OIL & GAS Risk Based Inspection (RBI) Planning for Hull Structures of Mobile Offshore ... · 2020....
Transcript of OIL & GAS Risk Based Inspection (RBI) Planning for Hull Structures of Mobile Offshore ... · 2020....
-
SAFER, SMARTER, GREENER
OIL & GAS
Risk Based Inspection (RBI) Planning for Hull Structures of Mobile Offshore Units (MOUs)
1
Madhu Parlapalli, David Baxter
DNV GL, Aberdeen
-
Why do we inspect ?
Does the structure
have adequate
structural integrity ?
Degradation
(fatigue cracking,
corrosion…)
Accidental loads
or damage
Environmental
overload
Fabrication defects or
installation damage
Design errors
2
-
When do we inspect MOU Hull Structures?
Time based approach
• Fixed interval
• Key areas
• CLASS or other prescriptive regulatory
requirements
• Renewal survey every 5 years, or every
2.5 years beyond fatigue design life
Risk based approach
• Variable intervals (probability and
consequence of failure)
• Qualitative or Quantitative approaches
• DNVGL-RP-C210: Fatigue cracks
• DNVGL-OTG-17: Coating and Corrosion
3
-
What do we inspect ?
4
-
What do we inspect ?
5
-
Where do we inspect ?
6
-
How do we inspect ?
7
-
What else can be done ?
8
http://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=0CAcQjRxqFQoTCNWywp-JsMgCFUF3PgodMoMFeA&url=http://www.twi-global.com/technical-knowledge/job-knowledge/fatigue-testing-part-3-080/&psig=AFQjCNFpLX8kM31B17NR4lgULZtKYgmlVQ&ust=1444297361280035
-
Background : DNVGL-RP-C210
• Specific guidance for:
• Jackets
• Semisubmersibles
• FPSOs
• Focuses on fatigue cracking
• SN-Data
• Fracture Mechanics
• Assessment of Probability of
Fatigue Failure
• Target Reliability
• Probability of Detection
• Determining requirements for NDT
at fatigue hotspots
• Inspection Planning
• Examples
9
-
Background: DNVGL-OTG-17
10
• Guidelines for how to apply RBI Methodology for
MOUS classed in DNVGL worldwide for the following:
• Semi-submersible
• Jack-up
• FPSO
• Deep Draft Floater
• Buoy
• TLP
• The developed in-service inspection plan (IIP) shall
be approved by DNVGL CLASS before the plan is
implemented as part of the class in-service inspection
plan.
• Damage Degradation Mechanisms:
• Fatigue (Section 3)
• Coating and Corrosion (Section 4)
-
Typical methodology (overview) for Fatigue RBI
• Data Collection:
• Drawings, inspection
history, operating history,
etc.
• Perform structural analysis (if
not already available)
• Perform quantitative RBI at
selected hotspots
(low fatigue life, history of
cracking or high
consequence)
• NDT intervals determined by
RBI
• Visual inspection of other
details
Data Gathering
Structural Analysis
Consequence
of failure
Fatigue Life
Crack History
Probabilistic
Analysis
GVI CVI NDT
low
yes
high
low
medium
11
-
Qualitative Screening - Overview
12
• Develop structural hierarchy of inspectable elements
• Identify relevant failure modes and damage mechanisms
• Evaluate Probability ranking for relevant mechanisms
• linked to inspection history, results of fatigue analysis, type of coating, fluid
corrosivity etc.
• Evaluate Consequence ranking for safety, environment or business impact
• Evaluate Risk level using agreed risk matrix
• Develop inspection plan for different inspection activities based on risk
• Typical Inspection intervals are as shown below
-
Quantitative Analysis
13
• For high risk items or where there is a desire to optimize inspection
frequencies beyond those given by screening assessment
• Quantitative assessment of fatigue in accordance with DNVGL-RP-C210
• Quantitative assessment of corrosion in accordance with DNVGL coating
degradation model and DNVGL-OTG-17
-
14
Quantitative analysis of fatigue
• Fracture mechanics Model
• Probabilistic approach: probability of
failure as a function of time, taking
account of uncertainties in loading and
material properties, as well as inspection
history
• Calibrated against S-N life to ensure
consistency
• Optimum inspection frequency
to keep POF below target level
(target level depends on
consequence of failure)
• Structural improvements, toe-
grinding etc., where
appropriate
-
Probabilistic Crack Growth Analysis
Optimal inspection plans for details with different fatigue lives (time between
NDT inspections)
15
-
Quantitative analysis for Corrosion / Coating degradation
16
POF associated with coating degradation modelled using parameters such as:
• Target life (depending on coating specification)
• Surface preparation factor (cleanliness)
• Environmental exposure factor (type and temperature of tank contents)
• Osmosis factor (salts on surface or solvents in paint)
• Cathodic disbondment (type of coating and test results)
• Other factors: such as flexibility of coating, resistance to shrinkage etc.
Inspection frequency:
• Time at which expected coating quality falls
below target
• Time at which the probability of unacceptable
coating quality exceeds target
• Cost optimal criteria depending on cost of
re-coating and/or patch repairs
= input parameter = calculated value
Coating Specification
Target Life (years)fractile
level %± # Std
Tl ife Uncertainty Value
Type III 15 3 90 % 1.645
Installation Year = 2004
Coating Year = 2004
Planning Year = 2004
Definition of Target life Mean value StDTlife
15.0 1.82
Repair Strategy Q
Limit for coating repair 0.98 LRepair
Limit for re-coating 0.95 LRe-coat
Definition of Target life 0.9 LLife
Reducion factors for coating life Mean CoV
Surface preparation factor FCoating 1 0.05
Steel flexibility factor FFlexibility 1 0
Environmental factor FEnviroment 1 0.1
Osmosis factor FOsmosis 1 0.05
Cathodic disbonding factor FCP 1 0.05
Coating generic type factor FGeneric type 1 0.1
Total Life Reduction factor FTotal 1.00
Repaired area coating breakdown
factorFRepair 0.800 0.1
Total Tank Area (m3) 100 000
IMR Costs (relative or absolute)
Expected
cost Variable
Inspection cost 50 000 C Inspection N insp : Number of inspections
Repair cost pr. area unit (m3) 40 C Repair P Repair : Probability of performing repair during inspection
Re-coating cost pr. area unit (m3) 40 C Re-Coating P Re-Coating : Probability of performing re-coating during inspection
Discount rate (%) 6 r
INPUT parameters for Coating Specification & Re-coating Requirements
Coating System
Limit State Models :
- Q(t) > LRepair => No Repair or Re-coating
- LRepair > Q(t) > LRe-coat => Coat Repair
- Q(t) < LRe-coat => Re-coating
Q : Coating Quality measured as 1 - D
D : Ratio of coating degradation
-----------------------------------------------------------------------------
Uncertainties:
Tl ife : Standard deviation of coating life is specified by given
fractile level i.e. 90% fractile level => ± 1.645 StD
CoV : All reduction factors for coating life can be modelled
as uncertain with given CoV
Type III
Coating Degradation
80 %
82 %
84 %
86 %
88 %
90 %
92 %
94 %
96 %
98 %
100 %
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16Years
Q =
1 -
Co
ati
ng
bra
ked
ow
n (
%)
.Bets estimate Coating quality
L_repair
L_re-coat
L_Life
=
−−
+
++=
insp
i
N
it
iCoatingReiCoatingReiRepairiRepairiInspection
Totalr
tCtPtCtPtCC
1 )1(
)()()()()(
-
Methodology – Corrosion / Coating degradation
17
• Update of Models based on Inspection Data
• Update of inspection plans for future condition
• Inspection updating procedure is similar to that used for fatigue cracks
• Classification of the condition of the coating in the ballast tanks based
on IACS:
• GOOD: Condition with only minor spot rusting
• FAIR: Condition with local breakdown at the edges of stiffeners
and weld connections and/or light rusting over 20% or more of
areas under consideration, but less than as defined for POOR
condition.
• POOR: Condition with general breakdown of coating over 20% or
more of areas or hard scale at 10% or more of areas under
consideration
-
Methodology – Corrosion / Coating degradation
18
• Acceptance criteria for inspection planning:
• Expected coating condition falling below threshold, or probability of
certain condition exceeding the target
• Alternatively, plans can be developed to minimize cost (while
maintaining adequate safety)
• cost of inspection versus cost of patch repairs, re-coating, etc.
-
Case Study – FPSO
Phase 1
• Global structural model and analysis
• Global screening
• Refined fatigue analysis at selected
hotspots
Phase 2
• Detailed fatigue analysis of additional
hotspots
• Quantitative RBI analysis
• Recommendations
19
-
Case Study – FPSO
• Hotspots models
20
-
Case Study – FPSO
• Hotspot Model C – Termination of Stiffeners - Frame #A - #B Starboard
21
-
Case Study – FPSO 1
• Hotspot Model C – Repaired
22
-
Case Study – FPSO 1
• RBI Analysis: Fatigue Degradation Mechanism
• Generic inspection plans
23
-
Case Study – FPSO 1
• RBI Analysis: Fatigue Degradation Mechanism
• Detailed inspection plans produced for each hotspot area (e.g. C5 to
C8)
• Repair from one side in 2004 moves C1 hotspot onto outside
• EC unable to detect cracks on outside. Crack growth analysis (of
through wall crack) demonstrates 6-monthly GVI (leak detection) is
sufficient in this instance.Tank Hotspot Side of
plateFatigue life (Original)
Fatigue life (Repaired)
Inspection Method
Insp. 1 Insp. 2 Notes
A C1 Internal 8 32 GVI* Crack / repair 2004Toe grinding 2006
External 4137 20 GVI*
C5 Internal 8 12 EC 2010 -
C7 Internal 11 11 EC 2010 -
C8 Internal 2 2 EC 2010 2012
24
-
Case Study – FPSO 1
• RBI Analysis: Coating Degradation Mechanism
• Total life correction factor: product of the critical coating factors, i.e.
surface preparation, environmental stress, Osmosis, Cathodic disbonding,
Coating type and maintenance factor
• Repair of coating and repair of the corrosion damages in this tank is
advised if the unit shall be used for the next 5 to 10 years
25
Figure: Simulation of Coating Degradation for
COT A
Assumptions:
• Inspection in 2018, coating
breakdown as FAIR, between 5% -
20%
-
Summary
• Background info for RBI
• Brief Methodology
• Fatigue (DNVGL-RP-C210)
• Corrosion (DNVGL-OTG-17)
• Sample Case Study
26
-
SAFER, SMARTER, GREENER
www.dnvgl.com
Thank you for your attention…
27