基于信号共振稀疏分解与HFE的往复压缩机故障诊断

往复压缩机结构复杂,内部激励源众多,振动信号呈现强烈的非平稳、非线性的特性。针对这一问题,提出了信号共振稀疏分解与层次模糊熵(HFE)结合的往复压缩机故障诊断方法。与传统的频带划分信号分解方法不同,信号共振稀疏分解方法根据品质因子的不同,可将中心频率相近且频带相互重叠的信号有效分离,形成高、低共振分量;层次模糊熵既分析信号的低频成分,又分析信号的高频成分,可准确、全面的描述信号特征。利用信号共振稀疏分解方法提取故障信号,然后使用层次模糊熵构建特征向量;以SVM作为模式分类器,诊断故障类型。实验及结果表明,该方法可有效诊断往复压缩机轴承故障。     

压缩机气阀故障诊断

在简述压缩机气阀故障原因的基础上,总结了压缩机气阀故障的原因、信号特征、诊断方法及防治措施。     

往复压缩机气阀故障诊断的信息融合方法研究

针对使用单一信号对往复压缩机气阀故障进行诊断时所存在的不足．提出一种适合气阀故障诊断的信息融合方法。该方法将气阀的压力信号和振动信号特征进行信息融合，再通过RBF神经网络进行故障诊断。诊断实例表明该方法能更准确地诊断出气阀的各种故障。     

往复压缩机气阀泄漏故障试验模拟与诊断分析

气阀是往复压缩机最容易出故障的部件,因高频开启关闭、受力情况复杂时常导致泄漏等故障的发生,对机组高效、稳定运行产生不利影响。针对往复压缩机气阀泄漏故障,通过建模仿真获得气阀泄漏理论数据,掌握故障早期特征变化规律;基于试验台故障模拟试验,融合压力、温度数据分析故障特征,验证理论仿真结论;最后提出了一种融合多参数实现气阀泄漏故障分析诊断方法。研究成果为实际往复压缩机故障报警限值的确定和自动诊断提供了支持。     

往复式压缩机气阀故障的振动诊断方法

介绍了往复式压缩机气阀故障的振动诊断方法，并给出了气阀故障信号的物理解释。     

活塞压缩机气阀故障的高阶谱特征研究

研究了基于高阶谱的活塞压缩机气阀故障特征提取方法，分析了2D12活塞压缩机气阀正常、阀片断裂故障、弹簧损坏故障及阀片有缺口故障4种状态下的高阶谱特征。结果表明：高阶谱方法能够较准确地提取气阀的各种故障特征。     

关联维数在往复压缩机气阀故障诊断中的应用

针对传统方法难以有效进行往复压缩机气阀故障诊断的现状，提出了应用关联维数对往复压缩机气阀故障进行诊断的新方法。通过对气阀常见故障类型的诊断，表明了关联维数可以有效地分析出往复压缩机气阀的常见故障。     

基于RBF神经网络的往复压缩机气阀故障诊断

提出了一种往复压缩机气阀的故障诊断方法.把往复压缩机气阀的振动信号作为识别故障的特征向量,送入RBF神经网络中,进行故障类别的自动识别.试验结果表明,该诊断模型对往复压缩机气阀故障诊断具有良好的诊断效果,系统不仅能够检测到往复压缩机气阀故障的存在,而且能够比较准确地识别往复压缩机气阀的故障模式.     

滚动轴承故障的非线性诊断方法

反映滚动轴承故障的特征周期信号处于较低频带内，容易被噪声淹没，难以检测。针对当前的诊断方法没有充分利用其故障特征频带的问题，提出了基于混沌振子的滚动轴承故障的非线性诊断方法。该方法直接检测故障的低频特征信号，完成故障诊断。讨论了该方法的适用场合，并将其应用在滚动体剥落故障的诊断上，取得了良好的效果。     

近似熵在往复压缩机气阀故障诊断中的应用

针对传统方法难以有效进行往复压缩机气阀故障诊断的现状,提出了应用近似熵对往复压缩机气阀故障进行诊断的新方法。通过对气阀常见故障类型的诊断,表明了近似熵可以有效地分析出往复压缩机气阀的常见故障。     

基于小波分析的发动机气门漏气故障诊断研究

对发动机气门漏气声学特性及其振动诊断机理进行了研究 ,在模拟实验的基础上对测取的燃爆段缸盖振动信号进行小波变换 ,比较了三种工况下各尺度小波分解信号能量与总能量之比 ,验证了理论研究的正确性。提取高频带内能量与总能量之比作为特征参数 ,有效地诊断了发动机气门漏气故障。     

液压电磁阀故障机理分析与瞬态特性仿真

液压电磁阀作为液压系统中的关键元件,是电控系统和液压系统的控制中枢,在大型武器装备中使用数量多、分布广,且故障率较高,对其实施在线监测诊断对于确保液压系统的正常运行十分重要.在收集整理电磁阀常见故障的基础上,深入研究了各类故障的发生机理;通过对电磁阀常见故障和特征信号的可测性分析,提出了一种基于磁场和振动敏感的电磁阀非介入式测试诊断技术;通过对电磁阀工作过程中的磁路分析和阀芯受力分析,建立了电磁阀的瞬态响应仿真模型,并进行了实验验证,为电磁阀故障的快速检测诊断奠定了基础.     

Condition monitoring of engine timing system by using wavelet packet decomposition of a acoustic signal

A change in the technical condition of mechanical components of internal combustion engines may not be detected by on-board diagnostic systems installed in vehicles. In similar cases, measurements and analyses of vibroacoustic signals being recorded prove useful. Since there are certain limitations to vibration measurements, including those related to the vibration transmission and the engine’s high temperature at measurement points, the authors of this paper have proposed that measurements and analyses of acoustic signals should be applied for the sake of assessment of the internal combustion engine technical condition. However, such an assessment requires new acoustic signal processing methods to be developed, and so this subject has been elaborated in the paper. The article provides a discussion on the option of applying a wavelet packet decomposition while filtering the internal combustion engine’s acoustic signal in order to diagnose an excessive valve clearance. The authors prepared an algorithm enabling selection of the chosen details and approximation of the wavelet analysis to low-frequency components, which constitute the noise, as well as high-frequency components comprising information on the possible enlarged engine valve clearance. Next, based on the selected high-frequency acoustic signal components, a method for automatic detection of enlarged clearance valves was developed, assuming that energy participations of the acoustic signal being emitted were to be determined while opening and closing individual valves. Under the study discussed, identification tests were conducted using two 4-cylinder internal combustion engines featuring valves of different clearances to consequently confirm the efficiency of the algorithm developed for the acoustic signal filtration and automatic detection of enlarged clearance valves.     

Output waveform analysis of an electro-hydraulic vibrator controlled by the multiple valves

The existing research of the electro-hydraulic vibrator mainly focuses on system stability, working frequency width and output waveform distortion. However, this high frequency performance of the electro-hydraulic vibrator is difficult to be improved greatly due to fast insufficiently frequency response of the servo valve itself and limited compensation capability of the control structure in the vibrator system. In this paper, to realize high frequency vibration, an improved two-dimensional valve （here within defined as a 2D valve） as a main control component is adopted to the parallel connection with a servo valve to control the electro-hydraulic vibrator, Because the output waveforms of this electro-hydraulic vibrator are incapable to be verified through timely feedback as in the conventional electro-hydraulic servo system, the analysis to the output waveform becomes crucial to the design and control of the electro-hydraulic vibrator. The mathematical models of hydraulic actuation mechanism and the orifice area of the parallel valves connection are established first. And then the vibration process is divided into two sections in terms of the direction of the flow, the analytical expression of the excited waveform is solved. Based on relationships exist between working states and the control parameters the analytical results, the vibration boundary positions and the are derived. Finally an experimental system was built to validate the theoretical analysis. It is verified that this electro-hydraulic vibration system could achieve high working frequency, up to 2 000 Hz. The excited waveform is similar to the sinnsoidal waveform. And the ascent and decent slopes of the waveforms are somewhat asymmetrical. This asymmetry is not only caused by the change of the direction of the elastic force but also dependent on the bias position of the vibration. Consequentky the distortion of effective working waveform is less tha~ 10%. This electro-hydraulic vibrator controlled by the multiple valves could not only greatly enhance the working frequency but also precisely control the vibration characteristic variables such as waveform shape.     

基于粗集理论的往复泵泵阀故障诊断方法

提出了一种基于粗集理论的往复泵泵阀故障诊断方法。该方法可以直接从经过小波包处理的泵阀振动信号中提取故障诊断观测，并由此建立基于规则的泵阀故障诊断系统，该系统不仅可以对发生故障的单个泵阀进行诊断，而且还能对同时发生故障的多个泵阀进行诊断，试验结果表明了这种方法的有效性。这种方法的可行性也为其它复杂机械的故障诊断提供了新思路。     

基于峰度的电力变压器铁芯松动故障在线监测方法

本文提出一种对电力变压器铁芯松动故障进行在线监测和估计的方法。通过改变变压器铁芯的压紧力得到模拟松动故障情况下的铁芯振动信号,使用谱峰度方法验证并分析了变压器铁芯振动信号中高频分量的冲击特性,在此基础上采用小波变换对铁芯振动信号进行分解,对分解得到高频分量计算其峰度值并用以监测铁芯压紧力的减小趋势。对实验变压器不同测点的多组数据分析表明,谱峰度能有效检测铁芯振动信号中的较高频率冲击成分及铁芯压紧力发生较大变化的情况,铁芯振动信号小波域中高频成分的峰度值能够很好的反映出铁芯压紧力变化的趋势。本方法准确的表述了铁芯振动信号中与松动故障相关的统计特性,为电力变压器铁芯的故障监测提供了一种可行的方法。     

CEEMD和小波半软阈值相结合的滚动轴承降噪

针对滚动轴承振动信号降噪处理时如何保证信号边缘信息完整性的问题,提出将互补集合经验模态分解(complementary ensemble empirical mode decomposition,简称CEEMD)与小波半软阈值相结合的信号降噪方法,对滚动轴承故障高频振动信号进行降噪处理。首先,采用CEEMD方法对故障振动信号进行分解,针对信号特点自适应获取不同频段模态分量;其次,将对包含噪声污染的高频信号模态分量进行相关性分析,得到含噪成分较高的高频模态分量,进一步采用小波半软阈值进行降噪处理;最后,将降噪后的模态分量同残余分量进行信号重构,完成降噪过程。分析结果表明,相对于传统小波阈值降噪和CEEMD强制降噪方法,提出的方法能够有效去除高频信号的噪声,且最大程度地保证了原始信号的完整性,降噪效果更好。     

用概率分析研究克服流体诱发调节阀振动

按流体诱发振动的流体力学理论 ,应用概率分析和数理统计方法 ,研制了异径多束流调节阀。使用表明 ,其克服流体诱发调节阀的振动效果十分显著     

基于小波包分解与主流形识别的非线性降噪

为解决工程实际中强噪声、非线性且频率成分复杂的振动信号降噪问题,提出了基于小波包分解和主流形识别的非线性降噪方法。采用小波包分解将原始振动信号正交无遗漏地分解到各频带范围内,根据各子频带中信噪空间分布,分别采用相应参数对小波包分解系数进行相空间重构;采用局部切空间排列(local tangent space alignment,LTSA)主流形识别方法在高维相空间中实现信号与噪音的分离,并重构出降噪后的一维小波包分解系数,最后进行小波包分解重构得到降噪后的振动信号。通过仿真实验和实例应用对本文所提方法的有效性进行了验证,试验结果表明本文方法具有良好的非线性降噪能力。     

基于改进经验小波变换的行星齿轮箱故障诊断

行星齿轮箱振动信号具有复杂多分量和调幅-调频的特点。幅值解调和频率解调方法能够避免传统Fourier频谱中的复杂边带分析,有效识别故障特征频率。经验小波变换通过对信号Fourier频谱的分割构造一组正交滤波器组,能提取具有紧支撑Fourier频谱的单分量成分,再对单分量成分运用Hilbert变换即可实现信号的解调分析。经验小波变换能够有效分离出调幅-调频成分,不存在模态混叠现象,具有完备的理论基础,自适应性好、算法简单、计算速度快。将改进的经验小波变换应用于行星齿轮箱振动信号的解调分析;提出了一种单分量个数的估算方法,解决了经验小波变换中的Fourier频谱划分问题;给出了对故障敏感的信号分量的选取方法,提高了分析的针对性。将改进方法应用于行星齿轮箱振动仿真信号和实验信号分析,验证了该方法的有效性。