Risks in DP Shuttle Tanker Offloading Operations Dr Haibo Chen 29 May 2013
CeSOS Highlights in 10 Years!
Main Contributors (1999 – 2013) Dr. Haibo Chen Scandpower Inc., Lloyd’s Register Beijing, China
Prof. Torgeir Moan CeSOS, Dept. of Marine Technology NTNU, Trondheim, Norway
Dr. Sverre Haver, Mr. Harald Kleppestø Mr. Kjell Larsen Statoil, Norway
Dr. Jan Erik Vinnem Preventor AS, Stavanger
Capt. Helge Samuelsen Mr. Arve Lerstad Ship Modelling and Simulation Centre Trondheim, Norway
Mr. Kåre Breivik Sevan Marine Arendal, Norway
We would like to thank Mr. Kjell Helgøy from Teekay for information on DP safety and incidents, and Mr. Torbjørn Hals from Kongsberg Maritime for information on DP control system.
What are offloading concepts: late 1970s - Present
•
Indirect offloading: – A small offloading platform – Submerged offloading systems such as OLS, SAL, STL
•
Direct offloading: – Tandem/alongside from ship-shaped units (weather vane), typical 80 m distance – Normally with hawser and hose connections
What are risks: Collision and Oil Spill • DP shuttle tanker position loss • Towards installation: Collision risk – DP shuttle tanker collision had occurred on Emerald FSU, Gryphon FPSO, Captain FPSO, Schiehallion FPSO, Norne FPSO, Njord FSU and several loading platforms in the past 15 years in the North Sea. – The highest shuttle tanker speed at contact had reached 2.4 knots. – The impact energy involved in these collision ranges from a few MJ to around 100 MJ. In one of the collision incident flare tower supporting structure located on the stern of FPSO suffered damage.
• Away from installation: Hose rupture and oil spill risk
Has oil spill happened before? •
On the 20th August 1980: One fatality on the bow of a shuttle tanker performing offloading. The hose was ruptured after hawser breakage and timely shutdown of crude offloading was not achieved. The crude oil pumped to the shuttle tanker bow area and caught fire. (station-keeping problems)
•
On the 12th December 2007: Statfjord A loading buoy A rupture in the hose resulted in around 4,400 m3 of crude oil being pumped into the sea. (hose integrity problems)
Collision Frequency Model (1999 – 2003) • Practical frequency model between DP shuttle tanker and FPSO P(collision ) = P(drive − off ) × P (Failure of recovery drive − off ) – P(drive-off): Frequency of shuttle tanker drive-off forward which has the potential to cause collision. – P(Failure of recovery | drive-off): Failure probability of recovery initiated from shuttle tanker, given a shuttle tanker drive-off forward.
• Identify vulnerable situations for drive-off: surging and yawing • Recognize human recovery element and investigate how to improve The modelling work was summarized in Journal of Reliability Engineering and System Safety 84 (2004) 169-186.
Drive-off Forward: Causes and Frequency •
DP2 shuttle tanker generic drive-off forward frequency is around 0.02 to 0.05 per year (assume 24 hours operation, 1 offloading per week) – Incidents information from 1980s to 2009. Estimated DP shuttle tanker offloading hours based on industrial statistical sources. – Assumptions related to DP2 vs. DP1 shuttle tankers
•
Main causes – Controllable pitch propeller, e.g. failure in pitch control, or feedback – Position reference systems, e.g. common mode failure of DARPS – Sensors, e.g. failure of one wind sensor, hawser tension sensor, or draught sensor – DP software, e.g. software error, hidden bugs – Human operators, e.g. human errors by DP operator
Vulnerable Situation: Relative Motions in Offloading • Tanker is vulnerable to drive-off when relative motions between FPSO and tanker is excessive. Surging
Yawing Normal
Normal
Hawser FPSO
Tanker
Hawser Tanker
FPSO
Hose
Wind
Hose
Fishtailing
Wind, Wave, Current
Heading Deviation
Wave, Current Tanker
FPSO
FPSO
Surging
Yawing
60 m
Tanker 80 m
Hs = 5.5 m loading / 4.5 m connection
SIMO Simulation Work (1999 – 2003) •
Objectives: – Predict how likely excessive surging and yawing will happen. – Identify effective measures to reduce these occurrences.
•
Simulation work in cooperation with Statoil and Marintek – SIMO time-domain simulation of FPSO and tanker – Vessel and environmental data provided by Statoil – Validation by full scale motion measurements on FPSO and tanker
•
Recommendations : – FPSO should minimize surge motion, and tanker should avoid following a moving target on FPSO stern for positioning. – Operational coordination on FPSO and tanker for mean heading. – FPSO should minimize yaw motion. The modelling work was summarized in Journal of Offshore Mechanics and Arctic Engineering, August 2004, Vo.126, 235-242.
DP Operator Recovery Actions
FPSO Stern
FPSO Stern
(1)
(2)
FPSO Stern
(3)
Primary Secondary
Time Constraint for Successful Intervention To stop tanker within a short separation distance in drive-off scenario, recovery has to be initiated very early. Big tanker mass vs. Short distance Propulsion response & effects Separation Distance (m)
50
80
150
Time window for recovery (s)
37
53
81
How much time is needed by a human operator?
Modelling of Operator Intervention Drive-off First Abnormal Initiation Signal
Drive off Confirmation
Action Initiation
Information time: 0-Ta
DECISION
Decision time: Ta-Td
State Evaluation Task Formulation
Data
0
Ta
Execution time: Td-T1
INFORMATION
EXECUTION
Observation Detection
Muscle Command
Td
Action
T1
Time
Reference: Wickens’ model & Step-ladder model for human information processing stages
Investigate Operator Action Time • Incidents indicated that the action initiation time range from one to two minutes. • Expert judgment by Simulator instructor – Simulator training with experienced tanker DP operators – An average 29 s in INFORMATION stage and 56 s in DECISION & EXECUTION stages, and in total 85 s for action initiation Information Stage
Decision and Execution Stages
Time (sec)
0 - 10
10 - 20 20 - 30 30 - 50
0 - 20 20 - 30 30 - 60 60 - 90
No. out of 100 times Training
10
20
20
50
0
20
30
50
Probability
0.1
0.2
0.2
0.5
0.0
0.2
0.3
0.5
DP Operator Questionnaire Survey in 2002 • 10 captains and 7 DP officers • 16 Feedbacks of time estimation, 1093 tandem offloadings experience behind. • Questionnaire formulated based on human action model Reasonable Time Needed To Initiate Recovery Action 150
Time (sec)
120 Execution Tim e
90
Decision Tim e Inform ation Tim e
60 30 0 1
2
3
4
5
6
7
8
9
Feedback ID
10 11 12 13 14 15 16
Simulator Observation of DP Operator Emergency Intervention (2003 – 2006) • DP operator reaction time to drive-off in simulator • 66 records in 2003-2006. • Human action vs. probability curve fitted, representing best available knowledge. DP Operator Reaction Time in Drive-off Scenario Probability of Reaction within Time
(2003-2006, 66 simulator observations) 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0
20
40
60
80
100
Time (s)
120
140
160
180
200
Direct Offloading Development (2007 – 2013) • Shuttle tanker offloading operations from fixed or geostationary floating offshore installations in the North Sea. • Offloading from Kristin Platform, 2007, 12-days • Offloading from geostationary Sevan FPSOs in the Hummingbird/Chestnut fields in the North Sea • Offloading from Sevan FPSO in the Goliat field in the Barents Sea (2015)
What’s new in direct offloading?
• Weathervane with inherent safe heading philosophy – Heading pivot point – Min. 150 m no entry zone
• No hawser between installation and shuttle tanker
Wind
150 m
• Separation distance 250 m
Heading pivot point
250 m
Fixed / Geostationary Offshore Installations
hose
Illustration drawing: The size and distance are not to scale.
Risks in Offloading: Collision and Oil Spill Zone 2
Wind
• Towards installation = Collision risk
150 m
• DP shuttle tanker position loss!
Heading pivot point
250 m
Fixed / Geostationary Offshore Installations
• Away from installation = Hose rupture and oil spill risk
Zone 1
hose
The 29th, 30th Int. Conference on Ocean, Offshore and Arctic Engineering (OMAE): - OMAE2010-21185 for collision risk analysis - OMAE2011-50344 for oil spill risk analysis
Zone 1 Zone 2
Illustration drawing: The size and distance are not to scale.
Drive-off with Potential for Collision
Drive-off No Collision
Wind / Wave
Installation
Drive-off into Collision Zone
Illustration drawing: The size and distance are not to scale.
Inherent Operational Safety Barriers
• 250 m distance = Time window for recovery action by tanker DP operator is 3 minutes vs. 1 minute.
DP Operator Reaction Time in Drive-off Scenario (2003-2006, 66 simulator observations) 1.0
Probability of Reaction within Time
• Shuttle tanker positioning strategy = less than 1 hour drive-off collision risk exposure time vs. 20 hours.
0.8 0.6 0.4 0.2 0.0 0
20
40
60
80
100 Time (s)
120
140
160
180
200
Collision Frequency: Direct vs. Tandem Offloading • The collision frequency in the direct offloading (6.43·10-5 per year) is much lower than the equivalent tandem offloading from FPSO (1.62·10-2 per year). 1.0E-01 Direct of f loading f rom geostationary FPSO Tandem of f loading f rom ship-shaped FPSO 1.0E-02
3.04E-03
Frequency per Year
Premises: Offloading once per week, 24 hours operational time.
1.17E-02
1.21E-03
1.0E-03
1.62E-04 1.0E-04
4.66E-05
4.05E-05 1.20E-05
1.0E-05
4.82E-06 negligible
1.0E-06
0-22 MJ
22-49 MJ
49-87 MJ
87-136 MJ
negligible
>136 MJ
Collision Impact Energy (relevant to FPSO damage)
Summary • Collision frequency model with “human element” into equation. – Study of failure prone situation of drive-off: relative motions – Investigate human action time under emergency situations
• Simulator observation of human action and timing • Direct offloading from geostationary FPSO/installation: from concept to real operation.
References • • • • • • •
•
Chen, H. and Moan, T.: "Human Intervention of Tanker Drive-off in Tandem Offloading Operation", The 2nd International Conference on Human Factors in Ship Design and Operation, RINA HQ, London, UK, 2002 Chen, H. and Moan, T.: "FPSO - Shuttle Tanker Collision Risk Reduction", OMAE2003-37108, Proceedings of the 22nd OMAE Conference, Cancun, Mexico, 2003 Chen, H. and Moan, T.: "Probabilistic Modeling and Evaluation of Collision between Shuttle Tanker and FPSO in Tandem Offloading", Journal of Reliability Engineering and System Safety, Vol. 84 (2004) 169-186 Chen, H., Moan, T., Haver, S., and Larsen, K.: "Prediction of Relative Motions and Probability of Contact between FPSO and Shuttle Tanker in Tandem Offloading Operation", Journal of Offshore Mechanics and Arctic Engineering, Vol. 126, 235-242, August 2004. Chen, H., Moan, T. and Vinnem, J. E.: “Safety of shuttle tanker offshore loading operations with emphasis on the human barrier”, European Safety & Reliability Conference (ESREL2007), 24-27 June 2007. Chen, H. and Moan, T.: “Human element in the safety modeling of offshore marine operations”, Proc. of the Marine Operations Specialty Symposium 2008, Singapore, 5-7 March 2008. Chen, H., Lerstad, A. and Moan, T.: “Probabilistic Evaluation of Collision between DP Shuttle Tanker and Geostationary FPSO in Direct Offloading”, OMAE2010-21185, Proc. of the ASME 29th International Conference on Ocean, Offshore and Arctic Engineering, Shanghai, China, June 6-11, 2010. Chen, H., Moan, T., Breivik K., Lerstad, A.: “Analysis of Oil Spill Risk in DP Shuttle Tanker Direct Offloading Operations”, OMAE2011-50344, Proc. of the ASME 30th International Conference on Ocean, Offshore and Arctic Engineering, Rotterdam, The Netherlands, June 19-24, 2011.
For more information, please contact: Dr. Haibo Chen Managing Director Scandpower Asia Operations T E W w
+86 138-0132-0200
[email protected] www.scandpower.com www.lr.org
