Acoustic Emission Research Articles

Classification of Crack Coordinates through Wavelet - Analysis and CNN: Analysis of Data Preparation and Network Structure

For the two-dimensional localization of acoustic emission events (AE), artificial intelligence (AI) was applied. AE events were generated by breaking pencil leads on a granite plate measuring with dimensions of 50 cm × 50 cm × 6 cm. A total of 128 measurements were carried out at each of the 81 positions on the top surface of the granite plate. The recorded signals with 16 channels were processed using continuous wavelet transformation (CWT). The resulting RGB images served as input data for a convolutional neural network (CNN). The transformed images of the measured AE signals were converted into a four-dimensional input by considering them as deeper input, allowing for improved data capture. The result was analysed using a binary classifier according to the one-versus-all method. However, due to limited computational capacity, processing the four-dimensional data posed a challenge, resulting in significant programming effort. The results of this investigation provide insights into the potential of using AI for the localization of AE events.

Acoustic emission-based leakage detection for gas safety valves: Leveraging a multi-domain encoding learning algorithm

Safety valves in gas distribution systems are crucial for preventing excessive pressure buildups. Leakage in these valves compromises system stability, causing equipment damage and environmental pollution. Acoustic emission signal analysis from the valve body is essential for detecting leakage, but these signals are often weak and easily disrupted by noise, complicating detection. This paper proposes a multi-domain encoding learning algorithm to address these challenges. Unlike conventional single-domain methods, it combines two encoding images, the Gramian angular difference field and the Markov transition field, to enhance the comprehensive expression of leakage features. A lightweight convolutional neural network reduces computing resource dependency, and a convolutional block attention mechanism improves feature identification. Experimental results demonstrate that our method detects gas leakage with an accuracy of at least 96.77%. It maintains superior performance even under intense noise interference, offering a promising solution for detecting weak gas leakages in safety valves amidst high background noise.

Quantum machine learning for additive manufacturing process monitoring

Machine learning is useful for analyzing and monitoring complex manufacturing processes. However, it has several limitations including the curse-of-dimensionality and lack of training data. In this paper, we propose a quantum machine learning strategy to tackle these challenges. Quantum support vector machine is applied to identify the states of machines in fused filament fabrication process based on acoustic emission data. Quantum convolutional neural network is used to detect spatters in laser powder bed fusion process based on coaxial optical images. Our results show that quantum machine learning can achieve the similar accuracy levels of predictions by classical machine learning counterparts, but with exponentially fewer parameters.

Optimising Sensor Placement for Tool Condition Monitoring: A Comparative Analysis of Acoustic Emission Data

In the monitoring of machining processes and in the context of tool condition monitoring, the position of acoustic emission sensors affects the quality of the resulting information. This drives this comparative study which presents a comparison of acoustic emission data recorded simultaneously by a sensor attached to the side of the milling table and another sensor near the spindle during a tool wear test on a CNC milling machine. The signal information is clearly influenced by the different positions. Whereas the signals near the spindle are dominated by vibrations of the spindle motor during machining, these strong contributions do not occur in the recorded data of the machining process detected by the sensor mounted to the table. However, the position of the sensor near the spindle has the advantage of maintaining a constant distance to the source of the acoustic emissions of the machining process, as the sensor moves with the tool. In contrast, the sensor attached to the milling table is affected from the fact that the distance to the milling tool is constantly changing. This results in different dominant frequencies, that are analysed in this study. To analyse both the entire frequency range and the higher frequencies, two distinct approaches are employed, one without and one with the use of a high-pass filter. Furthermore, the recorded signals were used to calculate the characteristic features, namely the Partial Powers, thereby enabling assessment of the condition of the tool. The results demonstrate that this single feature type is sufficient for training a machine learning model to predict the tool wear, focusing on the flank wear width. This study analyses the efficiency of wear prediction with respect to both sensors and frequency spectra, and compares the results. In this context, the most accurate predictions were achieved with Gaussian process regression models.

Experimental study of type-I crack propagation in rock monitored by fiber Bragg grating

The occurrence of rock mass engineering disasters can be reduced significantly by obtaining the key information in the process of rock internal fracture accumulation and evolution through advanced monitoring means and taking certain preventive measures. Through fiber Bragg grating monitoring tests of the type-I crack propagation in rock, the spectral characteristics of the fiber Bragg grating in the process of rock failure were analyzed, the bandwidth expansion was extracted to characterize the local tensile strain during the crack propagation process, and the damage variable based on the bandwidth expansion was calculated to describe the type-I crack propagation process. The research results showed the following: (1) The fiber Bragg grating generated a non-uniform local tensile strain during type-I crack propagation. The bandwidth expansion of the wave spectrum was positively correlated with the local tensile strain. The change in the characteristics of the wave spectrum could provide the local real strain in the process of crack propagation. (2) During the steady propagation of the type-I crack, the bandwidth increased obviously, while the number of acoustic emission events was relatively small, and most of them were in the process zone. The gentle increase in the bandwidth corresponded to the unstable growth stage of the type-I crack, which could be used as a precursor of rock sample instability. During loading, the maximum tensile strain of the type-I crack was 6 %, while the maximum shear strain reached 1.8 %, primarily attributed to tensile failure. The average value of fracture toughness KIC, based on the equivalent linear elastic fracture mechanics model, was determined to be 1.21 MPa·m1/2, and the average value of fracture energy Gf was calculated as 57.7 N/m. The variation of peak reflectivity in fiber Bragg grating was associated with the closure of pre-existing defects and the propagation of microcracks in rocks, resulting in unstable shear strain at crack extension sites. (3) The damage variable D defined by the bandwidth broadening began to become concentrated and accumulated in the crack steady growth stage, and it increased slowly in the crack intensification and evolution stage. The grating was sensitive to changes when it was pasted to the surface at 90°, but it could easily fracture or become unable to capture spectral signals in time. Based on a comprehensive comparison, the monitoring effect was the best when the grating was pasted at 60° relative to the crack.

New Methods and Challenges for IPM in Archives, Modern and Historic Libraries

ABSTRACT In this article the most important aspects of IPM, new challenges like climate change, new pests and updated methods of detection are presented. The grey silverfish has already resulted in many problems in archives and libraries across Europe. To set priorities and evaluate a large storage building with many rooms, we present a pest index for detecting high infestation rates and hotspots in the building. For the biscuit beetle we present two new detection methods: The use of a scent detecting (or sniffing) dog trained to detect the larvae of the beetle and the Insect Activity Detection System (IADS), a new technology that uses the acoustical emissions of the larvae feeding on the material. The last aspect of this paper is climate change: in the future higher temperatures, extreme weather events and introduced species will be new challenges for libraries and archives.

Contact damage response of alumina-based layered architectures under different loading configurations

Hertzian contact damage upon spherical indentation is investigated in alumina-based ceramics designed with embedded textured compressive layers. The effect of the location of the embedded layer and the testing configuration (parallel or normal to the layer orientation) on the sub-surface damage evolution is explored. An acoustic emission system for crack detection and post-mortem cross sectioning are employed to evaluate the onset and extension of damage in the different samples. Distinct acoustic signals detected at different contact loads are correlated with the onset of the ring crack, cone crack propagation and quasi-plastic deformation within the textured layer. The distinct damage mechanisms are influenced by the location of the embedded textured compressive layer and the loading orientation with respect to the layers. The design of layered ceramics for contact applications should consider the loading orientation as well as the contact stress field with respect to the “protective” layer to enhance damage tolerance.

A Hybrid GAN-Inception Deep Learning Approach for Enhanced Coordinate-Based Acoustic Emission Source Localization

In this paper, we propose a novel approach to coordinate-based acoustic emission (AE) source localization to address the challenges of limited and imbalanced datasets from fiber-optic AE sensors used for structural health monitoring (SHM). We have developed a hybrid deep learning model combining four generative adversarial network (GAN) variants for data augmentation with an adapted inception neural network for regression-based prediction. The experimental setup features a single fiber-optic AE sensor based on a tightly coiled fiber-optic Fabry-Perot interferometer formed by two identical fiber Bragg gratings. AE signals were generated using the Hsu-Nielsen pencil lead break test on a grid-marked thin aluminum plate with 35 distinct locations, simulating real-world structural monitoring conditions in bounded isotropic plate-like structures. It is demonstrated that the single-sensor configuration can achieve precise localization, avoiding the need for a multiple sensor array. The GAN-based signal augmentation expanded the dataset from 900 to 4500 samples, with the Wasserstein distance between the original and synthetic datasets decreasing by 83% after 2000 training epochs, demonstrating the high fidelity of the synthetic data. Among the GAN variants, the standard GAN architecture proved the most effective, outperforming other variants in this specific application. The hybrid model exhibits superior performance compared to non-augmented deep learning approaches, with the median error distribution comparisons revealing a significant 50% reduction in prediction errors, accompanied by substantially improved consistency across various AE source locations. Overall, this developed hybrid approach offers a promising solution for enhancing AE-based SHM in complex infrastructures, improving damage detection accuracy and reliability for more efficient predictive maintenance strategies.

An Improved Convolutional Neural Network for Pipe Leakage Identification Based on Acoustic Emission

Oil and gas pipelines are the lifelines of the energy market, but due to long-term use and environmental factors, these pipelines are prone to corrosion and leaks. Offshore oil and gas pipeline leaks, in particular, can lead to severe consequences such as platform fires and explosions. Therefore, it is crucial to accurately and swiftly identify oil and gas leaks on offshore platforms. This is of significant importance for improving early warning systems, enhancing maintenance efficiency, and reducing economic losses. Currently, the efficiency of identifying leaks in offshore platform pipelines still needs improvement. To address this, the present study first established an experimental platform to simulate pipeline leaks in a marine environment. Laboratory leakage signal data were collected, and on-site noise data were gathered from the “Liwan 3-1” offshore oil and gas platform. By integrating leakage signals with on-site noise data, this study aimed to closely mimic real-world application scenarios. Subsequently, several neural network-based leakage identification methods were applied to the integrated dataset, including a probabilistic neural network (PNN) combined with time-domain feature extraction, a Backpropagation Neural Network (BPNN) optimized with simulated annealing and particle swarm optimization, and a Long Short-Term Memory Network (LSTM) combined with Mel-Frequency Cepstral Coefficients (MFCC). Corresponding models were constructed, and the effectiveness of leak detection was validated using test sets. Additionally, this paper proposes an improved convolutional neural network (CNN) leakage detection technology named SART-1DCNN. This technology optimizes the network architecture by introducing attention mechanisms, transformer modules, residual blocks, and combining them with Dropout and optimization algorithms, which significantly enhances data recognition accuracy. It achieves a high accuracy rate of 99.44% on the dataset. This work is capable of detecting pipeline leaks with high accuracy.

Acoustic emission failure detection and localisation in historic fibrous plaster ceiling

Fibrous plaster (FP) is a composite material that has been used for suspended decorative ceilings in historical buildings since the late 19th century. Such ceilings are susceptible to failures, particularly in hard-to-access structural hangers (‘wads’). Due to a limited understanding of structural behaviour of FP ceilings, maintenance and reconstruction decisions are based on visual and tactile inspections. Such inspections are costly, intrusive and ultimately subjective. Non-destructive techniques are needed to detect and localise failures in FP ceilings. An Acoustic Emission system was recently developed to detect and localise damage sources in FP ceilings. The method processes data from AE sensor clusters with one-dimensional convolutional neural networks(1D-CNNs) trained to account for wave propagation anisotropy. This paper presents a case study conducted at the Institution of Civil Engineers (ICE) building in London (UK) to validate the method. The results confirm the ability of the method to localise failures under real site conditions in 450 mm-by-450 mm monitoring areas. The method may enable non-intrusive detection of failures from the underside of ceilings during periodic inspections. The possibility of conducting pre-training in laboratories and covering large monitoring areas up to 1500mm-by-1500mm is established and may facilitate practical applications in the future.

Experimental investigations on the failure characteristics of the slit-contained circular opening under biaxial compression: Insights into the rockburst prevention

The slit-cut method for the rockburst prevention and control is believed effective with its easiness in operation, adjustment and compatibility. However, there is limited advance knowledge of the physics of the slit-cut method, which is vital for the engineering designs. In this study, the biaxial compression tests with a synchronous AE (acoustic emission)-DIC (digital image correlation) monitoring are creatively carried out on the slit-contained circular opening specimens with different slit configurations to demonstrate the academic thoughts and mechanisms of the slit-cut method. The experiments document the two typical failure types namely the internal crack propagation and the dynamic rockburst, which occupy different AE hit rate characteristics and different entropy properties. With the occurrence of rockburst, the AE hit rate presents a bouncing ascend-descend trend, and a higher disorder and chaos is faithfully exhibited. Depending on the slit parameters, the slit-cut method can efficiently mitigate rockburst in terms of the occurrence frequency and magnitude. The underlying mechanism lies in the enhancement of the shear mechanism and the development of the internal cracks through which the stored energy can be greatly dissipated. However, due to the unrestricted shear failure in the slit-contained opening specimens, the opening-scale inward instability can be triggered by the internal crack coalescence, thus posing a threat to the safety of the opening. Implementing the slit-cut method with a consideration of the in-situ stress conditions is evidently essential for the safe excavation.

Correlation and regression analysis of natural electromagnetic and acoustic emission activity in rock samples of the Oktyabrsky deposit

Correlation and regression analysis of natural electromagnetic and acoustic emission activity in rock samples of the Oktyabrsky deposit

Brittle failure geometry analysis based on acoustic emission locations in the Khibiny deposit rock samples

Brittle failure geometry analysis based on acoustic emission locations in the Khibiny deposit rock samples

Fracturing and acoustic emission characteristics of saturated reservoir rocks under constant-amplitude-cyclic loading

Fracturing and acoustic emission characteristics of saturated reservoir rocks under constant-amplitude-cyclic loading

Damage Status and Failure Precursors of Different Coal Impact Types Based on Comprehensive Monitoring of Infrared Radiation and Acoustic Emission

Different coal failure impact types exhibit different damage statuses and failure modes, resulting in distinct signal characteristics of infrared radiation (IR) and acoustic emission (AE). This paper combines IR and AE monitoring methods to innovatively establish coal damage and failure precursor warning models and obtains the IR and AE precursor characteristics for different coal failure impact types. This research shows that there is a good correspondence between IR and AE timing and spatial distribution of different coal impact types. As the impact tendency increases, the intensity of IR and AE signals increases with coal failure, and the AE positioning points and IR high-temperature areas tend to concentrate. The coal body gradually changes from tensile failure to shear failure. The shear cracks in the failure stage of coal with no, weak, and strong impact are 39.9%, 50.9%, and 53.7%, respectively. The IR and AE instability precursor point of coal with no, weak, and strong impact occurred at 55.2%, 66.3%, and 93.4% of coal failure, respectively. After the IR and AE combined instability precursor point, the dissipated energy and combined damage variable increase rapidly, and the coal body will undergo instability and failure. The research results provide a theoretical basis for comprehensive monitoring of coal body failure and rock burst.

Damage and failure assessment of banana/ramie/epoxy composites using acoustic emission monitoring

In recent years, Composite materials that include natural fibers have played a significant role in aerospace, automotive industries, and other engineering applications. Due to its enhancing properties, the usage of composites has owing to an increase in day-to-day life. This research aims to develop a sustainable alternative material pronounced for environmental sustainability for surpassing synthetic fibers. Therefore, real-time monitoring and detection of the damage mechanisms in natural fiber composites are indispensable. In this regard, the acoustic emission (AE) technique lends itself to identifying the microscopic damage mechanisms in composite materials. Compression tests were performed in banana/ramie/epoxy composite specimens, and following consequences, the AE parameters (frequency, amplitude, counts, duration, energy rise time, etc.) were recorded. The novelty of this article is to detect the damage evolution and failure mechanisms of banana/ramie/epoxy composites under quasi-static compression with AE signal analysis for the first time. Findings showed that the stacking of hybrid reinforcing banana and ramie fiber improves the stress distribution, leading to enhanced load-carrying capacity (30.56 %), energy absorption (26.78 %), and stiffness (35.24 the compressive strength reached the maximum strength of the banana/ramie/epoxy composites%).This increase represents a substantial 30 % increase in AE energy absorption prior to failure, suggesting that the composite specimens experienced significant damage or extensive failure mechanisms Further, the AE parameters also showed that the AE energy under compression is relatively low and that frequency disperses between 70 and 150 kHz.

Failure behaviour of various rocks coated new thin spray-on liner with different curing time under uniaxial compression: Insights from AE and DIC

Thin spray-on liners (TSL) is increasingly popular in underground mine support operations. In this study, a new type of TSL material is proposed, and its macro-micro physical and mechanical properties are characterized. A systematic investigation on the deformability behaviour of six rocks (i.e., coal rock, soft red sandstone, marble, hard sandstone, granite, and limestone) covered TSL materials with different curing ages (i.e., 0-day, 3-day, 7-day, 14-day, 21-day, and 28-day) under uniaxial compression circumstance is performed using acoustic emission (AE) and digital image correlation (DIC) methods. The results show that TSL improves the peak strain of each rock sample, except hard sandstone. The impact of TSL on the peak compressive strength of coal rock, soft red sandstone, and marble is nearly negligible, while it diminishes the strength of hard sandstone, granite, and limestone. As the TSL ages, the peak strength and elastic modulus of limestone initially increase and then decrease, reaching the minimum at 14-day of age, with values 89.81 MPa and 23.13 GPa respectively. However, for other rock samples, the effect of TSL age is not significant. The AE signal produced inside the rock diminishes during the elastic-plastic stage, while the ratio of shear cracks increases upon failure of each rock specimen. The macroscopic failure modes of coal rock and limestone exhibit prominent axial splitting characteristics, whereas marble and hard sandstone undergo mixed failure processes involving shear, tension, and shear-tension. The test data highly correspond to the macroscopic fracture characteristics. The research findings can serve as guidance for TSL support in both soft and hard rock scenarios.

Use of welded specimens testing for assessment of acoustic emission equipment predictive properties

Use of welded specimens testing for assessment of acoustic emission equipment predictive properties

FE-based machine learning model for predictive damage assessment in bonded composite joints via acoustic emission

This study systematically evaluated the mechanical performance of bonded composite single-lap joints with varying overlap lengths under tensile load. By integrating Acoustic Emission (AE) signals with Finite Element (FE) modeling, this study established a predictive machine learning model to enhance the understanding and prediction of failure mechanisms. Firstly, AE signal features were extracted and utilized as inputs for a hierarchical clustering model to classify AE signals into matrix cracking, fiber breakage, and adhesive debonding. AE signals classified as adhesive debonding were further analyzed by accumulating counts, amplitude, and energy. A cohesive zone-based Finite Element (FE) model was developed to simulate progressive debonding damage in the adhesive layer. After calibration with experimental results, various physical quantities such as stress, the damage initiation indicator, and the stiffness degradation indicator were extracted for correlation analysis. The experiments and simulations unravel the missing relationship between the AE and FE data. A strong correlation was observed between the cumulative count and cumulative amplitude with the damage initiation indicator, and between cumulative energy with the damage propagation indicator. Finally, based on the disclosed relationship between AE and FE data, a Support Vector Regression (SVR) model was established to predict the damage indicator. The developed model accurately predicted testing datasets within low absolute error ranges, underscoring its high predictive accuracy. Comprehensive stress analysis proposed a quantitative damage initiation indicator of 0.32 for structural fractures and a damage indicator of 0.12 for allowable stress levels. Consequently, combining AE techniques with FE modeling provided an effective tool for evaluating damage evolution in bonded composite joints.

Investigation on mechanical property and failure mechanism of the carbon/glass hybrid laminate bolted joints using acoustic emission

In this study, tensile tests and acoustic emission (AE) monitoring were performed on single lap specimens of carbon/glass hybrid laminates with different carbon fiber layer angles. The results indicated that the carbon fiber layer angle not only affects the bearing strength and bearing stiffness of hybrid laminates, but also influences the macroscopic extension direction of damage. The study also found that carbon fiber and glass fiber damage exhibit different centroid frequencies. Building on these findings, the analysis identified the primary fiber type involved in fiber damage to different hybrid fiber laminates. Additionally, it was noted that the friction generated between the bolt and the laminate under load triggers AE signals with a high centroid frequency.