Automated real-time damage detection strategy using raw dynamic measurements

Cardoso, Rharã de Almeida; Cury, Alexandre; Barbosa, Fernando Policarpo

doi:10.1016/j.engstruct.2019.109364

Cited by 42 publications

(16 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the tuning phase, a pre-defined novelty detection system of limit-state functions' detection thresholds are dynamically tuned based on the user-defined parameters (i.e., T L , D L , and V L ) with reliability analysis. The tuning is vital for any method since non-proper parameters can lead to many false alarms, or undetected novelties [22,23,24]. The Gan-generated data objects are utilized for the tuning phase to decrease the detection's sensitivity to user-defined parameters.…”

Section: Methodsmentioning

confidence: 99%

“…As discussed in Soleimani et al [9], the features to vary in a fixed-range is beneficial for the GAN's training. The second feature (F II) is a reduced F I representation, made of quartiles of vibrational energy in each time-series data object, which is tried on prior studies [23]. Since the method is designed to be general and avoid time-averaging with specified windows, a Periodogram power spectral density estimation is employed.…”

Section: Feature Extractionmentioning

confidence: 99%

“…User-defined (detection) parameters such as the number of time-series data object to undergo the detection process in each detection-iteration (i.e., detection window length) [23], or thresholds for having a dynamic number of clusters [22] can result in a variety of clusters, or alarms differently from the ground truth ones. The detection window length is negatively correlated to the novelty detection power and positively correlated with having less novelty detection while the system is intact (false alarms) [24,23]. In studies dealing with the detection window length, different lengths were studied [24,23,3], and the one with the best detection result was introduced as the best candidate.…”

Section: Introductionmentioning

confidence: 99%

“…The detection window length is negatively correlated to the novelty detection power and positively correlated with having less novelty detection while the system is intact (false alarms) [24,23]. In studies dealing with the detection window length, different lengths were studied [24,23,3], and the one with the best detection result was introduced as the best candidate. Nevertheless, tuning the detection-window length is impractical in a real-time detection process with unseen data, as it must be defined before the initiation of the analysis.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

System-reliability based multi-ensemble of GAN and one-class joint Gaussian distributions for unsupervised real-time structural health monitoring

Soleimani-Babakamali,

Sepasdar,

Nasrollahzadeh

et al. 2021

Preprint

View full text Add to dashboard Cite

Unsupervised health monitoring has gained much attention in the last decade as the most practical real-time structural health monitoring (SHM) approach. Among the proposed unsupervised techniques in the literature, there are still obstacles to robust and real-time health monitoring. These barriers include loss of information from dimensionality reduction in feature extraction steps, case-dependency of those steps, lack of a dynamic clustering, and detection results' sensitivity to user-defined parameters. This study introduces an unsupervised real-time SHM method with a mixture of low-and high-dimensional features without a case-dependent extraction scheme. Both features are used to train multi-ensembles of Generative Adversarial Networks (GAN) and one-class joint Gaussian distribution models (1-CG). A novelty detection system of limit-state functions based on GAN and 1-CG models' detection scores is constructed. The Resistance of those limit-state functions (detection thresholds) is tuned to user-defined parameters with the GAN-generated data objects by employing the Monte Carlo histogram sampling through a reliability-based analysis. The tuning makes the method robust to user-defined parameters, which is crucial as there is no rule for selecting those parameters in a real-time SHM. The proposed novelty detection framework is applied to two standard SHM datasets to illustrate its generalizability: Yellow Frame (twenty damage classes) and Z24 Bridge (fifteen damage classes). All different damage categories are identified with low sensitivity to the initial choice of user-defined parameters with both introduced dynamic and static baseline approaches with few or no false alarms.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Feature Extractionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

System-reliability based multi-ensemble of GAN and one-class joint Gaussian distributions for unsupervised real-time structural health monitoring

Soleimani-Babakamali,

Sepasdar,

Nasrollahzadeh

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Lately, features extracted directly from vibration data are being used to recognize and predict structural behaviors during monitoring, thus avoiding the evaluation of a finite element model (FEM) of the structure, needed in modal parameter-based methods, 15,16 and decreasing the complexity of the analyses. For instance, Cardoso et al 2,17 proposed an automated methodology in which raw acceleration signals are transformed into symbolic objects and those features are used in a pattern recognition procedure to detect structural damage. Such an approach was tested with small-to large-scale structures and showed an adequate sensitivity even in scenarios with small damage levels.…”

mentioning

confidence: 99%

A fast and efficient feature extraction methodology for structural damage localization based on raw acceleration measurements

Alves

Cury

2021

Struct Control Health Monit

Self Cite

View full text Add to dashboard Cite

Many damage detection strategies have been developed within the field of structural health monitoring, showing promising results in real-world applications. Most of them rely on the use of modal parameters to locate damage in structures. However, they are highly dependent on the process of extracting such characteristics, which contains uncertainties and needs to be adapted specifically for each structure. Recently, a special focus is being given to techniques based only on the use of raw structural acceleration measurements due to their relatively lower computational complexity manipulation. Thus, this work proposes an efficient feature extraction methodology for structural damage localization based on raw vibration signals. Furthermore, its ability to indicate damage quantification is also investigated. The proposed approach consists in extracting sensitive features from time, frequency, and quefrency domains. Then, percentile intervals are defined for each feature regarding the structures' healthy state. Finally, a damage index is estimated based on the number of outlier features of the structures' damaged state. In this sense, the sensor with the highest number of outlier features indicates the damage location. To assess this methodology, four different applications are studied: a numerical two-dimensional frame, a simply supported beam tested in laboratory, a three-dimensional frame also tested in laboratory, and the Tianjin Yonghe Bridge in China. Results show that the proposed approach is not only able to correctly indicate damage locations but also to give insights about their magnitudes.

show abstract

Real-Time Unsupervised Detection of Early Damage in Railway Bridges Using Traffic-Induced Responses

Meixedo

Ribeiro

Santos

et al. 2021

Structural Integrity

View full text Add to dashboard Cite

This chapter addresses unsupervised damage detection in railway bridges by presenting a novel AI-based SHM strategy using traffic-induced dynamic responses. To achieve this goal a hybrid combination of wavelets, PCA and cluster analysis is implemented. Damage-sensitive features from train-induced dynamic responses are extracted and allow taking advantage not only of the repeatability of the loading, but also, of its large magnitude, thus enhancing sensitivity to small-magnitude structural changes. The effectiveness of the proposed methodology is validated in a long-span bowstring-arch railway bridge with a permanent structural monitoring system installed. A digital twin of the bridge was used, along with experimental values of temperature, noise, trains loadings and speeds, to realistically simulate baseline and damage scenarios. The methodology proved highly sensitive in detecting early damage, even in case of small stiffness reductions that do not impair structural safety, as well as highly robust to false detections. The ability to identify early damage, imperceptible in the original signals, while avoiding observable changes induced by environmental and operational variations, is achieved by carefully defining the modelling and fusion sequence of the information. A damage detection strategy capable of characterizing multi-sensor data while being sensitive to identify local changes, is proposed as a tool for real-time structural assessment of bridges without interfering with the normal service condition.

show abstract

Automated real-time damage detection strategy using raw dynamic measurements

Cited by 42 publications

References 25 publications

System-reliability based multi-ensemble of GAN and one-class joint Gaussian distributions for unsupervised real-time structural health monitoring

System-reliability based multi-ensemble of GAN and one-class joint Gaussian distributions for unsupervised real-time structural health monitoring

A fast and efficient feature extraction methodology for structural damage localization based on raw acceleration measurements

Real-Time Unsupervised Detection of Early Damage in Railway Bridges Using Traffic-Induced Responses

Contact Info

Product

Resources

About