Page 1
© 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
Page 2
© ESR Technology | 04/12/2018 | PAGE 2
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!
Page 3
© ESR Technology | 04/12/2018 | PAGE 3
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
Page 4
© ESR Technology | 04/12/2018 | PAGE 4
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
Page 5
© ESR Technology | 04/12/2018 | PAGE 5
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
Page 6
© ESR Technology | 04/12/2018 | PAGE 6
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
Page 7
© ESR Technology | 04/12/2018 | PAGE 7
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
Page 8
© ESR Technology | 04/12/2018 | PAGE 8
HOIS CUI & scab inspection project
Page 9
© ESR Technology | 04/12/2018 | PAGE 9
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)
Page 10
© ESR Technology | 04/12/2018 | PAGE 10
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
Page 11
© ESR Technology | 04/12/2018 | PAGE 11
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
Page 12
© ESR Technology | 04/12/2018 | PAGE 12
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)
Page 13
© ESR Technology | 04/12/2018 | PAGE 13
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
Page 14
© ESR Technology | 04/12/2018 | PAGE 14
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
Page 15
© ESR Technology | 04/12/2018 | PAGE 15
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
Page 16
© ESR Technology | 04/12/2018 | PAGE 16
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
Page 17
© ESR Technology | 04/12/2018 | PAGE 17
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
Page 18
© ESR Technology | 04/12/2018 | PAGE 18
Examples of trials in progress
Page 19
© ESR Technology | 04/12/2018 | PAGE 19
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
Page 20
© ESR Technology | 04/12/2018 | PAGE 20
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
Page 21
© ESR Technology | 04/12/2018 | PAGE 21
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
Page 22
© ESR Technology | 04/12/2018 | PAGE 22
HOIS UAV image quality project
Page 23
© ESR Technology | 04/12/2018 | PAGE 23
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
Page 24
© ESR Technology | 04/12/2018 | PAGE 24
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
Page 25
© ESR Technology | 04/12/2018 | PAGE 25
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”
Page 26
© ESR Technology | 04/12/2018 | PAGE 26
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)
Page 27
© ESR Technology | 04/12/2018 | PAGE 27
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?
Page 28
© ESR Technology | 04/12/2018 | PAGE 28
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
Page 29
© ESR Technology | 04/12/2018 | PAGE 29
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
Page 30
© ESR Technology | 04/12/2018 | PAGE 30
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.
Page 31
© ESR Technology | 04/12/2018 | PAGE 31
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?
Page 32
© ESR Technology | 04/12/2018 | PAGE 32
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
Page 33
© ESR Technology | 04/12/2018 | PAGE 33
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)
Page 34
© ESR Technology | 04/12/2018 | PAGE 34
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
Page 35
© ESR Technology | 04/12/2018 | PAGE 35
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
Page 36
© ESR Technology | 04/12/2018 | PAGE 36
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
Page 37
© ESR Technology | 04/12/2018 | PAGE 37
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
Page 38
© ESR Technology | 04/12/2018 | PAGE 38
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
Page 39
© ESR Technology | 04/12/2018 | PAGE 39
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
Page 40
© ESR Technology | 04/12/2018 | PAGE 40
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