Subsea Blowout-Preventer Systems: Reliability and Testing
- Per Holand (SINTEF Safety and Reliability)
- Document ID
- Society of Petroleum Engineers
- SPE Drilling Engineering
- Publication Date
- December 1991
- Document Type
- Journal Paper
- 293 - 298
- 1991. Society of Petroleum Engineers
- 1.6.10 Running and Setting Casing, 4.2.4 Risers, 1.7 Pressure Management, 5.1.2 Faults and Fracture Characterisation, 1.6 Drilling Operations, 1.10 Drilling Equipment, 5.6.4 Drillstem/Well Testing
- 2 in the last 30 days
- 671 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
A comprehensive study of subsea blowout-preventer (BOP) performance in the North Sea during 1978-89 was conducted to identify BOP reliability problems. Evaluations of the BOP as a barrier to blowouts, together with rig downtime caused by BOP failures and malfunctions, showed that BOP reliability has improved significantly during the last few years.
During 1982-90, S carried out a comprehensive reliability study of subsea BOP systems for various off companies operating in the Norwegian sector of the North Sea and the Norwegian Petrolem Di (NPD). The project was divided into five phases, with final reporting done after each phase.
Phase 1. Analysis of failure data from 61 wells and BOP system analysis.
Phase 2. Analysis of failure data from 99 wells and mechanical report on control-systems
Phase 3. Evaluation of BOP test procedures and operational control.
Phase 4. Analysis of failure data item 58 wells drilled by fairly new rigs. Evaluation of failure causes. Estimation of blowout probabilities based on a fault-tree model.
Phase.5. Analysis of 47 exploration wells drilled during 1987-89. BOP failures and tests were recorded and analyzed.
This paper focuses on the results from Phase 5. Results from Phases 2 through 4 serve as references for trends in BOP reliability and testing. Results from Phase 1 have been presented elsewhere and are not included here.
Data for Reliability Study
Fig. 1 shows a typical subsea BOP system used on the Norwegian Continental Shelf. The BOP has a through-bore of 18% in., with a rated working pressure of 10,000 or 15,000 psi. It is normally stacked with two annular preventers and four ram-type preventers. The typical system has three choke/kill-line outlets that are connected to the choke or kill line by two redundant fail]-safe valves. The BOP close/open functions are controlled by a dual hydraulic control system and an acoustic backup emergency control system for the most important functions.
For the failure recording period, the BOP's were function and pressure tested before running, after landing, after running casing, and about once a week during operations, according to NPD regulations
Data regarding BOP failures and malfunctions were collected by reviewing daily running reports (Intl. Assn. of Drilling Contractors format or daily telex), BOP test reports, final well reports, and equipment failure from 256 exploration wells drilled from semisubmersible drilling rigs in the Norwegian sector of the North Sea from 1978 to 1989.
A comprehensive failure-event data base was established with dBASE III Tm on an EBM PS/2. In total, 831 failure events were recorded during 61 years in service. These failures have caused a total rig downtime of 15,100 hours. The following information was recorded for each failure event: rig name; field and well number; water depth; failure date; type of operation when the failure/malfunction was discovered; faded component, type, and make; why and how the failure/malfunction was discovered; what was done to repair or restore the failures; rig downtime caused by the failure/malfunction; and additional comments. Selected failures were analyzed in more detail by use of additional information from rig owners and manufacturers.
Failure-Rate Calculation. A BOP failure is a failure associated with one of the BOP components or subsystems. The failure rate, X, for a BOP component is estimated by
where n=number of failures and r=total time in service.
In practice, it is difficult to determine the total time that a component actually has been functioning and exposed to stress. From discussions with drilling engineers, however, it was found appropriate to relift the wellhead in service to the time the BOP was first to the wellhead until the drilling was completed and the BOP was disconnected from the wellhead.
Downtime Calculations. The rig downtime includes time lost because of failures and malfunctions-i.e., securing the well; pulling the stack; and repairing, rerunning, and drilling out the well barrier. Regular BOP testing, handling, and maintenance are not included.
BOP Performance Results
Trends in BOP Reliability. Fig. 2 shows the year-to-year failure rate for subsea BOP stacks used during floating drilling. The average annual value is marked with an x, calculated by Eq. 1, and the line indicates a 90% confidence interval. A long line indicates that large statistical uncertainties exist in the data because few data were recorded.
As Fig. 2 shows, fewer BOP failures were recorded during the early 1980's, and the failure rate seems to have stabilized at a certain level during 1994-85. In studies like this one that are based on a limited amount of data, variations from year to year are expected; however, here the decreasing trend is clear. Better equipment for BOP handling and running and some minor design changes had a positive effect on the failure rate. Other factors that contributed to this improvement concern BOP maintenance and include better cellar dock design, better tools, more allocated personnel, and better maintenance system and procedures.
Fig. 3 shows a year-to-year overview of the average rig downtime per day in service caused by BOP failures and malfunctions. The decreasing trend shown for the future rate (Fig. 2) cannot be observed for the average downtime per day in service (Fig. 3) because the average downtime is influenced by the duration of single failures. The high average downtime per day in service for 1989 is based on relatively short times in service and is strongly influenced by one annular preventer failure. Thus, in evaluations of downtime, the average downtime is misleading because relatively few failures of long duration dominate it. Together, the six most failures one half the total observed rig down time in Phase 5 of the study.
The downtime distribution for Phases 2 through 5 equated to 25 % of the failures less than 1.51 hours, 50% of the failures less than 7.5 hours, 75 % of the failures less than 25 hours, and 90 % of the failures less than 80 hours. Data from Phases 2 and 4 of our BOP study also confirmed that the average downtime is much higher during winter than during summer drilling. No significant trends in the failure rate or downtime were observed with respect to different water depths or different areas of the North Sea.
|File Size||1 MB||Number of Pages||6|