SPE 63112 Application of a Completion Equipment Reliability Database in Decision Making

Einar Molnes, ExproSoft and Geir-Ove Strand, SINTEF Petroleum Research

Copyright 2000, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 2000 SPE Annual Technical Conference and Exhibition held in Dallas, Texas, 14 October 2000. This paper was selected for presentation by an SPE Program Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

potential benefits which can be drawn from such databases. Some industry cases are now established which have demonstrated the cost saving potential of such databases. Examples of applications of reliability data are: Risk and reliability studies LCC/LCP analysis Tender evaluations and purchasing decisions Rig contracting strategies Incentive based contract definitions Downhole barrier acceptance criteria definitions

Abstract A study (JIP) on reliability of well completion equipment (Wellmaster Phase III) was completed by SINTEF in November 1999. This has resulted in a comprehensive database on well completion equipment, with a total of 8000 well-years of completion experience represented and more than 1000 downhole failures included, given as input from the 16 funding oil companies of this JIP. The database represents all categories of downhole equipment, from tubing hanger level down. The paper points towards the major contributors to well interventions and downtime, indicating industry average and benchmark failure rates of the most vital completion components. A historical evolution in reliability of Subsurface safety valves (SCSSV) is demonstrated, and the industry wide effect of reliability improvements is shown through specific examples. In the North Sea, reliability data has gained widespread acceptance for use in decision making. The paper lists several cases where reliability data of downhole equipment has been used with a major impact on field development and subsequent operational expenditures.

Cautiously defined and consistent reliability data collection requirements is a prerequisite for successful 1 reliability databases. The new ISO 14224 standard constitutes a valuable reference in this context. The Wellmaster Phase III project objective has been to contribute to improvement in completion equipment reliability through systematic collection, analysis and feedback of reliability data to participating oil companies and equipment manufacturers. The main deliverable from the project has been the new Wellmaster data collection software for completions with an integrated analysis tool, an updated database on completion equipment and reliability statistics and a 2 summary report on main findings. Data analysis has focused on in-service equipment failures, defined as failures occurring from 6 days after landing the tubing hanger on the wellhead. Failures occurring prior to that are defined as installation failures, and a fair amount of these failures have also been reported. All failure reported are also listed in a webapplication where the Wellmaster JIP member companies have access.

1. Introduction Reliability data has gained widespread use in the offshore business due to industry studies like OREDA, Wellmaster and others. The introduction of statutory codes and regulations in a number of oil producing countries has also strongly accelerated this development. During the last decade, offshore industry managers have become increasingly aware of the

EINAR MOLNES AND GEIR-OVE STRAND

SPE 63112

Work process description By means of the Wellmaster software, a completion schematic is built to represent the completion configuration and give details on the equipment in the well. Failure data are captured directly via this schematic by pointing at the failed item and entering detailed data. Failure modes unique to each item are predefined and can be selected when entering information on new failures. In this way, a comprehensive and consistent database can be generated. An integrated processing package is then used to prepare a range of different reliability reports from the database (MTTF, failure mode distribution, MTTW, run time distributions etc.). Upon reporting of failure data, an e-mail connection can be established with the manufacturer with the possibility for the manufacturer to feed back information on the likely cause of equipment failure. Well and equipment performance data is normally collected by the operating company itself, or through assistance from a contractor. The data is then checked for consistency and quality in accordance with the agreed data collection requirement. Upon data quality compliance, the data are merged into a master database, which in turn is fed back to the contributing member companies on CD-ROMs. A subset of the data (equipment failure data) can be viewed by the JIP member companies in a dedicated Internet browser, WellWatch. The different steps of the data collection, QA and data feedback cycle is shown in Figure 1.

but a significant amount of data from water injectors and gas/condensate producers are also included. A breakdown of well type vs. well service time is shown in Figure 3. With respect to completion type, the data are dominated by data from fixed offshore platforms (79.6 %), but a considerable proportion also exists for subsea completed wells (11.4 %). The remaining data are from TLP wells (6.1 %) and onshore wells (2.9 %).

2. Case histories (benefits) Heidrun TLP The first TLP (Tension Leg Platform) on the Norwegian continental shelf was installed in Saga Petroleum's Snorre field. The production risers for the Snorre TLP are fitted with passive fire protection. The decision to use passive fire protection on the risers was based on detailed regulations issued by the Norwegian Petroleum Directorate (NPD) and a series of risk analyses performed by Saga Petroleum. In 1992, the NPD issued the new risk analysis 3 regulations . These regulations state that a risk analysis should be performed for all major potential risks associated with offshore field development and operation. The results from the analysis shall be measured against risk acceptance criteria which have to be pre-defined by the operating oil company. If the risk level as demonstrated by the risk analysis is below the pre-defined acceptance criteria, the results from the risk analysis may in some cases override the requirements given in more detailed regulations concerning certain safety systems. The Heidrun platform is the 2nd TLP installation on the Norwegian Continental shelf. For this development, several risk analyses were performed which addressed the need for passive fire protection of the risers. SINTEF was requested to do a 3rd party verification of these studies, in order to produce input to the final decision concerning the issue of fire protection of the Heidrun risers. The decision point in these risk analyses was the blowout escalation risk, i.e the risk that a blowout on one of the oil production wells would escalate to additional wells - thus jeopardizing the entire Heidrun TLP. The blowout escalation risk is directly related to the rate of critical failures of the downhole safety valve (DHSV/SCSSV). In the 3rd party verification study, a review and update of the studies with the latest reliability data for SCSSVs

Database contents The Wellmaster database is currently the most comprehensive completion equipment database worldwide, with participation from 16 major oil companies in Phase III. Key figures on the database scope are as follows: More than 71000 completion string items represented in equipment database A total of 1002 equipment failures included for a total of 5 different completion equipment categories A total of 1613 wells with 1921 completions are included, representing a total of some 8000 completion-years of experience

The majority of data in the Wellmaster database is from the North Sea. Figure 2 illustrates the data distribution by region. In Phase III, a fair amount of data from the Gulf of Mexico has also been added, whereas earlier project phases have included almost exclusively North Sea data. The database is dominated by data from oil producers,

SPE 63112

APPLICATION OF A COMPLETION EQUIPMENT RELIABILITY DATABASE IN DECISION MAKING

was performed. This clearly indicated that the blowout escalation risk for the Heidrun TLP was within the predefined acceptable level. This led to an approval from the NPD to develop the Heidrun field without passive fire protection of the TLP production risers. This led to a cost saving of minimum NOK 750 million (ca USD 120 million) for the Heidrun TLP when compared to the Snorre TLP. Without the availability of updated and recognized, independent SCSSV reliability data this cost saving would not have been achieved.

U.K., with a study underway in Australia as this paper is written. In the authors opininon, there is no general answer to this question. This issue has to be addressed on the basis of local/regional regulations concerning requirements on external protection of subsea wellheads and x-mas trees and the risk picture (dropped object risk, trawlboard impact risk and other potential external damage factors). Studies have also been performed where nonconventional completion configurations have been studied, typically on removal of annulus safety systems from gas lifted wells.

Alternative SCSSV leakage acceptance criteria 4 A paper presented earlier this year summarized the findings from a study which has looked at alternative leakage acceptance criteria for SCSSVs. The basis for the widely used API RP 14B was reviewed and compared with an alternative method to define acceptance criteria (leak rate levels). This alternative method utilizes principles from risk and reliability analysis to suggest a systems oriented approach for primary and secondary barriers (the x-mas tree master valve and the SCSSV) in combination - rather than looking at these items on an isolated basis. Rather than applying the API RP 14B criteria for all well types, the paper suggests the use of a matrix with recommendations on well type specific risk levels and corresponding acceptance criteria. Compared to todays practice, this implies a certain relaxation of acceptance criteria without compromising the overall safety level. This is achieved through increasing the test frequency whenever a situation arises in which one of the main barriers (PMV/SCSSV) has failed according to the existing API RP 14 B criteria. The direct implication of this is that considerable cost savings can be achieved during the wells lifetime. The cost saving potential is greatest for subsea producers, due to the high cost of interventions. The results from the work described in this paper are now carried forward and implemented into a new NORSOK standard for risk based acceptance criteria for SCSSVs. This standard will be applicable for the Norwegian Continental Shelf, but can also be applied internationally. The standard is expected to be ready by the end of 2000.

2. Results The Wellmaster Phase III project extended previous historical data provided by SINTEF. The Ekofisk Bravo blowout in Norway in April 1977 paved the way for a collective effort on improvement of safety levels for Norwegian offshore installations and was the basis for SINTEFs initial reliability study on SCSSVs which was published in 1983. Since then, an unbroken chain of historical data on performance of both SCSSVs and other completion equipment data has followed. Figure 4 illustrates the historical evolution in SCSSV reliability. A significant improvement in SCSSV performance has resulted, from an initial Mean Time to Failure (MTTF) of 14.2 years (1983) to the most recent result of 36.7 years (1999). This represents a tremendous boost in well production availability and availability of the SCSSV as a safety barrier. Morever, downhole reliability data has become an important instrument in communication with interpretation of auhorities rules and regulations in many offshore regions around the world. The challenge for many operators is the time lag from demonstration of performance improvement to revisions of governmental regulations. A distinct trend in well completions is the increased preference towards the use of single rod piston, flapper type tubing retrievable safety valves without equalizing feature. This trend towards design standardization is paying off when looking at this purely from a SCSSV reliability and well intervention standpoint. However, as a result of the reduction in piston area, the control pressure needs to be increased. This has some negative effects for subsea completions, with high control umbilical pressures, increased probability of hydraulic leaks in control pods, subsea hydraulic connections and other. High completion equipment reliability is particularly important for subsea completed wells, particularly in

SCSSV removal from subsea completions? A hot issue over the last 1-2 years involving extensive use of completion and subsea equipment reliability data in risk analysis is the issue of SCSSV removal from subsea completions. Studies have been performed on this issue in Brazil, Gulf of Mexico (through) and in the

EINAR MOLNES AND GEIR-OVE STRAND

SPE 63112

deep waters. This is illustrated by some examples from the North Sea, which were reported in the Wellmaster Phase III project: Well A experienced a failure (leakage in closed position) of the TR-SCSSV in June 1995. Upon failure diagnosis of the well, a tubing to annulus communication due to a leaking GLV was noted. The well was shut down and a subsea workover followed. Due to problems with rig availability and production allocation restraints, the well was off production until October 1996. During the workover, the failed GLV was replaced and an insert valve was run inside the failed TR-SCSSV. Total well downtime was 476 days. Well B also involved a failure of the SCSSV. A critical failure (leakage in closed position) occurred on the TRSCSSV during initial completion, in August 1997. The failure was caused by a coiled tubing bottom hole assembly hanging up while pulling out of hole at TRSCSSV depth, indicated through flapper and hinge pin damage during the subsequent workover. Workover 1 took place during January/February 1998. The tubing hanger was found to be stuck in the wellhead and failed to release. Dolomite particles from a kill pill was found to be jamming the tubing hanger lock/unlock mechanism. The workover was aborted, and preparations were made to re-enter the well through Workover 2, where the tubing hanger was to be milled out. Workover 2 was performed during April/May 1999, when the tubing hanger was successfully milled, the tubing pulled and a new completion string run. The well was brought back on production in May 1999 after a shut-in period of 482 days. In both cases, the shut-in period was extended due to limitations in rig availability. Both wells are prolific producers, causing significant loss of revenues in this period.

throughout the organizations and more frequent updating of the historical data now high on the agenda of the participating oil companies. Collection, analysis and feedback of reliability data is an effective means of communicating performance of equipment and operations across organizations boundaries. The fact that reliability data is now being utilized for definition of new standards and governmental regulations, is another indication of the need for continuous collection and analysis of reliability data by independent third parties.

ACKNOWLEDGEMENT The authors wish to thank the participants of the Wellmaster Phase III project for permission to publish this paper. The participants were A/S Norske Shell, Amerada Hess Norge A/S, BHP Petroleum Pty. Ltd., BP Exploration Operating Company Ltd., British Gas Exploration and Production plc, Enterprise Oil plc Mobil Exploration Norway Inc., Norsk Agip A/S Norsk Hydro ASA, Norske Conoco AS,TOTAL Norge A.S, Saga Petroleum ASA, Statoil, Chevron Petroleum Technology Company, Exxon Production Research Company and Texaco Group Inc.

NOMENCLATURE GLV MTTF MTTW NORSOK OREDA SCSSV TR WR Gas lift valve Mean time to failure Mean time to workover The competitive standing of the Norwegian Continental Shelf Offshore Reliability Data Surface Controlled Subsurface Safety Valve Tubing retrievable Wireline retrievable

3. Conclusions Reliability databases have been present in the offshore industry for some two decades, with the number of applications of reliability data in decision making continually growing. The key to successful reliability databases in the offshore sector lies in populating the databases with data in sufficient quality and quantity. In the current low cost era of the industry, it is difficult to convey the benefits of these databases to management. A wider acceptance of the benefits of the results now seems to be emerging, with current focus on stronger implementation of the data collection software REFERENCES /1/ ISO 14224: Petroleum and natural gas industries Collection of reliability and maintenance data for equipment. Issued by ISO/TC 67/WG 4 N5. Molnes, E. and Strand, G.-O.: Reliability of Well Completion Equipment Phase III Main Report. SINTEF Petroleum Research Report 32.0898.00/04/99 (Confidential). Trondheim, November 1999.

/2/

SPE 63112

APPLICATION OF A COMPLETION EQUIPMENT RELIABILITY DATABASE IN DECISION MAKING

/3/

The Norwegian Petroleum Directorate: Regulations concerning implementation and use of risk analysis in the petroleum activities. Molnes, E. and Strand, G.-O.: Towards risk based acceptance criteria for downhole safety valves. Paper presented to Petrobras VI Technical Meeting Reliability Engineering, Rio de Janeiro, 28-30 March 2000.

/4/

EINAR MOLNES AND GEIR-OVE STRAND

SPE 63112

Buffer Database Manufacturer

Failure report duplicate

Failure cause comment

Moderator/ QA

Wellmaster Client version 2.5

E-Mail

Wellmaster Windows NT Agent Online WellWatch Database

ExproSoft Wellmaster Server

Company Database

Generate automatic reply and confirmation

WellWatch HTML Web Browser

Users - Oil companies

Users - Manufacturers

Figure 1 Wellmaster/WellWatch information flow chart.

Distribution Wells by Region

Adriatic 0.1 % GoM 8.1 % Africa 2.7 %

S.E. Asia/Australia 2.5 %

Other 0.1 %

North Sea 86.5 %

North Sea

Other

GoM

Adriatic

Africa

S.E. Asia/Australia

Figure 2 Distribution of wells by region.

SPE 63112

APPLICATION OF A COMPLETION EQUIPMENT RELIABILITY DATABASE IN DECISION MAKING

Service Time (well years)

5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0 Oil production Water well injection well Gas production well Gas injection well Other

Welltype
Figure 3 Distribution of well service time by welltype.

50 45 40 35 30 25 20 15 10 5 0

36.69

MTTF (years)

20.2 14.2 16.4 12.7

19.55

SCSSV I (1983)

SCSSV II (1986)

SCSSV III (1989)

SCSSV IV (1992)

Wellmaster II Wellmaster III (1996) (1999)

Figure 4 Historical development of TR-SCSSV (flapper valve type) reliability during the period 1983-1999.