Learn more at

Transcription

1

2 1 TAHITI ONLINE MONITORING SYSTEM FOR STEEL CATENARY RISERS AND FLOWLINES OTC Tahiti Online Monitoring System for Steel Catenary Risers and Flowlines Metin Karayaka, Jen-hwa Chen, Curtiss Blankenship, Chevron ETC, Wolfgang Ruf, 2H Offshore Inc., Mat Podskarbi, Schlumberger Copyright 2009, Offshore Technology Conference This paper was prepared for presentation at the 2009 Offshore Technology Conference held in Houston, Texas, USA, 4 7 May This paper was selected for presentation by an OTC program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect any position of the Offshore Technology Conference, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Offshore Technology Conference 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 OTC copyright. Abstract The Riser and Flowline Monitoring (RFM) project deployed one of the most comprehensive subsea structural monitoring systems to date on a Tahiti infield (production) Steel Catenary Riser (SCR) and associated flowline. State-of-the-art motion and strain measurement devices are optimally placed along the SCR to continuously measure and store real-time full scale riser response. In addition, RFM project is the first to implement monitoring devices on a flowline to measure the flowline buckling a phenomenon that is predicted during repeated start up/shut down. The project goals are two-fold: 1. Understand fundamental hydrodynamic behavior of SCRs and flowlines, specifically, floater motion induced response of catenary risers, Vortex Induced Vibration of catenary risers, riser behavior at the pull tube exit region, riser-soil interaction at the touchdown region, flowline buckling, flowline axial walking, and flow assurance characteristics of infield flowlines. The information generated will be used in future riser designs. 2. The information will be used to validate Tahiti riser and flowline system robustness and conduct health checks on the fatigue critical risers and flowlines, particularly after significant environmental or operational events. This paper describes the monitoring system configuration, the technology deployed, and the installation methods. INTRODUCTION The Tahiti field development is located in the Gulf of Mexico at 4,000 feet of water. More detailed information about the field and the riser systems can be found in References [1], and [2], respectively. Steel Catenary Risers (SCRs) have a proven track record and have been successfully installed in water depth of up to 8,000 feet. However, there is industry wide acceptance that a reasonable level of uncertainty exists with respect to the ability to predict riser performance. Calibration of analytical SCR response prediction models with full scale field measurements will improve design methods, design allowables, and operation integrity. RFM project s main drivers for development of a full scale monitoring system are to understand fundamentals of riser hydrodynamics with full-scale field measurements, calibrate design tools, and validate Tahiti riser system robustness. The monitoring devices and locations are specifically selected to characterize SCR behavior with respect to: 1. Vessel Induced Motion (VIM) 2. Vortex Induced Vibration (VIV) 3. Effectiveness of VIV suppression devices (strakes) 4. Riser behavior at the pull tube exit regions 5. Riser-soil interaction at the touchdown region 6. Flowline buckling, axial walking 7. Flow assurance characteristics of infield flowlines

3 OTC KARAYAKA, CHEN, BLANKENSHIP, RUF, and PODSKARBI 2 HANGOFF REGION TOUCHDOWN REGION FLOWLINE REGION Figure 1 - General arrangement of Riser and Flowline Monitoring system

4 OTC KARAYAKA, CHEN, BLANKENSHIP, RUF, and PODSKARBI 3 The general RFM system configuration is schematically illustrated in Figure 1. The monitoring system devices are placed at three independent regions of the riser. Based on their location, these regions are described as follows: Hangoff Region: Riser joints immediately after the pull tube exit where the riser deformation is affected due to interface with the hull and pull tubes. Touchdown Region: Riser joints immediately before and after the riser seabed touchdown point where the riser deformation is affected by riser seabed interaction. Flowline Region: Flowline joints approximately one mile away from the Touchdown Region where the riser weight is reduced with buoyancy modules to promote controlled buckling. Figure 2 presents RFM system block diagram for power and data communication with the monitored regions. All monitoring devices can be powered and interrogated from the topside facilities. The topside console is accessible from onshore for in-situ response monitoring and transferring collected data. The system is configured with emphasis on system and component redundancy. The system redundancy is improved by placing Hangoff Region on an independent circuit than the Touchdown Region and Flowline Region devices. Selecting different types monitoring devices and placing them at multiple locations increase component redundancy. Figure 2 - RFM system block diagram Communication with the devices at the Hangoff Region is facilitated with electrical conductors. Topside cabling is preinstalled during onshore construction. A conduit is build inside the Spar centerwell and dedicated power and communication cable is pulled inside the conduit before the hull is transported to GOM. The hull cable is terminated at the Spar soft tank with an ROV mate-able connection bracket. All riser monitoring devices are mounted on the riser joints at onshore. An ROV installed 200-ft long jumper connected the hull cable with the monitoring devices on the riser (Figure 3). The Touchdown Region and Flowline Region are also directly connected to the topside facilities. Communication with these regions is facilitated with fiber optic lines as well as electrical conductors. Similar to the Hangoff Region, topside cabling is pre-installed. Power to the subsea is provided with one of the four Tahiti umbilicals..redundant electrical conductors in the Tahiti umbilical, dedicated to RFM devices, facilitated connection from topside umbilical termination unit to subsea umbilical termination unit (SUTA). A RFM project specific 7,000-ft long hybrid electric/fiber optic flying lead (RFM-01) connects the SUTA to RFM Hub (Figure 4) that transfer power to Touchdown Region and Flowline Region monitoring devices. The power to Touchdown Region is provided with a 750-ft long pressure balanced hose electrical flying lead. Termination of the flying lead to the riser Touchdown Region is same as that of the Hangoff Region (Figure 3). A hybrid fiber optic/electrical flying lead, identical to RFM-01, connects the RFM Hub to the Flowline Region. The flowline monitoring devices have fiber optic sensors but low voltage electrical power is required for interrogation of the sensors. The data communication is facilitated with fiber optic lines inside the RFM-02 back to the RFM Hub. The data communication between the RFM Hub and the Touchdown Region devices is facilitated by electrical conductors. All data communication between the RFM Hub and SUTA is facilitated with fiber optic lines in RFM-01. The fiber optic lines in the Tahiti umbilical completed the communication of all subsea monitoring systems with the topside facilities.

5 4 KARAYAKA, CHEN, BLANKENSHIP, RUF, and PODSKARBI OTC Figure 3 - Hangoff riser bracket with flying lead Figure 4 RFM Hub umbilical termination unit Riser Monitoring Devices The monitoring devices for Hangoff Region and Touchdown Region are configured based on measuring riser response (strain or accelerations) of discrete points along the riser. Interpretation of dynamic response of the entire riser from discrete data points is complex and requires careful consideration in the instrumentation design process. The data obtained at discrete locations are extrapolated along the whole riser which requires time domain or frequency domain data processing techniques as discussed in [3], [4] and [5]. A total of 16 measurement stations are installed on the risers. Four (4) stations have motion sensors, six (6) stations have strain sensors, and six (6) stations have combined motion and strain sensors. A measurement station arrangement with combined strain and motion measurement devices is shown in Figure 5. The monitoring stations are designed compact to mitigate potential interference with installation equipment and provide ease of handling. The INTEGRIpod motion sensors, selected for the RFM project, features 3 axis accelerations, 2 axis angular rates, and 1 axis inclination. Table.1 provides a summary of key technical features of the INTEGRIpod motion sensors. Figure 5 INTEGRIstick and INTEGRIpod, attached to the riser with protection frame. Figure 6 - INTEGRIstick dynamic curvature sensor Measurement station on riser

6 5 TAHITI ONLINE MONITORING SYSTEM FOR STEEL CATENARY RISERS AND FLOWLINES OTC Table.1 - INTEGRIpod motion sensor specification Resolution Range 3-axis Acceleration 0.001g +/- 1.2g 2 axis Angular rate +/- 0.01deg/s +/- 10deg/s 1 axis Inclination +/- 0.01deg/s +/- 10deg/s The INTEGRIstick Dynamic Curvature Sensor (Figure 6) is selected to measure bending strains on fatigue critical areas of the riser. These strain sensors are directly installed on top of the insulation and measure curvature changes in the riser. The device measures in-, and out of plane curvatures and corresponding bending strains can be calculated with a resolution of 3 micro strain. This high sensitivity allows capture of lowest bending moment at any sea-state. A comprehensive qualification program is conducted to confirm performance of the strain measurement devices under RFM system design conditions. The qualification tests included a full scale 4-point bending test conducted on a full scale riser joint. The test was used to ensure that applied bending loads measured with the strain sensor are accurate, repeatable, and creep free. The riser monitoring system design need to address the conflict among technical needs, system integration requirements, and overall system cost. From technical point of view, event characterization accuracy increases with increase in number of sensors. The system integration requirements impose limits on the communication methods with the devices and location of the devices on the risers. Overall cost of the system increase with increase in number of sensors. It should be noted that the cost of the sensor devices are small compared to the knock-on cost increases in connectors, cables, and installation. In the ensuing sections, the selection of number and location of sensors are briefly summarized. Characterization of VIV induced riser motions The number and location of measurement stations depends upon the range of modes expected to be excited and level of accuracy required. In principle, to capture the VIV response, spatial extent of the instrumentation should enable to capture at least a quarter wave length of the shape of lowest mode number expected. To capture the shape of the highest mode expected, there should be at least two instruments available to capture the quarter wave length. The motion sensors placed are placed at five location over 320-ft long region of the riser at the exit of the pull tube. The selected configuration is suitable for characterizing all significant VIV modes. The sensors placed at the Touchdown Region increases the resolution of VIV characterization. A technique to obtain an optimum instrumentation locations and the number required is discussed in [6]. Characterization of vessel induced riser motions As shown in Figure 7, the entire range of fatigue seastates is analyzed and SCR motion response is extracted along the entire riser length. The motion sensors are distributed along the critical locations of the riser. The sensor accuracy of m/s 2 is chosen to reflect such that the majority of the fatigue loading can be captured. This accuracy level is suitable to detect riser motions even at the lowest defined fatigue seastates. Note that the monitoring devices are selectively placed at highly accelerated regions and the spacing is suitable to capture the acceleration profile over the region.

7 6 KARAYAKA, CHEN, BLANKENSHIP, RUF, and PODSKARBI OTC TDZ EXCURSION Chevron Tahiti ACCELERATION OF DESIGN SEA-STATES 0.08 Acceleration (m/s^2) MKII Sensor Noise 0 Floor Location from Nominal TDP (ft) Bin1 Bin2 Bin3 Bin4 Bin5 Bin6 Bin7 Bin8 Bin9 Bin10 Bin11 Bin12 Bin13 Bin14 Bin15 MKII Figure 7 - Wave vessel motion induced riser Acceleration with motion sensor accuracy Characterization of Hull and Seabed interaction The SCRs have discontinuities at the hull attachment and riser contact points; therefore, these locations are the most critical riser regions due to elevated strains. Figure 8 shows strain distribution along the riser length at the Touchdown Region for a selected fatigue seastate. Although there is a clear spike in the maximum strain, its location depends on the hull offset due to environment direction. The strain monitoring devices needs to be distributed over a large region to characterize the overall shape of the strain disturbance at the Touchdown Region for all offset conditions. Total number and distribution of the monitoring stations are determined by studying strain profile for a wide range of environmental conditions. A total of 10 strain sensor are installed to capture highest bending loads expected regardless of vessel offset and soil interaction response. The motion sensors are placed at 4 locations. The sensors are placed over 640 ft, along four separate quad joints. The riser strain field is also disturbed at the pull tube exit. However, unlike the Touchdown Region, the strain disturbance is confined to a smaller region near the pull tube and the location is not effected by hull motions. For the Hangoff Region five measurement stations are installed. Two of these stations measure combined strain and motions, three of the stations measure riser motions. These measurements along with pull tube monitoring device measurements, which are installed through a separate Tahiti monitoring project, help characterize pull tube and riser response with full scale measurements.

8 7 TAHITI ONLINE MONITORING SYSTEM FOR STEEL CATENARY RISERS AND FLOWLINES OTC Chevron Tahiti SCR Monitoring TDZ STRAIN - FATIGUE SEASTATE BIN 8, 24.6ft Hs Strain ( s) Location from the Nominal TDP (ft) 4 Quad Joints 3 Quad Joints 2 Quad Joints Far - 2.2% WD Offset Bin8-24.6ft Hs Near - 2.2% WD Offset Figure 8 Touchdown Region curvature versus monitoring locations Flowline Monitoring Devices Tahiti high temperature, high pressures (HPHT) flowlines are expected to experience thermal expansion and contraction during operation. For flowlines below a critical length, the expansion can lead to buckling and possible failure. To accommodate sizeable expansions and contractions of high temperature high pressure flowlines during start up and shut down, buckling is initiated at controlled locations [2]. On Tahiti flowlines this controlled buckling zone is introduced through installation of the buoyancy modules at two locations along the flowline. Buoyancy modules reduce the weight of the flowline in water, decreasing the friction between the flowline and seabed allowing for lateral buckling. High stresses occur during buckling and repeated cycling can lead to failure by fatigue. Typical inspection methods use sidescan sonar or ROV survey to capture a snapshot of the buckle at a single moment in time. These sonar images provided limited resolution and furthermore reveal nothing about the actual historical behavior of the buckle between snapshots. Continuous monitoring gives a complete history of the buckle response to operational cycles and shut downs. This enables a far more accurate determination of accumulated fatigue, and allows any accumulated buckle growth or movement to be monitored over the life of the field providing crucial data for the integrity management and understanding buckling phenomenon. The flowline monitoring system measures curvature over four quad joints, approximately 640 ft, with subc-strip devices (Figure 9) as well as temperature, axial and hoop strain measurements via subc-hat device (Figure 10). The subc-strip is 160 ft long I section of the composite with fibers laid axially along it. It is attached to the flowline on the outside of the insulation via series of straps. The subc-strip is following the flowline lateral shape transferring the curvature into fiber Bragg gratings (FBG). The fibers are laid within the strip separated by the width of the strip. The measurement of the axial curvature at the opposite sides of the strip at number of locations along the flowline enables calculation of curvature profile in the buckle zone as shown in Figure 9. The subc-hat is installed under the insulation half-wrapping around the flowline in order to take measurements of pipe wall temperature, axial curvature to calculate axial force and hoop curvature to calculate internal pressure. These measurements allow for comparing the effects (curvature, buckling) with causes (axial force, internal pressure and temperature) to study the mechanism of flowline buckling under internal forces and expansion due to changing flow conditions. This is shown graphically in Figure 11. Flowline measurement devices are based on fiber optic Bragg grating (FBG) technology. The FBG can be considered as a linear optical stain gauge. It is responsive only to axial curvature and when aligned appropriately can be used for measuring hoop, axial, bending and torsional curvatures as well as temperature effects. Being linear and absolute, free from thermal balance issues it does not require repeated recalibration nor is it susceptible to crosstalk issues. It is possible to place several sensors at different locations in a single optical fiber, commonly referred as multiplexing. By multiplexing sensors in this way, over a hundred sensors can be monitored from a single instrument using a single connection, as was done in case of Tahiti. The principles of FBG technology are shown in Figures 12 and 13.

9 OTC KARAYAKA, CHEN, BLANKENSHIP, RUF, and PODSKARBI 8 Figure - 10 Flowline pressure, temperature and axial curvature measurement Figure 9 Riser monitoring joints on barge

10 OTC KARAYAKA, CHEN, BLANKENSHIP, RUF, and PODSKARBI 9 Figu12 Relation between curvature and wavelength shift Figure 13 Typical layout of FBG curvature measurement system Figure 11 Flowline buckling characterization with monitoring devices MONITORING SYSTEM INSTALLATION The Tahiti SCRs are installed with J-Lay installation method. Riser joints (quad joints) are made of four individual pipes and J-Lay collar to facilitate installation. The quad joints are prepared onshore and attached together offshore at the lay barge. As discussed in the ensuing sections, each monitoring region consists of devices spread over multiple quad joints. The RFM project evaluated various installation methods to adapt to J-Lay requirements. The selected installation method designed and attached all monitoring devices on individual quad joints at onshore facilities. Then, the devices on each quad joint are connected to each other with dry-mate connectors during J-Lay installation. This approach reduced potential installation damage risk to the monitoring equipment, but required interfaces with a wide range of project disciplines and planning during the early phases of the Tahiti project. The RFM system could not be configured unless the Tahiti project disciplines are not engaged during early construction phase. During the design phase of the project many requirements had to be considered to allow for a smooth and successful installation both onshore and offshore. The project had to be executed alongside a major offshore development project without interfering with progress or requirements to change existing equipment or installation processes. The RFM project had five major installation campaigns: 1. Installation of Hangoff Region cables in hull conduits at Spar construction yard. 2. Attachment of monitoring equipment to risers and flowlines at multijoint onshore construction yard. 3. Installation of interconnecting cables between riser quads during J-Lay process 4. Installation of subsea cables connecting the monitored regions to the facility 5. Topside commissioning and system start up The principle of the installation process was to install as much of the equipment onshore as practical and to minimize any impacts on offshore installation process. The installation of the riser monitoring system onto the riser has to consider all obstruction and interferences during handling, transportation and J-Lay installation. The quad joints are positioned with a crane and supported in selected areas. The Tahiti fatigue sensitive riser quads are end-matched; therefore RFM quads are radially pre-aligned to allow for optimum concentric overlap. Following strake installation, the measurement stations monitoring devices are placed onto the riser and secured with corrosion resistant metal straps. Interconnecting cables are routed through the strakes and securely strapped on to the riser. Following installation completion of the riser monitoring devices the monitoring quads are transported to the offshore installation vessel via barge (Figure 14). The quads are transported from the barge via spreader bar and picked up from a strongback clamp (Figure 15). The strongback orients the quads from a horizontal, - to a vertical position. Many pinch-points had to be considered in the design stage to ensure system health after installation.

11 10 KARAYAKA, CHEN, BLANKENSHIP, RUF, and PODSKARBI OTC During welding of the quads in the weld-house no cables are acceptable that would interfere with the welding operation. Early stages of the RFM system design considered the use of ROV installable flying lead to be installed between every quad joint. This design however was not practical and cost effective due to the requirements of additional connectors and flying leads - thus driving potential failure point and increased costs. The restriction was resolved with the use of cable storage baskets that are placed about 20 feet from end of each quad joint as shown in Figure 16. The cables are equipped with dry mate-able connectors. A steel messenger wire is connected to the end of the connector, after the weld and field joint application is completed the weld table, pedestal and strong back will be opened. The messenger wire is used to pull up the stored cable in the basket and connection is made with the connector from the previous joint. Figure 14 Riser monitoring joints on barge Figure 16 Cable basket with dry-mate able connector and messenger wire in weld house Figure 15 Instrumented Flowline Region quads at DB50 deck Figure 17 Instrumented quad in the DB50 strongback

12 OTC KARAYAKA, CHEN, BLANKENSHIP, RUF, and PODSKARBI 11 The hoop, axial strain measurement device was installed on the pipe onshore and coated with insulation in order to maintain thermal insulation integrity along the pipeline. Final connection of the flowline monitoring devices and other components of the system was done offshore during J-lay installation. Installation of the subc-strip assemblies was more challenging due to its length. First step was to install 160 ft long subc-strips on reels for transportation from manufacturing facilities to test locations and then to pipe yard for onshore installation activities. The subc-strips were installed on 4 joints together with buoyancy modules and located on the barge in a designated area as shown in Figure 17. From the DB50 deck, instrumented quads were transported to the J-lay tower via a spreader bar and then through a strongback. Final connection of the subc-strips between the joints and its connection to the subc-hat and ROV connector was done in the weld-house during J-lay process. As with the riser system, particular care had to be taken in order to ensure that the monitoring system survives the transportation and installation process. Number of pinch-points at the strong-back and J-lay stinger presented had to be taken into account during design of the equipment and definition of the installation procedure. Data Acquisition and Software A comprehensive data acquisition system was installed as part of the project scope. The data acquisition system is housed in a data cabinet in the main control room of the Tahiti facility. The functionality of the Data acquisition system is summarized as follows: 1. Data acquisition from the Hangoff, Touchdown, and Flowline Region sensor stations 2. Data storage on fixed and recoverable media 3. Display monitoring system status, operation log and warnings; 4. Provide LAN access from onshore location to fully service the system The data acquisition system is comprised hardware components such as UPS power supplies, data storage servers for riser and flowline monitoring data, Server with LCD screen and interface units that control all monitoring location. The data acquisition system is connected to the hurricane bus of the vessel. These features enable the monitoring system to stay alive even during severe events such as hurricanes Figure 19 Flowline Monitoring GUI Figure 18 Riser Monitoring Data Acquisition Software GUI Conclusions The RFM project successfully installed on-line monitoring devices on one of the Tahiti production risers. At the preparation time of this paper, the Hangoff Region devices are commissioned. The flying lead (RFM-01) for the seabed monitoring devices is scheduled for installation. The RFM project demonstrated that a technology driven project can be integrated into a major capital project without interfering with the project schedule and activities. The RFM project interfaced with a large number of project disciplines;

13 12 KARAYAKA, CHEN, BLANKENSHIP, RUF, and PODSKARBI OTC operations, topsides, hull construction, subsea umbilical design, riser installation subsea installation, and seabed layout. Successful integration of the RFM project and management of interfaces are accomplished by defining project objectives and framework during early construction phases of the Tahiti project. For technology projects with size and complexity similar to RFM project, this timing appears to be optimal. Earlier engagement would lead to recycle in the technology project. Late engagement would limit flexibility. Acknowledgements The authors are thankful to Chevron Energy Technology Company and Tahiti Project for project funding. The project acknowledges contributions of the following Tahiti Project team members for help managing interfaces; Mike Brashear, Walt Clingo, Ken Farnsworth, Paul Griffin, Bob Kirchhoff, Jim Reiners, Scott Springman, and Earl White. ODI provided connectors and valuable information to system design. J Ray McDermott helped integration of the RFM system to J-Lay riser installation process. The authors also acknowledges contributions of RFM project team members Dr. Pei An, Sandip Ukani, Hernan Castro, Daniel Reagan, Michael Ritchie of 2H Offshore Inc. and Dillwyn David, Matthew Sell, and Damon Roberts of Insensys. REFERENCES [1] Varnado, B.R., Tahiti Development Overview, OTC OTC, Houston, TX, USA, May [2] Thompson, H,, Reiners, J., Brunner, M., DeLack, K., Qi, X.,Noel, C.D., Tahiti Project Fatigue Sensitive Flowlines- Design and Installation OTC 19858, OTC, Houston, TX, USA, May 2009 [3] Cook, H., Dopjera, D., Thethi, R., Williams, L.: Riser Integrity Management for Deepwater Developments, OTC, Houston, TX, USA, 1-4th May [4] Campbell, M., Shilling, R., Howells, H. Drilling Riser VIV Analysis Calibration Using Full Scale Field Data, Deepwater Offshore Technology, Vitoria, Brazil, 8-10th Nov [5] Thethi, R., Howells, H., Natarajan, S., Bridge, C., A Fatigue Monitoring Strategy & Implementation on a Deepwater Top Tensioned Riser, OTC, Houston, TX, USA, 2-5th May [6] Podskarbi M., Thethi, R., & Howells, H.: Fatigue Monitoring of Deepwater Drilling Risers, Subsea Rio, Rio de Janeiro, Brazil, 8-10th June [7] Karayaka, M., Ruf, W., Natarajan, S., Steel Catenary Riser Response Characterization with On-Line Monitoring Devices, OMAE [8]. Campbell, M., Shilling, R., Howells, H. Drilling Riser VIV Analysis Calibration Using Full Scale Field Data, Deepwater Offshore Technology, Victoria, Brazil, 8-10 th Nov [9]. Thethi, R., Howells, H., Natarajan, S., Bridge, C., A Fatigue Monitoring Strategy & Implementation on a Deepwater Top Tensioned Riser, OTC, Houston, TX, USA, 2-5 th May [10]. Natarajan, S., Howells, H., Deka, D., Walters, D., Optimization of Sensor Placement to Capture Riser VIV Response, OMAE, Hamburg, Germany, 4-9 th June [11]. Karayaka, M., Podskarbi, M., Walters, D., & Hatton, S., Design Consideration of Monitoring Systems for Deepwater Catenary Risers, ISOPE, Lisbon, Portugal, 1-5 th July 2007.