This paper proposes a novel technique for leak detection and localization in industrial fluid pipelines. Artificial intelligence-based supervised methods are a popular adoption for pipeline leak ...identification. However, supervised techniques need prior knowledge about pipeline failure for training purposes. To address this challenge, a new method for pipeline leak detection independent of prior knowledge is proposed. First, acoustic emission signals from the pipeline are acquired. Then, a multiscale Mann-Whitney test is developed, and the test output statistics are used as a pipeline leak state indicator. After detecting the leak, the proposed method localizes the leak by using a newly developed method called acoustic emission event tracking. A major challenge in leak localization is the false alarms that are generated by the poor identification of leak-related acoustic emission events. The acoustic emission event tracking presented in this work precisely determines the leak-related acoustic emission events. The new method first detects acoustic emission hits by using the variability index constant false alarm rate algorithm. Then, short-term energy is calculated in the hit perceived variability index constant false alarm rate windows. The high-energy acoustic emission events are separated into an event bank. Leak-related acoustic emission events are filtered out from the event bank using the theory of wave propagation. The filtered events obtained from the proposed method elucidate leaks and thus reduce the error in leak localization. The results obtained from the proposed method outperformed the reference methods in terms of accuracy for leak detection and localization under variable pressure and leak conditions.
Offshore wind turbines play a vital role in transferring wind energy to electricity, which could help relieve the energy crisis and improve the global climate. In general, offshore wind turbines are ...installed open sea to avoid the potential interruption of people’s daily life. In such kind of harsh operating environment, the wind turbine transmission system is prone to failure, especially for the rolling bearings. Therefore, it is crucial to conduct condition monitoring of rolling bearings to ensure the safe and efficient operation of offshore wind turbines. Intelligent fault diagnosis is a research hotspot for condition monitoring of rolling bearings. However, the existing intelligent fault diagnosis techniques have some limitations. For example, most of the existing techniques were developed based on single sensory data, which can lead to inaccurate and unstable diagnostic results. Moreover, most existing techniques implicitly assume that there are sufficient labeled samples for classifier training. This may not be the case for offshore wind turbines where the labeled samples are limited. To address the aforementioned issues, an intelligent fault diagnosis technique by integrating an information stream fusion and a semi-supervised learning approach is proposed in this study. In the proposed method, a coupled convolutional residual network is proposed to realize the information streams fusion, in which the vibration signal and acoustic emission signal are served as the inputs of the proposed network, and then a concatenation operation is used to fuse the features obtained from two information streams. Meanwhile, a semi-supervised learning approach is also proposed, which can utilize the labeled samples, the correctly predicted samples, and the unlabeled samples to improve diagnostic accuracy. The diagnostic result on the experimental offshore wind turbine bearing dataset demonstrates that the proposed method achieves the highest diagnostic accuracy compared to existing comparative methods.
Comparisons of ANN prediction of collision energy and particle size with DEM simulations.
Display omitted
•An ANN model was proposed to predict particle flow characteristics in rotating drums.•The ...model was based on the acoustic emission signals generated from DEM simulations.•The key features of the signals were obtained through principal component analysis.•Flow properties included filling level, particle size and energy distributions.•ANN predictions compared well with DEM simulations.
Rotating drums are widely used in industries for particle mixing, granulation and grinding. Linking internal particle flow condition with externally measured variables is crucial to online process monitoring and control. This work proposed a modelling framework to use an artificial neural network (ANN) model for quick prediction of particle flow based on the acoustic emission (AE) signals generated from the discrete element method (DEM) simulations. In total 131 DEM simulations were conducted under different conditions (i.e., different particle size distributions and filling levels). The AE signals on the drum surface were then obtained based on the simulated particle–wall collisions. Through FFT transformation and principal component analysis (PCA), 5 principal components (PCs) were obtained and, together with power draw, fed into the ANN model to predict to the unmeasurable internal flow conditions, including filling level and the distributions of particle size and internal collision energy. The back propagation neural network was adopted in the model. After being trained with 90 datasets, the ANN model was able to predict those internal variables with reasonable accuracy (R2 > 0.95). Finally, the potentials and limitations of the model to the optimal operation of drums were discussed.
Display omitted
•Feature characteristics vary with the bearing’s rotational speed.•This paper proposes a reliable fault diagnosis scheme based on acoustic spectral imaging (ASI) of acoustic emission ...signals.•ASI provides a visual representation of acoustic emission spectral features in images.•The proposed approach provides a robust classifier technique with high diagnostic accuracy.
Incipient fault diagnosis of a bearing requires robust feature representation for an accurate condition-based monitoring system. Existing fault diagnosis schemes are mostly confined to manual features and traditional machine learning approaches such as artificial neural networks (ANN) and support vector machines (SVM). These handcrafted features require substantial human expertise and domain knowledge. In addition, these feature characteristics vary with the bearing’s rotational speed. Thus, such methods do not yield the best results under variable speed conditions. To address this issue, this paper presents a reliable fault diagnosis scheme based on acoustic spectral imaging (ASI) of acoustic emission (AE) signals as a precise health state. These health states are further utilized with transfer learning, which is a machine learning technique, which shares knowledge with convolutional neural networks (CNN) for accurate diagnosis under variable operating conditions. In ASI, the amplitudes of the spectral components of the windowed time-domain acoustic emission signal are transformed into spectrum imaging. ASI provides a visual representation of acoustic emission spectral features in images. This ensures enhanced spectral images for transfer learning (TL) testing and training, and thus provides a robust classifier technique with high diagnostic accuracy.
•Acoustic emission sensing system based on demodulating the phase of fiber Bragg grating reflected light is applied to material health detection.•The detections of continuous acoustic emission ...signals and burst acoustic emission signals are achieved.•The maximum signal-to-noise ratio can reach 77 dB and the sensor phase sensitivity is 0.012 rad/με.•Detecting weak burst acoustic emission signals emitted during concrete damage is achieced, with a signal-to-noise ratio of 21.5 dB.
Acoustic emission detection is widely employed in the field of material health monitoring as an important non-destructive testing method. An acoustic emission sensing system based on phase demodulation in fiber Bragg grating is demonstrated in experiment to solve the problems existing in acoustic emission sensor, such as low sensor sensitivity and poor anti-interference capabilities. According to experimental results, the system can effectively detect both continuous and burst acoustic emission signals from 100 kHz to 700 kHz. The signal-to-noise ratio can reach 77 dB and the sensor phase sensitivity is 0.012 rad/με. The dynamic strain resolution of the system is 3.77 nε/√Hz. The system is also able to detect weak burst acoustic emission signals emitted during concrete damage, and the signal-to-noise ratio is 21.5 dB in the acoustic emission sensing in corrosion of reinforcement.
Aiming at the difficulty of predicting the remaining fatigue life of mechanical parts, a similar fatigue life prediction method based on acoustic emission signals is proposed considering the failure ...of vibration signals. Moreover, this paper innovatively introduces the temperature signal as a degradation feature to assist the acoustic emission signal feature for fatigue life prediction. The proposed fatigue life prediction method involves multiple processes such as feature extraction, feature smoothing, feature selection, feature compression and health index construction. A variational autoencoder structure combined with a long short-term memory neural network is used to achieve feature compression and retain feature trend. A tensile fatigue test bench was developed to collect the degradation signal from health to fatigue fracture to validate the proposed method. The validation results show that the proposed method can accurately predict the remaining fatigue life. In addition, the role of various data processing methods and the applicability of the proposed method in different working conditions are also discussed.
•A method is provided for feature extraction and processing of acoustic emission signals in fatigue life prediction.•Introducing a monotonicity loss in the proposed VAE ensures that the compressed features retain the original trend.•The multi-condition fatigue life prediction using the temperature signal to assist the acoustic emission signal is realized.
This study concerns with fault diagnosis of low speed bearing using multi-class relevance vector machine (RVM) and support vector machine (SVM). A low speed test rig was developed to simulate various ...types of bearing defects associated with shaft speeds as low as 10
rpm under several loading conditions. The data was acquired from the low speed bearing test rig using acoustic emission (AE) and accelerometer sensors under a constant load with different speeds. The aim of this study is to address the problem of detecting an incipient bearing fault and to find reliable methods for low speed machine fault diagnosis. In this paper, two methods of multi-class classification techniques for fault diagnosis through RVM and SVM are presented and the effectiveness of using AE and vibration signals due to low impact rate for low speed diagnosis. In the present study, component analysis was performed initially to extract the features and to reduce the dimensionality of original data features. The classification for fault diagnosis was also conducted using original data feature and without feature extraction. The result shows that multi-class RVM produces promising results and has the potential for use in fault diagnosis of low speed machine.
This study involved full-life monitoring of high-cycle fatigue damage evolution in stainless steel cruciform joints with undercuts and their control groups, using acoustic emission technique. Based ...on continuum damage mechanics, the wavelet packet subband energy of acoustic emission signals was used as the damage variable to investigate the relationship between fatigue damage and acoustic emission signals, to quantitatively assess the extent of fatigue damage. The research findings indicated that the wavelet packet subband energy effectively captured the fatigue damage evolution process in both defective and non-defective specimens. The fatigue damage evolution strips obtained from processing acoustic emission signal data using nonlinear normalization method demonstrated insensitivity to fatigue stress amplitudes. The wavelet packet subband energy exhibited excellent identification ability for early-stage fatigue damage and its development before macroscopic crack formation. This study's outcomes were expected to provide strategies for identification and quantitative assessment of fatigue damage in welded structures throughout their entire service life, as well as predicting remaining life.
•A modified fatigue damage evolution model considering initial defects was proposed.•Wavelet packet subband energy effectively identified micro-scale damage before crack formation.•Wavelet packet subband energy distinguished initial damage from weld defects effectively.•The nonlinearly normalized data exhibited insensitivity to the stress amplitude.
•Studies on the scattering of acoustic emission signals in stiffened plates.•The proposed propagation model guides the optimal positions of sensors.•Understanding of the propagation mechanism of ...signals in stiffened plates.•Reflection and transmission coefficients evaluate the scattering of signals.
Mechanical equipment with the stiffener has a strong interference with the propagation of acoustic emission (AE) signals from faults, reducing the accuracy of fault detection. This paper conducts an in-depth study of the interaction between AE signals and the stiffener. The installation constraints, that can separate the direct signal, signals scattered from the stiffener and signals reflected from the boundary in the time domain, for sensors are deduced based on the multipath propagation model of AE signals in the stiffened plate. On this basis, the scattering characteristics of AE signals with different frequencies in different height stiffened plates are predicted by simulations. Moreover, the reflection and transmission coefficients are calculated to quantify the scattering characteristics. The results show that the signal, undergoing a “T-shaped” transformation at the stiffener, generates various modes, among which the transmission signal accounted for the largest proportion. In addition, experiments are performed to verify the numerical simulations, and the results are in good agreement with the numerical simulations. This work clarifies the propagation characteristics of AE signals in stiffened plates, and the research can optimize the spatial arrangement for sensors.