Machine learning techniques for structural health monitoring of heritage buildings: A state-of-the-art review and case studies

Mishra, Mayank

doi:10.1016/j.culher.2020.09.005

Cited by 136 publications

(33 citation statements)

References 178 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the latest advances in neurosciences and high-capability computing devices, machine learning (ML) algorithms based on Artificial Neural Networks (ANNs) have rapidly emerged as a promising tool to solve damage/defects identification and localization problem [8,10]. Various techniques, including mode shapes, natural frequencies, strain history and many others [11][12][13][14][15], have been used for damage identification during the years for different application fields. This was achieved by utilizing different supervised or unsupervised machine learning techniques for damage recognition [16][17][18].…”

Section: Introductionmentioning

confidence: 99%

Damage Detection in Flat Panels by Guided Waves Based Artificial Neural Network Trained through Finite Element Method

et al. 2021

View full text Add to dashboard Cite

Artificial Neural Networks (ANNs) have rapidly emerged as a promising tool to solve damage identification and localization problem, according to a Structural Health Monitoring approach. Finite Element (FE) Analysis can be extremely helpful, especially for reducing the laborious experimental campaign costs for the ANN development and training phases. The aim of the present work is to propose a guided wave-based ANN, developed through the use of the Finite Element Method, to determine the position of damages. The paper first addresses the development and assessment of the modeling technique. The FE model accuracy was proven through the comparison of the predicted results with experimental and analytical data. Then, the ANN was developed and trained on an aluminum plate and subsequently verified in a composite plate, as well as under different damage configurations. According to the results herein proposed, the ANN allowed to detect and localize damages with a high level of accuracy in all cases of study.

show abstract

Section: Introductionmentioning

confidence: 99%

Damage Detection in Flat Panels by Guided Waves Based Artificial Neural Network Trained through Finite Element Method

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Intelligent monitoring techniques have been used for damage localization and quantification in numerous operational bridges [10], buildings [11], and aerospace structures [12], by identifying the deviation from the optimal working conditions and determining remaining life. Mishra [13] systematically reviewed the advantages of combining test data with machine learning for structural health monitoring and damage prognosis, to ensure the longevity of heritage buildings. Gopinath et al [14] also reviewed the various long-term and short-term techniques for damage identification and localization in heritage structures.…”

Section: Introductionmentioning

confidence: 99%

Real-Time Structural Health Monitoring and Damage Identification Using Frequency Response Functions along with Finite Element Model Updating Technique

Singh

Sehgal

Prakash

et al. 2022

Sensors

View full text Add to dashboard Cite

Throughout service, damage can arise in the structure of buildings; hence, their dynamic testing becomes essential to verify that such buildings possess sufficient strength to withstand disturbances, particularly in the event of an earthquake. Dynamic testing, being uneconomical, requires proof of concept; for this, a model of a structure can be dynamically tested, and the results are used to update its finite element model. This can be used for damage detection in the prototype and aids in predicting its behavior during an earthquake. In this instance, a wireless MEMS accelerometer was used, which can measure the vibration signals emanating from the building and transfer these signals to a remote workstation. The base of the structure is excited using a shaking table to induce an earthquake-like situation. Four natural frequencies have been considered and six different types of damage conditions have been identified in this work. For each damage condition, the experimental responses are measured and the finite element model is updated using the Berman and Nagy method. It is seen that the updated models can predict the dynamic responses of the building accurately. Thus, depending on these responses, the damage condition can be identified by using the updated finite element models.

show abstract

“…Therefore, researchers began to apply machine learning (ML) algorithms to increase the accuracy of imagebased crack inspection techniques. Many researchers have proposed ML-based crack detection methods [27,28]. In ML-based crack detection methods, features are extracted through IPTs and then classified according to whether they contain cracks.…”

Section: Introductionmentioning

confidence: 99%

Two-stage method based on the you only look once framework and image segmentation for crack detection in concrete structures

Mishra

Jain

Singh

et al. 2022

Archit. Struct. Constr.

Self Cite

View full text Add to dashboard Cite

Detecting the presence of cracks and identifying their severity are crucial tasks for determining the structural health of a concrete building. In this study, we develop a two-stage automated method based on the You Only Look Once (YOLOv5) deep learning framework for the identification, localization, and quantification of cracks in the concrete structures. In the first stage, cracks are identified and localized using bounding boxes, while in the second stage, the length of cracks and, therefore, the damage severity are determined. The developed deep learning model is trained using 4500 annotated images from a total of 40000 images of size 227 × 227 pixel, which are obtained from an open-source dataset collected at various campus buildings of Middle East Technical University (METU). The concept of transfer learning (i.e., pre-trained weights) is used for the training, which drastically reduces the training time. The detection and localization accuracy of this model is measured in terms of the average precision, average recall, and F1-score. The YOLOv5 model achieves the mean average precision (mAP_0.5) of 95.02%. A ResNet model is also developed just to capture the supremacy of the YOLOv5 model. The proposed method can help in identifying structural anomalies through real-time monitoring that must be urgently repaired and thus can be used in high-quality civil infrastructure monitoring systems.

show abstract

Machine learning techniques for structural health monitoring of heritage buildings: A state-of-the-art review and case studies

Cited by 136 publications

References 178 publications

Damage Detection in Flat Panels by Guided Waves Based Artificial Neural Network Trained through Finite Element Method

Damage Detection in Flat Panels by Guided Waves Based Artificial Neural Network Trained through Finite Element Method

Real-Time Structural Health Monitoring and Damage Identification Using Frequency Response Functions along with Finite Element Model Updating Technique

Two-stage method based on the you only look once framework and image segmentation for crack detection in concrete structures

Contact Info

Product

Resources

About