• 首页 | 期刊简介 | 编委会 | 投稿须知 | 出版道德规范 | 下载专区 | English
夏桂然,魏敦涛,刘泽佳*,周立成,刘逸平,汤立群.基于深度学习的黏钢构件黏接层损伤识别方法[J].实验力学,2024,39(4):399~412
基于深度学习的黏钢构件黏接层损伤识别方法
Damage identification method of adhesive layer of adhesive steel members based on deep learning
投稿时间:2023-12-20  修订日期:2024-02-05
DOI:10.7520/1001-4888-23-268
中文关键词:  深度学习  黏钢加固结构  超声检测  损伤识别
英文关键词:deep learning  reinforced steel structure  ultrasonic detection  damage identification
基金项目:广东省自然科学基金项目(2023A1515012942)
作者单位
夏桂然 华南理工大学 土木与交通学院 广东广州 510641 
魏敦涛 广州公路工程集团有限公司 广东广州 510075 
刘泽佳* 华南理工大学 土木与交通学院 广东广州 510641 
周立成 华南理工大学 土木与交通学院 广东广州 510641 
刘逸平 华南理工大学 土木与交通学院 广东广州 510641 
汤立群 华南理工大学 土木与交通学院 广东广州 510641 
摘要点击次数: 1843
全文下载次数: 533
中文摘要:
      由于传统的黏钢加固结构检测方法过分依赖于人工观察和经验判断,其结果容易受到检测人员主观因素的影响。为了提高黏钢结构黏接层损伤检测的准确性、可靠性,本文提出了一种基于深度学习的黏钢构件黏接层损伤识别方法。首先,通过黏钢构件超声检测实验验证了有限元模型的有效性,并在此基础上生成大量损伤工况下的超声时程响应数据;随后,将不同工况下的超声时程响应作为深度学习模型的输入,分别对黏钢构件黏接层进行损伤分类、缺陷大小和位置识别,以及老化程度的定量识别;最后,比较和分析了BiLSTM、BiGRU和1D-CNN模型在各个识别任务下的准确率和鲁棒性,并以此为依据建立了基于深度学习的黏钢构件黏接层损伤识别方法。结果表明:所选用的3种模型均能精准实现损伤类型的区分,准确率达到100%;在20dB强噪声条件下,BiLSTM模型在2种缺陷定量精度需求和1种缺陷定位精度需求中,准确率仍分别可达93.33%、86.48%和73.31%;而在相同噪声条件下,1D-CNN模型在3种老化程度检测精度需求下,准确率仍分别可达93.95%、88.09%和79.85%。
英文摘要:
      Due to the traditional inspection methods of cemented steel structures rely too much on manual observation and empirical judgment, and the results are easily affected by subjective factors of inspection personnel. In order to improve the accuracy and reliability of adhesive layer damage detection of adhesive steel structure, a method of adhesive layer damage identification based on deep learning was proposed in this paper. Firstly, the validity of the finite element model was verified by ultrasonic testing of the steel members, and a large number of ultrasonic time-history response data under damaged conditions were generated on this basis. Then, the ultrasonic time-history response under different working conditions was used as the input of the deep learning model to classify the damage, identify the size and location of defects, and quantitatively identify the degree of aging of the bonding layer of the adhesive steel member respectively. Finally, the accuracy and robustness of BiLSTM, BiGRU and 1D-CNN models in each identification task were compared and analyzed, and a deep learn-based damage identification method for the bonding layer of adhesive steel components was established. The results show that the three models can accurately distinguish damage types with an accuracy of 100%. Under the condition of 20dB strong noise, the accuracy of BiLSTM model can reach to 93.33%, 86.48% and 73.31% respectively in two kinds of defect quantitative accuracy requirements and defect location. Under the same noise condition, the accuracy of 1D-CNN model can reach to 93.95%, 88.09% and 79.85% respectively under three kinds of aging degree detection accuracy requirements.
查看全文  下载PDF阅读器
关闭

网站版权:《实验力学》编辑部
您是本站第 65849147 位访问者,今日一共访问943次,当前在线人数: 0
技术支持:本系统由北京勤云科技发展有限公司设计