基于共振解调技术的滚动轴承故障自动诊断系统

介绍了基于共振解调技术的滚动轴承故障诊断原理、特点及实现方法 ,开发了可在线自动检测和诊断滚动轴承故障的软件系统     

基于FPGA的滚动轴承故障在线检测系统设计北大核心CSCD

针对传统滚动轴承故障诊断方案不能实时在线诊断的问题,设计了基于FPGA的滚动轴承故障在线检测系统。在硬件方面,设计了信号调理电路,实现了A/D转换,完成了滚动轴承信号在线采集,设计了上位机界面,能够实时显示检测结果,实现了结果的可视化;在算法方面,利用共振解调技术,完成了滚动轴承故障频率的提取。最后在QPZZ-Ⅱ实验台对系统进行测试,并与软件检测方法对比检验系统实时性能。结果表明:系统能有效提取滚动轴承故障特征频率,解决了传统检测方案效率低的问题,满足滚动轴承故障实时在线检测的需要。     

基于共振解调的滚动轴承故障诊断的研究与实现

滚动轴承故障诊断是机械故障检测中的一个重要方面,通过对滚动轴承典型故障机理及实际振动特征的分析,发现共振解调是一种有效分析滚动轴承故障信号的方法,而且具有一定的现实意义。     

浅析滚动轴承故障自适应谐振解调技术

本文提出一种利用共振进行共振解调的方法。运用经验模式分析方法,将故障划分成多个固有模式,以最小的特征值为主要特征,以最佳特征值为最大共振频率,并利用自适应选取不同的频段。利用希尔伯特变换,对滤波后的信号进行了理解、调和分析,得到包含故障特征的低频包络,并进行了辨识。与现有的谐波解调方法相比,该方法更具实际应用价值。     

实现滚动轴承损伤类故障自动诊断的一种简便方法

提出了一种用共振解调技术实现滚动轴承损伤类故障自动诊断的简便方法。     

滚动轴承的自动监测及故障诊断系统

介绍了基于PC的滚动轴承监测以及故障诊断系统的原理。在系统监测方面使用轴承振动信号的时域参数判断法进行质量的检测；在故障诊断理论方面论述了共振解调以及小波包络谱分析等诊断方法，并用试验中采集到的轴承振动信号进行了系统的仿真，说明两种方法在轴承故障诊断中的有效性，同时还介绍了系统软件及硬件的架构方案。     

基于平方包络谱相关峭度的最优共振解调诊断滚动轴承故障

利用峭度指标识别滚动轴承共振频带,结合包络分析解调故障特征,是滚动轴承故障诊断的常用方法。峭度指标虽然能够表征瞬态冲击特征的强弱,却无法利用瞬态冲击特征循环发生的特点,导致其难以区分脉冲噪声和循环瞬态冲击,无法准确识别共振频带,进而容易导致错误的故障诊断结果。受峭度和信号自相关的启发,重新定义相关峭度,提出平方包络谱相关峭度新指标;并结合Morlet小波滤波和粒子群优化算法,提出一种滚动轴承最优共振解调方法。通过与峭度、谱峭度等进行对比,仿真和试验分析结果表明平方包络谱相关峭度能够准确识别循环瞬态冲击;最优共振解调能够稳健确定共振频带的最优中心频率和带宽,准确解调诊断滚动轴承故障,验证了平方包络谱相关峭度在检测循环瞬态冲击和识别最优共振频带中的有效性和优越性。     

滚动轴承表面损伤故障智能诊断新方法

本文针对目前基于小波变换的滚动轴承故障诊断研究中普遍存在小波变换参数选取和故障特征计算无法自动完成的问题,提出了一种基于小波包变换的滚动轴承故障特征自动提取技术,实现了小波函数参数的自动选取和故障特征的自动提取.最后,基于结构自适应神经网络方法建立了滚动轴承的集成神经网络智能诊断模型,利用实际的滚动轴承实验数据进行了验证,结果表明了本文方法的有效性.     

滚动轴承故障诊断及经验总结

介绍了滚动轴承故障的基本形式、国内新近的监测方法及诊断原理,并针对滚动轴承故障进行了经验总结。     

Application of an improved kurtogram method for fault diagnosis of rolling element bearings

Kurtogram, due to the superiority of detecting and characterizing transients in a signal, has been proved to be a very powerful and practical tool in machinery fault diagnosis. Kurtogram, based on the short time Fourier transform (STFT) or FIR filters, however, limits the accuracy improvement of kurtogram in extracting transient characteristics from a noisy signal and identifying machinery fault. Therefore, more precise filters need to be developed and incorporated into the kurtogram method to overcome its shortcomings and to further enhance its accuracy in discovering characteristics and detecting faults. The filter based on wavelet packet transform (WPT) can filter out noise and precisely match the fault characteristics of noisy signals. By introducing WPT into kurtogram, this paper proposes an improved kurtogram method adopting WPT as the filter of kurtogram to overcome the shortcomings of the original kurtogram. The vibration signals collected from rolling element bearings are used to demonstrate the improved performance of the proposed method compared with the original kurtogram. The results verify the effectiveness of the method in extracting fault characteristics and diagnosing faults of rolling element bearings.     

滚动轴承故障诊断的神经网络方法

总结分析了轴承的故障形式及原因,给出了振动频率,阐述了Bp网络的结构及算法,并对实例建立BP神经网络。     

滚动转子式压缩机故障的在线检测

探讨了滚动转子式压缩机故障的在线检测技术,通过对其工作过程的分析,确定了用壳体振动作为故障分析信号,并结合小波包和神经网络方法将正常与异常压缩机区分开来,现场的检测结果表明本文方法准确可靠,具有较高的检测效率。     

Active magnetic bearings used as exciters for rolling element bearing outer race defect diagnosis

The active health monitoring of rotordynamic systems in the presence of bearing outer race defect is considered in this paper. The shaft is assumed to be supported by conventional mechanical bearings and an active magnetic bearing (AMB) is used in the mid of the shaft location as an exciter to apply electromagnetic force to the system. We investigate a nonlinear bearing-pedestal system model with the outer race defect under the electromagnetic force. The nonlinear differential equations are integrated using the fourth-order Runge–Kutta algorithm. The simulation and experimental results show that the characteristic signal of outer race incipient defect is significantly amplified under the electromagnetic force through the AMBs, which is helpful to improve the diagnosis accuracy of rolling element bearing׳s incipient outer race defect.     

Vibration monitoring of rolling element bearings by the high-frequency resonance technique — a review

Vibration monitoring of rolling element bearings by the high-frequency resonance technique is reviewed. It is shown that the procedures for obtaining the spectrum of the envelope signal are well established, but that there is an incomplete understanding of the factors which control the appearance of this spectrum. Until the envelope spectrum can be fully explained, use of the technique is limited     

Grease lubrication of rolling element bearings — role of the grease thickener

The contribution of grease thickener to lubricant film formation was examined in this paper. Lubricant film thickness and friction were measured for different grease thickener types in a bearing simulation device. The results showed that the greases formed thick (20–80nm), low friction surface layers at low speeds, which were much greater than the corresponding base oil film. These films appeared to be formed by the physical deposition of thickener in the track during overrolling of the grease. This was confirmed by infrared reflection analysis, which showed the deposited films to have increased thickener content. The ability of grease to form renewable physically deposited solid films has significant implications for optimising lubricant formulation for certain applications, e.g. bearings operating at high temperatures and low speeds where a conventional elastohydrodynamic lubricating film would be inadequate. Copyright © 2007 John Wiley & Sons, Ltd.     

Sync-period frequency analysis and its application to the diagnosis of multiple element defects of rolling bearings

A method for diagnosing multiple element defects in rolling bearings has been investigated. The method combines the time-synchronous averaging and envelope spectral analysis techniques to produce spectra of synchronously averaged envelope signals with a range of synchronous frequencies. The spectra are displayed in the synchronous period versus frequency domain, to result in the sync-period versus frequency distribution. The distribution separates the characteristic defect frequencies and their associated sidebands in the synchronous period axis. This analysis technique makes it possible to detect and diagnose multiple defects appearing in different elements of rolling bearings. Another main benefit of the method is the significant noise reduction by both the enveloping and the synchronous averaging processes. Results from both computer synthesised data and experimental simulated data are presented.     

改进掩码自编码器的滚动轴承半监督故障诊断

针对滚动轴承在不同转速条件下数据分布不同以及实际工程应用中标签样本不足导致故障诊断精度低的问题,将领域适配模块融入掩码自编码器(MAE)中,提出了改进掩码自编码器(IMAE)的滚动轴承半监督故障诊断方法。首先,对滚动轴承振动信号进行连续小波变换(CWT)得到反应信号时频特征的二维时频图,然后对时频图随机掩码,利用无标签样本进行掩码自编码器预训练,获得数据中复杂的内在特征,减少对有标签样本的依赖;其次将领域适配模块引入到预训练后的编码器中,使用少量有标签源域数据对IMAE进行微调,在希尔伯特空间中利用最小化最大均值差异减小因转速不同造成的源域与目标域间数据分布差异;最后在Softmax分类层下实现滚动轴承半监督故障诊断。通过滚动轴承数据集实验验证,所提方法检测精度均达到94%以上,证明了该方法的可行性与有效性。     

一种改进CNN在轴承故障诊断中的应用

针对轴承智能故障诊断过程中的特征自适应提取和在变工况下诊断能力差的问题,提出了一种基于特征通道权重调整的“端对端”一维卷积神经网络(Squeeze-Excitation Convolutional Neural Network,SECNN)滚动轴承故障诊断模型。首先采用一维卷积神经网络自适应地从原始振动信号中提取特征进行分类;然后通过增加特征通道权重模块来获取通道全局信息,学习特征通道之间的依赖关系;再据此对特征通道权重进行调整,增强滚动轴承故障诊断模型在变工况下的特征自适应提取能力。通过轴承实验台数据的验证结果表明:SECNN在多个变载荷工况下的故障诊断准确率均值达到97%,相比于传统方法提高了20%左右。同时利用t-SNE技术可视化特征提取过程,进一步验证了所提取的诊断模型的有效性。     

基于自动编码器数据降维的滚动轴承故障诊断研究

针对滚动轴承振动信号多域特征数据维数较高的问题,采用自动编码器(Auto-Encoder,AE)对特征数据进行降维处理,实现故障诊断.该方法首先提取滚动轴承振动信号中的特征数据,其次通过自动编码器对特征数据进行降维,最后将降维后的数据用于训练BP(Back Propagation)神经网络,并进行故障诊断.为验证自动编...