1.中国海洋大学工程学院,山东 青岛 266100
2.青岛科技大学信息学院,山东 青岛 266100
郑文言(1997—),男,硕士研究生。主要从事结构抗爆性能研究。E-mail: zhengwenyan@stu.ouc.edu.cn
方辉(1980—),男,教授,博导,博士。主要从事结构损伤方面的研究。E-mail: fanghui@ouc.edu.cn
收稿:2021-11-22,
修回:2022-03-21,
纸质出版:2023-08-28
移动端阅览
郑文言,段利亚,方辉.近距离爆炸下钢筋混凝土柱损伤的图像识别及快速评估[J].防灾减灾工程学报,2023,43(04):712-720.
ZHENG Wenyan,DUAN Liya,FANG Hui.Rapid Assessment of Reinforced Concrete Column Damage After Near‑field Explosion Based on Image Recognition[J].Journal of Disaster Prevention and Mitigation Engineering,2023,43(04):712-720.
郑文言,段利亚,方辉.近距离爆炸下钢筋混凝土柱损伤的图像识别及快速评估[J].防灾减灾工程学报,2023,43(04):712-720. DOI: 10.13409/j.cnki.jdpme.20211122002.
ZHENG Wenyan,DUAN Liya,FANG Hui.Rapid Assessment of Reinforced Concrete Column Damage After Near‑field Explosion Based on Image Recognition[J].Journal of Disaster Prevention and Mitigation Engineering,2023,43(04):712-720. DOI: 10.13409/j.cnki.jdpme.20211122002.
快速准确评估工事爆炸损伤,对于战时指挥具有重要意义。传统上,爆后结构损伤评估多采用有限元法,可靠性得到广泛认可,但计算耗时较长,无法满足战场快速评估要求。为提高评估效率,提出一种由体积判断网络和损伤判断网络组成的全新卷积神经网络模型,基于体积损失率评估近距离爆炸下的损伤。为降低训练成本,采用有限元法获取爆后带损伤的钢筋混凝土柱图像制作包含8 700张图片的数据集。新提出的模型在包含1 740张图片的测试集上取得99.71%的准确率。在钢筋混凝土柱材料参数调整时,微调后的模型在不同材料参数的数据集上取得82.97%以上的准确率。在三维打印的模型测试中,该模型在24张图片上取得70.83%的正确率,平均每次评估测试耗时0.05 s。结果表明,提出的卷积神经网络模型准确率较高,计算用时远小于有限元方法,损伤评估流程可与结构理论对照解释。
During wartime, rapid and accurate assessment of fortification explosion damage is of great significance. Traditionally, the finite element method (FEM) is used for damage assessment after the near-field explosion, and its reliability is widely recognized. However, the calculation takes a long time and cannot meet the requirement of rapid battlefield assessment. In order to improve the evaluation efficiency, a new convolution neural network(CNN) model composed of a volume assessment network and a damage judgment network is established to evaluate the damage based on the volume loss rate. In order to reduce the training cost, a damaged reinforced concrete (RC) column model after the explosion is simulated by FEM. The images of the RC column are automatically obtained by script, and a data set containing 8 700 pictures is made. The CNN model achieves an accuracy of 99.71% on a test set containing 1 740 pictures. The accuracy of the CNN model after fine-tuning is more than 82.97% on the data set of different material parameters. In the test of the 3D printing model, the accuracy of the model is 70.83% on 24 pictures, and the average time of each evaluation test is 0.05 s. The accuracy of the proposed CNN model is reliable, and the calculation time is much less than that of FEM. The damage assessment process can be explained in comparison with the structural theory.
Shi Y C , Hao H , Li Z X . Numerical derivation of pressure‑impulse diagrams for prediction of RC column damage to blast loads [J]. International Journal of Impact Engineering , 2008 , 35 ( 11 ): 1213 - 1227 .
李殷 . 建筑物爆炸破坏快速评估技术研究 [D]. 长沙 : 国防科学技术大学 , 2015 .
Li Y . Study on rapid assessments method of damage to buildings under blast loading [D]. Changsha : National University of Defense Technology , 2015 . (in Chinese)
Bao X L , Li B . Residual strength of blast damaged reinforced concrete columns [J]. International Journal of Impact Engineering , 2009 , 37 ( 3 ): 295 - 308 .
阎石 , 刘蕾 , 齐宝欣 , 等 . 爆炸荷载作用下方钢管混凝土柱的动力响应及破坏机理 [J]. 防灾减灾工程学报 , 2011 , 31 ( 5 ): 477 - 482 .
Yan S , Liu L , Qi B X , et al . Dynamic response and failure mode analysis of concrete infilled rectangular steel tube columns under blasting loading [J]. Journal of Disaster Prevention and Mitigation Engineering , 2011 , 31 ( 5 ): 447 - 482 . (in Chinese)
Cui J , Shi Y , Li Z , et al . Failure analysis and damage assessment of RC columns under close-in explosions [J]. Journal of Performance of Constructed Facilities , 2015 , 29 ( 5 ): B4015003 .
师燕超 , 李绍琦 , 李忠献 , 等 . 基于实测频率的钢筋混凝土柱爆炸损伤快速评估方法 [J]. 建筑结构学报 , 2021 , 42 ( 11 ): 155 - 164 .
Shi Y C , Li S Q , Li Z X , et al . Rapid evaluation method for blast damage of reinforced concrete columns based on measured frequency [J]. Journal of Building Structures , 2021 , 42 ( 11 ): 155 - 164 . (in Chinese)
Hu J , Shen L , Albanie S , et al . Squeeze-and-Excitation Networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2020 , 42 ( 8 ): 2011 - 2023 .
Xie S , Girshick R , Dollár P , et al . Aggregated residual transformations for deep neural networks [C]∥ 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Honolulu, HI, USA : IEEE , 2017 : 5987 - 5995 .
Howard A , Sandler M , Chen B , et al . Searching for mobileNetV3 [C]∥ 2019 IEEE/CVF International Conference on Computer Vision (ICCV) . Seoul, South Korea : IEEE , 2019 : 1314 - 1324 .
Dusenberry D , Schmidt J , Hobelmann P , et al . Blast protection of buildings, ASCE/SEI 59-11 [M]. Reston, VA : American Society of Civil Engineers , 2011 .
李明鸿 . 双层钢管混凝土组合墩柱爆炸损伤机理和评估方法研究 [D]. 南京 : 东南大学 , 2020 .
Li M H . Damage mechanism and performance assessment of concrete-Filled double-Skin steel tubular columns subjected to blast loads [D]. Nanjing : Southeast University , 2020 . (in Chinese)
Canny J . A computational approach to edge detection [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 1986 , PAMI-8( 6 ): 679 - 698 .
Kingma D P , Ba J . Adam: A method for stochastic optimization [J]. Journal of Machine Learning Research , 2017 , 18 ( 1 ): 1 - 37 .
Rumelhart D , Hinton G E , Williams R J . Learning Representations by Back Propagating Errors [J]. Nature , 1986 , 323 ( 6088 ): 533 - 536 .
Krizhevsky A , Sutskever I , Hinton G . ImageNet classification with deep convolutional neural networks [J]. Advances in Neural Information Processing Systems , 2012 , 25 ( 2 ): 84 - 90 .
0
浏览量
0
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
苏公网安备32010202012147号
