Acoustic Emission Source Location Research Articles

Method of Locating Loose Particles Inside Aerospace Equipment Based on Parameter-optimized XGBoost

The complex internal structure and heterogeneous composition material of aerospace equipment make it difficult to apply traditional acoustic emission source localization methods to the loose particle localization research. In previous studies, the authors transformed the loose particle localization problem into the multi-classification problem in machine learning, trained the loose particle localization model, but there were problems of low classification accuracy and low practicality. In this paper, the authors first introduced the ensemble learning idea into the loose particle detection field, designed a complete loose particle localization scheme, analyzed the above problems from multiple perspectives, and proposed new algorithms or strategies to enhance the superiority and practicability of the loose particle localization model. Specifically, in view of the low classification accuracy, the authors carried out the research on pulse preprocessing, feature engineering and model training from the four perspectives of signal, feature, data set and classifier, respectively. The zero-pulse-filling pulse matching algorithm and the channel-weighting-based feature selection method was newly proposed, Mel-Frequency Cepstral Coefficients features was newly extracted, feature optimization scheme was designed, and XGBoost ensemble classifier was trained. Thus, high-quality loose particle signals, high-quality localization data sets and high-performance loose particle localization models were obtained, respectively. Test results show that, the classification accuracy achieved by the new loose particle localization model was 96.80%, which was a significant improvement compared to the 83.53% achieved in previous studies. In view of the low practicality, the authors built the loose particle localization experimental system, gave the specific implementation steps of the loose particle localization method, as well as the general procedures for applying the loose particle localization model for physical testing. Meanwhile, taking into account the requirements of aerospace engineering applications, the authors added the majority voting strategy to convert classification results into localization results, thus newly proposed the definition of equipment-level loose particle localization accuracy. Several physical testing results show that, the loose particle localization model achieved a localization accuracy of 90.91%, which effectively verified the feasibility and stability of the proposed method. This study is an important supplement to the loose particle detection research, and of great significance in improving the reliability of aerospace systems.

Acoustic Emission Characterization and Cracking Characteristics of Reinforced Concrete Pipe

Acoustic emission (AE) and digital image correlation (DIC) technology were used in this study to monitor the behavior of reinforced concrete pipe (RCP) in the optimized three-edge bearing test (TEBT). At the same time, the AE source location of different pipe sections was simulated. The results show that the whole compressive loading process for RCP could be divided into 3 to 4 stages. Both the AE counts and AE energy could well characterize the cracking of concrete, and AE energy was more sensitive to the breaking of steel bar. The amplitude and duration of AE signal produced by the breaking of steel bar were larger than those produced by the cracking of concrete. The AE frequency for concrete cracking was relatively scattered. The AE frequency for steel bar breaking was small and concentrated, and this frequency distribution showed an exponential decreasing trend. The results of optimized AE source location showed that the generation and development of cracks occur in the initial loading stage. The results of multi-index analysis showed that the cracks in the top (loading position) and the bottom (bearing position) of the RCP show that the cracks extend from the inner wall to the outer wall, while the cracks on the left and right sides perpendicular to the loading direction show that the cracks extend from the outer wall to the inner wall.

Arrival-Time Detection With Multiscale Wavelet Analysis and Source Location of Acoustic Emission in Rock

In order to overcome the difficulties of traditional arrival time-picking techniques applied to Acoustic Emission (AE) events with low signal-to-noise ratio (SNR) and complex waveforms, this work proposed a new time-picking method. It combines the advantage of the wavelet transform technique, the ratio of Short-Term Average to Long Term Average (STA/LTA) method, and the Akaike Information Criterion (AIC), which is called the Wavelet-STA/LTA-AIC (WSA) picker. This work assessed the performance of picking arrival times of the WSA picker by locating the AE events from pencil-lead breaking on the marble panel (artificial sources) and tensile cracking sources on the granite. For all experiment results, the proposed WSA picker shows less location error compared with the AIC picker, regardless of geometry and source type. For location results of the pencil-lead breaking events on the marble panel, the average location errors are 1.4cm (AIC picker) and 0.6cm (WSA picker). For location results of the tensile cracking events on granite, the WSA picker is more concentrated in the region of the fracture surface and has fewer wrong-locating points at the specimen edge when compared with the AIC picker. It demonstrates that the proposed WSA picker can improve the accuracy and capability of AE location and be a viable method for arrival-time picking.

Cluster-based sensor selection framework for acoustic emission source localization in concrete

The acoustic emission (AE) method can determine the location of damage initiation and progression in large-scale structures using an array of sensors. As the location accuracy depends on proper identification of time of arrival and wave velocity, the method is more successful in homogeneous and isotropic materials than in heterogeneous materials such as concrete. The heterogeneity causes dispersive and attenuative properties such that the source-sensor distance and angle control the AE signal characteristics influencing the source location accuracy. Generally, using more than the minimum number of sensors in the location algorithm results in more accurate source localization. If the AE signal of a sensor is significantly different from the other sensors, its time of arrival may not contribute beneficially to the source localization algorithm. On the contrary, this may increase errors in determining arrival time due to signal distortion caused by dispersion, low signal-to-noise ratios resulting from attenuation, or sensors being occupied with detecting nearby noise signals. In this paper, a new cluster-based sensor selection framework is developed for selecting the best sensor combinations before applying the source localization algorithm. The framework involves selecting the best combination of sensors to input into the source localization algorithm based on the cross-correlation characteristics of signal features for single AE events or cluster analyses for large datasets. Both approaches identify the signal similarity of sensors to be used in the location sensor group. As two-dimensional source localization requires a minimum of three sensors, the number of sensor outputs extracted from the sensor selection framework is bounded by three. The source localization accuracy is evaluated using the data collected impact excitation with the known source location and structural tests of a Basalt Fiber-Reinforced Polymer (BFRP) reinforced concrete slab. The impact experiments show that the new framework to select the best three-sensor combination increases the localization accuracy to 93%, unlike the results obtained by using all six sensors (80%) and the first three sensors in the hit sequence (75%). The framework is applied to the AE data recorded from actual damage initiation and progression in a simply supported BFRP-reinforced concrete slab. The cluster of events accumulates more densely at the mid-span, where the crack initiation is validated with crack sensors and images. It is demonstrated that selecting the sensors for two-dimensional source localization using similarity analyses improves the accuracy of the source location in concrete.

Acoustic Emission Source Localization using Deep Transfer Learning and Finite Element Modeling–based Knowledge Transfer

This paper presents a novel data-driven approach to localize two types of acoustic emission sources in an aluminum plate, namely a Hsu-Nielsen source, which simulates a crack-like source, and steel ball impacts of varying diameters acting as the impact source. While deep neural networks have shown promise in previous studies, achieving high accuracy requires a large amount of training data, which may not always be feasible. To address this challenge, we investigated the applicability of transfer learning to address the issue of limited training data. Our approach involves transferring knowledge learned from numerical modeling to the experimental domain to localize nine different source locations. In the process, we evaluated six deep learning architectures using tenfold cross-validation and demonstrated the potential of transfer learning for efficient acoustic emission source localization, even with limited experimental data. This study contributes to the growing demand for running deep learning models with limited capacity and training time and highlights the promise of transfer learning methods such as fine-tuning pretrained models on large semi-related datasets.

Cyclic tensile behavior of ultra-high performance concrete with hybrid steel-polypropylene fiber: Experimental study and analytical model

The present study deals with the effect of steel-polypropylene hybrid fiber on the tensile stress–strain behavior of ultra-high performance concrete (UHPC) subjected to monotonic and cyclic loading. Eight batches of UHPC samples with hybrid fibers were tested, with the failure process monitored by acoustic emission (AE) technique synchronously. The failure mechanism of UHPC and reinforcing mechanism of hybrid fibers were analyzed based on the combination of AE source location and scanning election microscope (SEM) observation results. The results indicated that the inclusion of hybrid steel-polypropylene fibers in UHPC matrix contributes to the enhancement of tensile mechanical performance with respect to the tensile strength, peak strain and especially the tensile toughness. Moreover, the relationship between plastic strain accumulation and envelope unloading strain during cyclic tension presents a linear pattern and appears insensitive to the variation of fiber parameters. However, the hybrid fibers significantly contribute to alleviation of the elastic stiffness degradation of UHPC during the cyclic tension. Based on the stiffness degradation process, a damage evolution law characterized by an exponential equation associated with fiber reinforcement indexes is proposed. Notably, the failure of UHPC specimen stems from concurrent macro-crack propagation and multiple cracking behavior throughout the whole loading process, which is improved due to the multi-scale reinforcement effect of hybrid fibers. Finally, in consideration of the effect of hybrid fibers, an analytical constitutive model is developed to generalize the tensile behavior of UHPC, and the comparisons yield a close estimation with test results in current research and literature.

A Joint Acoustic Emission Source Localization Method for Composite Materials.

Damage localization methods for composite materials are a popular research topic at present. The time-difference-blind localization method and beamforming localization method are often individually utilized in the localization of the acoustic emission sources of composite materials. Based on the performances of the two methods, a joint localization method for the acoustic emission sources of composite materials is proposed in this paper. Firstly, the performance of the time-difference-blind localization method and the beamforming localization method were analyzed. Then, with the advantages and disadvantages of these two methods in mind, a joint localization method was proposed. Finally, the performance of the joint localization method was verified using simulations and experiments. The results show that the joint localization method can reduce the localization time by half compared with the beamforming localization method. At the same time, compared with the time-difference-blind localization method, the localization accuracy can be improved.

Influence of the Pre-Existing Defects on the Strain Distribution in Concrete Compression Stress Field by the AE and DICM Techniques

This research investigates the influence of the pre-existing defects within concrete taken from the in-service irrigation structure on the strain distribution. The X-ray Computed Tomography (CT) technique is employed to investigate the internal concrete matrix and evaluate the defect distribution in it. The cracking system in a concrete matrix is detected as a damage type caused by the severe environment, and it is varied by the different degrees in all samples. The geometric properties of defects and their spatial location are obtained by image processing of CT images. The compression test with Acoustic Emission (AE) and Digital Image Correlation (DIC) measurements is conducted to analyze the fracture processes and acquire the damage spatial information. The AE signal descriptors are effective parameters for real-time detection and potential local damage monitoring. Moreover, the analysis of the DICM strain and displacement fields reveals the most potential fracture zones. The AE source location analysis indicated a connection between pre-existing defects and strain localization. The AE events and strain are high in the defect areas. Additionally, the amplitude and frequency of the AE events correlated with the location of the defects indicating that the structure weakness at that point leads to concentrated deformation development.

Application of Classified Elastic Waves for AE Source Localization Based on Self-Organizing Map

Acoustic emission (AE) source localization has been used to visualize progress failures generated in a wide variety of materials. In the conventional approaches, AE source localization algorithms assume that the AE signal is propagated as a straight line. However, it is supposed that progress failures form heterogeneity of elastic wave velocity distributions. Hence, with the conventional source localization, it is expected that the localization accuracy is reduced in heterogeneous materials since diffraction and refraction waves are generated. Thus, if the straight propagation waves are classified, conventional source localizations are performed in the heterogeneous materials. The self-organizing map (SOM) is one of the unsupervised learning methods, and the SOM has potential to classify straight propagation waves for the source localizations. However, the application of classified AE signals in source localization is not popular. If classified AE signals are applied to the time difference of arrival (TDOA) method, which is the popular localization method, it is expected that number of visualized sources are decreased because the algorithm does not consider the selection of the propagated wave. Although ray tracing has potential to localize a larger number of sources than the TDOA method, it is expected that the localized sources are less accurate in comparison with results of the TDOA method. In this study, classified waves were applied to two of the source localizations, and model tests based on pencil-lead breaks (PLBs) generating artificial AE sources were conducted to validate the performance of the source localizations with classified waves. The results of the validation confirmed that the maximum error in the TDOA method is larger in comparison with ray tracing conducted with 20 mm intervals of source candidates. Moreover, ray tracing localizes the same number of sources as the number of PLB tests. Therefore, ray tracing is expected to more practically localize AE sources than the TDOA method if classified waves are applied.

A probabilistic framework for source localization in anisotropic composite using transfer learning based multi-fidelity physics informed neural network (mfPINN)

The practical application of data-driven frameworks like deep neural network in acoustic emission (AE) source localization is impeded due to the collection of significant clean data from the field. The utility of the such framework is governed by data collected from the site and/or laboratory experiment. The noise, experimental cost and time consuming in the collection of data further worsen the scenario. To address the issue, this work proposes to use a novel multi-fidelity physics-informed neural network (mfPINN). The proposed framework is best suited for the problems like AE source detection, where the governing physics is known in an approximate sense (low-fidelity model), and one has access to only sparse data measured from the experiment (high-fidelity data). This work further extends the governing equation of AE source detection to the probabilistic framework to account for the uncertainty that lies in the sensor measurement. The mfPINN fuses the data-driven and physics-informed deep learning architectures using transfer learning. The results obtained from the data-driven artificial neural network (ANN) and physics-informed neural network (PINN) are also presented to illustrate the requirement of a multi-fidelity framework using transfer learning. In the presence of measurement uncertainties, the proposed method is verified with an experimental procedure that contains the carbon-fiber-reinforced polymer (CFRP) composite panel instrumented with a sparse array of piezoelectric transducers. The results conclude that the proposed technique based on a probabilistic framework can provide a reliable estimation of AE source location with confidence intervals by taking measurement uncertainties into account.

Structural phenomenological concept of monitoring the load-bearing capacity of structural elements of composite materials

The authors demonstrate the potential of a methodological approach for monitoring the damage kinetics and evaluating the load-bearing capacity of composite structural elements based on the method of acoustic emission. This methodology was developed by specialists of the Blagonravov Mechanical Engineering Research Institute of the Russian Academy of Sciences. The methodology was implemented for a fatigue test of an aircraft panel represented by a complex hybrid structure with a lining made of layered composite and a cellular internal structure of aluminum foil with polymeric filler that has a continuous layered structure in the gripping area. Fatigue tests were carried out at 4 Hz from the zero-loading cycle with an amplitude of 145 kN. To record the accumulation of damages, R15–AST transducers by Mistral (USA) and an A-line 32D 8-channel acoustic emission system by Interunis-IT LLC were used. Comparing the weight content of location impulses with their threshold values in energy clusters characterizing the kinetics of micro-, meso-, and macro-damages of the composite material structure made it possible to determine the bearing capacity of the aircraft panel at respective stages of damage kinetics. The results of the acoustic emission diagnostics are given for the examined aircraft panel during cyclic loading, including the coordinate location of acoustic emission sources, accumulation of acoustic emission events by channels, the dynamics of partial activity changes, and weight content of location impulses at damage evolution stages of a composite material. Therefore, the acoustic emission diagnostics of the aircraft panel made it possible to identify the areas of intensive damage accumulation in the hybrid structure of the panel and control the actual bearing capacity by defining the extent of its damages at various scale and structural levels during cyclic loading. The described methodology of using acoustic emission diagnostics for monitoring the damage kinetics and actual bearing capacity of highly loaded composite elements extends the research potential of the acoustic emission method.

A transfer learning approach for acoustic emission zonal localization on steel plate-like structure using numerical simulation and unsupervised domain adaptation

The detection and localization of damage in metallic structures using acoustic emission (AE) monitoring and artificial intelligence technology such as deep learning has been widely studied. However, a current challenge of this approach is the difficulty of obtaining sufficient labeled historical AE signals for the training process of deep learning models. This problem can be approached through the implementation of transfer learning. The innovation of this paper lies in the development of a transfer learning approach for AE source localization on a stainless-steel structure when no historical labeled AE signals are available for training. A finite element model is developed to generate numerical AE signals for the training. Unsupervised domain adaptation (UDA) technology is utilized to reduce the distribution difference between the numerical and the realistic AE signals and to derive the localization results of the unlabeled realistic AE signals. The results suggest that the proposed approach is capable of localizing AE signals with high accuracy in the absence of labeled training data.

Multi-dimensional non-uniform deformation and failure of siltstone determined using acoustic, 3D-digital image correlation, and computed tomography

Uniaxial compression tests monitored using linear variable differential transformers (LVDTs), digital image correlation (DIC), computed tomography (CT), acoustic emission (AE), and ultrasonic velocity were utilized to study the multi-dimensional non-uniform deformation and failure characteristics of siltstone. Under one-dimensional (1D) conditions, the characteristic stress of the siltstone was determined by the AE counts, ultrasonic velocity, and volumetric strain measured by the extensometer. Under two-dimensional (2D) conditions, zonal disintegration was observed on the apparent fracture of the siltstone specimen. The crack nucleation process, stress release characteristics, and crack type were analyzed in the “broken area,” “penetrating crack area,” and “non-penetrating crack area” under different characteristic stress states. Subsequently, the evolution of the non-uniformity coefficient of the 2D strain field with the 1D characteristic stress was established. Under three-dimensional (3D) conditions, the fracture size of the siltstone was divided according to the equivalent length of fracture volume, and the spatial distributions of macro-, meso-, and microfractures were analyzed. The siltstone 3D zonal disintegration phenomenon was further quantitatively described by calculating the fractal dimensions of fractures in different radius ranges. Finally, the AE source localization method was improved by modifying the wave velocity. The transformation of the primary and secondary bearing areas in different stress stages was revealed based on the spatio-temporal evolution of the AE sources with different energy levels.

Acoustic Emission Source Localization in Heterogeneous Rocks with Random Inclusions Using a PRM-Based Wave Velocity Model

Acoustic Emission Source Localization in Heterogeneous Rocks with Random Inclusions Using a PRM-Based Wave Velocity Model

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

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

Development of novel acoustic emission source location method utilizing wavelet transform and akaike information criterion

The acoustic emission (AE) method enables real-time monitoring of damage initiation and progression. Recently, AE analysis using machine learning has become widely popular; however, the AE source location is often located manually to ensure reliability and accuracy. It is desirable to ensure the AE source location is fully automated with high accuracy when used for machine learning and other applications, or in a situation that generates huge amounts of AE signals. This study proposes a novel AE source location method that can accurately and automatically locate AE sources. First, a wavelet transform was applied to an AE signal to extract the wavelet coefficient of a specific frequency. Then, the Akaike information criterion is applied to the time transient of wavelet coefficient to identify the initial wave arrival time. The localized AE source accuracy of the method is compared with a conventional method. The result of the verification, comparing source location error is nearly same with manual initial detection and the developed method. In addition, about 30 times faster than the conventional visual method. Thus, the developed method is an excellent AE source location method in terms of both accuracy and speed of analysis.

Monitoring Of Fresh Concrete With Superabsorbent Polymers (Saps) Using Acoustic Emission (AE)

In fresh concrete, early-age shrinkage is affected by many factors, such as thermal deformations moisture loss and the hydration reaction. Novel admixtures for internal curing, like SuperAbsorbent Polymers (SAPs) can be used to mitigate the shrinkage phenomenon. These admixtures strongly modify the microstructure influencing the Acoustic Emission (AE) activity. Acoustic emission has been proven adequate to monitor activities during the curing of fresh cement-based materials. However, so far, it is difficult to distinguish the original mechanisms due to the overlapping nature of the processes that take place. In addition, localization of AE sources is not possible due to the heterogeneous and damping nature of the fresh material. The motivation of this study is to obtain real-time information on the different ongoing processes in fresh concrete, like settlement and shrinkage cracking, using AE and comparing the results to concrete containing SAPs. A preliminary study for 3D source localization in fresh concrete is also performed. The goal is to control concrete curing and confirm suitable final mechanical properties, resulting therefore in more sustainable materials.

Review on acoustic emission source location, damage recognition and lifetime prediction of fiber-reinforced composites

Review on acoustic emission source location, damage recognition and lifetime prediction of fiber-reinforced composites

ENTROPY APPROACH FOR FILTERING ACOUSTIC EMISSION SIGNALS

The acoustic emission (AE) method has been successfully used in recent years to monitor the condition of industrial and civil infrastructure structures. In AE, time-of-arrival (ToA) estimation is considered a key parameter for the accurate location of a growing defect. This paper describes an entropy-based filtering approach to improve the ToA estimation of a signal in a noisy environment compared to the commonly used Akaike Information Criterion (AIC). The proposed method consists of coarsening the input data using the Crutchfield-Packard algorithm and calculating the local (instantaneous) entropy. In the present study, it is demonstrated that the local entropy of the noise component of the signal the local entropy of the useful (informative) part can be distinguished. As a result, the approach allows filtering out the noise component by selecting appropriate filtering parameters. The proposed approach has been tested on experimental data obtained on an aluminium plate used to pinpoint an AE source. The entropic approach demonstrates higher accuracy in final target location of AE sources at significant noise levels compared to classical AIC.

Classification of Located Acoustic Emission Events Using Neural Network

Location of acoustic emission (AE) events is one of the main evaluation tools in AE analysis. Reliable location of AE sources enables accurate investigation of the mechanisms that led to a crack in the material. It is known that the location errors are influenced by several factors, including the accuracy of the elastic wave arrival time reading, the geometric distribution of the AE sensors, and most importantly, by the physical properties of the propagation medium. The aim of this study is the application of a neural network to classify clustered AE events, which were detected during six hydraulic fracturing tests in massive salt rock. A fully connected feed forward network was used for pattern recognition and classification of the input events according to target classes. For input data the signal arrival time profiles of the longitudinal (L) and transversal (T) elastic waves were used to train, to test, and to validate the neural network. In total 765 AE events were classified in various target classes. Receiver operating characteristic analysis (ROC) was applied for analyzing the result of the neural network approach. The neural network classified clustered events correctly, while few spatially scattered events outside the region of a cluster could not be matched to any cluster. Bootstrap analysis showed that the results are robust and demonstrates the high potential of Deep Learning (DL) methods in the location of AE events.