Investigation of Frequency-Domain Dimension Reduction for A<sup>2</sup>M-Based Bridge Damage Detection Using Accelerations of Moving Vehicles
Recent decades have witnessed a rise in interest in bridge health monitoring utilizing the vibrations of passing vehicles. However, existing studies commonly rely on constant speeds or tuning vehicular parameters, making their methods challenging to be used in practical engineering applications. Add...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Materials |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1944/16/5/1872 |
| _version_ | 1797614970717863936 |
|---|---|
| author | Zhenkun Li Yifu Lan Weiwei Lin |
| author_facet | Zhenkun Li Yifu Lan Weiwei Lin |
| author_sort | Zhenkun Li |
| collection | DOAJ |
| description | Recent decades have witnessed a rise in interest in bridge health monitoring utilizing the vibrations of passing vehicles. However, existing studies commonly rely on constant speeds or tuning vehicular parameters, making their methods challenging to be used in practical engineering applications. Additionally, recent studies on the data-driven approach usually need labeled data for damage scenarios. Still, getting these labels in engineering is difficult or even impractical because the bridge is typically in a healthy state. This paper proposes a novel, damaged-label-free, machine-learning-based, indirect bridge-health monitoring method named the assumption accuracy method (A<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mrow></mrow><mn>2</mn></msup></semantics></math></inline-formula>M). Initially, the raw frequency responses of the vehicle are employed to train a classifier, and K-folder cross-validation accuracy scores are then used to calculate a threshold to specify the bridge’s health state. Compared to merely focusing on low-band frequency responses (0–50 Hz), utilizing full-band vehicle responses can significantly improve the accuracy, meaning that the bridge’s dynamic information exists in the higher frequency ranges and can contribute to detecting bridge damage. However, raw frequency responses are generally in a high-dimensional space, and the number of features is much greater than that of samples. To represent the frequency responses via latent representations in a low-dimension space, appropriate dimension-reduction techniques are therefore, needed. It was found that principal component analysis (PCA) and Mel-frequency cepstral coefficients (MFCCs) are suitable for the aforementioned issue, and MFCCs are more damage-sensitive. When the bridge is in a healthy condition, the accuracy values obtained using MFCCs are primarily dispersed around 0.5, but following the occurrence of damage, they increased significantly to 0.89–1.0 in this study. |
| first_indexed | 2024-03-11T07:19:51Z |
| format | Article |
| id | doaj.art-aaea683255c646708df995ba0dfc9535 |
| institution | Directory Open Access Journal |
| issn | 1996-1944 |
| language | English |
| last_indexed | 2024-03-11T07:19:51Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Materials |
| spelling | doaj.art-aaea683255c646708df995ba0dfc95352023-11-17T08:04:04ZengMDPI AGMaterials1996-19442023-02-01165187210.3390/ma16051872Investigation of Frequency-Domain Dimension Reduction for A<sup>2</sup>M-Based Bridge Damage Detection Using Accelerations of Moving VehiclesZhenkun Li0Yifu Lan1Weiwei Lin2Department of Civil Engineering, Aalto University, 02150 Espoo, FinlandDepartment of Civil Engineering, Aalto University, 02150 Espoo, FinlandDepartment of Civil Engineering, Aalto University, 02150 Espoo, FinlandRecent decades have witnessed a rise in interest in bridge health monitoring utilizing the vibrations of passing vehicles. However, existing studies commonly rely on constant speeds or tuning vehicular parameters, making their methods challenging to be used in practical engineering applications. Additionally, recent studies on the data-driven approach usually need labeled data for damage scenarios. Still, getting these labels in engineering is difficult or even impractical because the bridge is typically in a healthy state. This paper proposes a novel, damaged-label-free, machine-learning-based, indirect bridge-health monitoring method named the assumption accuracy method (A<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mrow></mrow><mn>2</mn></msup></semantics></math></inline-formula>M). Initially, the raw frequency responses of the vehicle are employed to train a classifier, and K-folder cross-validation accuracy scores are then used to calculate a threshold to specify the bridge’s health state. Compared to merely focusing on low-band frequency responses (0–50 Hz), utilizing full-band vehicle responses can significantly improve the accuracy, meaning that the bridge’s dynamic information exists in the higher frequency ranges and can contribute to detecting bridge damage. However, raw frequency responses are generally in a high-dimensional space, and the number of features is much greater than that of samples. To represent the frequency responses via latent representations in a low-dimension space, appropriate dimension-reduction techniques are therefore, needed. It was found that principal component analysis (PCA) and Mel-frequency cepstral coefficients (MFCCs) are suitable for the aforementioned issue, and MFCCs are more damage-sensitive. When the bridge is in a healthy condition, the accuracy values obtained using MFCCs are primarily dispersed around 0.5, but following the occurrence of damage, they increased significantly to 0.89–1.0 in this study.https://www.mdpi.com/1996-1944/16/5/1872bridge health monitoringdamage detectionindirect methoddimension reduction |
| spellingShingle | Zhenkun Li Yifu Lan Weiwei Lin Investigation of Frequency-Domain Dimension Reduction for A<sup>2</sup>M-Based Bridge Damage Detection Using Accelerations of Moving Vehicles Materials bridge health monitoring damage detection indirect method dimension reduction |
| title | Investigation of Frequency-Domain Dimension Reduction for A<sup>2</sup>M-Based Bridge Damage Detection Using Accelerations of Moving Vehicles |
| title_full | Investigation of Frequency-Domain Dimension Reduction for A<sup>2</sup>M-Based Bridge Damage Detection Using Accelerations of Moving Vehicles |
| title_fullStr | Investigation of Frequency-Domain Dimension Reduction for A<sup>2</sup>M-Based Bridge Damage Detection Using Accelerations of Moving Vehicles |
| title_full_unstemmed | Investigation of Frequency-Domain Dimension Reduction for A<sup>2</sup>M-Based Bridge Damage Detection Using Accelerations of Moving Vehicles |
| title_short | Investigation of Frequency-Domain Dimension Reduction for A<sup>2</sup>M-Based Bridge Damage Detection Using Accelerations of Moving Vehicles |
| title_sort | investigation of frequency domain dimension reduction for a sup 2 sup m based bridge damage detection using accelerations of moving vehicles |
| topic | bridge health monitoring damage detection indirect method dimension reduction |
| url | https://www.mdpi.com/1996-1944/16/5/1872 |
| work_keys_str_mv | AT zhenkunli investigationoffrequencydomaindimensionreductionforasup2supmbasedbridgedamagedetectionusingaccelerationsofmovingvehicles AT yifulan investigationoffrequencydomaindimensionreductionforasup2supmbasedbridgedamagedetectionusingaccelerationsofmovingvehicles AT weiweilin investigationoffrequencydomaindimensionreductionforasup2supmbasedbridgedamagedetectionusingaccelerationsofmovingvehicles |