Denoising Fault-Aware Wavelet Network:A Signal Processing Informed Neural Network for Fault Diagnosis

Deep learning(DL) is progressively popular as a viable alternative to traditional signal processing(SP) based methods for fault diagnosis. However, the lack of explainability makes DL-based fault diagnosis methods difficult to be trusted and understood by industrial users. In addition, the extraction of weak fault features from signals with heavy noise is imperative in industrial applications. To address these limitations, inspired by the Filterbank-Feature-Decision methodology, we propose a new Signal Processing Informed Neural Network(SPINN) framework by embedding SP knowledge into the DL model. As one of the practical implementations for SPINN, a denoising fault-aware wavelet network(DFAWNet) is developed, which consists of fused wavelet convolution(FWConv), dynamic hard thresholding(DHT),index-based soft filtering(ISF), and a classifier. Taking advantage of wavelet transform, FWConv extracts multiscale features while learning wavelet scales and selecting important wavelet bases automatically;DHT dynamically eliminates noise-related components via point-wise hard thresholding;inspired by index-based filtering, ISF optimizes and selects optimal filters for diagnostic feature extraction. It’s worth noting that SPINN may be readily applied to different deep learning networks by simply adding filterbank and feature modules in front. Experiments results demonstrate a significant diagnostic performance improvement over other explainable or denoising deep learning networks. The corresponding code is available at https://github. com/alber tszg/DFAWn et.

Dynamics and Fault Diagnosis of Railway Vehicle Gearboxes:A Review

The railway vehicle gearbox is an important part of the railway vehicle traction transmission system which ensures the smooth running of railway vehicles.However,as the running speed of railway vehicles continues to increase,the railway vehicle gearbox is exposed to a more demanding operating environment.Under both internal and external excitations,the gearbox is prone to faults such as fatigue cracks,and broken teeth.It is crucial to detect these faults before they result in severe failures and accidents.Therefore,understanding the dynamics and fault diagnosis of railway vehicle gearbox is needed.At present,there is a lack of systematic review of railway vehicle gearbox dynamics and fault diagnosis.So,this paper systematically summarizes the research progress on railway vehicle gearbox dynamics and fault diagnosis.To this end,this paper first summarizes the latest research progress on the dynamics of railway vehicle gearboxes.The dynamics and vibration characteristics of the gearbox are summarized under internal and external excitations,as well as faulty conditions.Then,the stateof-the-art signal processing and artificial intelligence methods for fault diagnosis of railway vehicle gearboxes are reviewed.In the end,future research prospects are given.

Compound Fault Diagnosis for Rotating Machinery:State-of-the-Art,Challenges,and Opportunities

Compound fault,as a primary failure leading to unexpected downtime of rotating machinery,dramatically increases the difficulty in fault diagnosis.To deal with the difficulty encountered in implementing compound fault diagnosis(CFD),researchers and engineers from industry and academia have made numerous significant breakthroughs in recent years.Admittedly,many systematic surveys focused on fault diagnosis have been conducted by reputable researchers.Nevertheless,previous review articles paid more attention to fault diagnosis with several single or independent faults,resulting in that there is still lacking a comprehensive survey on CFD.Therefore,to fulfill the above requirements,it is necessary to provide an in-depth overview of fault diagnosis methods or algorithms for compound faults of rotating machinery and uncover potential challenges or opportunities that would guide and inspire readers to devote their efforts to promoting fault diagnosis technology more effective and practical.Specifically,the backgrounds,including the related definitions and a new taxonomy of CFD methods,are detailed according to the way of implementing compound fault recognition.Then,the stateof-the-art applications of CFD are overviewed based on relevant publications in the past decades.Finally,the challenges and opportunities associated with implementing CFD are concluded and followed by a conclusion for ending this survey.We believe that this review article can provide a systematic guideline of CFD from different aspects for potential readers and seasoned researchers.

A New Method for Rolling Element Bearing Fault Diagnosis Based on Cyclostationary Theory

The theory of cyclostationary and its application are very important for the analysis and processing of a non stationary signal. The paper introduces second order cyclostationary statistics, with emphass on discussion of cyclic periodogram arithmetic. Comparing the time smoothed cyclic periodogram with the frequency smoothed cyclic periodogram, we found that the former is more useful to extract the feature of cyclostationary signals. The method has been applied to analyze the vibration signal of a rolling element bearing measured on a test bench, and proved to be effective. Meanwhile, we have compared it with traditional power spectral density analysis, and the results prove that the time smoothed cyclic periodogram is more available to diagnose the fault of a rolling element bearing.

PERFORMANCE ANALYSIS OF SECOND-ORDER STATISTICS FOR CYCLOSTATIONARY SIGNALS

The second order statistics for cyclostationary signals were introduced, and their performance were discussed. It especially researched the time lag characteristic of the cyclic autocorrelation function and spectral correlation characteristic of spectral correlation density function. It was pointed out that those functions can be available to extract the time vary information of the kind of non stationary signals. Using the relations of time lag cyclic frequency and frequency cyclic frequency independently, vibration signals of a rolling element bearing measured on test bed were analyzed. The results indicate that the second order cyclostationary statistics might provide a powerful tool for the feature extracting and fault diagnosis of rolling element bearing.

Gear fault diagnosis using gear meshing stiffness identified by gearbox housing vibration signals

Gearbox fault diagnosis based on vibration sensing has drawn much attention for a long time.For highly integrated complicated mechanical systems,the intercoupling of structure transfer paths results in a great reduction or even change of signal characteristics during the process of original vibration transmission.Therefore,using gearbox housing vibration signal to identify gear meshing excitation signal is of great significance to eliminate the influence of structure transfer paths,but accompanied by huge scientific challenges.This paper establishes an analytical mathematical description of the whole transfer process from gear meshing excitation to housing vibration.The gear meshing stiffness(GMS)identification approach is proposed by using housing vibration signals for two stages of inversion based on the mathematical description.Specifically,the linear system equations of transfer path analysis are first inverted to identify the bearing dynamic forces.Then the dynamic differential equations are inverted to identify the GMS.Numerical simulation and experimental results demonstrate the proposed method can realize gear fault diagnosis better than the original housing vibration signal and has the potential to be generalized to other speeds and loads.Some interesting properties are discovered in the identified GMS spectra,and the results also validate the rationality of using meshing stiffness to describe the actual gear meshing process.The identified GMS has a clear physical meaning and is thus very useful for fault diagnosis of the complicated equipment.

Time-frequency Feature Extraction Method of the Multi-Source Shock Signal Based on Improved VMD and Bilateral Adaptive Laplace Wavelet

Vibration signals have the characteristics of multi-source strong shock coupling and strong noise interference owing to the complex structure of reciprocating machinery.Therefore,it is difficult to extract,analyze,and diagnose mechanical fault features.To accurately extract sensitive features from the strong noise interference and unsteady monitoring signals of reciprocating machinery,a study on the time-frequency feature extraction method of multi-source shock signals is conducted.Combining the characteristics of reciprocating mechanical vibration signals,a targeted optimization method considering the variational modal decomposition(VMD)mode number and second penalty factor is proposed,which completed the adaptive decomposition of coupled signals.Aiming at the bilateral asymmetric attenuation characteristics of reciprocating mechanical shock signals,a new bilateral adaptive Laplace wavelet(BALW)is established.A search strategy for wavelet local parameters of multi-shock signals is proposed using the harmony search(HS)method.A multi-source shock simulation signal is established,and actual data on the valve fault are obtained through diesel engine fault experiments.The fault recognition rate of the intake and exhaust valve clearance is above 90%and the extraction accuracy of the shock start position is improved by 10°.

储能变流器信号高精度故障诊断方法

随着能源转型和碳中和的全球发展趋势,储能变流器关键组件的稳定性变得至关重要。特别是其功率器件和散热器在实际运行中的稳定性直接关系到整个系统的可靠性。关注储能变流器功率模组振动信号的故障诊断问题,传统诊断方法处理复杂信号时往往面临挑战,需要频繁地调整参数。此外,由于储能变流器的工作环境复杂,现有深度学习诊断方法的性能也不尽如人意。为此,提出一种基于大模型知识和通道注意力网络的储能变流器功率模组故障诊断方法LLMCAN。首先通过预训练的大规模语言模型,在特征提取过程中利用丰富的领域知识,增强模型对复杂功率模组振动信号的分析能力。其次引入通道注意力网络使模型能够自适应学习信号中不同通道之间的关系,提高故障诊断的准确性。在包含1000条真实工况数据的储能变流器信号数据集上进行验证,其中包括正常工况和9种故障模式。实验结果表明,该方法在多种度量指标下均显示出优越性能,其中诊断准确率高达99.8%,远超传统方法,为储能变流器功率模组的故障诊断提供一个高效、准确的解决方案。

针对冲击性故障信号的谱融合特征提取算法

利用盲解卷积方法在时域中进行故障信号特征提取时,常会出现多个信号混淆分离结果,但以往的研究中只强调了分离的部分,而很少对分离后的信号进行进一步的处理,给实际应用造成不便。这里在盲解卷积和谱融合的基础之上,使用核改进的模糊c均值聚类算法,针对机械故障信号的脉冲特性,提出一种针对冲击性故障信号处理的实用型算法。计算机仿真实验证实了该算法的有效性。此算法优化了以往的聚类筛选方法,可以有效排除反卷积后诸多无用信号的干扰,将故障脉冲信号的特征准确提取出来,能提高故障诊断的效率。

MCKD在一种新型随机共振系统下的转动体故障诊断研究

为解决由高阶项限制引起的输出饱和问题,利用分段势函数抗饱和的优良特性,提出了一种新的非饱和三稳二阶随机共振(UTSOSR)系统。首先,通过仿真实验验证了该系统能够显著改善经典三稳二阶随机共振系统的输出饱和问题。其次,基于绝热近似理论,推导出UTSOSR系统的稳态概率密度,平均首次通过时间和功率谱放大因子(SA),并通过分析系统各参数对这些性能指标的影响,来更加深入地探究系统的动力学行为。将SA和信噪比增益(Gsnr)作为评价指标,通过数值仿真验证了UTSOSR系统具有更优越的信号增强和抗噪声性能。同时,为了获得更优的输出性能,将最大相关峭度解卷积(MCKD)与UTSOSR系统相结合,提出MCKD-UTSOSR方法对目标信号特征进行提取。最后,联合遗传算法和变步长网格优化算法寻找MCKD-UTSOSR方法的最优参数,并应用于转动体微弱故障信号检测。数据分析结果表明,MCKD-UTSOSR方法相比于其他方法,其信噪比提升了1.1289~23.5854 dB,谱峰峰值提升了88.423~7488.118133,为实际工程中高效的信号处理和故障检测提供了创新和可靠的解决方案。

基于SO-PAA-GAF和AdaBoost集成学习的高压断路器故障诊断

针对在小样本和复杂工况下高压断路器故障诊断识别精度不高的问题,提出一种基于振动信号处理和Ada Boost集成学习的高压断路器故障诊断方法。首先,搭建高压断路器实验平台并采集8种工况下的分闸振动信号。其次,对振动信号进行绝对值处理后,使用分段聚合近似(piecewise aggregate approximation,PAA)进行分段平均,将输出的新序列采用格拉姆角场(Gramian angular field,GAF)转换成图片,并使用Relief F方法对提取的高维图片特征进行重要度排序。最后,将保留的重要特征输入到Ada Boost集成学习模型进行故障诊断,并用蛇优化算法确定最优PAA分段步长和输入分类器特征数量,以进一步提高故障诊断精度。通过分析多种信号处理方式及分类模型可知,图片信号和Ada Boost集成学习模型能够有效处理振动信号并准确判断故障类型,为准确、可靠地诊断高压断路器故障提供了新途径。

柴油机多源冲击振动信号稀疏表示及其故障诊断应用

柴油机在船舶、核电、车辆等领域应用广泛,对其进行监测与故障诊断具有重要意义。随着设备健康监测技术的发展,数据存储压力日益显著,信号稀疏表示成为一种有效的解决措施。针对柴油机振动信号具有强冲击、非平稳的特点,提出一种基于分解信号(Decomposed Signal,DS)字典的柴油机多源冲击信号稀疏表示方法,并以稀疏系数作为特征应用于柴油机气门间隙异常故障诊断。首先,采用变分时域分解(Variational Time-domain Decomposition,VTDD)对信号进行处理获得分解信号。然后,将分解信号组成DS字典。接着,通过正交匹配追踪(Orthogonal Matching Pursuit,OMP)算法实现原信号和分解冲击信号的稀疏表示。最后,以稀疏系数作为特征进行柴油机气门间隙异常故障诊断。测试结果表明,所提方法具有较好的应用效果,故障诊断准确率高于90%。

IMPROVED SINGULAR VALUE DECOMPOSITION TECHNIQUE FOR DETECTING AND EXTRACTING PERIODIC IMPULSE COMPONENT IN A VIBRATION SIGNAL

Vibration acceleration signals are often measured from case surface of arunning machine to monitor its condition. If the measured vibration signals display to have periodicimpulse components with a certain frequency, there may exist a corresponding local fault in themachine, and if further extracting the periodic impulse components from the vibration signals, theseverity of the local fault can be estimated and tracked. However, the signal-to-noise ratios (SNRs)of the vibration acceleration signals are often so small that the periodic impulse components aresubmersed in much background noises and other components, and it is difficult or inconvenient for usto detect and extract the periodic impulse components with the current common analyzing methods forvibration signals. Therefore, another technique, called singular value decomposition (SVD), istried to be introduced to solve the problem. First, the principle of detecting and extracting thesignal periodic components using singular value decomposition is summarized and discussed. Second,the infeasibility of the direct use of the existing SVD based detecting and extracting approach ispointed out. Third, the approach to construct the matrix for SVD from the signal series is improvedlargely, which is the key program to improve the SVD technique; Other associated improvement is alsoproposed. Finally, a simulating application example and a real-life application example ondetecting and extracting the periodic impulse components are given, which showed that the introducedand improved SVD technique is feasible.

旋转机械故障诊断中的振动信号模型综述

对旋转机械振动信号进行信号处理,能够有效提取特征进行故障诊断。振动信号模型来自旋转机械运动学和动力学机理,以数学形式表达,可以指导信号处理方法的设计。随着故障机理研究和信号处理方法研究的推进,研究人员对信号模型进行了发展,并基于这些信号模型设计了相应的信号处理方法。首先,介绍了一般化的信号模型,包括周期信号模型、循环平稳信号模型、自适应谐波模型、波形函数模型、任意阶谐波模型等,以及对应的信号处理方法;其次,分别介绍定工况和变工况条件下针对轴承和齿轮的典型振动信号模型及对应信号处理方法;最后,对振动信号模型的研究发展进行总结和展望,旨在回顾旋转机械故障诊断所涉及的信号模型,并说明其在信号处理算法设计和故障诊断特征提取中的价值和意义。

基于改进层次斜率熵(IHSloE)的信号低频和高频故障特征提取方法

采用传统的基于粗粒化处理的多尺度特征提取方法,无法提取故障信号中的高频部分的故障信息,导致其提取到的故障特征难以准确地表征滚动轴承的故障状态和动态特性,无法保证故障诊断的可靠性和准确性。针对该缺陷,提出了一种基于改进层次斜率熵(IHSloE)和随机森林(RF)的滚动轴承故障诊断方法。首先,利用改进层次化处理代替粗粒化处理,实现了信号的多尺度分析目的,基于斜率熵,提出了改进层次斜率熵的非线性动力学指标;随后,利用IHSloE方法提取了滚动轴承振动信号的故障特征,建立了表征滚动轴承故障特性的故障特征;最后,基于RF模型建立了多故障分类器,并将故障特征输入至RF分类器进行了训练和测试,以实现滚动轴承的故障识别目的;利用滚动轴承数据集进行了实验,并将其与其他的故障特征提取指标进行了对比。研究结果表明:IHSloE方法采用改进的层次化处理,能够快速有效地提取出振动信号中的高频故障特征,诊断准确率达到了99%,而特征提取时间仅为149.35 s;相较于采用粗粒化处理和层次处理的特征提取方法,其准确率至少提高了2%和1%,证明该方法适用于滚动轴承的故障诊断。

Fault Feature Extraction of Diesel Engine Based on Bispectrum Image Fractal Dimension

Fault feature extraction has a positive effect on accurate diagnosis of diesel engine. Currently, studies of fault feature extraction have focused on the time domain or the frequency domain of signals. However, early fault signals are mostly weak energy signals, and time domain or frequency domain features will be overwhelmed by strong back?ground noise. In order consistent features to be extracted that accurately represent the state of the engine, bispectrum estimation is used to analyze the nonlinearity, non?Gaussianity and quadratic phase coupling(QPC) information of the engine vibration signals under different conditions. Digital image processing and fractal theory is used to extract the fractal features of the bispectrum pictures. The outcomes demonstrate that the diesel engine vibration signal bispectrum under different working conditions shows an obvious differences and the most complicated bispectrum is in the normal state. The fractal dimension of various invalid signs is novel and diverse fractal parameters were utilized to separate and characterize them. The value of the fractal dimension is consistent with the non?Gaussian intensity of the signal, so it can be used as an eigenvalue of fault diagnosis, and also be used as a non?Gaussian signal strength indicator. Consequently, a symptomatic approach in view of the hypothetical outcome is inferred and checked by the examination of vibration signals from the diesel motor. The proposed research provides the basis for on?line monitoring and diagnosis of valve train faults.

基于改进EMD-Kurtogram法的滚动轴承早期故障诊断研究

实现滚动轴承早期故障准确诊断的关键是得到故障部位有效振动信息,但实际工程中所采集到的轴承振动信号常含有噪声、干扰成分,给有效信息的选择带来了困难。带通滤波是解决该问题的有效方法之一,但不合理的滤波器参数会降低诊断结果的准确性。为此,提出一种基于改进EMD-Kurtogram法的滚动轴承早期故障诊断方法。该方法首先对EMD方法处理后的采样信号进行重构,再根据快速谱峭度图得到带通滤波器所需要的最优参数,最后经过带通滤波及时频域分析得到故障频率。通过实验平台验证及相关算法的对比得出,所提方法得到的故障倍频信息更加充分、清晰,所含噪声干扰更少,证明了该方法的有效性和先进性。

基于改进格拉姆角场和注意力机制的滚动轴承故障诊断

针对传统的卷积神经网络在噪声环境下特征辨识性差且难以充分挖掘数据信息的问题,提出基于改进的格拉姆角和场(IGASF)和注意力机制的滚动轴承故障诊断模型。首先,根据轴承转速和采样频率计算单个故障周期包含的信号点数,对单个故障周期内采集到的振动信号进行分段聚合,利用IGASF进行编码生成相应特征图;然后,将特征图输入卷积神经网络(CNN)进行滚动轴承故障特征提取,并引入注意力模块实现特征的自适应加权;最后,输入到Softmax层完成滚动轴承故障分类。对比试验结果表明,该方法具有更好的抗噪能力和更高的诊断准确率。

Fault Detection and Identification Using Deep Learning Algorithms in Induction Motors

Owing to the 4.0 industrial revolution condition monitoring maintenance is widely accepted as a useful approach to avoiding plant disturbances and shutdown.Recently,Motor Current Signature Analysis(MCSA)is widely reported as a condition monitoring technique in the detection and identification of individual andmultiple Induction Motor(IM)faults.However,checking the fault detection and classification with deep learning models and its comparison among them selves or conventional approaches is rarely reported in the literature.Therefore,in this work,wepresent the detection and identification of induction motor faults with MCSA and three Deep Learning(DL)models namely MLP,LSTM,and 1D-CNN.Initially,we have developed the model of Squirrel Cage induction motor in MATLAB and simulated it for single phasing and stator winding faults(SWF)using Fast Fourier Transform(FFT),Short Time Fourier Transform(STFT),and Continuous Wavelet Transform(CWT)to detect and identify the healthy and unhealthy conditions with phase to ground,single phasing and in multiple fault conditions using Motor Current Signature Analysis.The faults impact on stator current is presented in the time and frequency domain(i.e.,power spectrum).The simulation results show that the scalogram has shown good results in time-frequency analysis for fault and showing its impact on the energy of current during individual fault and multiple fault conditions.This is further investigated with three deep learning models(i.e.,MLP,LSTM,and 1D-CNN)for checking the fault detection and identification(i.e.,classification)improvement in a three-phase induction motor.By simulating the three-phase induction motor in various healthy and unhealthy conditions in MATLAB,we have collected current signature data in the time domain,labeled them accordingly and created the 50 thousand samples dataset for DL models.All the DL models are trained and validated with a suitable number of architecture layers.By simulation,the multiclass confusion matrix,precision,recall,and F1-score are obtained in several conditions.The result shows that the stator current signature of the motor can be used to detect individual and multiple faults.Moreover,deep learning models can efficiently classify the induction motor faults based on time-domain data of the stator current signature.In deep learning(DL)models,the LSTM has shown better accuracy among all other three models.These results show that employing deep learning in fault detection and identification of induction motors can be very useful in predictive maintenance to avoid shutdown and production cycle stoppage in the industry.

基于相空间重构与PCA的潜水泵故障诊断

针对潜水泵信号存在较大干扰影响诊断效果问题,提出相空间重构与主成分分析(PCA)相结合的方法,将潜水泵一维时序信号通过C-C算法计算得到嵌入维数和延迟时间进行相空间重构,并对重构数据进行PCA,将满足贡献累加值的主成分提取出来并还原为一维空间,通过频谱分析得到数据频谱图。通过具体实验,将所提方法处理后的正常状态与故障状态频谱图进行对比,对潜水泵故障进行分析与诊断。实验结果表明,该方法得到的频谱图能够凸显信号故障频率,在一定程度提高了故障诊断准确率。