Unsupervised learning-based framework for indirect structural health monitoring using adversarial autoencoder

Hurtado, A. Calderon; Kaur, K.; Alamdari, Mehrisadat Makki; Atroshchenko, Elena; Chang, Kai-Chun; Kim, C.W.

doi:10.1016/j.jsv.2023.117598

Cited by 11 publications

(7 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is because the size of the damage is equated to the element size for modeling damage through element stiffness loss. In this study, the chosen size for the damaged element is 0.5 m, a dimension commonly employed in the literature [10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28]. Consequently, a total of 50 elements constitute the 25 m simply supported beam.…”

Section: Numerical Validation Of the Residual Cp Responsementioning

confidence: 99%

“…Consequently, a total of 50 elements constitute the 25 m simply supported beam. In addition, a simulation time step of 0.001 s is also a common choice in VBI studies [10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26][27][28]. The first third frequencies of the bridge are ω b1 = 3.80 Hz, ω b2 = 15.22 Hz, and ω b3 = 34.24 Hz, respectively.…”

Section: Numerical Validation Of the Residual Cp Responsementioning

confidence: 99%

“…However, the application of the VSM in identifying bridge modal shapes and damage is primarily limited to theoretical investigations [15][16][17][18]. More recently, some laboratory tests on scaled VBI systems have attempted to fill this gap for validating the VSM in identifying bridge mode shapes and damage, but they are often conducted under ideal conditions [19][20][21][22]. The significant challenge in identifying bridge modal properties using the VSM arises from extracting weak bridge vibrational signals from vehicle responses amid the pollution caused by vehicle-induced vibrations and road surface roughness.…”

Section: Introductionmentioning

confidence: 99%

“…In the exploration of applying the VSM for indirect bridge damage identification, various studies have examined algorithms such as mode shape difference, wavelet transform, and deep learning [19][20][21]27]. Zhang et al [26] introduced the metric of instantaneous amplitude squared (IAS), derived from the Hilbert transform of isolated bridge vibration signals from vehicle responses, to detect bridge damage.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Theoretical and Experimental Investigations of Identifying Bridge Damage Using Instantaneous Amplitude Squared Extracted from Vibration Responses of a Two-Axle Passing Vehicle

Liu,

Zhou,

Zhang

et al. 2024

Buildings

View full text Add to dashboard Cite

Identifying bridge damage using a movable test vehicle is highly regarded for its mobility, cost-effectiveness, and broad monitoring coverage. Previous studies have shown that the residual contact-point (CP) response between connected vehicles is free of the impact of vehicle self-vibrations and road roughness, making it particularly suitable for the indirect extraction of bridge modal properties. However, most experimental campaigns regarding contact-point (CP) responses focus on a single-axle testing vehicle within a non-moving state. This study aims to theoretically and experimentally identify bridge damage using the instantaneous amplitude squared (IAS) extracted from the residual CP response of a two-axle passing vehicle. First, the closed-form solution of the residual CP acceleration was derived for a two-axle vehicle interacting with a simply supported beam. The IAS index was constructed from the driving frequency of the residual CP acceleration. Then, numerical investigations using finite element simulation were conducted to validate using the IAS index for indirect bridge damage identification. The application scope of the approach under various vehicle speeds and road roughness grades was examined. Finally, a laboratory vehicle–bridge interaction system was tested to validate the approach. Numerical studies demonstrated that bridge damage could be directly determined by observing the IAS abnormalities, which were baseline-free. The IAS from the residual CP response outperformed the IAS from CP responses in identifying bridge damage. However, it was better to use the IAS when the vehicle speed was no greater than 2 m/s and the grade of the road surface roughness was not high. Laboratory tests showed that it was possible to identify bridge damage using the IAS extracted from the residual CP acceleration under perfect road surfaces. However, it fell short under rough road surfaces. Hence, further experiments are required to fully examine the capacity of the IAS for bridge damage identification in practical applications.

show abstract

Section: Numerical Validation Of the Residual Cp Responsementioning

confidence: 99%

Section: Numerical Validation Of the Residual Cp Responsementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Theoretical and Experimental Investigations of Identifying Bridge Damage Using Instantaneous Amplitude Squared Extracted from Vibration Responses of a Two-Axle Passing Vehicle

Liu,

Zhou,

Zhang

et al. 2024

Buildings

View full text Add to dashboard Cite

show abstract

“…Gradually, Non-Destructive Testing (NDT) methods, such as Acoustic Emission (AE) [9], ultrasonic [10], X-ray [11], magnetic particle inspection [12], eddy current [13], and other detection methods, are frequently utilized for monitoring the key component of the structure [13]. Currently, the most commonly used SHM methods are vibration-based SHM [14][15][16][17][18], AEbased SHM [9,19,20], Guided Wave (GW) based SHM [21][22][23][24][25], and Electro-Mechanical Impedance (EMI) based SHM [26,27]. Among them, vibration-based SHM is the most popular one by analyzing the relationship between vibration characteristics and damage states.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning for Structural Health Monitoring: Data, Algorithms, Applications, Challenges, and Trends

Jia,

2023

Sensors

View full text Add to dashboard Cite

Environmental effects may lead to cracking, stiffness loss, brace damage, and other damages in bridges, frame structures, buildings, etc. Structural Health Monitoring (SHM) technology could prevent catastrophic events by detecting damage early. In recent years, Deep Learning (DL) has developed rapidly and has been applied to SHM to detect, localize, and evaluate diverse damages through efficient feature extraction. This paper analyzes 337 articles through a systematic literature review to investigate the application of DL for SHM in the operation and maintenance phase of facilities from three perspectives: data, DL algorithms, and applications. Firstly, the data types in SHM and the corresponding collection methods are summarized and analyzed. The most common data types are vibration signals and images, accounting for 80% of the literature studied. Secondly, the popular DL algorithm types and application areas are reviewed, of which CNN accounts for 60%. Then, this article carefully analyzes the specific functions of DL application for SHM based on the facility’s characteristics. The most scrutinized study focused on cracks, accounting for 30 percent of research papers. Finally, challenges and trends in applying DL for SHM are discussed. Among the trends, the Structural Health Monitoring Digital Twin (SHMDT) model framework is suggested in response to the trend of strong coupling between SHM technology and Digital Twin (DT), which can advance the digitalization, visualization, and intelligent management of SHM.

show abstract

Drive-by scour damage detection in railway bridges using deep autoencoder and different sensor placement strategies

Fernandes,

Lopez,

Ribeiro

2024

J Civil Struct Health Monit

View full text Add to dashboard Cite

Unsupervised learning-based framework for indirect structural health monitoring using adversarial autoencoder

Cited by 11 publications

References 52 publications

Theoretical and Experimental Investigations of Identifying Bridge Damage Using Instantaneous Amplitude Squared Extracted from Vibration Responses of a Two-Axle Passing Vehicle

Theoretical and Experimental Investigations of Identifying Bridge Damage Using Instantaneous Amplitude Squared Extracted from Vibration Responses of a Two-Axle Passing Vehicle

Deep Learning for Structural Health Monitoring: Data, Algorithms, Applications, Challenges, and Trends

Drive-by scour damage detection in railway bridges using deep autoencoder and different sensor placement strategies

Contact Info

Product

Resources

About