No MODUs Adrift
OBJECTIVE: Mooring failures during hurricanes Ivan, Katrina, and Rita caused a number of MODU’s to go adrift. The objective of this research project is to develop technical solutions to technical solutions to (1) prevent or mitigate a MODU going adrift in the event of the failure of its mooring system, and (2) mitigate the drift of an unmoored MODU in a hurricane.INTRODUCTION: Mooring systems failures caused 16 deepwater MODU’s to go adrift during hurricanes Ivan, Katrina, and Rita.
Drifting MODU’s can potentially damage other critical elements of the offshore oil and gas infrastructure, e.g., colliding with floating or fixed production systems and transportation hubs, or damaging pipelines by dragging anchors. The health of this infrastructure has become a matter of national significance because of the importance of deepwater production for the US oil and gas supply and its influence on worldwide process.
The industry is currently conducting a JIP to assess methods to immediately strengthen MODU mooring systems before the next hurricane season. The JIP will also address the hazards and risks of drifting MODUs, assess current API standards to determine if mooring design criteria should be increased, and develop and recommend new criteria if warranted.
The project described here is complementary. While the above JIP is focused more on criteria and standards, this project will focus on technical solutions to prevent a MODU from going adrift should the mooring system fail in a hurricane, and methods to control, reduce, or stop a MODU that has gone adrift in a hurricane.
Industry representatives will be engaged through a Project Advisory Board, meetings, and workshops to bring their experience and knowledge to bear on this problem.
The OTRC has developed considerable knowledge in recent studies that is applicable to this issue, including
For more information on these studies, see the OTRC website at otrc.tamu.edu.
- Synthetic Mooring Lines
- Foundation and Mooring System Risks
- Suction Caissons and Vertically Loaded Anchors,
BENEFITS TO MMS & INDUSTRY: This project will provide cost-effective technical options to mitigate MODU drift due to mooring failures in hurricanes and reduce the hazards and risks to the offshore infrastructure (e.g., floating and fixed production structures, pipelines and flowlines, subsea well systems).
DEPLOYMENT OF RESULTS: This project will provide technical options to (1) prevent drift-offs in the event of a MODU mooring system failure, and (2) reduce or stop the drift of a MODU in the event of loss of station keeping due to mooring system failure during a hurricane. Project Advisory Board members (MMS and industry representatives) will be engaged throughout the project for guidance and input. Other industry members (operators, consultants & contractors) will be invited to open Workshops to suggest options for study, review results, and provide input to preliminary design analyses.
At appropriate stages of study, it is anticipated that various individual technical options will be pursued by industry for final design and implementation.
ANTICIPATED NUMBER OF PHASES: 2
PROJECT PLAN FOR PHASE 1 (2006-2007):
Scope and Plan: The tasks planned for this Phase are discussed below. The task schedule is shown in the Project Schedule. A Project Advisory Board of industry experts will be established to provide ongoing review and guidance. Quarterly meetings are anticipated.
Task 1 Data Gathering - Gather available data on MODU mooring system failures during hurricanes that can be used to understand failures of mooring lines and foundation components, the failure pattern, and validate analysis tools to predict the performance of a MODU & its mooring system during the progressive failure of the mooring system. Potential sources include MMS sponsored study of MODU Failures during Ivan & Lili (M. Sharples), 2005 Hurricane Readiness & Recovery Conference, API/OOC JIP on MODU Moorings, industry sources.
Gather available data on GPS positions of drifting MODUs during hurricanes.
Obtain access to hindcast wind, wave, & current data for hurricanes Ivan, Katrina, & Rita from other MMS sponsored studies.
Task 2 Case Studies - Analyze mooring data to investigate failures of mooring line components & foundation elements (suction piles or anchors) and the pattern of progressive failures that leads to the loss of station keeping and the MODU becoming adrift.
Task 3 Validate Global Analysis Tools to Predict Mooring Failure - Analyze failure cases with existing global analysis tools (e.g., WINPOST, CABLE3D, ORCAFLEX) to validate capability to model progressive mooring system failures during a hurricane (Task 2).
Task 4 Generate Ideas to Prevent MODU Going Adrift - Based on observed or inferred progressive mooring failures (Task 2) and validated analysis tools (Task 3), generate & qualify ideas to prevent total loss of station-keeping (MODU going adrift) by intervention either during or immediately following the failure of a mooring system. Intervention means could include extra mooring lines that were normally slack, a high strength composite tether that was normally slack, highly elastic elements in mooring lines to absorb shock of dynamic loads & prevent line failure, etc.
Task 5 Validate Global Analysis Tools to Predict MODU Drift - Analyze drift cases with existing global analysis tools (e.g., WINPOST, CABLE3D, ORCAFLEX) and hindcast data to validate capability to model MODU drift.
Task 6 Generate Ideas to Slow or Stop a Drifting MODU - Based on drift data (Task 2) and validated analysis tools (Task 5), generate & qualify ideas to either slow or stop a MODU that has gone adrift after a mooring system failure in a hurricane. Ideas could include sea anchors, drag anchors, or torpedo piles & mooring lines that could be deployed remotely.
Task 7 Screen Ideas - Share ideas generated in Tasks 4 & 6 with industry and solicit their input and other ideas. Conduct a screening study of all ideas generated in Tasks 4 & 6 and received from industry. Ideas to be screened to establish technical feasibility and estimate effectiveness, impact on normal MODU operation, costs, etc. Document results of screening study in a report to provide a basis for systematically comparing alternatives.
Task 8 Review Alternatives & Select Promising Alternatives for Design - Distribute screening study report to interested parties in MMS & industry. Conduct a workshop, in which the screened ideas are discussed and the most promising alternatives to (1) prevent MODU going adrift as a result of a mooring system failure and (2) slowing or stopping a MODU adrift in a hurricane are selected.
Task 9 Final Report for Phase 1 - Complete and issue final report for Phase 1.
Task 10 Design Studies - See Phase 2.
Deliverables for Phase 1: Assuming a September 1 start, a Status Report summarizing progress and interim results will be due on February 1 and June 1. A Final Report will be due December 1. A report will summarize the screened concepts and be provided to MMS & OTRC sponsors and interested industry members that are invited to the Workshop.
PROJECT PLAN FOR PHASE 2 (2007 – 2008):
Scope of Work:
Task 9 Design Studies - Complete detailed analyses and preliminary designs of promising systems to (1) prevent MODUs going adrift as a result of a mooring system failure and (2) slowing or stopping a MODU adrift in a hurricane.
Industry consultants & contractors may be involved to assist in completing the preliminary design analyses.
Anticipated Results: Preliminary designs of promising systems to (1) prevent MODUs going adrift as a result of a mooring system failure and (2) slowing or stopping a MODU adrift in a hurricanePRINCIPAL INVESTIGATOR (S) & OTHERS INVOLVED IN PROJECT:
PI(s): M.H. Kim, J. Zhang, C. Aubeny, R.M. Gilbert, E.G. Ward
Others: 2 Graduate Students under the direction of Drs. Kim and Zhang
Date: June, 2007Project Title: No MODUs Adrift
MMS Project: 574 TO Number: 39735
PI: E.G. Ward, Jun Zhang, M.H. Kim
COTR: M. Quinney
Estimated Completion Date: December 2007
Project Description:
The objective of this research project is to develop technical solutions to (1) prevent or mitigate a MODU going adrift in the event of the failure of its mooring system, and (2) mitigate the drift of an unmoored MODU in a hurricane. The project is structured in the following tasks:
Task 1: Data Gathering
Task 2: Case Studies of Mooring and Foundation Failures
Task 3: Validate Global Analysis Tools to Predict Mooring Failure
Task 4: Generate Ideas to Prevent MODU Going Adrift
Task 5: Validate Global Analysis Tools to Predict MODU Drift
Task 6: Generate Ideas to Slow or Stop a Drifting MODU
Task 7: Screen Ideas
Task 8: Review Alternatives & Select Promising Alternatives for Design
Task 9: Final Report for Phase 1Progress:
Work continued to focus on developing and validating computational tools to:
- model progressive mooring system failures (Task 3), and
- model the drift of a MODU that has lost its mooring during a hurricane (Task 5).MODU Mooring Progressive Failure Model (Task 3):
An existing vessel-mooring-riser coupled dynamic time-domain analysis program has been extended to analyze mooring line breakage and the corresponding transient motions. The numerical tool has been used to study the progressive failure of the mooring lines on a generic MODU model to simulate the progressive failure of a MODU mooring system. The MODU model being analyzed is a generic model with rig particulars and mooring arrangements (chain-wire-chain or CWC) similar to an actual MODU (MODU1). The purpose of this analysis is to validate the software’s capability to predict the progressive failure of a MODU’s mooring system during a hurricane by studying the mooring failure and drift off of MODU1 during hurricane Ivan.Results to date include the following:
1. The performance of MODU1 and its moorings was simulated for 1 hour exposures to several hurricane metocean conditions with collinear winds, waves, and currents that included the API 10-yr and 100-yr conditions (API 21st Edition), and the maximum hurricane Ivan conditions hindcast at the MODU1 location.
The MODU survived the API 10-yr and API 100-yr metocean conditions, but failed in both of the more severe Ivan conditions. As expected, the taut side mooring lines broke first, then lateral lines failed, and finally the slack-side lines after the platform moved in the direction of environmental loading over those moorings. The sequence of line failure agrees well with the result of reported forensic analyses. After all lines broke, the platform starts to drift in the direction of the wind and current.
2. The performance of MODU1 was also analyzed in Ivan conditions with several assumptions of non-collinear wind, wave, and current directions. In non-collinear environments, the pattern of mooring line failures was not as straightforward as for the collinear cases, i.e., weather side first and lee side last. The pattern becomes more complex and the mooring system can survive wider spreads of angles as shown below.
An analysis of performance under the non-collinear hindcast conditions for the MODU1 location in Ivan will be completed.
3. The performance of the MODU1 model with the CWC mooring was also compared with a chain-polyester-wire of CPW mooring system for collinear hurricane conditions analyzed above. The CPW system in general functioned better and survived even the Ivan case, in which the CWC failed. The advantage of CPW mooring is expected to become more significant as water depth increases.
Results to date illustrate that the program can predict important features of the progressive failure of a MODU’s mooring system in severe hurricane conditions, and additional studies are underway to further confirm this. A similar analysis for a generic model of another deepwater MODU (MODU2) is also being completed.
The next phase of the work is to use this model to investigate the feasibility of devices that could intervene and prevent a MODU whose mooring system has started to fail from a total mooring failure and drift off (Task 4).
MODU Drift Model (Task 5):
The numerical program DRIFT developed earlier was used to predict the path of a second MODU (MODU2) that went adrift during hurricane Katrina. The predicted path compared well with the track measured by an onboard GPS. The predicted and measured paths for the two MODUs are compared in Figure 1. The path of hurricane Katrina, along with the times that the hurricane and MODUs were at specific locations, is also shown. Note that the map has a distorted scale.MODU2 went adrift at about time 03 hours as the hurricane passed about 40 n mi to the east. The simulation of the drift was started shortly afterwards. The predicted and measured tracks compared well as it drifted eastward about 25 n mi during ~ 4.5 hours, at which time the predicted and measured locations were within about 2.2 n mi.
As previously reported, MODU1 went adrift as the eye approached the MODU. The initial drift was to the west and the eye passed over the MODU. During this period, the MODU was subjected to rapidly changing and poorly defined winds in the eye of the storm. The simulation began after the eye had passed northward and predicted and measured tracks compared well as it drifted about 25 n mi over ~ 5 hours. The predicted location was within about 1.3 n mi of the measured location at that time.
These comparisons illustrate that program DRIFT can satisfactorily predict a MODU’s drift in severe winds near a hurricane.
In the next phase of this work, DRIFT will be used to study the feasibility of slowing or stopping a MODU that has gone adrift in a hurricane (Task 6). The existing program COUPLE will be used to compute the force on the MODU of a device to slow its drift speed, and then that force will be applied to the MODU in a simulation of the MODU adrift in a hurricane. The same combination of software can also be used to estimate the impact of force due to dragging anchors on the drift of a MODU.
Reports and Publications:
1. G. V. Tahchiev, (2007) ‘Numerical Prediction of MODUs’ Drift During Hurricanes’, MS thesis, Texas A&M University.2. G. V. Tahchiev and Jun Zhang (2007) ‘Numerical Prediction of MODUs Drift during Hurricane Katrina’, Proceedings of OMAE 2007, paper No. 29606.
Date: December, 2006Project Title: No MODUs Adrift
MMS Project: 574 TO Numbers: 39735
PI: E. G. Ward, J. Zhang, M.H. Kim
COTR: M. Quinney
Estimated Completion Date: June 2007
Project Description:
The objective of this research project is to develop technical solutions to (1) prevent or mitigate a MODU going adrift in the event of the failure of its mooring system, and (2) mitigate the drift of an unmoored MODU in a hurricane. The project was structured in the following tasks:
Task 1 - Data Gathering
Task 2 - Case Studies of Mooring & Foundation Failures
Task 3 - Validate Global Analysis Tools to Predict Mooring Failure
Task 4 - Generate Ideas to Prevent MODU Going Adrift
Task 5 - Validate Global Analysis Tools to Predict MODU Drift
Task 6 - Generate Ideas to Slow or Stop a Drifting MODU
Task 7 - Screen Ideas
Task 9 - Final Report for Phase 1
Task 8 - Review Alternatives & Select Promising Alternatives for DesignProgress:
Work to date has focused on data gathering (Tasks 1) and developing and validating computational tools to model the drift of a MODU that has lost its mooring (Task 5) and to model progressive mooring system failures (Task 3) during a hurricane.
Data Gathering
Data gathering has been completed. The validation of computational tools to model the drift (Task 5) and progressive mooring system failures (Task 3) will utilize data from hurricanes Ivan, Katrina, and Rita. We have• obtained Oceanweather’s hindcasts for these 3 storms,
• obtained drift tracks for a number of MODUs that went adrift during Katrina and Rita,
• obtained hydrodynamic and wind response characteristics for generic MODU hull shapes that are representative of the MODU cases we will use to validate the computational tools,
• accessed public domain information on mooring system failures during these storms. This information will be used to validate the model for progressive mooring failure.MODU Drift Model (Task 5)
A computational model for MODU drift behavior has been developed. The model considers the wind, wave, and current forces on the structure. The wind load is based on the hindcast 1-hour averaged speed at 10 m elevation. The oscillatory wave forces are neglected, and the steady surge and sway wave forces are based on the unidirectional wave spectra. The relative current-structure velocity is used in computing the net current force. As shown in Figure 1, the first comparisons showed good agreement between the of measured and predicted drift tracks. The simulation starts just after the MODU has gone adrift from a location near the eye of the hurricane, and continues for about 5 hours during which it drifts about 25 miles. Figure 2 shows the wind and wave force vectors on the MODU along the drift track. The wind force is much more dominant than the wave force (note the approximately 4-to-1 difference in the force vector scales). The current force is negligibly small. The predicted drift track is biased toward the northwest and the final predicted position is about 3 nautical miles to the northwest of the actual position. Further examination suggests that this results from characterizing the wave direction as that of the peak of the spectra rather than the higher frequency components that contribute more strongly to the steady forces.Overall, the agreement is remarkable for this case that represents a MODU going adrift from a location near the right front quadrant of the hurricane and drifting to the northeast as the storm moved northward. Additional cases will be considered to further validate the model in different situations.
MODU Mooring Progressive Failure Model (Task 3)
An existing vessel-mooring-riser coupled dynamic analysis program was extended to include the capability of modeling line breakage and the corresponding transient motion in the time domain. The program can also analyze the sequence of line breakage and where each line breaks. The program was first tested for a TLP tendon failure episode and reasonable results were obtained. Other tests showed that the transient response of a moored rig just after line breakage can be unexpectedly large and accelerate further damage of other mooring lines.The MODU model being analyzed is a generic model with rig particulars and mooring arrangements (chain-wire-chain or CWC) similar to an actual MODU (MODU1).The numerical tool is now being used to simulate the progressive failure of MODU 1 mooring lines during hurricane Ivan using input data for a similar generic rig and a similar mooring arrangement. The simulated rig motions and progressive mooring line failure will be compared with the forensic analysis reported for that MODU during hurricane Ivan. We are also developing an animation capability to illustrate the pattern of progressive mooring line failures.
Progress Reports: June 2007 December 2006
