基于谱相干滤波冲击增强的轴承故障诊断

针对优化特征改进包络谱(IESFO)存在早期故障弱特征提取能力较弱和对频率分辨率要求较高的不足,提出了基于谱相干滤波冲击增强的轴承故障特征提取方法。首先使用IESFO算法选取优化解调频带并进行带通滤波;然后对滤波后信号使用多点优化最小熵反褶积(MOMEDA)算法增强信号中轴承故障产生的冲击;最后进行包络分析。基于实测信号的研究结果表明,和现有方法相比,本方法在轴承性能退化过程中可以提前提取到轴承早期故障信息,且可用于提取滚动轴承复合故障信息。     

自适应VMD联合MOMEDA的滚动轴承故障提取

针对强背景噪声下滚动轴承微弱故障特征提取问题,提出了一种基于参数自适应优化变分模态分解(VMD)与多点最优最小熵解卷积(MOMEDA)相结合的轴承故障特征提取方法。首先对滚动轴承时域振动信号进行VMD分解,然后基于自相关函数脉冲谐波噪声比指标(AIHN)最大化原则进行挑选得到最佳模态分量(BIMF)并对其进行MOMEDA滤波,包络解调后得到故障特征频率,最后将本文所提方法体应用于数值仿真信号上可以明显观察到故障特征频率131.1Hz,应用于实际轴承故障信号可以有效识别轴承故障特征频率294.5Hz,与原始包络谱提取的311Hz以及MCKD提取的320Hz相比更加接近理论故障特征频率294Hz。     

电动机滚动轴承的故障诊断

本文根据滚动轴承发生损伤故障时振动信号的特点，利用带通数字滤波和希尔伯特变换，对电动机轴承振动信号进行处理，然后对包络信号作谱分析，再从包络谱中提取故障特征频率分量，以诊断电动机轴承故障。实验结果表明，这种诊断方法是很有效的。     

自适应陷波理论及其在轴承故障诊断中的应用

滚动轴承故障冲击特征易被工频载波信号淹没,而传统的信号降噪方法对工频干扰不具有针对性,所以将工频陷波理论引入到轴承故障诊断中。由于陷波的窄带滤波特性,其对中心频率及带宽参数变化较为敏感,通过粒子群多参数寻优,以时域峭度最大原则对陷波器中心频率及带宽进行自适应选取,以时域波形匹配方差作为评价指标验证陷波对故障冲击特性的还原能力。试验分析表明自适应陷波可以有效地从工频调制信号中解调出故障冲击特征,对陷波后信号进行包络谱分析,其故障特征谱线得到增强,辅助以集合经验模态分解(EEMD)、变分模态分解(VMD)去噪方法,可以得到更理想的效果。     

基于FastICA-BAS-MCKD的滚动轴承复合故障特征提取方法

针对强背景噪声下轴承复合故障特征难以分离提取的问题,提出了一种基于快速独立成分分析-天牛须-最大相关峭度 解卷积算法(FastICA-BAS-MCKD)的滚动轴承复合故障特征提取方法。 首先,引入 FastICA 对滚动轴承多通道故障信号进行盲 源分离;其次,利用 BAS 算法同步优化 MCKD 算法的解卷积周期 T、滤波器长度 L 和移位数 M,构建基于 BAS-MCKD 的滚动轴 承振动信号自适应分析方法;然后,应用 BAS-MCKD 方法处理分离后的信号,实现分离信号的降噪和特征增强;最后,应用希尔 伯特解调方法对 MCKD 处理后的信号进行包络谱分析,实现滚动轴承不同类型故障的识别。 仿真和实测信号的分析结果表 明,所提方法能清晰地从复合故障信号中提取出单一故障特征频率,为滚动轴承复合故障特征提取提供了一种有效的解决 方案。     

基于EEMD的声阵列滚动轴承故障诊断

研究针对滚动轴承故障诊断中的类型和位置分析问题,提出了一种基于集合经验模态分解(EEMD)的声阵列滚动轴承故障诊断分析方法。以EEMD分解信号的峭度和能量作为评价指标,提取包含故障信息的IMF分解信号,根据滚动轴承理论故障频率及其倍频分析对分解信号进行窄带滤波后通过Hilbert包络谱实现故障类型判断,通过对窄带滤波后的分解信号使用声阵列技术进行声像分析实现滚动轴承故障定位分析。最后通过试验进行了方法验证,结果表明过使用基于EEMD分解的阵列分析方法,可更为直观确定故障位置和故障类型,有利于有轨机车等多轴承驱动系统轴承故障的快速和实时诊断,对于确定检修、制定合理维修决策、改进维修质量具有十分重要指导意义。     

Transformer partial discharge signal reconstruction and feature extraction method based on time–frequency matrix

Effectively extracting power transformer partial discharge (PD) signals is of great significance for monitoring the power transformer insulation state. However, practical and effective extraction methods have been lacking. This paper proposes a novel method for the extraction PD signals feature by the analysis of their time–frequency matrix. First, empirical mode decomposition (EMD) is carried out for raw signals to obtain the complete Hilbert time–frequency spectrum. Second, the frequency band partition is carried out. And the component of raw signals on each frequency band is constructed after filtering by a Hilbert–Huang transform (HHT) band‐pass filter. Then the time–frequency matrix is constructed by different frequency band components, and singular value decomposition is carried out. Using its singular value energy spectrum, the PD signal is reconstructed, further realizing feature extraction. Finally, the analysis result from actual examples indicates that this method can effectively extract PD signal feature frequency spectrum, and meanwhile also can eliminate the strong background interference and retain the higher time–frequency resolution. © 2017 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.     

基于VMD与快速谱峭度的滚动轴承故障诊断

针对滚动轴承故障信号易受环境噪声干扰,故障特征信息获取相对困难的问题,提出了基于变分模态分解(VMD)与快速谱峭度的滚动轴承故障特征提取方法.首先将轴承信号分解为若干个固有模态分量(IMF),然后利用最大相关峭度解卷积算法对各阶模态分量进行计算,选取相关峭度值相对较大的几个IMF分量作为故障信息最突出的研究对象,并对其...     

Generation of band‐pass constant‐envelope signals with a pre‐assigned spectrum: a synthesis procedure

The recent, increasing adoption of spread‐spectrum techniques, either in communications or in other fields, makes the design of band‐pass sources a topical subject. We consider the generation of constant‐envelope (i.e. constant power) band‐pass signals via a frequency modulation technique. In particular, we propose a general solution to the synthesis problem of delivering a constant envelope signal with a pre‐assigned band‐pass spectrum. Both quasi‐stationary and non‐stationary modulations are discussed and a novel optimization algorithm is proposed for the latter case. A design example concludes the presentation. Copyright © 2002 John Wiley & Sons, Ltd.     

基于形态优化滤波的轴承故障特征提取方法

针对滚动轴承故障特征难以提取等问题,提出利用判别指标最大准则选择最优形态滤波算子进行轴承故障特征提取新方法,以达到在噪声干扰下更优的故障冲击信号提取效果。首先利用6种形态滤波器以不同尺度结构元素对轴承故障信号进行降噪处理;其次计算滤波信号的判别指标,以判别指标最大原则获取最佳形态滤波算子;然后利用最佳形态滤波算子处理滚动轴承实例故障信号;最后借助特征频率强度系数、峭度和偏斜度评价滤波质量,将该方法与传统方法进行比较。测试结果表明,该方法能够更好地提取轴承故障特征信息,有效抑制噪声实现轴承故障精确诊断。     

考虑频率相关参数的直流输电线路故障特征分析

目前高压直流输电线路利用行波波头的上升速率构成保护判据,由于长距离线路频率相关特性对高频信号的衰减作用较强,导致保护的灵敏性不足。利用零极点法分析线路故障暂态信号的频率特性,发现故障电气量的高频分量中包含位置主频和边界主频。高频暂态量由边界主频和位置主频共同作用产生。区内、外故障的特征差异取决于反射系数频率特性的通带和透射系数频率特性的阻带,因此特征频带应为透射系数阻带的截止频率与反射系数通带的最高频率间的频带。由PSCAD仿真结果证明,区内、外故障时电流在此特征频带内差异明显。     

Analytical synthesis of voltage‐mode even/odd‐nth‐order differential difference current conveyor and fully differential current conveyor II‐grounded resistor and capacitor universal filter structures

A complete definition of an odd/even‐nth‐order notch or band‐reject filter transfer function is presented. Based on the differences between the input voltage and (i) an nth‐order high‐pass; (ii) a traditional nth‐order notch; and (iii) an nth‐order all‐pass filtering transfer function, a systematic method has been proposed to derive a universal filter structure that can realize voltage‐mode odd/even‐nth‐order low‐pass, band‐pass, high‐pass, all‐pass and traditional notch filters. The intrinsic capability of voltage‐mode addition and subtraction of the two active elements, differential difference current conveyors and fully differential current conveyors, is used to advantage in the aforementioned synthesis procedure. Based upon the definition of an nth‐order notch or band‐reject filter transfer function proposed in this paper, the aforementioned universal one has been further extended to the newly defined nth‐order band rejection filter. The voltage and current tracking errors of the two active elements are compensated by varying the resistances of the proposed filter. Filtering feasibility, stability, component sensitivities, linear and dynamic ranges, power consumption, and noise are simulated using H‐Spice with 0.35 µm process. Compared to some of the recently reported universal biquads, the new one is shown to enjoy the lowest component sensitivities and the best output accuracy for all‐pass signals. Moreover, Monte Carlo and two‐tone tests for intermodulation linearity simulations are also investigated. Copyright © 2014 John Wiley & Sons, Ltd.     

基于COT-SSD的变转速滚动轴承微弱故障诊断

针对背景噪声干扰及转速波动工况下滚动轴承微弱故障识别困难这一问题,提出一种结合计算阶次追踪(COT)和奇异谱分解(SSD)的新型诊断方法。利用COT算法对采集的原始时域信号进行等角度重采样,继而利用SSD算法对重采样角域信号进行处理,通过自适应构建的轨迹矩阵的奇异值分解重组,将角域信号从高频至低频分解为若干个奇异谱分量,利用融合峭度指标筛选最佳奇异谱分量,选定最佳分量后对其进行进一步的包络解调运算,最终通过分析包络阶次谱中幅值突出的成分来准确判定滚动轴承运行状态。滚动轴承内外圈故障实测信号分析结果表明,所提方法能够有效提取出变速工况下滚动轴承的微弱故障特征信息。     

配电线路暂态保护故障行波特征的研究

配电线路区内、外故障时暂态分量频谱能量差异不明显,使得暂态保护原理难以应用于配电线路单端保护。通过行波网格法分析了配电线路区内、外故障行波的传播规律及特点,提出在线路边界处并联接入适当通频带的带通滤波装置,以此强化区内、外故障时故障暂态电压频谱能量差异。PSCAD/EMTDC仿真和模拟试验结果表明：安装了带通滤波装置的配电线路区外故障时,通频带内的故障行波流经带通滤波装置后发生较大衰减;区内故障时故障行波未通过带通滤波装置而几乎没有衰减,区内、外故障时暂态电压分量频谱能量差异非常明显。利用这种差异可以准确区分区内、外故障,这为暂态保护原理在配电线路单端保护中的应用提供了有效的理论依据和解决方案。     

基于VMD与自相关分析的滚动轴承故障特征提取

滚动轴承故障信号多呈现非平稳、多分量调制特性,早期故障信号调制特性微弱、易受周围设备噪声干扰,导致轴承早期故障特征淹没在噪声信号中,故障特征难以提取。为此,提出一种变分模态分解(variational mode decomposition,VMD)与自相关分析相结合的轴承故障特征提取方法。首先利用自相关分析消除故障信号中噪声干扰,提取周期成分;然后再用VMD算法将消噪信号分解成若干本征模态分量(intrinsic mode function,IMF),运用能量算子对相关系数及峭度值较大分量进行解调分析;最后通过能量解调谱来判别滚动轴承故障类型。将该方法应用到滚动轴承仿真故障数据和实测数据中,结果表明,该方法可降低了噪声的干扰,有效提取故障特征频率,能够实现滚动轴承故障的精确诊断。     

基于最优Morlet小波自适应包络解调的弱故障特征提取方法

为自适应实现Morlet小波与故障冲击特征成分的最优匹配,采用基于Shannon小波熵的方法优化带宽参数设计最优Morlet小波。针对最佳尺度求取的难题,利用谱峭度与小波熵均能敏感反映冲击性的特性,提出了基于峭熵比求取最佳尺度。基于此,提出基于最优Morlet小波自适应包络解调的弱故障特征提取方法,该方法首先对信号进行最优Morlet连续小波变换;然后,依据峭熵比自适应地求取最佳尺度并提取最佳尺度的小波系数;最后,对最佳尺度的小波系数取模即可实现对最优频带的包络解调,得到包络谱,从而实现微弱故障特征的提取。实例分析表明:该方法克服了传统包络解调需要人为设定带通滤波器参数的不足,能有效地从强噪背景中提取微弱故障特征。     

小波包分析在滚动轴承故障诊断中的应用

结合小波包分析和能量谱分析方法 ,提出一种新的滚动轴承故障诊断方法。利用小波包对滚动轴承振动加速度信号进行分解 ,求出各频率段的能量 ,并以此作为滚动轴承发生故障的特征量对振动信号进行重构 ,比较正常信号和故障信号的差异 ,从而识别滚动轴承的故障。通过对于实测信号的分析证明该方法具有特征参量少、故障特征突出等优点 ,可有效地用于滚动轴承的故障诊断。     

ITT变换在风电机组滚动轴承故障诊断中的应用

为实现风电机组滚动轴承微弱故障诊断,提出了基于改进的时时(ITT)变换的风电机组滚动轴承故障诊断方法。由时时(TT)变换可得到一维轴承故障振动信号的TT变换矩阵,实现滚动轴承振动信号的二维TT表示。提取该TT变换矩阵的对角线元素可滤除低频干扰信号,起到增强故障特征的效果。鉴于噪声对TT变换分析效果具有重要影响,提出基于能量熵准则的奇异值分解降噪方法改进TT变换,以提高TT变换的抗噪能力,实现强背景噪声条件下轴承微弱故障特征提取。仿真、实验及工程应用实例结果均表明所提方法可以有效诊断出风电机组滚动轴承的故障类型。     

A reconfigurable high ultimate rejection inductorless band‐pass filter by the use of N‐path passive mixers

In this paper, a band‐pass filter with a tunable bandwidth and the center frequency is introduced, which employs N‐path and N × M‐path passive mixer structures, for multiband multistandard wireless receivers. The center frequency of the proposed filter is tunable from 0.1 to 1 GHz, while its bandwidth is also adjustable from 6% to 34% of the center frequency at 100 MHz. The passband ripple is reduced by applying a Miller compensation technique, resulting in a worst‐case ripple of only 1.6 dB over the entire tuning range. An additional eight‐path filter is also utilized at the input of the circuit, which highly improves the out‐of‐band rejection of the filter as well as its out‐of‐band linearity. The noise figure and the input return loss are, respectively, better than 5 and 10 dB, and depending on the desired center frequency, the total power consumption of the proposed filter varies from 41 to 70 mW.     

多尺度形态滤波模态混叠抑制方法

针对Hilbert-Huang变换(HHT)中存在的模态混叠现象,依据数学形态学理论,提出多尺度平均组合形态滤波方法,并构建了多尺度平均组合形态滤波器对原始振动信号进行降噪预处理,以实现对模态混叠的抑制。并以滚动轴承的振动信号为原始数据进行故障特征频率提取实验,将所提方法与集合经验模态分解(EEMD)方法对模态混叠的抑制效果进行对比。结果表明,所提的多尺度平均组合形态滤波方法耗时仅为EEMD的1/10,且特征频率提取的误差率比EEMD低0.16%。最后,将多尺度平均组合形态滤波与HHT相结合进行滚动轴承故障特征提取的现场试验,特征频率提取结果与理论值的误差率为0.26%。