基于WOA-BP-LSTM自编码器的CFRP薄壁C柱轴压响应预测

  • 牟浩蕾 ,
  • 张贾 ,
  • 冯振宇 ,
  • 白春玉
展开
  • 1. 中国民航大学
    2. 西安交通大学
    3. 中国飞机强度研究所

收稿日期: 2025-06-03

  修回日期: 2025-07-30

  网络出版日期: 2025-07-30

基金资助

国家自然科学基金项目;天津市应用基础研究多元投入基金项目;中央高校基本科研业务费项目;中国民航大学研究生科研创新项目

Prediction of axial crushing response for CFRP thin-walled C-columns based on WOA-BP-LSTM autoencoder

  • MOU Hao-Lei ,
  • ZHANG Jia ,
  • FENG Zhen-Yu ,
  • BAI Chun-Yu
Expand

Received date: 2025-06-03

  Revised date: 2025-07-30

  Online published: 2025-07-30

Supported by

National Natural Science Foundation of China;Tianjin Applied Basic Research Multi-Input Fund Project;Fundamental Research Funds for the Central Universities

摘要

针对航空器货舱下部碳纤维增强复合材料(CFRP)薄壁C柱在准静态轴压下的力-位移响应预测问题,提出了一种融合鲸鱼优化算法(WOA)、反向传播(BP)神经网络和长短期记忆(LSTM)自编码器的智能预测模型(WOA-BP-LSTM自编码器模型)。通过CFRP薄壁C柱准静态轴压试验验证了有限元模型可靠性,其轴压响应评价指标误差均小于10%,基于该模型构建了包含700组变截面几何参数的力-位移响应数据集。采用LSTM自编码器实现力-位移响应特征降维与重建,随后采用BP神经网络对力-位移响应进行预测,并采用WOA进行神经网络参数优化。结果表明,LSTM自编码器实现了力-位移响应的高精度重建,测试集初始峰值压溃力和能量吸收的重建误差均小于3%,80%样本误差小于1%;优化后预测模型的力-位移响应预测精度显著提升,测试集平均绝对误差(MAE)降低17.55%,均方误差(MSE)降低31.77%,均方根误差(RMSE)降低17.47%,初始峰值压溃力和能量吸收的预测误差均小于8%,80%样本误差小于5%。该智能预测模型实现了变截面CFRP薄壁C柱轴压响应的快速精准预测并降低了计算成本,为其轴压响应研究提供了一种高效的参数-性能映射工具

本文引用格式

牟浩蕾 , 张贾 , 冯振宇 , 白春玉 . 基于WOA-BP-LSTM自编码器的CFRP薄壁C柱轴压响应预测[J]. 航空学报, 0 : 1 -0 . DOI: 10.7527/S1000-6893.2025.32349

Abstract

Aiming to predict the force-displacement responses of carbon fiber reinforced plastic (CFRP) thin-walled C-columns in the aircraft sub-cargo area under quasi-static axial crushing, an intelligent prediction model (WOA-BP-LSTM autoencoder model) integrating the Whale Optimization Algorithm (WOA), Back Propagation (BP) neural network, and Long Short-Term Memory (LSTM) autoencoder was proposed. The reliability of the finite element model of CFRP thin-walled C-columns was validated through quasi-static axial crushing tests, with axial crushing response evaluation indicators showing errors within 10%. A dataset comprising 700 force-displacement response samples with variable cross-sectional geometric parameters was constructed based on the model. The LSTM autoencoder was employed for dimensionality reduction and reconstruction of the force-displacement responses. Subsequently, the BP neural network was used for force-displacement responses prediction, with WOA optimizing the neural network parameters. The results showed that the LSTM autoencoder achieved high-precision reconstruction of force-displacement responses, where the errors for initial peak crushing force and energy absorption in the test set were both less than 3%, and 80% of the samples had errors within 1%. The optimized prediction model significantly improved prediction accuracy, reducing the test set’s mean absolute error (MAE) by 17.55%, mean squared error (MSE) by 31.77%, and root mean squared error (RMSE) by 17.47%. Prediction errors for the initial peak crushing force and energy absorption were both less than 8%, with 80% of samples showing errors within 5%. This model enables rapid and accurate prediction of axial crushing responses for variable cross-section CFRP thin-walled C-columns while reducing computational costs, providing an efficient parameter-performance mapping tool for the study of its axial crushing response.

参考文献

[1]ABRAMOWICZ W.Thin-walled structures as impact energy absorbers[J].Thin-Walled Structures, 2003, 41(2-3):91-107 [2] BAROUTAJI A, SAJJIA M, OLABI A G.On the crashworthiness performance of thin-walled energy absorbers: Recent advances and future developments[J]. Thin-Walled Structures, 2017, 118: 137-163. [3]JACOB G C, FELLERS J F, SIMUNOVIC S, et al.Energy absorption in polymer composites for automotive crashworthiness[J].Journal of Composite Materials, 2002, 36(7):813-850 [4] GUIDA M, MARULO F, ABRATE S.Advances in crash dynamics for aircraft safety[J]. Progress in Aerospace Sciences, 2018, 98: 106-123. [5] MOU H L, XIE J, FENG Z Y, et al.Review on the crashworthiness design and evaluation of fuselage structure for occupant survivability[J]. Progress in Aerospace Sciences, 2024, 148: 101001. [6]MAMALIS A G, ROBINSON M, MANOLAKOS D E, et al.Crashworthy capability of composite material structures[J].Composite Structures, 1997, 37(2):109-134 [7]HULL D.A unified approach to progressive crushing of fibre-reinforced composite tubes[J].Composites Science and Technology, 1991, 40(4):377-421 [8]PALANIVELU S, VAN PAEPEGEM W, DEGRIECK J, et al.Parametric study of crushing parameters and failure patterns of pultruded composite tubes using cohesive elements and seam,Part I: Central delamination and triggering modelling[J].Polymer Testing, 2010, 29(6):729-741 [9] LIU X C, BAI C Y, XI X L, et al.Impact response and crashworthy design of composite fuselage structures: An overview[J]. Progress in Aerospace Sciences, 2024, 148: 101002. [10] MOU H L, XIE J, LIU Y, et al.Impact test and numerical simulation of typical sub-cargo fuselage sec-tion of civil aircraft[J]. Aerospace Science and Technology, 2020, 107: 106305. [11] MOU H L, XIE J, FENG Z Y.Research status and future development of crashworthiness of civil aircraft fuselage structures: An overview[J]. Progress in Aerospace Sciences, 2020, 119: 100644. [12] CHIU L N S, FALZON B G, RUAN D, et al.Crush responses of composite cylinder under quasi-static and dynamic loading[J]. Composite Structures, 2015, 131: 90-98. [13] XIN Z B, DUAN Y G, ZHOU J, et al.Effect of tailored plies on the energy absorption capability of square CFRP tubes with discontinuous fibers[J]. Composite Structures, 2019, 209: 150-159. [14] WANG Y F, FENG J S, WU J H, et al.Effects of fiber orientation and wall thickness on energy absorption characteristics of carbon-reinforced composite tubes under different loading conditions[J]. Composite Structures, 2016, 153: 356-368. [15] LIU X, BELKASSEM B, JONET A, et al.Experimental investigation of energy absorption behaviour of circular carbon/epoxy composite tubes under quasi-static and dynamic crush loading[J]. Composite Structures, 2019, 227: 111266. [16] MCGREGOR C, VAZIRI R, POURSARTIP A, et al.Axial crushing of triaxially braided composite tubes at quasi-static and dynamic rates[J]. Composite Structures, 2016, 157: 197-206. [17] ZHANG J F, WANG L M, PENG D M, et al.Crashworthiness optimization of a sandwich tube filled with CFRP sinusoidal corrugated board[J]. Aerospace Science and Technology, 2023, 132: 108065. [18] SIROMANI D.Crashworthy design and analysis of aircraft structures[M]. Drexel University, 2013: 116-142 [19] LV R, REN Y R.Improving the crashworthiness of carbon fiber-reinforced composite channel section structure through controllable failure mechanisms[J]. Aerospace Science and Technology, 2023, 141: 108582. [20] RICCIO A, RAIMONDO A, DI CAPRIO F, et al.Experimental and numerical investigation on the crashworthiness of a composite fuselage sub-floor support system[J]. Composites Part B: Engineering, 2018, 150: 93-103. [21] RICCIO A, SAPUTO S, SELLITTO A, et al.On the crashworthiness behaviour of a composite fuselage Sub-floor component[J]. Composite Structures, 2020, 234: 111662. [22]牟浩蕾, 刘兴炎, 刘冰, 等.不同缘条宽度复合材料C型柱轴向压缩吸能特性[J].航空材料学报, 2023, 43(6):107-116 [23] FERABOLI P, SPETZLER M.Design of energy-absorbing CFRP stanchions for the cargo floor structure of transport category airplanes[C]//JAMS 2013 Technical Review Meeting, Washington D.C., 2013: 1-21. [24]解江, 宋山山, 宋东方, 等.复合材料 型柱轴压失效分析的层合壳建模方法[J].航空学报, 2019, 40(2):522395- [25]FENG Z Y, XIE J, SONG S S, et al.The failure mechanism and energy-absorbing characteristics of composite thin-walled C-channels subject to low-speed axial compression[J].Journal of Composite Ma-terials, 2019, 53(16):2249-2259 [26]解江, 宋山山, 牟浩蕾, 等.典型螺栓连接薄壁型柱轴压失效行为:试验及数值模拟[J].复合材料学报, 2021, 38(10):3361-3372 [27]汪洋, 吴志斌, 刘富.复合材料货舱地板立柱压溃响应试验[J].复合材料学报, 2020, 37(9):2200-2206 [28] OSTLER D, BLESSING E, PERL M.A building block approach for crashworthiness testing of composites[C]//JAMS 2019 Technical Review Meeting, Washington D.C., 2019: 1-30. [29] REINER J, VAZIRI R, ZOBEIRY N.Machine learning assisted characterisation and simulation of compressive damage in composite laminates[J]. Composite Structures, 2021, 273: 114290. [30] SHIROLKAR N, PATWARDHAN P, RAHMAN A, et al.Investigating the efficacy of machine learning tools in modeling the continuous stabilization and carbonization process and predicting carbon fiber properties[J]. Carbon, 2021, 174: 605-616. [31]YANG Z Z, YU C H, BUEHLER M J.Deep learning model to predict complex stress and strain fields in hierarchical composites[J].Science Advances, 2021, 7(15):eabd7416- [32] ZHANG Z, ZHANG K, ZHANG G, et al.Low-velocity impact simulation of carbon fiber reinforced composite laminate using IFF-criterion based on BP-ANN[J]. Aerospace Science and Technology, 2024, 148: 109095. [33]杨智春, 杨特.动载荷识别物理嵌入式神经网络模型与方法[J].航空学报, 2025, 46(5):531450- [34]樊云翔, 艾化楠, 王明振, 等.基于深度学习的水上飞机非定常水载荷重构[J].航空学报, 2024, 45(20):129882- [35] WANG W H, WANG H, ZHOU J N, et al.Machine learning prediction of mechanical properties of braided-textile reinforced tubular structures[J]. Materials & Design, 2021, 212: 110181. [36] LIANG H Y, LU C D, PU Y F, et al.Data-driven initial peak crushing force prediction of hybrid tubes[J]. International Journal of Mechanical Sciences, 2024, 271: 109141. [37]MAHDI E S, EL KADI H.Crushing behavior of laterally compressed composite elliptical tubes: Experiments and predictions using artificial neural networks[J].Composite Structures, 2008, 83(4):399-412 [38] XU T, XU P, YANG C X, et al.An LSTM-stacked autoencoder multisource response prediction and constraint optimization for scaled expansion tubes[J]. Applied Soft Computing, 2024, 153: 111285. [39] MIRJALILI S, LEWIS A.The whale optimization algorithm[J]. Advances in Engineering Software, 2016, 95: 51-67. [40]WANG B R, GAO D Y, GU H Y, et al.Guided Wave Based Composite Structural Fatigue Damage Moni-toring Utilizing the WOA-BP Neural Network[J].Computers, Materials & Continua, 2025, 83(1):455-473 [41] JIANG J Q, XU G B, WANG H Z, et al.High-accuracy road surface condition detection through multi-sensor information fusion based on WOA-BP neural network[J]. Sensors and Actuators A: Physical, 2024, 378: 115829. [42] GAO Z T, JIN Y H, ZHANG Y, et al.Static EIS multi-frequency feature points combined with WOA-BP neural network for Li-ion battery SOH estimation[J]. Measurement, 2025, 253: 117780. [43]HOCHREITER S, SCHMIDHUBER J.Long short-term memory[J].Neural Computation, 1997, 9(8):1735-1780 [44] SRIVASTAVA N, MANSIMOV E, Salakhudinov R.Unsupervised learning of video representations using lstms[DB/OL]. arXiv preprint :1502.04681, 2015.
文章导航

/