-
复合材料层压板低速冲击损伤的脉冲与超声红外热波成像检测试验及检测热图处理流程可分为以下步骤,如图1所示。
-
依据ASTM D7136/D 7136M-12标准制备了碳纤维增强环氧树脂和玻璃纤维增强聚丙烯复合材料层压板试件,试件铺层均为[45/0/-45/90]4S交叉铺层,试件规格150 mm×100 mm×4.8 mm。采用Rapid Scan 2型滚轮式超声相控阵探伤仪对试件进行出厂检测,结果显示试件表面质量良好,内部无分层及夹杂等缺陷。两类试件尺寸图及实物图如图2所示。
-
从上述制备的两类试件中分别选取2块,依次编号为1#-4#,依据ASTM D7136标准进行冲击试验。采用图1所示的Instron Dynatup 9250 HV落锤冲击试验机进行低速冲击试验,冲击器质量为12.527 kg,锤头选用直径16 mm半球形锤头,另配有150 mm×100 mm的敞开窗口的相框夹具用于放置试件,夹具周围的四个橡胶夹头用来固定试件以防止试件冲击过程中的震颤,同时试验机配有气动回弹装置防冲头二次冲击。
分别对两类层压板1#-4#试件进行两种能量水平(15 J与30 J)的冲击试验,试验过程中通过改变冲击器的高度来调整冲击能量。观察试件冲击过程中各响应参数与时间的关系曲线(以2#试件为例),如图3所示,冲击过程可分为冲头的自由下落、冲头与试件的相互挤压和冲头回弹三个阶段。在自由下落阶段,冲击器自由落下,其势能转化为动能,当与试件接触的前一刻其速度最大;在挤压阶段,冲头开始接触并挤压试件,导致其速度开始降低,冲头与试件之间的接触力开始增加,冲头动能逐渐转化为冲击能量,当冲击过程持续一段时间后,速度下降至0,接触力达到最大,动能全部转化为冲击能,在此期间冲击力-时间曲线的波动意味着试件损伤的生成,而曲线的第一次骤降被认为是大面积分层的开始,称之赫兹破坏[25-26];在弹回阶段,由于试件本身的弹性恢复能力,冲头被弹回,冲头与试件之间开始脱离,接触力逐渐降低,弹回动能增加,当冲头与试件完全脱离的一刻,二者接触力为0,弹回速度达到最大,至此冲击过程完成。对冲击后的试件进行超声C扫检测,分层损伤结果如图4所示,分层面积分别为1077.72 mm2,1523.96 mm2,831.56 mm2和1281.74 mm2。
-
主动红外热成像检测系统主要包括三个部分:热激励源、红外热像仪和计算机控制处理系统,其检测原理如图5所示。通过脉冲或超声热激励源对被测试件施加可控的热激励源,当热流在试件内部传播,由于损伤的热属性差异,导致热量在损伤处形成累积,从而改变试件表面的温度场,采用红外热像仪监测这种变化的温度场,可以实现对试件内部损伤的判读。
采用课题组自主开发的脉冲和超声激励热成像检测系统进行主动热成像检测试验,如图6所示,两试验共用红外热像仪和工控计算机。红外热像仪采用德国InfraTec公司制造的Vhr 680 型红外热像仪,其可探测红外光谱范围为7.5~14 μm,可测量温度范围为−40~+1200 ℃,测量精度为±2% (<0或>100 ℃)或±1.5 ℃(0~100 ℃),最大采集频率50 Hz,试验中也均设置为50 Hz;脉冲热激励源由两只并行排列的自然冷却式线性脉冲闪光灯组成,可释放宽度为2 ms的高能瞬时脉冲,最大输出能量为4800 J,蓄能时间间隔为5 s;超声热激励源采用诺威尔公司生产的UST-200超声枪,其输出功率为1 kW,频率为20 kHz,激励时间为200 ms。由于医用胶布在有效保护试件的同时还能削弱驻波共振现象,提高热图损伤对比度,因此采用其作为超声热成像检测耦合剂。
-
依据上述检测参数对1#-4#试件进行了脉冲热成像检测试验。检测结果如图7~图10所示,图中亮度较暗的区域为非损伤区域,较亮的中心区域为冲击损伤区域。从图中可以看出,在脉冲激励触发之后,试件首先被全局加热,当热流向层压板内部传播遇到损伤时,损伤会阻碍其传播左右从而导致损伤处热量堆积,温度下降趋势变缓,而在无损伤区域,由于无任何阻碍,热流传播速度较快,温度趋势不变,由此产生了表面温差,以1#试件为例,整个过程的温度变化示意图见图11。
-
关于超声红外热成像,当被测试件收到超声激励后,由于损伤内表面摩擦、塑性变形、粘弹性阻尼效应等因素影响,使振动的机械能转换为热能,从而改变试件表面的温度场。Renshaw [27]等指出摩擦生热主要位于损伤内表面之间的区域,是最主要的生热形式。四块试件的超声热成像检测结果如图12~15所示。与上述脉冲热成像现象类似,亮度较暗的区域为非损伤区域,较亮的区域为损伤区域。
图 12 1#试件(TS-15J)超声热波检测结果
Figure 12. Test results of 1# specimen (TS-15J) by ultrasonic thermalgraphy
图 13 3#试件(TP-15J)超声热波检测结果
Figure 13. Test results of 3# specimen (TP-15J) by ultrasonic thermalgraphy
图 14 2#试件(TS-30J)超声热波检测结果
Figure 14. Test results of 2# specimen (TS-30J) by ultrasonic thermalgraphy
图 15 4#试件(TP-30J)超声热波检测结果
Figure 15. Test results of 4# specimen (TP-30J) by ultrasonic thermalgraphy
在激励瞬间,各试件靠近表面的分层损伤经不断摩擦后在表面呈现一块非规则圆形热斑,随着激励的持续,摩擦作用不断增强,远离试件的分层损伤摩擦生热后的温度不断传递到试件表层,因此观察到的热斑面积和亮度不断增加,且热量耗散速度较慢。另外超声枪与试件的接触区域由于超声刺激作用也会产生较高的温度,称之为激励热区。
关于不同试件在超声激励下的时序变化特征,为消除环境干扰和时间表面特性影响,利用最大表面温差来分析不同试件在超声激励下的表面温度的时序变化特征,如公式(1)所示:
$$\Delta T_t^{\max } = T_t^{d\max } - T_t^b$$ (1) 式中:
$\Delta T_t^{\max }$ 为t时刻试件表面损伤区域温度最大值$ T_t^{d\max } $ 与背景区域温度平均值$T_t^b$ 的差值。利用公式(1),得到1#-4#的最大表面温差-时间曲线如图16(b)所示。从图16中可以看出,TP试件在超声激励下的最大表面温差整体要大于TS试件,这与图12-15中的热图结果相一致。这可能是在超声激励时TP试件良好的韧性使得超声枪在激励时与试件贴合程度更高,超声波的传递效率增加使得损伤区域之间的摩擦作用增加,生热量增高,分层损伤越明显,而TS试件由于表面刚性较大从而与超声枪激励时的贴合度不高,导致其检测效果相比于TP试件不够明显。
-
要实现对主动红外热波成像无损检测能力的有效评估,关键在于能否实现损伤面积参数的定量识别。因此,为准确解读检测热图中所含的损伤信息,需要采用有效的数据处理和数字图像处理方法对热图进行处理和分析,从而实现损伤参数的定量识别。为此,从图像预处理、损伤提取及阈值分割、损伤定量识别三个步骤对热图进行处理,下面以1#试件为例。
-
为提高热斑损伤阈值分割的稳健性,首先对图像进行预处理,具体可分为如下两步:一是先对热成像检测热图序列进行自适应对比度增强(CLAHE),突出损伤边缘,提高损伤边缘对比度,放大损伤与背景之间的像素差异[28-29];二是通过高斯滤波法消除图像噪声,改善图像的清晰度和视觉效果[30],其原理如公式(2)所示:
$$ {I_{\text{g}}}(x,y) = \frac{{\displaystyle\sum\limits_{(i,j) \in {n_{x,y}}} {{n_d}(i,j)I(i,j)} }}{{\displaystyle\sum\limits_{(i,j) \in {n_{x,y}}} {{n_d}(i,j)} }} $$ (2) 式中:nx,y表示中心像素(x,y)的M·M(M为奇数)大小的邻域;nd为相似度权重因子。
-
经上述热图预处理操作后,热图质量提高的同时突出了分层损伤的特征信息,但要实现损伤的定量评估,还需要将损伤从热图中分割提取出来。图像分割算法一般基于强度值的两个基本属性:不连续性和相似性。前者是基于强度水平的不连续变化,如阈值分割和边缘检测,但由于严重依赖于高阶偏微分方程和滤波器操作使其不易于进行热图损伤区域提取。而后者即图像强度值的相似性属性使其可以有效区别背景区域和损伤区域,因此,文中提出了基于图像强度值相似性理论的红外热图生长区域算法。
通过分析热图序列,发现损伤区域热斑灰度差值变化较小,而损伤区域热斑与背景区域相交边界处周围灰度差值变化较大。假设种子点与其四联通区域灰度值为f和
${\delta _i}$ ,种子点灰度值与其四连通区域之间的强度值之差为d。三者关系如公式(3)所示:$$ d = f(x,y) - {\delta _i} $$ (3) 设为区域生长的基本停止参数,即生长因子。当种子点与四连通区域的灰度差值小于
$ {\delta _{\lim }} $ 时,生长区域以种子点为中心不断向外扩张,每扩张一次判断其生长因子条件,若不满足则以其四连通区域为新的种子点并继续生长,直至损伤区域被全部提取,当种子点与四连通区域的灰度值大于$ {\delta _{\lim }} $ 时,即所在区域达到损伤区域热斑与背景区域相交边界处时,种子点停止生长。其原理如公式(4)和图17所示。$$ {\text{img }}(x,y) =\left\{ \begin{gathered} 1,{d_{\min }} < {\delta _{\lim }} \\ 0,{d_{\min }} \geqslant {\delta _{\lim }} \\ \end{gathered} \right. $$ (4) 以1#为例,脉冲和超声红外热成像检测损伤区域提取和分割后的图像结果如图18和图19所示。由结果可知该算法损伤区域分割效果较好,非常适用于红外热图序列的损伤提取。
-
试件中的损伤信息是根据表面温度的差异性反映出来的,这为利用原始热图中的温度数据获取损伤参数提供了依据。此外,由上述热图处理方法研究可知,热图经过分割后成为二值图像,即损伤区域的像素值为1,而非损伤区域的像素值为0,这为利用二值链码技术确定损伤参数奠定了基础。
试件中的实际损伤面积可以根据边界内部的像素点进行换算得到[31]。设试件在热图中的面积为Si,试件的实际面积为Ss(Ss=150 mm×100 mm),则可以根据公式(5)换算得到实际的损伤面积S。
$$ S = \frac{{{S_d}}}{{{S_i}}} \times {S_s} $$ (5) 式中:
$ {S_d} = \sum\limits_{(i,j \in R)} 1 $ ,为试件在热图中的损伤区域面积。通过此步骤可以算出1#-4#试件在脉冲和超声热成像下所检测到的实际损伤面积,从而为两项技术在热固性/热塑性复合材料损伤检测中的定量分析和能力评估奠定基础。 -
通过上述定量识别算法计算出各试件损伤检测结果,并将计算结果与超声C扫描结果进行对比得到相对误差,如表1所示。
表 1 脉冲/超声检测方法下TP/TS试件分层损伤面积及其相对误差
Table 1. The delamination damage area and its relative error of TP/TS specimens under pulse/ultrasonic detection method
Sample number Damage area/mm2 Relative error Pulse Ultrasonic C-scanning Pulse Ultrasonic 1# 875.29 1048.16 1077.72 18.7% 2.74% 2# 1068.70 1476.83 1523.96 29.8% 3.09% 3# 63.75 791.69 831.56 92.3% 4.79% 4# 209.65 1084.70 1281.74 83.6% 15.3% 由表1可知,对于热固性CFRP,脉冲红外热成像相对于超声C扫描方法分层损伤检测结果误差在30%以内,超声红外热成像相对于超声C扫描误差在5%以内。对于热塑性GFRP,脉冲红外热成像对其损伤检测效果较差,与超声C扫相对误差最高达到了92.3%,而超声红外热成像相对于超声C扫描误差在20%以内。而且发现对于冲击能量较小的试件,损伤面积计算值的相对误差就越小。由此可知:脉冲红外热成像对于热固性复合材料低速冲击分层损伤的检测能力较好,但并不适用于热塑性复合材料低速冲击分层损伤检测;而超声红外热成像对于两种类型复合材料分层损伤都具有较好的检测能力,且检测精度整体优于脉冲红外热成像。
Evaluation of infrared thermal wave detection capability for delamination damage of thermosetting/thermoplastic composites
-
摘要: 主动红外热成像技术在不同基体复合材料分层损伤中的检测能力尚未被评估。文中通过设计制作两种典型热固性/热塑性复合材料层压板,分别采用脉冲红外热成像、超声红外热成像、超声C扫描三种方法对不同冲击能量下的分层损伤进行了检测研究。以超声C扫描结果为参照,对比了两种红外热成像技术的检测结果,同时针对热图序列损伤区域的阈值分割提取开发了基于图像强度值相似性理论的区域生长算法。损伤的定量识别结果表明:脉冲热成像对热固性复合材料的分层损伤检测效果较好,但其不适用于热塑性复合材料损伤检测,超声热成像对于两类复合材料分层损伤均有较好的检测能力且整体检测精度优于脉冲热成像。期间对不同损伤检测效果的深层次机理进行了分析,并提出了分别针对两种基体类型复合材料的红外热成像技术评估流程和标准。Abstract: The ability of active infrared thermography to detect delamination damage in different matrix composites has not been evaluated. Two typical thermosetting/thermoplastic composite laminates were designed and manufactured in this paper. The three methods of pulse infrared thermography, ultrasonic infrared thermography, and ultrasonic C-scan were used to detect and study the delamination damage under different impact energy. The scanning results of ultrasonic C-scan were for reference, and the detection results of the two infrared thermography were compared. At the same time, a region growth algorithm based on the similarity theory of image intensity value was developed for the threshold segmentation and extraction of damaged regions of heat map sequence. Accuracy and quantitative damage identification results show that pulse infrared thermography is effective in detecting layered damage of thermoset composites, but it is not suitable for damage detection of thermoplastic composites. Ultrasonic thermal thermography is effective for both types of delamination damage of composites. It has better detection capability and overall detection accuracy is better than pulse infrared thermography. The deep mechanism of different damage detection effects was analyzed, and the evaluation process and standard of infrared thermography for the two matrix types of composites were proposed.
-
表 1 脉冲/超声检测方法下TP/TS试件分层损伤面积及其相对误差
Table 1. The delamination damage area and its relative error of TP/TS specimens under pulse/ultrasonic detection method
Sample number Damage area/mm2 Relative error Pulse Ultrasonic C-scanning Pulse Ultrasonic 1# 875.29 1048.16 1077.72 18.7% 2.74% 2# 1068.70 1476.83 1523.96 29.8% 3.09% 3# 63.75 791.69 831.56 92.3% 4.79% 4# 209.65 1084.70 1281.74 83.6% 15.3% -
[1] Li Wei, Guo Quanfeng. Application of carbon fiber composites to cosmonautic fields [J]. Chinese Optics, 2011, 4(3): 201-212. (in Chinese) doi: 10.3969/j.issn.2095-1531.2011.03.001 [2] Zhang Junqiang, Yan Changxiang. Application of carbon/epoxy composites to support structures of space payloads [J]. Chinese Optics, 2009, 2(2): 79-84. (in Chinese) doi: 10.3969/j.issn.2095-1531.2009.02.002 [3] Chen Yuan, Yu Fujie, Hu Mingduo, et al. Design and implementation of combination cam-type Z-pin pre-insertion device for aeronautical materials [J]. Optics and Precision Engineering, 2019, 27(2): 379. (in Chinese) doi: 10.3788/OPE.20192702.0379 [4] Shah S Z H, Karuppanan S, Megat-Yusoff P S M, et al. Impact resistance and damage tolerance of fiber reinforced composites: A review [J]. Composite Structures, 2019, 217: 100-121. doi: 10.1016/j.compstruct.2019.03.021 [5] Richardson M O W, Wisheart M J. Review of low-velocity impact properties of composite materials [J]. Composites Part A: Applied Science and Manufacturing, 1996, 27(12): 1123-1131. doi: 10.1016/1359-835X(96)00074-7 [6] Meo M, Polimeno U, Zumpano G. Detecting damage in composite material using nonlinear elastic wave spectroscopy methods [J]. Applied Composite Materials, 2008, 15(3): 115-126. doi: 10.1007/s10443-008-9061-7 [7] Lei L, Ferrarini G, Bortolin A, et al. Thermography is cool: defect detection using liquid nitrogen as a stimulus [J]. NDT & E International, 2019, 102: 137-143. [8] Deane S, Avdelidis N P, Castanedo C I, et al. Application of NDT thermographic imaging of aerospace structures [J]. Infrared Physics & Technology, 2019, 97: 456-466. doi: 10.1016/j.infrared.2019.02.002 [9] Li Yin, Tian Gan, Yang Zhengwei, et al. Detection capability evaluation of low velocity impact damage in composites using ultrasonic infrared thermography [J]. Chinese Journal of Scientific Instrument, 2016, 37(5): 1124-1130. (in Chinese) [10] Liu Junyan, Liu Liqiang, Wang Yang. Experimental study on active infrared thermography as a NDI tool for carbon–carbon composites [J]. Composites Part B: Engineering, 2013, 45(1): 138-147. doi: 10.1016/j.compositesb.2012.09.006 [11] Kou Guangjie, Yang Zhengwei, Jia Yong, et al. Detection on cracks in blades with complex profile based on ultrasonic infrared thermal imaging [J]. Infrared and Laser Engineering, 2019, 48(12): 1204002. (in Chinese) doi: 10.3788/IRLA201948.1204002 [12] Tian Gan, Yang Zhengwei, Zhu Jietang, et al. Vibration characteristics and acoustic chaos analysis in ultrasonic infrared thermal wave detection [J]. Infrared and Laster Engineering, 2016, 45(3): 0304003. (in Chinese) [13] Yang Zhengwei, Kou Guangjie, Zhou Wei, et al. Theoretical model for heat generation of crack on different preload force under ultrasound excitation [J]. Infrared and Laster Engineering, 2020, 49(S1): 20200158. (in Chinese) [14] Maldague X P V. Introduction to NDT by active infrared thermography [J]. Materials Evaluation, 2002, 60(9): 1060-1072. [15] Usamentiaga R, Venegas P, Guerediaga J, et al. Infrared thermography for temperature measurement and non-destructive testing [J]. Sensors, 2014, 14(7): 12305-12348. doi: 10.3390/s140712305 [16] Adams R D, Cawley P, Pye C J, et al. A vibration technique for non-destructively assessing the integrity of structures [J]. Journal of Mechanical Engineering Science, 1978, 20(2): 93-100. doi: 10.1243/JMES_JOUR_1978_020_016_02 [17] Milne A J M, Reynolds W N. The non-destructive evaluation of composites and other materials by thermal pulse video thermography [C]//Proceedings of SPIE, 1985, 520(12): 119-122. [18] Avdelidis N P, Almond D P, Dobbinson A, et al. Aircraft composites assessment by means of transient thermal NDT [J]. Progress in Aerospace Sciences, 2004, 40(3): 143-162. doi: 10.1016/j.paerosci.2004.03.001 [19] Ibarra-Castanedo C, Avdelidis N P, Grinzato E G, et al. Delamination detection and impact damage assessment of GLARE by active thermography [J]. International Journal of Materials and Product Technology, 2011, 41(1-4): 5-16. [20] Gaudenzi P, Bernabei M, Dati E, et al. On the evaluation of impact damage on composite materials by comparing different NDI techniques [J]. Composite Structures, 2014, 118: 257-266. doi: 10.1016/j.compstruct.2014.07.048 [21] Chulkov A O, Vavilov V P, Nesteruk D A, et al. Synthesizing data of active infrared thermography under optical and ultrasonic stimulation of products made of complex-shaped CFRP [J]. Russian Journal of Nondestructive Testing, 2020, 56(7): 595-601. doi: 10.1134/S1061830920070037 [22] Li Y, Zhang W, Yang Z, et al. Low-velocity impact damage characterization of carbon fiber reinforced polymer (CFRP) using infrared thermography [J]. Infrared Physics & Technology, 2016, 76: 91-102. [23] Pracht M, Świderski W. Experimental testing of aramid composite applied in ballistic armour by ultrasonic IR thermography [J]. Problems of Mechatronics Armament Aviation Safety Engineering, 2019, 10(2): 79-88. doi: 10.5604/01.3001.0013.2117 [24] Quattrocchi A, Freni F, Montanini R. Comparison between air-coupled ultrasonic testing and active thermography for defect identification in composite materials [J]. Nondestructive Testing and Evaluation, 2021, 36(1): 97-112. doi: 10.1080/10589759.2019.1699084 [25] Panettieri E, Fanteria D, Montemurro M, et al. Low-velocity impact tests on carbon/epoxy composite laminates: a benchmark study [J]. Composites Part B: Engineering, 2016, 107: 9-21. doi: 10.1016/j.compositesb.2016.09.057 [26] Ishikawa T, Aoki Y, Suemasu H. Pursuit of mechanical behavior in compression after impact (CAI) and open hole compression (OHC) [C]//Proceedings of 15th International Conference on Composite Materials (ICCM-15), 2005. [27] Renshaw J, Chen J C, Holland S D, et al. The sources of heat generation in vibrothermography [J]. NDT & E International, 2011, 44: 736-739. [28] Upadhyay J, Jaiswal A. A joint implementation of adaptive histogram equalization and interpolation [J]. Optik, 2015, 126(24): 5936-5940. doi: 10.1016/j.ijleo.2015.08.150 [29] Vyavahare A J, Thool R C. Segmentation using region growing algorithm based on CLAHE for medical images[C]//Fourth International Conference on Advances in Recent Technologies in Communication and Computing (ARTCom2012), 2012: 182-185. [30] Shin D H, Park R H, Yang S, et al. Block-based noise estimation using adaptive Gaussian filtering [J]. IEEE Transactions on Consumer Electronics, 2005, 51(1): 218-226. doi: 10.1109/TCE.2005.1405723 [31] Dudzik S. A simple method for defect area detection using active thermography [J]. Opto-Electronics Review, 2009, 17(4): 338-344. doi: 10.2478/s11772-009-0016-9