Deep learning can help to improve the guided-wave-based damage detection of composite structures; however, it needs a large number of damage samples. Based on a large number of simulated damage samples and a small number of real ones, a domain adaptive damage identification model is designed to realize the migration from simulated damage detection to real damage detection. Firstly, guided-wave signals of faked damage are collected extensively in the form of mass attachment on to the structure surface, and corresponding deep learning model based on convolutional-timing-sequential hybrid neural network is designed to achieve a high accuracy of damage detection. Secondly, a certain amount of guided-wave signals of real damage are collected, and a domain adaptive module is adopted by the model, which approximates the data distribution law of simulated damage and real damage in the feature space. With this framework, the model could detect the real damage without the labelling process in advance. The experimental results demonstrate the detection accuracy of 85.7%, which is ahead of other traditional deep learning models.
WANG Yupeng
,
LYU Shuaishuai
,
YANG Yu
,
LI Jiaxin
,
WANG Yezi
. Damage recognition of composite structures based on domain adaptive model[J]. ACTA AERONAUTICAET ASTRONAUTICA SINICA, 2022
, 43(6)
: 526752
-526752
.
DOI: 10.7527/S1000-6893.2022.26752
[1] 袁慎芳. 结构健康监控[M]. 北京:国防工业出版社, 2007:39-49. YUAN S F. Structural health monitoring and damage control[M]. Beijing:National Defense Industry Press, 2007:39-49(in Chinese).
[2] CRAVEN R, IANNUCCI L, OLSSON R. Delamination buckling:a finite element study with realistic delamination shapes, multiple delaminations and fibre fracture cracks[J]. Composites Part A:Applied Science and Manufacturing, 2010, 41(5):684-692.
[3] YUAN S F, CHEN J, YANG W B, et al. On-line crack prognosis in attachment lug using Lamb wave-deterministic resampling particle filter-based method[J]. Smart Materials and Structures, 2017, 26(8):085016.
[4] KULAKOVSKYI A, MESNIL O, CHAPUIS B, et al. Statistical analysis of guided wave imaging algorithms performance illustrated by a simple structural health monitoring configuration[J]. Journal of Nondestructive Evaluation, Diagnostics and Prognostics of Engineering Systems, 2021, 4(3):1-13.
[5] QIU L, YUAN S F, ZHANG X Y, et al. A time reversal focusing based impact imaging method and its evaluation on complex composite structures[J]. Smart Materials and Structures, 2011, 20(10):105014.
[6] WORLTON D C. Experimental confirmation of lamb waves at megacycle frequencies[J]. Journal of Applied Physics, 1961, 32(6):967-971.
[7] MEI H F, YUAN S F, QIU L, et al. Damage evaluation by a guided wave-hidden Markov model based method[J]. Smart Materials and Structures, 2016, 25(2):025021.
[8] SBARUFATTI C, MANSON G, WORDEN K. A numerically-enhanced machine learning approach to damage diagnosis using a Lamb wave sensing network[J]. Journal of Sound and Vibration, 2014, 333(19):4499-4525.
[9] 王彬文, 吕帅帅, 杨宇. 基于能量图谱和孪生网络的导波损伤诊断方法[J]. 振动测试与诊断, 2021, 41(1):182-189, 208. WANG B W, Lyu S S, YANG Y. Guided-wave based damage diagnosing method with energy spectrum and Siamese network[J]. Journal of Vibration, Measurement & Diagnosis, 2021, 41(1):182-189, 208(in Chinese).
[10] ATASHIPOUR S A, MIRDAMADI H R, HEMASIAN-ETEFAGH M H, et al. An effective damage identification approach in thick steel beams based on guided ultrasonic waves for structural health monitoring applications[J]. Journal of Intelligent Material Systems and Structures, 2013, 24(5):584-597.
[11] GANG Y. A Bayesian approach for damage localization in plate-like structures using Lamb waves[J]. Smart Materials and Structures, 2013, 22(3):035012.
[12] SU Z Q, YE L. Lamb wave-based quantitative identification of delamination in CF/EP composite structures using artificial neural algorithm[J]. Composite Structures, 2004, 66(1-4):627-637.
[13] AGARWAL S, MITRA M. Lamb wave based automatic damage detection using matching pursuit and machine learning[J]. Smart Materials and Structures, 2014, 23(8):085012.
[14] SHEN Q, YUE S Y, LU W, et al. Ultrasonic guided wave damage detection method for stiffened plates based on deep learning[J]. Journal of Physics:Conference Series, 2021, 1894(1):012069.
[15] CUI R T, AZUARA G, LANZA DI SCALEA F, et al. Damage imaging in skin-stringer composite aircraft panel by ultrasonic-guided waves using deep learning with convolutional neural network[J]. Structural Health Monitoring, 2021:1-16.
[16] KHAN A, SHIN J K, LIM W C, et al. A deep learning framework for vibration-based assessment of delamination in smart composite laminates[J]. Sensors (Basel, Switzerland), 2020, 20(8):23-35.
[17] ZHANG B, HONG X B, LIU Y. Multi-task deep transfer learning method for guided wave-based integrated health monitoring using piezoelectric transducers[J]. IEEE Sensors Journal, 2020, 20(23):14391-14400.
[18] TZENG E, HOFFMAN J, ZHANG N, et al. Deep domain confusion:maximizing for domain invariance[J]. Computer Science, 2014, arXiv:1412.3474.
[19] 杨宇, 王彬文, 吕帅帅, 等. 一种基于深度学习的复合材料结构损伤导波监测方法[J]. 航空科学技术, 2020, 31(7):102-108. YANG Y, WANG B W, LYU S S, et al. A deep-learning-based method for damage identification of composite laminates[J]. Aeronautical Science & Technology, 2020, 31(7):102-108(in Chinese).
[20] 刘国强, 肖迎春, 张华, 等. 复合材料加筋壁板损伤识别的概率成像方法[J]. 复合材料学报, 2018, 35(2):311-319. LIU G Q, XIAO Y C, ZHANG H, et al. Probability-based diagnostic imaging for damage identification of stiffened composite panel[J]. Acta Materiae Compositae Sinica, 2018, 35(2):311-319(in Chinese).
CJA