A new strategy for data-driven damage diagnosis of shear structures adapted to ambient vibration

被引:2
作者
Zhang, Xuan [1 ]
Li, Luyu [1 ,2 ]
机构
[1] Dalian Univ Technol, State Key Lab Coastal & Offshore Engn, Dalian, Peoples R China
[2] Dalian Univ Technol, Dalian, Peoples R China
关键词
Structural health monitoring; Data-driven; Ambient excitation; Unsupervised learning; ARMAX model; Mel-frequency cepstral coefficients; STATISTICAL PATTERN-RECOGNITION; FEATURE-EXTRACTION; ARMAX MODEL;
D O I
10.1016/j.measurement.2024.114257
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Since many critical infrastructures of large civil engineering are shear structures and are exposed to ambient excitation most of the operational time, it is crucial to utilize ambient vibration data for damage diagnosis of shear structures. However, some autoregressive moving average with exogenous inputs (ARMAX) model methods will face challenges under ambient excitation due to assumptions about external loads. To address this issue, we propose a new data -driven strategy. Firstly, pseudo -free responses of shear structures are calculated from correlation signals. Subsequently, substructure ARMAX models are constructed using pseudofree responses, and model residuals are extracted as damage -sensitive features (DSFs). Furthermore, to improve the accuracy of the anomaly diagnosis, the Mel -frequency cepstral coefficients (MFCCs) are employed for dimensionality reduction of model residuals and are combined with Mahalanobis-squared distance (MSD) to form a hybrid distance method. Finally, the practicality and superiority of the strategy are validated using numerical and experimental benchmark structures.
引用
收藏
页数:16
相关论文
共 56 条
[1]   A fast and efficient feature extraction methodology for structural damage localization based on raw acceleration measurements [J].
Alves, Victor ;
Cury, Alexandre .
STRUCTURAL CONTROL & HEALTH MONITORING, 2021, 28 (07)
[2]   A structural health monitoring strategy using cepstral features [J].
Balsamo, L. ;
Betti, R. ;
Beigi, H. .
JOURNAL OF SOUND AND VIBRATION, 2014, 333 (19) :4526-4542
[3]   Integrated ARMA model method for damage detection of subsea pipeline system [J].
Bao, Chunxiao ;
Hao, Hong ;
Li, Zhong-Xian .
ENGINEERING STRUCTURES, 2013, 48 :176-192
[4]   Efficient FPGA-based architecture of an automatic wheeze detector using a combination of MFCC and SVM algorithms [J].
Boujelben, Ons ;
Bahoura, Mohammed .
JOURNAL OF SYSTEMS ARCHITECTURE, 2018, 88 :54-64
[5]   Natural excitation technique and eigensystem realization algorithm for phase I of the IASC-ASCE benchmark problem: Simulated data [J].
Caicedo, JM ;
Dyke, SJ ;
Johnson, EA .
JOURNAL OF ENGINEERING MECHANICS, 2004, 130 (01) :49-60
[6]   New supervised learning classifiers for structural damage diagnosis using time series features from a new feature extraction technique [J].
Chegeni, Masoud Haghani ;
Sharbatdar, Mohammad Kazem ;
Mahjoub, Reza ;
Raftari, Mahdi .
EARTHQUAKE ENGINEERING AND ENGINEERING VIBRATION, 2022, 21 (01) :169-191
[7]   A data-driven approach for drive-by damage detection in bridges considering the influence of temperature change [J].
Corbally, Robert ;
Malekjafarian, Abdollah .
ENGINEERING STRUCTURES, 2022, 253
[8]   Structural health monitoring using high-dimensional features from time series modeling by innovative hybrid distance-based methods [J].
Daneshvar, Mohammad Hassan ;
Gharighoran, Alireza ;
Zareei, Seyed Alireza ;
Karamodin, Abbas .
JOURNAL OF CIVIL STRUCTURAL HEALTH MONITORING, 2021, 11 (02) :537-557
[9]   Early damage detection under massive data via innovative hybrid methods: application to a large-scale cable-stayed bridge [J].
Daneshvar, Mohammad Hassan ;
Gharighoran, Alireza ;
Zareei, Seyed Alireza ;
Karamodin, Abbas .
STRUCTURE AND INFRASTRUCTURE ENGINEERING, 2021, 17 (07) :902-920
[10]  
Dyke S.J., 2003, P 16 ASCE ENG MECH C