Page 121 - 《应用声学》2022年第4期

第 41 卷 第 4 期 邱文等： Lamb 波多特征参数的复合材料损伤程度评估方法 619

fuzzy decision making[J]. Inverse Problems in Science and
5 结论 Engineering, 2020, 28(1): 21–46.
[8] Hossain M S, Ong Z C, Ismail Z, et al. Artificial neural
本文对复合材料的不同程度损伤进行评估，并 networks for vibration based inverse parametric identifi-
设计了相应的配套系统。通过结构健康监测技术进 cations: a review[J]. Applied Soft Computing, 2017, 52:
203–219.
行损伤的在线监测，利用信号分析方法中的小波包
[9] de Oliveira M A, Inman D J. Performance analysis of sim-
分析法对损伤信号进行时域与频域上的分析，提取 plified fuzzy ARTMAP and probabilistic neural networks
了时域特征参数 (波形特征 Wf、波峰特征 Wp) 和 for identifying structural damage growth[J]. Applied Soft
Computing, 2017, 52: 53–63.
频域特征参数 (能量分布 Ed、能量百分比 E)，建立
[10] Wang P, Shi Q. Damage identification in structures based
损伤特征向量，构建了损伤信息标准库，利用两种神 on energy curvature difference of wavelet packet trans-
经网络对不同程度的损伤进行评估。 form[J]. Shock & Vibration, 2018, 2018(2): 1–13.
针对系统的准确性与可靠性，设计实验进行了 [11] Zhang W, Sun L, Zhang L. Local damage identification
method using finite element model updating based on a
验证。分别采集了通孔、裂纹两种损伤进行验证。 new wavelet damage function[J]. Advances in Structural
裂纹损伤分为 3 种不同程度，从实验的结果可以得 Engineering, 2018, 21(10): 1482–1494.
[12] 邓菲, 刘洋, 诸葛霞, 等. 变化环境下的超声导波结构健康监
出：BP 神经网络评估正确率在 75% 以上，GA-BP
测研究进展 [J]. 机电工程学报, 2016, 52(18): 1–7.
神经网络的损伤程度评估正确率达到 87.5% 以上， Deng Fei, Liu Yang, Zhuge Xia, et al. Progress on the
具有较高的正确率。通孔损伤则分为6 种不同程度， research of ultrasonic guided wave structural health mon-
GA-BP 神经网络与 BP 神经网络的综合均评估正 itoring in the changing ambient[J]. Journal of Mechanical
Engineering, 2016, 52(18): 1–7.
确率分别为78.85%、68.9%，表明在一定程度上也能 [13] Wu W, Zhang H, Jia F, et al. Surface effects on frequency
够完成损伤程度的预测。两种损伤的准确率出现的 dispersion characteristics of Lamb waves in a nanoplate[J].
差距主要是由损伤程度种类的不同而产生了较大 Thin Solid Films, 2020: 697.
[14] Qi L, Feng Y, Sun L, et al. Leak source beam-forming
的影响。从整体上看，该损伤程度评估系统能够对 location of spacecraft in orbit based on dispersion charac-
未知损伤进行预测，降低安全隐患，为维修提供指导 teristics of lamb wave[C]. Advanced Science and Industry
意见。 Research Center: Science and Engineering Research Cen-
ter, 2020: 1770–1775.
参 考 文 献 [15] 吕文瀚, 吴先梅, 陈家熠. 金属材料疲劳损伤检测的非线性声
学方法 [J]. 应用声学, 2018, 37(6): 874–881.
[1] 陈雪峰, 杨志勃, 田绍华, 等. 复合材料结构损伤识别与健康 Lyu Wenhan, Wu Xianmei, Chen Jiayi. Nonlinear acous-
监测展望 [J]. 振动, 测试与诊断, 2018, 38(1): 203–212. tic method for fatigue damage detection of metal materi-
[2] Andrzej K. Nondestructive damage assessment of compos- als[J]. Journal of Applied Acoustics, 2018, 37(6): 874–881.
ite structures based on wavelet analysis of modal curva- [16] 王强, 胥静, 王梦欣, 等. 结构裂纹损伤的 Lamb 波层析成像
tures: state-of-the-art review and description of wavelet- 监测与评估研究 [J]. 机械工程学报, 2016, 52(6): 30–36.
based damage assessment benchmark[J]. Shock and Vi- Wang Qiang, Xu Jing, Wang Mengxin, et al. Lamb
bration, 2015, 2015: 1–19. wave tomography technique for crack damage detection[J].
[3] Yang Z B, Radzienski M, Kudela P, et al. Fourier spectral- Journal of Mechanical Engineering, 2016, 52(6): 30–36.
based modal curvature analysis and its application to [17] 黄桥生, 任德军, 章亚林, 等. P92 钢焊接接头蠕变损伤的非
damage detection in beams[J]. Mechanical Systems and 线性超声检测研究 [J]. 应用声学, 2020, 39(3): 366–371.
Signal Processing, 2017, 84(84): 763–781. Huang Qiaosheng, Ren Dejun, Zhang Yalin, et al. Nonlin-
[4] Yang Z B, Radzienski M, Kudela P, et al. Two- ear ultrasonic detection of creep damage in welded joints
dimensional modal curvature estimation via Fourier spec- of P92 steel[J]. Journal of Applied Acoustics, 2020, 39(3):
tral method for damage detection[J]. Composite Struc- 366–371.
tures, 2016, 148: 155–167. [18] 夏小松, 郑艳萍. 基于 Lamb 波时间反转法的复合材料损伤检
[5] 张玉祥, 张鑫, 陈家照, 等. 基于压电阻抗法的结构损伤检测 测 [J]. 中国机械工程, 2021, 32(1): 26–31, 53.
技术进展 [J]. 无损检测, 2016(1): 69–74. Xia Xiaosong, Zheng Yanping. Damage detection in com-
Zhang Yuxiang, Zhang Xin, Chen Jiazhao, et al. Devel- posite based on time reversal Lamb waves method[J].
opment on detecting technique of structure damage based China Mechanical Engineering, 2021, 32(1): 26–31, 53.
on EMI[J]. Nondestructive Testing, 2016(1): 69–74. [19] 陈军, 王庆冬. 嵌入式超声传感器的混凝土损伤非线性检测研
[6] 左春愿. 基于机电阻抗技术的结构损伤识别方法研究 [D]. 大 究 [J]. 应用声学, 2018, 37(4): 481–487.
连: 大连理工大学, 2016. Chen Jun, Wang Qingdong. Concrete damage nonlinear
[7] Alexandrino P D S L, Gomes G F, Sebastião Simões detection based on embedded ultrasonic sensors[J]. Jour-
Cunha Jr. A robust optimization for damage detection us- nal of Applied Acoustics, 2018, 37(4): 481–487.
ing multiobjective genetic algorithm, neural network and