Transcript

Xavier Maldague

[email protected]

http://mivim.gel.ulaval.ca

Infrared thermography for NDT:

Applications and more

Chaire de recherche du CanadaTitulaire : Xavier Maldague

Xavier Maldague

November 2013

Mor

e In

fo a

t Ope

n A

cces

s D

atab

ase

ww

w.n

dt.n

et/?

id=

1516

9

2

Outline

1. Infrared vision;

2. Non-Thermal infrared NDT;

3. Thermal infrared NDT: Passive thermography;

4. Thermal infrared NDT : Active thermography;

5. Thermal infrared NDT : More Applications;

6. Conclusions.

3

1. Infrared vision1. Infrared vision

4

Infrared vision

� Infrared vision is a

subdivision of computer

vision (i.e. the use of

computers to emulate

human vision), which

employs (low-, mid- and

high-level) computerized

processes to "make sense"

of images generated in the

infrared part of the

electromagnetic (EM)

spectrum.

Thermal emissions

Non-thermal reflections

Reflectography/

transmittography

Thermography

THzTerahertz imaging

Electromagnetic

spectrum

Planck’s Law

5

Planck’s Law

-2 .3 .8 3 5 8 14 400 400010

-10

10-5

100

105

1010

λ [µm]

Nλ,

b [W

/ m

2 µ

m]

40,000 K: Blue star

5,800 K: The Sun

3,000 K: Common light bulb (100 W)

800 K: An object starts to shine

310 K: Human body's normal temperature

273.15 K: 0oC

2.735 K: Background Cosmic Radiation9.3 µm

visible midIR

longIR

Wien

displacement

law

1 µm

Lo

i d

e W

ien

2897/6000 ~ 0.5 µm

6

Infrared

thermograms

Infrared

thermograms

Visible

photographs

Visible

photographs

Planck’s Law

Self-emissionSelf-emission

Need

illumination

Need

illumination

Non-thermal IR vision

7

NIR/SWIR

reflections or

transmissionsReflectography

Transmittography

8

Thermal IR vision

Thermal

emissions

IR vision: summary of characteristics

�NIR (0.7-1.1 µm) or

SWIR(1.1-2.5 µm)

� Reflections/transmissions

� Single images

�No advanced processing

required

�Qualitative results

�MWIR (3-5 µm) or

LWIR (7-14 µm)

� Thermal emissions

� 3D thermogram sequences

� Advanced processing

techniques

�Qualitative and quantitative

9

Non-thermal vision Thermal vision

10

2. Non-thermal Infrared NDT2. Non-thermal Infrared NDT

NIR reflectography/transmittography of GFRP

11

900 – 1700 nm

NIR: reflection NIR: transmissionvisible images

GFRP: Glass Fiber Reinforced Plastic

front

back

12

3. Thermal Infrared NDT: Passive Thermography3. Thermal Infrared NDT: Passive Thermography

13

Passive thermography: introduction

� The passive approach is used when the object of

interest has enough thermal contrast with

respect to the background in order to be

detected with an infrared sensor. Typical

applications include: surveillance, people

tracking, humidity assessment in buildings,

liquid levels in storage tanks, insulation

problems, electrical components, etc.

Sources : http://www.x20.org/thermal/ http://www.temperatures.com/thermalimaging.html

Aeronautical application: (1/4)

Water ingress detection in honeycomb

14

AA BB CC DD

EE FF GG HH

frontbackVolume, V

[ml]

F 1 0.2

B 1 0.4

G 1 0.6

C 1 0.8

E 10 2

D 10 4

H 10 6

A 10 9

Cells filled

with water

Defective

area

Section of a military

aircraft component

10 cells with 9 ml of water

not frozen frozen

Aeronautical application: (2/4)

Water ingress detection in honeycomb

� The impact of water volume

15

(a) Early thermogram at t=295 s showing all defects (A to H);

(b) Thermogram at t=518 s showing all defects except defect F;

(c) Thermogram at t=1315 s showing only defects A, D and H.

Aeronautical application: (3/4)

Water ingress detection in honeycomb

� The impact of water volume

16

The sound area (dotted

black line) warms up

following a logarithmic

growth with respect to

time

In the presence of water,

temperature profiles

diverge from logarithmic

behavior (since it takes

longer to warm up

water).

The divergent time for all defects is approximately

the same (roughly around 180 s), regardless of the

water extend and volume

0 200 400 600 800 1000 1200 1400800

1000

1200

1400

1600

1800

2000

2200

2400

2600Temperature profiles, Telops HD

t [s]

T [a

rbitra

ry u

nits

]

Sound area

0.2 ml

0.4 ml

0.6 ml

0.8 ml

2 ml

4 ml

6 ml

9 ml

Aeronautical application: (4/4)

Water ingress detection in honeycomb

�Data correlation

17

0 200 400 600 800 1000 1200 1400-100

0

100

200

300

400

500

600

700

800Temperature profiles, Telops HD

t [s]

T [a

rbitr

ary

un

its]

9 ml

6 ml

4 ml

2 ml

0.8 ml

0.6 ml

0.4 ml

0.2 ml

Time for maximum

contrast = 336.2 s

V = f ( tmax )

Maximun

contrast =

146.8

V = 3.2E-08 t 2.685

R² = 0.9911

0

2

4

6

8

10

0 250 500 750 1000 1250 1500

Wa

ter

ingre

ss v

olu

me

[ml]

Time for maximum contrast [s]

Thermograms at different distances

18

268

269

270

271

272

273

274

275

276

277

278

270

275

280

285

290

295

275

280

285

290

295

275

280

285

290

295

300

305

1.5 m from target 4 m from target

10 m from target 20 m from target

0.4 ml

defect

19

4. Thermal Infrared NDT: Active Thermography4. Thermal Infrared NDT: Active Thermography

20

Active thermography for NDT

� Active thermography for NDT is based on the detection and

recording by an infrared camera of thermal radiations emitted by

object surface.

� To detect defects, it is sometimes necessary to destabilize the

object thermal state through heating or cooling (→ active

thermography ).

� The presence of an internal defect reveals itself on surface as a

temperature perturbation above this defect.

Active thermography for NDT

Main advantages:

� Possibility to perform one-sided inspection (in reflection configuration);

� Carried out in real-time;

� Appropriate on most composites materials and multi-layer structures,

including porous materials and industrial lines;

� Relatively unaffected by the object’s geometry, and well adapted for

the inspection of large surfaces.

21

Active thermography for NDT

Main problems:

� Sensible to heating sources (type, duration, location);

� Response time (very fast for metals => need for fast acquisition rates);

� Affected by the object’s surface condition and thickness;

22

23

Non-uniform heating

+

=

24

Advanced signal processing techniques

� Thermal contrast-based techniques (max. contrast, FWHM, etc.)

� Differential Absolute Contrast, DAC

� Thermographic Signal Reconstruction, TSR

� Principal Component Thermography, PCT

� Pulsed Phase Thermography, PPT

( ) ( )te

QT πln

2

1lnln −

=∆

A=USVT

)()()( tTtTtTaSd

−=∆

( ) nn

N

k

Nnkjn tkTtF ImReexp

1

0

)2( +=∆∆= ∑−

=

− π

( )tTt

ttTT ddac ′⋅

′−=∆ )(

012 =

∂⋅−∇

t

TT

α

( )te

QTtT

π+=

0,0

3D diffusion equation

1D solution for a Dirac pulse

25

defect

Sa1

Sa2

Sa3

Sa4

0 1 2 3 4 5 6 -0.2 0 0.2 0.4 0.6

.01 .1 1 6.3

-4

-3

-2

-1

0

1

t [s]

∆T

[o

C]

Sa1

Sa2

Sa3

Sa4

.01 .1 1 6.3

-1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

t [s]

∆T

[o

C]

Sa1

Sa2

Sa3

Sa4

(a) (b)

Absolute contrast DAC

Ibarra-Castanedo C., Bendada A. and Maldague X. “Image and signal processing

techniques in pulsed thermography,” GESTS Int'l Trans. Computer Science and Engr.,

22(1): 89-100, November 2005 available online: http://www.gests.org/down/2709.pdf.

Example: DAC on CFRP

26

Partial Least-Squares Thermography

Time Series

Data

Scores

Highest variance

Loadings Residuals

PLS

Factor 1

��

�′�

PLS

Factor 2

��

�′�

PLS

Factor 3

��

�′�

PLS

Factor a

��

�′�

. . .

How many factors should be included in the

PLS model? Answer: Cross-validation

27

Partial Least-Squares Thermography

0.00637 s 0.318 s 1.726 s

0.8 mm depth

1.0 mm depth

Raw thermograms with

non-uniform heating

Enhanced with PLST,

after non-uniform

heating suppression

28

Active thermography: approaches, techniques

29

Pulsed thermography

30

Pulsed thermography, PT

�Metal corrosion, crack detection, disbonding, impact damage in

composites, turbine blades, delaminations, porosity, defect

characterization: depth, size, thermal properties, artworks.

31

Active thermography: approaches, techniques

Lock-in thermography, LT

32

Phase

permanent regimesine wave heating

• same frequency• temporal shift

Thermal waves

input:

output:

33

Lock-in thermography, LT

� Crack identification, disbonding,

impact damage, cultural heritage

inspection, artworks, cultural

buildings.

34

Active thermography: approaches, techniques

35

Eddy current (or Inductive) thermography, ECT

� Crack detection in electro-conductive materials, detection of impact

damage in composites, inspection of soldering joints.

36

Inspection of honeycomb sandwich structures

Movie: Eddy current thermography

Crack inspection: simulation

37

The approach for crack detection

SimulationGeometry of the specimen

Crack inspection: experimental

38

Experimental setup Coil and specimen

Segmentaion result

39

Active thermography: approaches, techniques

40

Vibrothermography, VT

� Coating wear, fatigue test, crack detection.

41

Open microcracks in thermally-sprayed-coatings

The coating (~100-200 µm) is

formed by a mixture of Tungsten-

Carbide and Cobalt powder

accelerated and heated in a

plasma jet and sprayed onto a 1

mm thick steel substrate.

12.8

mm

16 mm

~0.8 mm ~0.8 mm

42

Optical PT Optical LT

Burst VT Line-scan ECT

Real crushed core produced

during VT inspection

Paint detached

from the surface

Comparative example: PT, LT, VT, ECT

Inspection of CF-18 rudders (1/3)

43

Inspection of CF-18 rudders (2/3)

� Impact of de-noising with synthetic data

f=0.015 Hz f=0.04 Hz f=1.2 Hz

PPT from raw

pulsed data

PPT from

synthetic

pulsed data

44

Inspection of CF-18 rudders (3/3)

�Depth retrieval with

phase profiles

0 0.2 0.4 0.58-0.05

0

0.05

0.1

0.15

f [Hz]

∆φ

[ra

d]

z1,raw

z2,raw

z1,synt

z2,synt

z1=0.5 mm

z2=2 mm

Sa

z1

z2

45

46

5. Thermal Infrared NDT: More Applications5. Thermal Infrared NDT: More Applications

47

Road and bridges inspection (1/4)

Notre-Dame street

Montreal, Canada

November 4th, 2008

Interstate 35

Minneapolis,

August 3rd, 2007

Viaduc de la Concorde,

Montreal, Canada

September 30th, 2006

48

Road and bridges inspection (2/4)

Reinforcement

using composite

layers

Traditional

inspection with

the "tap testing"

techniqueToutry Bridge, France

49

Road and bridges inspection (3/4)

Thermal stimulation

Images acquisition

Bridge inspection: St-Urbain Bridge

50

Movie: Bridge Inspection

Fiber distribution and orientation

� Randomly-Oriented Strands (ROS)

51

Heat + Pressure

Fiber distribution and orientation

� Complex-shaped parts

52

� Fiber orientation measurement

Strength and stiffness

Fiber distribution and orientation: point scan

� Laser point scan experimental setup

53

Laboratory setup

Results show the ellipses major axes

indicating the fiber direction for the

same area at two different positions

rotated 90o

6.37°

-85.19°

Error of 1.5°

Artworks inspection (1/3)

54

Artworks inspection (2/3)

Overlay

Hidden drawingsNIR camera (0.9-1.7 µm)

incandescent lamp 90 V

Visible photograph

Artworks inspection (3/3)

Overlay

Defects

Thermal camera (3-5 µm)

PPT phase f=75 mHz

Visible photograph

57

6. Conclusions6. Conclusions

58

Conclusions

� Infrared reflectography and transmittography employ the non-

thermal part of the infrared spectrum, where the

opacity/transparency of materials, subjected to a specific infrared

radiation, are exploited to detect internal anomalies in materials.

� Infrared thermography works in the thermal part of the infrared

spectrum under the principle that dissimilar materials provide

different thermal signatures, useful for surface/subsurface defect

detection.

� Passive thermography is typically used in the field of security and

surveillance, biological applications, the inspection of electrical and

electronic components, and buildings among others.

59

Conclusions (cont.)

� Active thermography is widely used in aerospace and automobile

industries and is finding new applications such as the inspection of

bridges and roads and the assessment of artworks and cultural

heritage.

� Data processing techniques are required to enhance contrast, to

improve the spatial resolution and to increase the signal-to-noise ratio

of the infrared signal.

� Continual technological progress in commercial infrared cameras and

computers , as well as the constant development of new processing

techniques, have promoted the appearance of new and innovative

applications for infrared vision.

�� �

60

Thank you for your attention !


Recommended