基于VMD与IFWA-SVM的滚动轴承故障诊断研究北大核心

为有效提取非平稳性、复杂性的滚动轴承振动信号特征,提出一种基于变分模态分解、改进烟花算法(IFWA)优化支持向量机(SVM)的滚动轴承故障诊断方法。利用VMD对原始信号进行分解,计算得到各IMF的样本熵,将原始信号的时域特征与其结合组成特征矩阵。为提高故障诊断效率,采用IFWA优化SVM,建立IFWA-SVM模型。使用训练集特征矩阵训练诊断模型,实现滚动轴承的故障诊断。利用实测信号验证该方法,并与粒子群算法优化进行比较。结果表明:利用该方法进行诊断,正确率提高了3.33%、训练时间缩短了21.55 s,验证了该方法的可行性。     

基于改进VMD形态谱和FCM的滚动轴承故障诊断方法

针对滚动轴承故障信号的非线性特性及不同故障类型信号具有不同形态特征的特点,提出一种基于改进变分模态分解（VMD）形态谱和模糊C均值聚类（FCM）算法相结合的故障诊断方法。采用VMD方法对滚动轴承振动信号进行分解,针对分解过程中关键参数的选取,提出相关参数选择方法,并计算各固有模态函数(IMF)的能量波动系数,以获得对信号特征信息敏感的模态分量进行重构。计算重构信号的形态谱以反映信号的形态特征。通过FCM算法实现滚动轴承工作状态和故障类型的诊断。运用该方法对实测滚动轴承振动信号进行分析,并将所提方法同基于原始振动信号、经验模态分解、总体经验模态分解形态谱的故障特征提取方法进行对比。结果表明：所提方法能够更加有效提取滚动轴承信号的故障特征,实现故障类型的准确诊断。     

基于VMD-SPWVD-CNN的滚动轴承故障智能诊断

针对传统的故障诊断方法依赖经验进行人工提取特征的问题,基于变分模态分解(VMD)和平滑伪魏格纳分布(SPWVD)并结合卷积神经网络(CNN)提出了VMD-SPWVD-CNN模型用于滚动轴承故障诊断.首先,利用VMD对轴承振动信号进行处理,分解为多个具有不同中心频率的模态分量;其次,对信号的每个模态分量分别进行SPWVD...     

基于变分非线性调频模态分解的滚动轴承多故障诊断方法研究

基于变分非线性调频模态分解,提出一种滚动轴承多故障诊断方法。对滚动轴承多故障振动信号分量的瞬时频率和瞬时幅值进行估计;在此基础上,通过最小化信号的带宽实现滚动轴承故障信号的重构。对比所提方法与变分模态分解法,通过实验案例对所提方法进行验证。结果表明：所提方法明显优于变分模态分解方法,即使在强噪声背景下,仍能有效地实现滚动轴承多故障信号的分解与重构,可以有效地诊断滚动轴承多故障。     

基于VMD和Teager能量增强谱的滚动轴承故障诊断方法

针对强噪声干扰下滚动轴承故障特征难以提取的问题,提出一种变分模态分解和Teager能量增强谱的滚动轴承故障诊断方法。该方法首先通过变分模态分解(Variational Mode Decomposition,VMD)将非平稳的轴承故障振动信号分解成一系列平稳的窄带分量;然后根据峭度-相关性最大准则挑选包含故障特征信息最丰富的窄带分量作为主分量;最后对选取的主分量进行Teager能量增强谱,提取滚动轴承的故障特征。通过仿真和实例分析的结果表明:该方法能有效地提取出滚动轴承早期故障特征,且能够抑制强烈的噪声干扰和增强故障冲击特征,优于传统包络谱分析和基于经验模态分解(Empirical Mode Decomposition,EMD)和Teager能量谱的方法的分析结果。     

基于多域空间状态矩阵奇异值与局部保持投影的滚动轴承故障特征提取方法

针对滚动轴承振动信号复杂且难以从中提取有效故障特征的问题,提出了一种总体经验模态分解(EEMD)、奇异值分解(SVD)和局部保持投影(LPP)相结合的故障特征提取方法。首先,对振动信号进行EEMD分解,利用EEMD分解后的固有模态分量(IMF)分别构造时域、频域和时频域空间状态矩阵;其次,利用SVD提炼时域、频域和时频域空间状态矩阵中的故障信息,筛选其中累加百分比大于90%的奇异值组成多域有效奇异值数组,构造多域奇异值特征矩阵;然后,利用LPP约简多域奇异值特征矩阵,提取低维、高区分度的故障特征;最后,利用支持向量机(SVM)对提出的故障特征提取方法进行评估。实验结果证明了该方法提取的故障特征可有效反映滚动轴承的故障状态。     

基于改进EEMD的滚动轴承故障特征提取技术

在对集合经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)进行研究之后,提出了一种利用改进的EEMD进行滚动轴承故障特征提取的方法。该方法根据EEMD的分解过程中信号和加入的白噪声的特点来选择EEMD的参数,并且对分解后所得到的的固有模态函数(IMF)分量进行阈值处理后再重构,以降低噪声的干扰。对重构后的信号进行包络谱分析,提取其故障特征,最后将该方法与通用的EEMD方法进行对比,研究结果表明EEMD是一种很有效的滚动轴承故障特征提取方法。     

基于VMD与分层极限学习机的滚动轴承故障诊断方法研究

针对滚动轴承信号的不规则特性,致使信号故障特征难提取及难以辨识的问题,为实现滚动轴承故障的智能诊断,提出基于VMD排列熵与分层极限学习机的滚动轴承故障诊断方法。首先将测得振动信号进行变分模态分解(Variational Mode Decomposition,VMD),利用排列熵进一步提取各模态特征组成高维特征向量集;其次利用自动编码器(Automatic Encoder,AE)对极限学习机的隐含层进行分层,且使隐含层节点的输入权值和阈值满足正交条件;最后将构建的特征向量作为H-ELM算法的输入,通过训练建立H-ELM滚动轴承故障分类模型。实验结果表明:H-ELM滚动轴承故障分类模型比SVM、ELM故障分类模型具有更高的精度、更强的稳定性。     

基于参数优化变分模态分解的滚动轴承微弱故障诊断研究

为了提高变分模态分解(VMD)对滚动轴承微弱故障特征提取的准确性,提出了一种基于参数优化VMD与奇异值分量及其熵相结合的滚动轴承故障诊断方法。该方法通过寻优算法确定VMD的模态数K和二次惩罚因子α;根据余弦-标准差指标提取VMD典型本征模态分量(IMF);计算IMF奇异值及其熵,并利用计算结果分别判断滚动轴承的不同故障状态。结合美国西储大学轴承振动信号数据,实验结果表明:相比经验模态分解奇异值故障诊断方法,基于参数优化VMD奇异值故障诊断方法能更明显地识别滚动轴承的不同故障类型,为区分滚动轴承微弱故障提供了一种可行的诊断思路。     

基于VMD与时间序列分析的滚动轴承故障特征提取方法

为了在非线性、非平稳的滚动轴承故障振动信号中有效提取出敏感的故障特征,提出了基于变分模态分解(VMD)与时间序列分析相结合的特征提取方法。首先通过VMD将原始信号分解为不同预设尺度的本征模态分量(IMF),对各个IMF分量建立时间序列预测模型,通过叠加重构得到最终的预测模型,比较评价指标确定最优参数的选取。最后,通过仿真信号与滚动轴承实际故障数据分析,并与经验模式分解(EMD)进行对比,结果表明该方法能够有效的提取到故障特征频率。     

Influence of metal ions on the formation of artificial zinc rusts

The artificial rust particles were prepared from ZnCl₂ solutions dissolving Al(III), Fe(III), Fe(II), Ni(II), Co(II) and Mg(II) at different atomic ratios from 0 to 0.3 in metal/Zn. With increasing metal/Zn the crystal phases of the products turned following as ZnO → a mixture of ZnO and Zn₅(OH)₈Cl₂ · H₂O (ZHC) → ZHC. Al(III) most facilitated the formation of ZHC but Mg(II) and Fe(III) produced no ZHC. The morphology of the formed particles varied following as agglomerate → fine → rod → sheet → irregular with the increase of metal/Zn. The sheet and irregularly shaped particles were identified as ZHC and the other particles as ZnO.     

Effect of dislocation density on the low temperature aging behavior of an ultra low carbon bake hardening steel

Strain aging was studied in an ultra low carbon (ULC) steel with a total carbon content of 20 ppm (wt.%) in order to identify the process stages and mechanism of bake hardening in this type of steel. The effects of dislocation density, varied by means of uniaxial tensile prestraining (1–10%) on the aging kinetics were investigated within an aging temperature range of 50–170°C. The aging was evaluated by means of strength measurements and the determination of interstitial carbon content after aging using a piezoelectric composite oscillator operating at 40 kHz. The interaction between interstitial carbon and dislocations was examined through amplitude dependent internal friction measurements. The influence of dislocation density on the aging behavior have been discussed with reference to the kinetics and mechanism of the aging process.     

A corrosion study of the main constituent phases of AZ91 magnesium alloys

The different constituents of an AZ91 alloy (α,β, and MnAl phases) were synthesized and their corrosion resistance was studied by electrochemistry in ASTM D1384 water, pH 8.3. The pure phases were characterised through the corrosion potential, the polarisation resistance, and polarisation curves, then systematically coupled to assess the galvanic corrosion occurring in the AZ91 alloy. The aluminium content of the oxide film was obtained by X-ray photoelectron spectroscopy measurements. The corrosion rate of the α solid solution alloys depends closely on their Al content. Aluminium enhances the corrosion resistance of the α-phase through the formation of an Al enriched superficial layer. The β-phase is 150 mV nobler than the α-phase, but their corrosion rates are similar. The galvanic currents are low (below 20 μA cm⁻²) whatever the implemented couples and close to the corrosion current previously measured for the AZ91 alloys.     

Effect of solution heat treatment on galvanic coupling between intermetallics and matrix in AA7075-T6

Susceptibility to localised corrosion is strongly affected by heat treatments performed on Al-Zn-Cu-Mg alloys. In order to study how galvanic coupling between intermetallics and matrix is affected by solution heat treatment, AA7075-T6 and solution heat treated AA7075 have been characterised by means of scanning electron microscopy and scanning Kelvin probe force microscopy. Solution heat treatment strongly increased the Volta potential difference between the intermetallics and the surrounding matrix showing a strong increase in galvanic coupling. This is explained by Zn and Mg enrichment of the matrix caused by dissolution of strengthening particles during solution heat treatment.     

Characterizing corrosion behavior under atmospheric conditions using electrochemical techniques

AC and DC electrochemical experiments were performed as a function of humidity and contaminant concentration in an effort to identify the range of atmospheric environments where corrosion processes could be detected and possibly quantified. AC measurements exhibited two time constants at 25% relative humidity (RH), possibly indicating the ability to resolve both electrolyte resistance and interfacial impedance. Galvanic current measurements were sensitive to the presence of Cl_2(g) at 30% RH and electrochemical transients were detected at both 30% and 50% RH levels, also indicating sensitivity to interfacial processes. Higher humidity levels allowed better quantification due to decreasing electrolyte and interfacial impedances.     

硬质合金Hc、Com、D、HV_3的初步探讨

通过研究硬质合金的矫顽磁力、钴磁、密度和硬度,阐述了这些性能之间的相互关系,以及影响这些性能的主要原因,说明了其对生产实践中质量控制的重要意义。     

Electrochemical behaviour of amorphous and nanoquasicrystalline Zr-Pd and Zr-Pt alloys in different environments

The corrosion behaviour of melt spun amorphous and nanoquasicrystalline Zr₇₀Pd₃₀ and Zr₈₀Pt₂₀ alloy ribbons has been investigated using potentiodynamic polarization study in NaCl, H₂SO₄ and NaOH solutions at different concentrations. The amorphous and nanoquasicrystalline alloys show better corrosion resistance than Zr in all the solutions studied. Both the alloys are susceptible to chloride attack and pitting has been observed. Complete passivation has been observed in H₂SO₄, while gradual break down of passivating layer occurs in NaOH. In general, nanoquasicrystalline state in both the alloys shows better corrosion resistance than amorphous state in all the solutions studied.     

A study on the deactivation of an IrO2–Ta2O5 coated titanium anode

The deactivation of an IrO₂–Ta₂O₅ coated titanium anode was studied during an accelerated life test at 2 A cm⁻² in 1 mol dm⁻³ H₂SO₄ solution using CV, EIS, SEM and EDX. The changes of voltammetric charge, double layer capacitance, oxide film resistance and charge transfer resistance of oxygen evolution with time during the electrolysis were monitored. The morphology and surface composition of the oxide anode before and after electrolysis test were analysed. A comprehensive process of deactivation of the oxide anode was proposed based on the test results and analysis.     

加速发展我国的快速成形技术

介绍了第二届北京国际快速成形及制造会议概况，分析了快速成形（RP）技术的发展趋势，指出RP技术研究领域的重大进展及学科发展的主攻方向是：由RP向快速制造（RM）发展，由RP向快速模具（RT）发展，RP技术在生物医学领域的应用。     

Corrosion behaviour of powder metallurgical and cast Al-Zn-Mg base alloys

The behaviour of Al-Zn-Mg base alloys produced by powder metallurgy and casting has been studied using potentiodynamic polarisation in 0.3% and 3% NaCl solutions. The influence of alloy production route on microstructure has been examined by scanning electron microscopy, Auger electron spectroscopy and secondary ion mass spectrometry. An improvement in performance of powder metallurgy (PM) materials, compared with the cast alloy, was evident in solutions of low chloride concentration; less striking differences were revealed in high chloride concentration. Both powder metallurgy and cast alloys show two main types of precipitates, which were identified as Zn-Mg and Zr-Sc base intermetallic phases. The microstructure of the PM alloys is refined compared with the cast material, which assists understanding of the corrosion performance. The corrosion process commences with dissolution of the Zn-Mg base phases, with the relatively coarse phases present in the cast alloy showing ready development of corrosion.