Research Project for Ph.D.

Thesis

Title:

Reliability and Risk Analysis of Station Keeping System of

Mobile Offshore Drilling Units (MODUs)

Ph.D. Candidate: Zhen Gao

Supervisor: Professor Torgeir Moan

Norwegian University of Science and Technology (NTNU)

Faculty of Engineering and Technology
Department of Marine Technology

Trondheim, Norway
June, 2004
1. Background

Mobile offshore drilling units (MODUs) are vessels capable of engaging in drilling
operations for the exploration for or exploitation of resources beneath the sea-bed
such as liquid or gaseous hydrocarbons, sulphur or salt (1989 MODU Code [1]). The
MODU considered in this project is a semi-submersible, consisting of the main vessel,
its station keeping system, risers and sub-sea equipments. During the drilling work, a
MODU should be able to be restrained within pre-defined limits by its station keeping
system. It’s extremely crucial to the safety of operation. Failures in station keeping
system will make the vessel apart from the desired position which might lead to
collision with another installation in close proximity, damage to risers, BOP (Blowout
Preventer), sub-sea equipments, and more seriously sub-sea blowout. The final
consequence will be direct loss of property and lives, significant downtime and
pollution in environment.

Station keeping system is designed to maintain the position and heading of a floating
structure within prescribed limits under maximum design environmental condition
and acceptable condition during operation. The main function of such system is to
counteract the steady and low-frequency (LF) forces due to environmental loads, such
as current, wind and second-order mean and slowly varying wave loads [2]. The
motions concerned in a station keeping modeling for a floater are horizontal motions,
i.e. surge, sway and yaw.

Mooring system has been widely used for permanently moored floaters. Usually,
there’re two mooring configurations, spread mooring and single point mooring.
Mooring system consists of fairleads, mooring lines, connections and anchors. The
component of a mooring line is either chain, wire rope, synthetic rope (e.g. polyester)
or a combination of them. Mooring line strength and anchor capacity are the main
concerns in mooring analysis. Moorings should also be designed to sustain the sudden
failure of any single mooring line without causing excessive load to be developed in
remaining lines and without causing conditions in the remaining lines which would
endanger the security of anchors.

Structural reliability analysis can be utilized to evaluate the safety level of mooring
system by a strength-load effect probabilistic model. The general methods have been
introduced and illustrated in detail in many famous books, such as Melchers [3], 1999
and Madsen [4], 1986. Simplified methods (FORM, SORM) and simulation methods
(Monte Carlo simulation, importance sampling) are frequently used in practice. A lot
of mooring reliability studies have been carried out. Hovde [5] (1995) formulated the
component and system failure probability of mooring lines under ultimate loads and
fatigue loads in a frequency domain. Haver et al [6] (1999) carried out a system

2
reliability study on the mooring system of a semi-submersible using different
probabilistic models. However, most of these studies utilized the simplified methods
in frequency domain and time domain methods were only adopted for transient
response analysis. This is because the simulation methods are very time-consuming
for complex nonlinear systems. But, the number of simulations can be extremely
reduced by auxiliary domain method for short-term analysis, which is planned to be
done in this project. Another stubborn problem appeared in mooring analysis is the
combination effect of LF and WF vessel motions. Stansberg [7] (1991) studied the
phenomena of LF vessel motion and gave out formulae to evaluate the maxima and
extreme responses including the damping effect. However, this difficulty still remains,
especially in simplified frequency domain methods.

An alternative positioning technology is use of dynamic positioning (DP). DP system

is a system which automatically controls a vessel’s position and heading exclusively
by means of thrusters. A typical DP system consists of control console, reference
systems, such as position, heading and environment references, and power system
including power generation, distribution and consumption (IMCA M 103 [8], 1999).
Compared with the passive mooring system, the DP system is more active, which
involves automatic control and human-machine interaction. Stability of the control
system and reliability of human actions will be key factors for the operation safety.
Redundancy should therefore be applied in a DP vessel to maintain an acceptable
safety level and clear operational procedure is needed. Records of vessel position and
heading, reference status and available power are needed in such system.

The design of control system is very important and directly influences the
performance of DP system under environmental conditions. For a full DP vessel, the
coupled dynamic equations of LF surge, sway and yaw motions can be set up as a
base for appropriate control schemes (Balchen et al [9], 1980 and Sørensen et al [10],
1996). The thruster component reliability can be determined by a capacity and
demand model using classical reliability theory. Failure probability of a reference
system depends on the principle it’s based on. Power generation and distribution
system is a fundamental but complicated system consisting of a great number of
components. Incidents and accidents data of DP vessels have been recorded and
investigated by many organizations, e.g. IMCA, HSE (Health and Safety Executive),
and are useful to a proper estimation of failure rate of complex technical systems. But
such data for DP drilling vessels are very scarce. Risk assessment should be carried
out where the human error is included in the modeling. Direct human-machine
interaction will happen in an emergency when some technical failures occur and lead
to loss of position of the vessel. Whether the DP operator is able to react in a correct
way within a limited time window during the latter scenario to save the situation from

3
worsening is a crucial question to the safety. Consequence analysis is then performed
considering different failure modes and if necessary risk reduction measures (e.g.
ESD (Emergency ShutDown)) will be taken. Chen [11][12] (2002) studied the safety
against collision for a DP vessel in tandem offloading operation with a two-stage
probabilistic method, i.e. initiation of loss of position and recovery action. Three
failure modes, namely drive-off, drift-off and force-off have been identified.
Moreover, a DP operator reaction time model is proposed to analyze the robustness of
recovery and quantified according to expert judgement and a questionnaire survey,
which is also verified latter by SMS’s (Ship Manoeuvring Simulator center) training
records of DP operators in drive-off scenarios. The safety assessment of DP drilling
operation taking different failure modes into account will be carried out in the project
based on the above methodology.

Thruster assisted mooring is a combined system, which is adopted to reduce the

tension in mooring line by thruster force instead of pure mooring system or full DP
system, and potentially a good choice in some situation due to the flexibility of
thruster in changing heading of the vessel. However, in such system, the mutual effect
between mooring lines and thruster system should be carefully considered in order to
reduce line tension by thrusters appropriately during the normal drilling work,
especially when active thruster system (DP system) is involved [13]. And also the effect
of sudden failure of single mooring line or DP system fault and the corresponding
consequences should be investigated by using FMEA. Recently, a Statoil drilling plan
in the Barents Sea is considering to use the thruster assisted mooring system. The
combined system needs the coupled hydrodynamic loads calculation and applied
control scheme which also considers the responses from mooring lines and risers.
Another issue is that continuous monitoring shall be provided of important parameters,
such as position, heading, mooring line tension and available electrical power. During
the adjustment of position and heading of the vessel, the coupled effects should also
be integrated into the operational procedure in such a way that the manual or
automatic adjustment will not enlarge line tension and riser tension.

2. Objectives and Scope

The main objective of this research is to develop a new method for reliability and risk
analysis of different station keeping systems of MODUs, including mooring system,
DP system and thruster assisted mooring system. The attention will be concentrated
on the station keeping system for offshore drilling operation. Although the safety
consideration is quite different from other offshore activity, for example diving
support, the general methodology which is expected from this work can be applied

4
similarly. The main tasks during the Ph.D. research work are listed in the following:

a) Investigate the key procedure in offshore drilling activities with a MODU (a

semi-submersible) in medium or deep water and model mathematically the main
concerned safety limitations, such as riser angle limit, operation requirement, etc.

b) Study and do computations of hydrodynamic loads acting on MODU using the

state-of-art programs or experimental data, combined with different control schemes
for a DP vessel.

c) Model and analyze a mooring line component and system reliability problem
formulated by typical limit states (overload, fatigue and wear deterioration)
considering different mooring line types, such as chain and fiber rope. A case
calculation of system failure probability will be done for a semi-submersible using
spread mooring. Auxiliary domain simulation method will be used in comparison with
the simplified FORM method. Updating of reliability may not be touched upon here.

d) Identify and set up different failure modes (drive-off, drift-off, and force-off) of a
DP vessel caused by reference errors, thruster failure, energy supply system failure
and human error. Possible experience database will be chosen and investigated and a
specified capacity-demand model for thruster failure analysis will be developed and
adopted in case study.

e) Model uncertainties in human actions in DP operations given specific failure mode

with a human reaction time model and possibly further develop it. The training
database will be chosen and analyzed to get more accurate results in quantification.

f) Case study a thruster assisted mooring system of a MODU with different failure
criteria mentioned above and determine the corresponding reliability and risk level.

3. Research Methods

The general reliability theory and risk methodology will be adopted in this research
work and some special considerations and simplifications suitable for mooring and
DP system will be proposed and adopted in analysis. The following research methods
are specified to fulfill the objectives c), d), e) and f).

1) Failure mode of mooring line system

For a mooring line, the ultimate and fatigue strength are of importance and

5
quasi-static method or dynamic method can be chosen to analyze the line tension. The
dynamic model is complex but more accurate, especially for the line tension due to
the wave-frequency (WF) vessel motions. Mooring line analysis and anchor force
evaluation will be given in this work for a mooring system with chains or fiber ropes.
Catenary mooring system (CMS) with chains provides mooring forces by its own
weight, which might be an obvious disadvantage of negative buoyancy and large
footprint in deep water, while taut mooring system (TMS) with ropes pre-tensioned is
more desirable for such condition, especially TMS of synthetic material. However,
careful considerations in design should be given to the anchors in TMS due to more
uplift load taken by the anchors than those in conventional CMS (Dahlberg [14], 2002).

Basically, the reliability analysis of a mooring system is a time-dependent problem

which means the failure probability should be given by a function of time. However,
for overload reliability analysis, a simplified method using extreme value estimation
of loads in a reference period, e.g. one year or the whole service life time, can be
applied and are commonly adopted in design codes with, for example annual failure
probability, as an interpretation of safety level. Typically, only a single sea state is
considered in such method for design purpose. Herein, the short-term (for example 3
hours) extreme line tension will be first evaluated considering the combined LF and
WF vessel motions and integrated in a long-term (foe example one year) analysis with
respect to different sea states for characterized by environmental parameters, such as
significant wave height, peak wave period, mean current and wind speeds, and
environment directions, and finally combined with time-independent random
variables in a nested reliability analysis (Wen et al [15], 1987). Possibly some rules for
load combination will be discussed and illustrated. Fatigue reliability analysis based
on similar formulation considering all environmental conditions will be also touched
upon and especially attention will be paid to short-term and long-term fatigue damage
of different mooring line types using S-N approach or fracture mechanics method.

2) Thruster failure analysis

Classical reliability method, especially a capacity-demand model, will be also applied
for thruster reliability analysis. Here, thrusters are main propellers, tunnel thrusters,
azimuth thrusters, combined with rudders. The capacity of thruster system instead of
mooring line strength is of much concern, but it quite depends on the environment
conditions, especially the wave condition and current direction. Loss of propulsion
force is observed where severe cavitation occurs due to the decreasing immersion
depth of propeller in waves and the interaction effect with current. Thrusters in a DP
vessel are automatically controlled against the external forces. As a consequence, the
control scheme should be involved in thruster system design together with the number
of thrusters, the layout and the thrust allocation.

6
3) Failure mode of DP drilling operation
Based on the redundancy level of equipments, a DP vessel is classified either with
Class 1, 2 or 3 according to IMO (International Maritime Organization) and some
classification societies. For example, a Class 3 DP vessel should reach the following
requirement: "Loss of position should not occur from any single failure including a
completely burnt fire sub division or flooded watertight compartment" (IMO MSC
Circular 645 [16], 1994). The safety of drilling operation with a DP vessel depends on
the equipment reliability and human reactions. Incidents are often initiated by
technical failures and developed into disastrous accidents with no saving actions
executed in time or with some performed in wrong ways. The DP operation safety can
be modeled through a two-stage method (Chen [17], 2004) characterized by resistance
to loss of position and robustness of recovery. Three failure modes of loss of position
have been identified, i.e. drive-off, drift-off and temporary large excursion from the
incident and accident investigation. Reference system error, thruster failure, failure in
control and power system, and human error are possible causes from which the
frequency of loss of position in different scenarios can been derived. Relevant
database will be utilized in the analysis.

4) Human reaction model

Whether an incident, even an accident will happen given loss of position depends on
the detection of such situation and proper recovery action initiated by the DP operator.
The human reaction time model can be used in analyzing the robustness of recovery.
A series of human reactions are formulated in a Detection-Decision-Execution way
and the total required time window for a successful recovery can be derived by adding
the time spans spent in these stages. This model and possible development in detail
will be used and expected in the current study.

5) Risk assessment
Risk analysis is defined in the NORSOK (Norwegian offshore standardization
organization) standard as an analysis which includes a systematic identification and
description of risk to personnel, environment and assets (NTS (Norwegian
Technology Standards institution) [18], 1998). Compared with the classical structural
reliability, risk analysis has a broad definition with focus on the cause and
consequence of a system failure. However, risk assessment meets difficulty in
quantification to represent the safety level appropriately. Risk quantification is often
characterized by a mixture of the classical statistical approach and the Bayesian
(subjective) approach. And the qualitative risk assessment is frequently adopted due to
the simplicity. A typical risk assessment involves identification of initiating event,
cause analysis and consequence analysis (Vinnem [19], 1999). The estimation of risk

7
should take into account both the occurrence frequency of a hazard and the potential
consequence resulting from it. Probably, some risk methodologies (e.g. FMEA and
ETA) will be adopted in DP safety assessment. An important issue in identifying
possible causes for hazards and modeling failure frequency of complex systems or
unclear phenomena is the investigation and analysis of incident and accident data.
IMCA and HSE publish such data every year together with a brief description and
analysis. Such data will be investigated.

6) Integrated approach
For a thruster assisted mooring system, the integrated approach is important. The
state-of-art modeling of the hydrodynamic loads relevant to station keeping
performance, with focus on LF and WF load combination and statistical characteristic
of extreme responses, and control systems for positioning purpose with linear or
nonlinear schemes will be applied in the formulation of reliability and risk problem
with a cooperation with other members in CeSOS (Center for Ships and Ocean
Structures) and representatives from industry. And also the state-of-art software will
be chosen for scenario simulation and numerical calculation.

4. Expected Results

The final goal of this research work is to develop a method for reliability analysis and
risk assessment of station keeping system considering the safety requirements related
to the offshore drilling activities using a MODU.

Mainly, modeling and analysis of a mooring system with different line types
considering overload and fatigue reliability will be performed, and the identification
and formulation of DP operation failure modes, i.e. loss of position with different
scenarios initiated by reference error, thruster and power system failure, etc., and the
modeling and analysis of human recovery actions will be carried out.

Identification of failure modes of mooring line system and DP system during drilling
operation will point out the potential areas where risk reduction measures might be
taken to improve the safety level. The methodology can also be a foundation for
deciding whether a risk reduction measure is reasonably practicable based on a
cost-benefit analysis, which gives assessment of the cost of implementing a design
change or modification and the likely reduction in fatalities that this would be
expected to give.

8
5. Work Plan

This research project has been started on August 8, 2003 and will be terminated on
August 31, 2006 with a period of three years. The planned main tasks are listed in the
following schedule:

Main tasks 2003 2004 2005 2006

Course work
1
(structural reliability, hydrodynamics, etc.)
Literature studies
2
(risk analysis, control theory, DP operation, etc.)
Mooring system reliability modeling and
3
analysis, and Case study
Limitation consideration and modeling for
4
drilling operations with a MODU

5 Thruster capacity and reliability analysis

Possible further development of human reaction

6
time model
Failure modes identification and modeling of DP
7
operation safety, and Case study
Thruster assisted mooring system analysis, and
8
Case study

9 Thesis writing

10 Thesis defence

Professor Torgeir Moan is the main supervisor during this research work. A close
cooperation with Haibo Chen at Scandpower Risk Management AS and other Ph.D.
students at CeSOS is planned.

6. References

[1] IMO, Code for the Construction and Equipment of Mobile Offshore Drilling
Units, 1989.
[2] Pinkster, J. A., Low-frequency phenomena associated with vessels moored at
sea, Society of Petroleum Engineers Journal, Vol. 4, pp. 87-94, December 1975.
[3] Melchers, R. E., Structural Reliability Analysis and Prediction, Ellis
Horwood Ltd., 1999.
[4] Madsen, H. O., Krenk, S. and Lind, N., Methods of Structural Safety,

9
Prentice-Hall, 1986.
[5] Hovde, G. O., Fatigue and Overload Reliability of Offshore Structural
Systems, Considering the Effect of Inspection and Repair, Dr.Ing. Thesis, Department
of Marine Technology, NTNU, May 1995.
[6] Haver, S., Meling, T. S. and Larsen, K., System Failure Analysis of
Veslefrikk B Mooring, Statoil Report No. 99s97*5292, Stavanger, 1999.
[7] Stansberg, C. T., A simple method for estimation of extreme values of
non-Gaussian slow-drift responses, Proceedings of the First International Offshore
and Polar Engineering Conference, Vol. 3, pp. 442-451, Edinburgh, 1991.
[8] IMCA, Guidelines for the Design and Operation of Dynamically Positioned
Vessels, IMCA M 103, February 1999.
[9] Balchen, J. G., Jenssen, N. A., Mathisen, E. and Sælid, S., A dynamic
positioning system based on Kalman filtering and optimal control, Modeling,
Identification and Control, Vol. 1, No. 3, pp. 135-163, 1980.
[10] Sørensen, A. J., Sagatun, S. I. and Fossen, T. I., Design of a dynamic
positioning system using model-based control, Control Engineering Practice, Vol. 4,
No. 3, pp. 359-368, 1996.
[11] Chen, H. and Moan, T., Collision risk analysis of FPSO-tanker offloading
operation, OMAE2002-28103, The 21st International Conference on Offshore
Mechanics and Artic Engineering, Oslo, 2002.
[12] Chen, H., Probabilistic Evaluation of FPSO-Tanker Collision in Tandem
Offloading Operation, Dr.Ing. Thesis, Department of Marine Technology, NTNU,
January 2003.
[13] HSE, Station Keeping, Offshore Technology Report 2001/050, 2002.
[14] Dahlberg, R. and Mathisen, J., Consistent design codes for anchors and
mooring lines, OMAE2002-28424, The 21st International Conference on Offshore
Mechanics and Artic Engineering, Oslo, 2002.
[15] Wen, Y. K. and Chen, H. C., On fast integration for time variant structural
reliability, Probabilistic Engineering Mechanics, Vol. 2, No. 3, pp. 156-162, 1987.
[16] IMO, Guidelines for Vessels with Dynamic Positioning Systems, MSC
Circular 645, 1994.
[17] Chen, H., Moan, T. and Verhoeven, H., Safety of dynamic positioning
operation on mobile offshore drilling units, The 23rd International Conference on
Offshore Mechanics and Artic Engineering, Vancouver, 2004.
[18] NTS, Risk And Emergency Preparedness Analysis, NORSOK Standard
Z-013, Rev. 1, March 1998.
[19] Vinnem, J. E., Offshore Risk Assessment – Principles, Modeling and
Applications of QRA Studies, Kluwer Academic Publishers, 1999.

10