基于多维度排列熵与支持向量机的轴承早期故障诊断方法

针对许多现有方法无法有效诊断滚动轴承早期故障的问题,引入排列熵的方法对轴承振动信号进行早期故障分析。通过研究嵌入维数和延迟时间对信号排列熵的影响,提出多维度排列熵的特征提取方法。利用多维度排列熵方法所提取的特征,建立了基于支持向量机的轴承早期故障智能诊断模型。对轴承不同类型、不同程度的故障数据进行分析,证明了多维度排列熵方法可以有效提取轴承不同状态的特征信息,与支持向量机结合的智能诊断模型可以精确地诊断轴承不同类型的早期故障,具有很强的通用性;该模型在贫样本的情况下,依然具有很高的诊断精度,适用于滚动轴承早期故障状态的在线监测。     

改进多尺度幅值感知排列熵与随机森林结合的滚动轴承故障诊断

针对滚动轴承故障识别准确率较低的问题,本文提出了一种新型滚动轴承故障诊断方法。该方法能够在准确识别滚动轴承故障类型的基础上,进一步分析故障的严重程度。首先,通过固有时间尺度分解提取滚动轴承振动信号的最佳固有旋转分量,突显故障信号的冲击特征;然后,利用改进多尺度幅值感知排列熵对信号幅值和频率变化敏感的特性,计算不同时间尺度下的幅值感知排列熵作为故障特征向量,改善了多尺度分析中的粗粒化过程,提升了故障特征提取的稳定性;最后,利用故障特征集构建随机森林多分类器,实现对滚动轴承不同故障类型的识别及严重程度分析,具有较强的泛化能力。实验结果表明,与现有滚动轴承故障诊断方法相比,平均故障识别准确率达到99.25%。该方法能够稳定而有效地提取滚动轴承的故障特征且具有较好的实时性。     

基于二叉树的支持向量机旋转机械故障诊断方法

支持向量机理论最初是针对两类模式识别问题而提出的.在故障诊断领域,多类故障诊断问题更为普遍.针对支持向量机常用的多类分类算法进行了分析,在此基础上提出了一种基于聚类思想的二叉树多类分类算法,并运用该算法对转子模拟试验台几种典型的故障进行了模式识别,实验结果表明新方法能有效地、准确地识别故障模式,比较符合实际工程要求,具有较高的推广性能.     

基于局域均值分解包络谱和SVM的滚动轴承故障诊断方法研究

论述了局域均值分解(Local mean decomposition,LMD)的定义和算法。结合局域均值分解、包络分析和支持向量机(Support vector machine,SVM)的各自特点,提出了一种基于LMD包络谱和SVM的滚动轴承故障诊断方法,该方法先对滚动轴承振动信号进行分解,得到一系列的生产函数分量,然后,再对前面几个生产函数分量进行包络分析,从包络谱中提取特征幅值比作为特征向量输入到SVM分类器中进行识别。实验结果验证了提出的方法的有效性,可以有效地识别滚动轴承的不同故障。     

一种基于动态剪枝二叉树SVMs的高炉故障诊断新方法

高炉故障诊断是一个多类分类问题,且各个故障类别间具有一定的关系,在识别其中某一类故障时,并不需要区分全部故障的类别,为此提出了基于剪枝二叉树的支持向量机改进算法,每次识别时都去除相对没有价值的支持向量,根据类间相似度重新构造二叉树,剪掉没有价值的枝节,减少支持向量机个数,加速识别过程。通过对高炉故障模型的仿真实验,比较不同多类分类算法的性能,证明该方法能够在保证识别准确率的情况下提高故障诊断速度。     

A roller bearing fault diagnosis method based on hierarchical entropy and support vector machine with particle swarm optimization algorithm

Targeting the non-linear dynamic characteristics of roller bearing faulty signals, a fault feature extraction method based on hierarchical entropy (HE) is proposed in this paper. SampEns of 8 hierarchical decomposition nodes (e.g. HE at scale 4) are calculated to serve as fault feature vectors, which takes into account not only the low frequency components but also high frequency components of the bearing vibration signals. HE can extract more faulty information than multi-scale entropy (MSE) which considers only the low frequency components. After extracting HE as feature vectors, a multi-class support vector machine (SVM) is trained to achieve a prediction model by using particle swarm optimization (PSO) to seek the optimal parameters of SVM, and then ten different bearing conditions are identified through the obtained SVM model. The experimental results indicate that HE can depict the characteristics of the bearing vibration signal more accurately and more completely than MSE, and the proposed approach based on HE can identify various bearing conditions effectively and accurately and is superior to that based on MSE.     

A rolling bearing fault diagnosis strategy based on improved multiscale permutation entropy and least squares SVM

A novel rolling bearing fault diagnosis strategy is proposed based on Improved multiscale permutation entropy (IMPE), Laplacian score (LS) and Least squares support vector machine-Quantum behaved particle swarm optimization (QPSO-LSSVM). Entropy-based concepts have attracted attention recently within the domain of physiological signals and vibration data collected from human body or rotating machines. IMPE, which was developed to reduce the variability of entropy estimation in time series, was used to obtain more precise and reliable values in rolling element bearing vibration signals. The extracted features were then refined by LS approach to form a new feature vector containing main unique information. By constructing the fault feature, the effective characteristic vector was input to QPSO-LSSVM classifier to distinguish the health status of rolling bearings. The comparative test results indicate that the proposed methodology led to significant improvements in bearing defect identification.     

A novel fault diagnosis model for gearbox based on wavelet support vector machine with immune genetic algorithm

A novel intelligent diagnosis model based on wavelet support vector machine (WSVM) and immune genetic algorithm (IGA) for gearbox fault diagnosis is proposed. Wavelet support vector machine is a powerful novel tool for solving the diagnosis problem with small sampling, nonlinearity and high dimension. Immune genetic algorithm is developed in this study to determine the optimal parameters for WSVM with the highest accuracy and generalization ability. Moreover, the feature vectors for fault diagnosis are obtained from vibration signal that preprocessed by empirical mode decomposition (EMD). The experimental results indicate that this proposed approach is an effective method for gearbox fault diagnosis, which has more strong generalization ability and can achieve higher diagnostic accuracy than that of the artificial neural network and the SVM which has randomly extracted parameters.     

A feature extraction method based on HLMD and MFE for bearing clearance fault of reciprocating compressor

According to the nonlinearity, nonstationarity and multi-component coupling characteristics of reciprocating compressor vibration signal, a feature extraction method based on hermite local mean decomposition (HLMD) and multiscale fuzzy entropy (MFE) is proposed for the diagnosis of reciprocating compressor oversized bearing clearance faults. Firstly, aiming at the strong nonstationary characteristic of vibration signal, a novel HLMD algorithm was given by using the monotone piecewise cubic hermite interpolation (MPCHI) instead of cubic spline interpolation (CSI) to construct the envelopes. Secondly, HLMD was performed on the vibration signals in each state and a series of PF components are produced, and the highlighted PF components which contain the main information of fault state were chosen with the correlation coefficient. Thirdly, MFE of the selected PF components were calculated to form the eigenvectors matrix, and the eigenvectors which have the best divisibility were extracted based on the average euclidean distances of each scale factor. Finally, four bearing clearance fault states were extracted by the proposed method, and taken SVM as a pattern classifier, the faults were diagnosed accurately. Furthermore, the comparison of recognition results with other three feature extraction methods demonstrates the superiority of this method.     

A roller bearing fault diagnosis method based on the improved ITD and RRVPMCD

Targeting that the measured vibration signal of roller bearing contains the characteristics of non-stationary and nonlinear, and the extraction features may contain smaller correlation and redundancy characteristics in the roller bearing fault diagnosis, the vibration signal processing method based upon improved ITD (intrinsic time-scale decomposition) and feature selection method based on Wrapper mode are put forward. In addition, in the design of the classifier, targeting the limitation of existing pattern recognition method, a new pattern recognition method-variable predictive model based class discriminate (VPMCD) is introduced into roller bearing fault identification. However, the parameters are fitted by using least squares in VPMCD method, while least squares regression is sensitive to “abnormal value”. Therefore, a robust regression-variable predictive mode-based class discriminate (RRVPMCD) method is proposed in this paper, robust regression is adopted to estimate parameters and the effect of “abnormal value” in the estimation of parameters would be reduced by giving each feature a weight. Firstly, improved ITD method and feature selection method based on Wrapper mode are combined to extract the fault features of roller bearing vibration signals, and feature vector matrixes are established, then a predictive model is built through the method of RRVPMCD, finally, the established predictive model is used for pattern recognition. Experimental results show that the model based on the improved ITD, the Wrapper feature selection and RRVPMCD method can effectively identify work status and fault type of roller bearing.     

Multi-fault diagnosis study on roller bearing based on multi-kernel support vector machine with chaotic particle swarm optimization

A novel intelligent fault diagnosis model based on multi-kernel support vector machine (MSVM) with chaotic particle swarm optimization (CPSO) for roller bearing fault diagnosis is proposed. Multi-kernel support vector machine is a powerful new tool for roller bearing fault diagnosis with small sampling, nonlinearity and high dimension. Chaotic particle swarm optimization is developed in this study to determine the optimal parameters for MSVM with high accuracy and great generalization ability. Moreover, the feature vectors for fault diagnosis are obtained from vibration signal that preprocessed by time-domain, frequency-domain and empirical mode decomposition (EMD) and the typical manifold learning method LTSA is used to select salient features. The experimental results indicate that this proposed approach is an effective method for roller bearing fault diagnosis, which has more strong generalization ability and can achieve higher diagnostic accuracy than that of the single kernel SVM or the MSVM which parameters are randomly extracted.     

基于MCKD和改进HHT多尺度模糊熵的齿轮故障诊断方法

针对齿轮故障信号常伴有大量噪声,故障特征难以提取的问题,提出一种基于最大相关峭度解卷积(MCKD)和改进希尔伯特-黄变换(HHT)多尺度模糊熵的故障诊断方法。首先采用MCKD算法对采集到的齿轮振动信号进行降噪处理,以提高信号的信噪比;然后利用自适应白噪声完备经验模态分解(CEEMDAN)对降噪后信号进行分解,获得一系列不同尺度的固有模态函数(IMF),并通过相关系数-能量的虚假IMF评价方法选取对故障敏感的模态分量;最后计算敏感IMF分量的模糊熵,将获得的原信号多尺度的模糊熵作为状态特征参数输入最小二乘支持向量机(LS-SVM)中,对齿轮的故障类型进行诊断。实测信号的诊断结果表明,该方法可实现齿轮故障的有效诊断。     

An improved complementary ensemble empirical mode decomposition with adaptive noise and its application to rolling element bearing fault diagnosis

A novel time–frequency analysis method called complementary complete ensemble empirical mode decomposition (EEMD) with adaptive noise (CCEEMDAN) is proposed to analyze nonstationary vibration signals. CCEEMDAN combines the advantages of improved EEMD with adaptive noise and complementary EEMD, and it improves decomposition performance by reducing reconstruction error and mitigating the effect of mode mixing. However, because white noise mixed in with the raw vibration signal covers the whole frequency bandwidth, each mode inevitably contains some mode noise, which can easily inundate the fault-related information. This paper proposes a time–frequency analysis method based on CCEEMDAN and minimum entropy deconvolution (MED) for fault detection of rolling element bearings. First, a raw signal is decomposed into a series of intrinsic mode functions (IMFs) by using the CCEEMDAN method. Then a sensitive parameter (SP) based on adjusted kurtosis and Pearson’s correlation coefficient is applied to select a sensitive mode that contains the most fault-related information. Finally, the MED is applied to enhance the fault-related impulses in the selected IMF. The fault signals of high-speed train axle-box bearing are applied to verify the effectiveness of the proposed method. Results show that the proposed method can effectively reveal axle-bearing defects’ fault information. The comparisons illustrate the superiority of SP over kurtosis for selecting the sensitive mode from the resulted signal of CCEEMEDAN. Further, we conducted comparisons that highlight the superiority of our proposed method over individual CCEEMDAN and MED methods and over two other popular signal-processing methods, variational mode decomposition and fast kurtogram.     

基于支持向量机属性约简集成的模拟电路故障诊断

针对模拟电路故障数据存在大量无关或冗余特征的特点,为进一步提高故障诊断准确率,提出支持向量机属性约简集成的模拟电路故障诊断新方法.首先证明一致决策表属性约简与集合覆盖的等价性,将最优属性约简问题转化成最小集合覆盖问题;然后在结合混沌优化产生初始信息素分布和进行混沌扰动的基础上,设计求解最小集合覆盖问题的混沌蚁群算法;最后给出基于属性约简集成的模拟电路故障诊断模型.用双二次滤波电路对算法进行验证,取得97.8%的故障诊断准确率,与其他方法进行比较,结果显示了本文方法的优越性.     

A novel intelligent gear fault diagnosis model based on EMD and multi-class TSVM

In the condition monitoring of gear reducer, the labeled fault samples are sparse and expensive, while the unlabeled samples are plentiful and cheap. How to diagnose the faults occurring in complex and special gear reducer effectively becomes a troublesome problem in case of insufficient labeled samples or excess unlabeled samples. This paper presents a novel model for fault diagnosis based on empirical mode decomposition (EMD) and multi-class transductive support vector machine (TSVM), which is applied to diagnose the faults of the gear reducer. The experimental results obtain a very high diagnosis accuracy. Even though the number of unlabeled samples is 50 times as that of labeled samples, the mean of testing accuracy of the proposed novel method can reach at 91.62%, which distinctly precedes the testing success rates of the other similar models in the same experimental condition.     

A multi-step progressive fault diagnosis method for rolling element bearing based on energy entropy theory and hybrid ensemble auto-encoder

It is meaningful to efficiently identify the health status of bearing and automatically learn the effective features from the original vibration signals. In this paper, a multi-step progressive method based on energy entropy (EE) theory and hybrid ensemble auto-encoder (HEAE), systematically blending the statistical analysis approach with the deep learning technology, is proposed for rolling element bearing (REB) fault diagnosis. Firstly, a preliminary detection about the REB health status is performed by the statistical analysis technique integrated with the EE theory. Secondly, if fault exists in REB, a new HEAE is constructed based on denoising auto-encoder and contractive auto-encoder to strengthen the feature learning ability and automatically extract the deep state features from the raw data. Subsequently, a modified t-distributed stochastic neighbor embedding (M-tSNE) algorithm is developed to achieve the features reduction to further improve the diagnosis efficiency. Finally, the low-dimensional representations after features reduction are as the inputs of softmax classifier to recognize the fault conditions. The proposed method is applied to the fault diagnosis of REB. The results confirm the effectiveness and superiority of the proposed method, and it is more suitable for the actual engineering applications compared with other existing methods.     

A novel scheme for fault detection of reciprocating compressor valves based on basis pursuit,wave matching and support vector machine

A scheme for fault detection of compressor valves based on basis pursuit (BP), wave matching and support vector machine (SVM) is presented. BP is applied to extract the main vibration component in the signal and suppress background noise. Wave matching is a new feature extraction method proposed in this paper. Instead of extracting features through commonly used indicators such as statistic measures or information entropy, wave matching extracts features by matching the vibration signal with parameterized waveform optimized by differential evolution (DE) algorithm. It only produces a small number of features and the features have clear physical meaning. SVM is employed in the fault classification because of its superiority in dealing with small sample problems. The results of real compressor valve signal analysis confirm that the proposed scheme can differentiate compressor valve faults with high accuracy and reliability.     

A hybrid fault diagnosis method based on second generation wavelet de-noising and local mean decomposition for rotating machinery

In order to extract fault features of large-scale power equipment from strong background noise, a hybrid fault diagnosis method based on the second generation wavelet de-noising (SGWD) and the local mean decomposition (LMD) is proposed in this paper. In this method, a de-noising algorithm of second generation wavelet transform (SGWT) using neighboring coefficients was employed as the pretreatment to remove noise in rotating machinery vibration signals by virtue of its good effect in enhancing the signal–noise ratio (SNR). Then, the LMD method is used to decompose the de-noised signals into several product functions (PFs). The PF corresponding to the faulty feature signal is selected according to the correlation coefficients criterion. Finally, the frequency spectrum is analyzed by applying the FFT to the selected PF. The proposed method is applied to analyze the vibration signals collected from an experimental gearbox and a real locomotive rolling bearing. The results demonstrate that the proposed method has better performances such as high SNR and fast convergence speed than the normal LMD method.