Multiscale singular value manifold for rotating machinery fault diagnosis

Time-frequency distribution of vibration signal can be considered as an image that contains more information than signal in time domain. Manifold learning is a novel theory for image recognition that can be also applied to rotating machinery fault pattern recognition based on time-frequency distributions. However, the vibration signal of rotating machinery in fault condition contains cyclical transient impulses with different phrases which are detrimental to image recognition for time-frequency distribution. To eliminate the effects of phase differences and extract the inherent features of time-frequency distributions, a multiscale singular value manifold method is proposed. The obtained low-dimensional multiscale singular value manifold features can reveal the differences of different fault patterns and they are applicable to classification and diagnosis. Experimental verification proves that the performance of the proposed method is superior in rotating machinery fault diagnosis.     

Rotating machinery diagnosis using wavelet packets-fractal technology and neural networks

This paper presents a new fault diagnosis procedure for rotating machinery using the wavelet packets-fractal technology and a radial basis function neural network. The main purpose is to investigate different fault conditions for rotating machinery, such as imbalance, misalignment, base looseness and combination of imbalance and misalignment. In this study, we measured the non-stationary vibration signals induced by these fault conditions. Applying wavelet packets transform to these signals, the fractal dimension of each frequency channel was extracted and the box counting dimension was used to depict the failure characteristics of the fault conditions. The failure modes were then identified by a radial basis function neural network. An experiment was conducted and the results showed that the proposed method can detect and recognize different kinds of fault conditions. Therefore, it is concluded that the combination of wavelet packets-fractal technology and neural networks can provide an effective method to diagnose fault conditions of rotating machinery.     

Non-dominated solution set based on time–frequency infograms for local damage detection of rotating machines

The determination of an index to balance the impulsiveness and cyclostationarity of an expected component is an interesting research topic in mechanical health monitoring. Many proposed indices focus on evaluating the impulsiveness or cyclostationarity of an analyzed signal, thus resulting that an expected component cannot dominate other modes using these indices under the interferential circumstance of occasional shocks and harmonic signals. Moreover, several parameters have to be determined in advance, thereby leading to taxing consumption even if some methods can achieve a trade-off between impulsiveness and cyclostationarity via multi-objective optimization algorithms. A non-dominated solution set based on time–frequency infograms is then proposed in this study to preferably synthesize the impulsiveness and cyclostationarity of detecting the local damage of rotating machines. Specifically, the proposed method is conducted through four steps: (a) decompose the measured signal into certain levels, (b) construct the time–frequency infograms, (c) define the non-dominated solution set by Pareto filtering, and (d) perform Hilbert transform on these non-dominated solutions to diagnose the health condition of the rotating machines. Several case studies, including a simulated signal, and experimental data from defective bearing and gear respectively, were conducted to validate the proposed method. Analysis and comparison results indicated that the proposed method outperforms the fast kurtogram, infograms and some advanced indices in identifying fault features from rotating machines.     

Compressed sparse time–frequency feature representation via compressive sensing and its applications in fault diagnosis

Feature extraction in time–frequency domain is wildly used in fault diagnosis of rotating machines. However, it needs more time and space to store the time–frequency information, which restricts its practical applications, especially for remote health monitoring. A novel parallel FISTA-like proximal decomposition algorithm was proposed for reconstruction of sparse time–frequency representation (TFR) from the limited noisy observations based on the recently developed compressive sensing. The effectiveness of recovering buried sparse signatures was demonstrated by numerical simulations. The proposed method yielded better results than those obtained by the traditional RecPF method. A novel framework for remote machine health condition monitoring was then developed via the proposed algorithm and the advancements in wireless communication. The effectiveness of the new proposed method for the sparse TFR in detecting bearings and gears defects in rotating machines is further verified using many practical cases. These results illustrate the proposed method can well retain TF signatures without clearly artifacts in the recovered TFR using only very limited measurements.     

基于解析模态分解的机械故障诊断方法

针对旋转机械故障诊断问题,提出了一种基于解析模态分解(AMD)的旋转机械故障诊断方法。只要知道信号的频率成分,AMD方法就可以将含不同频率成分的信号分解为单频率信号,尤其能够分解有紧密间隔频率成分的信号。对于可预知故障特征频率的旋转机械的故障诊断,可利用AMD方法提取机械振动信号中故障特征频率所在频段的信号,并求该段信号的频谱,若频谱中含有故障特征频率,则说明机械振动信号中存在该故障。通过对滚动轴承故障信号和转子不对中故障信号的分析以及和经验模态分解(EMD)方法的对比,证明了AMD方法的有效性,且AMD方法比EMD方法更快速、准确。     

一维深度子领域适配的不同转速下旋转机械 复合故障诊断

旋转机械复合故障与单一故障样本间相关性高易造成错分类,且旋转机械转速往往不同,进一步加剧了旋转机械复合故障诊断的难度。针对上述问题,提出一维深度子领域适配的不同转速下旋转机械复合故障诊断方法。首先,以旋转机械复合故障的频域信号作为网络的输入,最大程度保留信号特征;其次,搭建领域共享的一维卷积神经网络,对不同转速下旋转机械复合故障的频域信号特征进行学习;然后,添加局部最大均值差异形成子领域适配层,对齐每对子领域分布以避免单一故障和复合故障的特征混合,并通过最小化局部最大均值差异值缩小两域子领域特征分布差异,以减少不同转速所带来的干扰;最后,在子领域适配层后添加softmax分类层,实现对目标数据的故障状态识别。通过不同转速旋转机械复合故障诊断实验证明了所提方法的有效性。     

Imbalanced fault diagnosis of rotating machinery using autoencoder-based SuperGraph feature learning

Existing fault diagnosis methods usually assume that there are balanced training data for every machine health state. However, the collection of fault signals is very difficult and expensive, resulting in the problem of imbalanced training dataset. It will degrade the performance of fault diagnosis methods significantly. To address this problem, an imbalanced fault diagnosis of rotating machinery using autoencoder-based SuperGraph feature learning is proposed in this paper. Unsupervised autoencoder is firstly used to compress every monitoring signal into a low-dimensional vector as the node attribute in the SuperGraph. And the edge connections in the graph depend on the relationship between signals. On the basis, graph convolution is performed on the constructed SuperGraph to achieve imbalanced training dataset fault diagnosis for rotating machinery. Comprehensive experiments are conducted on a benchmarking publicized dataset and a practical experimental platform, and the results show that the proposed method can effectively achieve rotating machinery fault diagnosis towards imbalanced training dataset through graph feature learning.     

Fault diagnosis of rotating machinery based on the statistical parameters of wavelet packet paving and a generic support vector regressive classifier

The fault diagnosis of rotating machinery has attracted considerable research attention in recent years because such components as bearings and gears frequently suffer failure, resulting in unexpected machine breakdowns. Signal processing-based condition monitoring and fault diagnosis methods have proved effective in fault identification, but the revelation of faults from the resulting signals requires a high degree of expertise. In addition, it is difficult to extract the fault-induced signatures in complex machinery via signal processing-based methods. In this paper, a new intelligent fault diagnosis scheme based on the extraction of statistical parameters from the paving of a wavelet packet transform (WPT), a distance evaluation technique (DET) and a support vector regression (SVR)-based generic multi-class solver is proposed. The collected signals are first pre-processed by the WPT at different decomposition depths. In this paper, the wavelet packet coefficients at different decomposition depths are referred to as WPT paving. Statistical parameters are then extracted from the signals obtained via the WPT at different decomposition depths. In selecting the sensitive fault features for fault pattern expression, a DET is employed to reduce the dimensionality of the feature space. Finally, a SVR-based generic multi-class solver is proposed to identify the different fault patterns of rotating machinery. The effectiveness of the proposed intelligent fault diagnosis scheme is validated separately using datasets from bearing and gearbox test rigs. In addition, the effects of different wavelet basis functions on the performance of the proposed scheme are investigated experimentally. The results demonstrate that the proposed intelligent fault diagnosis scheme is highly accurate in differentiating the fault patterns of both bearings and gears.     

A sequential fuzzy diagnosis method for rotating machinery using ant colony optimization and possibility theory

This study proposes a novel intelligent fault diagnosis method for rotating machinery using ant colony optimization (ACO) and possibility theory. The non-dimensional symptom parameters (NSPs) in the frequency domain are defined to reflect the features of the vibration signals measured in each state. A sensitive evaluation method for selecting good symptom parameters using principal component analysis (PCA) is proposed for detecting and distinguishing faults in rotating machinery. By using ACO clustering algorithm, the synthesizing symptom parameters (SSP) for condition diagnosis are obtained. A fuzzy diagnosis method using sequential inference and possibility theory is also proposed, by which the conditions of the machinery can be identified sequentially. Lastly, the proposed method is compared with a conventional neural networks (NN) method. Practical examples of diagnosis for a V-belt driving equipment used in a centrifugal fan are provided to verify the effectiveness of the proposed method. The results verify that the faults that often occur in V-belt driving equipment, such as a pulley defect state, a belt defect state and a belt looseness state, are effectively identified by the proposed method, while these faults are difficult to detect using conventional NN.     

基于假设检验和支持向量机的旋转机械故障诊断方法

针对旋转机械故障诊断中存在的早期非平稳微弱故障信号特征提取困难、故障诊断不准确等问题,提出了一种基于自适应假设检验滤波和支持向量机（SVM）的故障诊断方法。该方法采用统计学假设检验原理来评估参考信号(噪声信号)和原始信号(故障信号)在频域上的相似性,删除具有高相似性的频域成分;通过粒子群优化算法获得最佳的显著性水平α;定义评估因子Ipq来评价假设检验滤波的效果。最后通过SVM来逐次诊断轴系构造异常。验证结果表明该方法能够有效地诊断出传动轴不对中和不平衡的故障类型。     

一种基于冲击信号聚类分析的轴承故障诊断方法

为了对旋转机械内轴承的运行状态进行故障监测和诊断,在对振动冲击信号进行分段截取的基础上,提出了基于分段信号时、频域特征提取结合模糊K聚类的滚动轴承故障诊断方法,并将该方法应用于NU205轴承故障诊断中。     

基于等角度采样的列车频带变化类故障诊断方法研究

列车轮对部件故障常会受到频带变化因素的影响,如采用一般旋转机械故障监测方法,因谱线整体迁移模糊,可能导致误诊或漏诊。针对这一问题,研究了各影响因素与故障特征之间的动态响应,基于故障机理建立了频带变化类故障的动力学模型,分析了影响因素下的故障特征。提出了基于等角度采样的监测诊断方法,结合列车轮对组件实际情况制定了监测诊断方案,并通过实例应用对该诊断方法的准确性进行了验证。      

模糊Kohonen神经网络的旋转机械故障诊断

针对传统故障诊断方法不能解决旋转机械故障诊断的模糊性问题,提出一种基于模糊Kohonen神经网络的故障诊断模型,通过模糊量化处理故障样本模式和在Kohonen网络中使用邻域函数自动调整权重程度的改进学习算法,较大提高了网络的学习速度和聚类能力,能对具有模糊性的复合故障进行诊断,是一种适合于复杂旋转机械故障诊断的有效可行的方法。     

Deep residual learning-based fault diagnosis method for rotating machinery

Effective fault diagnosis of rotating machinery has always been an important issue in real industries. In the recent years, data-driven fault diagnosis methods such as neural networks have been receiving increasing attention due to their great merits of high diagnosis accuracy and easy implementation. However, it is mostly difficult to fully train a deep neural network since gradients in optimization may vanish or explode during back-propagation, which results in deterioration and noticeable variance in model performance. In fault diagnosis researches, larger data sequence of machinery vibration signal containing sufficient information is usually preferred and consequently, deep models with large capacity are generally adopted. In order to improve network training, a residual learning algorithm is proposed in this paper. The proposed architecture significantly improves the information flow throughout the network, which is well suited for processing machinery vibration signal with variable sequential length. Little prior expertise on fault diagnosis and signal processing is required, that facilitates industrial applications of the proposed method. Experiments on a popular rolling bearing dataset are implemented to validate the proposed method. The results of this study suggest that the proposed intelligent fault diagnosis method for rotating machinery offers a new and promising approach.     

Transient modeling and parameter identification based on wavelet and correlation filtering for rotating machine fault diagnosis

At constant rotating speed, localized faults in rotating machine tend to result in periodic shocks and thus arouse periodic transients in the vibration signal. The transient feature analysis has always been a crucial problem for localized fault detection, and the key aim for transient feature analysis is to identify the model and its parameters (frequency, damping ratio and time index) of the transient, and the time interval, i.e. period, between transients. Based on wavelet and correlation filtering, a technique incorporating transient modeling and parameter identification is proposed for rotating machine fault feature detection. With the proposed method, both parameters of a single transient and the period between transients can be identified from the vibration signal, and localized faults can be detected based on the parameters, especially the period. First, a simulation signal is used to test the performance of the proposed method. Then the method is applied to the vibration signals of different types of bearings with localized faults in the outer race, the inner race and the rolling element, respectively, and all the results show that the period between transients, representing the localized fault characteristic, is successfully detected. The method is also utilized in gearbox fault diagnosis and the effectiveness is verified through identifying the parameters of the transient model and the period. Moreover, it can be drawn that for bearing fault detection, the single-side wavelet model is more suitable than double-side one, while the double-side model for gearbox fault detection. This research proposed an effective method of localized fault detection for rotating machine fault diagnosis through transient modeling and parameter detection.     

基于多元经验模态分解的旋转机械早期故障诊断方法*

针对旋转机械早期微弱故障诊断问题,提出了基于多元经验模态分解的旋转机械早期故障诊断新方法。首先将多个加速度传感器合理布置在轴承座的关键位置,同步采集多通道振动信息;再利用多元经验模态分解同时对多通道振动信号进行自适应分解,得到一系列多元IMF分量;最后,依据峭度准则和相关系数从中选取包含故障主要信息的IMF分量进行信号重构,提取故障特征。多元经验模态分解方法克服了EMD等方法在进行多通道数据融合时缺乏理论依据的局限性。仿真信号和旋转机械故障信号的实验结果表明,该方法明显优于EEMD方法,对齿轮和滚动轴承故障的检测精度更高,可以在强背景噪声情况下更好地提取出故障冲击特征。     

旋转机械振动故障诊断中的几个实际例子

在旋转机械振动故障诊断研究中，提供丰富的案例是非常必要的，既可为理论研究提供素材，又可支持理论研究，还可作为广大实际测试人员的参考资料。本文介绍了旋转机械振动故障诊断实践中的几个典型例子，包括联轴器故障的振动频域特征，滑动轴承及滚动轴承某些故障的振动特征，由转子严重失衡引起的１／２亚谐波共振问题等。上述问题对现场故障诊断人员具有一定的参考价值     

Fault diagnosis of rotating machine by thermography method on support vector machine

Feature-based classification techniques consist of data acquisition, preprocessing, feature representation, feature calculation, feature selection, and classifiers. They are useful for online, real-time condition monitoring and fault diagnosis / features, which are now available with the development of information technologies and various measurement techniques. In this paper, an intelligent feature-based fault diagnosis is suggested, developed, and compared with vibration signals and thermal images. Fault diagnosis is performed using thermal imaging along with support vector machine (SVM) classification to simulate machinery faults, resulting in an accuracy level comparable to vibration signals. The observed results show that fault diagnosis using thermal images for rotating machines can be applied to industrial areas as a novel intelligent fault diagnostic method with plausible accuracy. It can be also proposed as a unique non-contact method to analyze rotating systems in mass production lines within a short time.     

SEQUENTIAL DIAGNOSIS FOR A CENTRIFUGAL PUMP BASED ON FUZZY NEURAL NETWORK

A sequential diagnosis method is proposed based on a fuzzy neural network realized by ＂the partially-linearized neural network （PNN）＂, by which the fault types of rotating machinery can be precisely and effectively distinguished at an early stage on the basis of the possibilities of symptom parameters. The non-dimensional symptom parameters in time domain are defined for reflecting the features of time signals measured for the fault diagnosis of rotating machinery. The synthetic detection index is also proposed to evaluate the sensitivity of non-dimensional symptom parameters for detecting faults. The practical example of condition diagnosis for detecting and distinguishing fault states of a centrifugal pump system, such as cavitation, impeller eccentricity which often occur in a centrifugal pump system, are shown to verify the efficiency of the method proposed in this paper.