Highlights from the HOIS JIP: NDT methods for CUI ... · Unmanned aerial vehicles (UAV) for external RVI recent and growing area: •Information on the external condition of piping,

© ESR Technology | 04/12/2018 | PAGE 1

Highlights from the HOIS JIP: NDT methods for CUI inspection and use of UAV (drones) for remoteexternal visual inspection

Joint Offshore Energy and NDT Technical Group Meeting

21 November 2018:

Non‐Destructive Testing and Inspection in the Offshore Energy IndustrySteve Burch (ESR Technology)

Courtesy Jim McNab, Oceaneering


Introduction to HOIS Joint Industry Project (JIP)

HOIS is a major well established (>30 years) JIP on good

practice for NDT/NDE in the oil & gas industry

40 Members comprise:

• Oil and Gas producers - operators

• NDT service companies

• NDT equipment vendors

• NDT notified bodies

• A regulatory authority (UK HSE)

• The Oil & Gas Technology Centre (OGTC)

Managed by ESR Technology

Global representation: Americas, UK, Europe, Middle East &

Far East

Follow us on LinkedIn!


Key HOIS aims & activities

Improve effectiveness of in-service inspection for the oil & gas industry and promotes

good practice

• More reliable in-service inspection – better detection of in-service degradation

• More accurate measurements of remaining wall thickness

• Provide asset owners with improved information on asset condition

• Reduce leaks and near misses

This is done by:

• Developing good practice documents for specific inspection applications – recommended

practices and contribution to international standards which feed into company inspection

procedures

• Assessment of the performance of current and developing inspection techniques by means

of rigorously controlled blind trials

• Industry forum for latest developments – what’s new and what works

• Some development of novel inspection techniques for specialised applications


Latest HOIS Guidance documents & RPs

Title Date Publicly available?

HOIS Guidance on Image Quality for UAV/UAS based external remote visual inspection in the oil & gas industry (Open version)

July 2018 Yes

HOIS Human Factors guidance for manual UT for internal corrosion and wall-thickness loss

May 2018 No

HOIS Human Factors guidance for visual inspection for external corrosion May 2018 No

HOIS guidance for effective pipework inspection April 2018 No

HOIS recommended practice for precision thickness measurements for corrosion monitoring

Nov 2017 No

HOIS guidance on IVI reporting protocol and proforma for pressure vessels June 2017 No

HOIS general guidance on mitigation of human factors on offshore and onshore inspections

April 2017 No

HOIS Guidance for in-situ inspection of corrosion under insulation (CUI) Nov 2016 No

HOIS Guidance on inspection of corrosion under pipe supports (CUPS) Sept 2016 No

Recommended Practice for the in-service inspection of wall loss in pipes by digital radiography

April 2015 Yes

Publicly available documents can be downloaded from www.hoispublications.com

http://www.hoispublications.com/


Current & recent HOIS projects

• Inspection of pipe supports (CUPS)

• CUI & external corrosion inspection

• Trunnion inspection

• Non intrusive inspection of internally CRA clad vessels

• Updating DNV GL RP G103 Non Intrusive inspection

• Inspection of composite repairs

• Guidelines for UAV/UAS based external remote visual inspection

(RVI)

• Guidance for effective inspection of pipework

• Precision thickness measurements for more reliable corrosion

monitoring

• Subsea inspection trials

• Human factors aspects of NDT in offshore oil & gas


ISO 20769:1 & 2, 2018 – Stop press!

Very recent publication of two significant standards for in-service radiography:

ISO 20769: Non-destructive testing -- Radiographic inspection of corrosion and deposits in pipes by X- and gamma rays

-- Part 1: Tangential radiographic inspection

-- Part 2: Double wall radiographic inspection

Based on major HOIS in-service radiography project

• Trials started in 2006

• Initial draft of first European standard 2010

• Publication of EN 16407 in 2014

• Major revision to incorporate new HOIS findings on localised internal and external wall loss.

• Publications of EN 20769 November 2018


HOIS highlights

Illustrate work of HOIS JIP by highlighting work on two projects covering different aspects of good practice in NDT for oil and gas.

1. Major project on NDT for inspection of external corrosion both through insulation (CUI) and on uninsulated pipes (Scabs)• Co-funding by the OGTC to expand and accelerate the project

• Blind evaluation trials currently in progress

• Final objectives to refine existing HOIS guidance on CUI inspection and generate a new guidance document

2. Project to provide guidance on UAV (drones) image quality for remote external visual inspection• Guidance document produced and publicly available

• Project now complete


HOIS CUI & scab inspection project


CUI – costs to industry

Corrosion costs UK £4 billion per annum (source

OGTC)!

CUI has caused ruptures and fires leading to damage

costing 10s of millions of pounds on single assets

(OGTC)

Estimated that 60% of piping failures due to CUI (OGA

source)

CUI is unpredictable “finds us before we find it”

For one operator alone, UK annual maintenance costs

of at least £10 million (OGTC)


NDT for CUI

Main current NDT methods for CUI:

• Guided wave testing (GWT)

• Radiography (real time, digital detector arrays etc)

• Pulsed eddy current (PEC)

Existing NDT methods can be effective in certain limited situations but

• Lack of independently validated information on detection & sizing performance

• All NDT methods can miss areas of CUI

No “silver bullet” for NDT of CUI

Economic issues:

• Cost compared to stripping & replacing insulation

• NDT costs can be additional to those of existing maintenance programmes

• If CUI detected, then need to strip off old insulation and replace anyway.

End result is that many companies do not use advanced NDT for CUI – rely on periodic removal of insulation, using an RBI approach


Advances in inspection and monitoring technologies

Recent advances in electromagnetic methods:

• Pulsed Eddy current – how do the newer systems compare with

the older ones?

• Pulsed Eddy current arrays – increased productivity

• Other low frequency eddy current type approaches (multiple

sensors)

Monitoring is a different approach

• GWT – higher sensitivity to change than detection during

inspection

• Other novel methods being developed can require installation

prior to (re-)installation of insulation


HOIS/OGTC trials

Aim is to assess detection and sizing performances of different NDT

methods on the same test samples

• Pulsed-Eddy current systems from 3 different equipment manufacturers

• Advanced EM systems from USA, Canada and New Zealand

• Radiography (real time and DDA)


HOIS Project structure – main tasks

Test pipe fabrication and assembly• Ex-service pipes

• Manufacture of pipes with realistic areas of wall loss (CNC machining)

Benchmarking available pipes• 3-D Laser scanning

• Development of an internal UT scanner

Trials of inspection methods for CUI inspection and external corrosion scabs• Independently hosted and assessed blind trials of NDT methods

HOIS guidance documents• Update CUI document

• Development of new document for uninsulated scabs


Test components for trials

10” manufactured pipes & assemblies with areas of wall loss introduced by CNC machining

• Accurately replicate morphology of ex-service “scabs” mapped previously using internal UT scanning

Areas of wall loss filled with compound containing “real” corrosion product

• Has magnetic properties

Manufacturing allows control of locations and severity of all areas of wall loss

• Can create different test samples with similar overall populations of “defects”

• Allow effects of other variables to be reliably assessed

Also ex-service small bore pipes


Schematic of CUI trial samples

10” sch 20

10” sch 20

3m

Rockwool; SS cladding

Rockwool; Galvanised cladding

Rockwool & Chicken wire; SS cladding

Heat tracing, Rockwool; SS cladding

10” sch 80

Rockwool; SS cladding

First set of trials 100mm insulation;

Second set at 50mm insulation

Hanger support

Hanger support


10" test assemblies & pipes – after welding & insulation

10” OD pipes currently with 100mm insulation manufactured using CNC machining to accurately replicate morphology of ex-service “scabs” previously scanned with internal UT probes


HOIS/OGTC CUI trials on 10" pipes

100mm trials complete.

Schedule in place for 50mm trials – last trial start of Jan 2019. 11 trials planned in total

Thanks to all trial participants for their excellent

level of commitment and

support

Trial participant Technology Insulation Thickness

Eddyfi/Bilfinger LYFT, LYFT array 100mm, 50mm

ETher NDE/Maxwell PECT 100mm, 50mm

TUV Sonopec 100mm, 50mm

Exxam Systems MFECT array 100mm, 50mm

Russell NDE Bracelet Probe array 100mm, 50mm

RRI GMR array 50mm


Examples of trials in progress


Results from trials

Aim will be to compare reported locations of corrosion with benchmark

• Assess probability of detection (POD) and false calls for each method

Also compare reported sizing information with benchmark (remaining thickness/wall loss)

Aim to investigate performance as a function of key trial variables including

• Insulation thickness

• Cladding type (Stainless steel or galvanised)

• Pipe wall thickness

• Severity of corrosion (i.e. max wall loss)

Other issues:

• Elbows – scanned or not?

• Coverage gaps caused by supports & hangers

• Chicken wire in insulation and heat tracing on pipe


HOIS/OGTC CUI guidance document

Results from trials to be fed into guidance documents

Existing HOIS guidance document on CUI inspection – confidential to HOIS

members

Results of current trials to inform substantial revision to this document

Will give guidance on applicability and performance of NDT methods for CUI

Limits of applicability of different NDT methods

Finalised version to be published due to public funding provided by the OGTC


CUI project summary

Aim of HOIS/OGTC project is to better quantify the inspection

performance of key methods for CUI inspection

• Based on test components containing realistic simulations of CUI

Investigate effects of key variables including pipe wall thickness,

insulation thickness, cladding type, geometry etc

Increased understanding of performance of key methods of NDT

for CUI

• Comparative results on same test components

Will inform major revision of guidance for inspection of CUI

Better assess role of NDT for CUI integrity management


HOIS UAV image quality project


Need for external remote visual inspection (RVI) guidelines

Visual inspection probably the oldest NDT method

Unmanned aerial vehicles (UAV) for external RVI recent and growing area:

• Information on the external condition of piping, vessels, tanks etc

• For engineering assessments of condition

Reports of substantial variations in the quality of the digital images that comprise the final deliverables

External RVI not covered in any detail by International Standards. Brief reference only in

• ASME V Article 9 which states that RVI needs to be demonstrated to an equivalent resolution to CVI/DVI

No evidence of usage of this standard in current UAV based RVI


Aims and scope of HOIS UAV project

Focus on quality of arising RVI imagery

Not addressing UAV pilot certification, nor site deployment/operational aspects

• Covered by others elsewhere

• e.g. Unmanned Aircraft Systems (UAS) Operations Management Standards and Guidelines, Issue 1, Oil & Gas UK January 2017.

External RVI only

Visible spectrum only (not IR)

Emphasis on stills not videos

Aim was to develop guidance for the oil & gas industry on minimum quality of UAV imagery needed for different applications


Challenges

Visual Inspection traditionally performed by a human

inspector

Usage of UAVs that generate digital images as the

end deliverable relatively new

Current visual inspection standards lack quantitative

measures of image quality

• Unlike other forms of NDT, such as radiography and

ultrasonics

Wide range of UAV applications

• Potentially different quality standards for each one

• “One size will not fit all”


HOIS members survey: application areas & quality measures

Obtained feedback to establish which UAV based RVI applications

have highest priority for guidance on image quality:

• Close/direct visual inspection (CVI/DVI)

• Assessment of coating condition to ISO 4628

• Flare-tip inspection

Quality measures emerging from survey were:

• Spatial resolution

• Noise level in images (signal to noise ratio)

Similar to digital radiography standards (e.g. ISO 17636:2)


Spatial resolution

An important measure of image quality in RVI

Needs to be specified on the object

NOT in terms of number of pixels in camera sensor

Best measured using line pairs per mm

Also use reciprocal mm/lp – similar to radiographic unsharpness

Key issues

• What resolution is needed in UAV based RVI

• How can this resolution be achieved?


Measure resolution – use a standard chart: USAF 1951

Different bar patterns arranged

in groups and elements within

each group

Referenced in BS EN

13927:2003

Is widely used for assessing

resolution of various optical

systems

Works well in practice


Resolution needed to match Close Visual Inspection (CVI)

ASME V Article 9, and BS EN 17637 state: • Surface being inspected should be no further than 600mm from the

human eye

• Inspector’s eyesight must meet a specified acuity level

Placed USAF 1951 test chart ~600mm from various human eyes that all passed the acuity test

Consensus was that resolution just discernible on the chart was ~3 line pairs/mm

Note: requirement in these standards to be able to discern a fine line 1/32 in (0.8mm) in width is NOT a measure of spatial resolution.

• Two closely separated lines would be needed for that


Close visual inspection

UAV does not provide the same information as traditional human based CVI, even if spatial resolution similar

UAVs have limitations including:• No surface preparation

• No tactile response

• Cannot measure dimensions directly nor view surface closer up with a magnifier

UAVs have some advantages:• Provide digital photographic record of inspected area

Unwise to claim that UAV imagery is fully equivalent to CVI – but at least ensure amount of detail visible is similar if spatial resolution is as good as eye at 600mm.


What is needed to achieve required resolutions using a UAV?

Potentially several variables affect resolution

• Distance from target

• Lens focal length

• Sensor size

• Number of pixels in sensor

• Quality of optics and camera

• Camera shake. Out of focus. Depth of field etc

Which are most important and how can they be related?


HOIS members USAF 1951 trials

Asked members to perform trials using airborne UAVs

to image USAF 1951 charts at different distances

Instructions to participants:

• Use imaging devices normally deployed for RVI

• Start at min distance normally used for RVI

• Provide images (& videos) for analysis


Results from members trials

Wide variations in

imaging systems used

Camera sensor sizes

• 6mm to 35mm

Lens focal lengths

• Wide angle to telephoto

Number of pixels in

sensors

• 2 Mp to >40 Mp

R = Resolution in mm/lp

(similar to radiographic unsharpness)


Analysis in terms of pixel size on the chart

Key variable is pixel

size, Δx, on the chart.

Two trendlines with

different gradients

• Unclear why

R ~ 3 Δx

R ~ 4 Δx

Δx


Interpretation of trial results

Key variable that determines resolution is pixel size on the

target.

Two relationships found:

• R ~ 3 Δx (most cameras)

• R ~ 4 Δx (a few other cameras and image types)

• Other cameras may give different results

Need to assess each new camera/lens combination to find/check

relationship between resolution and pixel size

• Using resolution chart imaged at different distances

• Determine constant c that relates resolution to pixel size:

R ~ c Δx


Distances needed for CVI (3 lp/mm)

Provided imaging system has “standard” relationship between pixel size and resolution R ~ 3 Δx

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

0 50 100 150 200 250 300 350 400 450 500 550 600

Dis

tan

ce f

rom

len

s to

tar

get

(m)

35mm equivalent Lens Focal Length (mm)

CVI imaging distances for different focal length lenses

N = 2000

N = 3000

N = 4000

N = 5000

N = 6000

N = 7000

N = 8000

N = 8000

N = 2000

N is number of pixels across

image

Results from lens theory


Verification of resolution

How to verify that the required resolution has

been achieved:

• Include a resolution chart in field of view – practical

issues?

OR

• Determine image scale (pixel size on target) and infer

resolution from that

• View image at 1:1 magnification and check for any lack

of sharpness due to motion blur, out of focus etc


Status of HOIS and other guidance

Issue 1 of guidance circulated to HOIS members in

June

Decision to make publicly available to increase uptake

Can be downloaded from www.hoispublications.com –

along with other freely available HOIS guidance

document

Raised awareness via article in NDT News

Collaboration with ASME

• Planning to revise ASME V article 9 to include usage of USAF

1951, following HOIS work

http://www.hoispublications.com/


Summary of HOIS UAV project

Recently increased usage of unmanned aerial vehicles (UAV) for external RVI:

• To obtain information on the external condition of piping, vessels, tanks and structures

• So that engineering assessments can be made of their condition

Lack of guidance has led to substantial variations in the quality of the digital images that comprise the end product of the inspection

Very wide range of image devices deployed on UAVs. No consensus over what is needed.

HOIS project has examined key image quality measures by means of members’ trials.

Developed guidance that seeks to achieve more standardisation in quality of the end deliverables (i.e. images)

Key parameter is spatial resolution achieved on the object


Conclusions

HOIS is a major JIP for good practice in NDT in the oil and gas industry

Established reputation for hosting of well managed, independently assessed

blind evaluation trials

• Precise benchmarking of test component

• Impartial analysis and reporting of results

Development of authoritative, trial based guidance for good practice in many

challenging NDT applications

Partnership with the OGTC where we align with their two key themes:

• NDT for CUI

• NII of pressure vessels

Documents

Highlights from the HOIS JIP: NDT methods for CUI ... · Unmanned aerial vehicles (UAV) for external RVI recent and growing area: •Information on the external condition of piping,