mud weights. Furthermore, Dual Density drilling methods are likely to become commercially available in the next two years. It is likely that a well drilled to a formation using Dual Density methods could not be killed by a relief well using any other drilling method. Investigation of dynamic kill with Dual Density drilling will be included in the proposed study.
4) Costs of intervention – We propose to develop a cost estimate template for ultra-deepwater blowout intervention.APPROACH:
The proposed work is a multi-year, multi-phase project and has been broken down into five separate tasks, some of which could be performed independently of each other. Tasks 1, and 2 could be performed concurrently. Except for the literature and data gathering, Task 3 cannot begin until Task 2 is sufficiently complete so that the model could be utilized to validate the methods developed in Task 3. Task 4 cannot be completed until Task 3 is nearly complete. Task 5 will be completed after Tasks 1-4 are complete. A separate budget has been submitted for each task (as well as a travel and meeting budget). Each task is described below.
These five tasks that make up the entire project will be performed in two Phases. Phase I will consist of Tasks 1 and 2, complete. The initial work (literature review and data gathering) on Task 3 will begin in Phase I. Phase II will include the completion of Task 3 as well as Tasks 4 and 5. The U.S. Department of Interior – Minerals Management Service and the Offshore Technology Research Center has funded Phase I in its entirety. Phase II funding must come from industry and other sources through GPRI.PHASE I (MMS/OTRC FUNDING)
Task 1 – Bridging of blowouts in the GOM and tools for evaluation.
High flow rate blowouts sometimes cause the well bore to collapse and bridge. When this occurs the well will often self kill, resulting in probably the fastest and least expensive method of blowout containment. Bridging usually occurs in poorly consolidated sandstones, and reactive shales, which are common in the Gulf of Mexico. This project proposes to study the formations likely to be encountered in ultra-deep waters of the GOM to determine the conditions in which well bores will collapse and bridge. The project will also determine if there are ways in which the likelihood of bridging could be increased. We will also investigate the cases with long open hole intervals where bridging high in the hole may not be advisable because of the possibility for cross flow below the bridge.
Task 2 – Dynamic kill model for conventional and dual density Deep Water Blowouts (surface and underground) and investigation of pump rates to kill wells.
Dynamic kill models have been developed in the past, however these models may not be adequate for blowouts in water depths as great as 10,000 feet, nor are they designed to model dual density operations. A dynamic kill model will be developed which can be used for both conventional drilling and dual density operations. Both cases will have the capability of predicting kill rates for circulation through the drillstring in the blowout well as well as from relief wells. Returns will be modeled for circulation up the marine riser, choke or kill line, through seafloor pumps and return line (for dual density) all back to the surface, as well as exiting the well bore into the water column at the seafloor. The model will also have the ability to analyze underground blowouts. Modeling of underground blowouts with consideration for thief zone characteristics is not available in many current dynamic kill models.
Task 3 – Phase I - Develop blowout control methods based on Task 2 to include mechanical hookup alternatives.
During Phase I, a study will be made of the state of the art in blowout containment methods and equipment that is presently available. Data from actual blowout cases will be gathered and then used to validate the Bridging and Dynamic kill simulators developed in Tasks 1 and 2. We will begin to use the dynamic kill simulator to evaluate the hydraulic requirements needed to dynamically kill ultra-deepwater blowouts for hypothetical cases. This will be used to determine the capabilities of the simulator and identify any necessary improvements. During our study of actual blowout cases we will analyze which actions were more successful in regaining control of these blowouts.
In our search for actual blowout cases we will also gather data on ultra-deepwater kicks. From this study we will develop a “best practices” model for handling of ultra-deepwater kicks.
The results of Phase I will by catalogued and included in a Phase I report available through the MMS web site conference proceedings, and trade publications.
PHASE II (FUNDING FROM INDUSTRY THROUGH GPRI)
Task 3 – Phase II - Develop blowout control methods based on Task 2 to include mechanical hookup alternatives.
After the simulators have been validated with comparisons to actual cases, new blowout kill methods will be developed, and analyzed by use of the Dynamic kill and Bridging simulators. These new methods will be made available to the MMS and the petroleum industry. At the end of Task 3, we will have made a complete study of ultra-deepwater kicks and blowouts. We will have analyzed which actions worked best in certain situations, and will make recommendations on how to improve handling of ultra-deepwater kicks and how best to regain control of ultra-deepwater blowouts. The results of this task will be included in the Phase II final report in Task 5.
Task 4 - Cost estimate for deepwater intervention.
After Tasks 1-3 are nearing completion, work will begin on a cost estimation for deepwater intervention based on the results of these first three tasks. This cost estimation will aid the industry on the risk and consequences of ultra deepwater blowout. This cost estimate will be included in the final report.
Task 5 - Final report and administrative meetings.
Administrative meetings and workshops will be conducted throughout the project. The MMS, DOE, EPA, and individuals from the petroleum industry will be invited to the workshops, where the results of the research will be presented. Input from the attendees will be used to guide the research team in completing the individual tasks outlined above. After all the tasks are completed, a written report, bridging simulator, and dynamic kill simulator will be published in an electronic format and made available to the MMS (free of charge), to industry participants on a cost of publication basis and to industry non-participants on a fee basis.
DEPLOYMENT OF RESULTS:
MMS would have in hand a useful document for evaluation of ultra-deepwater well control risk and knowledge of methods necessary for successful intervention.
Industry would have access to a document that could guide well planning, contingency plan development and ultra-deepwater blowout intervention operations should that ever become necessary.
At the completion of this project, the following deliverables will have been met.
· The industry will be provided with a study which will determine the likelihood of a well bridging during a deepwater blowout, and ways to induce bridging and the consequences of undesirable bridging that may result in cross flows below the bridge. A bridging simulator will be available to forecast bridging in blowouts. This simulator will also have the capability of forecasting sand production problems in producing wells as well as well bore stability problems during underbalanced or near balanced drilling operations.
· A dynamic kill simulator with the ability to model:
· conventional and dual density wells
· circulation paths through the a drillstring located in the blowout well and relief wells
· returns to the surface via the drilling riser, choke and kill line, seafloor pumps and return line, or returns to the ocean at the seafloor,
· and underground blowouts.
· A manual cataloging the state of the art in blowout containment equipment and methodology. This will include mechanical hookup alternatives.
· Blowout control methods for dual density wells.
· Cost estimate for deepwater intervention.
· A final report in electronic format which can be used in risk analysis, contingency planning, and as a manual for containment of deepwater blowouts.During the project a series of forums will be held with representatives from the industry sponsors, MMS, and OTRC, as well as others with a vested interest in the results of the project.
ANTICIPATED PROJECT DURATION:
27 to 33 months depending on the scheduling of the tasks and the level of effort of each member of the team for each task during each budget period. The total man-months will not change.
PROJECT PLAN FOR YEAR 1 (2001-2002): Phase I
Scope of Work:
For fiscal year 2001-2002 we intend to begin work on Phase I, Tasks 1, 2, and 3.
Task 1: Bridging tendencies, we will start our literature search for pertinent publications on well bore bridging. We will also begin to gather data from operators active in the deep waters of the Gulf of Mexico so that we can begin our study of the well bore stability of the anticipated formations that would be encountered in the ultra-deep waters.
Task 2: We will begin the literature search to find the current status on dynamic kill models. We will also begin to develop the framework of the dynamic kill simulator which will not only have the capability of modeling conventionally drilled wells, but also wells drilled utilizing dual gradient technology.
Task 3: We will begin the literature search for ultra-deepwater blowout containment.
Anticipated Results:
At the end of fiscal year 2001-2002 most of the literature search for Tasks 1 and 2 should be complete, and work should have begun on development of the dynamic kill simulator (Task 2), and the study of the bridging tendencies of the formations that operators and drilling contractors are likely to drill through in the ultra-deep waters of the Gulf of Mexico. The literature search for Task 3 will be substantially complete. A workshop will be held to report on our progress and to gather input from industry.
PROJECT PLAN FOR YEAR 2 (2002-2003):
Scope of Work:
For fiscal year 2002-2003 we will complete Task 1(bridging tendencies) and Task 2 (dynamic kill simulator) and Task 3 (Phase I). Our plan is to begin Phase II (Blowout control methods, cost estimator and Final Report respectively)
Task 1: The study of bridging tendencies will be completed by the end of this fiscal year.
Task 2: During this fiscal year, the rheological models and multiphase flow models that will be utilized in the dynamic kill model will have be developed programmed into the dynamic kill simulator.
Task 3: A literature search for the current state of blowout control will be completed, and work will begin on development of new blowout control methods will begin. The dynamic kill simulator will be utilized to validate the procedures that are included in the blowout control methods. The Phase II portion of Task 3 will begin. We will have studied and reported on the “best practices” utilized in handling ultra-deepwater kicks and blowouts.
The Phase I report will be delivered to the MMS. The writing of the final report will begin with completion of Task 1. There will also be workshops held to report and discuss the progress of the project.
Anticipated Results:
By the end of this fiscal year, the results Task 1 (bridging tendencies) will be complete and made available to the sponsors, and will be utilized in the development of blowout control methods being developed in Task 3. The dynamic kill simulator being developed in Task 2 will be complete enough that it will be tested against actual blowouts. The dynamic kill simulator will be available for use in validating the new procedures involved in blowout control methods in Phase II - Task 3. Best Practices for handling ultra-deepwater kicks and blowouts will be documented in a Phase I report issued to the MMS.
PROJECT PLAN FOR YEAR 3 (2003-2004):
Scope of Work: For fiscal year 2003-2004 we will complete Tasks 3 through 5.
Task 2: We will complete the dynamic kill simulator. It will include any needed changes that may be identified in the development of blowout control methods in Task 3.
Task 3: We will complete the blowout control methods for wells being drilled in ultra-deep waters, and will have validated them with the dynamic kill simulator developed in Task 2.
Task 4: We will begin and complete a template for cost estimations of deepwater intervention of blowout wells. The results from Tasks 1-3 will be utilized in this task.
Task 5: We will complete the final report and manual and make the results available to the sponsors as well as the rest of the industry. The results of Tasks 1-4 will be reported and discussed in a series of workshops with guests invited from the sponsoring entities.
Anticipated Results: The final product of this project is described above in the section entitled "Deployment of Results."
PRINCIPAL INVESTIGATOR (S) & OTHERS INVOLVED IN PROJECT:
Following is a brief description of the qualifications of the key personnel and a description of their role in the project.
Mr. Curtis Weddle, III, P.E. – Cherokee Offshore Engineering
Curtis E. Weddle, III, PE, has 22 years of drilling and well control experience. Mr. Weddle will be the industry advisor and co-author of this study. He is currently team leader for well control methods development in the MudLift Drilling JIP, a project to develop a dual density drilling system for ultra-deepwater. He is a principal of Cherokee Offshore Engineering, a consulting firm for well control, project management and drilling. Prior to that he was responsible for well control operations worldwide for BP Exploration. His experience includes specification, design, commissioning and trouble shooting of ultra-deepwater BOP systems and several kick control operations in ultra-deepwater. He is currently on the executive committee for the IADC Deepwater Well Control Guidelines publication and was a founder of that ongoing project. He has been a member and chair of for projects such as BOP Test Frequency Justification, Sustained Annular Pressure Mitigation, Deepwater Rig Availability for Relief Wells and Prevention of Unplanned Disconnects. He has been chair of the IADC Deepwater Well Control Conference on two occasions and spoken or presented papers at that meeting for the last 5 years. Other experience includes major ultra-deepwater project development and evaluation in the Gulf of Mexico, deep high pressure gas drilling in the United States and work in Colombia, Venezuela, Alaska, Papua New Guinea, Indonesia, Vietnam, North Sea, NW Australia, Algeria and Azerbaijan.
Mr. Weddle will work on deepwater intervention methods and case simulations for deepwater blowouts. He will also provide industry liaison and focus to complete a final product that is useful to the industry. For the five Tasks in the project he will work as follows:
· Task 1 – Peer review of work, creation of cases for evaluation, contribution to report as to practicality of encouraging bridging and problems with cross flows that may be created by bridging.
· Task 2 – User input and output development, quality assurance and proofing of the model, representation of the end user.
· Task 3 – Peer review of current practice, sorting of successful vs. unsuccessful practices, incorporation of current deepwater construction practice into a collection of options for mudline intervention in the event of a blowout, hydraulic modeling, rig requirements for deepwater intervention, incorporate dual density equipment requirements and capabilities into the final report.
· Task 4 – Aid in creation of the cost estimate template and population of same.
· Task 5 – Co-author of final report as well as co-chair of industry meetings with Dr. Schubert.Jerome J. Schubert, Ph.D., P.E. – Texas A&M University, Harold Vance Department of Petroleum Engineering
Dr. Jerome J. Schubert, P.E. will be the Principal Investigator for this project. Dr. Schubert has a B.S. (1978), M.Eng. (1995), and Ph.D. (1999) all in Petroleum Engineering from Texas A&M University, and is currently employed as Lecturer/Assistant Research Engineer by the Harold Vance Department of Petroleum Engineering at Texas A&M University. Dr. Schubert has worked as a Drilling Engineer for over eight years with Pennzoil Company and Enron Oil & Gas, over four years as a Well Control Instructor with the University of Houston/Victoria, and as a faculty member at Texas A&M University for over six years. At Texas A&M University, Dr. Schubert is involved in teaching graduate and undergraduate drilling courses and in drilling research. Related research activities that Dr. Schubert has been involved with are kick detection, shallow water flows, and development of well control procedures for the MudLift Drilling JIP. He also serves on the IADC Training and Well Control Committees, and on the IADC WellCAP Review Panel.
Dr. Schubert will provide supervision of graduate students working on this project. He will provide guidance in their research and will evaluate the results of their work. Dr. Schubert will co-author all papers, reports and manuals developed from the project.
· Task 1 - Dr. Schubert will provide input to Dr. Valko as to blowout behavior. He will aid Dr. Valko in supervising his graduate student that will be working on this project.
· Task 2 - Dr. Schubert will supervise a graduate student in development of the Dynamic Kill model, and will provide insight into dual gradient drilling, and blowout behavior.
· Task 3 - Dr. Schubert will supervise a graduate student in gathering and cataloguing the current state of the art in blowout containment methods and equipment. He will work with Mr. Weddle in developing new blowout containment methods for ultra deepwater blowouts and dual density blowout control methods.
· Task 4 - Dr. Schubert will work with Mr. Weddle in estimating the containment cost of ultra deepwater blowouts.
· Task 5 - Dr. Schubert will help prepare the final report, and organize all meetings and workshops.Peter P. Valko, Ph.D. – Texas A&M University, Harold Vance Department of Petroleum Engineering
Dr. Peter Valko will be a co-PI for Task 1 – Bridging of blowouts in the GOM and tools for evaluation. Dr. Valko has a B.S. in Chemical Engineering from Veszprem University (1973) in Hungary, an M.S. in Applied Mathematics from Veszprem University (1975), and a Ph.D. in Chemical Engineering from the Institute of Catalysis (1981), in Novosibirsk, Russia. Dr. Valko has extensive teaching and research experience in Petroleum Engineering, in the areas of hydraulic fracturing and sand control, where he studied well bore mechanics, rock mechanics, and well bore stability, all useful in determining caving and bridging tendencies during extended periods of pressure drawdown as during blowouts.
Dr. Valko’ role in this project will be to co-supervise his graduate student along with Dr. Schubert in the study of bridging tendencies. This task will determine the parameters in which bridging is likely.
Contacts:
Dr. Jerome J. Schubert Mr. Curtis E. Weddle, III
Department of Petroleum Engineering Cherokee Offshore Engineering
Texas A&M University, M.S. 3116 28403 Teal Court
College Station, TX 77843-3116 Magnolia, TX 77355
979/862-1195, j-schubert@.tamu.edu 281/356-9139, cweddle@kropla.com
fax 979/845-1307
Date: June 2004Project Name: Development of a Blowout Intervention Method and Dynamic Kill Simulator for Blowouts Occurring in Ultra-Deepwater
Project Number: 408 Task Order: 18132
Principal Investigators: Dr. Jerome J. Schubert and Dr. Peter Valko, TAMU
Mr. Curtis Weddle, III, Cherokee Offshore EngineeringEstimated Completion Date: December 2004
Project Description: The original project plan included the following 5 tasks and was to be funded by MMS and industry. In Phase I, MMS funding is sufficient to complete Tasks 1 and 2, and make substantial progress on Task 3. Industry funding is being pursued to complete the project. Task 4 has been abandoned based on MMS and industry input. The remainder of Task 3 and Task 5 will be completed with industry funds in Phase II.
Task 1 - Bridging tendencies in ultra-deepwater blowouts
Task 2 - Dynamic kill investigation of ultra-deepwater blowouts and simulator development
Task 3 - Development of ultra-deepwater blowout control methods
Task 4 - Costs of intervention (abandoned)
Task 5 - Final report, progress meetings, and workshopsProgress:
Task 1 - Bridging tendencies in ultra-deepwater blowouts The wellbore collapse and bridging simulator (version 1) has essentially been completed, and the focus is now on model verification. .
The model will be verified with two groups of blowout scenarios to simulate and analyze the control of each. The first is an actual field case and the second is a hypothetical blowout with input data from a real well configuration and reservoir. John Wright Company provided the actual field case data for a bridged blowout. Additional case histories for use in model verification are also being gathered. The hypothetical blowout scenarios will be developed in Task 3. The results will be compared and discussed.
Task 2 - Dynamic kill investigation of ultra-deepwater blowouts and simulator development The code for the dynamic kill simulator has been completed, and is currently being debugged. Verification of the simulator with available case histories has begun, and additional case histories are being gathered and evaluated for use as further verification cases.
Task 3 - Development of ultra-deepwater blowout control methods The development of kick and blowout scenarios and cataloging them for future modeling with the dynamic kill and wellbore bridging simulators continues. Actual case histories of well control events in deep water are being gathered. In addition to being used to validate the dynamic kill simulator and wellbore collapse and bridging simulator, these case histories will be utilized to develop a best practices report for ultra-deepwater well control. The calibrated simulators will then be used to model the kick and blowout scenarios and develop detailed plans of actions.
Task 4 Costs of intervention (abandoned)
Task 5 - Final report, progress meetings, and workshops (will be completed in Phase II)
Reports & Publications:
Jourine, S., Karner, S L, Kronenberg, A K, Chester, F M.: Influence of Intermediate Stress on Yielding of Berea Sandstone Eos Trans. AGU, 84(46), Fall Meet. Suppl., Abstract T41D-0249, 2003.
Jourine S., Schubert J.J, Valkó P.P.: Saturated Poroelastic Hollow Cylinder Subjected To Non-stationary Boundary Pressure – Model and Laboratory Test. Submitted to Gulf Rocks '04, 6th North American Rock Mechanics Symposium (NARMS).
Oskarsen, R. T., and Schubert, J.J., “Development of a Dynamic Kill Simulator for Drilling in Ultra-deep Water,", Presented at the AADE National Technical Conference
Jourine, S., and Schubert, J. J., “Wellbore Bridging as a Possible Alternative to Blowout Control in Ultra-Deepwater Wells,” Presented at the 2003 AADE National Technical Conference, Houston, TX. April 1-3, 2003
Date: December 2003Project Name: Development of a Blowout Intervention Method and Dynamic Kill Simulator for Blowouts Occurring in Ultra-Deepwater
TEES Project Number: 32558-5888K MMS Task Order: 18132 MMS Project Number: 408
Principal Investigators: Dr. Jerome J. Schubert (TAMU), Dr. Peter Valkó (TAMU),
Mr. Curtis Weddle, III (Cherokee Offshore Engineering)Estimated Completion Date: 06/30/2004
Project Description:
The original project plan included the following 5 tasks and was to be funded by MMS and industry. In Phase I, MMS funding is sufficient to complete Tasks 1 and 2, and make substantial progress on Task 3. Industry funding is being pursued to complete the project. Task 4 has been abandoned based on MMS and industry input. The remainder of Task 3 and Task 5 will be completed with industry funds in Phase II.
Task 1 - Bridging tendencies in ultra-deepwater blowouts
Task 2 - Dynamic kill investigation of ultra-deepwater blowouts and simulator development
Task 3 - Development of ultra-deepwater blowout control methods
Task 4 - Costs of intervention (abandoned)
Task 5 - Final report, progress meetings, and workshopsProgress:
Task 1 - Bridging tendencies in ultra-deepwater blowouts Wellbore collapse (bridging) is the fastest, least-expensive and possibly only method of blowout control in deep water. Understanding of the main bridging mechanisms together with a quantitative model can provide a solid basis to determine the likelihood of a well bridging during a deepwater blowout, find ways to induce collapse, avoid undesirable packing, and move the bridging into the category of active blowout control technologies. A study of current wellbore bridging concepts was performed, and a numerical model to predict blowout self-killing has been developed. The model incorporates elements that describe reservoir inflow performance, wellbore hydraulics, and wellbore stability. The model can be used to evaluate bridging in both open hole drilling and cased hole completions in a variety of geological conditions. The model incorporates elements that describe reservoir inflow performance, wellbore hydraulics, and wellbore stability (shear, tensile, and erosion failures). The model elements have been partially tested with published data and laboratory experiments. The overall model will be verified on two groups of blowout scenarios to simulate and analyze the control of each.
The model will be utilized on two groups of blowout scenarios to simulate and analyze the control of each. The first is an actual field case and the second is a hypothetical blowout with input data from a real well configuration and reservoir. John Wright Company provided the actual field case data for bridged blowout. The hypothetical blowout scenarios will be developed during Task 3 of the current project. The results will be compared and discussed.
Task 2 - Dynamic kill investigation of ultra-deepwater blowouts and simulator development Most of the subroutines needed for the dynamic simulator have been completed and Graphical User Interfaces are being developed. The dynamic simulator will be validated with case histories and integrated with the wellbore collapse and bridging simulator (see Task 1 above) by the end of May, 2004.
Task 3 - Development of ultra-deepwater blowout control methods The potential failure points above and below the wellhead during a deepwater blowout (during drilling, completion or well intervention) have been identified and used to plan and design the dynamic kill simulator. A number of kick, blowout and intervention scenarios are being developed to describe the well kill. This process is complex since each initial blowout scenario can lead to multiple pathways during intervention and killing. The initial effort has focused on open hole blowouts during drilling. Completion, production, and workover scenarios will follow in that order. These scenarios are being cataloged for future modeling with the dynamic kill and wellbore bridging simulators.
Information from actual case histories of well control events in deep water will be gathered. These case histories will be utilized to develop the best practices report for ultra-deepwater well control and to validate the dynamic kill simulator and wellbore collapse and bridging simulator. After validation, the simulator can be used to study various kick, blowout, and intervention scenarios and develop well control methods.Substantial progress is being made on the above Task 3 activities. A plan and prioritization to use the simulator to develop well control methods for the various kick, blowout, and intervention scenarios will be developed based on the industry funds available in Phase II.
Task 4 Costs of intervention (abandoned)
Task 5 - Final report, progress meetings, and workshops (will be completed in Phase II)
Reports & Publications:
Jourine, S., Karner, S L, Kronenberg, A K, Chester, F M.: Influence of Intermediate Stress on Yielding of Berea Sandstone Eos Trans. AGU, 84(46), Fall Meet. Suppl., Abstract T41D-0249, 2003.
Jourine S., Schubert J.J, Valkó P.P.: Saturated Poroelastic Hollow Cylinder Subjected To Non-stationary Boundary Pressure – Model and Laboratory Test. Submitted to Gulf Rocks '04, 6th North American Rock Mechanics Symposium (NARMS).
Oskarsen, R. T., and Schubert, J.J., “Development of a Dynamic Kill Simulator for Drilling in Ultra deep Water,", Presented at the AADE National Technical Conference
Jourine, S., and Schubert, J. J., “Wellbore Bridging as a Possible Alternative to Blowout Control in Ultra-Deepwater Wells,” Presented at the 2003 AADE National Technical Conference, Houston, TX. April 1-3, 2003
OTRC PROJECT STATUS REPORTDate: June, 2003
Project Name: Development of a Blowout Intervention Method and Dynamic Kill Simulator for Blowouts Occurring in Ultra-Deepwater
Project Number: 32558 - 5888K Task Order: 18132
Principal Investigators: Dr. Jerome J. Schubert, TAMU
Dr. Peter Valko, TAMU
Mr. Curtis Weddle, III, Cherokee Offshore EngineeringEstimated Completion Date: 06/30/2004
Project Description:
Task 1. Bridging tendencies in ultra-deepwater blowouts
Task 2. Dynamic kill investigation of ultra-deepwater blowouts and simulator development
Task 3. Development of ultra-deepwater blowout control methods
Task 4. Costs of intervention
Task 5. Final report, progress meetings, and workshopProgress:
Task 1. By Serguei Jourine.Mathematical model of solid production.
I have been studying applicable geomechanics models for solid production from rock formations, which could be coupled with the developed fluid flow model through solid mass flow rate. My current approaches assume that bridging probability is defined by time independent formation strength and wellbore outflow performance, taking in account time dependent solid production from the damaged zone.1. The most conservative linear elastic deformational model was selected to predict the stress concentrations and onset of failure. A good qualitative agreement was obtained with calculated data and laboratory tests with hollow sandstone cylinders. The test results are consistent with my assumptions that wall failure is primarily controlled by rock heterogeneity even for a visible homogeneous sample, wall failure can stimulate secondary fractures, isolated washouts and it can produce the rock debris with wide size distributions (Fig.1).
Thus, the stress-strength calculations can provide only the bounds of possible mass solid production because of the heterogeneity and complex change of wellbore geometry during formation failure. In my opinion more complex geomechanics models cannot improve the quality of predictions without reliable rock strength and stress state data, which usually are not available under routine drilling operations.
Currently I am working on development of default dimensionless axisymmetrical FEA meshes to estimate the stress concentration for a clean wellbore section, wellbore bottom and the bridged wellbore section. The calculations with this mesh allow fast estimating of the maximum possible volume of solid production based on well geometry, in-situ stress field and known formation properties (Fig.2).
2. The hydro mechanical sand production model, proposed by Vardoulakis, was investigated to predict transient solid mass flow rate of debris transported to wellbore. Assuming that debris larger than the wellbore diameter can not be transported by fluid flow and the overburden weight is redistributed on the intact formation regions, all deformation and strength characteristics of the rocks within the unstable zone are suppressed, and the emphasis is solely on mass transport.
The mathematical model is based on mass balance and particle transport considerations, including one-dimensional Darcy flow in inhomogeneous formations. The coupled fluid-flow and erosion process was solved numerically by a finite difference scheme. Numerical results of the original model were recalculated, a parametric study was performed, and solution convergence was estimated (Fig.3).
Despite some oversimplifying model assumptions and poor physically defined parameters, the erosion model provides the results, which is qualitatively consistent with observed sand production history. It can explain the high bridging probability observed in early times and the low probability of this occurring in later times. Currently I am working on model improvement.
Acknowledgements: I thank Dr. A. Kronenberg and Dr. F. Chester for assistance in rock testing and Dr. D. Mamora for construction of sample X-ray cross-section regarding this study.
Task 2. By Ray Tommy Oskarsen
Since December 2002 my work has been concentrating on improving my computer modeling skills by taking classes in numerical methods for fluid dynamics and object oriented programming. These two classes will aid immensely in finishing the dynamic kill models and writing code for the simulator. Since most of my time since December has been in improving my modeling and programming skills, the progress on the dynamic kill simulator has slowed, however, I have made some progress in developing some of the models that will be used in the dynamic simulator.
For parts of the summer I will be interning with Anadarko. When the internship is completed the project will be my full-time employment. I expect to complete the dynamic kill simulator and be ready to graduate by the end of the year 2003.Task 3. By Steve Walls.
The project team began its effort by identifying all potential failure points during a deepwater blowout, from any mode of operation including, drilling, completion or well intervention, and sorted those points into two areas: above and below the wellhead. Areas below the wellhead have been further categorized depending upon where they were located relative to the last casing shoe or production packer.
From this initial scope, scenarios have been developed by which the intervention can begin and proceed throughout the well kill. This process is ongoing, as each beginning scenario has multiple pathways during a blowout intervention and each of those pathways have individually been defined and developed.
The initial effort has been focused on the occurrence of open hole blowouts during the drilling phase and the next step is considering the logical following point of the completion phase, insomuch as completion scenarios involve their own specific failure points. Specific scenarios related to well production and workover phases will then be developed. Much progress has been made in developing these complex and unpredictable scenarios. Now being negotiated, anticipated in-kind assistance from the only blowout response company with actual deepwater experience will enhance the final development of each scenario.
To give an example of a specific scenario during the drilling phase and how it has been developed, consider the following possible sequence: A well is being drilled with a relatively deep intermediate casing string has penetrated multiple horizons, which were expected during the planning. A new, high-pressure reservoir has been unexpectedly drilled, resulting in a wellbore kick that was detected and shut in. Difficulty in determining the actual kick time and volume is complicated by the reservoir itself being relatively low permeability and pressures slowly rising during the shut-in period, as the kick data is being evaluated and well kill decisions are being taken.
Approximately two hours after the initial shut-in, flow from the wellbore is observed and it is determined that the annular preventer, which was used to shut in the well, has begun leaking. Although its failure is not catastrophic, some ten additional barrels of influx have been allowed into the well and pressures which had been rising fairly steadily, are now observed to be rising at a somewhat slower rate now that a set of pipe rams have been closed.
As wellbore pressures are increasing, the ROV has noted some stirring of sediments from around the blowout preventer. Investigation has determined that there is a small amount of leakage from what appears to be the wellhead connector area. A quick bottom survey detects no other disturbance around the seafloor. As this was being investigated, the wellbore pressure suddenly becomes lower and then seems to stabilize, although it is still very high. The wellhead was being observed at this time by the ROV unit and no discernible increase in leakage occurred when the annular pressure decreased and then stabilized, and there has been only a small decrease in the drillpipe pressure.
The determination of a failure in most likely an exposed underground formation in conjunction with a wellhead seal failure increases the severity of the well control incident and the emergency response plan is activated, although the rig is determined to be in minimal danger as this point and does not active its shearing sequence to abandon the well. While waiting on electric logging tools to arrive to help determine the potential downhole failure point, the well is put on circulation using the driller’s method to initially clean the wellbore of formation contaminants in preparation to repair the failure. Blowout specialists are also en route to the site, as well as to the company office to liaise with the emergency management team.
During this initial circulation, increasing amounts of gas are noted at the surface but are handled easily by the gas-handling equipment. Definite bubbles are also noted at the wellhead area by the ROV operators, as well as slightly increasing amounts of fluid flow. No surface area of contamination can be seen by auxiliary vessels.
Extra support vessels have been acquisitioned by the emergency management team, which include auxiliary ROV and diving support, potential clean-up response and overflow personnel who do not need to be on the rig itself, but need to be at the site. This vessel is equipped as a standby incident command center in case the rig requires abandonment.
Higher density mud is circulated through the wellbore even as increasing amounts of leakage are being reported at the wellhead area. Pressures are slowly decreasing on the annulus as the mud is brought further around during the circulation. Losses are reported to be approximately 25 barrels per hour. As the heavier mud is brought past the blowout preventers on its way to the surface, the leakage at the wellhead is noted to be fluid only with no apparent gas bubbles.
Lost circulation material is pumped through the drillstring and around to the casing shoe and circulation is stopped to evaluate the results. Losses have decreased to <5 barrels per hour and the flow from the wellhead seems to be slowed somewhat but consistently leaking. Wellbore annular pressure has been steadily dropping but still is registering. Fluid densities in the riser, choke and kill lines are adjusted to match the open hole densities
The planning team recommends at this point to determine whether or not the drill string can be moved in order to isolate the high-pressure formation from the failure point in the wellbore, as annular pressures have been lowered enough to enable closure on the annular preventer. After the stack has been cleared of possible contaminants, the switch is made but the drill string cannot be moved. Wireline measurements indicate that the string is stuck somewhere below the casing shoe. The drill string is severed and the ability to circulate the well is unencumbered. Several circulations lower the wellhead pressure to 0 psi and the well is opened. Losses continue at the wellhead and are observed to be at the same 5 barrel per hour rate as before the string was severed.
The drill string is recovered and an RTTS test tool is run into the well. After the tool is set and the bypass closed, the loss is determined to be at the wellhead only. Subsequent operations entail the setting of a drillable packer through which cement is squeezed and then another RTTS tool is used to hang off the drillstring 100 feet below the wellhead. The blowout preventer is then picked up from the wellhead and the rig moves itself slightly off location. The ROV, moving into the wellhead area, determines that the seal area is damaged along the seal area of the wellhead connector. The BOP is then set back down onto the wellhead and the ROV remotely releases the seal ring, at which point the BOPs are then lifted back up. The ROV picks up the seal ring and brings it to the surface and an emergency seal ring is then run back and installed. Subsequent landing and testing of the BOP stack determine that the emergency seal ring has been able to seal the damaged area.
Recovery of the wellbore begins at this point with the retrieval of the RTTS and the hung-off drill string. No pressures are encountered above the cemented drillable packer and the wellbore is circulated clean. The RTTS tool is retrieved and the previously run cleanout assembly is used to drill up the packer and cement plugs. Upon reaching the area above the severed drill string, losses of < 5 barrels per hour are observed while circulating but no discernible losses can be seen by the ROV at the wellhead area or the surrounding bottom.
This particular scenario can be further developed into successful fishing of the stuck drill string and cleanup of the well bore with evaluation and casing setting as the best case, and unsuccessful fishing attempts resulting in an uncontrolled underground flow along with a renewed wellhead leak along the emergency seal area as the worst. Even though this scenario has little initial emergency impact, the potential economic impact of reservoir loss is very significant. Other scenarios require critical initial response actions due to their immediate endangering of the rig and personnel.
Each of the scenarios must be developed with this kind of detail, in order that the areas of commonality can be distilled and determined to be applicable to any problem, dependent upon its potential severity. From this, a cohesive plan that considers resource requirements can be developed which then only needs to be adjusted to local logistics (and market availability) conditions.
As this work has been developed, it has been noted that there are critical early response actions that may require the adjustment of industry emergency action plans. In particular, a rig operation may need to disengage from the wellhead if there is no backup rig capable of performing intervention work whether it be top (into the present wellbore) or side intervention (relief well drilling). Also, present experience by deepwater blowout personnel has shown the industry that ROV support during a mud line broach is impossible with presently envisioned equipment. We are now evaluating the application of other technology with regard to its use in performing subsea tasks.
Task 4. Costs of intervention. From industry input, it appears that there is not much interest in a blowout cost estimator from the operators and drilling contractors. It will be difficult to obtain funding from these groups. We have decided to abandon this task.
Task 5. Final report, progress meetings, and workshops. Since December, one company (ConocoPhillips) has signed the GPRI AFE for Phase II, two companies (Petrobras and Saipem) have included funds to support the project in their 2003 budgets but have not signed the AFE as of the writing of this report, we are negotiating with one company (Wild Well Control) to supply time from some of their top Engineers as payment in kind for this project.
Date: December 19, 2002
Project Name: Development of a Blowout Intervention Method and Dynamic Kill Simulator
for Blowouts Occurring in Ultra-DeepwaterProject Number: 32558 - 5888K PE Task Order: 18132
Principal Investigators: Dr. Jerome J. Schubert, TAMU
Dr. Peter Valko, TAMU
Mr. Curtis Weddle, III, Cherokee Offshore EngineeringEstimated Completion Date: 06/30/2004
Project Description:
Task 1. Bridging tendencies in ultra-deepwater blowouts.
Task 2. Dynamic kill investigation of ultra-deepwater blowouts and simulator development.
Task 3. Development of ultra-deepwater blowout control methods.
Task 4. Costs of intervention.
Task 5. Final report, progress meetings, and workshops.Progress:
Task 1. Results: Mathematical model of dynamic packing is developed.
From our work it has been determined that the method for the prediction of blowout self-killing is based on an analysis of reservoir performance data and wellbore hydraulics. Wellbore hydraulics performance curves are generated for different circulation rates of a solid-fluid mixture and compared with the reservoir performance curve. Any wellbore-hydraulics curve for a given circulation rate that intersects or falls below the reservoir performance curve will reach a stable flow condition, with the solid being lifted out of the well. The minimum solid load to achieve a self-killing condition has a performance curve that is totally above (not intersecting) the reservoir performance curve. See Figure 1.In the early part of this project, we described the outflow curve construction using the steady state approach. During the current period of activity we developed the mathematical model for the transient fluid flow region below terminal velocity but above the chocking velocity. The region is shown in Figure 1.
In this region the bridge can be formed due to gravity segregation or fluid flow.
The model is based on the following assumptions:
1. The flow is one-dimensional.
2. Fluid is incompressible.
3. All the particles have the same size.
4. The irregular motion and collision of particles are ignored.
5. The friction on the wall is neglected, for it is far less than the gravity of particles.
6. Particle acceleration is ignored.
Figure 1. Outflow performance curves for solid-liquid flow7. Buoyancy force is ignored.
The numerical algorithms for solving this system of equation using fully implicit and centered-in-distance/backward-in-time method as the part of Outflow Calculation Subroutine are being tested.
Testing results shows that the developed model allows us to describe the dynamics of bridge formation and to estimate the likelihood and duration of bridge.Task 2. Dynamic kill simulator – We have begun to develop the model for the dynamic kill
simulator. We are continuing to investigating the capabilities/limitations of existing dynamic kill models that are available to the petroleum industry to compare to our needs. We have made considerable progress in developing the core structure of the program and have completed some of the required subroutines in Visual Basic.The past three months we have been studying advanced numerical methods and computer programming. Our initial plan was to write the code for the dynamic kill simulator in Visual Basic. Through consultation with experts in programming in the computer science department regarding the program we are developing. It was strongly recommended that we write the program in Java. Java has the advantage of being platform independent. This enables a Java program to run on any operating system, which makes Java the preferred programming language to use with the Internet. We envision the dynamic kill simulator to be available to the users over the TAMU servers. Thus, we have abandoned the VB program and have started converting the VB to Java. Much of the past three months has been spent getting more familiar with Java
Task 3. Development of ultra-deepwater blowout control methods. We have continued gathering information on the latest blowout containment technology that has been utilized by the industry. It has become clearer that many of the techniques that have been used in the past may not be applicable in ultra deep waters. We have begun progress in our ultra-deepwater kick containment and are preparing to gather information from weekly drilling reports issued to the MMS. The information we will gather is on kicks and blowouts that have occurred in ultra-deepwater wells. From this information we will develop “best practices” in kick containment and blowout control.
Task 4. Costs of intervention. From industry input, it appears that there is not much interest in a blowout cost estimator from the operators and drilling contractors. It will be difficult to obtain funding from these groups. However, there has been some interest from insurance underwriters. We are negotiating with these groups to obtain funding for this task
Task 5. Final report, progress meetings, and workshops. We conducted a workshop in August, 2002 to convey to the industry, OTRC, and the MMS the status of our project up to that time. We will use input from industry representatives present at the workshop to help guide us in our project. The workshop was used to highlight why this project is necessary, and why industry needs to fund the remainder of the project. There were approximately 90 attendees at the workshop, and all agreed that this was a worthwhile project. There were six representatives that expressed a strong interest in funding the project. As of today we have firm commitments from two and four others who have submitted the project to their respective budget committees.
Reports & Publications: None
Date: June 2002
Project Name: Development of a Blowout Intervention Method and Dynamic Kill Simulator
for Blowouts Occurring in Ultra-DeepwaterTask Order: 18132 Project Number: 5888K
Principal Investigators: Dr. Jerome J. Schubert, TAMU
Dr. Peter Valko, TAMU
Mr. Curtis Weddle, III, Cherokee Offshore EngineeringEstimated Completion Date: 06/30/2004
Project Description:
Task 1. Bridging tendencies in ultra-deepwater blowouts.
Task 2. Dynamic kill investigation of ultra-deepwater blowouts and simulator development
Task 3. Development of Dual Density blowout control methods
Task 4. Costs of intervention
Task 5. Final report, progress meetings, and workshopsProgress:
The literature search for the items 1,2, and 3 are substantially completed, and graduate students have begun to work on Tasks 1 and 2.
Task 1. Bridging tendencies – We have begun work on a wellbore stability model which will be used to forecast wellbore collapse and bridging tendencies during blowouts. The model should also be of use in forecasting sand production problems for producing wells as well as determining minimum allowable wellbore pressures for underbalanced drilling.
a. The Model concept has been developed
b. Subroutines and equations for inflow, outflow and wellbore stability have been written and are being tested.
Task 2. Dynamic kill simulator – We have begun to develop the model for the dynamic kill simulator. We are investigating the capabilities/limitations of existing dynamic kill models that are available to the petroleum industry to compare to our needs. We are currently finishing up the core structure of the program and have completed some of the required subroutines.
Task 3. We have been gathering information on the latest blowout containment technology that has been utilized by the industry. We are putting together our arguments that many of the techniques that have been used in the past may not be applicable in ultra deep waters.
Task 5. We are planning and organizing a workshop to be held in August, 2002 to convey to the industry, OTRC, and the MMS the present status of our project. We will use input from industry representatives present at the workshop to help guide us in our project. Finally the workshop will also be used to highlight why this project is necessary, and why industry needs to fund the remainder of the project.
Date: December 4, 2001Project Name: Development of a Blowout Intervention Method and Dynamic Kill Simulator for Blowouts Occurring in Ultra-Deepwater
Task Order: 18132 Project Number: 5888K
Principal Investigators: Dr. Jerome J. Schubert, TAMU
Dr. Peter Valko, TAMU
Mr. Curtis Weddle, III, Cherokee Offshore EngineeringEstimated Completion Date: June 2004
Project Description:
1) Bridging tendencies in ultra-deepwater blowouts – Gulf of Mexico and other ultra-deepwater sediments are generally poorly consolidated. Many believe that a high rate ultra-deepwater blowout will bridge and self kill. We will investigate the likelihood of this and define the parameters for evaluation of bridging including conditions with open hole drilling and cased hole completions.
2) Dynamic kill investigation of ultra-deepwater blowouts – we would develop a dynamic kill model for deepwater blowouts and investigate methods and pump rates necessary to kill the blowout from the existing well bore or from one or more relief wells.
3) Development of Dual Density blowout control methods – In the event that a deepwater blowout results in loss of the riser or a disconnect it may not be possible or safe to reconnect the riser and divert flow to the surface. If that is the case, dynamic kill could only be accomplished from a relief well using Dual Density mud weights. Furthermore, Dual Density drilling methods are likely to become commercially available in the next two years. It is likely that a well drilled to a formation using Dual Density methods could not be killed by a relief well using any other drilling method. Investigation of dynamic kill with Dual Density drilling will be included in the proposed study.
4) Costs of intervention – We propose to develop a cost estimate template for ultra-deepwater blowout intervention.
5) Final report, progress meetings, and workshops – We propose to hold progress meetings and workshops throughout the project with representatives of the sponsors and selected participants from industry not only to report the progress that the research team has made, but to solicit input from industry experts. The final report will contain the documentation of the work that was done as well as a copy of the dynamic kill simulator and cost estimator. The document will contain information as to recommended procedures in the event of an ultra-deepwater blowout as well as a index of where equipment and services can be locatedProgress:
The literature search for the items 1,2, and 3 have begun, and prospective graduate students have been identified to work on the project.
