Acoustic-based Diagnosis Research Articles

A recursive multi-head self-attention learning for acoustic-based gear fault diagnosis in real-industrial noise condition

Acoustic-based diagnosis (ABD) is a promising method for rotating machinery fault detection in real-industrial fields due to its advantage of non-contact measurement by air-couple. However, most ABD methods, which rely on single function module of denoising or anti-noise diagnosis, lacks sufficient ability against the strong and highly non-stationary background noise interference in practical industrial application. To deal with the obstacle, a novel two-stage ABD system, which performs denoising and anti-noise diagnosis in a step-by-step fashion, is proposed in this paper. In our method, the denoising sub-system can adaptively track and suppress real-industrial background noise from collected multi-channel signal by our designed recursive multi-head self-attention (RMHSA) mechanism, which take into account the global and long-range information of signal to recursively exclude noise component at each position in signal sequence for overcoming the limitation of traditional method in balancing denoising performance and computational complexity and discontinuity in diagnosis procedure. Subsequently, a novel anti-noise method is explored to automatically track the residual noise components after initial suppression and estimating the noise interference probability within time-frequency (T-F) unit of each signal sample by our designed spatial multi-head self-attention (SMHSA) mechanism in recursive manner. Simultaneously, the anti-noise protocol interacts with diagnosis model based on estimated probability in anti-noise diagnosis sub-system to further improve the diagnosis performance of ABD system under the interference of background noise. Experiment result in both real-industrial noise condition and simulated noise conditions with extremely worse SNRs indicate that the proposed two-stage ABD system is effective in dealing with gear fault diagnosis task under noise condition.

A hierarchical adversarial multi-target domain adaptation for gear fault diagnosis under variable working condition based on raw acoustic signal

Acoustic-based diagnosis (ABD) relied on air-coupled measurement is received great attention in recent years due to its ability in overcoming the limitation of vibration-based diagnosis on contact measurement. However, most of the ABD approaches are only focus on a stable working condition for analysis and research. Main challenge of ABD task is the unstable working condition including variable load and speed in real-industrial scenarios, which involve variation in amplitude and phase of acoustic signal to cause domain shift problem for fault pattern detection. In this condition, to perform ABD task for meeting the requirement of scalability in dealing with multiple targets under continuously variable working condition, a novel hierarchical adversarial multiple-target domains adaptation (HAMT-DA) learning framework is proposed in this paper. Following the naturally solution idea, a target-compress adversarial adaptation (TCAA) mechanism is developed in the first-level of framework to compress multiple-target domains by minimizing the feature distribution discrepancy over all domains, and narrow the divergence between source and target domains concurrently. Then, domain invariant feature refining (DIFR) strategy is designed to take responsibility for second-level domain adaptation in the intermediate of framework by refining domain invariant attribute and encoding the domain invariant feature into high-level semantic representation space. Subsequently, a class-separate adversarial adaptation (CSAA) method is explored in the roof of the hierarchical framework for the exploration of fault-discriminative property by forming third-level domain adaptation. Thus, the entire HAMT-DA learning framework can be obtained for single source to multiple-target (SSMT) cross-domain ABD task. The experiment result in variable working conditions indicate that the proposed method has more than 94% average diagnosis performance for gear fault patterns in five cross-domain validation tasks and achieve the highest recognition accuracy among four verifications, which is about 7% average improvement than that of the optimal comparison method. It verifies the superiority of our method on robustness and generalization for acoustic-based gear fault diagnosis under SSMT setting.

Gear Fault Diagnosis Under Variable Load Conditions Based on Acoustic Signals

The health condition of gears has been a topic of study in the past few decades due to the importance of gears for the transmission system. In recent years, some studies have used acoustic signals for gear diagnosis, which can overcome the limitation of vibration signals through noncontact measurement by air-couple. Although many acoustic-based diagnosis (ABD) methods have achieved good diagnosis performance of gear in stable working conditions, these methods suffer from effectiveness loss as the change of working load condition in the actual industry causes the domain shift problem. To overcome the above shortcoming, a domain-adversarial neural network (DANN) with a temporal attention mechanism (TAM) and a high dropout mechanism (HDM) is proposed in this article, which uses the acoustic signal of gears as the input of the model to detect gear health condition. First, the confrontation between the feature extractor and the discriminator in DANN is used to extract domain-invariant features for solving the domain shift problem. Then TAM is introduced into the feature extractor in DANN to refine domain invariant features for further enhancing domain adaptation ability to improve the diagnostic performance. Finally, HDM is utilized to erase the neurons of the input of the classifier with a random high probability to enhance the generalization ability of the model for further improving the classification performance. The experimental results show that the proposed method is effective to solve the domain shift problem of acoustic signals under variable load conditions.

An adaptive anti-noise network with recursive attention mechanism for gear fault diagnosis in real-industrial noise environment condition

Acoustic-based diagnosis (ABD) is a promising method for machinery fault detection due to its ability of non-contact measurement by air-couple. However, most of the ABD methods are constrained by strong and highly non-stationary background noise interference in practical industrial application. To address the shortcoming, a novel anti-noise ABD method based on recursive attention mechanism (RAM) is proposed in this paper. In proposed method, a multi-stage attention module (MSAM) is firstly designed as fundament of RAM to automatically estimate the noise interference probability within time–frequency (T-F) unit of each signal sample. Simultaneously, a recursive learning strategy is introduced to construct RAM by reusing the MSAM for multiple blocks to gradually refine the estimated probability and adaptively simulated noise interference in diagnosis model for enhancing anti-noise diagnosis ability. Then, based on RAM, a domain adaption method is established to endow the model with good cross-domain ability for further improving the anti-noise performance of the diagnosis model. The experiment result in both real-industrial noise condition and stimulated noise conditions with different SNRs indicate that the proposed method has stronger robustness and better generalization ability than other popular methods in dealing with gear fault diagnosis task under noise condition.

A Recursive Denoising Learning for Gear Fault Diagnosis Based on Acoustic Signal in Real Industrial Noise Condition

Acoustic-based diagnosis (ABD) is a promising method for machinery fault detection due to its ability to overcome the limitation of vibration measurement through non-contact measurement by air couple. However, most of the ABD approaches are not widely used in real industrial scenario due to the limitation of strong and highly non-stationary background noise interference. To address the shortcoming, a novel ABD method based on recursive denoising learning (RDL) is proposed in this article. In the proposed method, a new multistage attention mechanism is designed as the fundament of RDL for adaptive tracking and estimating non-stationary industrial background noise and automatic suppressing noise. Based on the multistage attention mechanism, a novel recursive learning strategy is introduced to further improve the performance of noise suppression by recursive tracking noise component and gradual denoising in coarse-to-fine manner. Then, an information fusion method, which is based on an improved tiny-shuffle network (TSN), is adopt to increase the discriminative representation of fault feature through fusion of multi-channel denoising information for improving diagnosis accuracy. Afterward, an RDL-based fault diagnosis method is finally obtained by combining with a standard fault diagnosis model, and it eventually achieves good performance for detection gear fault pattern in noise interference environment. The experimental results in both real industrial background noise condition and additive white Gaussian noise (AWGN) condition with different signal-to-noise ratios (SNRs) indicate that the proposed method performs better than all other popular methods in noise suppression and gear fault pattern detection, which verify the effectiveness of the proposed ABD method in dealing with gear fault diagnosis task under noise condition.

An efficient acoustic-based diagnosis of inter-turn fault in interior mount LSPMSM

This paper investigates the use of acoustic signals in diagnosing stator inter-turn faults in Line Start Permanent Magnet Synchronous Motors (LSPMSMs). Singular spectrum analysis (SSA) is proposed to analyze the acoustic signal for fault detection. The proposed methodology involves collection of experimental acoustic data using a smartphone from an interior mount LSPMSM under different loading levels, healthy as well as faulty motor cases where different number of turns are shorted to emulate different fault levels. Acoustic signal for each case is analyzed using SSA, which results in decomposition of the signal into periodic components and noise. Fast Fourier Transform (FFT) of each component is carried out to find the unique frequency representative for each case. The results with the proposed methodology show that it is capable to detect the occurrence of inter-turn fault under different loading levels with the capability of distinguishing fault severity with readily available acoustic sensor of a smart phone. In addition, it is possible to differentiate between the load and no-load modes of operation.

Learning Attention Representation with a Multi-Scale CNN for Gear Fault Diagnosis under Different Working Conditions.

The gear fault signal under different working conditions is non-linear and non-stationary, which makes it difficult to distinguish faulty signals from normal signals. Currently, gear fault diagnosis under different working conditions is mainly based on vibration signals. However, vibration signal acquisition is limited by its requirement for contact measurement, while vibration signal analysis methods relies heavily on diagnostic expertise and prior knowledge of signal processing technology. To solve this problem, a novel acoustic-based diagnosis (ABD) method for gear fault diagnosis under different working conditions based on a multi-scale convolutional learning structure and attention mechanism is proposed in this paper. The multi-scale convolutional learning structure was designed to automatically mine multiple scale features using different filter banks from raw acoustic signals. Subsequently, the novel attention mechanism, which was based on a multi-scale convolutional learning structure, was established to adaptively allow the multi-scale network to focus on relevant fault pattern information under different working conditions. Finally, a stacked convolutional neural network (CNN) model was proposed to detect the fault mode of gears. The experimental results show that our method achieved much better performance in acoustic based gear fault diagnosis under different working conditions compared with a standard CNN model (without an attention mechanism), an end-to-end CNN model based on time and frequency domain signals, and other traditional fault diagnosis methods involving feature engineering.

Bearing Fault Diagnosis Based on Spatial Features of 2.5 Dimensional Sound Field

The traditional acoustic‐based diagnosis (ABD) technique based on single‐channel testing has a significant engineering value. Since its diagnosis robustness is sensitive to sound signal acquisition location, it develops slowly. To solve this problem, the 2‐dimensional (2D) sound field variation near the machine is adopted for diagnosis by the near‐field acoustic holography (NAH)‐ based fault diagnosis method with array measurement. However, its performance is limited due to the neglect of the sound field normal change information. To dig the sound field fault information further, a 2.5‐dimensional (2.5D) acoustic field diagnosis method is presented in this paper and its performance compared with the 2D technology is verified by the bearing diagnostic test. Different from the 2D technique with only one source image, the 2.5D acoustic field model consists of source image, holographic sound image, and the differences between them, and its effective feature model is constructed by Gabor wavelet feature extraction and random forest feature reduction algorithm. The diagnostic effect of the 2.5D technique compared with the 2D technique increases more than 11% in the bearing diagnostic test. It provides new ideas for the development of the NAH‐based fault diagnosis method, and further improves the ABD technique‐based array measurement.

Fault detection in rotating machines with beamforming: Spatial visualization of diagnosis features

Rotating machines diagnosis is conventionally related to vibration analysis. Sensors are usually placed on the machine to gather information about its components. The recorded signals are then processed through a fault detection algorithm allowing the identification of the failing part. This paper proposes an acoustic-based diagnosis method. A microphone array is used to record the acoustic field radiated by the machine. The main advantage over vibration-based diagnosis is that the contact between the sensors and the machine is no longer required. Moreover, the application of acoustic imaging makes possible the identification of the sources of acoustic radiation on the machine surface. The display of information is then spatially continuous while the accelerometers only give it discrete. Beamforming provides the time-varying signals radiated by the machine as a function of space. Any fault detection tool can be applied to the beamforming output. Spectral kurtosis, which highlights the impulsiveness of a signal as function of frequency, is used in this study. The combination of spectral kurtosis with acoustic imaging makes possible the mapping of the impulsiveness as a function of space and frequency. The efficiency of this approach lays on the source separation in the spatial and frequency domains. These mappings make possible the localization of such impulsive sources. The faulty components of the machine have an impulsive behavior and thus will be highlighted on the mappings. The study presents experimental validations of the method on rotating machines.

A Novel Denoising Method for an Acoustic-Based System through Empirical Mode Decomposition and an Improved Fruit Fly Optimization Algorithm

Generally, the sound signal produced by transmission unit or cutting unit contains abundant information about the working state of a machine. The acoustic-based diagnosis system presents some distinct advantages in some severe conditions particularly due to its unique non-contact measurement and unlimited use at the installation site. However, the original acoustic signal collected from manufacture process is always polluted by various background noises. In order to eliminate noise components from machinery sound effectively, an empirical mode decomposition (EMD) threshold denoising method optimized by an improved fruit fly optimization algorithm (IFOA) is launched in this paper. The acoustic signal was first decomposed by the adaptive EMD to obtain a series of intrinsic mode functions (IMFs). Then, the soft threshold function was applied to shrink the IMF coefficients. While the threshold of each IMF was determined by statistical estimation and empirical value for traditional EMD denoising, the denoising effect was often not desired and time-consuming. To solve these disadvantages, fruit fly optimization algorithm (FOA) was introduced to search global optimal threshold of each IMF. Moreover, to enhance the group diversity during production of the next generation of fruit flies and balance the local and global searching ability, a variation coefficient and a disturbance coefficient was introduced to the basic FOA. Then, a piece of simulated acoustic signal produced by the train was applied to validate the proposed EMD and IFOA threshold denoising (EMD-IFOA). The simulation results, which decreased 35.40% and 18.92% in mean squared error (MSE) and percent root mean square difference (PRD) respectively, and increased 40.36% in signal-to-noise ratio improvement (SNRimp) compared with basic EMD denoising scheme at SNR = 5 dB, illustrated the effectiveness and superiority of the proposed approach. Finally, the proposed EMD-IFOA was conducted on an actual acoustic-based diagnosis system for cutting state recognition of the coal mining shearer to demonstrate the practical effect.

A gearbox fault diagnosis scheme based on near-field acoustic holography and spatial distribution features of sound field

Vibration signal analysis is the main technique in machine condition monitoring or fault diagnosis, whereas in some cases vibration-based diagnosis is restrained because of its contact measurement. Acoustic-based diagnosis (ABD) with non-contact measurement has received little attention, although sound field may contain abundant information related to fault pattern. A new scheme of ABD for gearbox based on near-field acoustic holography (NAH) and spatial distribution features of sound field is presented in this paper. It focuses on applying distribution information of sound field to gearbox fault diagnosis. A two-stage industrial helical gearbox is experimentally studied in a semi-anechoic chamber and a lab workshop, respectively. Firstly, multi-class faults (mild pitting, moderate pitting, severe pitting and tooth breakage) are simulated, respectively. Secondly, sound fields and corresponding acoustic images in different gearbox running conditions are obtained by fast Fourier transform (FFT) based NAH. Thirdly, by introducing texture analysis to fault diagnosis, spatial distribution features are extracted from acoustic images for capturing fault patterns underlying the sound field. Finally, the features are fed into multi-class support vector machine for fault pattern identification. The feasibility and effectiveness of our proposed scheme is demonstrated on the good experimental results and the comparison with traditional ABD method. Even with strong noise interference, spatial distribution features of sound field can reliably reveal the fault patterns of gearbox, and thus the satisfactory accuracy can be obtained. The combination of histogram features and gray level gradient co-occurrence matrix features is suggested for good diagnosis accuracy and low time cost.

A fault diagnosis scheme of rolling element bearing based on near-field acoustic holography and gray level co-occurrence matrix

Vibration signal analysis is the most widely used technique in condition monitoring or fault diagnosis, whereas in some cases vibration-based diagnosis is restrained because of its contact measurement. Acoustic-based diagnosis (ABD) with non-contact measurement has received little attention, although sound field may contain abundant information related to fault pattern. A new scheme of ABD for rolling element bearing fault diagnosis based on near-field acoustic holography (NAH) and gray level co-occurrence matrix (GLCM) is presented in this paper. It focuses on applying the distribution information of sound field to bearing fault diagnosis. A series of rolling element bearings with different types of fault are experimentally studied. Sound fields and corresponding acoustic images in different bearing conditions are obtained by fast Fourier transform (FFT) based NAH. GLCM features are extracted for capturing fault pattern information underlying sound fields. The optimal feature subset selected by improved F-score is fed into multi-class support vector machine (SVM) for fault pattern identification. The feasibility and effectiveness of our proposed scheme is demonstrated on the good experimental results and the comparison with the traditional ABD method. Considering test cost, the quantized level and the number of GLCM features for each characteristic frequency is suggested to be 4 and 32, respectively, with the satisfactory accuracy rate 97.5%.

Diagnosing Rolling Bearing Faults Using Spatial Distribution Features of Sound Field

The vibration-based feature extraction is the main approach for mechanical fault diagnosis,whereas,in some conditions vibration signal is not easily measured because of its contact-measuring.Acoustic-based diagnosis(ABD) can overcome this disadvantage.However,for traditional ABD it is hard to choose proper measuring positions and the acoustic signals acquired based on single channel measurement can be used only for local analysis.Based on near-field acoustic holography(NAH),a new feature extraction method by using sound field distribution for rolling bearing fault diagnosis is presented.Firstly,sound fields in different bearing conditions are reconstructed by NAH.Using gray level co-occurrence matrix(GLCM) features extracted from acoustic images,the inner relationship between bearing conditions and sound fields is established.These features are fed into support vector machine(SVM) classifier for fault diagnosis.The effectiveness of our proposed method is demonstrated on the experimental investigation.The method provides a new reference for mechanical fault diagnosis.