«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《建筑科学与工程学报》[ISSN:1673-2049/CN:61-1442/TU]

卷:

42卷

期数:

2025年02期

页码:

27-38

栏目:

建筑结构

出版日期:

2025-03-20

文章信息/Info

Title:

Intelligent detection of post earthquake damage anomaly of RC frame structure based on wavelet scattering transform

文章编号:

1673-2049(2025)02-0027-12

作者:

康帅¹,王自法²,周荣环³,贺东青¹,俞叶¹,李一民¹

(1. 河南大学 建筑工程学院,河南 开封 475004; 2. 中国地震局工程力学研究所,黑龙江 哈尔滨 150080; 3. 中建八局发展建设有限公司,河南 郑州 450000)

Author(s):

KANG Shuai¹, WANG Zifa², ZHOU Ronghuan³, HE Dongqing¹, YU Ye¹, LI Yimin¹

(1.School of Civil Engineering and Architecture, Henan University, Kaifeng 475004, Henan, China; 2. Institute of Engineering Mechanics, China Earthquake Administration, Harbin 150080, Heilongjiang, China; 3. China Construction Eighth Bureau Development and Construction Co., Ltd., Zhengzhou 450000, Henan, China)

关键词:

小波散射变换;  异常检测;  震后损伤;  深度学习;  框架结构

Keywords:

wavelet scattering transform;  abnormal detection;  post earthquake damage;  deep learning;  frame structure

分类号:

TU375

DOI:

10.19815/j.jace.2023.10033

文献标志码:

A

摘要:

基于传统方法的地震损伤评估存在精度和效率低的问题,为实现基于结构地震响应观测数据的震后结构损伤快速评估,并有效解决数据类型不平衡的问题,提出基于时频分析与卷积自编码(CAE)模型的损伤异常数据检测方法。首先应用小波散射变换对原时域信号进行处理,生成时频数据; 然后建立相应的卷积自编码网络模型,将时频数据输入CAE模型进行重构训练,根据重构误差确定异常判断阈值; 基于该阈值精准区分数据中的异常值,并将计算结果与基于时域输入方法的计算结果进行对比分析,最后验证该方法在噪声环境下的有效性。结果表明:基于CAE模型的异常检测方法可以很好地识别损伤数据集中的异常序列,召回率达到了90%以上,且在小波散射变换作用下的异常检测效果更好; 时频分析结合CAE模型的损伤异常数据检测方法极大地提升了震后损伤评估的效率,基于小波散射变换的耗时仅为基于传统时域输入方法的1/3,该方法在噪声作用下也表现出较高的检测精度。

Abstract:

Earthquake damage assessment based on traditional methods has the problems of low accuracy and efficiency. In order to achieve rapid assessment of post earthquake structural damage based on structural seismic response observation data and effectively solve the problem of data type imbalance, a damage abnormal data detection method based on time-frequency analysis and convolutional autoencoder(CAE)model was proposed. Firstly, the wavelet scattering transform was applied to process the original time domain signal to generate time-frequency data. Then, the corresponding convolutional autoencoder network model was established, and the time-frequency data was input into the CAE model for reconstruction training. The anomaly judgment threshold was determined according to the reconstruction error. Using this threshold, anomalies in the dataset were accurately distinguished, and the results were compared with those obtained using direct time-domain input method. Finally, the effectiveness of this method under noisy conditions was validated. The results show that the anomaly detection method based on the CAE model can well identify abnormal sequences in the damage data set, achieving a recall rate of over 90%, and the anomaly detection effect under the action of wavelet scattering transform is better. The proposed damage anomaly detection method integrating time-frequency analysis with the CAE model can significantly enhance the efficiency of post earthquake damage assessment. The computation time using wavelet scattering transform is only one-third of that using traditional time-domain input methods, and the method maintains high detection accuracy even in noisy environment.

参考文献/References:

[1] 刘洪波,佟 瑶,蒋垚俊,等.RC框架结构地震易损性分析方法研究进展[J].世界地震工程,2020,36(3):141-150.
LIU Hongbo, TONG Yao, JIANG Yaojun, et al. Recent development of seismic fragility analysis methods for RC frame structures[J]. World Earthquake Engineering, 2020, 36(3): 141-150.
[2]FEMA. Multi-hazard loss estimation methodology HAZUS-MH 2.1 advanced engineering building module(AEBM)technical and user's manual[M]. Washington DC: Federal Emergency Management Agency, 2012.
[3]LU X Z, HAN B, HORI M, et al. A coarse-grained parallel approach for seismic damage simulations of urban areas based on refined models and GPU/CPU cooperative computing[J]. Advances in Engineering Software, 2014, 70: 90-103.
[4]LU X Z, CHENG Q L, XU Z, et al. Real-time city-scale time-history analysis and its application in resilience-oriented earthquake emergency responses[J]. Applied Sciences, 2019, 9(17): 3497.
[5]WU R T, JAHANSHAHI M R. Deep convolutional neural network for structural dynamic response estimation and system identification[J]. Journal of Engineering Mechanics, 2019, 145(1): 04018125.
[6]YU Y, WANG C Y, GU X Y, et al. A novel deep learning-based method for damage identification of smart building structures[J]. Structural Health Monitoring, 2019, 18(1): 143-163.
[7]戴 伦,张文达,田石柱.桥梁线弹性地震反应的卷积神经网络估计[J].地震工程与工程振动,2021,41(4):188-195.
DAI Lun, ZHANG Wenda, TIAN Shizhu. Convolutional neural network estimation of bridge linear elastic seismic response[J]. Earthquake Engineering and Engineering Dynamics, 2021, 41(4): 188-195.
[8]AHMED B, MANGALATHU S, JEON J S. Seismic damage state predictions of reinforced concrete structures using stacked long short-term memory neural networks[J]. Journal of Building Engineering, 2022, 46: 103737.
[9]周 玉,孙红玉,房 倩,等.不平衡数据集分类方法研究综述[J].计算机应用研究,2022,39(6):1615-1621.
ZHOU Yu, SUN Hongyu, FANG Qian, et al. Review of imbalanced data classification methods[J]. Application Research of Computers, 2022, 39(6): 1615-1621.
[10]卓 琳,赵厚宇,詹思延.异常检测方法及其应用综述[J].计算机应用研究,2020,37(增1):9-15.
ZHUO Lin, ZHAO Houyu, ZHAN Siyan. Overview of anomaly detection methods and their applications[J]. Application Research of Computers, 2020, 37(S1): 9-15.
[11]来 杰,王晓丹,向 前,等.自编码器及其应用综述[J].通信学报,2021,42(9):218-230.
LAI Jie, WANG Xiaodan, XIANG Qian, et al. Review on autoencoder and its application[J]. Journal on Communications, 2021, 42(9): 218-230.
[12]李清勇,王建柱,祝叶舟,等.基于结构相似深度卷积自编码的异常扣件检测模型[J].交通运输工程学报,2022,22(4):186-195.
LI Qingyong, WANG Jianzhu, ZHU Yezhou, et al. Anomalous fastener detection model based on deep convolutional autoencoder with structural similarity[J]. Journal of Traffic and Transportation Engineering,2022,22(4):186-195.
[13]AN J, CHO S. Variational autoencoder based anomaly detection using reconstruction probability[R]. Seoul: SNU Big Data AI Center, 2015.
[14]NIU Z J, YU K, WU X F. LSTM-based VAE-GAN for time-series anomaly detection[J]. Sensors, 2020, 20(13): 3738.
[15]邵世宽,张宏钧,肖钦锋,等.基于无监督对抗学习的时间序列异常检测[J].南京大学学报(自然科学),2021,57(6):1042-1052.
SHAO Shikuan, ZHANG Hongjun, XIAO Qinfeng, et al. Time series anomaly detection based on unsupervised adversarial learning[J]. Journal of Nanjing University(Natural Science), 2021, 57(6): 1042-1052.
[16]SOLEIMANI-BABAKAMALI M H, SOLEIMANI-BABAKAMALI R, SARLO R, et al. On the effectiveness of dimensionality reduction for unsupervised structural health monitoring anomaly detection[J]. Mechanical Systems and Signal Processing, 2023, 187: 109910.
[17]MALLAT S. Group invariant scattering[J]. Communications on Pure and Applied Mathematics, 2012, 65(10): 1331-1398.
[18]BRUNA J, MALLAT S. Invariant scattering convolution networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(8): 1872-1886.
[19]ANDEN J, MALLAT S. Deep scattering spectrum[J]. IEEE Transactions on Signal Processing, 2014, 62(16): 4114-4128.
[20]WIATOWSKI T, BOLCSKEI H. A mathematical theory of deep convolutional neural networks for feature extraction[J]. IEEE Transactions on Information Theory, 2018, 64(3): 1845-1866.
[21]RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[J]. Nature, 1986, 323: 533-536.
[22]RASTIN Z, GHODRATI AMIRI G, DARVISHAN E. Unsupervised structural damage detection technique based on a deep convolutional autoencoder[J]. Shock and Vibration, 2021, 2021(1): 6658575.
[23]常晓燕.基于深度神经网络的多维时间序列异常检测方法研究[D].上海:东华大学,2022.
CHANG Xiaoyan. Research on anomaly detection method of multidimensional time series based on deep neural network[D]. Shanghai: Donghua University, 2022.
[24]张 锐,李宏男,王东升,等.结构时程分析中强震记录选取研究综述[J].工程力学,2019,36(2):1-16.
ZHANG Rui, LI Hongnan, WANG Dongsheng, et al. Selection and scaling of real accelerograms as input to time-history analysis of structures: a state-of-the-art review[J]. Engineering Mechanics, 2019, 36(2): 1-16.
[25]乔云龙.框架结构地震反应的不确定性分析[D].哈尔滨:中国地震局工程力学研究所,2020.
QIAO Yunlong. Uncertainty analysis of seismic response of frame structure[D]. Harbin: Institute of Engineering Mechanics, China Earthquake Administration, 2020.
[26]王东超.结构地震易损性分析中地震动记录选取方法研究[D].哈尔滨:哈尔滨工业大学,2016.
WANG Dongchao. Study on selection method of ground motion records in structural seismic vulnerability analysis[D]. Harbin: Harbin Institute of Technology, 2016.
[27]Global topics report on the prestandard and commentary for the seismic rehabilitation of buildings: FEMA 357[S]. Washington DC: ASCE, 2000.
[28]杨 光.中美抗震规范RC结构层间位移角限值的对比研究[D].广州:华南理工大学,2018.
YANG Guang. Comparative study on the limit value of story drift angle of RC structures in Chinese and American seismic codes[D]. Guangzhou: South China University of Technology, 2018.
[29]吕大刚,于晓辉,王光远.基于单地震动记录IDA方法的结构倒塌分析[J].地震工程与工程振动,2009,29(6):33-39.
LÜ Dagang, YU Xiaohui, WANG Guangyuan. Structural collapse analysis based on single-record IDA method[J]. Journal of Earthquake Engineering and Engineering Vibration, 2009, 29(6): 33-39.
[30]缪惠全.加速度基线漂移时域处理方法的对比研究[J].地震工程与工程振动,2022,42(2):135-150.
MIAO Huiquan. Comparative study of time-domain processing methods of acceleration baseline drift[J]. Earthquake Engineering and Engineering Dynamics, 2022, 42(2): 135-150.

相似文献/References:

备注/Memo

备注/Memo:

收稿日期:2023-10-15
基金项目:国家自然科学基金项目(51978634); 河南省科技公关项目(232102321076)
作者简介:康 帅(1983-),男,工学博士,副教授,E-mail: kangshuai@henu.edu.cn。
通信作者:王自法(1965-),男,工学博士,教授,E-mail: zifa@iem.ac.cn。
Author resumes: KANG Shuai(1983-), male, PhD, associate professor, E-mail: kangshuai@henu.edu.cn; WANG Zifa(1965-), male, PhD, professor, E-mail: zifa@iem.ac.cn.

常用功能

更新日期/Last Update: 2025-03-20