基于改进Elman神经网络的CFRP补强钢板界面脱粘预测研究

PDF(2446 KB)

振动与冲击 ›› 2024, Vol. 43 ›› Issue (3) : 120-127.

论文

王庆松1，张玉1，张洪雨1，陈柏桦2

作者信息 +

Prediction of interfacial debonding of CFRP reinforced steel plate based on improved Elman neural network

WANG Qingsong1, ZHANG Yu1, ZHANG Hongyu1, CHEN Baihua2

Author information +

文章历史 +

摘要

针对碳纤维复合材料（carbon fiber reinforced polymer, CFRP）补强钢结构出现内部界面脱粘损伤后难以观测的问题，结合Lamb波检测方法和神经网络提出了一种界面脱粘预测方法。搭建了基于Lamb波的CFRP补强钢板信号分析实验平台，利用ABAQUS软件建立了CFRP补强钢板的机电耦合有限元模型，并通过实验验证了有限元模型的准确性。将长方形和圆形两种脱粘形状的信号在时域和频域内进行分析，基于自适应遗传算法改进的Elman神经网络建立了CFRP补强钢板脱粘预测模型，并将与脱粘面积相关性较高的信号特征数据作为预测模型的特征数据。对预测模型进行性能测试，脱粘形状为长方形和圆形预测值的平均绝对百分比误差分别为3.03%和8.06%，结果表明改进的Elman网络对于脱粘损伤具有较好的预测精度。

Abstract

Aiming at the problem that it is difficult to observe the internal interface debonding damage of carbon fiber composite (CFRP) strengthening steel structures, a method for interface debonding prediction using Lamb wave and neural networks was proposed. The experimental platform for the CFRP strengthening steel plate signal analysis based on Lamb wave was built. The electromechanical coupling finite element model was established using ABAQUS software, and the accuracy of the finite element model was verified by experiment. The signals obtained from the models with rectangular and circular debonding shapes were analyzed in the time and frequency domain. Based on the Elman neural network improved by the adaptive genetic algorithm, the prediction model for the debonding area of steel plate reinforced with CFRP was established. The signal feature data with a high correlation with the debonding area was used as the feature data for the prediction model. The performance of the prediction model is verified, and the mean absolute percentage error of the predicted debonding area for rectangular and circular debonding shapes is 3.03% and 8.06%, respectively, which show that the improved Elman network has better prediction accuracy for debonding.

导出引用

王庆松1，张玉1，张洪雨1，陈柏桦2. 基于改进Elman神经网络的CFRP补强钢板界面脱粘预测研究[J]. 振动与冲击, 2024, 43(3): 120-127

WANG Qingsong1, ZHANG Yu1, ZHANG Hongyu1, CHEN Baihua2. Prediction of interfacial debonding of CFRP reinforced steel plate based on improved Elman neural network[J]. Journal of Vibration and Shock, 2024, 43(3): 120-127

参考文献

[1] 金岩，郝志勇. 针对通过噪声的空滤器声学特性研究与改进[J]. 浙江大学学报：工学版，2006, 40(8):1143-1145. JIN Yong, HAO Zhiyong. Investigation and improvement of air-infilter acoustic performance towards pass-by noise [J]. Journal of Zhejiang University (Engineering Science)，2006,40(8):1143-1145. [1] 罗伟豪. 碳纤维复合材料修复补强技术研究[D].湖南大学,2020.DOI:10.27135/d.cnki.ghudu.2020.001254. LUO Zhiwei. Study on CFRP repair and reinforcement technology [D]. Hunan University,2020.DOI:10.27135/d.cnki.ghudu.2020.001254. [2] 王元清,宗亮,石永久,等.钢结构加固技术研究进展与标准编制[J].建筑结构学报,2022,43(10):29-40.DOI:10.14006/j.jzjgxb.2022.0095. WANG Yuanqing, ZONG Liang, SHI Yongjiu, et al. Research advances and design specification codification of retrofitting techniques for steel structures [J]. Journal of Building Structures,2022,43(10):29-40.DOI:10.14006/j.jzjgxb.2022.0095. [3] Worlton D C. Experimental confirmation of Lamb waves at megacycle frequencies[J]. Journal of applied physics, 1961,32(6): 967-971. [4] Deng M, Xiang Y, Liu L. Time-domain analysis and experimental examination of cumulative second-harmonic generation by primary Lamb wave propagation[J]. Journal of Applied Physics, 2011, 109(11): 113525. [5] 张燕,龚立娇.Lamb波频散特性的数值仿真研究[J].压电与声光,2014,36(05):701-704. ZHANG Yan, GONG Lijiao. Numerical Simulation Study on Lamb Wave Dispersion Curves [J]. Piezoelectrics & Acoustooptics, 2014,36(05):701-704. [6] Metya A K, Tarafder S, Balasubramaniam K. Nonlinear Lamb wave mixing for assessing localized deformation during creep[J]. NDT & E International, 2018, 98: 89-94. [7] 阚婷婷,阎守国,张碧星.板状结构Lamb波频散补偿与聚焦接收成像检测[J].应用声学,2022,41(01):96-102. KAN Tingting, YAN Shouguo, ZHANG Bixing. Lamb wave dispersion compensation and focused receiving imaging detection of plate structure [J]. Journal of Applied Acoustics,2022,41(01):96-102. [8] De Luca A, Sharif-Khodaei Z, Aliabadi M H, et al. Numerical simulation of the Lamb wave propagation in impacted CFRP laminate[J]. Procedia Engineering, 2016, 167: 109-115. [9] Gorgin R, Luo Y, Wu Z. Environmental and operational conditions effects on Lamb wave based structural health monitoring systems: A review[J]. Ultrasonics, 2020, 105: 106114. ,2021.DOI:10.27342/d.cnki.gscdu.2021.000784. [10] 张旭,刘宏业,李璇,等.基于分层模型的复合材料耦合Lamb波频率-波数域特性研究[J/OL].测控技术:1-7[2022-11-26].DOI:10.19708/j.ckjs.2022.07.281. ZHANG Xu, LIU Hongye, LI Xuan, et al. Research on frequency-wavenumber domain characteristics of coupled Lamb waves in composite laminates based on layered model [J/OL]. Measurement & Control Technology: 1-7 [2022-11-26].DOI:10.19708/j.ckjs.2022.07.281. [11] Alleyne D N, Cawley P. Optimization of Lamb wave inspection techniques[J]. Ndt & E International, 1992, 25(1): 11-22. [12] Sarvanos D, Birman V, Hopkins D. Detection of delaminations in composite beams using piezoelectric sensors[C]// 35th Structures, Structural Dynamics, and Materials Conference. Hilton Head,SC：SSDMC,1994. [13] Senyurek V Y. Detection of cuts and impact damage at the aircraft wing slat by using Lamb wave method[J]. Measurement, 2015, 67: 10-23. [14] 胡晓丹,周世圆,赵明华,等.不锈钢折弯板裂纹Lamb波检测技术[J].应用声学,2021,40(05):785-793. HU Xiaodan, ZHOU Shiyuan, ZHAO Minghua, et al. Detection of crack defects in stainless steel bending plates on Lamb wave [J]. Journal of Applied Acoustics, 2021,40(05):785-793. [15] 王壮杰,严刚,郭树祥,等.超声Lamb波监测铝合金结构疲劳裂纹扩展实验研究[EB/OL].(2022-03-17)[2023-02-17]. https://doi.org/10.13433/j.cnki.1003-8728.20220084. [16] 阎石,王鲁,崔赛杰,等.基于非线性Lamb波的板状结构微裂缝识别机理与算法[J].建筑科学与工程学报,2022,39(05):113-121.DOI:10.19815/j.jace.2021.02039. YAN Shi,WANG Lu,CUI Sai-jie,et al.Microcrack recognition mechanism and algorithm of plate structures based on nonlinear Lamb wave[J].Journal of Architecture and Civil Engineering,2022,39(5):113-121. [17] Song F, Huang G L, Hu G K. Online debonding detection in honeycomb sandwich structures using multi-frequency guided waves[C]//Second International Conference on Smart Materials and Nanotechnology in Engineering. SPIE, 2009, 7493: 434-441. [18] Sikdar S, Fiborek P, Kudela P, et al. Effects of debonding on Lamb wave propagation in a bonded composite structure under variable temperature conditions[J]. Smart Materials and Structures, 2018, 28(1): 015021. [19] Ng C T, Veidt M. Scattering characteristics of Lamb waves from debondings at structural features in composite laminates[J]. The Journal of the Acoustical Society of America, 2012, 132(1): 115-123. [20] Allen J C P, Ng C T. Debonding detection at adhesive joints using nonlinear Lamb waves mixing[J]. NDT & E International, 2022, 125: 102552. [21] 焦东方,孙志华.空气质量指数回归分析[J].中国海洋大学学报(自然科学版),2018,48(S2):228-234.DOI:10.16441/j.cnki.hdxb.20180258. JIAO Dongfang, SUN Zhihua. Regression analysis of air quality index [J]. Periodical of Ocean University of China,2018,48(S2):228-234.DOI:10.16441/j.cnki.hdxb.20180258. [22] 李斌,王大国,何治良,等.三轴条件下岩石抗剪强度参数线性回归法的修正[J].岩土力学,2022,43(10): 2689-2697.DOI:10.16285/j.rsm.2022.0637. LI Bin, WANG Daguo, HE Zhimin, et al. Modification of linear regression method for rock shear strength parameters under triaxial condition [J]. Rock and Soil Mechanics, 2022,43(10): 2689-2697.DOI:10.16285/j.rsm.2022. 0637. [23] 冯启林. 基于聚合神经网络的4D航迹预测及可视化系统开发[D].四川大学,2021.DOI:10.27342/d.cnki.gscdu.2021.000784. FENG Qilin. 4D trajectory prediction method based on the combinatorial neural network and visualization system development [D]. Sichuan University [24] Chala G T, Negash B M. Artificial Neural Network and Regression Models for Predicting Intrusion of Non-Reacting Gases into Production Pipelines[J]. Energies, 2022, 15. [25] 中华人民共和国国家质量监督检验检疫总局，中国国家标准化管理委员会. GB/T 8923.1-2011: 涂覆涂料前钢材表面处理表面清洁度的目视评定第1部分：未涂覆过的钢材表面和全面清除原有涂层后的钢材表面的锈蚀等级和处理等级[S]. 北京: 中国标准出版社, 2011. General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China, China National Standardization Administration. GB/T 8923.1-2011: Preparation of steel substrates before application of paints and related products-Visual assessment of surface cleanliness-Part 1:Rust grades and preparation grades of uncoated steel substrates and of steel substrates after overall removal of previous coatings [S]. Beijing: Standards Press of China, 2011. [26] Alleyne D N, Cawley P. Optimization of Lamb wave inspection techniques[J]. Ndt & E International, 1992, 25(1): 11-22. [27] Lee Rodgers J, Nicewander W A. Thirteen ways to look at the correlation coefficient[J]. The American Statistician, 1988, 42(1): 59-66. [28] Elman J L. Finding Structure in Time[J]. Cognitive Science, 1990, 14(2):179-211. [29] Li L, Xie X, Gao T, et al. A Modified Conjugate Gradient-based Elman Neural Network[J]. Cognitive Systems Research, 2021, 68(2). [30] Aziz M, Mostafa S A, Foozy C, et al. Integrating Elman Recurrent Neural Network with Particle Swarm Optimization Algorithms for an Improved Hybrid Training of Multidisciplinary Datasets[J]. Expert Systems with Applications, 2021, 183(12–2):115441. [31] （英）尼格尔•刘易斯著；沙瀛译. 深度学习实践指南基于R语言[M]. 北京：人民邮电出版社, 2018.06. [32] Cheng Y, Qi W M, Cai W Y. Dynamic properties of Elman and modified Elman neural network[C]//Proceedings. International Conference on Machine Learning and Cybernetics. IEEE, 2002, 2: 637-640