Gearbox Fault Feature Extraction Research Articles

Noise reduction method for wind turbine gearbox vibration signals based on CVMD-DRDSAE

Abstract Wind turbine gearbox fault feature extraction is difficult due to strong background noise. To address this issue, a noise reduction method combining comprehensive learning particle swarm optimization-variational mode decomposition (CLPSO-VMD) and deep residual denoising self-attention autoencoder (DRDSAE) is proposed. Firstly, the proposed CLPSO-VMD algorithm is used to decompose the noisy wind turbine gearbox vibration signals. Subsequently, the intrinsic mode functions highly correlated with the original signals are selected through the Spearman correlation coefficient and utilized for signal reconstruction, thereby filtering out high-frequency noise outside the fault frequency band in the frequency domain characterization. Secondly, the improved DRDSAE is utilized to learn the latent representations of data in the first-level denoised signal, further reducing the residual noise within the fault frequency band while retaining important signal features. Finally, the envelope spectrum highlights the weak feature of the wind turbine gearbox vibration signal. Experimental results demonstrate the effectiveness of the proposed method in denoising wind turbine gearbox vibration signals under strong noise.

基于图滤波理论的齿轮箱故障特征提取应用研究

基于图滤波理论的齿轮箱故障特征提取应用研究

Gearbox Fault Diagnosis Based on Improved Variational Mode Extraction.

Gearboxes are widely used in drive systems of rotating machinery. The health status of gearboxes considerably influences the normal and reliable operation of rotating machinery. When a gearbox experiences tooth failure, a vibration signal with impulse features is excited. However, these impulse features tend to be relatively weak and difficult to extract. To solve this problem, a novel approach for gearbox fault feature extraction and fault diagnosis based on improved variational mode extraction (VME) is proposed. Since the initial value of the desired mode center frequency and the value of the penalty parameter in VME must be assigned, a short-time Fourier transform (STFT) was performed, and a new index, the standard deviation of differential values of envelope maxima positions (SDE), is proposed. The feasibility and effectiveness of the proposed approach was verified by a simulation signal and two datasets associated with a gearbox test bench. The results demonstrate that the VME-based approach outperforms the variational mode decomposition (VMD) approach.

A gearbox fault feature extraction method based on wingsuit flying search algorithm-optimized orthogonal matching pursuit with a compound time-frequency atom dictionary

A gearbox fault feature extraction method based on wingsuit flying search algorithm-optimized orthogonal matching pursuit with a compound time-frequency atom dictionary

Gearbox Fault Feature Extraction and Diagnosis

Gearboxes, as an essential connection in mechanical equipment as well as a universal component for transmitting power, have a wide range of applications in modern industrial development. Gear systems can be damaged to different degrees during operation, such as gear breakage, pitting, wear, etc., all of which generate periodic impulse impact force and produce the modulation phenomenon of vibration signal, and extract modulation information from the signal and analyze its intensity and frequency to determine the degree and parts of the parts damage. This paper mainly discusses the analysis and processing of the vibration data of gears by MATLAB, based on the snowflake diagram obtained by analyzing the SDP (Symmetrized Dot Pattern) method, the images are filtered and binarized separately, and the images obtained are compared with the similarity with the standard images, thus greatly leveling the fault identification purpose.

Application of a Novel Improved Adaptive CYCBD Method in Gearbox Compound Fault Diagnosis

Recently, cyclostationary blind deconvolution (CYCBD) is often used in gearbox fault feature extraction, and it is more effective in recovering single periodic pulse. However, in the engineering application of CYCBD, the filter length ( $L$ ) and cycle frequency ( $\alpha$ ) need to be verified and set through a large number of experiments, and the efficiency is very low; Moreover, the effect is not good when it is used to extract composite fault features under the background of strong noise. In order to overcome the above limitations, empirical mode decomposition (EEMD) is used to preprocess the composite fault. EEMD can remove the high-frequency noise and weak correlation components in the sampled signal. The strong correlation component is reconstructed to obtain the modal function closer to the fault frequency. The chimpanzee intelligent algorithm is applied to the determination of $L$ and $\alpha $ of CYCBD by optimization to form an adaptive CYCBD. Adaptive CYCBD takes the dispersion entropy of envelope spectrum as the fitness function of chimpanzee optimization algorithm (CHOA), and finds the optimal $L$ and $\alpha $ through iteration. The optimal parameter value is applied to CYCBD, and the reconstructed modal function is deconvoluted to obtain the optimal inverse filter, so as to accurately separate the fault characteristic components. Simulation and experimental results show that this method is effective for gearbox composite fault diagnosis and extraction under strong noise background.

Application of a New Enhanced Deconvolution Method in Gearbox Fault Diagnosis

When the mechanical transmission mechanism fails, such as gears and bearings in the gearbox, its vibration signal often appears as a periodic impact. Considering the influence of noise, however, the fault signal is often submerged in the noise, so it is necessary to propose a feasible and effective fault extraction method. MOMEDA (multipoint optimal minimum entropy deconvolution adjusted) overcomes the tedious iterative process of MED (minimum entropy deconvolution) and overcomes the resampling trouble in MCKD (maximum correlated kurtosis deconvolution). It is suitable for dealing with periodic impact signal. Besides, aiming at the poor ability of MOMEDA to capture the deconvolution result of target function in a strong noise environment, this paper proposes an improved MOMEDA gearbox fault feature extraction method. Considering that MOMEDA has poor anti-noise performance and can easily cause misdiagnosis in a strong noisy environment, this paper constructs an autoregressive mean sliding model to improve the noise immunity of MOMEDA. Firstly, the stability of the test signal is judged by the autocorrelation coefficient (ACF) and the partial correlation coefficient (PACF). Secondly, the ARMA (autoregressive moving average) model is constructed and a set of optimal model coefficients are obtained to filter the signal, which greatly improves MOMEDA’s ability to capture fault features. Thirdly, the fault feature is extracted by MOMEDA, and the fault information is extracted accurately under a strong noise environment. Finally, compared with AR-MED, ARMAMED, and other methods, the advantages of ARMAMOMEDA are verified. Moreover, the effectiveness and superiority of the proposed method are verified by simulation signals and experimental data from the Case Western Reserve University Bearing Data Center.

Fault feature extraction of wind turbine gearbox under variable speed based on improved adaptive variational mode decomposition

This paper proposes a study on gearbox fault feature extraction of wind turbine under variable speed condition using improved adaptive variational mode decomposition (VMD). Frequent changes in wind speed and critical noise interference make the vibration signal exhibit non-stationary characteristics. Although computational order tracking can transform non-stationary signals in the time domain into stationary angular signals, and then extract fault features from order spectrum obtained by FFT, the obtained order spectrum is liable to be polluted by noise or to appear order aliasing. To avoid order aliasing and eliminate noise interference, the original non-stationary signal is firstly processed using the improved adaptive VMD method called adaptive differential evolution – VMD (ADE-VMD). ADE-VMD can not only utilise the advantages of traditional VMD but also adaptively select narrow-band intrinsic mode function (NBIMF) to construct the reconstructed signal with less noise and without order aliasing. In the experiment, we compared the ADE-VMD method with other VMD methods such as GA-VMD, PSO-VMD and DE-VMD, and the results showed that ADE-VMD has excellent adaptive processing ability, and its convergence and optimisation speed are more remarkable. ADE-VMD can effectively filter the noise inference and avoid the order aliasing, so it is well suitable for fault feature extraction under variable speed.

A Hybrid Approach for Weak Fault Feature Extraction of Gearbox

A novel hybrid fault diagnosis method based on ensemble empirical mode decomposition and weighted adaptive multi-scale morphological analysis (WAMMA) is proposed to detect the early damage of gearboxes. In this method, we propose a characteristic frequency ratio (CFR) method to determine the weighted coefficient for each scale of AMMA. First, multiple scales are obtained using the AMMA method. Second, the weighted coefficient of each scale in the AMMA method is calculated using the CFR. Third, the final results can be obtained by multiplying the weighted coefficients and filtering results with all scales. Since the performance of each scale of AMMA is evaluated using the CFR, the demodulation ability can be effectively improved. However, the WAMMA is easily disturbed by heavy noise when extracting early fault feature directly. A method combined EEMD with the WAMMA is proposed. The effectiveness of the proposed method has been verified using two experimental vibration signals of gearboxes. The results demonstrate that the proposed method has a superior performance in the extraction of weak fault characteristics of gearboxes in comparison with the WAMMA and EEMD-AMMA methods.

Application of an Improved Multipoint Optimal Minimum Entropy Deconvolution Adjusted for Gearbox Composite Fault Diagnosis.

The fault feature extraction of gearbox is difficult to achieve under complex working conditions, and this paper presents a hybrid fault diagnosis method for gearbox based on the combining product function (CPF) and multipoint optimal minimum entropy deconvolution adjusted (MOMEDA) methods. First, ensemble local mean decomposition (ELMD) is utilized to reduce the noise in original signal, and get a series of product functions (PFs), through the correlation coefficient method to remove false components and residual components. Then, multi-point kurtosis of the definition is achieved by calculating the multi-point kurtosis spectrum of each layer PF, and the fault feature period is extracted and the PFs without periodic impact are removed. After that, in order to maintain the integrity of the original signal, the PFs with the same period are recombined by the combined product function method. Finally, the different cycle interval is configured, reduce the noise through MOMEDA on the combined signal, to further extract the fault feature. The method is applied to the feature extraction of gear box composite fault to verify the feasibility of this method.

EEMD-Based cICA method for single-channel signal separation and fault feature extraction of gearbox

This paper proposes a novel fault feature extraction method with the aim of extracting the fault feature submerged in the single-channel observation signal. The proposed method integrates the strengths of the constrained independent component analysis (cICA) extracting only the signals of interest (SOIs) with the advantage of ensemble empirical mode decomposition (EEMD) alleviating the mode mixing. The method, which is named EEMD-based cICA, not only enables gear fault feature extraction but also offers a new independent component analysis (ICA) mixing model with source noise and measured noise for the single-channel observation signal. The efficiency of the proposed method is tested on simulated as well as real-world vibration signals acquired from a multi-stage gearbox with a missing tooth and a chipped tooth, respectively.

Mine gearbox fault diagnosis based on multiwavelets and maximum correlated kurtosis deconvolution

As the mine gearbox working conditions are poor, the fault signal is weak and usually drowned in background noise when gearbox occurring fault. The fault feature is very difficult to extract. Aiming at solving this problem, this paper proposed a mine gearbox fault feature extraction method which combines multiwavelets decomposition with maximum correlated kurtosis deconvolution (MCKD).The component of multiwavelets decomposing was processed by MCKD method, MCKD suppress the noise in the signal and enhance the weak impact feature of fault signal, the envelope of its deconvolution signal was calculated, then the fault could be judged by analyzing the prominent frequency component of envelope spectrum. Thus, the experiment analysis and engineering application verify the effectiveness of the proposed method.

Application of EMD-WVD and particle filter for gearbox fault feature extraction and remaining useful life prediction

Fault feature extraction and remaining useful life (RUL) prediction are important to condition based maintenance (CBM). In order to realize the fault feature extraction of gearbox vibration signal presenting nonlinear and non-Gaussian, the integration of empirical mode decomposition (EMD) and Wigner-Ville distribution (WVD) are proposed in this paper. Taking the kurtosis as standard, the WVD is applied to some IMFs with larger kurtosis to calculate the time-frequency distribution, with an effective suppress on mode mixing and the cross-term interference. Afterwards, particle filter (PF) with the state space model based on Wiener process is proposed to predict the RUL of gearbox considering degradation feature, gearbox teeth wear and nonlinear and non-Gaussian system. The gearbox life cycle test shows that the EMD-WVD method can extract the valued characteristics of vibration signal accurately, and the particle filter can provide an effective way to predict the RUL of gearbox.

A novel gearbox fault feature extraction and classification using Hilbert empirical wavelet transform, singular value decomposition, and SOM neural network

There are growing demands for condition monitoring and fault diagnosis of rotating machinery to lower unscheduled breakdown. Gearboxes are one of the fundamental components of rotating machinery; their faults identification and classification always draw a lot of attention. However, non-stationary vibration signals and low energy of weak faults makes this task challenging in many cases. Thus, a new fault diagnosis method which combines the Hilbert empirical wavelet transform (HEWT), singular value decomposition (SVD), and self-organizing feature map (SOM) neural network is proposed in this paper. HEWT, a new self-adaptive time-frequency analysis was applied to the vibration signals to obtain the instantaneous amplitude matrices. Then, the singular value vectors, as the fault feature vectors were acquired by applying the SVD. Last, the SOM was used for automatic gearbox fault identification and classification. An electromechanical model comprising an induction motor coupled with a single stage spur gearbox is considered where the vibration signals of four typical operation modes were simulated. The conditions include the healthy gearbox, input shaft slant crack, tooth cracking, and tooth surface pitting. Obtained results show that the proposed method effectively identifies the gearbox faults at an early stage and realizes automatic fault diagnosis. Moreover, performance evaluation and comparison between the proposed HEWT–SVD method and Hilbert–Huang transform (HHT)–SVD approach show that the HEWT–SVD is better for feature extraction.

Fault diagnosis of gearbox using empirical mode decomposition and multi-fractal detrended cross-correlation analysis

A gearbox vibration signal is non-stationary and non-linear and has multiple components and multi-fractal properties, which make it difficult to effectively extract gearbox fault features. This paper proposes a method for gearbox fault feature extraction based on empirical mode decomposition (EMD) and multi-fractal detrended cross-correlation analysis (MFDCCA). First, EMD, a time-frequency analysis method, was employed to decompose the gearbox vibration signal into a number of intrinsic mode functions (IMFs). Second, the multi-fractal features hidden in the non-linear vibration signal were extracted by applying the MFDCCA to the selected major IMFs, thus highlighting the multi-fractality information which can be used to characterize different fault modes and severities of the gearbox. Third, for each IMF, three multi-fractal feature parameters sensitive to gearbox faults were selected from the multi-fractal features, further constructing the multi-fractal fault feature vector. Then, the principal component analysis (PCA) was introduced to reduce the dimensions of the extracted multi-fractal fault feature vectors and to enhance the accuracy of diagnosis. Finally, a radial basis function neural network was utilized to classify gearbox faults. Several commonly occurring faults were used to validate the proposed method in this study. Experimental results provide evidence that the extracted multi-fractal fault features can effectively distinguish different fault modes, even under slight variation in working conditions. Simultaneously, the results of comparison show that the performance of the proposed EMD-MFDCCA-PCA method outperforms that of EMD-MFDFA (multi-fractal detrended fluctuation analysis) combined with the traditional PCA.

Sparse representation of transients in wavelet basis and its application in gearbox fault feature extraction

Vibration signals from a defective gearbox are often associated with important measurement information useful for gearbox fault diagnosis. The extraction of transient features from the vibration signals has always been a key issue for detecting the localized fault. In this paper, a new transient feature extraction technique is proposed for gearbox fault diagnosis based on sparse representation in wavelet basis. With the proposed method, both the impulse time and the period of transients can be effectively identified, and thus the transient features can be extracted. The effectiveness of the proposed method is verified by the simulated signals as well as the practical gearbox vibration signals. Comparison study shows that the proposed method outperforms empirical mode decomposition (EMD) in transient feature extraction.

Sparsity-enabled signal decomposition using tunable Q-factor wavelet transform for fault feature extraction of gearbox

Localized faults in gearboxes tend to result in periodic shocks and thus arouse periodic responses in vibration signals. Feature extraction has always been a key problem for localized fault diagnosis. This paper proposes a new fault feature extraction technique for gearboxes by using sparsity-enabled signal decomposition method. The sparsity-enabled signal decomposition method separates signals based on the oscillatory behavior of the signal rather than the frequency or scale. Thus, the fault feature can be nonlinearly extracted from vibration signals. During the implementation of the proposed method, tunable Q-factor wavelet transform, for which the Q-factor can be easily specified, is adopted to represent vibration signals in a sparse way, and then morphological component analysis (MCA) is employed to estimate and separate the distinct components. The corresponding optimization problem of MCA is solved by the split augmented Lagrangian shrinkage algorithm (SALSA). With the proposed method, vibration signals of the faulty gearbox can be nonlinearly decomposed into high-oscillatory component and low-oscillatory component which is the fault feature of gearboxes. To evaluate the performance of the proposed method, this paper investigates the effect of two parameters pertinent to MCA and SALSA: the Lagrange multiplier and the penalty parameter. The effectiveness of the proposed method is verified by both the simulated and practical gearbox vibration signals. Results show the proposed method outperforms empirical mode decomposition and spectral kurtosis in extracting fault features of gearboxes.

Fault feature extraction of gearbox by using overcomplete rational dilation discrete wavelet transform on signals measured from vibration sensors

Gearbox fault diagnosis is very important for preventing catastrophic accidents. Vibration signals of gearboxes measured by sensors are useful and dependable as they carry key information related to the mechanical faults in gearboxes. Effective signal processing techniques are in necessary demands to extract the fault features contained in the collected gearbox vibration signals. Overcomplete rational dilation discrete wavelet transform (ORDWT) enjoys attractive properties such as better shift-invariance, adjustable time–frequency distributions and flexible wavelet atoms of tunable oscillation in comparison with classical dyadic wavelet transform (DWT). Due to these advantages, ORDWT is presented as a versatile tool that can be adapted to analysis of gearbox fault features of different types, especially in analyzing the non-stationary and transient characteristics of the signals. Aiming to extract the various types of fault features confronted in gearbox fault diagnosis, a fault feature extraction technique based on ORDWT is proposed in this paper. In the routine of the proposed technique, ORDWT is used as the pre-processing decomposition tool, and a corresponding post-processing method is combined with ORDWT to extract the fault feature of a specific type. For extracting periodical impulses in the signal, an impulse matching algorithm is presented. In this algorithm, ORDWT bases of varied time–frequency distributions and varied oscillatory natures are adopted, moreover an improved signal impulsiveness measure derived from kurtosis is developed for choosing optimal ORDWT bases that perfectly match the hidden periodical impulses. For demodulation purpose, an improved instantaneous time–frequency spectrum (ITFS), based on the combination of ORDWT and Hilbert transform, is presented. For signal denoising applications, ORDWT is enhanced by neighboring coefficient shrinkage strategy as well as subband selection step to reveal the buried transient vibration contents. The proposed fault feature extraction technique is applied in a range of engineering applications, and the processing results demonstrate that the ORDWT-based feature extraction technique successfully identifies the incipient fault features in the cases where DWT and empirical mode decomposition method are less effective.

Study and Improving on Adaptive Filter

In this paper the principle of adaptive filter and various least mean square (LMS) adaptive filter algorithm is studied, based on the related hyperbolic tangent function LMS algorithm is presented, referred to as CTanh-LMS algorithm. Simulation results show that, compared with other adaptive filter algorithm, this method has better denoising ability, and the algorithm is simple, fast convergence rate, and can satisfy the gearbox vibration signal denoising requirements. The proposed algorithm can not only solve the gearbox fault feature extraction, and give adaptive filter algorithm research provides a new means, has important theoretical significance and practical value.

Gearbox Fault Feature Extraction Using Hilbert Transform, S-Transform, and a Statistical Indicator

Abstract Vibration analysis has been widely used in machine fault feature extraction and diagnosis of rotating machinery. The effects of modulation and nonstationarity in vibration signals collected from a faulty gearbox present challenges for fault feature extraction. Hilbert transform and S-transform have the ability to address these issues through demodulation and time-frequency analysis, respectively. In this paper, we propose to use a feature indicator to represent the S-transform coefficients. The proposed method integrates the advantages of Hilbert transform, S-transform, and the proposed feature indicator (FI). The validity of the proposed method is verified with a simulated signal and real gearbox vibration signals. Comparison studies show that the proposed method is more effective and does not require the operators to have a lot of diagnostic experience.