Deepwater_Horizon_Accident_Investigation_Report_Appendices_ABFGH

Appendix A.

Transocean Deepwater Horizon Rig Incident Investigation Into the Facts and Causation (April 23, 2010)

The President of BP Exploration & Production Inc. agreed the following Terms of Reference and has requested that Mark Bly, Group Head of Safety & Operations, lead the investigation team 1:

The scope of the investigation to find the facts surrounding the uncontrolled release of hydrocarbons and efforts to contain the release aboard the Transocean drillship Deepwater Horizon, located approximately 40 miles south of Venice, LA at Mississippi Canyon 252, BP's Macondo prospect is as follows:

1.Collect evidence surrounding the incident 2

2.Determine the actual physical conditions, controls, and operational regime related to the incident to understand: a.Sequence of relevant events

b.Reasons for initial release

c.Reasons for fire

d.Efforts to control flow at initial event

3.Prepare a report to include: a.Background

b.Timeline

c.Description of incident

d.Critical factors i.

Immediate Causes ii.System Causes

e.Proposed Recommendations 4.Administrative a.All activities of the fact-finding teams will be approved in advance by the respective

team leader b.Retain all incident investigation team documents (including notes, drafts, electronic

documents and emails) relating to the fact-finding and the incident

c.Maintain confidentiality of our discussions

d.BP person at each interview

e.No questions or tasks to BP contractors without BP approval

1

James Lucari, Managing Attorney – BP Legal HSSE and Regulation, has been assigned to provide legal

advice and counsel to Mr. Bly in his role as investigation team lead. 2

Given the business and regulatory relationships involved in this context, the Incident Investigation team’s

efforts will be informed by physical evidence, data and information that is in the custody and/or control of third parties. The incident investigation team’s access to this information may affect its ability to complete the terms of reference set forth herein. A Deepwater Horizon Accident Investigation Report Appendix A. Transocean Deepwater Horizon Rig Incident Investigation Into the Facts and Causation

193A p p e n d i x A

194Deepwater Horizon Accident Investigation Report

Deepwater Horizon Accident Investigation Report Appendix B. Acronyms, Abbreviations and Company Names 195A p p e n d i x

B Appendix B. Acronyms,

Abbreviations and Company Names

Appendix B. Acronyms, Abbreviations and Company Names

196Deepwater Horizon Accident Investigation Report

Appendix B. Acronyms, Abbreviations and Company Names

197A p p e n d i x

B

Deepwater Horizon Accident Investigation Report

Appendix B. Acronyms, Abbreviations and Company Names

198Deepwater Horizon Accident Investigation Report

Appendix B. Acronyms, Abbreviations and Company Names

199A p p e n d i x

B

Deepwater Horizon Accident Investigation Report

Appendix B. Acronyms, Abbreviations and Company Names

200Deepwater Horizon Accident Investigation Report

Appendix F. Roles and Responsibilities for Macondo Well

207A p p e n d i x F

Appendix F. Roles and Responsibilities for Macondo Well

Major Parties Involved

Deepwater drilling programs are large, complex ventures that require the involvement of several organizations. This appendix provides an overview of the roles and responsibilities of the following entities involved in the Macondo well project. (Refer to Appendix B. Acronyms, Abbreviations and Company Names.)

Minerals Management Service (MMS)

MMS had regulatory oversight for offshore oil and gas leasing activities on the outer

continental shelf, including the area of the Macondo well. Federal regulations set forth the

MMS program for regulatory oversight of offshore operations and governed the process for

review and approval of design, drilling operations and rig maintenance plans. An exploration

plan was submitted for approval to the MMS before exploratory drilling began on the

Mississippi Canyon Block 252 lease. An Application for Permit to Drill (APD) was also submitted for approval before drilling began on the Macondo well. Subsequent Applications for Permit

to Modify (APM) were submitted for approval based on the actual conditions encountered

while drilling the well. MMS inspectors visited the rig during drilling operations to monitor

compliance with regulations, permits and statutes.

Since June 18, 2010, the MMS’s regulatory role has been exercised by the Bureau of Ocean

Energy Management, Regulation and Enforcement (BOEMRE).

BP

As the lease operator, BP was responsible for geologic assessment of subsurface formation,

engineering design of the well and obtaining regulatory approvals required for construction of

the well. BP was also responsible for retaining and overseeing the contractors who supported

well design and various aspects of the drilling operations. During drilling operations, BP had

staff on the rig who represented the company’s interests in operations related to the well;

this did not include operations related to non-well-related activities such as rig maintenance or

marine systems.

BP retained contractors to provide engineering and operations services, drilling equipment

and personnel. These contractors typically operated under their own management systems,

supplying their own (or sub-contracted) personnel, equipment and materials as needed.

Contracts issued by BP generally covered multi-year terms.

Deepwater Horizon Accident Investigation Report

Appendix F. Roles and Responsibilities for Macondo Well

T ransocean

Transocean provided the drilling rig, Deepwater Horizon, and the personnel to operate it.

Transocean was solely responsible for operation of the drilling rig and for operations safety. It

was required to maintain well control equipment and use all reasonable means to control and

prevent fire and blowouts. Transocean provided, maintained and operated a range of subsea

equipment, including the Cameron blowout preventer (BOP) and associated control systems.

BP retained the right of inspection and approval of the work performed on its behalf, although

the actual performance and supervision of the work was Transocean’s responsibility.

Halliburton

Halliburton provided engineering services, materials, testing, mixing and pumping for

cementing operations. This included both onshore engineering support and offshore equipment

and personnel based on Deepwater Horizon. Halliburton was responsible for and provided

technical advice as to the design, modeling, placement and testing of the cement that was

pumped into place behind the casing string and in the shoe track to isolate the hydrocarbon

zone(s) from the wellbore at the Macondo well site.

Sperry-Sun is a Halliburton company that was responsible for mudlogging equipment

(including downhole drilling tools) and personnel. Sperry-Sun mudlogging personnel were

responsible for monitoring the well and advising the driller in the areas of mud pit volume

changes, mud flow in and out of the well, mud gas levels and any pressure fluctuations.

M-I SWACO

M-I SWACO provided mud products, engineering services and mud supervisory personnel.

M-I SWACO was responsible for specifying the mud additives to be mixed, measuring mud

properties and submitting daily reports to BP regarding mud conditions and effectiveness. At

the well site, the Transocean deck and drill crews conducted mud handling at the direction of

M-I SWACO’s mud engineers.

Weatherford

Weatherford provided the casing components including the float collar, shoe and centralizers.

Weatherford also provided the personnel and equipment for running the casing and casing

components into the wellbore.

Dril-Quip

Dril-Quip provided wellhead equipment, including the casing hangers, seal assembly and

lockdown sleeve used on the Macondo well. Dril-Quip also provided personnel responsible for

supervising the installation of this equipment and any subsequent servicing, modifications and

maintenance to the wellhead equipment.

208Deepwater Horizon Accident Investigation Report

Appendix F. Roles and Responsibilities for Macondo Well

209A p p e n d i x F

Oceaneering

Oceaneering provided remote operated vehicle (ROV) equipment and personnel. Oceaneering had responsibility for transporting, piloting and maintaining the ROV assigned to the Macondo well. Oceaneering used the ROV to inspect the wellhead, BOP stack, riser and seafloor.

Deepwater Horizon Accident Investigation Report

210Deepwater Horizon Accident Investigation Report

Appendix G. Analysis Determining the Likely Source of In-flow

211A p p e n d i x G

Appendix G. Analysis Determining the Likely Source of In-flow

This appendix describes how the hydrostatic pressure observations, OLGA? well flow modeling results and static well kill results support the report’s conclusion that the in-flow was from the

formation, through the annulus cement barrier, through the shoe track barriers and into the

production casing. Analysis 5A. Well Integrity Was Not Established or Failed of this report refers to other considerations on which the conclusion was also based—those other considerations are not discussed in this appendix.

The investigation team reviewed real-time surface and subsurface data, witness accounts and

modeling results to determine the most likely route of in-flow to the well.

Two barrier failure scenarios were considered:

Flow through the casing hanger seal assembly in the wellhead.

Flow through the shoe track barriers (shoe track cement and double-valve float collar).

The possibility of hydrocarbon ingress through the production casing was a third scenario

considered in Analysis 5A, but as explained there, a design and installation review of the

9 7/8 x 7 in. casing was completed by the investigation team, and no integrity issues were

identified. Furthermore, the positive-pressure test completed following the casing installation

confirmed the integrity of the production casing.

Engineering analyses were conducted for the casing hanger seal assembly and the shoe track

barriers. Although analysis indicated that casing hanger seal assembly failure was not likely, this flow path scenario is included in the hydrostatic pressure and well modeling analyses to further explain why this scenario was determined not to be credible.

The investigation team conducted hydrostatic pressure calculations to match the 1,400 psi

pressure observed during the negative-pressure test and performed dynamic flow modeling

(OLGA?) to match pressure responses observed prior to the explosion.

This discussion addresses the following topics:

T he Hydrostatic Pressure Effect – a brief review of how pressure changes as fluid moves through the wellbore.

C hanging Cross-section Across the Riser and Production Casing Annulus – an illustration of the

changes in system pressure created as fluids flow through the wellbore.

H ydrostatic Pressure Calculations – an analysis of known reservoir and surface pressure

measurements relevant to determining flow path.

D ynamic Modeling of Drill Pipe Pressure Response – simulations of the two scenarios

considered and their conformance with known data.

Deepwater Horizon Accident Investigation Report

Appendix G. Analysis Determining the Likely Source of In-flow

The Hydrostatic Pressure Effect

Hydrostatic pressure is the pressure exerted by a fluid due to the force of gravity. As shown in

Figure 1, pressure at any point in a fluid column is generated by the height of liquid above that

point. The pressure is calculated as follows:

Pressure = fluid density x fluid height x gravity.

Schematic of How the Same Volume of Fluid Can Create Different Pressures.

Figure 1.

As shown in Figure 2, a fluid with a higher density (heavier fluid) generates more pressure than

a lighter fluid if they are in cylinders of the same height.

212Deepwater Horizon Accident Investigation Report

Appendix G. Analysis Determining the Likely Source of In-flow

213A p p e n d i x G

Figure 3 shows that when two different fluid columns of the same height are connected, the

heavier fluid will transmit pressure to the lighter fluid. If the lighter fluid is trapped at the top, it will generate pressure, which is labeled as P1 in Figure 3

.

Figure 3. Pressure Transmission Through Connected Fluid Columns.

Deepwater Horizon Accident Investigation Report

Appendix G. Analysis Determining the Likely Source of In-flow

Changing Cross-section Across the Riser and

Production Casing Annulus

Figure 4 shows selected cross-sections through the length of the Macondo well. (The well

intervals shown are for illustrative purposes. For well design, refer to Appendix C. Macondo Well

Components of Interest.) The annular space between the drill pipe and the production casing and

between the drill pipe and the riser is the focus of the analysis conducted in the next section of

this appendix. The different cross-sections in this region each generate a different hydrostatic

pressure for a barrel of fluid occupying that space, as follows:

E ach barrel of fluid (assuming the same fluid) in the annular space between the 3 1/2 in. drill

pipe and the production casing would generate approximately 5.6 times more hydrostatic

pressure than a barrel of fluid in the riser annulus.

E ach barrel of fluid (assuming the same fluid) in the annular space between the 5 1/2 in. drill

pipe and the production casing would generate approximately eight times more hydrostatic

pressure than a barrel of fluid in the riser annulus.

214Deepwater Horizon Accident Investigation Report

Appendix G. Analysis Determining the Likely Source of In-flow

215A p p e n d i x G

Hydrostatic Pressure Calculations

While the negative-pressure test was being conducted on the kill line, a pressure of 1,400 psi

on the drill pipe was observed by the rig team. The 1,400 psi observation was confirmed by the investigation team using real-time data. Of the known pressures in the various sands, this pressure best matched the M56A sand at 17,788 ft.

Reservoir Data

The reservoir pressure of the M56A sand was measured at 12,038 psi with the Schlumberger

Modular Formation Dynamics Tester (MDT) logging tool. This pressure is equivalent to a

13.1 ppg mud weight, using subsea measurements as the depth reference.

The hydrocarbons in the M56A sand could either be oil or gas. The sonic log signature of the

M56A sand was indicative of oil-bearing sands. However, based on the M56A position directly

above the boundary of the thermogenic front, it could be gas. Performing the calculation using

oil rather than gas would not change the conclusion of the hydrostatic pressure analysis.

The equivalent densities of the sands measured are:

14.1 ppg for M57C (measured by Geotap?).

13.1 ppg for M56A (measured by MDT).

12.6 ppg for M56E (measured by MDT).

Fluids in Wellbore

The fluids used for the calculations are:

8.6 ppg seawater in the drill string.

14.17 ppg synthetic oil-based mud in the wellbore.

19 bbls of in-flow fluid in the wellbore.

Witness accounts indicated that in preparation for the negative-pressure test on Deepwater

Horizon, approximately 15 bbls of seawater were bled from the drill pipe, which was

substantially above the compressibility of the system. Including seawater compressibility,

approximately 3.5 bbls should have been bled to remove pressure from the well. (Refer to

Appendix R. Fluid Compressibility Calculations.) Therefore, the investigation team concludes

that this additional fluid was in-flow fluid. According to witness accounts, an additional 3 bbls to

4 bbls was bled from the kill line, which gives an estimated total in-flow volume of 19 bbls.

The in-flow fluid is assumed by the investigators to be hydrocarbons with an estimated density

of 5.18 ppg at 239°F at 12,000 psi.

Note: Illustrations used in the following discussion show a sharp interface between different

fluids in the well; this was done for simplicity of illustration. In reality, there would be varying

degrees of fluid mixing at each interface.

Deepwater Horizon Accident Investigation Report

216Deepwater Horizon Accident Investigation Report Appendix G. Analysis Determining the Likely Source of In-flow

After the volume was bled, the drill pipe pressure increased to 1,400 psi. Figure 5 shows the expected position of fluids, assuming in-flow through the annulus cement barrier, through the shoe track barriers and into the production casing.

Using the 19 bbls total bleed volume, the in-flow was calculated to occupy the 7 in. casing up to 17,558 ft. The top of the 14.17 ppg mud in the drill pipe was calculated at 7,200 ft. using the 15 bbls volume bled from the drill pipe. The remainder of the drill pipe column was

8.6 ppg seawater.

Assuming the fluids configuration shown in Figure 5, the resulting hydrostatic pressure at 18,304 ft. is:

1,400 + (7,200 x 8.6 + [17,558 – 7,200] x 14.17 + [18,304 – 17,558] x 5.18) x 0.052 = 12,453 psi which corresponds to an equivalent density of:

.The calculated equivalent density of 13.08 ppg best matches the M56A sand.

However, some witness accounts suggested that the total volume bled back could have been more than 19 bbls. If an in-flow of more than 19 bbls occurred, the calculated shut-in pressure would be higher and therefore better match the M56A sand.

Calculations were also made for the casing hanger seal assembly path, but a match could not be made. However, using hydrostatic calculation alone, flow through the casing hanger seal assembly could not be ruled out because of the uncertainties associated with parameters affecting pressure on the assembly, such as bleed volume uncertainty, fluid density

complications with the 16 ppg spacer below the BOP and weight gradients within that spacer caused by mixing at the seawater interface.

Conclusions

The 1,400 psi drill pipe pressure observed during the negative-pressure test best matched communication with the M56A sand through the annulus cement barrier and shoe track

barriers. Hydrostatic pressure calculations alone were not sufficient to conclusively determine that this was the failure mode that allowed flow to the surface. Uncertainty with fluid density in the annulus, inconsistent bleed volumes from witness accounts and the presence of

other sand stringers with unknown pressure were outstanding unresolved variables.

Additional analysis was performed to further assess the most likely flow path and is

discussed below.

Dynamic Modeling of Drill Pipe Pressure Response The investigation included preparation of an MC 252 well flow simulation for both scenarios using the OLGA ? well flow model. Outputs for each simulation were compared to real-time data and witness accounts. The scenario for hydrocarbon flow through the annulus cement barrier and shoe track barriers was validated; the scenario for flow through the casing hanger seal assembly could not be modeled.

12,4530.052 x 18,308

13.08 ppg =

Deepwater Horizon Accident Investigation Report Appendix G. Analysis Determining the Likely Source of In-flow 217A p p e n d i x

G Figure 5. Mud Displacement Caused by In-flow

Through the Shoe Track Barriers at Approximately 18:00 Hours.

218Deepwater Horizon Accident Investigation Report

Appendix G. Analysis Determining the Likely Source of In-flow

Real-time Data and Witness Accounts of Increasing Pressure During the Sheen T est

Figure 6 shows the pressure response on the drill pipe from 21:00 hours to 21:50 hours correlated to mud pump strokes per minute (spm).

At 21:08 hours, the spacer reached the top of the riser, and the mud pumps were shut down to conduct a static sheen test before commencing the displacement of the spacer overboard. From 21:08 hours to 21:14 hours, the period during which the mud pumps were shut down, the drill pipe pressure increased from 1,017 psi to 1,263 psi. The investigation team concluded that the pressure increase was probably caused by an increase in hydrostatic pressure.Figure 7 and Figure 8 illustrate the two scenarios considered for flow from the formation into the wellbore: through the shoe track and through the casing hanger seal assembly,

respectively. Evaluation of these scenarios indicates that the pressure increase in the drill pipe was caused by flow through the shoe track, as described below.

Figure 7 illustrates flow through the shoe track displacing the 14.17 ppg mud past the bottom of the drill pipe and up the annulus around the drill pipe. Each barrel of 14.17 ppg mud that filled this annulus above the bottom of the drill pipe displaced a lighter fluid, increasing

hydrostatic head. At the same time, a 16 ppg spacer was displaced out of the riser, reducing the hydrostatic pressure generated by the fluid column in the riser. As stated earlier, each barrel of fluid (assuming the same fluid) in the annular space between the 3 1/2 in. drill pipe and the production casing would generate approximately 5.6 times more hydrostatic pressure than a barrel of fluid in the riser annulus.

Recorded Data - Mud Pump Rates and Drillpipe Pressure

Figure 6. Drill Pipe Pressure Real-time Data from 21:00 Hours to 21:50 Hours.

Appendix G. Analysis Determining the Likely Source of In-flow

219A p p e n d i x G

Therefore, it could be concluded that the pressure increase generated by the 14.17 ppg mud

entering the (smaller) annulus at the bottom of the drill pipe would more than offset the

pressure reduction caused by 16 ppg spacer exiting the riser annulus, where the annulus cross

section is larger.

Figure 7. Model Predictions of Hydrostatic Pressure that Could Be Generated by In-flow Through

the Shoe Track Barriers from 21:09 Hours to 21:14 Hours.

Deepwater Horizon Accident Investigation Report

本文来源：https://www.bwwdw.com/article/48rq.html