Early Damage Detection Based on Pattern Recognition and Data Fusion

Santos, João; Crémona, Christian; Orcesi, André; Silveira, Paulo E. X.

doi:10.1061/(asce)st.1943-541x.0001643

Cited by 36 publications

(20 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are many applications in real life by using advanced validation algorithms such as clustering of social media data [22]- [24], bio-information [25], and damage detection [26] etc. However, most of the algorithms need to scale the original dataset with the normalization algorithm and then extract the major features by principal component analysis (PCA) before clustering.…”

Section: Related Workmentioning

confidence: 99%

Normalization-Based Validity Index of Adaptive K-Means Clustering for Multi-Solution Application

Endoh

2020

IEEE Access

View full text Add to dashboard Cite

Validity evaluation aims to analyze the quality of the clustering algorithm with different measurement criteria. A variety of assessment methods have been introduced in the application of pattern recognition and computer vision. Although it is well known that mining information of massive data is essential, most of the validity indices only provide a single partitioning scheme for clustering validation. Moreover, the conventional evaluation algorithm is susceptible to the density and dimension of the dataset, which leads to assessment failure. In this paper, a normalization-based validity index (NbVI) is proposed for validity evaluation of the adaptive K-means clustering from a multi-solution perspective. According to the concept of high-compact within clusters and high-separation among groups, NbVI attempts to find the maximum relative ratio between normalized inter-distance and normalized intra-distance. The experimental results demonstrate that the proposed NbVI method exhibits excellent performance for the clustering of the density-unbalanced dataset for multi-solution applications. Moreover, the NbVI validation shows high versatility using different clustering algorithms. INDEX TERMS NbVI, validity index, multi-solution, K-means clustering, unsupervised learning.

show abstract

Section: Related Workmentioning

confidence: 99%

Normalization-Based Validity Index of Adaptive K-Means Clustering for Multi-Solution Application

Endoh

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…To solve this problem, many authors suggested performing PCA on the raw data, but the vast majority of them point out to a supervised empirical framework. Santos et al also proposed a PCA‐based normalization strategy but using the broken‐stick rule for choosing the most appropriate number of principal components related to global effects, such as those caused by environmental and operational actions. This strategy relies on the premise that the higher eigenvalues of PCA are related to environmental and operational effects.…”

Section: Feature Classificationmentioning

confidence: 99%

“…Even the works addressing clustering methods, which are very well‐suited for unsupervised methods, report a supervised strategy for predefining cluster partitions to describe one or more known structural behaviors and, subsequently, compare them with new ones . However, some recently published works propose a time‐window procedure to extract sensitive features based on a fully unsupervised approach that uses clustering methods to perform feature classification. Such a methodology allows (a) detecting structural novelties without the necessity of any previous knowledge and (b) performing real‐time identification relying on the detection of novelties upon each newly acquired data.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised real‐time SHM technique based on novelty indexes

Cardoso

Cury

Barbosa

et al. 2019

Struct Control Health Monit

View full text Add to dashboard Cite

Structural health monitoring programs play an essential role in the field of civil engineering, especially for assessing safety conditions involving large structures such as viaducts, bridges, tall buildings, towers, and old historical buildings.Mostly, an SHM process needs to be based on a trustful strategy for detecting structural novelties or abnormal behaviors. Usually, such an approach is complemented with human inspection and structural instrumentation routines, where the latter requires proper hardware equipment and software tools. Recently, many advances were achieved regarding the hardware resources, such as wireless communication, remotely configurable sensors, and other data management devices. On the other hand, the software counterpart still is in its early developments. Several researches are in progress to fill this gap. In this context, this paper presents a novel online SHM identification method suitable to unsupervised real-time detection of abnormal structural behaviors. The proposed methodology includes the use of an original representation of raw dynamic signals, that is, in situ measured accelerations. To assess the proposed approach, numerical simulations and two experimental applications are studied: a railway viaduct, PK 075+317 in France and an old masonry tower in Italy. The results suggest that the proposed detection indexes are suitable for a wide range of SHM applications.

show abstract

“…The assessment observed that symbolic objects based on the interquartile intervals attain damage-sensitive information that can be utilized in classification algorithms as shown in the study by Alves et al (2015a). In the case of the study by Santos et al (2017), the dynamic cloud clustering algorithm was utilized, which is an adaptation of the popular k-means approach that converges to a solution quickly. This fast solution can cause issues however, as dynamic cloud clustering can converge to local minima.…”

Section: Non-modal-based Approaches To Damage Detectionmentioning

confidence: 99%

“…A disadvantage of the IDDM is that its assumption that for an undamaged state, all sources of vibration will equally cause all locations to produce the same variation in interpolation error will not be suitable to all bridge applications. Santos et al (2017) implemented a novel methodology to detect structural damage without the use of modal parameters. They proposed a data-driven technique that possesses real-time capabilities.…”

Section: Non-modal-based Approaches To Damage Detectionmentioning

confidence: 99%

Bridge Damage Detection Based on Vibration Data: Past and New Developments

Casas

Moughty

2017

Front. Built Environ.

View full text Add to dashboard Cite

Overtime, bridge condition declines due to a number of degradation processes such as creep, corrosion, and cyclic loading, among others. Traditionally, vibration-based damage detection techniques in bridges have focused on monitoring changes to modal parameters. These techniques can often suffer to their sensitivity to changes in environmental and operational conditions, mistaking them as structural damage. Recent research has seen the emergence of more advanced computational techniques that not only allow the assessment of noisier and more complex data but also allow research to veer away from monitoring changes in modal parameters alone. This paper presents a review of the current state-of-the-art developments in vibration-based damage detection in small to medium span bridges with particular focus on the utilization of advanced computational methods that avoid traditional damage detection pitfalls. A case study based on the S101 bridge is also presented to test the damage sensitivity to a chosen methodology.

show abstract

Early Damage Detection Based on Pattern Recognition and Data Fusion

Cited by 36 publications

References 34 publications

Normalization-Based Validity Index of Adaptive K-Means Clustering for Multi-Solution Application

Normalization-Based Validity Index of Adaptive K-Means Clustering for Multi-Solution Application

Unsupervised real‐time SHM technique based on novelty indexes

Bridge Damage Detection Based on Vibration Data: Past and New Developments

Contact Info

Product

Resources

About