基于小波分析ARMA模型的气体压力识别

提出了内燃机缸内气体压力的一种新的间接识别方法。利用缸盖振动响应信号建立盖系统的ARMA模型，并用小波包变换方法对缸盖振动响应信号进行去噪和分形，将重构后的缸盖振动响应信号作为缸内气体压力识别的原始数据，大大提高了识别精度。计算的结果与实测的结果吻合较好，本文的研究对内燃机的状态监测和故障诊断具有实用价值。     

基于小波包变换的内燃机气阀漏气诊断方法

探讨了利用缸盖振动信号诊断内燃机气阀漏气的一种新方法－－小波包方法。诊断机理分析和实验研究表明,缸盖 振动信号中关于气阀漏气的特征信息集中在高频段,故选用了小波包变换作为信号处理的基本方法,因为小波包变换较之小波变换在中高频段具有更高更均匀的频率分辨率;对实测信号的分析和处理结果表明了小波包在该领域的适用性;建立了基于马氏距离法的气阀漏气多指标诊断模型;进行了实机诊断,取得了预期的效果。     

分形几何在内燃机振动信号分析中的应用

本文将分形几何引入内燃机振动信号的分析之中,用分维数描述振动信号的复杂程度。对缸盖振动信号的谱分析表明,可以用分维数描述缸盖振动信号。当气门发生故障时,分维数将较正常时小,由此可判断气门状态正常与否及故障类型。     

Laplace小波相关滤波法与冲击响应提取

冲击响应信号的出现标志着机械设备发生松动，碰撞，冲击等故障。在复杂的振动信号中准确提取出冲击响应信息在机械故障诊断中具有重要的意义，介绍了用来识别冲击响应信号的Lapalce小波相关滤波法，通过对含有噪声的模拟冲击响应信号的识别，证明该方法能够在强大噪声干扰中准确捕捉到冲击响应信号。工程应用结果表明，相关滤波法可以从复杂的内燃机缸盖振动信号中准确定位进气阀关闭时冲击发生的时刻及频率，从而成功地诊断出因进气阀磨损而导致的漏气故障。     

内燃机KVMD-MHD振动谱图表征与TD-2DPCA编码诊断方法研究

为了直接对内燃机振动谱图像进行诊断识别,提出一种基于改进变分模态分解(VMD)、Margenau-Hill(MHD)时频分析与双向二维主成分分析(Two-directional,Two-dimensional PCA,TD-2DPCA)的内燃机振动谱图像识别诊断方法。该方法首先针对VMD分解过程中的层数选取问题,提出了一种中心频率筛选的VMD分解层数改进方法(KVMD),然后将内燃机振动信号利用KVMD分解成一组单分量模态信号,并对生成的各个单分量信号进行MHD处理后表征成振动谱图像;在此基础上,对生成的内燃机KVMD-MHD振动谱图像采用双向二维主成分分析形成编码矩阵,并采用最近邻分类器(KNNC)对上述编码矩阵直接进行模式识别,以实现内燃机振动谱图像的自动诊断。最后,将该方法应用在气阀机构4种工况下的缸盖表面振动信号诊断实例中,结果表明:该方法不仅改进了传统图像模式识别中的特征参数提取方法,而且能很好地消除时频分布中的交叉干扰项,使各时频分量物理意义明确,能有效诊断出内燃机气阀机构故障,为内燃机振动诊断探索了一条新途径。     

基于内燃机振动信号的可视化识别诊断

为提高故障识别诊断的精确度和实时性,有效解决内燃机多分量、非平稳振动信号特征提取困难的问题,提出一种基于改进局部二值模式(ILBP)与双向二维主成分分析(TD-2DPCA)的内燃机振动信号可视化故障识别诊断方法。针对传统时频方法在分析内燃机振动信号中,存在时频分辨率低及交叉干扰项的问题,将经验小波变换(EWT)与同步压缩小波变换(SST)应用到内燃机振动信号的时频图表征中;利用ILBP提取图像的纹理特征,并对ILBP图谱采用TD-2DPCA降维,将降维后的编码矩阵向量化后得到图像的特征参数;通过支持向量机(SVM)和最近邻分类器(NNC)分别特征向量进行训练、测试,实现内燃机的故障识别诊断。在内燃机气门间隙故障8种工况下缸盖振动信号的识别诊断试验中,均得到较高的分类精度;通过参数的合理优化,在保证了分类速率的同时,最高识别率达到96.67%,对比其他方法,充分表明该方法在内燃机故障诊断中的有效性。     

信号分析在柴油机气缸压力识别中的应用

论述了采用小波包算法对柴油机缸盖振动信号进行时频特性分析和信噪分离的方法，比较了不同的建模方法的特点并利用时间序列分析方法对缸盖振动信号和气缸压力信号分别建立时序模型，求取了缸盖系统的传递函数，达到了利用缸盖表面振动信号识别气缸压力的目的。     

利用缸盖噪声信息诊断柴油机失火故障

叙述了内燃机故障检测的新方法，即利用发动机运转的噪声信息来诊断内燃机失火故障，利用小波分析技术从发动机噪声信号中提取发动机的燃爆和气门开闭信息，相对于传统的基于振动的诊断方法，基于声学的诊断方法有着振动不可比拟的优点，这种方法不需接触发动机表面，不必粘贴传感器，不受发动机结构空间限制，而且检测系统简单，大大缩短了检测时间，提高了检测效率。     

基于EMD-WVD与LNMF的内燃机故障诊断

内燃机的振动信号是复杂非平稳信号,准确提取内燃机振动信号中的特征信息进行模式识别,是对振动信号进行故障诊断的关键。基于经验模态分解的维格纳时频分析方法,不但保留了维格纳分布的所有优良特,而且还避免了交叉项的干扰,能够有效地提取内燃机振动信号的特征信息;在此基础之上,针对传统非负矩阵分解非正交的基矩阵导致数据冗余性较大、影响后续故障分类准确率提高的问题,提出采用局部非负矩阵分解的方法,直接对EMD-WVD时频图像的矩阵进行分解,计算用于内燃机故障诊断的特征参数,并利用特征参数进行故障分类。对内燃机4种不同工况的振动信号进行实验,证明基于EMD-WVD与局部非负矩阵分解的方法对内燃机气门间隙的故障诊断的有效性。     

利用缸盖振动信号诊断气门机构故障

在分析缸盖振动信号时域和频域特征的基础上，详细地研究了气门间隙异常对缸盖振动信号时域和频域特征的影响，并利用时域和频域的变化诊断配气机构的故障，提出了用气门关闭时刻改变定量诊断气门异常间隙的分析方法。     

Fault Diagnosis for IC Engines Using Wavelet Packet and Image Processing

There are few applications of image processing technology for diagnosing and state monitoring for internal combustion (IC) engines, which is discussed in detail in this paper. The time-frequetwy distribution images of cylinder head vibration signals are obtained by decomposing them with a wavelet packet algorithm. It is the first time that toe look at time-frequency distribution images from the point of images. Based on this, a new method for applying image processing technology for diagnosing and state monitoring for internal combustion engines is presented in this paper. A valve fault diagnosis model is set up by image matching, which is realized on a four-stroke, six-cylinder diesel engine. At the same time,some notes are presented in this paper. It has been proved that it is of no good effect to diagnose with histograms of time-frequency images generated by cylinder head vibration signals that have been processed with a wavelet packet algorithm. The reason is given in this paper. Comparisons of diagnosing effect are carried out between noise-added signals and original signals. It has little effect on diagnosing results after signals have been added with noise. The results show that this method has a clear physical meaning and is of good engineering practicability, feasibility, good precision and high speed.     

Bearings Fault Diagnosis Using Vibrational Signal Analysis by EMD Method

ABSTRACT

Studying vibrational signals is one reliable method for monitoring the situation of rotary machinery. There are various methods for converting vibrational signals into usable information for fault diagnosis, one of which is the empirical mode decomposition method (EMD). This article is about diagnosing bearing faults using the EMD method, employing nondestructive test. Vibration signals are acquired by a bearing test machine. The discrete 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. Local Hilbert marginal spectrum can be obtained by applying thr EMD method to the envelope signal from which the faults in a roller bearing can be diagnosed and fault patterns can be identified. The results have shown bearing faults frequencies are easily observable. There is a variant of the EMD method called the ensemble EMD (EEMD), which overcomes the mode mixing problem which may occur when the signal to be decomposed is intermittent. The EEMD method is also applied to the acquired signals, and the two methods were compared. While the outcomes of both methods do not differ much, one important merit of the EMD is that it has much less computational processing time than EEMD.     

Mixed Fault Diagnosing of Rolling Bearing Under Noise

In field of rolling bearing fault diagnosis,the sampled bearing vibration signals will be generally disturbed with noise. In noisy environment,the conventional blind source separation method is not good for diagnosing bearing faults. In this paper,wavelet de-noising method and blind source separation technology have been combined. In order to achieve fault diagnosis of rolling bearing,firstly wavelet soft threshold de-noising method has been applied on sampled signals. Then the better robust JADE algorithm has been applied in signals blind source separation. At last,vibration signals bearing inner and outer faults of have been analyzed in this paper,and the corresponding bearing faults have been diagnosed successfully. The proposed research methods provide a new way for diagnosing rolling bearing's mixed faults under noise.     

A Method for Incipient Fault Diagnosis of Roller Bearings Based on the Wavelet Transform Correlation Filter and Hilbert Transform

Noise is the biggest obstacle that makes the incipient fault diagnosis results of roller bearings uncorrected; a new method for diagnosing incipient fault of roller bearings based on the Wavelet Transform Correlation Filter and Hilbert Transform was proposed. First, the weak fault information features are picked up from the roller bearings fault vibration signals by use of a de-noising characteristic of the Wavelet Transform Correlation Filter as the preprocessing of the Hilbert Envelope Analysis. Then, in order to get fault features frequency, de-noised wavelet coefficients of high scales which represent high frequency signal were analyzed by Hilbert Envelope Spectrum Analysis. The simulation signals and diagnosing examples analysis results reveal that the proposed method is more effective than the method of direct wavelet coefficients-Hilbert Transform in de-noising and clarifying roller bearing incipient fault.     

全矢小波包-包络分析方法及其在齿轮故障诊断中的应用

在旋转机械故障诊断中,针对传统单源信息采集的不全面性,提出了一种基于全矢谱技术的小波包-包络分析方法。首先对同源双通道信息分别采用小波包分解,根据需要选择频段的信息,并对提取的信号进行重构。然后采用全矢Hilbert解调分析方法对重构信号实现包络解调,并与两单源信息的包络解调相比较,说明了仅以单源信息为诊断依据的不足。利用全矢谱技术进行融合的全矢小波包-包络解调技术,不仅继承了小波包-包络分析方法的优势,而且更加全面地反映出了信号的真实性。最后通过仿真信号对其算法的可行性进行了验证,同时又以齿轮的故障振动信号为例,进一步表明了该方法在故障诊断中的有效性。     

Fault diagnosis method for an Aeroengine Based on Independent Component Analysis and the Discrete Hidden Markov Model

The vibration signals of an aeroengine are a very important information source for fault diagnosis and condition monitoring. Considering the nonstationarity and low repeatability of the vibration signals, it is necessary to find a corresponding method for feature extraction and fault recognition. In this paper, based on Independent Component Analysis (ICA) and the Discrete Hidden Markov Model (DHMM), a new fault diagnosis approach named ICA-DHMM is proposed. In this method, ICA separates the source signals ...     

基于EMD与神经网络的滚动轴承故障诊断方法

针对滚动轴承故障振动信号的非平稳特征，提出了一种基于经验模态分解(Empirical Mode Decomposition，简称EMD)和神经网络的滚动轴承故障诊断方法。该方法首先对原始信号进行了经验模态分解，将其分解为多个平稳的固有模态函数(Intrinsic Mode function，简称IMF)之和，再选取若干个包含主要故障信息的IMF分量进行进一步分析，由于滚动轴承发生故障时，加速度振动信号各频带的能量会发生变化，因而可从各IMF分量中提取能量特征参数作为神经网络的输入参数来识别滚动轴承的故障类型。对滚动轴承的正常状态、内圈故障和外圈故障信号的分析结果表明，以EMD为预处理器提取各频带能量作为特征参数的神经网络诊断方法比以小波包分析为预处理器的神经网络诊断方法有更高的故障识别率，可以准确、有效地识别滚动轴承的工作状态和故障类型。     

基于时频域模型的噪声故障诊断

为了避免传统的基于振动信号的内燃机主轴承磨损故障诊断中安装传感器以及提取故障特征频率的麻烦,采用一种基于内燃机工作噪声信号和时频域分析的方法。首先讨论了对内燃机噪声信号进行小波包络谱分析,得到可以判断主轴承磨损故障的特征频率。然后,进一步阐述了采用噪声信号小波包分解,可得到包含更多故障信息时-频分布图。基于此,运用图像处理技术建立基于图像匹配的内燃机主轴承诊断模型。结果表明此方法简单有效,充分利用了故障信息。     

基于时间-小波能量谱的齿轮故障诊断

振动信号中的冲击现象及其频率特征是诊断齿轮局部损伤故障的重要依据之一。针对齿轮故障特征提出了一种时间-小波能量谱信号处理方法,它能够有效提取振动信号中冲击成分的时域和频域特征。利用时间-小波能量谱方法分析了正常、磨损、断齿等三种状态的齿轮箱振动信号,并与传统频谱分析方法进行相比。结果表明：时间-小波能量谱不仅可以有效提取故障特征,识别出齿轮箱的故障存在,而且可以清晰地分辨出故障类型及故障元件。     

基于EEMD能量熵及LS-SVM滚动轴承故障诊断

针对滚动轴承振动信号的非平稳特性和现实中难以获得大量典型故障样本的实际情况,提出基于集合经验模态分解（EEMD）能量熵和最小二乘支持向量机（LS-SVM）的滚动轴承故障诊断方法。首先通过EEMD分解将非平稳的原始振动信号分解成若干个平稳的固有模态函数（IMF）;滚动轴承同一部位发生不同严重程度的故障时,在不同频带内的信号能量值会发生改变,因此可通过计算振动信号的EEMD能量熵判断发生故障的严重程度;从包含主要故障信息的IMF分量中提取的能量特征作为输入来建立支持向量机,判断滚动轴承的技术状态和故障严重程度,并选用不同核函数对诊断效果进行分析比较。实验结果表明,该方法能有效地应用于滚动轴承的故障诊断。