Acoustic emission-based damage localization using wavelet-assisted deep learning

Barbosh, Mohamed; Dunphy, Kyle; Sadhu, Ayan

doi:10.1186/s43065-022-00051-8

Cited by 18 publications

(7 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The input was a 2D image value obtained by the wavelet transformation. Conversely, [86] also employed 2D CNN for damage localization, however, the authors first exploited Empirical Mode Decomposition (EMD) to extract intrinsic mode functions (IMFs) from noisy raw measurements. These IMFs were then used to generate Continuous Wavelet Transform (CWT).…”

Section: Damage Localizationmentioning

confidence: 99%

From data to insight, enhancing structural health monitoring using physics-informed machine learning and advanced data collection methods

Rizvi,

Abbas

2023

Eng. Res. Express

View full text Add to dashboard Cite

Owing to recent advancements in sensor technology, data mining, machine learning and cloud computation, structural health monitoring (SHM) based on a data-driven approach has gained more popularity and interest. The data-driven methodology has proved to be more efficient and robust compared with traditional physics-based methods. The past decade has witnessed remarkable progress in machine learning, especially in the field of deep learning which are effective in many tasks and has achieved state-of-the-art results in various engineering domains. In the same manner, deep learning has also revolutionized SHM technology by improving the effectiveness and efficiency of models, as well as enhancing safety and reliability. To some extent, it has also paved the way for implementing SHM in real-world complex civil and mechanical infrastructures. However, despite all the success, deep learning has intrinsic limitations such as its massive-labelled data requirement, inability to generate consistent results and lack of generalizability to out-of-sample scenarios. Conversely, in SHM, the lack of data corresponding to a different state of the structure is still a challenging task. Recent development in physics-informed machine learning methods has provided an opportunity to resolve these challenges in which limited-noisy data and mathematical models are integrated through machine learning algorithms. This method automatically satisfies physical invariants providing better accuracy and improved generalization. This manuscript presents the sate of- -the-art review of prevailing machine learning methods for efficient damage inspection, discuss their limitations, and explains the diverse applications and benefits of physics-informed machine learning in the SHM setting. Moreover, the latest data extraction strategy and the internet of things (IoT) that support the present data-driven methods and SHM are also briefly discussed in the last section.

show abstract

Section: Damage Localizationmentioning

confidence: 99%

From data to insight, enhancing structural health monitoring using physics-informed machine learning and advanced data collection methods

Rizvi,

Abbas

2023

Eng. Res. Express

View full text Add to dashboard Cite

show abstract

“…Given an RGB image as depicted in Figure 2a, the number of channels is reduced by converting the RGB image to a greyscale image. As RGB images consist of a 3D matrix of size (M,N,c) where c is a channel associated with the red, blue, and green spectrum of the image, the image can be converted to grayscale [44] through Equation ( 6): P mn = 0.2989P R mn + 0.5870P G mn + 0.1140P B mn (6) where m is the integer value representing the location of the pixel along the length of the image, n is the integer value representing the location of the pixel along the width of the image, P R mn is the pixel intensity of the red channel at pixel location (m,n), P G mn is the pixel intensity of the green channel at pixel location (m,n), P B mn is the pixel intensity of the blue channel at pixel location (m,n), and P mn is the pixel intensity of the grayscale image at pixel location (m,n). Once the RGB image has been converted to grayscale, the centroid of the image can be determined as shown in Figure 2b based on the pixel intensity of the image as represented by the color bar.…”

Section: Cluster and Density Analysis Of Datasetsmentioning

confidence: 99%

Data Augmentation for Deep-Learning-Based Multiclass Structural Damage Detection Using Limited Information

Dunphy

Fekri

Grolinger

et al. 2022

Sensors

Self Cite

View full text Add to dashboard Cite

The deterioration of infrastructure’s health has become more predominant on a global scale during the 21st century. Aging infrastructure as well as those structures damaged by natural disasters have prompted the research community to improve state-of-the-art methodologies for conducting Structural Health Monitoring (SHM). The necessity for efficient SHM arises from the hazards damaged infrastructure imposes, often resulting in structural collapse, leading to economic loss and human fatalities. Furthermore, day-to-day operations in these affected areas are limited until an inspection is performed to assess the level of damage experienced by the structure and the required rehabilitation determined. However, human-based inspections are often labor-intensive, inefficient, subjective, and restricted to accessible site locations, which ultimately negatively impact our ability to collect large amounts of data from inspection sites. Though Deep-Learning (DL) methods have been heavily explored in the past decade to rectify the limitations of traditional methods and automate structural inspection, data scarcity continues to remain prevalent within the field of SHM. The absence of sufficiently large, balanced, and generalized databases to train DL-based models often results in inaccurate and biased damage predictions. Recently, Generative Adversarial Networks (GANs) have received attention from the SHM community as a data augmentation tool by which a training dataset can be expanded to improve the damage classification. However, there are no existing studies within the SHM field which investigate the performance of DL-based multiclass damage identification using synthetic data generated from GANs. Therefore, this paper investigates the performance of a convolutional neural network architecture using synthetic images generated from a GAN for multiclass damage detection of concrete surfaces. Through this study, it was determined the average classification performance of the proposed CNN on hybrid datasets decreased by 10.6% and 7.4% for validation and testing datasets when compared to the same model trained entirely on real samples. Moreover, each model’s performance decreased on average by 1.6% when comparing a singular model trained with real samples and the same model trained with both real and synthetic samples for a given training configuration. The correlation between classification accuracy and the amount and diversity of synthetic data used for data augmentation is quantified and the effect of using limited data to train existing GAN architectures is investigated. It was observed that the diversity of the samples decreases and correlation increases with the increase in the number of synthetic samples.

show abstract

“…Sikdar et al [25] proposed a convolutional neural network (CNN)-based algorithm for identification of the region of damage in a composite panel. Barbosh et al [26] utilised a combination of the CWT and a deep neural network to detect the location of damage in wooden beams, wooden plates and concrete beams. The application of ML in the monitoring of rails is rather sparse.…”

Section: Introductionmentioning

confidence: 99%

Acoustic emission source localisation for structural health monitoring of rail sections based on a deep learning approach

Mahajan

Banerjee

2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

Acoustic emission (AE) approach for non-destructive evaluation of structure is being developed for last two decades. In complex structures, one of the limitation of AE testing is to find the location of AE source. Time of flight and wave velocity are typically employed to localise AE sources. However, complex rail structure generates multiple wave modes travelling at varying speeds, making localization a difficult task. In this paper, the challenge of localisation has been split up into two parts: (i) identification of AE source zone i.e. head, web and foot (ii) identification of location along length of rail. AE events are simulated using pencil lead break (PLB) as source. Three models including artificial neural network (ANN), 1D and 2D convolutional neural network (CNN) are trained and tested using AE signals generated by PLB sources. The accuracy of zone identification is reported as 94.79% by using 2D CNN algorithm. For location classification it is also found that 2D CNN performed best with 73.12%, 79.37% and 67.50% accuracy of localizing AE source along the length in head, web and foot, respectively. For AE signal generating from actual damage in rail, bending test on inverted damaged rail section was then performed with load cases of 100kN, 150kN and 200kN. For all load cases, 2D CNN model results in accurate prediction of the zone of AE source, whereas, it accurately predicts the AE source location along the length for the load cases of higher intensity (150kN, 200kN). It is envisaged that the deep learning approach presented in this research work is expected to be helpful in developing real time monitoring system for rail inspection based on AE.

show abstract

Acoustic emission-based damage localization using wavelet-assisted deep learning

Cited by 18 publications

References 38 publications

From data to insight, enhancing structural health monitoring using physics-informed machine learning and advanced data collection methods

From data to insight, enhancing structural health monitoring using physics-informed machine learning and advanced data collection methods

Data Augmentation for Deep-Learning-Based Multiclass Structural Damage Detection Using Limited Information

Acoustic emission source localisation for structural health monitoring of rail sections based on a deep learning approach

Contact Info

Product

Resources

About