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