Temperature variation and damage characteristic of impacted CFRP laminate using infrared thermography: Experimental investigation

Accepted Manuscript Temperature variation and damage characteristic of impacted CFRP laminate using infrared thermography: Experimental investigation Yin Li, Wei Zhang, Lei Li, Pan-pan Bao, Zheng-wei Yang, Gan Tian, An-bo Ming PII: DOI: Reference:

S0142-1123(18)30093-8 https://doi.org/10.1016/j.ijfatigue.2018.03.009 JIJF 4610

To appear in:

International Journal of Fatigue

Received Date: Revised Date: Accepted Date:

25 December 2017 5 March 2018 7 March 2018

Please cite this article as: Li, Y., Zhang, W., Li, L., Bao, P-p., Yang, Z-w., Tian, G., Ming, A-b., Temperature variation and damage characteristic of impacted CFRP laminate using infrared thermography: Experimental investigation, International Journal of Fatigue (2018), doi: https://doi.org/10.1016/j.ijfatigue.2018.03.009

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Temperature variation and damage characteristic of impacted CFRP laminate using infrared thermography: experimental investigation Yin Lia, Wei Zhanga, Lei Lib, Pan-pan Baob, Zheng-wei Yanga, Gan Tiana, An-bo Minga a

Xi’an Research Inst. of Hi-Tech, Xi’an 710025, PR China

b

Aircraft Strength Research Institute, Xi’an, 710065, PR China

Abstract Infrared thermography is viewed as a promising method to investigate fatigue behavior and the objects studied in the available work are mainly focused on the metals and glass fiber reinforced plastic (GRFP), while the carbon fiber reinforced plastic (CFRP), especially the CFRP with initial impact damage, are rarely involved yet. In this work, experimental investigation on temperature variation and damage characteristic of impacted CFRP laminate is conducted. Several specimens are begun with impact test and compression static test, followed by compression-compression fatigue test. Then, real-time maximum surface temperature difference, ΔTmax, is defined for investigating the temperature variation during fatigue and the obtained result

shows

that

the

evolution

of Δ Tmax shows

three

stages

in

a

“rapidly-slowly-rapidly” manner. On this basis, a method to determine the cut-off point is proposed to facilitate the division of the three stages, by which the three stages occupy less than 5%, more than 85% and less than 10% of total fatigue life respectively. Finally, the damage characteristic during fatigue is investigated. the obtained results show that the main damage characteristic changes in a “matrix

crack-delamination-fiber rupture” manner as ΔTmax proceeds. Keywords Impacted CFRP; compression-compression fatigue test; temperature variation; damage characteristic; infrared thermography 1. Introduction In the past several decades, composite materials, especially the carbon fiber reinforced plastic (CFRP) laminate, have been widely used in the aircraft owning to their excellent mechanical performance such as high specific strength and specific stiffness over many other types of materials [1]. However, the CFRP laminate in service is unavoidably subjected to the fatigue cycle loading, resulting in the degradation of performance [2]. Additionally, the CFRP laminate is extremely sensitive to the impact; even the low energy impact loading may cause such initial damages as matrix crack, delamination, and even fiber breakage within the CFRP laminate [3, 4]. Therefore, it is vital to explore the fatigue behavior of the CFRP laminate within the initial impact damage to ensure the structures to satisfy the demands of engineering application. Many techniques have been employed to investigate the fatigue behavior, among which the most common methods are based on residual strength and residual stiffness [5-7]. Actually, the fatigue is an irreversible process, accompanied by the dissipation of energy; as a result, the surface temperature of structures in the fatigue process will constantly increase. Thus, many scholars attempted to investigate the fatigue behavior based on the changeable surface temperature. Infrared thermography (IR) technology

is viewed as a promising method to in-situ measure temperature with the advantages of full-field, non-destruction and non-contact [8-11] and has been gradually applied to investigate the fatigue behavior, which mainly contains the following three aspects: damage evolution in the fatigue process, determination of fatigue limit and evaluation of residual fatigue life (RLF). In terms of damage evolution, Ahsan Mian et al. [12] used the ultrasonic IR technology to detect the fatigue damage under different fatigue cycle numbers and preliminarily explored the damage evolution by analyzing the vibration displacement of node, which was selected on the specimen surface. To deeply investigate the damage evolution of composite materials, M.M. Khonsari et al. [13] utilized IR and acoustic emission (AE) technologies to analyze the damage evolution of G10/FR4. The obtained results show that the damage evolution mainly contains three stages, which occupy the 10~20%, 70~75% and 10~20% of total fatigue life respectively. Moreover, they also explored the damage evolution by analyzing the entropy production [14], heat dissipation, cyclic hysteresis energy etc. [15]. Additionally, Fan et al. [16] has successfully proven that the IR technology can be used for the damage evolution in the metal material and revealed the metal fracture mechanism. In terms of fatigue limit, Risitano et al [17] supposed that the changeable temperature was generated by fatigue loading owning to the intrinsic energy dissipative mechanisms and proposed an empirical method, i.e. one-curve method (OCM), to determine the fatigue limit of materials based on the temperature increase data and the corresponding various stress levels. On this basis, involving the intrinsic energy

dissipation both under low fatigue loading and high fatigue loading, Luong et al [18] proposed two-curve method (TCM), but this method has a difficulty of determining the fitting cut-off point. To solving this problem, Cura et al [19] proposed an iteration method to determine the fatigue limit. Based on these three methods, the temperature data obtained by IR technology have been used to determine the fatigue limit of such materials as composites [8, 20] and metals [21-23]. In terms of RLF, Fan et al [24] proposed an evaluation model of RLF based on energetic parameter, and achieved the RLF evaluation of Q235 steel with the error of 10%-15%. According to a series of fatigue results, M.M. Khonsari et al [25] pointed out that there existed a relation between the initial slope of temperature rise and fatigue cycles Nf, and on this basis, developed an empirical method to evaluate the RLF. But this method is only suitable for the specimens with no prior fatigue damage. As a result, they [26] improved this method to be propitious to the specimens with prior fatigue damage and summarized the procedures for the RLF evaluation. Afterwards, the improved method was successfully applied to predict RLF of composites [27] and metals [28] with excellent accuracy. From the above analysis, the IR technology has already become a very great method to investigate the fatigue behavior of materials. However, to the best of my knowledge, the objects studied in the available references were focused on the metals and glass fiber reinforced plastic (GRFP), while the CFRP, especially the CFRP with initial impact damage, has been rarely involved yet. Therefore, this work attempts to use IR technology to investigate the fatigue behavior of impacted CFRP laminate,

emphasizing the terms of damage evolution. This work is of importance as it paves the way for the RLF prediction of impacted CFRP using infrared thermography, which is a subject of future investigation. 2. Theory background 2.1. Thermomechanical coupling effect As indicated, the fatigue is an irreversible process accompanied by energy dissipation, which macroscopically leads to the temperature increase on the specimen surface. Combining the first and the second thermodynamic principle, the thermomechanical coupling equation [29] can be expressed as follows.

CT  k 2T =d1  sthe  sint  qe

(1)

With

  ):     

(2)

 2 : T 

(3)

 2 sint  T  t

(4)

d1  ( -

sthe  T

Where  is the density, C is the specific heat, k is the heat conduction coefficient, T is the temperature,  is the Cauchy stress tensor,  is the strain tensor,  is the internal variable describing the change of microstructure within the materials,  is the Helmholtz free energy and qe is the external heat source that can be removed using subtracting method. In Eqs. (1)-(4), d1 is the intrinsic dissipation caused by the irreversible plastic deformation and represents the heat generation of the specimen, resulting in the

increasing mean temperature on the surface of materials, i.e. the accumulated damage in the materials; sthe is the thermoelastic coupling effect between the temperature and the strain, which is a reversible conversion process and causes temperature fluctuation on the surface of specimen, but not leads to the accumulated damage, sint is the other thermomechanical coupling between temperature and internal variables, which is generally ignored. On the basis of the hypotheses proposed by T. Boulanger [30], the thermomechanical coupling equation can be simplified as follows.

C  k 2 =d1  sthe

(4)

Where θ=T-T0 is the temperature variation. From the above analysis, the damage accumulation in the fatigue process is mainly influenced by the intrinsic dissipation and thermoelastic coupling effect, and is characterized by the temperature evolution macroscopically, which paves the way for the analysis of fatigue behavior using temperature data. 2.2. Temperature measurement using IR From the infrared radiation framework, an object with temperature can radiate energy outward. The surface temperature of specimen will vary under fatigue loading due to the thermomechanical coupling effect. As a result, the specimen radiates the variable energy, which can be measured using IR technology. On this basis, the temperature can be obtained by Kirchoff’s law, as shown in Eq. (5).

W  E T 4

(5)

Where W is the radiant energy of an object, E is surface emissivity, δ is Stefan-Boltzmann constant and T is the surface temperature.

3. Experiment To investigate the temperature evolution of impacted CFRP laminate, such experiments as low energy impact, static compression and compression-compression fatigue are conducted in the following section. 3.1. Materials and experimental devices The specimen studied in this work, made of CFRP laminate, is stacked in 24 layers and extended in [450/00/-450/900]3s with the size of 150mm×100mm×4mm. The drop hammer impact testing device, as shown in Fig. 1, is used to conduct the low energy impact. The impact bar is made of stainless steel and the shape of impact tip is hemispheroidal with the diameter of 16mm. The specimen clamp is applied to fix the specimen and to prevent multiple impacts. The electro-hydraulic servo universal machine in Fig. 2 is applied to conduct the static test and fatigue test, which can impose the static / dynamic force on the specimen with the range of -100 kN to 100 kN. During the fatigue test, the infrared camera produced by Infra Tec, with the type of Vhr 680, is positioned about 20 cm from the specimen surface and applied to full-field monitor the specimen to gather the temperature data, which possesses such excellent performances as wide range of measurable temperature from -40℃ to 1200℃, high sensitivity of 0.03℃ at 30℃ and reading accuracy of 2%.

Fig. 1. The drop hammer impact testing device.

Fig.2. Electro-hydraulic servo universal machine and infrared camera. 3.2. Experimental details As indicated, this work focuses on the CFRP laminate with initial impact damage. For

the sake of simulating the natural impact damage in the CFRP laminate, the drop hammer impact device in Fig. 1 is applied to carry out impact test. The impact energy can be adjusted by changing the height of impact tip. In this work, all specimens are subjected to the impact test with the energy of 36 J. Afterwards, the static, displacement-controlled compression test is conducted to characterize the failure force under compression with ramp of displacement of 2 mm/min. In this work, three impacted specimens are randomly selected to conducted static compression test using the devices in Fig. 2 to obtain the ultimate compression strength (UCS), which is used to design the fatigue test of the specimens. Subsequently, the fatigue test is conducted at room temperature using the devices in Fig. 2. The tested specimen is fixed by the clamp, which is placed between the top grip and bottom grip. A series of compression-compression, load-controlled fatigue tests is performed and the constant sinusoidal fatigue loadings are applied at the frequency of 5 Hz and loading ratio R of 0.1, defined as the ratio of minimum to maximum load. In the fatigue test, the bottom grip is kept fixed and the sinusoidal fatigue load is imposed on the specimen through the top grip; meanwhile, the infrared camera is used for full-field monitoring the surface of specimens. Involving the data storage, the operator is selective in saving some thermographic images and gathering the temperature data. Summarily, the fatigue test details are listed in Table 1. Table 1. Fatigue test details. Specimens

Maximum stress δmax

Loading frequency f

Loading ratio R

1#

80% UCS

5 Hz

0.1

2#

80% UCS

5 Hz

0.1

3#

75% UCS

5 Hz

0.1

4#

70% UCS

5 Hz

0.1

4. Temperature analysis 4.1. Temperature distribution on the surface of specimen Fig. 3 shows the thermographic image during the fatigue test, in which the temperature presents different distributions on the surface of the specimen. Six sections with serial numbers of P1-P6 are adopted and the mean temperature of these sections is extracted respectively. The temperature in P1 zone, i.e. the impact zone, shows the maximum duo to the stress concentration in the fatigue test, while the temperature in P6 zone near the bottom grip shows the minimum. For the sake of characterizing the temperature variation caused by the intrinsic dissipation and thermoelastic coupling effect and eliminating the influence caused by environmental interferences, the real-time maximum surface temperature difference, which is defined as ΔTmax, i.e. the temperature difference between the P1 zone and the P6 zone under the same of fatigue cycle, is introduced to analyze the temperature variation. For example, for certain fatigue cycle, i, ΔTmax can be obtained by Eq. (6) and (7). Tmax,i  Tmax,i  Tmin,i

(6)

Tmax,i  TPavg 1,i  avg Tmin,i  TP 6,i

(7)

With

Where the subscript i represents the fatigue cycles, Tmax,i and Tmin,i represent the

average surface temperature in P1 zone and the average temperature in P6 zone at the fatigue cycle of i, respectively.

Fig. 3. Temperature distribution on the specimen surface during fatigue. 4.2. Temperature variation during fatigue To directly analyze the temperature variation of the tested specimens, the temperature data of the tested specimens with the numbers of 1#, 2#, 3# and 4# are gathered. Fig. 4 shows theΔTmax versus Nf of the tested specimens, from which we can find that the Nf increases with the decrease of the maximum stress amplitude δmax under the condition of the same loading frequency f and loading ratio R. In Fig. 4 a and b, although the specimens 1# and 2# with the same fatigue condition of δmax=80% UCS, f=5Hz, R=0.1, the Nf is different due to the material characteristics of specimen, clamp, operator, etc. in a complex manner, demonstrating the dispersion of fatigue test results,

which can be also obtained by fatigue results of specimens 3# and 4# in Fig. 4 c and d. Moreover, it is found that there is a phenomenon of temperature jump, i.e. the ΔTmax abruptly increases to about 20℃ marked by orange loop at the moment of ultimate rupture of specimen due to the occurrence of fibers rupture to release abundant heat. Additionally, thought the Nf presents different under different fatigue conditions, the

ΔTmax always presents three stages, i.e. the ΔTmax increases rapidly-slowly-rapidly till to ultimate rupture of specimens.

(a)

(b)

(c)

(d) Fig. 4. Variation of maximum surface temperature difference of the specimens. (a). 1#. (b). 2#. (c).3#. (d). 4#. 4.3. Determination of COPs For the sake of further investigating the three stages ofΔTmax regarding the impacted CFRP, an accompanying problem should be focused, i.e. how to exactly determine the cut-off points (COPs) between the stage I and stage II and between stage II and stage III as shown in Fig. 5 to facilitate the division of three stages. However, the three stages are recently divided using visual inspection with uncertainty. Thus, a method is proposed to determine the COPs in this section. Taking the specimen 4# as the example, the method is described in detail.

Fig. 5. The sketch map of COPs. From Fig. 5, ΔTmax increases rapidly-slowly-rapidly in the three stages, in other words, the slope of line in stage I and stage III is larger than that in stage II. Thus we can deduce that if theΔTmax proceeds from stage I to stage II, the slope may become small and smooth, while theΔTmax proceeds from stage II to stage III, the slope may become large and abrupt. On this basis, the procedures to determine the COPs are listed in the following three steps. Step 1: according to the curve of ΔTmax versus fatigue cycles, we can visually estimate the transition zone from stage I to stage II and that from stage II to stage III and extract these points in the transition zone, as shown in Fig. 6.

Fig. 6. ΔTmax versus fatigue cycles in the transition zones. (a). The whole fatigue process. (b). Transition from stage I to stage II. (c). Transition from stage II to stage III. Step 2: number the points in the transition zones as (N1, ΔTmax,1), (N2, ΔTmax,2), …, (Nj, ΔTmax,j) and calculate the slope of neighbouring points using Eq. (8)

T

max, j



Tmax, j 1  Tmax, j N j 1  N j

(8)

Step 3: according to the results obtained by step 2, as shown in Fig. 7, determine the first cut-off point (FCOP), from which the slope become small and smooth, and the second cut-off point (SCOP), from which the slope become large and abrupt.

(a)

(b) Fig. 7. Determination of COPs. (a). FCOP. (b). SCOP.

The COPs of the specimens are determined via the above three steps and the corresponding results are listed in table 2. For the specimen 1#, the stage I and stage III are both limited to a low fatigue cycle numbers, occupying the 4.67% (i.e. and 3.74% (i.e.

1000 ) 21401

21401  20600 ) of total fatigue life respectively, thus the stage II 21401

occupies 91.59% of total fatigue life. Similarly, for the specimens 2#-4#, the percent of the three stages are obtained in table 3. Table 2. The COPs of specimens. Specimens

Nf

FCOP (N, ΔTmax)

SCOP (N, ΔTmax)

1#

21401

(1000, 2.07)

(20600, 8.25)

2#

11777

(500, 2.48)

(10800, 8.35)

3#

63581

(2600, 1.55)

(60800, 5.15)

4#

77962

(2000, 1.92)

(72000, 5.14)

Table 3. The percent of three stages of specimens. Specimens

Nf

Stage I

Stage II

Stage III

1#

21401

4.67%

91.59%

3.74%

2#

11777

4.25%

87.45%

8.30%

3#

63581

4.09%

91.54%

4.37%

4#

77962

2.57%

89.78%

7.65%

From the above analysis, ΔTmax evolution versus Nf can be summarized in Fig. 8. For the impacted CFRP in this work, theΔTmax evolution shows three stages during fatigue, which is in agreement with that of the fresh specimen in the literature [13], whereas the stage I, stage II and stage III occupy less than 5%, more than 85%, and

less than 10% respectively.

Fig. 8. Δ Tmax evolution of impacted CFRP with the energy of 36J during compression-compression fatigue at the frequency of 5Hz and loading ratio of 0.1. 5. Damage characteristics To explore the damage characteristics during compression-compression fatigue test, the nondestructive detection of the impacted specimen is carried out before fatigue test and the corresponding result is shown in Fig. 9, which indicates that there already is matrix crack caused by the impacted test. So theΔTmax increases rapidly in stage I owning to the friction between cracked matrixes to generate heat. As fatigue test proceeds, the thermographic images are gathered as shown in Fig. 10. The hot spot on the specimen surface gradually extends, demonstrating that the internal delamination increases. In this case, the friction induced heat decreases and the ΔTmax increases

slowly in stage II. In stage III, the pinnacle marked by black rectangle appears in the thermographic image, as shown in Fig. 11 (a), indicating the fibers rupture is occurring, which can be validated by the results of scanning electron microscopy (SEM), as shown in Fig. 11 (b). Fig. 11 (c) shows the profile of specimen marked by white loop after fatigue failure, which is similar to the temperature profile. In this stage, the fiber rupture releases abundant heat, leading to the rapid increase of ΔTmax.

Matrix crack Fig. 9. The detection result of impacted specimen.

Fig. 10. The thermographic image sequence in stage II.

Fig. 11. The thermographic image and damage characteristic in stage III. (a). The thermographic image in stage III. (b). The fibers rupture obtained by SEM. (c). The profile of specimen after fatigue failure. From the phenomenological standpoint, for the impacted CFRP, there is not only evolution ofΔTmax, but also evolution of damage characteristics during fatigue, which is

summarized

in

Fig.

12.

Totally,

the

evolution

of

Δ Tmax presents

“rapidly-slowly-rapidly” and the corresponding main damage characteristic changes in a “matrix crack-delamination-fiber rupture” manner during fatigue.

Fig. 12. The evolutions of ΔTmax and damage characteristics during fatigue. 6. Conclusion In this work, temperature variation and damage characteristic of impacted CFRP during fatigue are investigated. The specimens are begun with the impact test and static compression test, followed by the compression-compression fatigue test. Afterwards, the temperature data and thermogrphic image sequence are analyzed, and the corresponding conclusions are listed as follows. (1) Real-time maximum surface temperature difference, Δ Tmax, is defined to investigate the temperature variation during fatigue, which presents three stages, i.e. theΔTmax increases rapidly-slowly-rapidly during fatigue. Additionally, there exists a phenomenon of temperature jump at the moment of ultimate rupture of specimen. (2) A method to determine the COPs is proposed to facilitate the division of the three

stages. On this basis, the stage I, stage II and stage III occupy less than 5%, more than 85%, and less than 10% of total fatigue life respectively. (3) Combined with nondestructive detection method and SEM, the evolution ofΔTmax corresponds to the change of main damage characteristic. In other words, the evolution of ΔTmax presents “rapidly-slowly-rapidly” and the corresponding main damage characteristic changes in a “matrix crack-delamination-fiber rupture” manner during fatigue. Acknowledgments This work was supported by National Natural Science Foundation of China (Grant nos. 51575516 and 51605481). References [1]

Soutis C. Fibre reinforced composites in aircraft construction [J]. Progress in Aerospace Sciences, 2005, 41(2):143-151.

[2]

Degrieck J, Van Paepegem W. Fatigue damage modeling of fibre-reinforced composite materials: Review [J]. Applied Mechanics Reviews, 2001, 54(4):279-299.

[3]

M.O.W. Richardson, M.J. Wisheart. Review of low-velocity impact properties of composite materials [J]. Composites Part A: Applied Science and Manufacturing, 1996, 27(12):1123-1131.

[4]

Shyr T W, Pan Y H. Impact resistance and damage characteristics of composite laminates [J]. Composite Structures, 2003, 62(2):193-203.

[5]

Lee L J, Yang J N, Sheu D Y. Prediction of fatigue life for matrix-dominated

composite laminates [J]. Composites Science & Technology, 1993, 46(1):21-28. [6]

Yang J N, Liu M D. Residual strength degradation model and theory of periodic proof tests for Graphite/Epoxy laminates [J]. Journal of Composite Materials, 1977, 11(2):176-203.

[7]

AIAA. A Stiffness Degradation model for Graphite/Epoxy laminates [J]. Journal of Composite Materials, 1990, 24(7):753-769.

[8]

Montesano J, Fawaz Z, Bougherara H. Use of infrared thermography to investigate the fatigue behavior of a carbon fiber reinforced polymer composite [J]. Composite Structures, 2013, 97(2):76-83.

[9]

Toubal L, Karama M, Lorrain B. Damage evolution and infrared thermography in woven composite laminates under fatigue loading [J]. International Journal of Fatigue, 2006, 28(12):1867-1872.

[10]

Minh Phong Luong. Fatigue limit evaluation of metals using an infrared thermographic technique [J]. Mechanics of Materials, 1998, 28(1–4):155-163.

[11]

Schmidt F, Rheinfurth M, Protz R, et al. Monitoring of multiaxial fatigue damage evolution in impacted composite tubes using non-destructive evaluation [J]. Composites Part A, 2012, 43(3):537-546.

[12]

Mian A, Han X, Islam S, et al. Fatigue damage detection in graphite/epoxy composites using sonic infrared imaging technique [J]. Composites Science & Technology, 2004, 64(5):657-666.

[13]

Naderi M, Kahirdeh A, Khonsari M M. Dissipated thermal energy and damage evolution of Glass/Epoxy using infrared thermography and acoustic emission

[J]. Composites Part B Engineering, 2012, 43(3):1613-1620. [14]

Naderi M, Khonsari, M.M. A thermodynamic approach to fatigue damage accumulation under variable loading [J]. Materials Science & Engineering A, 2010, 527(23):6133-6139.

[15]

Naderi M, Khonsari M M. On the role of damage energy in the fatigue degradation characterization of a composite laminate [J]. Composites Part B, 2013, 45(1):528-537.

[16]

Fan J L, Guo X L, Wu C W, et al. Research on fatigue behavior evaluation and fatigue fracture mechanisms of cruciform welded joints [J]. Materials Science & Engineering A, 2011, 528(29):8417-8427.

[17]

Rosa G L, Risitano A. Thermographic methodology for rapid determination of the fatigue limit of materials and mechanical components [J]. International Journal of Fatigue, 2000, 22(1):65-73.

[18]

Luong M P. Infrared thermographic scanning of fatigue in metals [J]. International Journal of Fatigue, 1995, 19(3):266-266.

[19]

Curà F, Curti G, Sesana R. A new iteration method for the thermographic determination of fatigue limit in steels [J]. International Journal of Fatigue, 2005, 27(4):453-459.

[20]

Peyrac C, Jollivet T, Leray N, et al. Self-heating method for fatigue limit determination on thermoplastic composites [J]. Procedia Engineering, 2015, 133:129-135.

[21]

Guo S, Zhou Y, Zhang H, et al. Thermographic analysis of the fatigue heating

process for AZ31B magnesium alloy [J]. Materials & Design, 2014, 65(30):1172-1180. [22]

Munier R, Doudard C, Calloch S, et al. Determination of high cycle fatigue properties of a wide range of steel sheet grades from self-heating measurements [J]. International Journal of Fatigue, 2014, 63(63):46-61.

[23]

Zhang H X, Wu G H, Yan Z F, et al. An experimental analysis of fatigue behavior of AZ31B magnesium alloy welded joint based on infrared thermography [J]. Materials & Design, 2014, 55(6):785-791.

[24]

Fan J, Guo X, Wu C. A new application of the infrared thermography for fatigue evaluation and damage assessment [J]. International Journal of Fatigue, 2012, 44:1-7.

[25]

Amiri M, Khonsari M M. Rapid determination of fatigue failure based on temperature evolution: fully reversed bending load [J]. International Journal of Fatigue, 2010, 32(2):382-389.

[26]

Amiri M, Khonsari M M. Nondestructive estimation of remaining fatigue life: thermography technique [J]. Journal of Failure Analysis & Prevention, 2012, 12(6):683-688.

[27]

Liakat M, Khonsari M M. Analysis and life prediction of a composite Laminate under cyclic loading [J]. Composites Part B, 2016, 84.

[28]

Liakat M, Naderi M, Khonsari M M, et al. Nondestructive testing and prediction of remaining fatigue life of metals [J]. Journal of Nondestructive Evaluation, 2014, 33(3):309-316.

[29]

Valès B, Munoz V, Welemane H, et al. Heat source estimation in anisotropic materials [J]. Composite Structures, 2016, 136: 287-296

[30]

Boulanger T, Chrysochoos A, Mabru C, et al. Calorimetric analysis of dissipative and thermoelastic effects associated with the fatigue behavior of steels [J]. International Journal of Fatigue, 2004, 26(3):221-229.

Highlights  Temperature evolution of impacted CFRP laminated is studied.  ΔTmax is defined to study temperature variation during fatigue.  A method to determine COPs is proposed to facilitate the division of three stages.  Damages characteristic of the impacted CFRP during fatigue is studied.