Upload
dinhdung
View
226
Download
5
Embed Size (px)
Citation preview
David Alleyne, Tomasz Pialucha and Brian Pavlakovic
6 September 2017
Guided Wave Testing and Monitoring
Over Long and Short Ranges
BINDT Telford
2Guided Ultrasonics Ltd. - 06/09/2017
Outline
Background
Guided Wave Testing (GWT) Concepts
Wave Propagation
Tools for predicting and measuring guided waves
Interaction of GW in plates and pipes
Practical implementation
Pipe testing and Monitoring
Quantitative Short Range Scanning development
Conclusions
GWT Background
GWT research at Imperial College, London from
1980’s. Application to plates and pipes
Commercial application from late1990’s and
ongoing specialised development and
advancement in guided wave testing for
industrial inspection since.
On-going Research at Imperial College, other
Universities and research groups globally.
Mostly for long range screening.
3Guided Ultrasonics Ltd. - 06/09/2017
Imperial College NDT Group, 1991?
Can Guided waves be useful for NDT?
Example Plants where guided waves…
Need to find corrosion…..
Pulse-echo test
Ultrasonic Testing (UT) Concept
GWT Screening Concept (LRUT)
• If the guided wave encounters a defect then part of the signal is reflected
• A single measurement inspects all the material in a long length of the structure
Transducer
Structure
Volts
Time, distance
Defect
End of structure
What happens if we just try......
Problems:
- Multiple modes
- Dispersion
- Sensitivity to defects varies with modes
"Dispersion curves" for a 1mm thick steel plate
Guided wave dispersion curves
L Rayleigh (1885); H Lamb (1917)
10
10Frequency (MHz)
a0
s0
a1
s1
s2
a2Phase
speed
(km/sec)
Development of the waveguide model
DISPERSE
Developments packaged in the modelling tool "DISPERSE"
SurfacesPlates
10
10Frequency (MHz)
a0
s0
a1
s1
s2
a2Phase
speed
(km/sec)
S0 at 100kHz (10mm Steel plate)
14Guided Ultrasonics Ltd. - 06/09/2017
S0 100 kHz 6 cycles Gaussian
0 mm
100 mm
500 mm
1 meter
1.5 meters
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
-25 25 75 125 175 225 275 325 375 425
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
-25 25 75 125 175 225 275 325 375 425
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
-25 25 75 125 175 225 275 325 375 425
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
-25 25 75 125 175 225 275 325 375 425
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
-25 25 75 125 175 225 275 325 375 425
S0 at 175kHz (10mm Steel plate)
15Guided Ultrasonics Ltd. - 06/09/2017
S0 175 kHz 6 cycles Gaussian
0 mm
100 mm
500 mm
1 meter
1.5 meters
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
-25 25 75 125 175 225 275 325 375 425
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
-25 25 75 125 175 225 275 325 375 425
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
-25 25 75 125 175 225 275 325 375 425
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
-25 25 75 125 175 225 275 325 375 425
-1.2
-0.8
-0.4
0.0
0.4
0.8
1.2
-25 25 75 125 175 225 275 325 375 425
......and also:
BarsPipes
Composites
Embedded or
immersed
Multiple layers
Solids, fluids, anisotropic....
Solids
or
fluids
Development of Finite Element Modelling
Tools
Rajagopal, Lowe. J Acoust Soc Am, 122, 2007.
Rajagopal, Lowe. J Acoust Soc Am, 124, 2008.Drozdz, Moreau, Castaings, Lowe, Cawley. Rev Prog QNDE, 25, 2006.
Drozdz, Skelton, Craster, Lowe. Rev Prog QNDE, 26, 2007.
Scattering of guided waves from a crack
Need to understand the
waves
Transducer
Pipe
Need to understand how they
reflect from defects
Need to do a lot of R&D, including:
...then develop transducers, instrumentation, controlling software, signal
processing, interpretation, operator training, formal procedures..... NDT inspector
training and qualification
Expertise, knowhow requirements...
Strategy for development of
Guided Wave Testing
• Select a guided wave mode that is sensitive to
defects of interest
• Select frequency and signal shape to control
dispersion
• Excite and receive specific mode(s)
• Control directionality
Dispersion Curves
T(0,1)
F(1,2)
F(2,3)
F(1,3)
L(0,2)
Guided waves for Sch 40 6” pipe
Alleyne, Lowe, Cawley. ASME J Appl Mech, 65, 1998.
Lowe, Alleyne, Cawley. ASME J Appl Mech, 65, 1998.
Reflection coefficient
of
torsional waves
Similar study done for extensional modes
0
10
20
30
40
50
0 10 20 30 40 50
Defect extent (% of circumference)
T(0,1)
F(1,2)
F(2,2)
Frequency 50 kHz
Reflection coefficient versus
circumferential length of defect
Alleyne, Pavlakovic, Lowe, Cawley. Insight, 43, 2001.
Reflection coefficient
of
torsional waves
Similar study done for extensional modes
0
20
40
60
80
100
0 20 40 60 80 100
Defect depth (% of thickness)
40
160
130
100
70
220Frequency (kHz)
190
6 inch pipe
Reflection coefficient versus
depth of defect
Early Research Transducers (early1990’s)
Wavemaker G4
Wavemaker G4mini
HD: High Definition
HT: High TemperatureLightweight Inflatable Ring
HT Solid RingInflatable Ring
Solid Ring
HT Inflatable Ring
HD Solid Ring HD Inflatable Ring
Low Profile Ring
Claw Transducers
Permanently Installed (gPIMS) Subsea Ring Rail Systems (G-Scan)
Extensive Product Range (2015)
Transducer RingsGuided Wave
Transduction Systems
Long range pipe screening
Why use Guided Waves
NPS 20” main gas gathering line
Rapid long range screeningFig 1: 200m coverage in < 5 minutes)
Screen inaccessible areasFig 2: Screening under sliding and clamped supports
Why use Guided Waves
NPS 6”Gas Line, North Africa
100% volumetric pipe coverageFig 1: C-scan provides precise circumferential
orientation of defects or features
Accurately locate defectsFig 2: Precise location of the defect allows easy follow
with more detailed NDT techniques (e.g. B-scan)
Internal Corrosion
detected 3.25m behind
test location at 6
o’clock orientation
OmniScan MX2 used to
carry out B-scan to confirm
dimensions.
6
3.25m
Testing offshore
Testing sub-sea
31
Guided Wave Testing Standards
• BSI – BS 9690-1:2011; BS 9690-2:2011
• ASTM – E2929-11
• ASME – Article 18
• NACE TG 410 – 2014
• IIW – Commission V – EN ISO (TC135/SC3)
• TUV certification: GUL GWT procedure certified
under EN standard 14748
Guided Ultrasonics Ltd. - 06/09/2017
32
Calibration critical for procedures
• Challenge is to develop GWT procedures and work
instructions using knowledge of the damage type that
is the major threat
– Equipment and certification requirements
– SNR and access restrictions
– Limitations
– Accuracy range and confidence
• Calibration is critical
Guided Ultrasonics Ltd. - 06/09/2017
33
Guided Waves Calibration
• Equipment calibration
– Manufacturer‘s calibration (electronic components)
– Automatic self-test functions
• User calibration
– Set-up of test parameters, e.g. ring size, transducer, test frequencies (mostly automated)
• Reference standard
test object = calibration test piece
– Distance calibration
– Comparison of the amplitude of indications with a reference
Guided Ultrasonics Ltd. - 06/09/2017
Summary
• Calibration methods until now sufficient for screening, but demanding applications require a more precise approach
• Direct measurement of outgoing amplitude possible with absolute calibration method (or of reflection coefficients with attenuation known)
• Removes most problems of current calibration methods
• Works also with other features (not just welds), even if defective
• Ultimately this leads to improved false-call rate and therefore reliability
34
Guided Waves Calibration
© by Guided Ultrasonics Ltd 2013 – DO NOT RE-DISTRIBUTE
35Guided Ultrasonics Ltd. - Screening and Monitoring Oct 2011
CUI and road crossing pipes
• Road crossings and
insulated pipe
inspected for CUI
• Inspections while
pipes operational
36Guided Ultrasonics Ltd. - Screening and Monitoring Oct 2011
CUI
• Insulated steam
pipe-work at
340°C
• Pipes tested while
operational
37
Corrosion Under Supports (CUPS)
Touch point corrosion detected using GWT. The CSC range can be
less than a few per cent, therefore control of the GWT critical for
success. More later
Guided Ultrasonics Ltd. - 06/09/2017
38
GWT strategy – Touch Point corrosion
➢Torsional mode axial propagation
– Select correct equipment options and set-up
– Screen many from on accessible position (cost efficiency)
– P/E test configuration with SNR limits (corrosion type)
– Accurate calibration for DAC setting
– Limitations, range and confidence
– Prove up (no matter how limited) where possible
– Compare results for improved accuracy and confidence
– Prove-up with QSR1
➢Circumferential or axial propagation QSR1…
Guided Ultrasonics Ltd. - 06/09/2017
39
Example – corrosion patch
Guided Ultrasonics Ltd. - 06/09/2017
-16dB
-20 0 20 40 6012
9
6
3
12
Clo
ck
-20 0 20 40 600.0
0.2
0.4
0.6
0.8
1.0
1.2
Distance (ft)
Am
p (
Lin
ear)
F
weld
ed p
late
Weld
Weld
Cat 1
Corr
osi
on
Support
Weld
Cat 3
Support-F1
40
Example – pitting corrosion
Guided Ultrasonics Ltd. - 06/09/2017-1
8dB
-100 -50 0 5012
9
6
3
12
Clo
ck
-100 -50 0 500.0
0.5
1.0
1.5
Distance (ft)
Am
p (
Lin
ear) W
eld
ed s
upport-F2
Conta
ct s
upport
Conta
ct s
upport
Weld
ed s
upport
weld
weld
ed s
upport
Corr
osi
on p
it
Weld
Hig
h c
onta
ct lo
adin
g
Elb
ow
41
Example – wear to pipe
Guided Ultrasonics Ltd. - 06/09/2017
-18dB
-10.0 0.0 10.0 20.012
9
6
3
12
Clo
ck
-10.0 0.0 10.0 20.00.0
0.5
1.0
1.5
Distance (ft)
Am
p (
Lin
ear)
-F1
+F1
+F2
Corrosion monitoring using guided
waves
▪ GWT needs access to the pipe – often this accounts for a large fraction ofthe inspection cost
▪ Safety implications of accessing the site, for example excavations,working at height, hazardous areas
➢ Recurrent testing for monitoring pipe condition can therefore be costly andcarry risks
Permanently installed sensors
gPIMS®
▪ Simplifies recurrent inspections
▪ Access to test point only required once, then reduces or eliminates
access costs
▪ Increased coverage and sensitivity
▪ Simplifies comparison of data by subtraction from baseline result
▪ Enables tracking and trending
Permanently installed sensors
gPIMS®
▪ Easy-to-install array sensor bonded to the pipe
▪ Connection box with cable leading to sensor
▪ Data collection via any 4th generation Wavemaker instrument
Corrosion Monitoring with gPIMS®
▪ Looking for differences between repeat tests
▪ Thousands of results each year
Data interpretation – what is the
difference?
Before After
Subtraction of two data sets
-
=
Residual
Before After
Advanced subtraction required
Compensation algorithm
▪ Baseline at a certain temperature
▪ Decide on some reference features such as welds
▪ Stretch current result trace so amplitude and position
of reference features aligned
▪ Then subtract
Managing changes in result traces
▪ Features now suppressed, leaving a small residual
-40dB (or 1% CSC)
-40 -20 0 20
65
2.5
1
0.5
0.25
0.1
0.05
0.025
Delt
a (
50dB
)
Refe
rence
: 2011-1
0-0
5 (
-627 d
ays
)
-40 -20 0 20
65
2.5
1
0.5
0.25
0.1
0.05
0.025
Distance (m)
gC
-244#1106 (
50dB
)
Test
ed: 2013-0
6-2
3
169+56,0
44'0
0" A1
W2
169+56,0
44'0
0" A1
Efficiency of baseline subtraction
▪ In order to achieve a residual of about -40dB or 1%, the
current result should not be more than 5 - 10°C different
from the baseline
➢ This generally necessitates more than one baseline
▪ Therefore the recommendation is to collect as many baseline
results within the first year of operation as possible
▪ Specially developed monitoring software can automatically
choose the optimum baseline to use for best subtraction
efficiency.
Corrosion monitoring example
▪ Simple subtraction – difficult to detect change
-100 0 100
1.4
1
0.5
0.25
0.1
0.05
0.025
Delt
a (
40dB
)
-100 0 100
1.4
1
0.5
0.25
0.1
0.05
0.025
Distance (ft)
G3-5
1#1583 (
40dB
)
1 1 1 1 111 1 1 1
209+60 209+00 208+40Ring +F1207+80 207+20
111 111 1 1 11
208+40E1W4W3209+00W2W1 E2 W5 207+80W6 207+20W7 W8209+60
▪ Basic subtraction
Corrosion monitoring example
▪ Advanced subtraction subtraction – much simpler change detection
-150 -100 -50 0 50 1000.00
0.10
0.20
0.30
0.40
0.50
Delt
a (
Lin
ear)
-150 -100 -50 0 50 1000.00
0.10
0.20
0.30
0.40
0.50
Distance (ft)
G3-5
1#1583 (
Scale
d L
inear)
1 1 1 1 1 11 1 1 1
209+60 209+00 208+40Ring +F1207+80 207+20
111 1 11 1 1 11
208+40E1W4W3209+00W2W1 E2 W5 207+80W6 207+20W7 W8209+60
53
Corrosion Under Pipe Supports
(CUPS)
Quantitative Measurement of
Corrosion Under Support
Different Concepts…
Aim of Short Range Testing
Tomography approach
Produce map or profile of depths of corrosion
Critical parameter is maximum depth
Tomography Attempt - 2015
Tomography approach
Produce map or profile of depths of corrosion
Critical parameter is maximum depth
Top Path
Bottom Path
ReceiverTransmitter
QSR1 measurement parameters
Circumferential Guided Waves used to Measure Wall Thickness.
Multiple families of modes
Dispersive modes
At each location scanned the QSR1 system automatically measures:
The pipe diameter
Distance between Transmitter and Receiver
Top Path wall thickness
Bottom Path wall thickness
Bottom Path Minimal Wall Thickness.
QSR1 - Short Range Scanning
Adapts to different pipe wall thickness.
Quantitative measure remaining minimum
wall1.
Quantitative measure of average wall
thickness.
Estimates where the wall loss location is.
Notes:1 : Down to about half of the nominal wall thickness (and report any areas with less
than half nominal wall
Quantitative Short Range - QSR1
Guided Ultrasonics are currently working on an innovative testing system QSR1 to quantitatively measure corrosion at supports.
QSR1 automatically measures the size of corrosion to the depths of up to half of the pipe wall thickness.
QSR1 also automatically measures pipe diameter and pipe wall thickness around the pipe circumference from a single location in a fraction of a second.
Quantitative Short Range - QSR1
Ex-service 12 inch pipe with the hidden Corrosion Under Pipe Support (CUPS) type defect was scanned by QSR1 in order to obtain the profile of the area.
Quantitative Short Range - QSR1
The scanned defect was then visually examined and the QSR1 scan was compared with the reference laser scan of the area.
Quantitative Short Range - QSR1
0
1
2
3
4
5
6
7
8
9
10
-200 -150 -100 -50 0 50 100 150 200
Rem
ainin
g W
all
Thic
knes
s (m
m)
Axial Position Along the Pipe (mm)
QSR1 measurement
Laser measurement
QSR1 top path measurement
Very good agreement was achieved between the laser scan of the defected area
(red line) and the QSR1 measurements (blue line).
Over the last 20 years pipe screening for corrosion has
become a powerful method in different industries.
Screening long lengths from a single location and in a
single test generates big savings when inspecting or
monitoring.
Continuous innovation and developing new devices to
improve sensitivity and coverage has been critical to
success.
The new Quantitative short range screening technology
together with cloud computing and artificial intelligence
will facilitate new levels of industrial adoption.
Conclusions