Early Detection Of Fatigue Cracks Research Articles

Decoding nonlinear ultrasonic time-frequency characteristics for fatigue crack quantification and localisation via CNN

ABSTRACT Early detection of fatigue cracks is crucial to guarantee the safe operation of engineering structures. Nonlinear ultrasonic technique (NUT) is widely used for closed crack characterisation as it breaks through the detection sensitivity limit upon closed defects that are usually invisible to linear ultrasonic techniques. The nonlinearity parameter is generally defined as the damage index (DI) to qualitatively detect these cracks. However, DI-based methods are underdetermined for quantifying and localising cracks in most circumstances. In this study, a deep learning method for intelligently decoding nonlinear ultrasonic time-frequency characteristics is developed to quantify and localise fatigue cracks. The nonlinear ultrasonic time-frequency spectrogram, which includes features characterising crack severity and location, is obtained by cleverly selecting the excitation frequency. The convolutional neural networks (CNNs) establish a mapping relationship between the nonlinear ultrasonic time-frequency characteristics and the crack length and location. It is worth mentioning that frequency-dependent convolutional kernels are proposed to more competitively decode nonlinear ultrasonic time-frequency characteristics. The results indicate that the CNNs with the frequency-correlation kernels are particularly robust and effective in both noiseless and noisy environments. The study demonstrated the potential of the proposed nonlinear ultrasonic time-frequency characteristic decoding method for accurate and efficient characterisation of damage.

Detection and monitoring of the fatigue crack growth on welds – Application-oriented use of NDT methods

Early detection of fatigue cracks and accurate measurements of the crack growth play an important role in the maintenance and repair strategies of steel structures exposed to cyclic loads during their service life. Observation of welded connections is especially of high relevance due to their higher susceptibility to fatigue damage. The aim of this contribution was to monitor fatigue crack growth in thick welded specimens during fatigue tests as holistically as possible, by implementing multiple NDT methods simultaneously in order to record the crack initiation and propagation until the final fracture. In addition to well-known methods such as strain gauges, thermography, and ultrasound, the crack luminescence method developed at the Bundesanstalt für Materialforschung und -prüfung (BAM), which makes cracks on the surface particularly visible, was also used. For data acquisition, a first data fusion concept was developed and applied in order to synchronize the data of the different methods and to evaluate them to a large extent automatically. The resulting database can ultimately also be used to access, view, and analyze the experimental data for various NDT methods. During the conducted fatigue tests, the simultaneous measurements of the same cracking process enabled a comprehensive comparison of the methods, highlighting their individual strengths and limitations. More importantly, they showed how a synergetic combination of different NDT methods can be beneficial for implementation in large-scale fatigue testing but also in monitoring and inspection programs of in-service structures - such as the support structures of offshore wind turbines.

Combined Acoustic Emission and Digital Image Correlation for Early Detection and Measurement of Fatigue Cracks in Rails and Train Parts under Dynamic Loading

Fatigue crack in rails and cyclic-loaded train parts is a contributory factor in multiple railroad accidents. We address the problem of crack detection and measurement at early stages, when total failure has not yet occurred. We propose to combine acoustic emission (AE) testing for prediction of crack growth with digital image correlation (DIC) for its accurate quantitative characterization. In this study, we imitated fatigue crack appearance and growth in samples of railway rail and two train parts by cyclic loading, and applied these two techniques for inspection. Experimental results clearly indicate the efficiency of AE in the early detection of fatigue cracks, and excellent DIC capabilities in terms of geometrical measurements. Combination of these techniques reveals a promising basis for real-time and non-destructive monitoring of rails and train parts.

Effect of Crack Closure on Magnitude of Modulated Wave

Fatigue cracks generated by repeated loads cause structural failures. Such cracks grow continuously and at an increasing speed owing to the concentration of stresses near the crack tips. Therefore, the early detection of fatigue cracks is imperative in the field of structural-health monitoring for the safety of structures exposed to dynamic loading. In particular, the detection of those cracks subjected to compression is known as a challenging problem in the nondestructive inspection area. The nonlinear ultrasonic modulation technique is effective for the detection of microcracks smaller than the size of a wavelength because this technique uses the deformation of waves passing through the crack surfaces. However, the technique has not been thoroughly verified for detecting cracks subjected to external forces. In this study, nonlinear ultrasonic modulation tests are performed on two types of crack specimens under compressive forces. The results show that in fatigue-cracked specimens, the cracks can be detected using modulated waves even under strong compressions. With artificial cracks, buckling occurs at a relatively low compression, and the amounts of modulated waves rapidly increase due to the bending of the specimen before buckling failure takes place. In this study, the crack detection methodology under compression is proposed and experimentally verified. The proposed method might be beneficial to find cracks under compression in various structural components.

Luminescence-based early detection of fatigue cracks

Classic non-destructive fatigue crack detection methods reveal the state of the fatigue damage evolution at the moment of application, generally not under operational conditions. The here introduced crack luminescence method realizes a clear visibility of the occurred and growing crack in loaded components during operation. Different established experiments show that due to the sensitive coating a crack formation can be detected even in early stage under the premise the crack reached the surface. The coating consists of two layers with different properties and functions. The bottom layer emits light as fluorescence under UV radiation. The top layer covers the fluorescing one and prevents the emitting of light in case of no damage at the surface. In case of surface crack occurrence, the luminescent light is clearly noticeable by visual observations and also by standard camera equipment which makes automated crack detection possible as well. It is expected that crack luminescence can increase structural safety as well as reduce costs and time for inspections and preventive maintenance.

Modeling of microscope images for early detection of fatigue cracks in structural materials

From the perspectives of health monitoring and life extension of structural materials, this paper addresses the problem of early detection of fatigue cracks in metallic materials (e.g., polycrystalline alloys). To this end, optical images have been collected from an ensemble of test specimens to construct computationally efficient models of crack evolution; these images are segmented into two major categories. The first category comprises images of (structurally) healthy specimens, while the second category contains images of specimens with cracks, including those in early stages of crack evolution. Based on this information, algorithms for early detection of crack formation are formulated in the setting of image classification, where the bag-of-words (BoW) technique has been used to develop models of the sensed images from a microscope, resulting in computationally efficient crack detection algorithms. To evaluate the performance of these crack detection algorithms, experiments have been conducted on a special-purpose fatigue testing apparatus, equipped with a computer-controlled and computer-instrumented confocal microscope system. The results of experimentation with multiple test specimens show excellent crack detection capabilities when the proposed BoW-based feature extraction is combined with quadratic support vector machine (QSVM) for pattern classification. Comparative evaluation with other classification tools establishes superiority of the proposed BoW/QSVM technique.

Damage evolution in the AMg6 alloy during high and very high cycle fatigue

Paper presents the “in situ” method for determining of irreversible fatigue damage accumulation, based on the analysis of nonlinear manifestations of the feedback signal in a closed system of an ultrasonic fatigue system. During very high cycle (gigacycle) fatigue, the anomalies of the elastic properties of the material are appear, which leads to a nonlinearity effect in the amplitude of oscillations. This effect increases with the initiation and growth of fatigue cracks. The technology was applied to samples of AMG-6 alloy with preliminary dynamic deformation at various levels of average stress to determine the moment of initiation and growth of the fatigue crack in very high cycle fatigue regime. This method is applicable for the early detection of fatigue cracks both on the surface and inside the material under cyclic loading in the ultrasonic mode. On the basis of wide-range defining relations for a deformable solid body with mesoscopic defects, a mathematical model has been proposed that can adequately describe behavior of the material during fatigue failure. The results of mathematical modeling are in good agreement with the experimental data

A vision-based method for crack detection in gusset plate welded joints of steel bridges using deep convolutional neural networks

Abstract There has been a growing demand for early detection of fatigue cracks in gusset plate joints in steel bridges. The present study develops a robust method for crack detection using the concept of transfer learning as an alternative to training an original neural network. Three standard deep learning methods of training a crack classifier using 1) a shallow convolutional neural network built from scratch, 2) the output features of the VGG16 network architecture previously trained on the general ImageNet dataset, and 3) the fine-tuned top layer of VGG16 are investigated. Data augmentation is used to reduce overfitting caused by the limited and imbalanced training dataset. The image dataset includes both fatigue test photographs and actual inspection photographs captured under uncontrolled distance, lighting, angle, and blurriness conditions. Data augmentation increased the accuracy index by 5%, 2%, and 5%, respectively, for the three methods. The fine-tuning methods achieved the best precision and recall scores and the best robustness. Slightly fine-tuning a well-trained fully connected layer together with the top convolutional layer of the VGG16 model in combination with data augmentation achieved the best performance for crack detection in gusset plate joints of steel bridges. Transfer learning methods using the output features of VGG16 achieved both improved accuracy and robustness and, therefore, are recommended for training a crack classifier.

Temperature profiles obtained in thermoelastic stress test for different frequencies

PurposeMaintenance is one of the most critical and expensive operations during the life cycle of metallic structures, in particular in the aeronautic industry. However, early detection of fatigue cracks is one of the most demanding operations in global maintenance procedures. In this context, non-destructive testing using image techniques may represent one of the best solutions in such situations, especially thermal stress analyses (TSA) using infrared thermography. The purpose of this paper is to access and characterize the main stress profile calculated through temperature variation, for different load frequencies.Design/methodology/approachIn this paper, a cyclic load is applied to an aluminum sample component while infrared thermal image is being acquired. According to the literature and experiments, a cyclic load applied to a material results in cyclic temperature variation.FindingsFrequency has been shown to be an important parameter in TSA evaluations, increasing the measured stress profile amplitude. The loading stimulation frequency and the maximum stress recorded show a good correlation (R2 higher than 0.995). It was verified that further tests and modeling should be performed to fully comprehend the influence of load frequency and to create a standard to conduct thermal stress tests.Originality/valueThis work revealed that the current infrared technology is capable of reaching far more detailed thermal and spatial resolution than the one used in the development of TSA models. Thus, for the first time the influence of mechanical load frequency in the thermal profiles of TSA is visible and consequentially the measured mechanical stress.

Use of acoustic emission and ultrasonic techniques for monitoring crack initiation/growth during ratcheting studies on 304LN stainless steel straight pipe

This paper describes the use of acoustic emission (AE) and ultrasonic techniques for monitoring crack initiation/growth during ratcheting studies on a straight pipe made of 304LN austenitic stainless steel under reversed four point bending. Generation of higher AE activity and AE events with higher peak amplitudes accompanied by AE cluster formation in a localized region between 826 and 845 cycles revealed the initiation of crack during the ratcheting test. Ultrasonic examination by using multiple skip technique conducted periodically at regular intervals of 25 cycles confirmed the formation of the crack only after 1000 cycles. Final rupture of the pipe occurred by through-thickness cracks after 1203 and 1225 cycles at two locations. In the range of cycles where the cracks grew, AE increased with the accumulated cycles, indicating good correlation between crack growth and AE. Acoustic emission was correlated with ballooning. While the ballooning rate decreases with cycles, count rate for different sensors increase. This has been attributed to the superposition of the generation of AE due to crack growth which increases continuously to that generated by circumferential change in strain. The superior sensitivity and reliability of acoustic emission for early detection of fatigue cracks during ratcheting test in pipe specimen is demonstrated.

Nonlinear Dynamics and Damage Assessment of a Cantilever Beam With Breathing Edge Crack

Failures in structures and machine elements can be prevented through early detection of fatigue cracks using various nondestructive testing methods. Vibration testing forms one of the most effective and recent one among these methods. There are mainly two approaches to crack detection through vibration testing: open crack model and breathing crack model. The present study is based on breathing crack model, in which the nonlinear vibration response under harmonic excitation is formulated through Volterra series and higher order frequency response functions. Bilinear stiffness characteristic of a cracked cantilever beam is approximated by a truncated polynomial series and response amplitudes of various harmonics are investigated for both qualitative and quantitative characterization. A new procedure is suggested whereby the presence of a breathing crack in a structure can be first identified and then the severity of the damage can be estimated through harmonic probing.

Sensitivity studies on sensor selection for crack growth investigation

This work focuses on an unsupervised, data driven, modifiedk-nearest neighborhood technique to detect and monitor fatigue crack growth in lug jointsamples using a surface mounted piezoelectric sensor network. A lug joint is an importantstructural hotspot in which damage initiates and progresses under fatigue loading. Earlydetection of fatigue cracks in a lug joint can help in taking preventive measures, thusavoiding any possible structural failure. The lug joint samples used in this study areprepared from an Al 6061 T6 plate with 0.25 inch thickness and are instrumented with asurface mounted piezoelectric actuator/sensor network. Experiments are conducted on lugsamples with a single notch and multiple notches that are symmetrically placed. For earlyinitiation of cracks, samples are notched at the shoulders. Under the influenceof fatigue loading, the crack growth rate is different even when the notches aresymmetrically placed. It is found that although cracks propagate from both thenotches, the sample fails from one of the shoulders once the critical crack length isreached. For the given sensor architecture, which is symmetric, the objective of thisstudy is to detect, isolate and monitor fatigue crack growth in each zone. Themethodology presented helps in identifying sensors that are most sensitive tothe presence of single and multiple cracks. Thus, the computational expense fordamage localization studies can be reduced by not making use of redundant sensors.