Note: Descriptions are shown in the official language in which they were submitted.
CA 02858866 2014-06-11
W02013/086626
PCT/CA2012/001161
METHOD AND SYSTEM FOR DETECTING AND LOCATING
DAMAGES IN COMPOSITE STRUCTURES
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims priority under 35 USC
119(e) of United States Provisional Patent Application No.
61/570,959 filed on December 15, 2011, the contents of which
are hereby incorporated by reference.
FIELD OF THE APPLICATION
[0001] The
present application relates to a method and
system for damage detection and location in composite
structures subjected to loading.
BACKGROUND OF THE ART
[0002] Polymer
matrix composites have found augmented use
in many important engineering structures, such as aircrafts.
This is due to their light weight, high strength, high
stiffness and good fatigue resistance. However, along with
advantages, there are many issues. One of the issues is the
ability to detect failure in the composite structures,
particularly for in-situ processes (i.e., detection and
location of damage during service). Many techniques have
been proposed to address this problem, yet with limited
success.
[0003] In the
technique using ultrasonics, ultrasonic
waves are sent through the thickness of the composite
laminate. If the
laminate is good, the time of travel of
the ultrasonic wave is short. If there are defects in the
laminate such as delaminations or matrix cracking, the time
of travel of the ultrasonic wave is altered. Scanning over
- 1 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
the surface area of the laminate is done and a map of images
of the time of travel is displayed. This map of images is
compared against a reference map of images taken over a good
laminate. Comparison between the actual map of images and
the reference map of images can indicate the area of defect.
pm
Ultrasonics appears to be the most widely used
technique. However, this technique poses several problems.
First, the scanning needs to be carried out in a laboratory
(in the case of small samples) or in a shop (for larger
samples). The fact
that the technique cannot be used
in situ puts a limitation to its usefulness.
[0005]
Moreover, there is a need for a transfer medium to
keep the ultrasonic beam coherent. Normally, water is used
as the transfer medium. This
presents the problem of
introduction of water and the mess that may be created due
to wetness. Ultrasonic laser has been introduced recently.
This helps the coherence of the beam to some extent.
However, the need for scanning still limits the usefulness
of the technique.
[0006] Another
technique is referred to as acoustic
emission. In
acoustic emission, an artificial ear (high-
frequency sensor) is attached to a composite structure.
When a crack occurs, this crack creates stress waves that
propagate in the structure. When the wave hits the sensor,
the sensor records and displays a signal. By reading the
signal displayed from the sensor, one may tell whether a
crack has been formed. By placing many sensors in a certain
geometrical pattern on the surface of the structure, the
crack may be located. This technique is however subjected
to the problem of the extraneous noise coming from many
sources, such as the reflection and reverberation from
stiffeners, free edges, etc., which can interfere with the
desired signal. Some
time unloading can also produce
signals due to the rubbing of the existing crack surfaces.
- 2 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
3
The interference of other signals confuses the information
and makes the technique intractable.
[0007] According
to another technique, X Rays are used.
In this technique, the sample is subjected to X Ray
diffraction.
Observation of the X Ray photograph can
differentiate the defect from the good material. This
technique can only be applied in laboratory.
[0008]
Thermography operates on the principle that the
heat emitted from the surface of a structure depends on the
stress state of the material below the surface. If there
are defects in the material below the surface, the
temperature distribution on the surface (down to a fraction
of a degree) will be non-uniform. By comparing the thermal
image of a reference sample to that of the sample studied,
one can determine whether there are defects in the sample.
Again this technique is applicable in laboratory and cannot
be used in situ.
[0009]
Shearography is an optical technique that involves
holography and speckle interferometry. An expanded laser
beam is used to illuminate the region to be examined on the
surface of the object. The surface can be used as is or a
layer of paint or powder layer can be applied on top. Light
scattered by the surface is recorded using a camera. The
camera has a glass wedge or shearing interferometer to
generate a double image of the object surface.
[0010] To detect
the deformation of the surface, a
reference interferogram is first recorded. Subsequently
interferograms are recorded either statically of dynamically
as the object is loaded. Either a single camera or multi-
camera sensor can be used to obtain the displacement
gradient components.
[0011] The
disadvantages of this technique include the
fact that a high precision location between the part and the
camera is required, which may not be suitable for field
- 3 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
4
applications. Also, it cannot be used for the measurement
of the bulk strain in the material.
[0012] As
another technique, optical fibers with gratings
(such as Bragg gratings) are either embedded inside the
composite structure or bonded on its surface. The
deformation of the material in the structure is transmitted
to the optical fiber. The
strain in the optical fiber
changes the spacing between the gratings. By sending light
waves in the fibers, the reflection of the light wave from
the gratings changes. By
observing the changes in the
reflected wave as compared to referenced signals, one can
determine the level of the strain. The technique is
interesting in that it can be used in situ. However the size
of the optical fibers is fairly large (about 100 microns).
Embedding these fibers in the composite structures with
fiber diameters in the order of 10 microns creates stress
concentration and may induce damage. Bonding the fibers on
the surface of the structure only allows it to detect
deformation locally. The
difference between the use of
optical fibers and strain gages needs to be proven.
Besides, the fibers are fragile and the equipment is bulky.
[0013] Over the
past few years, the electrical
conductivity of carbon fibers has been used as an indication
for the presence of damage. Since
carbon fibers are
electrically conductive along the fiber direction, by
applying an electric current over two probes at two points
along the direction of the fibers, the change in electric
voltage can be taken as an indication of damage in the
carbon fibers composites. The problem with this technique
is that since the resin is not conductive, one cannot use
the technique to detect resin cracks. The
majority of
damages at the relatively low loads is due to matrix
cracking and delamination, rather than to fiber breakage.
As such, the usefulness of this technique is limited. Also,
- 4 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
for composites made of fibers such as glass or Kevlar which
are not electrically conductive, the method does not work.
[0014] A number
of resistance-based sensors have been
developed for the sensing of deformation and damage in
composite structures. For
instance, a resistance-based
sensor consisting of wire contacts embedded at the edge of
carbon fiber composite has been developed. The sensor was
shown to be capable of detecting barely visible impact
damage, and accurately providing the damage location [L Hou
and S A Hayes "A resistance-based damage location sensor for
carbon-fibre composites" Smart Materials and Structures. 11
(2002) 966-969].
[0015] However,
the size of the panel is small (12 cm by
cm) and only one damage site was created. There have also
been developed sensing systems that consist of depositing
lines of conductive ink containing carbon nanofibers and
polymeric resin on the surface of the structure in the form
of a grid pattern. Electrical wires are connected to the
ends of these grid lines for measuring the changes in
resistances between grid lines [Rice, Brian P. "Sensing
system for monitoring the structural health of composite
structures" United States Patent Application 20050284232
Al]. A thin film containing carbon nanotubes intended for
detecting defects in structures has been developed [Lynch.
Jerome P., Huo. Tsung-chin, Kotov. Nicholas A., Kam. Nadine
Wong Shi, Loh, Kenneth J. "Electrical Impedance Tomography
of Nanoengineered Thin Films" United States Patent
Application 20090121727 Al]. Smits et al proposed a carbon
nanotube based sensor which consists of a number of
conductors and carbon nanotubes arranged in an array [Smits.
Jan M., Kite. Marlen T., Moore. Thomas C., Wincheski.
Russell A., Ingram. Joanne L.Watkins, Anthony N., Williams.
Phillip A. "Carbon nanotube-based sensor and method for
detection of crack growth in a structure" United States
- 5 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
6
Patent 7278324 B2, Oct.9, 2007]. Hayes and Jones presented
a composite material system where the fiber reinforcement
comprises electrically conductive fibers [Hayes. Simon,
Jones. Frank "Electrical damage detection system for a self-
healing polymeric composite" United States Patent
Application 20090294022 Al] .
[0016]
Recently, carbon nanotubes (CNTs) have been added
into resin such as epoxy to make the matrix electrically
conductive in composite structures. This
conductivity is
due to the formation of a network of the carbon nanotubes.
Upon the application of a mechanical load, the network is
stretched. If the load is high enough to create cracks in
the matrix material, the configuration of the network is
affected and there is a change in the conductivity. Chou
et al. added carbon nanotubes into epoxy matrix to make
glass/epoxy composite samples and showed that cracking in
tensile and impact samples corresponds with the increase in
electrical resistance of coupons of about 4 inches by 6
inches or smaller [Limin Gao, Tsu-Wei Chou, Erik T.
Thostenson, and Magali Coulaud "In situ sensing of impact
damage in epoxy/glass fiber composites using percolating
nanotubes networks" Carbon, 2011, 49, 3382-3385; Limin Gao,
Erik T. Thostenson, Zuoguang Zhang, and Tsu-Wei Chou
"Sensing of Damage Mechanisms in Fiber-Reinforced Composites
under Cyclic Loading using Carbon Nanotubes" Adv. Functional
Materials, 2009, 19, 123-130.; E. Thostenson and T. Chou
"Carbon nanotube networks: sensing
of distributed strain
and damage for life prediction and self-healing" Advanced
Material, 2006, 18, 2837-2841; E. Thostenson and T. Chou
"Real-time in situ sensing of damage evaluation in advanced
fiber composites using carbon nanotubes networks"
Nanotechnology 2008, 19, 215713; Thostenson, Erik T, Chou,
Tsu-wei "Method and system for detecting damage in aligned
- 6 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
7
carbon nanotube fiber composites using networks" U.S. Patent
No. 7,786,736-] .
[0017] S.V.Hoa et al. incorporated carbon nanotubes into glass
fiber reinforced epoxy composite to detect damage using
electrical resistance measurement and found that there is
correspondence between the change in electrical resistance
and damage accumulation during fatigue and tensile testing
in the small coupon specimen [ Mohammadreza Nofar, Suong V.
Hoa and Martin Pugh "Self sensing glass/epoxy composites
using carbon nanotubes" ICCM 17 Edinburgh, UK, 27-31 July
2009; M.Nofar, Suong V. Hoa and Martin Pugh "Failure
detection and monitoring in polymer matrix composites
subjected to static and dynamic loads using carbon nanotube
networks" Composites Science and Technology, August 2009,
Volume 69, Issue 10 ,Pages 1599-1606; H. Hena-Zamal and S.
V. Hoa "Fatigue damage behavior of glass/epoxy composite
using carbon nanotubes as sensors" 26 th ICAF Symposium -
Montreal, 1-3 June 2011;H. Hena-Zamal "Monitoring Fatigue
Damage Behavior of Glass/Epoxy Composites Using Carbon
Nanotubes as Sensors" Master thesis, Concordia University,
April 2011].
[0018] Athanasios Baltopoulos et al. described the
forward and inverse methods for detecting the location of
cracks in small specimen with size of 4 inch by 4 inch glass
fibers/epoxy/CNTs composite plate using Electric Resistance
Tomography (ERT) technique. They mounted electrodes around
the boundary of the specimen which fails the proposed
technique to detect damage at the center of the
plate.[Athanasios Baltopoulos, Nick Polydorides, Antonios
Vavouliotis, Vassilis Kostopoulos, Laurent Pambaguian
"sensing capabilities of multifunctional composite materials
using carbon nanotubes"IAC-10.C2.9.2,61st International
Astronautical Congress, Prague, CZ,2010].
- 7 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
8
[0019] Zhang et
al. embedded carbon nanotubes in graphite
fiber/carbon nanotube/epoxy laminates and showed that there
is a good correspondence between crack propagation and
increase in electrical resistance in the sample [Zhang W.,
Sakalkar V., Koratkar N. "In situ health monitoring and
repair in composites using carbon nanotubes additives" Appl
Phys Lett 2007; 91(31) :1-3] .
[0020] Proper
et al. embedded carbon nanotubes in epoxy
reinforced with Kevlar fibers. They then incorporated an
electrical grid on the surface of the sample and applied
electric potential across the grid. They showed that when
the sample (4 inch by 6 inch) is damaged by mechanical
impact, there is a correspondence between the change in
voltage across the grid and the impact damage. In their
arrangement, the distance between the grid points is fairly
small (0.25 inch). [A. Proper, W. Zhang, S.
Bartolucci,
A. Oberai and N. Koratkar "In-Situ Detection of Impact
Damage in Composites Using Carbon Nanotube Sensor Networks"
Nanoscience and Nanotechnology Letters 1, 3-7, 2009].
[0021] Wardle
et al. developed a system where carbon
nanotubes are grown radially from fibers made of alumina.
Resins are then incorporated into the fibers to make
laminates. Silver ink is then deposited to form electrode
lines on the surface of the laminate. The distance between
the electrode lines is about 0.118 inch. The
resistances
across the electrodes are measured, and it has been found
that there is correspondence between the changes in
resistance and damages created by impact loading. [Sunny
Wicks, Derreck Barber, Ajay Raghavan, Christopher T. Dunn,
Leo Daniel, Seth S. Kessler, Brian L. Wardle "Health
monitoring of carbon nanotube (CNT) hybrid advanced
composite for space applications" MIT; Ajay Raghavan,
Seth S. Kessler Christopher T. Dunn, Derreck Barber, Sunny
Wicks, Brian L. Wardle "Structural health monitoring using
- 8 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
9
carbon nanotube (CNT) enhanced composites", Proceedings of
the 7th International workshop on structural health
monitoring (IWSHM07), Stanford University, September 9-11,
2009].
[0022] Boger et
al. added carbon nanotubes and carbon
black in the glass/epoxy composites and found that there is
correspondence between the change in strain and the change
in electrical resistance [Boger L., Wichmann
M.H.G.,
Meyer, L.O., Schulte K., "Load and health monitoring in
glass fibre reinforced composites with an electrically
conductive nanocomposite epoxy matrix", Composites Science
and Technology 2008; 68:1886-1894].
[0on] Shang-
lin, Gao et al. deposited carbon nanotubes
onto glass fiber surfaces and found that epoxy composites
made using this fiber system may be used for in-situ sensing
of strain and damage [Shang-lin Gao, Rong-Chuan Zhuang, Jie
Zhang, Jian-Wen Liu, and Edith Mader "Glass Fibers with
Carbon Nanotube Networks as Multifunctional Sensors" Adv.
Functional Materials. 2010, 20, 1885-1893].
poN The
above works show very interesting and
innovative attempts to monitor deformation and damage in
composite materials. Many sensors have been developed and
show promise. However it remains to be seen whether these
sensors can win out over other existing sensors such as
strain gages, fiber optics, etc. The
incorporation of
carbon nanotubes in the epoxy used for the fiber/epoxy
composites is very interesting. However, the works in this
area have been limited to composite coupons of relatively
small sizes. Works
have been made on coupons of about
4 inches by 6 inches or smaller. Cracks
or damages
occurring in these samples would certainly interrupt the
flow of electrical current from one electrical probe to
another.
Extending the technique from small coupons to
larger structure runs into many challenging issues.
- 9 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
[0025] One
issue is the spatial non-uniformity in the
electrical conductivity over the surface of the structure.
Mixing the carbon nanotubes in the epoxy resin takes a lot
of effort. Incorporating the modified epoxy with the long
fibers requires the penetration of the nanoparticles in
between the long fibers. Many of the works cited above use
vacuum assisted resin transfer molding to make the
laminates. In this process, only vacuum is used and there
may not be sufficient pressure to compact the laminates
well. For
prepreg and autoclave curing, it is important
that the uniform distribution of the carbon nanotubes in the
whole of the structure to be obtained.
Otherwise, the
technique may not work.
SUMMARY OF THE APPLICATION
[0026] It is
therefore an aim of the present disclosure
to provide a method and system to detect and locate damages
in composite structures using carbon nanotubes.
[0027]
Therefore, in accordance with the present
application, there is provided a damage detection and
location method for a composite structure of the type
incorporating carbon nanotubes, the method comprising
receiving from a plurality of electrical contacts arranged
on a surface of the composite structure measurements of an
electrical property of the composite structure, calculating
a change between the received measurements and reference
values of the electrical property, and identifying a damage
if the change is above a predetermined threshold.
[0028] Still
further in accordance with the present
application, identifying the damage comprises locating the
damage by correlating the change to a location in the
composite structure of selected ones of the plurality of
electrical contacts having provided the measurements.
- 10 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
11
[0029] Still further in accordance with the present
application, receiving measurements of the electrical
property of the composite structure comprises receiving a
plurality of the measurements from a first grid of
electrically conductive points arranged on the surface of
the composite structure, each one of the plurality of the
measurements comprising an electrical resistance measured
between a first one and a second one of the electrically
conductive points.
[0030] Still further in accordance with the present
application, the composite structure is an electrically non-
conductive fiber reinforced composite structure and the
range of the electrical resistance is between 103 ohm and
106 ohm.
[0031] Still further in accordance with the present
application, calculating the change between the received
measurements and the reference values comprises calculating
for each one of the plurality of the measurements a
difference between the measured electrical resistance and a
reference electrical resistance between the first and the
second electrically conductive point.
[0032] Still further in accordance with the present
application, receiving measurements of the electrical
property of the composite structure comprises receiving a
plurality of the measurements from a second grid of
electrically conductive lines arranged in a first and a
second orientation on the surface of the composite
structure, each one of the plurality of the measurements
comprising a first electrical resistance measured between
two consecutive ones of the electrically conductive lines
arranged in the first orientation and a second electrical
resistance measured between two consecutive ones of the
electrically conductive lines arranged in the second
orientation.
- 11 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
12
[0033] Still further in accordance with the present
application, calculating the change between the received
measurements and the reference values comprises, for each
one of the plurality of the measurements, calculating a
first difference between the first measured electrical
resistance and a first reference electrical resistance
between the two consecutive electrically conductive lines
arranged in the first orientation, calculating a second
difference between the second measured electrical resistance
and a second reference electrical resistance between the two
consecutive electrically conductive lines arranged in the
second orientation, and computing an average of the first
and the second difference.
[0034] Still further in accordance with the present
application, receiving measurements of the electrical
property of the composite structure comprises receiving a
plurality of the measurements from a first grid of a first
set of electrically conductive points and a second grid of a
second set of electrically conductive points, the first and
second grids arranged on the surface of the composite
structure, the first set of electrically conductive points
for applying a constant electric current to the composite
structure and the second set of electrically conductive
points for measuring an electrical potential.
[0on] Still further in accordance with the present
application, receiving measurements of the electrical
property of the composite structure comprises receiving the
plurality of the measurements each comprising the electrical
potential measured between a first one and a second one of
the second set of electrically conductive points.
[0036] Still further in accordance with the present
application, calculating the change between the received
measurements and the reference values comprises calculating
for each one of the plurality of measurements a difference
- 12 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
13
between the measured electrical potential and a reference
electrical potential between the first one and the second
one of the second set of electrically conductive points.
[0037] In
accordance with the present application, there
is further provided a method for fabricating a composite
structure of the type incorporating carbon nanotubes, the
method comprising preparing a first amount of an epoxy
resin, preparing a second amount of a curing agent to the
epoxy resin, preparing a third amount of the carbon
nanotubes, mixing the first amount of the epoxy resin with
the second amount of the curing agent to produce the epoxy
matrix, dispersing the third amount of the carbon nanotubes
into the epoxy matrix to produce the modified epoxy matrix,
and incorporating the modified epoxy matrix into long
fibers.
[0038] Still further
in accordance with the present
application, incorporating the modified epoxy matrix into
the long fibers comprises incorporating the modified epoxy
matrix into one of electrically non-conductive long fibers
and electrically conductive long fibers.
[0039] Still further in
accordance with the present
application, the carbon nanotubes are provided in the
composite structure in a presence ensuring uniform
distribution of the carbon nanotubes in the composite
structure, electrical conductivity of the composite
structure, and detectability of a damage in the composite
structure.
[0040] In
accordance with the present application, there
is further provided a damage detection and location system
for a composite structure of the type incorporating carbon
nanotubes, the system comprising a plurality of electrical
contacts arranged on a surface of the composite structure, a
damage detection unit for detecting and locating a damage in
the composite structure, the damage detection unit
- 13 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
14
comprising an acquisition module for receiving from the
plurality of electrical contacts measurements of an
electrical property of the composite structure, a value
comparator for calculating a change between the received
measurements and reference values of the electrical
property, and for identifying a damage from the change
having a magnitude beyond a predetermined threshold, a
damage locator for determining a location of the damage, and
an output for providing data pertaining to the damage and
the location thereof.
[0041] Still further in accordance with the present
application, the damage locator is adapted to determine the
location of the damage by correlating the change to a
position on the composite structure of selected ones of the
plurality of electrical contacts having provided the
received measurements.
[0042] Still further in accordance with the present
application, the plurality of electrical contacts is
arranged on the surface of the composite as one of a first
grid of electrically conductive points deposited on the
surface of the composite structure and a second grid of
electrically conductive lines arranged in a first and a
second orientation on the surface of the composite
structure.
[0043] Still further in accordance with the present
application, the acquisition module receives a plurality of
the measurements each comprising an electrical resistance
measured between a first one and a second one of the
electrically conductive points.
[0044] Still further in accordance with the present
application, the value comparator calculates the change by
calculating for each one of the plurality of the
measurements a difference between the measured electrical
- 14 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
resistance and a reference electrical resistance between the
first and the second electrically conductive points.
[0045] Still further in accordance with the present
application, the acquisition module receives a plurality of
the measurements each comprising a first electrical
resistance measured between two consecutive ones of the
electrically conductive lines arranged in the first
orientation and a second electrical resistance measured
between two consecutive ones of the electrically conductive
lines arranged in the second orientation.
[0046] Still further in accordance with the present
application, the value comparator calculates the change by
calculating, for each one of the plurality of the
measurements, a first difference between the first measured
electrical resistance and a first reference electrical
resistance between the two consecutive electrically
conductive lines arranged in the first orientation, a second
difference between the second measured electrical resistance
and a second reference electrical resistance between the two
consecutive electrically conductive lines arranged in the
second orientation, and an average of the first and the
second difference.
[0047] Still further in accordance with the present
application, the plurality of electrical contacts is
arranged on the surface of the composite structure as a
first grid of a first set of electrically conductive points
and a second grid of a second set of electrically conductive
points, the first and second grids deposited on the surface
of the composite structure, the first set of electrically
conductive points for applying a constant electric current
to the composite structure and the second set of
electrically conductive points for measuring an electrical
potential.
- 15 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
16
[0048] Still further in
accordance with the present
application, the acquisition module receives a plurality of
measurements each comprising the electrical potential
measured between a first one and a second one of the second
set of electrically conductive points.
[0049] Still further in
accordance with the present
application, the value comparator calculates the change by
calculating for each one of the plurality of measurements a
difference between the measured electrical potential and a
reference electrical potential between the first one and the
second one of the second set of electrically conductive
points.
glom In
accordance with the present application, there
is further provided a computer readable medium having stored
thereon program code executable by a processor for damage
detection and location in a composite structure of the type
incorporating carbon nanotubes, the program code executable
for receiving from a plurality of electrical contacts
arranged on a surface of the composite structure
measurements of an electrical property of the composite
structure, calculating a change between the received
measurements and reference values of the electrical
property, and identifying a damage if the change is above a
predetermined threshold.
BRIEF DESCRIPTION OF THE DRAWINGS
[0051] Fig. la
is a flowchart of a damage detection and
location method for composite structures, in accordance with
an embodiment of the present disclosure;
[0052] Fig. lb
is a flowchart of the step of Fig. la of
fabricating composite structures;
- 16 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
17
[0053] Fig. lc is a flowchart of the step of Fig. la of
detecting, locating, and determining the severity of damages
in composite structures;
[0054] Fig. id is a flowchart of the step of Fig. lc of
monitoring damages of composite structures;
[0055] Fig. 2a is a graph of a distribution of electrical
resistances between measurements points in a 22x13"
composite laminate containing 0.20 wt % CNTs, showing a non-
uniform electrical resistance distribution;
[0056] Fig. 2b is a graph of a distribution of electrical
resistances between measurements points in a 22x13"
composite laminate containing 0.25 wt% CNTs;
[0057] Fig. 2c is a graph of a distribution of electrical
resistances between measurements points in a 22x13"
composite laminate containing 0.30 wt% CNTs;
[0058] Fig. 2d is a graph of a distribution of electrical
resistances between measurements points in a 22x13"
composite laminate containing 0.40 wt% CNTs;
[0059] Fig. 2e is a graph of a distribution of electrical
resistances between measurements points in a 22x13"
composite laminate containing 1 wt% CNTs;
[0060] Fig. 3a is a block diagram of a damage detection
and location system for composite structures, in accordance
with an embodiment of the present disclosure;
posq Fig. 3b is a flowchart of an algorithm implemented
by the analysis module of Fig. 3a for detecting and locating
damages in electrically non-conductive fibers reinforced
polymer composite structures containing CNTs;
[0062] Fig. 3c is a flowchart of an algorithm implemented
by the analysis module of Fig. 3a for detecting and locating
damages in electrically conductive fibers reinforced polymer
composite structures containing CNTs;
[0063] Fig. 4a is a schematic view of the system of Fig.
3a showing a matrix of measurement points;
- 17 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
18
[0064] Fig. 4b is a schematic view of an alternative
arrangement of the system of Fig. 3a, with sets of two grid
points on the surface of a carbon fibers/epoxy/CNTs
composite structure;
[0065] Fig. 4c is a schematic view of an alternative
arrangement of the system of Fig. 3a, showing a network of
electrical lines;
[0066] Fig. 5a is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after hole of size 1 (1/16) inch is
drilled;
V1067] Fig. 5b is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after two holes of sizes 1 and 2
(1/16 and 2/16) inch are drilled;
[0068] Fig. 5c is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after three holes of sizes 1, 2 and
3 (1/16, 2/16 and 3/16) inch are drilled;
g069] Fig. 5d a graph of an electrical resistance change
distribution of a 22x13" glass fibers/epoxy/ 0.30 wt % CNTs
composite laminate after holes of sizes 1, 2, 3 and 4
(1/16, 2/16, 3/16 and 4/16) inch are drilled;
[0070] Fig. 5e is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after five holes of sizes 1, 2, 3,
4 and 5 (1/16, 2/16, 3/16, 4/16 and 5/16) inch are drilled;
[0071] Fig. 5f is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after six holes of sizes 1, 2, 3, 4,
and 6 (1/16, 2/16, 3/16, 4/16, 5/16 and 6/16) inch are
drilled;
[0072] Fig. 6 is a graph showing a severity of damage due
to various sizes of holes;
- 18 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
19
[0073] Fig. 7a is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after impact damage 1 (78J) is made;
[0074] Fig. 7b is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after impact damage 1 and 2 (78J
each) are made;
[0075] Fig. 7c is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after impact damage 1, 2 and 3 (78J
each) are made;
[0076] Fig. 7d is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after impact damage 1, 2, 3 and 4
(78J each) are made;
[0077] Fig. 7e is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after impact damage 1, 2, 3, 4 and 5
(78J each) are made;
glom Fig. 7f is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after impact damage 1, 2, 3, 4, 5
and 6 (78J each) are made;
[0079] Fig. 8a is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after barely visible impact damage 1
(1J) is made;
[0080] Fig. 8b is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after barely visible impact damages
1 and 2 (1J and 2J) are made;
voirq Fig. Sc is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
- 19 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
CNTs composite laminate after barely visible impact damages
1, 2 and 3 (1J, 2J and 3J) are made;
[0082] Fig. 8d is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate 8 after barely visible impact
damages 1, 2 ,3 and 4 (1J, 2J, 3J and 4J) are made;
[0083] Fig. 8e is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after barely visible impact damages
1, 2, 3, 4 and 5 (1J, 2J, 33, 43 and 5J) are made;
[008] Fig. 8f is a graph of an electrical resistance
change distribution of a 22x13" glass fibers/epoxy/ 0.30 wt%
CNTs composite laminate after barely visible impact damages
1, 2, 3, 4, 5 and 6 (1J, 2J, 3J, 4J, 5J and 10J) are made;
[0085] Fig. 9 is a graph showing a severity of damage due
to various applied impact energies;
[0086] Fig. 10 is a graph of an average electrical
resistance change distribution of 22x13" glass fibers/epoxy/
0.30 wt% CNTs composite laminate after holes of sizes 1, 2,
3, 4, 5 and 6 (1/16, 2/16, 3/16, 4/16, 5/16 and 6/16) inch
located at A, B, C, D, E and F respectively are drilled;
[0087] Fig. lla is a graph of an electric potential
distribution of 22x13" carbon fibers/epoxy/ 0.3 wt % CNTs
composite laminate;
[0088] Fig. lib is a graph of an absolute electric
potential change distribution of 22x13" carbon fibers/epoxy/
0.30 wt% CNTs composite laminate after six holes of sizes 1,
2, 3, 4, 5 and 6 (1/16, 2/16, 3/16, 4/16, 5/16 and6/16) inch
respectively are drilled;
[0089] Fig. 11c is a graph showing the effect of hole
size on the change in electrical potential;
[0090] Fig. lid is a graph of an absolute electric
potential change distribution of 22x13" carbon fibers/epoxy/
0.30 wt% CNTs composite laminate after impact damage 1, 2,
- 20 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
21
3, 4, 5 and 6 are made using 318 mg aluminum particles
travelling at 700 m/sec (78J).; and
[0091] Fig. lie is a graph of an absolute electric
potential change distribution of 22x13" carbon fibers/epoxy/
0.30 wt% CNTs composite laminate after
barely visible
impact damages 1, 2 and 3 (1J, 2J and 3J) are made.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0092] Referring now to Fig. la, a method 100 for damage
detection and location in composite structures will now be
described. The composite structures are, for instance, of
relatively large scale, as commonly found in the aerospace
industry, the wind-turbine industry, the automotive
industry, the naval ships, the civil structures (composite
rebars), space structures, such as satellites, etc., to name
a few of many instances using composite structures. The
composite structures may consist of any polymer matrix
composite materials, with any appropriate type of resin
binding the fibers.
[0093] The method 100 illustratively
comprises
fabricating at step 102 composite structures using carbon
nanotubes (CNTs), depositing at step 104 a grid of
electrical contacts, such as electrically conductive
adhesive points or lines, as will be discussed further
below, on the surface of the composite structures
manufactured at step 102, and at step 106 detecting,
locating, and determining the severity of damages in the
composite structures in real-time and in-situ.
[0094] Referring to Fig. lb, the step 102 of fabricating
the composite structures comprises mixing at step 108 carbon
nanotubes in a polymer resin, such as an epoxy resin, and
incorporating at step 110 the modified epoxy resin with long
- 21 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
22
electrically conductive or electrically non-conductive
fibers to make the composite structures.
[0095] The step 108 may require preparing a predetermined
amount of polymer resins such as epoxy resin, preparing a
predetermined amount of curing agent for the epoxy resin,
preparing a predetermined amount of electrically conductive
nanoparticles such as carbon nanotubes, mixing the
predetermined amount of epoxy resin with the predetermined
amount of curing agent and obtaining an epoxy matrix which
is a combination of epoxy resin and curing agent, dispersing
the predetermined amount of carbon nanotubes into the epoxy
matrix to make the epoxy matrix electrically conductive, and
obtaining modified epoxy matrix which is a combination of
epoxy resin, curing agent and carbon nanotubes.
[0096] The step 110 may require preparing a predetermined
amount of long electrically non-conductive fibers such as
Kevlar and glass fibers or long electrically conductive
fibers such as carbon fibers, incorporating the modified
epoxy matrix with the predetermined amount of long
electrically nonconductive fibers such as glass fibers or
long electrically conductive fibers such as carbon fibers,
and obtaining smart structures made of fibers reinforced
polymer matrix composite materials.
[0097] The presence of carbon nanotubes in epoxy resin
provides electrical conductivity for the resin. According
to an embodiment, mixing may be done at step 108 by using a
three-roll calendering machine. Multiwall carbon nanotubes
may be used. The amount of the carbon nanotubes should not
be too large or too small. If the amount of the nanotubes
is too small, it will not provide a good percolation network
for electrical conductivity, or stability of the results.
If the amount of carbon nanotubes is too high, it creates
high viscosity which renders subsequent incorporation of
continuous fibers difficult. Also, too small an amount of
- 22 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
23
nanotubes will tend to give spatial variability in the
electrical conductivity. On the other hand, too large an
amount of carbon nanotubes will reduce the sensitivity of
the change in electrical resistance due to the occurrence of
damage.
[0098] It is
desirable for the incorporation at step 110
of epoxy containing CNTs into long fibers to take into
account uniformity of distribution of carbon nanotubes. Non-
uniform distribution of carbon nanotubes may not provide
good results due to the short-circuit phenomenon. For a
structure where there are regions of good conductivity and
regions of poor conductivity, the electrical current will
follow the path of good conductivity and avoid the path of
poor conductivity. When there is a defect in the region of
poor conductivity, the system cannot sense it due to the
lack of current flow. As such,
damage detection and
location thereof, as effected at step 106 of the method 100,
may not be efficient in the case of non-uniform CNT
distribution.
[0099] The
incorporation at step 110 of the modified
epoxy into the long fibers requires the penetration of the
nanoparticles in between the long fibers. Some of the prior
art works use vacuum-assisted resin transfer molding to make
the laminates. In this process, only vacuum is used and
there may not be sufficient pressure to compact the
laminates well. For
prepreg and autoclave curing, it is
desirable to obtain the uniform distribution of the carbon
nanotubes in the whole structure, as will be discussed
further with reference to Fig. 2a to Fig. 2e.
[00100]
Referring to Fig. lc, the step 106 of detecting,
locating, and determining the severity of damages in the
composite structures illustratively comprises the step 112
of coupling the grid deposited at step 104 of Fig. la to a
damage detection unit. For this purpose, electrical wires
- 23 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
24
may be attached to the grid to make electrodes. The
electrodes may in turn be connected to the damage detection
unit. The damage detection unit, which will be discussed
further below with reference to Fig. 4, may comprise an
acquisition module for collecting and storing information
regarding electrical properties of the composite structures
and an analysis module for analyzing the collected
information to detect, locate, and determine the severity of
damages in the composite structures.
[00101] Once the grid is coupled to the damage detection
unit at step 112, reference values of electrical properties
of the composite structures may be established at step 114.
For this purpose and as will be described further below, the
electrical properties, e.g. electrical resistance or
potential, between pairs of contacts on the grid, e.g. grid
points or grid lines, may be measured. In this manner,
reference values for all measurements points of the
composite structure containing carbon nanotubes may be
obtained. The reference values are, for instance, obtained
before the composite structure is used, or at a reset, for
instance after an inspection or a use.
[00102] The damages of the composite structures may then
be monitored in-situ and in real-time at step 116. In
particular, and referring to Fig. id, the step 116 of
monitoring the damages may comprise obtaining at step 118
present values during the use of the composite structure.
According to an embodiment, the present values are
continuously measured or punctually measured, at given time
intervals. A change or difference (A) between present values
and reference values may then be calculated for each set of
measurement points. If the difference is identified at step
120 as being above a given threshold or tolerance, this may
be an indication of damage. It may be desired to continue
monitoring the values of the set of measurement points, as
- 24 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
well as sets of measurement points neighbour to the
identified above-threshold set of measurement points, to
confirm that there is indeed a damage. It is considered to
set to zero the calculated difference of the identified
above-threshold set of measurement points and neighbour sets
of measurement points to confirm that there is a damage.
polim When damage is confirmed, the location of the
damage relative to the composite structure is provided at
step 122. This is done by correlating the identified above-
threshold set of measurement points to its location or
position on the composite structure. It is anticipated that
multiple values of the measured electrical properties, e.g.
electrical resistance or electrical potential, may be above
the threshold. In such case, the maximum values of the
measured electrical properties may be used to identify the
specific location of the damage.
[00104] The damage is then output at step 124. The damage
may be output in the form of a simple notification, an
identification of the location, and/or a quantification of
the damage. The output may be provided as a plot or any
other suitable presentation means. An example of output is
provided below in an exemplary embodiment.
gmq Referring to Fig. 2a to Fig. 2e and as discussed
above with reference to Fig. lb, in performing the step of
incorporating long fibers into the modified epoxy resin, it
is desirable to achieve uniform distribution of the carbon
nanotubes in the whole structure. Having an adequate
electrical conductivity further gives suitable results. If
the conductivity is too high, the occurrence of a defect may
be picked up as the change in electrical resistance between
contact points in the vicinity of the crack, for larger
cracks only. This was reflected in the cases of samples
containing 0.3 wt% CNTs and 1 wt% CNTs where the changes in
electrical resistance of 4.8% and 0.446% respectively were
- 25 -
CA 0213866 213106-11
WO 2013/086626
PCT/CA2012/001161
26
observed. Too small an amount of CNTs in the composite
structure A (even more than the percolation threshold of
CNTs for the case of small samples) will not guarantee
uniform distribution of electrical resistance. If the
distribution of electrical resistance is not uniform, it
makes it difficult to perform a comparison in electrical
resistance between new or present resistance values and
reference resistance values, leading to imprecise damage
detection and location.
[00106] In an experiment, the measured electrical
resistances show that a twenty-two (22) inch by thirteen
(13) inch composite laminate containing 0.10% CNTs does not
behave as a conductive material and it is approximately an
insulator. Fig. 2a to Fig. 2e show the distribution of the
electrical resistance between grid points in a composite
laminate (22 inch by 13 inch) made using different amounts
of CNTs (from 0.20 wt% to 1.0 wt%). It can be seen that the
distribution of electrical resistance is not uniform over
the surface of the laminated composite plates having
0.2 wt% CNTs and 0.25 wt% CNTs. This indicates that there is
a window of CNT percentages that provides damage detection
and location monitoring with high sensitivity and uniform
distribution of electrical resistance. In particular, the
amount of CNTs ranging between 0.3 and 0.4 wt% CNTs, and
more particularly about 0.30 wt% CNTs, provides good damage
detection and location for glass fibers/epoxy composite
structures. It should however be understood that the optimal
amount of CNTs may vary depending on the particular fiber
and/or resin used. The
weight ratio depends on the
specifications of the CNTs. For instance, when longer CNTs
are used, the ratio of CNT may be smaller for a same
conductivity than a higher ratio with smaller CNTs. Put in
terms of resistance, the range of resistance for
electrically non-conductive fibers reinforced composite
- 26 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
27
structures between 103 ohm to 106 ohm provides good damage
detection and location capability. These
values are
provided simply as an example and therefore non-exclusively,
and other values are considered as well.
N0107]
Referring now to Fig. 3a, a system 200 for damage
detection and location in a composite structure, generally
identified as A, will now be described. As discussed above,
the composite structure A is made of fibers (electrically
conductive or electrically non-conductive) reinforced
polymer matrix composite materials containing CNTs. The
damage detection system 200, and method 100 described
hereinabove, are of the type that can be used during
operation of the composite structure (e.g., flight
conditions of an aircraft). The damage detection system 200
has a damage detection unit 202, and within the composite
structure A either a matrix of measurement points generally
shown at 204, featuring measurement points labeled 1-40 in
Fig. 4a, a matrix of measurement points 204', featuring two
sets of grid points as shown in Fig. 4b, or a network of
electrical lines 205 as shown in Fig. 4c.
[00108]
Referring to Fig. 4a in addition to Fig. 3a, each
matrix point 1-40 can be made of conductive silver paste,
silver paint, or similar materials. A thin
film of grid
points may also be used. Each point serves as an electrical
contact point, and thus as a measurement point. It is
desirable to precisely set the distance between the
measurement points. Indeed,
if the distance between the
measurement points is too large, the change in the
resistance between the two measurement points may not
reflect the occurrence of damage that has occurred in the
vicinity of the measurements points. If the
distance
between the measurement points is too small, a very large
number of measurement points would be required for a
structure of a certain size. Too many measurement points
- 27 -
CA 0213866 213106-11
WO 2013/086626
PCT/CA2012/001161
28
may render the technique impractical. The distance between
measurement points 1-40 may thus be selected as a function
of the composite structure A.
[00109] In the
illustrated embodiment, for the matrix 204
of measurement points having a first dimension dl of
thirteen (13) inches and a second dimension d2 of twenty-two
(22) inches, the distance between the measurements points,
e.g. the distance d3 between measurement points 4 and 5 or
the distance d4 between measurement points 5 and 10, is set
to three (3) inches (or 76.20mm). Such a spacing was found
to be more sensitive to changes in electrical resistance
between adjacent points, thus resulting in a more accurate
detection and location of damages on the composite structure
A. Electrical conductive wires 206 are further
illustratively attached to the measurement points 1-40 for
electrical measurements and to the damage detection unit 202
for data gathering. The damage detection unit 202 then
applies a constant source voltage by electrodes of the wires
206 mounted on the surface of the composite structure A.
The electrical current is measured by the damage detection
unit 202 to calculate electrical resistance.
[00110]
Referring to Fig. 4b in addition to Fig. 3a,
another arrangement of grid points is generally illustrated
at 204', for being used as part of the damage detection
system 200. The
arrangement 204' of Fig. 4b depicts a
representation of a plurality of sets of two grid points.
Each point of the grid can be made of conductive silver
paste or similar materials. Each point of a set of two grid
points serves as an electrical contact point. The first set
of grid points, e.g. points 11-401, which is geometrically
similar to the grid points of Fig. 4a, is mounted on the
composite structure A. However, in this case, the first set
of grid points is used to apply a constant electrical
current through the composite structure. The second set of
- 28 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
29
grid points, e.g. points 1V-40V, is diagonally shifted
(e.g., by 5 mm) with respect to the first set and is mounted
to measure electric potential. The electrical conductive
wires 206 are attached to the grid points to make
electrodes. In particular, the wires 206 attached to first
grid points 11 to 401 are used to apply a constant current
through the composite structure. The wires 206 attached to
the shifted grid points 1V to 40V are used to measure
electric potentials. The electric potentials across the
second grid points monitored by the damage detection unit
202 are used as input values. In the illustrated embodiment,
for the matrix 204' with a first dimension dl' of thirteen
(13) inches and a second dimension d2' of twenty-two (22)
inches, the distance between the grid points, e.g. distance
d3' between measurement points 41 and 51 or the distance d4'
between measurement points SV and by, is set to three (3)
inches.
prill] As
schematically illustrated in Fig. 4c, a network
205 of electrical lines may be used as an alternative to
matrix points 204 or 204'. The network 205 comprises a
plurality of electrical contact lines in a first, e.g.
vertical, orientation (X lines) and a
plurality of
electrical contact lines in a second, e.g. horizontal,
orientation (Y lines). The electrical lines are mounted as
electrical contacts on the surface of the structure A, to
measure the electrical resistance using a two-probe method
(described further below). In this manner, it becomes
possible to detect damages of large electrically non-
conductive fibers such as glass-fiber and kevlar-fiber
reinforced epoxy composite structures containing CNTs. The
electrical conductive wires 206 are bonded to these grid
lines to make electrodes for electrical measurements. The
electrical conductive wires 206 are further attached to the
damage detection unit 202 for data gathering. A constant
- 29 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
voltage may be directly applied by the damage detection unit
202 via the mounted electrodes on the surface of the
composite structure A and the electrical current is measured
by the damage detection unit 202 to calculate the electrical
resistance. In such a case, insulator material may be used
at the junction of the vertical and horizontal lines, to
create non-contacting nodes as in 208. In the illustrated
embodiment, with the network 205 having a first dimension
dl' of thirteen (13) inches and a second dimension d2' of
twenty-two (22) inches, the electrical lines are spaced
apart by a distance d3" of three (3) inches along each
direction.
[00112]
Referring back to Fig. 3a, the damage detection
unit 202 comprises an acquisition module 220 that will
obtain the signals (e.g. electrical resistance or potential)
between measurement points of the network 204 (or the
network 204'), or from the lines of the network 205. The
signals are obtained in a reference state (e.g., prior to a
first use, at a reset, etc.), and during use as well. The
reference values can be further stored for future comparison
with a new set of values when damage occurs. The acquired
measurements may then be sent to an analysis module 221,
which may be used to analyze the received information for
the purpose of detecting, locating, and determining the
severity of damages in the composite structure A. For this
purpose, the analysis module 221 illustratively comprises a
value comparator 222 communicating with a database 224, a
damage locator 226, and an output module 228.
[00113] The
acquisition module 220 may comprise any data
acquisition unit suitable for acquiring measurements of a
given electrical property, e.g. electrical resistance or
potential. The analysis module 221 may be implemented as
program code stored in a memory (not shown) and executable
by a processor (not shown) of a computer (not shown). The
- 30 -
CA 0213866 213106-11
WO 2013/086626
PCT/CA2012/001161
31
program code, when executed, illustratively causes real-time
and in-situ determination of severities, detections, and
locations of damages in composite structures as in A. The
processor may be any device that can perform operations on
data. Examples are a central processing unit (CPU) and a
microprocessor. The memory accessible by the processor may
received and store data. The memory may be a main memory,
such as a high speed Random Access Memory (RAM), or an
auxiliary storage unit, such as a hard disk or flash memory.
The memory may be any other type of memory, such as a Read-
Only Memory (ROM), Erasable Programmable Read-Only Memory
(EPROM), or optical storage media such as a videodisc and a
compact disc.
[00114] One or more databases 224 may be integrated
directly into the memory. The database 224 described herein
may be provided as collections of data or information
organized for rapid search and retrieval by a computer. The
database 224 may be structured to facilitate storage,
retrieval, modification, and deletion of data in conjunction
with various data-processing operations. The database 224
may consist of a file or sets of files that can be broken
down into records, each of which consists of one or more
fields. Database information may be retrieved through
queries using keywords and sorting commands, in order to
rapidly search, rearrange, group, and select the field. The
database 224 may be any organization of data on a data
storage medium, such as one or more servers.
[00115] For the network 204 of measurement points
(Fig. 4a), an electrical resistance measurement technique is
used to detect both internal and surface damages of large
non-conductive fibers, such as kevlar and glass fibers,
reinforced epoxy composite structures containing CNTs. The
electrical resistance measurement technique adopts the
electrically conductive CNTs themselves as networks of in-
- 31 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
32
situ sensors and hence requires inexpensive equipment to
operate. As discussed above, a constant source voltage may
be directly applied by the mounted electrodes on the surface
of the composite material and the electrical current is
measured and received at the acquisition module 220 to
calculate the electrical resistance. In order to detect the
severities and locations of damage in the large-scale
composite structures, grid points may be examined based on
the electrical resistance measurement.
[00116] The Electrical Resistance Change (ERC) between two
measurement points of Fig. 4a is expressed using the
following equation:
¨RR.
AR(%) = _____________________ x 100 (1)
R1,14
[00117] Where is the
electrical resistance between
points i and j before damage, or reference resistance; and
[00118] RF,1,3 is the electrical resistance between points
i and j during use, also referred to present resistance.
[00119] Using the measurements of electrical current
received from the acquisition module 220, the value
comparator 222 calculates the present values of the
electrical resistance. Using equation (1), the value
comparator 222 then computes the ERC for the purpose of
comparing present values to reference values retrieved from
the database 224 of any appropriate type. As discussed
above, the complete set of electrical resistances between
grid point pairs before damage establishes the reference
resistance values. These reference values are illustratively
stored in the database 224 to enable the value comparator
222 to effect the above-mentioned comparison with the set of
present values.
polm For refereeing purposes, the electrical
resistances between adjacent pairs of measurement points are
- 32 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
33
obtained using for instance a two-probe method. For the
array of Fig. 4a, examples of electrical resistances between
adjacent pairs of measurement points corresponding to the
first cell are:
R1,2, R1,6, R2,7, R6,7,
[00121] Where R stands for the electrical resistance
between two measurement points represented by the subscript
numbers.
A similar approach may be adopted for the grid lines of
network 205 (Fig. 4c). Indeed, for the grid line arrangement
205 of Fig. 4c, the Average Electrical Resistance Change
(AERC) for each box between a foursome of grid lines is
expressed using the following equation:
',J+1+AYw
AERC(96) - (2)
2
[00122] Where and A
Y3,j+1 are the electrical
resistance changes between two consecutive vertical
(X lines) and horizontal electrical contact lines (Y lines),
respectively, which are defined in the following equations:
X
(%) = __ y x
100 (3)
1,1,t+1
Y
AY11-0 (%) = ___ 1'4'1+1 14'1+1 x 100 (4)
[00123] Where i is the number of the electrical contact
line in a first orientation, e.g. vertical in Fig 4c;
[00124] j is the number of the electrical contact line in
a second orientation, e.g., horizontal in Fig 4c;
polzq x1,1 is the initial electrical resistance
measured between two consecutive vertical electrical contact
lines before damage, or reference resistance;
- 33 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
34
[00126] XF,i,õ1 is the final electrical resistance measured
between two consecutive vertical electrical contact lines
during use, or present reference;
[00127] 171,4.1 is the initial electrical resistance
measured between two consecutive horizontal electrical
contact lines before damage, or reference resistance; and
[00128] 17F,3,j,/ is the final electrical resistance measured
between two consecutive horizontal electrical contact lines
during use, or present reference.
[00129] The complete set of electrical resistances between
measurement point pairs before damage establishes the
reference resistance values, stored in the database 224,
after the acquisition of measurements of the electrical
current by the acquisition module 220 and computation of the
electrical resistance by the value comparator 222. The
reference of electrical resistance values can be stored for
future comparison by the value comparator 222 with a present
set of electrical resistance numbers when damage occurs. The
comparison may be done by computing the AERC for each box
between a foursome of grid lines using equations (2), (3),
and (4).
[00130] According to the embodiment of the grid point
arrangement shown in Fig. 4b, an electric potential
measurement technique may be used to detect both the
internal and surface damages of large conductive fibers such
as carbon-fiber reinforced epoxy composite structures
containing CNTs. For this purpose, the electric potentials
between adjacent pairs of second grid points are measured
using a four-probe method.
[00131] The electric potential measurement technique
adopts the electrically conductive carbon fibers and CNTs
themselves as networks of in-situ sensors. Since there are
two electrically conductive networks made by carbon fibers
and CNTs additives into the composite, two sets of grid
- 34 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
points need to be mounted on the surface of carbon-fiber
reinforced epoxy composite containing CNTs to eliminate the
contact resistance of measuring electrodes to the structure.
As discussed above, a constant source current may be
directly applied by the first grid points mounted to the
surface of the composite laminate and the electric potential
was measured across the second grid points, using the four-
probe method. The Electric Potential Change percentage (EPC)
between the shifted grid points (Fig. 4b) is expressed using
the following equation:
V -V
AV(%) = _______________
x 100 ( 5)
[00132] Where VI,iv,jv: is the initial electric potential
before loading, or reference electric potential, between
grid points iv and jv; and
polm VF,Iv,jv: is the final electric potential after
loading, or present electric potential, between grid points
iv and jv.
polIq For the rectangular array with two grids shown in
Fig. 4b, examples of electric potentials between adjacent
pairs of shifted grid points corresponding to the first cell
are: Vlv, 2v / Vlv, 6v/ V2v, 7v/ v6v,7\r/ \fly, 7v/ V2v, 6v/ where V stands
for electric potential between two grid points represented
by the subscripts. The subscript numbers, separated by a
space represent the associated grid points. The electrical
wires 206 from the first grid points can be connected to the
data detection unit 202 for the collection and storing of
the information. The complete set of electric potentials
between pairs of shifted grid points before damage
establishes the reference potential values. This reference
of electric potential values can be stored for future
comparison with a new set of electric potential numbers when
damage occurs.
- 35 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
36
[00135] When the
composite structure is in operation, the
detection unit 202 is turned on and a scan rate of the
values of the resistance or potential can be done at a given
interval of time by the acquisition module 220. The present
set of values of the electrical resistance or potential is
compared against the reference values by the value
comparator 222. The difference, e.g. (LIR) as in equation (1)
for the network 104 of Fig. 4a or (L17) as in equation (5)
for the network 204' of Fig. 4c, between the new values of
the resistance or potential between each pair of measurement
points and the reference values is determined by the value
comparator 222. The value comparator 222 will then determine
if the difference exceeds a threshold value. If the
difference exceeds the threshold value, the damage detection
unit 202 will identify the difference as an indication of
damage.
[00136] The
Average Electrical Resistance Change as in
equation (2) may be used for the network 205 of grid lines
(Fig. 4c) to determine the average change of resistance.
AX+1 and A 1113,3+1 are used to provide the change of
resistance between each set of adjacent lines. Again, if the
difference exceeds the threshold value, the damage detection
unit 202 will identify the difference as an indication of
damage.
[00137] The
value comparator 222 may also use the
difference in electrical resistance (or potential) to
quantify the damages. A correlation is made between the
measured difference and the magnitude of the damages.
[00138] The
damage locator 226 will then indicate where
the damage is in the composite structure A, by relating the
identified difference to the location of the set of
measurement points having provided the measurements,
relative to the composite structure A, in the case of the
- 36 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
37
network 204 of Fig. 4a or the network 204' of Fig. 4b. In
the case of the network 205 of grid lines shown in Fig. 4c,
the location of the damage will be obtained using the values
of AXu and A in the
AERC. For instance, the box
section having the highest value of AERC as in equation (2)
may indicate the location of damage.
[00139] In
particular, in locating the damage in
electrically non-conductive fibers reinforced polymer
composite structures containing CNTs, the damage locator 226
may, upon identifying that the difference exceeds the
threshold value, e.g. a tolerance of 0.0005, detect the
maximum electrical resitance changes of rows and columns in
the matrix of points as in 204 of Fig. 4a or the network 205
of grid lines of Fig. 4c. The damage locator 226 may then
find the row and column corresponding to the maximum
electrical resistance changes, thereby identifying the
location of the damage.
[00140] In order
to locate the damage in electrically
conductive fibers reinforced polymer composite structures
containing CNTs, the damage locator 226 may, upon
identifying that the difference exceeds the threshold value,
e.g. a tolerance of 0.0005, detect the maximum electrical
potential changes of rows and columns in the matrix of
points as in 204' of Fig. 4b. The damage locator 226 may
then find the row and column corresponding to the maximum
electrical potential changes, thereby identifying the
location of the damage.
[00141] The
output module 228 of the damage detection unit
202 may then produce an output with an indication of damage,
a location of the damage, and/or a quantification of the
damage. The output may be any appropriate signal, interface,
alarm or report that provides such information.
- 37 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
38
[00142] Fig. 3b
and Fig. 3c show examples of algorithms
that may be implemented by the analysis module 221 of Fig.
3a to detect and locate damages in a composite structure in
the manner described above. The algorithm of Fig. 3b may be
used for electrically non-conductive fibers reinforced
polymer composite structures containing CNTs while the
algorithm of Fig. 3c may be used for electrically conductive
fibers reinforced polymer composite structures containing
CNTs.
[00143] The data
detection unit 202 may be configured for
real-time and in-situ determination of detection, location
and severity of damages in composite structures,
respectively. As detailed above, the data detection unit 202
can be used for both non-conductive fibers, such as glass
fibers and kevlar fibers, and conductive fibers, such as
carbon fibers reinforced epoxy composite structures, by
monitoring electric resistance and electric potential,
respectively as input values for real-time and in-situ
determination of severities, detections and locations of
damage in composite structures.
[00144] The
damage detection method 100 and system 200
have the ability to detect and locate the damages in
composite structures in situ.
Moreover, the damage
detection method 100 and system 200 enable the location of
the damage, without effects from extraneous sources.
[00145] In an
embodiment, the spacing between the
measurement points of the network 104 is three (3) inches,
as discussed above. This is compared with the spacing of
0.25 inch in prior-art cases. The
large spacing is
preferred to avoid using an excessively large array or
network. The
larger spacing is possible when an optimal
value of the CNT is used in the resin of the composite
structure A, as this allows maximum sensitivity of the
change in electrical resistance to the occurrence of damage.
- 38 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
39
Another reason is due to the use of the voltage-resistance
measurement method, rather than current-voltage methods used
in prior-art references.
[00146] Many examples are provided to demonstrate the
application of the proposed technique for determination of
detection, location, and severities of damages in
electrically non-conductive and conductive fibers reinforced
epoxy containing CNTs. The experimental results for glass
fibers and carbon fibers reinforced epoxy composites
containing CNTs are presented in the following:
[00147] Exemplary embodiment: Electrically non-conductive
fibers reinforced epoxy containing CNTs composite structures
using grid points
[00148] As discussed above, when an electrically non-
conductive fibers, e.g. glass-fibers, reinforced epoxy
composite structure containing CNTs is in operation, the
system 200 can be turned on and a scan of the values of the
electrical resistance can be done at a certain interval of
time using two-probe electrical resistance measurement. The
new set of values of the electrical resistance is compared
against the reference values. The difference between the new
values of the electrical resistance between each pair of
grid points and the reference values is determined. This can
be recorded and displayed.
[00149] According to a simulation, damages have been made
in a few composite plates and the changes in electrical
resistance have been measured and recorded. Two types of
damages have been created. A first type was done by drilling
holes of different sizes at different locations in the
plate. A second type was done by impacts caused by the
collision with high velocity projectiles and drop weights.
Fig. 5a to Fig. 5f show the locations and values of the
changes in electrical resistance due to the drilling of
holes in 22 inch by 13 inch glass fibers reinforced epoxy
- 39 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
composite structures containing 0.30 wt% CNTs. Fig. 6 shows
the effect of hole size on the change in electrical
resistance. In Fig. 6, the numbers below the curve represent
the pairs of electrical contact points for the measurement
of the electrical resistance. This
pair of electrical
contact points is closest to the hole. The numbers above the
curve represent the change in resistance in percentage. A
clear correlation is observed between hole size and change
in electrical resistance. Accordingly, the technique can
also be used to indicate the severity of damages, in
addition to the detection and location capabilities.
[00150] Fig. 7a to Fig. 7f show the locations and values
of the changes in electrical resistance due to the collision
with high velocity projectiles, e.g. a 318 mg aluminum
particle travelling at 700 m/sec (78J). It can be seen that
the damages can be detected and located distinctly. Fig. 8a
to Fig. 8f show the locations and values of the changes in
electrical resistance due to the collision with low velocity
projectiles. The energy levels vary from 1 J to 10 J as
produced by drop-weight impact tests. Some of the damages at
the lower energy levels are barely visible to the naked eye.
These locations correspond to the location of the damages
created and barely visible damages zone. Fig. 9 shows the
correlation between the change in electrical resistance and
the energy level. Again the severity of damages can be
shown.
[00151] Simulations may also be performed with a kevlar
fibers reinforced epoxy composite structure containing CNTs
to reach similar conclusions as above with respect to the
glass-fibers reinforced epoxy composite structure.
[00152] Exemplary embodiment: Electrically non-conductive
fibers reinforced epoxy containing CNTs composite structures
using grid lines
- 40 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
41
[00153] As discussed above, a variation of the technique
using grid points is to use lines of contacts. In one
simulation, lines of conductive paints are drawn on the
surface of a glass fibers reinforced epoxy composite
structure containing 0.30 wt% CNTs. Fig. 10 shows the
average changes in electrical resistance between the lines
when holes of different sizes have been drilled in the
composite laminate. In particular, holes of sizes 1 (1/16
inch), 2 (2/16 inch), 3 (3/16 inch), 4 (4/16 inch), 5 (5/16
inch), and 6 (6/16 inch) have been respectively drilled at
locations A, B, C, D, E, and F. The results of Fig. 10 can
be compared to those of Fig. 5f, where holes of the same
sizes have been drilled in the composite laminate, although
at different locations thereon. It can be seen that the grid
points technique enables to achieve better damage location
than the grid lines technique for glass fibers reinforced
epoxy composite structure containing 0.30 wt % CNTs.
[00154] Exemplary embodiment:
Electrically conductive
fibers reinforced epoxy containing CNTs composite structures
using two sets of grid points
[00155] As was the case for non-electrically conductive
fibers reinforced epoxy composite structures, electrically
conductive, e.g. carbon fibers reinforced epoxy composite
structures containing CNTs may be tested using the system
200. In this case, the system 200 can be turned on and a
scan rate of the values of the electrical potential can be
done at a certain interval of time. The new set of values of
the electrical potential is compared against the reference
values and the absolute difference between the new values of
the electrical potential between each pair of grid points
and the reference values is determined.
[00156] Damages, e.g. drilling of holes of different sizes
at different locations in the plates or impacts caused by
the collision with high velocity projectiles, may be made to
- 41 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
42
22 inch by 13 inch carbon fibers reinforced epoxy composite
laminates containing 0.30 wt% CNTs. Fig. ha indicates
electrical potential distribution of the composite laminate.
The changes in electrical potential are then be measured and
recorded.
[00157] Fig. llb shows the locations and values of the
changes in electric potential due to the drilling of holes
of sizes 1, 2, 3, 4, 5, and 6. Fig 11c shows the effect of
the hole size on the change in electrical potential. As in
the case of electrically non-conductive fibers reinforced
epoxy composite structures containing CNTs, a correlation
may be observed between the hole size and the change in
electrical potential. Accordingly, the technique can also be
used to indicate the severity of damages, in addition to the
detection and location capabilities.
polsq Fig. lld shows the locations and values of the
changes in electric potential due to the collision with high
velocity projectiles for the carbon fibers reinforced epoxy
CNTs composite plate. It can be seen that damages due to the
collision with high velocity projectiles can be detected and
located distinctly in electrically conductive fibers
reinforced epoxy composite structures containing CNTs. Fig.
lie shows the locations and values of the changes in
electric potential due to the collision with low velocity
projectiles. This shows that some of the damages at lower
energy levels (e.g. 1 to 3J) are barely visible to the naked
eye.
[00159] The structure illustrated is provided for
efficiency of teaching the present embodiment. It should be
noted that the present invention can be carried out as a
method, can be embodied in a system, or on a computer
readable medium. The embodiments of the invention described
above are intended to be exemplary only. The scope of the
- 42 -
CA 02858866 2014-06-11
WO 2013/086626
PCT/CA2012/001161
43
invention is therefore intended to be limited solely by the
scope of the appended claims.
- 43 -
