Crack Detection In Structures Research Articles

Cluster-based acoustic emission signal processing and loading rate effects study of nanoindentation on thin film stack structures

This paper presents a high-resolution, in-situ material testing system that integrates acoustic emission (AE) testing with a nanoindentation system for crack generation and detection in thin film stack structures. This is used to find the critical contact load during wafer probing of crack-sensitive backend-of-line (BEOL) structures in semiconductor integrated circuits. Scanning electron microscopy (SEM) and load–displacement curve analysis were used to confirm the formation and propagation of cracks in the multilayer structures. In order to improve the manual classification performance and understand the physical meaning of AE signals, this paper introduces a machine learning based signal processing approach based on a k-means clustering algorithm applied on collected AE signals. To obtain the optimal number of k-means clusters, Davies–Bouldin, Dunn, and Silhouette indices were calculated, and the individual ratings were cumulated based on a voting scheme. Multiple feature extraction methods, including raw time-domain AE signals, conventional AE extracted parameters, short-term signal energy, and representation features learned by the autoencoder, were used and evaluated by manually labeled clusters and binary confusion matrices. A supervised learning technique, the k-nearest neighbors algorithm, was also utilized on different AE signal datasets using different loading rates to further investigate the damage processes during nanoindentation and the physical meaning of different AE signals. The influences of loading rates on AE signals have been investigated, and loading rate effects on the critical load were observed – higher loading rates led to higher critical loads. This integrated test system and signal processing approach provides a high-resolution mechanical testing platform for studying and enabling automatic crack detection in wafer probing.

Concrete Cracks Detection and Monitoring Using Deep Learning-Based Multiresolution Analysis

In this paper, we propose a new methodology for crack detection and monitoring in concrete structures. This approach is based on a multiresolution analysis of a sample or a specimen of concrete material subjected to several types of solicitation. The image obtained by ultrasonic investigation and processed by a customized wavelet is analyzed at various scales in order to detect internal cracks and crack initiation. The ultimate objective of this work is to propose an automatic crack type identification scheme based on convolutional neural networks (CNN). In this context, crack propagation can be monitored without access to the concrete surface and the goal is to detect cracks before they are visible. This is achieved through the combination of two major data analysis tools which are wavelets and deep learning. This original procedure is shown to yield a high accuracy close to 90%. In order to evaluate the performance of the proposed CNN architectures, we also used an open access database, SDNET2018, for the automatic detection of external cracks.

Detection of Dangerous Defects and Damages of Steel Building Structures by Active Thermography

The article presents the results of a study of use thermographic non-destructive testing to search for cracks in steel structures. Theoretical substantiation of thermal non-destructive method of control for detection of cracks in steel structures. A practical study proving the possibility of using thermal non-destructive testing to detect cracks in steel structures is described. The problems that arise during the thermal non-destructive method of control and possible ways to solve them. The authors conclude that the method of thermographic control can be used during the inspection of steel structures for qualitative assessment of cracks.

A Defect Detection Method for the Surface of Metal Materials Based on an Adaptive Ultrasound Pulse Excitation Device and Infrared Thermal Imaging Technology

Ultrasonic excitation has been widely used in the detection of microcracks on metal surfaces, but there are problems such as poor excitation effect of ultrasonic pulse, long time to reach the best excitation, and difficult to find microcracks. In this paper, an adaptive ultrasonic pulse excitation device and infrared thermal imaging technology have been combined, as well as their control method, to solve the problem. The adaptive ultrasonic pulse excitation device adds intelligent modules to realize automatic adjustment of detection parameters, which can quickly obtain reliable excitation; the multidegree-of-freedom base realizes the three-dimensional direction change of the ultrasonic gun to adapt to different excitation occasions. When the appropriate ultrasonic excitation makes microcracks in the resonance state, the microcracks can be frictionated, which produce heat rise with the temperature. Then, the microcrack defect can be detected by the infrared thermal instrument through the different surface temperatures with imaging recognition method. Our detection experiments of the titanium alloy plates and the aluminum alloy profiles of marine engineering show that the method can get reliable detection parameters in a short time and measure the crack length effectively. It can be used in many aspects such as crack detection in mechanical structures or complex equipment operating conditions and industrial production processes.

Machine learning model for predicting the crack detection and pattern recognition of geopolymer concrete beams

One of the major challenges in the construction industry is the detection of cracks in concrete structures and identification of failure types of these structures that lead to their degradation. Manual quality checks are prone to human error, and require longer response time and specialist experience and knowledge. Therefore, visualizing the cracks and identifying failures in concrete structures using computer techniques is now a preferred option. The present work focuses on identifying the cracks using image processing and failure pattern recognition technique by employing suitable machine learning algorithms, and validating the techniques using Python programming. For this purpose, M30 grade geopolymer and conventional concrete beams were cast using Basalt Fibre Reinforced Polymer/Glass Fibre Reinforced Polymer and Steel bars. The beams were subjected to four point static bending test by varying the shear span to the effective depth ratio. The experimental images were used for image processing and failure pattern recognition in Python language. Employing six machine learning classifiers, the failures in the structures were classified into three classes namely, flexure, shear, and compression. The machine learning classifiers were also adopted to determine the confusion matrix, accuracy, precision, and recall scores. It was found that among the six classifiers used, the support vector classifier gave the best performance with 100% accuracy in identifying the failure patterns.

Parametric reduced order models for output-only vibration-based crack detection in shell structures

In this work parametric reduced order models (pROMs) for cracked shells are developed and applied to crack detection problems. Mesh morphing is employed to allow for parameterization of the models with respect to the crack location and size, while a clustering approach is adopted to partition the parameter space in sub-domains, for which efficient local reduced order models (ROMs) are constructed. Subsequently, the proposed ROMs are employed as forward simulators within an inverse problem setting, aiming to identify the location of the fault (crack). An output-only scheme is adopted, in the absence of information regarding the loading time history, based on transmissibility functions. The resulting scheme is tested through numerical examples involving realistic geometries, consisting of curved shells or multiple components.

Structural crack localization and identification based on FBG

The current structural crack detection still suffers from the problems of single detection target, inability to achieve true acoustic emission crack signal separation, and poor portability of the sensing system. To address this phenomenon, in this paper, based on the use of fast blind source signal separation, combined with the designed structural crack sensing detection model, we will first coarsely localize the crack location area, and within this area, we will use the designed structural multi-crack parameter identification algorithm to achieve the multi-crack location identification purpose. The experimental results show that the proposed method achieves an average localization accuracy of 1.95 cm and a localization time delay of about 25 ms. It provides a feasible method for structural crack detection.

Crack modeling of metal rod eigen-frequencies

The paper presents the development of approaches to the crack detection in metal rod structures based on the analysis of the lowest eigen-frequency modes. Full-scale experiments and numerical calculations are carried out, and the obtained results are compared. A vibration analyzer is used for full-scale experiments, and numerical calculations are performed by using Autodesk Inventor. With regard to the internal friction, the antinodes of various vibration forms were identified using a specially developed program. The model includes sensors for the the field experiment as masses affecting the frequency-response characteristics. The dependences are obtained for eigen-frequencies in the presence of cracks and for the crack locations. The polynomial dependences of the crack location on the lowest eigen-frequency modes of the rod can be used to analyze the crack position of in cantilever beams.

Validation of Fresnel–Kirchhoff Integral Method for the Study of Volume Dielectric Bodies

In this work, we test a nondestructive optical method based on the Fresnel–Kirchhoff integral, which could be applied to different fields of engineering, such as detection of small cracks in structures, determination of dimensions for small components, analysis of composition of materials, etc. The basic idea is to apply the Fresnel–Kirchhoff integral method to the study of the properties of small-volume dielectric objects. In this work, we study the validity of this method. To do this, the results obtained by using this technique were compared to those obtained by rigorously solving the Helmholtz equation for a dielectric cylinder of circular cross-section. As an example of the precision of the method, the Fresnel–Kirchhoff integral method was applied to obtain the refractive index of a hair by fitting the theoretical curve to the experimental results of the diffraction pattern of the hair measured with a CCD camera. In a same manner, the method also was applied to obtain the dimensions of a crack artificially created in a piece of plastic.

Machine learning based frequency modelling

Detection of cracks in structures has always been an important research topic in the industrial domain closely associated with aerospace, mechanical, marine and civil engineering. The presence of the cracks alters the dynamic response properties. Hence, it becomes crucial to locate these cracks in the structures to avoid any catastrophic failures and maintain structural integrity and performance. The study's objective is to propose two distinct statistical procedures for conducting the machine learning experiment for modelling the frequency and show the effect of experiment design on the results. In the study, the predictive performance of machine learning models and their ensembles is compared within each experiment design and between two experimental designs for the task of prediction of first six natural frequencies of a fixed ended cracked beam. The study highlights the significance of more than one experimental design to reduce the confirmation bias in the research and discusses the proposed methods' generalizability over the different modelling constraints and modelling parameters. The study also discusses a real-world implementation of the learned machine learning models from the perspective of Bayesian optimization.

Modal testing and evaluation of cracks on cantilever beam using mode shape curvatures and natural frequencies

This paper presented the process of implementing and developing a novel damage detection approach on beam like-structure using modal parameters. In Theoretical Modal Testing (TMT), the mode shapes, mode shape curvatures and natural frequencies were extracted by formulating differential equations of motion of un-cracked and cracked beams. In Experimental Modal Testing (EMT), the transient excitation method of modal testing is performed by using an impact hammer test. EMT involves an extraction of natural frequencies through frequency response functions at various surface cracks on the beam in a virtual instrumentation environment. Cantilever beams of Aluminum 6061 is considered for analysis with crack position at the interval of 50 mm from fixed end to free end at varying crack depths 20%, 40% and 60% of the beam total depth. The modal data from experimental modal testing was validated with theoretical modal testing data. It was observed that there was a minor change in lower modes; major change in natural frequencies for higher modes with the variation of damage location/depths. To overcome the difficulty in crack detection of structures based on mode shapes and frequencies individually, both the parameters mentioned are collectively used in this work as Frequency-Modeshape Based Damage Detection Technique (FMBDD) to evaluate the crack location as well as crack depth easily and accurately.

Structural Crack Detection and Recognition Based on Deep Learning

The detection and recognition of surface cracks are of great significance for structural safety. This paper is based on a deep-learning methodology to detect and recognize structural cracks. First, a training dataset of the model is built. Then, three neural networks, AlexNet, VGGNet13, and ResNet18, are employed to recognize and classify crack images. The tests indicate that the ResNet18 model generates the most satisfactory results. It is also found that the trained YOLOv3 model detects the crack area with satisfactory accuracy. This study also confirms that the proposed deep learning as a novel technology has the potential to be an efficient and robust tool for crack detection and recognition to replace traditional methods.

Performance Evaluation of Deep CNN-Based Crack Detection and Localization Techniques for Concrete Structures.

This paper proposes a customized convolutional neural network for crack detection in concrete structures. The proposed method is compared to four existing deep learning methods based on training data size, data heterogeneity, network complexity, and the number of epochs. The performance of the proposed convolutional neural network (CNN) model is evaluated and compared to pretrained networks, i.e., the VGG-16, VGG-19, ResNet-50, and Inception V3 models, on eight datasets of different sizes, created from two public datasets. For each model, the evaluation considered computational time, crack localization results, and classification measures, e.g., accuracy, precision, recall, and F1-score. Experimental results demonstrated that training data size and heterogeneity among data samples significantly affect model performance. All models demonstrated promising performance on a limited number of diverse training data; however, increasing the training data size and reducing diversity reduced generalization performance, and led to overfitting. The proposed customized CNN and VGG-16 models outperformed the other methods in terms of classification, localization, and computational time on a small amount of data, and the results indicate that these two models demonstrate superior crack detection and localization for concrete structures.

Short Review of the Use of Acoustic Emissions for Detection and Monitoring of Cracks

Proper detection and monitoring of cracks are crucial when evaluating the health of equipment and structures because of the catastrophic failures that can occur if they evolve without control. Among the group of nondestructive tests (NDT), acoustic emissions (AE) emerge as the most used technique for this purpose. This is due to its high sensitivity to low-energy events, which allows the incipient detection of cracks. In this paper, a short review is presented concerning some practical and research activities on crack detection and monitoring in structures using AE. It is shown that AE is a versatile technique for crack detection and monitoring, with applications in a broad range of materials such as concrete, composites, rocks, metals, and wood and in different types of civil and industrial applications.

Efficient crack detection and quantification in concrete structures using IoT

ABSTRACT Now-a-days public transit in our country has a significant usage of bridges and tunnels. Maintaining the safety of such structures becomes the need of the hour. Cracking can invite sudden failures of concrete structures. Within recent years, there has been an increase in the use of image processing techniques as Non-Destructive Testing (NDT) method to detect defects and anomalies in such structures. Hence, this work presents an efficient image processing model for identifying and quantifying the cracks in common structures using two algorithms. The first algorithm uses pattern classification method to classify the identified specimens as crack or not. This pattern classification algorithm is tested and evaluated with the test specimens of structures having induced cracks. The second algorithm called morphological processing uses an efficient thresholding strategy to improve the detection accuracy of the cracks. Here the aim is to detect even the faintest of cracks and magnify them. Thus, any one of the algorithms can be adopted depending on the application. The developed algorithms have been tested in real field at the final stage and achieved a detection accuracy of nearly 96.37%and all the obtained images have been stored digitally on a web server.

Exploring the Detection Accuracy of Concrete Cracks Using Various CNN Models

Automatic crack detection with the least amount of workforce has become a crucial task in the inspection and evaluation of the performances of concrete structure in civil engineering. Recently, although many concrete crack detection models based on convolutional neural networks (CNNs) have been developed, the accuracy of the proposed models varies. Up‐to‐date, the issue regarding the convolutional neural network architecture with best performance for detecting concrete cracks is still debated in many previous studies. In this paper, we choose three established open‐source CNN models (Model1, Model2, and Model3) which have been well‐illustrated and verified in previous studies and test them for the purpose of crack detection of concrete structures. The chosen three models are trained using a concrete crack dataset containing 40,000 images those with 227 × 227‐pixel in size. The performance of three different convolutional neural network (CNN) models was then evaluated. The comprehensive comparison result indicates that Model2 which used batch normalization is capable of the best performance amongst the three models as selected for concrete cracks detection, with recording the highest classification accuracy and low loss. In a conclusion, we recommend Model2 for a concrete crack detection task.

A new approach for crack detection in plate structures using an integrated extended finite element and enhanced vibrating particles system optimization methods

This study presents a new geometry-based crack detection approach for plate structures based on the integration of a dynamic extended finite element method (XFEM) and a physics-based optimization algorithm called enhanced vibrating particles system (EVPS). The so-called XFEM-EVPS method is applied to solve a forward problem. The problem is unraveled by applying traction loading phases and capturing environmental responses in two dimensions in each phase. Then, the EVPS algorithm is deployed to solve the inverse problem. The studied plate structures are analyzed by means of both fine and coarse meshing. Furthermore, each dissolved problem is similarly investigated through applying noise. The results indicate the efficiency of the proposed XFEM-EVPS method in detecting crack location. A comparative study is further conducted to benchmark the performance of the XFEM-EVPS method against a hybrid XFEM and particle swarm optimization approach.

Structural Crack Detection Using DPP-BOTDA and Crack-Induced Features of the Brillouin Gain Spectrum

Structural damage generally initiates in the form of structural cracks. Thus, developing efficient crack detection techniques is of great importance for the structural health monitoring. In this paper, a new crack identification method is proposed, which is based on the differential pulse-width pair Brillouin optical time domain analysis (DPP-BOTDA) technology and the irregular features of Brillouin gain spectrum (BGS) in the fiber due to structural cracks. The proposed method provides a new way to detect and quantify structural cracks without knowing the strain in the structure. First, the working mechanism of DPP-BOTDA is introduced to illustrate the reason that the DPP-BOTDA, compared to traditional BOTDA technique, can significantly improve the spatial resolution of distributed strain sensing, which is critical for structural crack detection. Then, the BGSs in the fiber with the presence of structural cracks, measured by the DPP-BOTDA, are numerically simulated, from which the crack-induced irregular features of the BGS are summarized. Based these irregular features, new structural crack detection and quantification methods are proposed, which are found to be independent of structural stain. Finally, an experiment is conducted on a simple supported reinforced concrete (RC) beam. The results demonstrate that by using the BGS measured by the DPP-BOTDA, the proposed structural crack identification method successfully detects the occurrence of structural cracks and relatively accurately predicts the crack widths.

Experimental research of thermographic control for detection of cracks in steel structures

The paper presents the results of experimental research of the possibility of using thermographic non-destructive testing for the search of cracks in steel structures. The basis of non-destructive thermal control is the registration of changes in the thermal field that occurs in the case of disturbance of the thermodynamic equilibrium of object with an environment that appears on the surface, and parameters of which allows to obtain the necessary information. The thermal control method is based on the interaction of the thermal field of object with thermodynamically sensitive elements (thermocouple, photodetector, liquid crystal element, bolometer) that convert the field parameters (intensity, temperature gradient, contrast, variability) into the electrical signal of the recording device. An experiment is described which proves the possibility of using thermal nondestructive testing for the detection of cracks in steel structures. As measuring instruments for the experiment were used thermal imagers with different resolution of the matrix of the IR image (thermal image), infrared thermohygrometer, luxmeter. The problems encountered during the use of thermal non-destructive testing and possible solutions are indicated. The authors conclude that the thermographic control method can be used in the examination of steel structures for the qualitative assessment of cracks.

An h-version adaptive FEM for eigenproblems in system of second order ODEs: vector Sturm-Liouville problems and free vibration of curved beams

Purpose This study aims to overcome the involved challenging issues and provide high-precision eigensolutions. General eigenproblems in the system of ordinary differential equations (ODEs) serve as mathematical models for vector Sturm-Liouville (SL) and free vibration problems. High-precision eigenvalue and eigenfunction solutions are crucial bases for the reliable dynamic analysis of structures. However, solutions that meet the error tolerances specified are difficult to obtain for issues such as coefficients of variable matrices, coincident and adjacent approximate eigenvalues, continuous orders of eigenpairs and varying boundary conditions. Design/methodology/approach This study presents an h-version adaptive finite element method based on the superconvergent patch recovery displacement method for eigenproblems in system of second-order ODEs. The high-order shape function interpolation technique is further introduced to acquire superconvergent solution of eigenfunction, and superconvergent solution of eigenvalue is obtained by computing the Rayleigh quotient. Superconvergent solution of eigenfunction is used to estimate the error of finite element solution in the energy norm. The mesh is then, subdivided to generate an improved mesh, based on the error. Findings Representative eigenproblems examples, containing typical vector SL and free vibration of beams problems involved the aforementioned challenging issues, are selected to evaluate the accuracy and reliability of the proposed method. Non-uniform refined meshes are established to suit eigenfunctions change, and numerical solutions satisfy the pre-specified error tolerance. Originality/value The proposed combination of methodologies described in the paper, leads to a powerful h-version mesh refinement algorithm for eigenproblems in system of second-order ODEs, that can be extended to other classes of applications in damage detection of multiple cracks in structures based on the high-precision eigensolutions.