鲸鱼算法优化VMD-CNN在滚动轴承故障诊断中的应用

针对滚动轴承工作环境恶劣且采集到的振动信号具有非线性、非平稳性等特征,为了自适应提取故障特征以及提高轴承故障智能诊断准确率,提出基于鲸鱼算法（Whale Optimization Algorithm,WOA）优化变分模态分解（Variational Mode Decomposition,VMD）与卷积神经网络（Convolution Neural Network,CNN）相结合的故障诊断方法。首先,使用鲸鱼优化算法对VMD超参数进行寻优,找到VMD最优的分解层数与惩罚因子,并利用优化后的VMD对轴承原始信号进行分解。其次,用连续小波变换将分解得到的一维本征模态信号转化为相应的二维时频图。最后,将二维时频图作为二维卷积神经网络的输入,并对其输入的时频图进行深层特征提取与模式识别。实验表明,所提出的方法能高效提取故障特征,准确率高达99.78%。     

基于包络谱稀疏度和最大相关峭度解卷积的滚动轴承早期故障诊断方法

滚动轴承处于早期故障阶段时,特征信号微弱,并且受环境噪声影响严重,因此故障特征提取困难。针对这一问题,将最大相关峭度解卷积算法应用于轴承故障诊断,并通过包络谱稀疏度来筛选最佳解卷积周期参数,提出了基于包络谱稀疏度和最大相关峭度解卷积的滚动轴承早期故障诊断方法。利用最佳参数相对应的最大相关峭度解卷积算法对原信号进行处理,得到解卷积信号后计算其包络谱,通过分析包络谱中幅值突出的频率成分来判断故障类型。早期故障仿真信号及实测全寿命数据分析结果表明,该方法可有效应用于轴承早期故障诊断。     

滚动轴承故障检测深度卷积稀疏自动编码器建模研究

针对机械设备故障诊断大多采用有监督学习提取故障特征,而有标签数据难以获取的现状,提出一种在稀疏自动编码器中嵌入卷积网络的深度神经网络。利用希尔伯特和傅里叶变换实现机械设备振动时间序列向Hilbert包络谱的转换,通过卷积网络中多组卷积核自动学习谱空间数据的不同特征,保证了特征提取的自动化、全面性和多样性,稀疏自动编码器搜索具有正交性数据特征的低维表示,并使得编码后的数据具有很强的聚类特性,实现设备的自动故障诊断。通过对滚动轴承振动信号进行分析实验,证明该方法在设备故障诊断中具有去标签化、自动化、鲁棒性等特点。     

感应电机轴承故障检测方法研究

分析了感应电机轴承发生故障时的振动信号的特性,利用带通滤波器和希尔伯特变换,对感应电机轴承振动信号进行处理,然后采用高分辨率谱估计算法--MUSIC（Multiple Signal Classification）算法对包络信号作谱分析,再从包络信号的MUSIC谱中提取故障特征频率分量.研究结果表明,该方法频率分辨率更高,故障检测更为准确.将该方法应用于电机轴承故障诊断,可准确提取轴承故障特征分量.     

基于MEWT‑ASCS的行星齿轮箱微弱故障特征提取

针对强噪声背景下行星齿轮箱早期微弱故障难以提取以及经验小波变换对信号频率区间边界划分不恰当以及不能有效确定模态数目的问题,提出了一种基于改进经验小波变换（modified empirical wavelet transform, 简称MEWT）和自适应稀疏编码收缩（adaptive sparse coding shrinkage,简称ASCS）的早期微弱故障特征提取方法。根据信号频谱的尺度空间表示,将原始故障信号自适应地分解为一系列的窄频带本征模态分量。利用包络谱峭度（envelope spectrum kurtosis, 简称ESK）值选择敏感分量,为了进一步凸显分量中的故障信息,使用ASCS算法对敏感分量进行稀疏降噪处理,从其包络谱中即可提取到清晰的故障特征频率成分。数值仿真和实际数据分析结果表明,本研究方法能够自适应地实现故障信号的模态分解并增强微弱的故障冲击特征。此外,与经验小波变换（empirical wavelet transform, 简称EWT）,EWT?ASCS和ASCS进行对比,本研究方法可有效提取包含故障信息丰富的分量,经ASCS处理后信号故障特征得到凸显,实现了行星齿轮箱早期微弱故障的准确识别。     

基于IITD和包络信号1.5维谱的轴承故障诊断

为了有效识别轴承的早期故障特征,提出了一种基于改进的本征时间尺度分解(IITD)结合包络信号1.5维谱的轴承故障诊断方法。IITD方法是将端点延拓引入到传统的本征时间尺度分解(ITD)当中,用于改善其端点效应。轴承振动信号经IITD分解后得到一组PR分量和一趋势项,对PR分量的包络信号进行1.5维谱分析。结果表明,IITD分解得到的PR分量包络信号的1.5维谱,可以准确提取轴的转动频率、内圈故障特征频率和外圈故障特征频率,从而实现了轴承故障的有效诊断,证明了该方法的有效性和实用性。     

基于FOA的变分模态分解在轴承故障诊断中的应用

变分模态分解(VMD)广泛应用于故障诊断中,从振动信号中提取故障特征是故障诊断过程中的关键部分。针对强背景噪声和脉冲干扰下滚动轴承早期故障特征难以提取的问题,提出了一种新的基于果蝇优化算法(FOA)的变分模态分解的轴承故障诊断方法。首先,利用果蝇优化算法自适应优化VMD的惩罚参数α和分解数K,获取最优参数组合;然后,对信号进行VMD分解,得到K个模态分量;最后,基于峭度最大化准则选取最优模态分量进行包络解调分析,提取出故障特征频率。通过仿真信号分析、实际故障轴承信号验证以及与基于果蝇优化算法的多分辨奇异值分解(MRSVD)方法进行对比,证明了所提方法的有效性。     

机械故障的稀疏流形聚类与嵌入诊断方法

传统流形学习算法中邻域尺寸是固定的,在故障诊断中并不恰当。本文中提出了一种基于新型流形学习算法稀疏流形聚类与嵌入(SMCE)的机械故障诊断方法来解决这个问题。SMCE通过求解稀疏优化问题自动确定邻域的大小,将传统流形学习中邻域尺寸选择变为优化问题的惩罚系数选择,进而从高维非线性观测数据中提取流形结构。利用SMCE从轴承和齿轮振动信号中提取特征进行诊断,实验表明,所提方法可以较好的提取故障信号内在的几何结构,应用无监督的谱聚类和有监督的支持向量机进行诊断准确率均高于98%。     

基于稀疏表示以及图谱理论的故障诊断方法

针对图傅里叶变换（graph Fourier transform,简称GFT）方法在提取轮对轴承故障特征信号的过程中,将信号中包含的部分噪声成分提取出来,从而对故障诊断结果产生影响这一问题,提出了一种基于稀疏表示以及图谱理论相结合的轮对轴承故障诊断方法。首先,根据具有局部损伤的滚动轴承振动信号特点构造合适的过完备字典库;其次,采用正交匹配追踪法求解系数实现对振动信号的稀疏表示;最后,通过图傅里叶变换方法将信号中含有的冲击分量集中到图谱域的高阶区域,从而对轮对轴承故障进行诊断。通过仿真数据以及试验数据处理结果,对提出方法的有效性进行了验证。     

基于信号共振稀疏分解和最大相关峭度解卷积的齿轮箱故障诊断

当齿轮箱内旋转零件发生故障时,其振动信号中的故障脉冲成分易被箱体中其他旋转部件的谐波信号和背景噪声所淹没,故障特征难以被有效提取。针对这一问题,提出了基于信号共振稀疏分解和最大相关峭度解卷积的故障诊断方法。该方法首先通过信号共振稀疏分解将信号中的低共振冲击成分从谐波分量和噪声中分离,然后对低共振分量进行最大相关峭度解卷积计算,进一步突出低共振分量中的周期脉冲成分,最后通过包络谱分析进行故障诊断。算法仿真、实验分析和工程应用结果表明,该方法能够有效提取强噪声信号中的周期性冲击成分,凸显故障特征,从而提供准确可靠的诊断结果。     

基于EMD和峭度的Hilbert包络解调在滚动轴承故障诊断中的应用分析

针对直接运用快速傅里叶变换（FFT）无法有效提取具有非线性非平稳特性的滚动轴承振动信号故障特征频率的问题,提出了一种基于经验模式分解和峭度指标的Hilbert包络解调方法.首先对滚动轴承的振动信号进行了经验模式分解（EMD）,得到了包含轴承故障特征信息的各阶本征模态函数（IMF）,再计算各阶IMF的峭度值,选取了峭度值较大的几阶IMF分量重构信号,并对重构信号进行了Hilbert包络解调分析,从而获得了滚动轴承的准确故障特征信息.分别对仿真模拟信号和实际滚动轴承发生内圈故障的振动信号进行了分析,清晰地得到了故障特征频率.研究结果表明,利用融合EMD、峭度系数和Hilbert包络解调的诊断方法能够快速、准确地提取滚动轴承的故障特征频率,从而可以对滚动轴承进行有效地故障诊断.     

FAULT DIAGNOSIS APPROACH FOR ROLLER BEARINGS BASED ON EMPIRICAL MODE DECOMPOSITION METHOD AND HILBERT TRANSFORM

Based upon empirical mode decomposition (EMD) method and Hilbert spectrum, a method for fault diagnosis of roller bearing is proposed. The orthogonal wavelet bases are used to translate vibration signals of a roller bearing into time-scale representation, then, an envelope signal can be obtained by envelope spectrum analysis of wavelet coefficients of high scales. By applying EMD method and Hilbert transform to the envelope signal, we can get the local Hilbert marginal spectrum from which the faults in a roller bearing can be diagnosed and fault patterns can be identified. Practical vibration signals measured from roller bearings with out-race faults or inner-race faults are analyzed by the proposed method. The results show that the proposed method is superior to the traditional envelope spectrum method in extracting the fault characteristics of roller bearings.     

基于信号共振稀疏分解与包络谱的齿轮故障诊断

当齿轮出现断齿、裂纹等局部故障时,其振动信号会出现周期性冲击脉冲。在齿轮故障早期,由于冲击脉冲微弱,常淹没在齿轮的啮合频率、转频等谐波成分以及噪声中,因此,对于齿轮早期故障,直接对齿轮振动信号做包络谱分析以诊断齿轮局部故障通常效果不佳。针对这一问题,将信号共振稀疏分解方法与包络谱分析相结合,提出了基于信号共振稀疏分解与包络谱的齿轮故障诊断方法。该方法采用信号共振稀疏分解将冲击脉冲从齿轮振动信号中分离出来,然后对冲击脉冲做Hilbert包络分析,获取冲击脉冲出现的周期,进而对齿轮状态和故障进行识别。仿真算例和应用实例证明了该方法的有效性。     

Hilbert-Huang变换在滚动轴承故障诊断中的应用

提出了一种新的滚动轴承故障诊断方法——基于小波系数包络信号的局部Hilbert边际谱方法，在Hilbert—Huang变换的基础上介绍了局部Hilbert谱和局部Hilbert边际谱，并将它应用于滚动轴承的故障诊断中。用小波基将滚动轴承故障振动信号分解，对高频段的小波系数用Hilbert进行包络分析得到包络信号，再对包络信号进行Hilbert—Huang变换求出局部Hilbert边际谱，从局部Hilbert边际谱中就可以判断滚动轴承的故障部位和类型。通过对滚动轴承具有外圈缺陷、内圈缺陷的情况下的振动信号的分析，说明该方法比传统的包络分析方法更能有效地提取滚动轴承故障特征。     

Weighted sparsity-based denoising for extracting incipient fault in rolling bearing

Given that the incipient fault is too weak for extraction, a novel approach that is based on sparse optimization is proposed for incipient fault diagnosis. The proposed optimization method consists of three steps: First, autocorrelation analysis is utilized to filter broadband random noise. Then, the weighted sparsity-based denoising method is proposed to extract periodic impulses. The prior knowledge that periodic impulses are sparse is used to constitute a penalty term; thus a novel weighted sparse optimization model is established. The majorization-minimization method is used to solve the optimization model. The high-pass filter in quadratic fidelity term is constructed by a Butterworth filter based on banded matrices, thus effectively improving computational efficiency. Lastly, the interval of periodic impulses, which corresponds to the fault frequency of rolling bearing, is obtained. Moreover, simulation and experimental results show that the proposed approach can successfully extract fault features from the signals of low signal to noise ratio.     

Quantitative diagnosis for bearing faults by improving ensemble empirical mode decomposition

In the bearing health assessment issues, using the adaptive nonstationary vibration signal processing methods in the time-frequency domain, lead to improving of early fault detection. On the other hand, the noise and random impulses which contaminates the input data, are a major challenge in extracting fault-related features. The main goal of this paper is to improve the Ensemble Empirical mode decomposition (EEMD) algorithm and combine it with a new proposed denoising process and the higher order spectra to increase the accuracy and speed of the fault severity and type detection. The main approach is to use statistical features without using any dimension reduction and data training. To eliminate unrelated components from faulty condition, the best combination of denoising parameters based on the wavelet transform, is determined by a proposed performance index. In order to enhance the efficiency of the EEMD algorithm, a systematic method is presented to determine the proper amplitude of the additive noise and the Intrinsic Mode Functions (IMFs) selection scheme. The fault occurrence detection and the fault severity level identification are performed by the Fault Severity Index (FSI) definition based on the energy level of the Combined Fault-Sensitive IMF (CFSIMF) envelope using the central limit theorem. Also, taking the advantages of a bispectrum analysis of CFSIMF envelope, fault type recognition can be achieved by Fault Type Index (FTI) quantification. Finally, the proposed method is validated using experimental data set from two different test rigs. Also, the role of the optimum denoising process and the algorithm of systematic selection of the EEMD parameters are described regardless of its type and estimating the consistent degradation pattern.     

基于RSGWPT-LCD的轴承信号故障特征提取及模式识别

为了有效提取滚动轴承振动信号的故障特征和提高分类识别精度,提出了一种基于冗余二代小波包变换-局部特征尺度分解(redundant second generation wavelet packet transform-local characteristic scale decomposition,简称RSGWPT-LCD)和极限学习机(extreme learning machine,简称ELM)相结合的故障特征提取和分类识别方法。首先,利用希尔伯特变换对原始振动信号进行处理,得到包络信号;其次,基于双层筛选机制,结合冗余二代小波包变换(redundant second generation wavelet packet transform,简称RSGWPT)和局部特征尺度分解(local characteristic-scale decomposition,简称LCD)方法对包络信号进行分解,筛选出包含主要信息的内禀尺度分量(intrinsic scale components,简称ISCs);然后,对提取的各ISCs分量构建初始特征矩阵并进行奇异值分解(singular value decomposition,简称SVD),将得到的奇异值作为表征各损伤信号的特征向量;最后,以提取的特征向量为输入样本,建立ELM模式分类器对滚动轴承损伤信号进行识别。信号仿真和实测数据表明,该方法可有效提取振动信号故障特征,提高分类识别精度,实现滚动轴承故障诊断。     

一种改进的基于小波变换的包络提取算法研究

结合“模极大值小波域去噪”原理，提出了一种基于小波变换和希尔伯特变换的包络提取算法。首先用小波变换作为包络分析的前置处理手段，再利用希尔伯特变换对处理后的信号进行包络提取。针对希尔伯特变换解包络的不足，采用模极大值小波域去噪算法对包络信号进行消噪，从而解决了一般算法难以解决的由于随机噪声的干扰造成的提取的包络轮廓信息粗糙的难题。实验结果表明，该方法可以精确提取信号的包络并通过包络细化谱分析全面获得信号所隐古的故障特征。     

Fault diagnosis of rolling element bearing using a new optimal scale morphology analysis method

Periodic transient impulses are key indicators of rolling element bearing defects. Efficient acquisition of impact impulses concerned with the defects is of much concern to the precise detection of bearing defects. However, transient features of rolling element bearing are generally immersed in stochastic noise and harmonic interference. Therefore, in this paper, a new optimal scale morphology analysis method, named adaptive multiscale combination morphological filter-hat transform (AMCMFH), is proposed for rolling element bearing fault diagnosis, which can both reduce stochastic noise and reserve signal details. In this method, firstly, an adaptive selection strategy based on the feature energy factor (FEF) is introduced to determine the optimal structuring element (SE) scale of multiscale combination morphological filter-hat transform (MCMFH). Subsequently, MCMFH containing the optimal SE scale is applied to obtain the impulse components from the bearing vibration signal. Finally, fault types of bearing are confirmed by extracting the defective frequency from envelope spectrum of the impulse components. The validity of the proposed method is verified through the simulated analysis and bearing vibration data derived from the laboratory bench. Results indicate that the proposed method has a good capability to recognize localized faults appeared on rolling element bearing from vibration signal. The study supplies a novel technique for the detection of faulty bearing.     

基于形态分量分析和包络谱的轴承故障诊断

滚动轴承出现局部损伤时,其振动信号往往由包含轴承自身振动的谐振分量、包含轴承故障信息的冲击分量及随机噪声分量构成。提出了基于形态分量分析和包络谱的滚动轴承故障诊断方法。该方法根据轴承振动信号中各组成成分的形态差异,利用改进的形态分量分析对滚动轴承故障振动信号中的谐振分量、冲击分量和噪声分量进行分离,然后对冲击分量进行Hilbert包络解调分析,根据包络谱诊断滚动轴承故障。算法仿真和应用实例表明,该方法能有效提取滚动轴承故障特征。