Weighted sparsity-based denoising for extracting incipient fault in rolling bearing

Given that the incipient fault is too weak for extraction, a novel approach that is based on sparse optimization is proposed for incipient fault diagnosis. The proposed optimization method consists of three steps: First, autocorrelation analysis is utilized to filter broadband random noise. Then, the weighted sparsity-based denoising method is proposed to extract periodic impulses. The prior knowledge that periodic impulses are sparse is used to constitute a penalty term; thus a novel weighted sparse optimization model is established. The majorization-minimization method is used to solve the optimization model. The high-pass filter in quadratic fidelity term is constructed by a Butterworth filter based on banded matrices, thus effectively improving computational efficiency. Lastly, the interval of periodic impulses, which corresponds to the fault frequency of rolling bearing, is obtained. Moreover, simulation and experimental results show that the proposed approach can successfully extract fault features from the signals of low signal to noise ratio.     

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.     

Use of autocorrelation of wavelet coefficients for fault diagnosis

This paper presents a novel time–frequency-based feature recognition system for gear fault diagnosis using autocorrelation of continuous wavelet coefficients (CWC). Furthermore, it introduces an original mathematical approximation of gearbox vibration signals which approximates sinusoidal components of noisy vibration signals generated from gearboxes, including incipient and serious gear failures using autocorrelation of CWC. First, the drawbacks of the continuous wavelet transform (CWT) have been eliminated using autocorrelation function. Secondly, the autocorrelation of CWC is introduced as an original pattern for fault identification in machine condition monitoring. Thirdly, a sinusoidal summation function consisting of eight terms was used to approximate the periodic waveforms generated by autocorrelation of CWC for normal gearboxes (NGs) as well as occurrences of incipient and severe gear fault (e.g. slight-worn, medium-worn, and broken-tooth gears). In other words, the size of vibration signals can be reduced with minimal loss of significant frequency content by means of the sinusoidal approximation of generated autocorrelation waveforms of CWC as reported in this paper.     

基于IEWT和MCKD的滚动轴承故障诊断方法

针对经验小波变换（Empirical wavelet transform,EWT）对强噪声环境中滚动轴承微弱故障诊断的不足,主要是傅里叶频谱分段不当的问题。提出一种基于最大相关峭度解卷积（Maximum correlated kurtosis deconvolution,MCKD）降噪与改进EWT相结合的滚动轴承早期故障识别方法。首先采用最大相关峭度解卷积算法以包络谱的相关峭度最大化为目标对原信号进行降噪处理、检测信号中的周期性冲击成分,然后根据信号Fourier频谱的包络极大值进行分段,通过分析各频段平方包络谱中明显的频率成分来诊断故障。新方法能有效降噪、增强信号中周期性冲击特征、降低单次偶然冲击的影响、抑制非冲击成分。通过对含外圈、内圈故障的滚动轴承进行试验分析,结果表明,相比于快速谱峭度图和小波包络分析方法,该方法提取出的特征更加明显,能有效实现滚动轴承早期微弱故障的识别。     

Blind vibration component separation and nonlinear feature extraction applied to the nonstationary vibration signals for the gearbox multi-fault diagnosis

Fault diagnosis of gearboxes, especially the gears and bearings, is of great importance to the long-term safe operation. An unexpected damage on the gearbox may break the whole transmission line down. It is therefore crucial for engineers and researchers to monitor the health condition of the gearbox in a timely manner to eliminate the impending faults. However, useful fault detection information is often submerged in heavy background noise. Thereby, a new fault detection method for gearboxes using the blind source separation (BSS) and nonlinear feature extraction techniques is presented in this paper. The nonstationary vibration signals were analyzed to reveal the operation state of the gearbox. The kernel independent component analysis (KICA) algorithm was used hereby as the BSS approach for the mixed observation signals of the gearbox vibration to discover the characteristic vibration source associated with the gearbox faults. Then the wavelet packet transform (WPT) and empirical mode decomposition (EMD) nonlinear analysis methods were employed to deal with the nonstationary vibrations to extract the original fault feature vector. Moreover, the locally linear embedding (LLE) algorithm was performed as the nonlinear feature reduction technique to attain distinct features from the feature vector. Lastly, the fuzzy k-nearest neighbor (FKNN) was applied to the fault pattern identification of the gearbox. Two case studies were carried out to evaluate the effectiveness of the proposed diagnostic approach. One is for the gear fault diagnosis, and the other is to diagnose the rolling bearing faults of the gearbox. The nonstationary vibration data was acquired from the gear and rolling bearing fault test-beds, respectively. The experimental test results show that sensitive fault features can be extracted after the KICA processing, and the proposed diagnostic system is effective for the multi-fault diagnosis of the gears and rolling bearings. In addition, the proposed method can achieve higher performance than that without KICA processing with respect to the classification rate.     

A novel method for real time gear fault detection based on pulse shape analysis

Early identification of faults in gearboxes is a challenging task, especially when the time is a critical factor. In this paper, a novel method for real time fault detection in gearboxes is proposed using adaptive features extraction algorithm to deal with non-stationary faulty signals. Moreover, integration of different techniques is presented in order to detect faults in a real time environment. Evolutionary algorithms are commonly used in different applications and have strong ability for optimization. However, they are inherently slow and not suitable for real time applications. The proposed method is based on a combination of conventional one-dimensional and multi-dimensional search methods, which showed high performance and accurate fault detection results compared with evolutionary algorithms. The effectiveness, feasibility and robustness of the proposed method have been demonstrated on experimental data. An average speed up factor of 87% has been successfully achieved with approximately 5% quality degradation in the results as compared with evolutionary algorithms like genetic algorithms.     

Separation method of rolling bearing compound fault characteristics based on Improved harmonic wavelet packet decomposition and fast ICA

Compound fault characteristics in single-channel vibration signals of rolling bearings are difficult to separate. On the basis of improved harmonic wavelet packet decomposition and fast independent component analysis (FICA), this study proposes a new method to address this problem. First, a series of mutually independent frequency bands are obtained after harmonic wavelet packet decomposition of the initial vibration signal to satisfy the requirement that the number of observed signals must be larger than the number of source signals in the FICA algorithm. Second, the optimal frequency bands are selected based on the maximum kurtosis index and used as the input matrix of the FICA algorithm to separate the compound fault characteristics further. Lastly, accurate separation and extraction of the compound fault characteristics of the rolling bearings are realized. Results show that the proposed method can effectively separate the compound fault characteristics in the single-channel vibration signals of the bearings.

     

Improved CICA algorithm used for single channel compound fault diagnosis of rolling bearings

A Compound fault signal usually contains multiple characteristic signals and strong confusion noise, which makes it difficult to separate week fault signals from them through conventional ways, such as FFT-based envelope detection, wavelet transform or empirical mode decomposition individually. In order to realize single channel compound fault diagnosis of bearings and improve the diagnosis accuracy, an improved CICA algorithm named constrained independent component analysis based on the energy method (E-CICA) is proposed. With the approach, the single channel vibration signal is firstly decomposed into several wavelet coefficients by discrete wavelet transform(DWT) method for the purpose of obtaining multichannel signals. Then the envelope signals of the reconstructed wavelet coefficients are selected as the input of E-CICA algorithm, which fulfills the requirements that the number of sensors is greater than or equal to that of the source signals and makes it more suitable to be processed by CICA strategy. The frequency energy ratio(ER) of each wavelet reconstructed signal to the total energy of the given synchronous signal is calculated, and then the synchronous signal with maximum ER value is set as the reference signal accordingly. By this way, the reference signal contains a priori knowledge of fault source signal and the influence on fault signal extraction accuracy which is caused by the initial phase angle and the duty ratio of the reference signal in the traditional CICA algorithm is avoided. Experimental results show that E-CICA algorithm can effectively separate out the outer-race defect and the rollers defect from the single channel compound fault and fulfill the needs of compound fault diagnosis of rolling bearings, and the running time is 0.12% of that of the traditional CICA algorithm and the extraction accuracy is 1.4 times of that of CICA as well. The proposed research provides a new method to separate single channel compound fault signals.     

基于ICA和小波变换的滚动轴承故障诊断方法研究

提出了结合独立分量分析(ICA)和小波变换进行滚动轴承故障诊断的方法。在设计的系统平台上,首先对冲击脉冲信号进行预处理,使信号较好地满足独立分量分析的前提条件。然后,应用独立分量快速算法分离故障轴承的冲击脉冲信号,通过小波快速算法完成信号重构,实现滚动轴承故障的识别。实验结果表明,利用独立分量分析方法提取的故障状态特征向量与小波快速算法相结合可以有效、准确地识别滚动轴承的故障信号。     

基于多普勒调制时移 Laplace 小波的列车轴承故障声 信号瞬态成分快速提取方法

提出了一种基于多普勒调制时移Laplace小波的列车轴承故障声信号瞬态成分快速提取方法,包含“先粗后精”两个步骤：1)瞬态参数粗估计,利用现有的多普勒调制等周期Laplace小波模型粗略估计瞬态参数;2)参数精确估计与瞬态成分提取,构造多普勒调制时移Laplace小波模型,使用逐个匹配的策略进行瞬态参数精确估计和瞬态成分的提取。所提方法具有以下优点：1)更高的精度,使用的多普勒调制时移Laplace小波模型在时域内仅有一个时延参数定位的小波成分,能够解决周期瞬态模型在提取伪周期瞬态成分时匹配误差问题;2)高效率,由于使用了周期瞬态模型粗略估计瞬态成分参数,因此在瞬态成分逐个提取的过程中小波参数的范围可以设的很小,实验对比分析结果显示,与直接提取方式相比效率提高了71.46%。本研究提供了一种从含有多普勒畸变的列车轴承故障声信号中精确地、高效率地提取瞬态成分的方法。     

应用分层自适应匹配追踪重构MEMS陀螺信号

对含噪微机械系统(MEMS)陀螺信号进行小波分解重构时,真实信号对应的小波系数很难选取,故本文提出一种分层自适应匹配追踪算法(LAMP)来解决上述问题。建立了含噪MEMS陀螺信号中信号小波系数稀疏提取架构,将信号小波系数提取问题转化为含噪信号中信号小波系数稀疏性的恢复问题。比较已有稀疏重构算法,采用一种新的LAMP算法,在各种可能的小波系数组合中挑选出分解系数最为稀疏的一组,以此消除信号中的噪声小波系数,进而重构MEMS陀螺信号。实验表明:提出的LAMP算法的稀疏重构效果优于其他迭代贪婪重构算法;基于LAMP的信号稀疏小波重构方法,可以有效去除MEMS陀螺信号的大量噪声;去噪前后,纯MEMS陀螺数据解算的方位角平均累积误差由10.060 2(°)/h减小到5.034 6(°)/h,优于传统小波阈值重构法平均累积误差8.596 8(°)/h,显示了较好的应用效果。     

Wavelet design for extracting weak fault feature based on lifting scheme

Weak fault features of mechanical signals are usually immersed in noisy signals. A new wavelet method based on lifting scheme to match weak fault characteristics is proposed. In this method, an initial set of finite biorthogonal filters is modified by a lifting and dual lifting procedure alternately, and different lifting operators and dual lifting operators are obtained. The properties of the initial wavelet is improved, and the new wavelet with particular properties is designed. Simulation and engineering results confirm that the proposed method is better than other wavelet methods for extracting weak fault feature. Modulus maxima of the detail signal in every operation cycle are extracted, the position and time that weak signal singularity occurs are clearly found, and slight rub-impact fault caused by axis misalignment and rotor imbalance of a heavy oil catalytic cracking set are desirably extracted. extracted. __________ Translated from Journal of Xi’an Jiaotong University, 2005, 39(5) (in Chinese)     

Particle swarm optimization algorithm to solve the deconvolution problem for rolling element bearing fault diagnosis

Extraction of the fault related impulses from the raw vibration signal is important for rolling element bearing fault diagnosis. Deconvolution techniques, such as minimum entropy deconvolution (MED), MED adjusted (MEDA) and maximum correlated kurtosis deconvolution (MCKD), optimal MED adjusted (OMEDA) and multipoint optimal MED adjusted (MOMEDA), are typical techniques for enhancing the impulse-like component in the fault signal. This paper introduces the particle swarm optimization (PSO) algorithm to solve the filter of deconvolution problem. The proposed approaches solve the filter coefficients of the deconvolution problems by the PSO algorithm, assisted by a generalized spherical coordinate transformation. Compared with MED, MEDA, and OMEDA, the proposed PSO-MED and PSO-OMEDA can effectively overcome the influence of large random impulses and tend to deconvolve a series of periodic impulses rather than a signal impulse. Compared with MCKD and MOMEDA, the proposed PSO-MCKD and PSO-MOMEDA can achieve good performances even when the fault period is inaccurate. The effectiveness of the proposed methods is validated by the simulated signals. The study of experimental bearing fault signal shows that the PSO based deconvolution methods delivered better performance for rolling element bearing fault detection than the traditional deconvolution methods. Additionally, the proposed methods are compared with the following two popular signal processing methods: the ensemble empirical mode decomposition (EEMD) and fast kurtogram, which are used to highlight the improved performance of the proposed methods.     

基于自适应随机共振和稀疏编码收缩算法的齿轮故障诊断方法

针对强背景噪声下齿轮故障冲击特征提取问题,提出了一种基于自适应随机共振和稀疏编码收缩算法的齿轮故障诊断方法。该方法选用相关峭度作为随机共振检测周期性冲击分量的测度函数,借助遗传算法实现信号中周期性冲击特征的自适应提取;在此基础上,利用稀疏编码收缩算法对随机共振检测结果做进一步降噪处理,从而凸显冲击特征,提高故障识别精度。试验和工程实例分析结果表明,该方法可实现齿轮故障冲击特征的增强提取,为齿轮故障诊断提供依据。     

基于PPCA-EWT的滚动轴承轻微故障诊断

针对经验小波变换(empirical wavelet transform,简称EWT)在强背景噪声下对轴承的轻微故障特征提取不足的问题,提出了概率主成分分析(probabilistic principal component analysis,简称PPCA)结合EWT的滚动轴承轻微故障诊断方法。首先,对信号做PPCA预处理,提取信号主要故障特征成分,去除强背景噪声干扰;然后,采用EWT方法分解轴承故障信号,按相关系数-峭度准则选出故障特征较为明显的分量,并将所选分量重构故障信号;最后,对信号采取包络分析,提取出轴承故障特征。仿真和实验结果表明,该方法能够有效地诊断出轴承故障且效果优于对信号进行EWT包络分析。     

基于Morlet小波的滚动轴承故障信号奇异性分析

针对滚动轴承故障信号分析中单一频域表征的问题,提出了将Morlet连续小波变换应用于故障信号奇异性提取和分析的新方法。在分析了滚动轴承故障信号的奇异性特征和奇异性信号小波检测机理的基础上,将Morlet连续小波用于对滚动轴承故障信息的提取与分析。试验证明,该方案能有效地对滚动轴承故障信号在时间和尺度平面进行分析,可以同时表征奇异性信号的时间和频率信息。     

A joint adaptive wavelet filter and morphological signal processing method for weak mechanical impulse extraction

Periodical impulses are vital indicators of rotating machinery faults. Therefore, the extraction of weak periodical impulses from vibration signals is of great importance for incipient fault detection. However, measured signals are often severely tainted by various noises, which makes the detection of impulses rather difficult. As such, a proper signal processing technique is necessary. In this paper, a hybrid method comprised of wavelet filter and morphological signal processing (MSP) is proposed for this task. The wavelet filter is used to eliminate the noise and enhance the impulsive features. Then, the filtered signal is processed by the morphological closing operator and a local maximum algorithm to isolate periodical impulses. To select the proper parameters of the joint approach, i.e., the center frequency, the bandwidth of wavelet filter, and the length of flat structuring elements (SE), a novel optimization algorithm based on differential evolution (DE) is developed. The results of simulated experiments and bearing vibration signal analysis verify the effectiveness of the proposed method.     

[设备健康管理]基于深度学习多样性特征提取与信息融合的行星齿轮箱故障诊断方法

针对行星齿轮箱振动信号噪声干扰大、单一分类器泛化能力不强的问题,提出了一种基于深度学习多样性特征提取与信息融合的行星齿轮箱故障诊断方法。利用多目标优化算法优化多个堆栈去噪自动编码器(SDAE)以获得多个性能优异的SDAE,并提取多样性的故障特征;采用多响应线性回归模型集成多样性故障特征实现信息融合,得到多目标集成堆栈去噪自动编码器(MO-ESDAE),最后将其应用于行星齿轮箱故障诊断。实验结果表明：该方法能有效提高故障诊断精度与稳定性,具有较强的泛化能力。     

Separation of fault features from a single-channel mechanical signal mixture using wavelet decomposition

This study reports a joint wavelet decomposition and Fourier transform approach to the separation of periodic mechanical source signals from single-channel signal mixture. With this method, the signal mixture is first decomposed to certain wavelet scales. The resulting wavelet coefficients are then Fourier transformed to extract the information pertinent to each signal source from these scales. Next, the number of signal sources is determined and the wavelet coefficients for each signal are constructed in all scale levels. Finally the source signals can be reconstructed using these wavelet coefficients. Since this method does not require the number of sources to be known a priori, it is particularly suitable for mechanical fault signal separation as the number of source signals varies with time and is unpredictable. It is also important to point out that the number of sources is determined without the commonly adopted sequential extraction/learning process and hence the proposed method can be used for on-line fault detection due to the reduced computing burden. The application of this method has been demonstrated using mixed bearing data containing both inner and outer race fault signals.     

A method for the compound fault diagnosis of gearboxes based on morphological component analysis

An improved morphological component analysis (MCA) method is proposed for the compound fault diagnosis of gearboxes. When gear fault and bearing fault occur simultaneously, the compound fault signal of the gearbox contains meshing components (related to the gear fault) and periodic impulse components (related to the bearing fault). The corresponding fault characteristics can be separated by MCA according to the morphological differences of the components. In the proposed method, the optimal dictionary, which can represent the characteristics of bearing faults, is first selected based on the principle of minimum information entropy. Then, the compound fault signal is decomposed into the meshing component and the periodic impulse component using MCA. Finally, the separated components are subjected to the Hilbert envelope spectrum analysis. The faults of the gear and the bearing can be diagnosed according to the envelope spectra of the separated fault signal components. Simulation and experimental studies validate the effectiveness of the proposed method for the compound fault diagnosis of gearboxes.