为快速构建并准确预测温度作用引起的斜拉桥主梁应变用于结构状态评估，基于某大跨度斜拉桥主梁超过1年的温度和应变监测数据，提出了一种基于迁移学习和双向长短时记忆(bi-directional long short-term memory, Bi-LSTM)神经网络的斜拉桥温度-应变映射模型建立方法。首先，利用解析模态分解(analytical mode decomposition，AMD)去噪应变数据，得到仅由温度引起的应变响应；其次，选择温度和某一测点应变数据构成数据集，采用Bi-LSTM神经网络训练该数据集，并通过网络结构和超参数优化建立温度-应变Bi-LSTM基准模型；最后，利用迁移学习方法，将已训练好的基准模型中部分参数迁移到其它温度-应变数据集，建立相应的温度-应变映射被迁移模型，并与未采用迁移学习的神经网络训练方法进行对比。研究结果表明：相比直接建立的温度-应变Bi-LSTM神经网络映射模型，采用迁移学习方法建立的被迁移模型，其拟合精度均高于所用的基准模型，且训练时间短，预测误差小。
Abstract
To rapidly construct and accurately predict the strain responses of the main girder induced by temperature in the long-span cable-stayed bridge for structural condition assessment, based on the measured temperature and strain data on the main girder of a long-span cable-stayed bridge over 1 year, a method of constructing the temperature-strain mapping model by using the transfer learning technique and the bidirectional long short-term memory (Bi-LSTM) neural networks is proposed in this study. Firstly, the analytical mode decomposition (AMD) is adopted to denoise the strain data to obtain the temperature-induced strain. Secondly, the temperature and the strain data at a particular measurement point were selected to form a dataset, and were fed to a Bi-LSTM neural network. Then a well-fitting neural network baseline model is constructed by optimizing the network structure and hyperparameters. Finally, using the transfer learning method, some parameters from the trained Bi-LSTM neural network model are transferred to other temperature-strain datasets to construct the transferred temperature-strain mapping models. Compared with the temperature-strain Bi-LSTM neural network models constructed directly from the datasets, the transferred temperature-strain Bi-LSTM neural network models built by using the transfer learning technique have higher fitting accuracy, shorter training time and smaller prediction error.
关键词
结构健康监测 /
大跨度斜拉桥 /
温度-应变映射模型 /
迁移学习 /
双向长短时记忆(Bi-LSTM)神经网络
{{custom_keyword}} /
Key words
structural health monitoring /
long-span cable-stayed bridge /
temperature-strain mapping model /
transfer learning /
bi-directional long short-term memory (Bi-LSTM) neural network
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 孙雅琼, 赵作周. 桥梁结构动应变监测的温度效应实时分离与动荷载识别[J]. 工程力学, 2019, 36(2): 186-194.
Sun Yaqiong, Zhao Zuozhou. Real-time separation of temperature effect on dynamic strain monitoring and moving load identification of bridge structure[J]. Engineering mechanics, 2019, 36(2): 186-194.
[2] Fan Z, Huang Q, Ren Y, et al. Real-time dynamic warning on deflection abnormity of cable-stayed bridges considering operational environment variations[J]. Journal of performance of constructed facilities, 2021, 35(1): 04020123.
[3] Catbas F N, Susoy M, Frangopol D M. Structural health monitoring and reliability estimation: long span truss bridge application with environmental monitoring data[J]. Engineering Structures, 2008, 30(9): 2347-2359.
[4] 吴佰建, 李兆霞, 王滢. 桥梁结构动态应变监测信息的分离与提取[J]. 东南大学学报（自然科学版）, 2008, 38(5): 767-773.
Wu Baijian, Li Zhaoxia, Wang Ying. Separation and extraction of bridge dynamic strain data[J]. Journal of Southeast University (English Edition), 2008, 38(5): 767-773.
[5] 周毅, 孙利民, 闵志华. 斜拉桥主梁应变监测数据分析[J]. 振动与冲击, 2011, 30(4): 230-235.
Zhou Yi, Sun Limin, Min Zhihua. Girder strain analysis of a cable-stayed bridge[J]. Journal of Vibration and Shock, 2011, 30(4): 230-235.
[6] 李苗, 任伟新, 胡异丁, 等. 基于解析模态分解法的桥梁动态应变监测数据温度影响的分离[J]. 振动与冲击, 2012, 31(21): 6-10.
Li Miao, Ren Weixin, Hu Yiding, et al. Separating temperature effect from dynamic strain measurements of a bridge based on analytical mode decomposition method[J]. Journal of Vibration and Shock, 2012, 31(21): 6-10.
[7] Xia Y, Chen B, Zhou X Q, et al. Field monitoring and numerical analysis of Tsing Ma Suspension Bridge temperature behavior[J]. Structural Control and Health Monitoring, 2013, 20(4): 560-575.
[8] Kromanis R, Kripakaran P. Predicting thermal response of bridges using regression models derived from measurement histories[J]. Computers & Structures, 2014, 136(3): 64-77.
[9] 刘泽佳, 陈溢涛, 周立成, 等. 桥梁长期健康监测大数据温度与应变特征及关联性分析[J]. 科学技术与工程, 2018, 18(35): 72-79.
Liu Zejia, Chen Yitao, Zhou Licheng, et al. Analysis of characteristics and correlation for temperature and strain based on long-term bridge health monitoring big data[J]. Science Technology and Engineering, 2018, 18(35): 72-79.
[10] 郑秋怡, 周广东, 刘定坤. 基于长短时记忆神经网络的大跨拱桥温度-位移相关模型建立方法[J]. 工程力学, 2021, 38(4): 68-79.
Zheng Qiuyi, Zhou Guangdong, Liu Dingkun. Method of modeling temperature-displacement correlation for long-span arch bridges based on long short-term memory neural networks[J]. Engineering mechanics, 2021, 38(4): 68-79.
[11] Zhao H W, Ding Y L, Li A Q, et al. Digital modeling on the nonlinear mapping between multi-source monitoring data of in-service bridges[J]. Structural Control and Health Monitoring, 2020, 27(11): e2618.
[12] Yue Z, Ding Y, Zhao H, et al. Case Study of Deep Learning Model of Temperature-Induced Deflection of a Cable-Stayed Bridge Driven by Data Knowledge[J]. Symmetry, 2021, 13(12): 2293.
[13] Pratt L Y. Discriminability-based transfer between neural networks[J]. Advances in neural information processing systems, 1993: 204-211.
[14] Zhuang F, Qi Z, Duan K, et al. A comprehensive survey on transfer learning[J]. Proceedings of the IEEE, 2020, 109(1): 43-76.
[15] Aliyari M, Droguett E L, Ayele Y Z. Uav-based bridge inspection via transfer learning[J]. Sustainability, 2021, 13(20): 11359.
[16] Lee J S, Kim H M, Kim S I, et al. Evaluation of structural integrity of railway bridge using acceleration data and semi-supervised learning approach[J]. Engineering Structures, 2021, 239: 112330.
[17] Lin Y, Nie Z, Ma H. Dynamics-based cross-domain structural damage detection through deep transfer learning[J]. Computer-Aided Civil and Infrastructure Engineering, 2022, 37(1): 24-54.
[18] 徐立芳, 莫宏伟. 深度学习[M]. 北京: 人民邮电出版社, 2020.
Xu Lifang, Mo Hongwei. Deep Learning[M]. Beijing: Posts & Telecom Press, 2020.
[19] Google. Tensorflow. https://www.tensorflow.org/. Accessed 2022.
[20] Chen G, Wang Z. A signal decomposition theorem with Hilbert transform and its application to narrowband time series with closely spaced frequency components[J]. Mechanical systems and signal processing, 2012, 28: 258-279.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(3440 KB)
