Surface Crack Detection for Carbon Fiber Reinforced Plastic (CFRP ...

IEEE SENSORS JOURNAL, VOL. 11, NO. 12, DECEMBER 2011

3261

Surface Crack Detection for Carbon Fiber Reinforced Plastic (CFRP) Materials Using Pulsed Eddy Current Thermography Liang Cheng and Gui Yun Tian, Senior Member, IEEE

Abstract—There is currently a requirement in many industries to inspect carbon fiber reinforced plastic (CFRP) components, such as those used in aircraft and for wind turbine blades to identify issues leading to potential failures. To detect surface cracks, pulsed eddy current (PEC) thermography is proposed as a powerful inspection technique, allowing the operator to observe the heating developed from the eddy current distribution in a structure using infrared imaging, detecting defects over a relatively wide area within a short time (of the order of milliseconds). In this paper, a PEC thermography inspection system for CFRP materials is studied and optimized. Using the system, the directional electrical conductivity of the CFRP material is observed through the surface heating pattern. Then, the normalized temperature rise and decay are investigated through the inspection of notches with varied depths and widths. The position invariance of the coil with respect to the notch along the fiber direction is also studied in the experiments. The work shows that PEC thermography can be used for defect detection and characterization through analysis of the surface heating pattern and the transient temperature change. Index Terms—Carbon fiber reinforced plastic (CFRP), nondestructive testing and evaluation (NDT&E), pulsed eddy current (PEC) thermography, surface defects.

I. INTRODUCTION N RECENT decades, there has been an increasing interest in the use of composite materials, particularly carbon fiber reinforced plastic (CFRP), in the aerospace and renewable energy industries, because of the low weight and improved mechanical properties compared with metals. Components made from CFRP, such as wind turbine blades and aircraft fuselage, have to be tested for quality evaluation after manufacturing and monitored during in-service operation to increase the component lifetime. To accomplish this, nondestructive testing and evaluation (NDT&E) techniques are used. Ultrasonic testing is one of the most widely used methods for composite material inspection [1], [2] and has the advantage that it can detect defects in the interior of material. However, this method suffers from a number of disadvantages, including

I

the need for a couplant required for introducing acoustic waves, lack of sensitivity to shallow surface breaking defects, and large attenuation of acoustic waves when propagating through the multilayered structure of composites. Other NDT&E techniques have also been applied to composite material inspection, such as X-ray [3], acoustic emission [4], [5], eddy current [6], [7], and microwave [8], [9]. Different NDT&E techniques have different characteristics, but the majority of methods have limitations with regard to large-scale sensing, imaging, and comprehensive measurement or have safety issues. Consequently, the integration of NDT techniques to achieve improved performances has been implemented. The major advantage of thermography over other techniques is the potential for rapid inspection of a large area within a short time, though currently it is mostly applied to samples in the lab instead of in situ structures [10]. However, there is a tradeoff between detectable defect size and inspection area. Thermography is also applicable to a wide range of materials, including glass fiber, carbon fiber composites, and metallic materials, where specific excitation techniques are suitable for different applications. To inspect defects over a large scale and at large standoff distances, integration of thermography and other NDE approaches have been investigated [11]–[15], e.g., flash thermography [16]–[18], vibrothermography, sonic thermography [19], laser thermography [2], optical thermography [20], and pulsed eddy current (PEC) thermography or induction thermography [21]. Among the thermography techniques mentioned previously, PEC thermography, combining PEC and thermography, has its own advantages. For composite materials, varied excitation direction can be used in PEC thermography to investigate different layers, since the electrical and thermal conductivity is the greatest along the fiber orientation. In addition, the application of heat is not limited to the sample surface, such as in the flash thermography; rather, it can reach a certain depth, which is governed by the skin depth or penetration depth formula (1)

Manuscript received March 01, 2011; revised April 27, 2011; accepted May 07, 2011. Date of publication May 23, 2011; date of current version November 02, 2011. This work was supported by the Engineering and Physical Sciences Research Council (EPSRC), U.K., under Grant EP/F06151X/1. The PEC thermography system was developed through the RCNDE (Research Centre of NDE) supported by EPSRC in collaboration with Rolls Royce, Alstom and University of Bath. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Okyay Kaynak. The authors are with the School of Electrical, Electronic and Computer Engineering, University of Newcastle, NE1 7RU Newcastle upon Tyne, U.K. (e-mail: [email protected]; [email protected]). Digital Object Identifier 10.1109/JSEN.2011.2157492

where is excitation frequency, is the electrical conductivity, and is the permeability of the material under inspection. The transient of electromagnetic (EM) distribution and heat diffusion can derive a depth profile of defects [22], compared with surface-heating techniques. PEC thermography involves the application of high-current EM pulse to the conductive material under inspection for a short period (typically less than 1 s). When the eddy currents encounter a discontinuity (e.g., notches or delaminations), they are

1530-437X/$26.00 © 2011 IEEE

3262


II. THEORETICAL CONSIDERATIONS In this paper, an analytical model is established for the eddy current (EM) and heat diffusion phenomena. A conductive material is heated by Joule heating, which is caused by resistive heating from the eddy currents. The generated resistive heat is proportional to the square of the eddy current density or electric field intensity . The relationship between , , and is governed by (2) where is the electrical conductivity. According to Maxwell’s equations, the magnetic vector potential can be calculated from (3) Fig. 1. 2-D geometry setup for PEC thermography simulation.

Subsequently, the electric field intensity vector eddy current density can be derived via (1) and forced to divert, leading to regions of increased and decreased eddy current density. The “notch” in this paper is defined as a surface crack over the full width of the sample, but finite in depth and width, as shown in Fig. 1. For the notch case, eddy current diversion results in increased eddy current density at the notch bottom, where higher levels of Joule (ohmic) heating is achieved [23]. Thus, the defect can be identified from a characteristic heat distribution in the thermal image/video. After the period of eddy current heating, the notch also affects the heat diffusion in the cooling phase. Therefore, the mixed phenomena of induction heating dominating the heating phase and heat diffusion dominating the cooling phase and their specific behavior is used for the quantitative NDE (QNDE) of defects. PEC thermography has been used to mainly inspect metals in previous studies [23]–[25]. Oswald-Tranta and Wally [24] investigated the temperature distribution around a crack with different penetration depths using FEM modeling and experiments with metallic materials. The results showed the crack can be identified by lower temperatures at surface edges of the crack and higher temperatures at the crack bottom in nonmagnetic materials with a large penetration depth. Taking CFRP material as an example, the penetration depth is much larger than that of a metallic material. Therefore, in this study the temperature at the bottom of the notch will be investigated because it is higher than that at the edge of the notch. Ramdane et al. carried out 3-D numerical simulations and experiments on both metallic and composite materials [26]. The results showed that delaminations in CFRP could be detected using PEC thermography. However, the other types of defect in composite material were not investigated. In this paper, PEC thermography is proposed and extended from surface defect detection in metallic components to CFRP components, via numerical simulations and experiments. This paper is organized as follows. Section II describes the analytical model for eddy current and heat diffusion mechanisms; the simulation studies on how notch depth and width impacts on the transient temperature change are described in Section III. Experiments on directional conductivity inspection, the impact of notch depth on the transient temperature rise, and notch position invariance are reported in Section IV.

and the

(4) The heat conduction equation of a specimen caused by a Joule heating source is governed by (5) where , , and are density, heat capacity, and thermal conductivity, respectively. III. NUMERICAL SIMULATION STUDIES A. Simulation Setup In this study, CFRP samples (dimensions 100 mm 38 mm 6 mm) provided by Exel Composites UK are used in the experiments. In the simulation, the thickness and width of the sample are set the same as the experimental sample, while the length of the sample is reduced for faster calculation since there is no need to model over the whole length. The single notch in the sample with varied depth or width is simulated, as shown in Fig. 1. Notches with width (0.1, 0.5, 1, or 2 mm) and depth (0.5, 1, or 2 mm) are simulated. The excitation frequency and current are set as 256 kHz and 380 to match the experimental setup, which is introduced in Section IV. A rectangular coil is used in experiments. However, the eddy currents are induced dominantly by the coil edge which is close to the sample. Thus, only one edge of the rectangular coil is simulated, drawn as a cylindrical wire in Fig. 1. CFRP is an anisotropic material. Both carbon fiber and the matrix (normally epoxy resin) contribute the electrical and thermal properties. They vary in different directions due to the fiber orientation. The electrical conductivities in the longitudinal, transversal and cross-ply directions are set at , , and S/m [27]. The thermal diffusivity of the CFRP as measured in the longitudinal, transversal, and cross-ply directions are , , m s, respectively, at 20 C [28]. The density of CFRP is set as 1540 kg/m . The specific heat is set at 850 J/kg K from the COMSOL Multiphysics

CHENG AND TIAN: SURFACE CRACK DETECTION FOR CFRP MATERIALS USING PEC THERMOGRAPHY

3263

Fig. 2. Simulation results of thermal images at the maximum heating (200 ms) mm, d 2 mm notch. for w

=1

=

simulation package (COMSOL for short) library for granite (granite: carbon content in excess of 99%). To solve the EM and heat diffusion problem, COMSOL is used to perform finite-element method (FEM)-based simulation via the AC/DC module. The geometry of sample with notch defects in the numerical simulation is the same as that in the experiment. B. Influence of Notch Depth Fig. 2 shows the temperature distribution in the area around the notch at the maximum PEC-induced heating time of 200 ms. More heating is observed at the bottom of the notch due to the increased eddy current density in that area caused by eddy current diversion around the bottom of the defect. The interaction between the notch and uniform eddy currents in the sample shows that the eddy currents will always follow the path of least resistance. Hence, in a sample without a defect, they will be mainly confined to the surface layer, as defined by the skin depth formula (1), along the sample thickness. When a discontinuity, such as a notch, interrupts in the eddy current path, they will be forced to divert around the bottom of the notch, which leads to an area of increased eddy current density and a resultant hot spot at the bottom of the notch. The schematic representation of eddy current behavior around different types of defect is also discussed in previous work by Newcastle University and the University of Bath [29]. As the electric and thermal conductivity of CFRP vary in different directions. As a consequence, variation of the applied EM field orientation through changes in the excitation coil direction can be used for the investigation of different layers. Since conductivity is greatest in the fiber direction and fibers are orientated in different directions in different layers, changes in coil orientation will cause increased current flow in different layers. Besides, the electric conductivity of CFRP is much smaller than that of metal. Therefore, the skin depth at 256 kHz (around 9.95 mm through the sample thickness) is much larger than that of metal (normally less than 0.1 mm for steel). For metallic materials, the skin depth is normally smaller than the notch depth. A smaller skin depth causes the eddy currents to be concentrated at the surface and much denser at the surface edge of the notch. Conversely, for CFRP, the skin depth is normally larger than notch depth. A large skin depth results in eddy currents at the

Fig. 3. Simulation results for transient temperature against time at “pos 1” shown in Fig. 1 at notch bottom for varied notch depth at notch width w 1 mm: (a) normalized responses and (b) nonnormalized (raw) responses.

=

bottom of the notch being denser than that at the surface edge of the notch, because the notch blocks the surface current flow. Thus, the temperature rise at the bottom of the notch is higher than the other regions in the CFRP sample, as shown in the simulation studies. For cases where skin depth is larger than sample thickness at the excitation frequency used in simulation and experiment, a deeper notch causes more eddy current diversion at the bottom of the notch than the shallower notch. This effect is illustrated in the simulation results shown in Fig. 3, where the temperature rise at the 4-mm-deep notch is larger than that at the shallower notch. The investigated point in Fig. 3 is marked as “pos1” in Fig. 1. These results agree with experimental results discussed in the Section IV. Derived from (2) and (5) using Green’s function solution for a finite body, the estimation of temperature against time for an infinite length and finite thickness plate can be expressed with the following two equations, respectively, for the heating phase [25], [30] and for the cooling phase [31], [32]: for

for

(6)

(7)

where is the thermal diffusivity and is the distance from the evaluated point to the rear side of the sample.

3264


Fig. 4. Amplitude of temperature rise (a) versus notch depth d.

For a defect-free sample, is the thickness of the sample; for a shown in Fig. 1 for the estimation sample with a notch, in this study; is the duration of the heating pulse. From (6) and (7), we can use coefficients and to describe the thermal response. Coefficient is the amplitude of temperature rise determined by local electric conductivity change (in this case, the eddy current density at the bottom of the notch), while is the normalized temperature decay rate, determined by local thermal property changes and notch dimensions etc. From Fig. 3(b), it can be seen that deeper notch has a greater temperature rise (coefficient ). The relationship between coefficient and notch depth is illustrated in Fig. 4. It implies that a deeper notch will interact with and change the course of more eddy currents. Therefore, a deeper notch leads to larger temperature rise in the heating phase. However, after the normalization of the transient temperature change with time for the same width and varied depth notches, a convergence of the transient temperature change is obtained for the heating phase, as shown in Fig. 3(a). It shows that notch depth does not affect the normalized transient heating behavior. As for coefficient , it is clearer to investigate after the normalization of the temperature curve, as shown in Fig. 3(a). The results indicate the deeper notch has a faster temperature decay in the cooling phase, as larger (smaller , where ) results in larger value, which matches (7). In conclusion, when the notch width is fixed, a deeper notch leads to the higher temperature rise in the heating phase at the bottom of the notch (coefficient ) and a faster temperature decay in the cooling phase (coefficient ). Although depth information can be derived from transient temperature change with time in both the heating and cooling phases, the coefficients and can also be used for width investigation. C. Influence of Notch Width As the simulation results presented in the previous section closely agree with the experimental results, it is feasible to use simulations to predict the impact of notch width on thermal responses. The transient temperature changes against time for varied notch widths are shown in Fig. 5(b), where the figure indicates that the maximum amplitude of temperature change increases as becomes smaller. Unlike the notch depth influence, the variation of notch width not only changes the amplitude of eddy current density (as notch depth influence in the Section III-B), but also the eddy current distribution at the

Fig. 5. Simulation results for transient temperature against time at “pos 1” shown in Fig. 1 at notch bottom for varied notch width at notch depth d 2 mm: (a) normalized responses and (b) nonnomalized (raw) responses.

=

Fig. 6. Experimental setup.

bottom of the notch due to the notch geometry change. This implies that the narrower notch will force eddy currents to divert around a narrower area at the notch bottom. Thus, a narrower notch leads to a greater temperature rise [seen in Fig. 5(b)], as well as a greater rate of change in temperature in the early stages of the heating phase [seen in Fig. 5(a)]. As for coefficient , it is clearer to investigate after the normalization of the temperature curve, shown in Fig. 5(a). The results also indicate that the narrower notch has faster temperature decay in the cooling phase, as larger results in smaller value. IV. EXPERIMENTS A. Induction Heating System The experimental setup is illustrated in Fig. 6. An Easyheat 224 from Cheltenham Induction Heating is used for coil excita-


3265

Fig. 7. Thermal image of sample at maximum heating for (a) horizontal and (b) vertical current excitation.

tion. The Easyheat has a maximum excitation power of 2.4 kW, and an excitation frequency a maximum current of 400 and 256 kHz are used in the range of 150–400 kHz (380 experiments). In general, high excitation frequencies will lead to high thermal contrast (or high temperature rise). The time-domain information will allow the derivation of defect profile information. The system has a quoted rise time (from the start of the heating period to full power) of 5 ms, which was verified experimentally. Water cooling of coil is implemented to counteract direct heating of the coil. The SC7500 is a Stirling cooled camera with a 320 256 array of 1.5–5 m InSb detectors, shown in Fig. 6. The camera has a sensitivity of 20 mK and a maximum full frame rate of 383 Hz. The maximum 383-Hz frame rate provides one frame every 2.6 ms, with the option to increase the frame rate with windowing of the image. In our studies, a high-speed thermal camera is used for investigation including feature optimization for QNDE. However, in real applications, it may be unnecessary to use such a high-end camera. The rectangular coil is constructed from a 6.35-mm high-conductivity hollow copper tube. The coil is selected to introduce parallel eddy currents in the direction of maximum conductivity in the sample, as determined in Section IV-B. A 200-ms heating duration is selected for inspection, which is long enough to elicit an observable heat pattern around the notch. The rectangular coil is fixed and the sample can be moved. In this study, notches with different depths are placed at the same position with respect to the coil for each test. In addition, normalization of transient temperature change is applied, using division of the transient temperature by the temperature at maximum heating. Therefore, the liftoff influence can be eliminated. B. Directional Conductivity Experiment As CFRP exhibits directional conductivity, dependent on the fiber orientation in the composite, coil orientation has a large impact on experimental results. Thus, before inspecting the sample for defects, the directional (horizontal and vertical) conductivity is first ascertained. This allows optimization of the direction of

the applied field and notch direction to achieve the best temperature contrast between the regions with and without defects. Two coil directions with respect to the sample surface, horizontal [Fig. 7(a)] and vertical [Fig. 7(b)], are investigated. For horizontal coil orientation, the eddy currents are following in a horizontal direction, hence, the conductivity of Exel sample in horizontal direction is investigated in this case. For vertical coil direction, the conductivity of Exel sample in the vertical direction is investigated. Fig. 7(a) and (b) shows the thermal image in terms of digital level (DL) after 2 s of heating using horizontal and vertical coil directions, respectively. As a reference, the temperature changes at the coil in Fig. 7(a) and (b) are similar. From the comparison of these two excitation directions, it can be seen that the increase in temperature at the sample surface with the coil orientated vertically is much larger than when the coil is orientated horizontally. According to (2), it can be concluded that the conductivity in the vertical direction is much larger than that in the horizontal direction. Thus, it can be ascertained the fiber orientation is in the vertical direction, as the conductivity is greater along the fiber orientation. With awareness of the fiber orientation, the coil orientation is fixed in the vertical direction in the following experiments. C. Influence of Notch Depth PEC thermography was used to inspect the sample shown in Fig. 1. The 350 mm 38 mm 6 mm CFRP sample contains three notches with a width of 1 mm, varying in depth from 0.5 to 2 mm. The notches are manufactured in the workshop at Newcastle University. The sample was heated for 200 ms using the rectangular coil shown in Fig. 6. Images were acquired for a total of 500 ms (200 ms heating followed by 300-ms cooling) at the maximum frame data acquisition rate of 383 fps. Notches with different depths were inspected while retaining the same positional relationship between notch and coil. As an example the thermal image at the maximum heating time for a 2-mm-deep notch is shown in Fig. 8. The transient temperature change at the same point at the bottom of the notch and close to the coil, marked as “pos 1”

3266


Fig. 8. Experimental results of thermal image at the maximum heating (200 ms) for a 2-mm-deep and 1-mm-wide notch; unit: digital level.

Fig. 10. Varied distance between the coil and 2-mm-deep notch: excitation current in length direction: (a) 2-cm coil-notch distance and (b) 8-cm coil-notch distance. (c) Diagram of the position between the coil and the notch.

Fig. 9. Experimental results for transient temperature change against time at “pos 1” shown in Fig. 7 at notch bottom for w = 1 mm and varied depth notches: (a) normalized responses and (b) nonnormalized (raw) responses.

shown in Fig. 8, is investigated. The thermal responses for varied notch depths are shown in Fig. 9. From the comparison of the thermal responses at the investigated point for three notches shown in Fig. 9, it can be ascertained that the deeper the notch is, the greater the increase in temperature, because the eddy currents are concentrated at the bottom of the notch when notch depth is smaller than skin depth. The relationship between notch depth and transient temperature change from experimental results agrees with that from simulation results illustrated in Fig. 3(a) and (b). From Fig. 9, we can find the amplitude of the temperature rise (coefficient ) for a 2-mm-deep notch is the largest due to highest eddy current density at the bottom of the notch. From the comparison of by the normalized transient temperature behavior shown in Fig. 9(a), we can see that the temperature decay rate for the 2 mm deep

notch is the largest. Because in (7) for a 2-mm-deep notch is the smallest ( ), the value for the 2-mm-deep notch is the smallest among the three notches. Therefore, the notch depth can be discriminated by coefficients and . D. Notch Position Invariance Along the Fiber Direction The variation in thermal response with a varied distance between the notch and the coil along the direction of the fiber is investigated in this section. As the fiber orientation is identified in Section IV-A, in the experiment, the mutual position between the notch and the coil is changed by moving the coil, shown in Fig. 10(c). Thermal videos are captured at different notch positions with respect to the coil. A 1-mm-wide and 2-mm-deep notch is tested in this experiment. When the distance between the coil and the 2-mm notch is increased to 8 cm, heating of the notch can still be seen, but the temperature variation is less than one third of that for the 2-cm distance between the coil and the notch, shown in Fig. 10(a) and (b). To compare the influence of notch location, 0-, 2-, and 8-cm coil-notch distances are tested. The normalized


Fig. 11. Thermal response at notch bottom versus distance between coil and notch: (a) normalized responses and (b) nonnormalized (raw) responses.

and nonnormalized thermal responses at the notch bottom are shown in Fig. 11. It can be ascertained from the results that, the location of the notch only influences the amplitude of the temperature change in the heating phase, as seen in Fig. 11(b). The interaction between eddy currents and the notch can still be seen when the coil-notch distance increases to 8 cm. However, the attenuation of the eddy current leads to a reduction in the amplitude of the temperature change against the distance between coil and notch [Fig. 10(b)]. The temperature rise and decay rate after normalization is not affected, as shown in Fig. 10(a), because the notch shape and dimensions are not changed. Therefore, the transient temperature change with time at varied notch positions in both the heating and cooling phase is not changed. Unfortunately, the time delay of thermal or eddy current propagation from the region beneath the coil to the notch cannot be observed in thermal videos, because the propagation velocities of the thermal wave and eddy currents are in the order of 10 m/s and 10 m/s, respectively. The time delay of either thermal wave or eddy current is much shorter than the minimum detectable time interval from the thermal camera (2.6 ms).

V. CONCLUSION AND FUTURE WORK In this paper, a PEC thermography system has been proposed and, for the first time, implemented for notch detection in CFRP

3267

samples. The method allows the user to observe the eddy current distribution in a structure using infrared imaging and detect defects over a relatively wide area. Both numerical simulations and experimental investigations have been performed. Directional conductivity in CFRP makes the eddy current distribution different from metallic materials. The notch as a surface crack was detected and observed using PEC thermography through both simulation and experiment. It has been proven that the PEC thermography technique is feasible for surface defect detection in low conductivity composite materials and is not limited to the sample surface, such as in flash thermography. Through the simulation and experimental results shown in Figs. 2 and 8, respectively, it can be seen that the heat is mainly generated at the notch. The influences of notch depth, width, and position are investigated in terms of the amplitude of the temperature rise and transient temperature behaviors in the heating and cooling phases, respectively. The conclusions can be drawn as follows. • Notch depth: the relationship between the notch depth and transient temperature change shows that a deeper notch leads to a greater temperature rise at the bottom of the notch. A deeper notch results in a faster normalized temperature decay rate in the cooling phase, but the normalized transient temperature behavior in the heating phase is not affected. • Notch width: The results show that narrower notches lead to not only a greater temperature rise, but faster temperature rise and decay rate at the beginning of the heating phase and in the cooling phase, respectively. • Notch position invariance with fiber direction: the same notch at different locations with respect to the coil only influences the amplitude of the temperature change. The normalized heating and cooling transient temperature behaviors are not changed. Thus, based on the amplitude of the temperature rise, the heating and cooling thermal response, the notch width and depth can be determined for feature extraction and QNDE. In future work, further investigations on natural cracks will be undertaken in the future. In addition, work on subsurface defects like delaminations and impact damages rather than surface defects (notches) will be carried out and the relationship between delamination size, location and transient temperature change will be investigated for QNDE. Finally, the maximum standoff distance will also be investigated further. ACKNOWLEDGMENT The authors would like to thank Prof. S. Yang, Sichuan University, China, and Associate Prof. Y. Li, Xi’an Jiaotong University, China, for their useful discussions. In addition, the authors also would like to thank Exel Composites UK for providing the samples used in the experiments. REFERENCES [1] R. Stoessel, “Air-coupled ultrasound inspection as a new non-destructive testing tool for quality assurance,” Ph.D. dissertation, Faculty of Eng. Design, Production Eng., and Automotive Eng., Univ. of Stuttgart, Stuttgart, Germany, 2004. [2] S. E. Burrows, A. Rashed, D. P. Almond, and S. Dixon, “Combined laser spot imaging thermography and ultrasonic measurements for crack detection,” Nondestructive Testing and Evaluation, vol. 22, no. 2–3, pp. 217–227, Jun. 2007.

3268

ˇ [3] E. Jasi¯unien˙e, R. Raiˇsutis, R. Sliteris, A. Voleiˇsis, A. Vladiˇsauskas, D. Mitchard, and M. Amos, “NDT of wind turbine blades using adapted ultrasonic and radiographic techniques,” Insight, vol. 51, no. 9, pp. 477–483, 2009. [4] A. Ghoshal, W. N. Martin, M. J. Schulz, A. Chattopadhyay, W. H. Prosser, and H. S. Kim, “Health monitoring of composite plates using acoustic wave propagation, continuous sensors and wavelet analysis,” J. Reinforced Plastics Composites, vol. 26, no. 1, pp. 95–112, 2007. [5] M. Giordano, A. Calabro, C. Esposito, A. D’Amore, and L. Nicolais, “An acoustic-emission characterization of the failure modes in polymer-composite materials,” Composites Sci. Technol., vol. 58, no. 12, pp. 1923–1928, 1998. [6] A. Sophian, G. Y. Tian, D. Taylor, and J. Rudlin, “Electromagnetic and eddy current NDT: A review,” Insight Non-Destructive Testing and Condition Monitoring, vol. 43, no. 5, pp. 302–306, May 2001. [7] J. Bowler and M. Johnson, “Pulsed eddy-current response to a conducting half-space,” IEEE Trans. Magn., vol. 33, no. 3, pp. 2258–2264, May 1997. [8] S. Kharkovsky, J. T. Case, M. A. Abou-Khousa, R. Zoughi, and F. Hepburn, “Millimeter wave detection of localized anomalies in the Space Shuttle external fuel tank insulating foam,” IEEE Trans. Instrum. Meas., vol. 55, no. 4, pp. 1250–1257, Aug. 2006. [9] S. Ganchev, N. Qaddoumi, E. Ranu, and R. Zoughi, “Microwave detection optimization of disbond in layered dielectrics with varying thicknesses,” IEEE Trans. Instrum. Meas., vol. 44, no. 2, pp. 326–328, Apr. 1995. [10] N. P. Avdelidis, B. C. Hawtin, and D. P. Almond, “Transient thermography in the assessment of defects of aircraft composites,” NDT&E Int., vol. 36, pp. 433–439, 2003. [11] C. Ibarra-Castenedo, J.-M. Piau, S. Guilbert, N. P. Avdelidis, M. Genest, A. Bendada, and X. P. V. Maldague, “Comparative study of active thermography techniques for the nondestructive evaluation of honeycomb structures,” Res. Nondestructive Eval. J., vol. 20, no. 1, pp. 1–31, 2009. [12] N. P. Avdelidis and D. P. Almond, “Through skin sensing assessment of aircraft structures using pulsed thermography,” NDT & E Int., vol. 37, pp. 353–359, 2004. [13] N. P. Avdelidis, C. Ibarra-Castanedo, X. Maldague, N. P. Marioli-Riga, and D. P. Almond, “A thermographic comparison study for the assessment of composite patches,” Infrared Phys. Technol., vol. 45, pp. 291–299, 2004. [14] G. Busse and P. Eyerer, “Thermal wave remote and nondestructive inspection of polymers,” Appl. Phys. Lett., vol. 43, no. 4, pp. 355–357, 1983. [15] R. Mulaveesala and V. S. Ghali, “Cross correlation based approach for thermal non destructive characterization of carbon fiber reinforced plastics,” Insight, vol. 53, no. 1, pp. 1–3, 2011. [16] R. Mulaveesala and S. Tuli, “Theory of frequency modulated thermal wave imaging for non-destructive sub-surface defect detection,” Appl. Phys. Lett., vol. 89, no. 19, pp. 191913–191913-3, 2006. [17] Y. Y. Hung, Y. S. Chen, S. P. Ng, L. Liu, Y. H. Huang, B. L. Luk, R. W. L. Ip, C. M. L. Wu, and P. S. Chung, “Review and comparison of shearography and active thermography for nondestructive evaluation,” Mater. Sci. Eng. R, Rep., vol. 64, no. 5–6, pp. 73–112, May 2009. [18] G. F. Nino, T. J. Ahmed, H. E. N. Bersee, and A. Beukers, “Thermal NDI of resistance welded composite structures,” Composites B, Eng., vol. 40, no. 3, pp. 237–248, Apr. 2009. [19] M. Morbidini and P. Cawley, “The detectability of cracks using sonic IR,” J. Appl. Phys., vol. 105, no. 9, pp. 093530-1–093530-9, 2009. [20] C. M. Zöcke, “Quantitative analysis of defects in composite material by means of optical lockin thermography,” Dr. Ing. thesis, der Naturwissenschaftlischen Technischen Fakultät III, Chemie, Pharmazie, Bio- und Werkstoffwissenschaften, Universitat des Saarlandes und der Ecole Doctorale MIM, der Université Paul-Verlaine/Metz, Saarbrucker , 2009. [21] I. Z. Abidin, G. Y. Tian, J. Wilson, S. Yang, and D. Almond, “Quantitative evaluation of angular defects by pulsed eddy current thermography,” NDT & E Int., vol. 43, no. 7, pp. 537–546, Oct. 2010. [22] I. Z. Abidin, C. Mandache, G. Y. Tian, and M. Morozov, “Pulsed eddy current testing with variable duty cycle on rivet joints,” NDT & E Int., vol. 42, no. 7, pp. 599–605, 2009. [23] S. Yang, G. Y. Tian, I. Z. Abidin, and J. Wilson, “Simulation of edge cracks using pulsed eddy current stimulated thermography,” J. Dynam. Syst., Meas., Control, vol. 133, 2011.


[24] B. Oswald-Tranta and G. Wally, “Thermo-inductive surface crack detection in metallic materials,” in Proc. 9th Eur. Conf. NDT, Berlin, Germany, 2006, paper We.3.8.3. [25] M. Noethen, K.–J. Wolter, and N. Meyendorf, “Surface crack detection in ferritic and austenitic steel components using inductive heated thermography,” in Proc. 33rd Int. Spring Seminar Electron. Technol. (ISSE), 2010, pp. 249–254. [26] B. Ramdane, D. Trichet, M. Belkadi, and J. Fouladgar, “3-D numerical modelling of the thermo-inductive technique using shell elements,” IEEE Trans. Magn., vol. 46, no. 8, pp. 3037–3040, Aug. 2010. [27] H. Menana and M. Féliachi, “3-D eddy current computation in carbonfiber reinforced composites,” IEEE Trans. Magn., vol. 45, no. 3, pp. 1008–1011, Mar. 2009. [28] J. K. Humphrey, A. G. Gibson, J. Carruthers, and M. Tobinson, “Carbon fibre reinforced plastic (CFRP) thermal diffusivity measurements,” in Proc. Graduate Conf. Sch. Mech. Syst. Eng., Newcastle upon Tyne, U.K., 2010. [29] J. Wilson, G. Y. Tian, I. Z. Abidin, S. Yang, and D. Almond, “Modelling and evaluation of eddy current stimulated thermography,” Nondestr. Test. Eval., vol. 25, no. 3, pp. 205–218, 2010. [30] D. P. Pritzkau, “RF pulsed heating,” Ph.D. dissertation, Stanford Linear Accelerator Center, Dept. Appl. Phys., Stanford Univ., Stanford, CA, 2001. [31] W. J. Parker, R. J. Jenkins, C. P. Butler, and G. L. Abbott, “Flash method of determining thermal diffusivity, heat capacity, and thermal conductivity,” J. Appl. Phys., vol. 32, pp. 1679–1684, 1961. [32] H. S. Carlslaw and J. C. Jaeger, Conduction of Heat in Solids, 2nd ed. Oxford, U.K.: Oxford Univ. Press, 1959.

Liang Cheng was born in Beijing, China, in 1985. He received the B.Sc. degree in electronics from Peking University, Beijing, China, in 2007, and the M.Sc. degree in communications and signal processing from Imperial College London, London, U.K., in 2008. He is currently working toward the Ph.D. degree in failure models and life cycle assessment of wind turbine systems at Newcastle University, Newcastle upon Tyne, U.K. He is currently working on future reliable renewable energy conversion systems and networks: a collaborative UK-China project, which is in collaboration with five U.K. universities (Durham University, Edinburgh University, Newcastle University, Nottingham University, and Warwick University), U.K. industry, five Chinese universities, one research institute, and three Chinese organizations.

Gui Yun Tian (M’01–SM’03) received the B.Sc. degree in metrology and instrumentation and M.Sc. degree in precision engineering from the University of Sichuan, Chengdu, China, in 1985 and 1988, respectively, and the Ph.D. degree from the University of Derby, Derby, U.K., in 1998. He then became a Research Fellow and Senior Research Fellow with the University of Derby and the University of East Anglia, U.K. From 2000 to 2006, he was a Lecturer, Senior Lecturer, Reader, Professor, and Head of the group of Systems Engineering, respectively, with the University of Huddersfield, U.K. Since 2007, he has been based at Newcastle University, Newcastle upon Tyne, U.K., where he has been Chair Professor in Sensor Technologies and M.Sc. Program Director of Advanced Sensor Technology. Currently, he is Group Head of Communications and Signal Processing in the School of Electrical, Electronic and Computer Engineering, Newcastle University. He has coordinated several research projects from the Engineering and Physical Sciences Research Council (EPSRC), Royal Academy of Engineering and FP7, on top of this he also has good collaboration with leading industrial companies such as Airbus, Rolls Royce, BP, nPower and TWI among others. Dr. Tian is a Fellow of the Institution of Electrical Engineers, U.K., and InstNDT. He was the recipient of the John Grimwade Award from the British Institute of Non-Destructive Testing. He is also a Chinese Changjiang scholar and as well as being on several editorial boards of international journals.