Collected Vibration Signals Research Articles

Deep transfer multi-wavelet auto-encoder for intelligent fault diagnosis of gearbox with few target training samples

Lack of typical fault samples remains a huge challenge for intelligent fault diagnosis of gearbox. In this paper, a novel approach named deep transfer multi-wavelet auto-encoder is presented for gearbox intelligent fault diagnosis with few training samples. Firstly, new-type deep multi-wavelet auto-encoder is designed for learning important features of the collected vibration signals of gearbox. Secondly, high-quality auxiliary samples are selected based on similarity measure to well pre-train a source model sharing similar characteristics with the target domain. Thirdly, parameter knowledge acquired from the source model is transferred to target model using very few target training samples. Transfer diagnosis cases for different fault severities and compound faults of gearbox confirm the feasibility of the proposed approach even if the working conditions have significant changes.

Deep Laplacian Auto-encoder and its application into imbalanced fault diagnosis of rotating machinery

Generally, the measured health condition data from mechanical system often exhibits imbalanced distribution in real-world cases. To enhance fault diagnostic accuracy of the imbalanced data set, a novel rotating machinery fault imbalanced diagnostic approach based on Deep Laplacian Auto-encoder (DLapAE) is firstly developed in this paper. First of all, the collected vibration signals are immediately entered into the constructed DLapAE algorithm for layer-by-layer feature extraction, afterwards the extracted deep discriminative sensitive features are flowed into Back Propagation (BP) classifier for health condition diagnosis. More specifically, it is well worth mentioning that Laplacian regularization term can be reasonably added into the original objective function of Deep Auto-encoder (DAE) for smoothing the manifold structure of data in DLapAE. Namely, the proposed DLapAE algorithm with Laplacian regularization can improve the generalization performance of this fault diagnosis framework and make it more suitable for feature learning and classification of imbalanced data. Last but not least, two case of the experimental bearing systems can prove the effectiveness of proposed methodology. Compared with other existing fault diagnosis methods based on deep learning, the proposed fault diagnosis method can effectively implement the accurate fault diagnosis for rotating machinery balanced and imbalanced datasets.

A Novel Method for Identifying Crack and Shaft Misalignment Faults in Rotor Systems under Noisy Environments Based on CNN.

Crack and shaft misalignment are two common types of fault in a rotor system, both of which have very similar dynamic response characteristics, and the vibration signals are vulnerable to noise contamination because of the interaction among different components of rotating machinery in the actual industrial environment, resulting in great difficulties in fault identification of a rotor system based on vibration signals. A method for identification of faults in the form of crack and shaft misalignments is proposed in this paper, which combines variational mode decomposition (VMD) and probabilistic principal component analysis (PPCA) to denoise the collected vibration signals from a test rig and then achieve signal feature extraction and fault classification with convolutional artificial neural network (CNN). The key parameters of the CNN are optimized and determined by genetic algorithm (GA) firstly, and the domain adaptability of the trained network is verified by the signals with different signal-to-noise ratio (SNR) values; then, the noisy vibration signals are decomposed into multiple band-limited intrinsic modal functions by VMD, and further data dimension reduction is performed by PPCA to realize the separation of the useful signals from noise; finally, the crack and shaft misalignment of the rotor system are identified by the optimized CNN. The results show that the proposed method can effectively remove the interference noise and extract the intrinsic features of the vibration signals, and the recognition rates of crack and shaft misalignment faults for the rotor system with different SNR values are more than 99%, which is considered to be very effective and useful.

Rolling-Bearing Fault-Diagnosis Method Based on Multimeasurement Hybrid-Feature Evaluation

To improve the accuracy of rolling-bearing fault diagnosis and solve the problem of incomplete information about the feature-evaluation method of the single-measurement model, this paper combines the advantages of various measurement models and proposes a fault-diagnosis method based on multi-measurement hybrid-feature evaluation. In this study, an original feature set was first obtained through analyzing a collected vibration signal. The feature set included time- and frequency-domain features, and also, based on the empirical-mode decomposition (EMD)-obtained time-frequency domain, energy and Lempel–Ziv complexity features. Second, a feature-evaluation framework of multiplicative hybrid models was constructed based on correlation, distance, information, and other measures. The framework was used to rank features and obtain rank weights. Then the weights were multiplied by the features to obtain a new feature set. Finally, the fault-feature set was used as the input of the category-divergence fault-diagnosis model based on kernel principal component analysis (KPCA), and the fault-diagnosis model was based on a support vector machine (SVM). The clustering effect of different fault categories was more obvious and classification accuracy was improved.

A Bionic Vibration Source Localization Device Inspired by the Hunting Localization Mechanism of Scorpions

The challenges we are faced with in localizing objects are the complex environments, such as tunnels, high-rise areas and underground parking lots. This paper develops a bionic vibration source localization device to estimate the direction of the object which is inspired by the unique and precise hunting localization mechanism of scorpions. The localization device uses the sensor array, which is patterned after the scorpions’ biological sensory structure, and imitates the coding mode of scorpions’ sensory neurons for determination of the prey (vibration source) bearing. To verify the effectiveness of the localization device, some experiments were performed through real collected vibration signals. The Average Estimated Error (AEE) and the Relative Estimated Error (REE) of the experimental results were calculated to be 3.64° ± 2.44° and −1.43° ± 4.14°, respectively. It indicates that the device has a good performance to estimate the bearings of vibration sources at different distances and azimuths. This bionic localization device lays the foundation for the development of locating the moving object in some special conditions.

Precision testing method of five‐axis NC machine tool based on ‘S’ specimen cutting

‘S’ specimen through its unique profile features makes it possible to fully mobilise the five-axis linkage function in the five-axis finishing stage. The continuous rotation direction of the rotating axis not only reflects the static error of the machine tool but also the servo system follow error, tool positioning error, profile error etc. The authors studied its influence on the accuracy of each axis of five-axis CNC machine tool by analysing the geometrical characteristics of a S specimen and using UG's CAM environment to process ‘S’ specimen. Finally, the corresponding program forming system for S specimen machining experiments is designed by LabVIEW, and the authors used macro instruction to do NC program trigger system to collect vibration signals segmentally, so that the processing and acquisition system can be synchronised. This provides a more complete basis for the test piece cutting method to evaluate performance of machine tools.

Multi-fault Condition Monitoring of Slurry Pump with Principle Component Analysis and Sequential Hypothesis Test

A new method about the multi-fault condition monitoring of slurry pump based on principal component analysis (PCA) and sequential probability ratio test (SPRT) is proposed. The method identifies the condition of the slurry pump by analyzing the vibration signal. The experimental model is established using the normal impeller and the faulty impellers where the collected vibration signals were preprocessed using wavelet packet transform (WPT). The characteristic parameters of the vibration signals are extracted by time domain signal analysis and the dimension of data was reduced by PCA. The principal components with the largest contribution rate are chosen as the inputted signal to SPRT to assess the proposed algorithm. The new methodology is reasonable and practical for the multi-fault diagnosis of slurry pump.

A phenomenological model for investigating unequal planet load sharing in epicyclic gearboxes

Abstract Vibration signals of epicyclic gearboxes often present abundant modulation sidebands in the spectrum, and investigations of these sidebands are helpful for the understanding of vibration mechanism and fault diagnosis of epicyclic gearboxes. In epicyclic gearboxes, both the carrier rotation and unequal load sharing among planets can lead to abundant sidebands near the meshing frequency. Modulation sidebands due to the carrier rotation have been extensively investigated, while additional sidebands caused by unequal load sharing among planet gears are seldom studied. Aiming at this issue, a phenomenological model of vibration signals of epicyclic gearboxes is developed considering unequal planet load sharing conditions. In this model, an angular shift of a planet gear is assumed to simulate the manufacturing or assembly errors in gearboxes, and different load sharing ratios among planets are calculated. After that, spectral structures of vibration signals of epicyclic gearboxes under unequal load sharing conditions are derived from the phenomenological model. Through the deduced results, it is shown that unequal planet load sharing can influence the spectral structures largely and the formation mechanisms of some additional sidebands in the spectrum are explained. Finally, experiments under three health conditions, i.e., healthy, cracked sun gear and cracked planet gear, are conducted on an epicyclic gearbox test rig. The effectiveness of theoretical deductions of phenomenological models is validated by the collected vibration signals from the test rig.

Agent discriminate model based optimization weighted method and its application in fault diagnosis of rolling bearings

In the fault diagnosis of rolling bearing, the vibration signals, which are collected from the field test, are often more complex because they unavoidably contain various noises and measurement errors, so ‘outliers’ may occur in the features extracted from the collected vibration signals. Aiming at the above problems, the agent discriminate model based optimization weighted (ADMOW) method is proposed. By using the entropy weight method (EWM), the entropy weights of the sample features are calculated first, and the features are then weighted to weaken the influence of ‘outliers’ on the modeling. Secondly, the particle swarm optimization (PSO) algorithm is used to optimize the parameters of the model, and a more accurate classification model is obtained. Eventually, ADMOW is applied to recognize defaults of rolling bearings. The test results indicate that by comparing several pattern recognition methods, the proposed method can effectively weaken the influence of ‘outliers’ and improve the recognition rate and the recognition accuracy.

Weak Fault Feature Extraction and Enhancement of Wind Turbine Bearing Based on OCYCBD and SVDD

The fault feature of wind turbine bearing is usually very weak in the early injury stage, in order to accurately identify the defect location, an original approach based on optimized cyclostationary blind deconvolution (OCYCBD) and singular value decomposition denoising (SVDD) is put forward to extract and enhance the fault feature effectively. In this diagnosis method, the fast spectral coherence is fused with the equal step size search strategy for the cyclic frequency parameter and the filter length parameter optimization, and a new frequency weighted energy entropy (FWEE) indicator which combining the advantages of the frequency weighted energy operator (FWEO) and the Shannon entropy, is developed for deconvolution signal evaluation during parameter optimization process. In addition, a novel singular value order determination approach based on fitting error minimum principle is utilized by SVDD to enhance the fault feature. During the process of defect identification, OCYCBD with the optimal parameters is firstly used to recover the informative source from the collected vibration signal. FWEO is further utilized to highlight the potential impulsive characteristics, and the instantaneous energy signal of deconvolution result can be acquired. The whole interferences contained in the instantaneous energy signal can’t be removed due to the weak fault signature and the severe background noise. Then, SVDD is applied to purify the instantaneous energy signal of deconvolution signal, by which the residual interference component is eliminated and the fault feature is strengthened immensely. Finally, frequency domain analysis is performed on the denoised instantaneous energy signal, and the defect location identification of wind turbine bearing can be achieved through analyzing the obvious spectral lines in the obtained enhanced energy spectrum. The collected signals from the experimental platform and the engineering field are both utilized to verify the feasibility of proposed method, and its superiority is further demonstrated through comparing with several well known diagnosis methods. The results indicate this novel method has distinct advantage on bearing weak feature extraction and enhancement.

The application of matching pursuit based on multi feature pattern set in the signal processing of rotating machinery

The vibration signal containing the fault feature is buried by the strong background noise usually (the noise is referring to the other signals not containing the fault feature in the general sense) when a fault arises in rotating machinery, so the collected vibration signal needs to be processed normally in order to get a correct diagnosis result. Sparse representation theory is a relatively new signal processing method and matching pursuit (MP) is the classical sparse representation method. However, in the traditional MP method only one kind of atom dictionary is constructed to match the fault feature hidden in the vibration signal, which will cause an unsatisfactory analysis result usually when several different kinds of components arise in the vibration signal synchronously. This paper proposes a novel fault feature strengthening method basing on MP to solve the above problem, and the proposed method constructs several different kinds of atom dictionary based on a multi feature pattern set to match the multi-components buried in the collected vibration signal of rotating machinery. Firstly, the effectiveness and advantages of the proposed method are verified by kinds of typical simulation signals of rotating machinery. Then the proposed method is extended to preprocess the vibration signals of bearing and gearbox, and the corresponding analysis results prove that the proposed method has obviously a better fault feature extraction effect on the vibration signal of rotating machinery than the traditional MP method, and the proposed method lays the foundation for further correct fault analysis.

An optimal variational mode decomposition for rolling bearing fault feature extraction

Rolling bearings usually work in tough conditions, which makes the collected vibration signals complex and the fault features weak. Hence, fault feature extraction methods for rolling bearings have become a research focus. In this paper, a new method termed optimal variational mode decomposition (VMD) is proposed to extract rolling bearing fault features. Firstly, since envelope entropy is very sensitive to fault signal features, envelope entropy is used as a fitness function, which is an objective function for the whale optimization algorithm (WOA). Secondly, the WOA has numerous merits, such as simple operation, fewer adjustment parameters and a strong ability for jumping out of the local optimum, and it is applied to the optimization of VMD. Finally, intrinsic mode function components are processed through a Teager energy operator. The proposed method is employed to analyze the experimental signal collected from rolling bearings. The comparison results show that the proposed method is more effective and demonstrates superiority over empirical mode decomposition, local mean decomposition and wavelet packet decomposition.

Vibration source classification and propagation distance estimation system based on spectrogram and KELM

Earth surface vibration signals source classification and propagation distance estimation attract increasing attention in recent years due to the wide applications in many areas. In this study, the authors develop a hybrid classification and propagation distance estimation algorithm for general earth surface vibration sources. The spectrogram (SPEC) feature characterising the energy distribution of vibrations is first developed for signal representation in this study. The kernel-based extreme learning machine (KELM) algorithm is then adopted for the vibration source classification and propagation distance estimation. Comparing with the conventional approaches, the proposed KELM + SPEC algorithm is not only effective in characterising the time- and frequency-domain features of vibrations, but also superior in accuracy and efficiency. To test the effectiveness of the proposed KELM + SPEC algorithm, experiments on real collected vibration signals are presented, where simulations on both periodic and aperiodic vibrations are carried out in the study. Comparisons to various existing vibration signal extraction and classification algorithms are provided to show the advantages of the proposed KELM + SPEC algorithm.

A New Morphological Filter for Fault Feature Extraction of Vibration Signals

Early faults in rolling bearings tend to result in periodic impulse components in the collected vibration signals. However, these fault features are always distorted by noise interferences. Feature extraction from vibration signals is an effective means to detect early defects in rolling bearings. A new morphological filter (MF), called enhanced difference morphological filter (EDMF) is proposed for vibration signal processing and then implementing bearing fault diagnosis. EDMF is capable of depressing noise and preserving effective impulsive components, where four new basic morphological operators are integrated effectively. The experimental results on simulation and bearing vibration signals verify that EDMF is effective for defect detection of rolling bearings. The comparison results show that the new MF can extract fault features more effectively from vibration signals with much noise than other typical MFs.

Transformer Incipient Hybrid Fault Diagnosis Based on Solar-Powered RFID Sensor and Optimized DBN Approach

This paper introduces a novel hybrid fault diagnosis method for power transformer. This method employs solar-powered radio-frequency identification (RFID) sensor for transformer vibration signal acquisition and deep belief network (DBN) for feature extraction. The customized RFID sensor employs solar panel as a power source, and a supercapacitor is adopted to be the stand-by power when the solar panel cannot work. A charging circuit is exploited to guarantee constant DC output voltage. The collected hybrid faults signal is characterized as nonlinear and nonstationary; moreover, it contains abundant noises and harmonic components, which makes it difficult to acquire succinct and robust features from the raw signals. Hence, the DBN is adopted to extract features from the collected vibration signal. In order to obtain optimum feature extraction performance, the quantum particle swarm optimization algorithm (QPSO) is employed to determine the hidden layer structure and learning rate of the DBN model. The experiments indicate that the proposed RFID sensor is able to realize reliable data acquisition and transmission. Besides, the optimized DBN achieves remarkable results in feature extraction for the hybrid fault signal and achieves high diagnosis accuracy.

Adaptive Reinforced Empirical Morlet Wavelet Transform and Its Application in Fault Diagnosis of Rotating Machinery

Identifying impact fault features from fault vibration signal is significantly meaningful for the fault diagnosis and condition monitoring of rotating machinery. Given defects and the working conditions, impact features are covered by background noise. A new method named empirical wavelet transform (EWT) has been receiving attention from the researchers and engineers. However, detecting boundaries by using the local maxima method from Fourier spectra and capturing the impact features through Meyer wavelet are the two crucial drawbacks of EWT. The former might be invalidated by the influence of non-stationary and noise frequency, and the latter is inappropriate for impact signal features. Therefore, reinforced empirical Morlet wavelet transform (REMWT) is proposed to overcome these shortcomings and efficiently diagnose fault features. In this method, the frequency spectrum boundaries are adaptively detected from the inner product of spectral kurtosis and Gaussian function via scale space representation, which can enhance the frequency character of impact features in vibration signals. Then, the constructed empirical Morlet wavelet serves as the adaptive filter bank for decomposing the signal into several empirical modes on the basis of spectrum boundaries. The meaningful component is selected via the maximum Pearson correlation coefficient method, and the envelope spectrum is used to accurately extract the fault features. The proposed method is then used to diagnose the fault features from the collected vibration signals. The results show its effectiveness and outstanding performance.

Condition Monitoring of Two Stroke IC Engine Ball Bearing Based on Vibration Signals Using Machine Learning Techniques

Ball bearing faults are foremost cause to the breakdown of rotating machine components. In general, ball bearing in the IC engine is one of the vital machine elements and plays major role in the rotating machine as it takes different kinds of stresses and dynamic loads under various working conditions. Monitoring of such ball bearing continuously is extremely needed to avoid damage to the components of IC Engine. Vibration analysis of IC engine is most commonly used to recognise the fault in ball bearing and to differentiate the types of ball bearing defects. In the present work, study has been conducted on condition monitoring of ball bearing of 2-stroke single cylinder IC engine using vibration signals and explains the use of statistical features and support vector machine as a tool for classification of the ball bearing conditions. The different cases of faults considered in the diagnosis of ball bearing are outer race (OR) defect, inner race (IR) defect, combined faults in outer and inner race (OR & IR) and ball defect. The vibration signals under good and different simulated defective conditions of ball bearing are collected using an accelerometer. The descriptive statistical set of features are found from the collected vibration signals. The significant features are carefully chosen by means of J48 decision tree technique and the features chosen from tree are given as input to the support vector machine (SVM). The experimental results show that, the SVM model provides 96% classification accuracy for the given vibration signals. The obtained result from the model is good and can be considered for diagnosis of rotating machinery. Hence, the J48 decision tree and SVM algorithm can be proposed for applications of condition monitoring of ball bearings.

Frequency-shift multiscale noise tuning stochastic resonance method for fault diagnosis of generator bearing in wind turbine

The wind energy industry has developed very fast, while the condition monitoring and fault diagnosis for wind turbine is increasingly becoming the focus of manufacturers and wind-farm operators. However, due to the influence of harsh operation environment and complex internal structure of wind turbine, collected vibration signals are corrupted by strong noise. Effective extraction of useful feature information submerged in strong noise that is indicative of structural defects has remained a major challenge. In this paper, a novel frequency-shift multiscale noise tuning stochastic resonance (SR) method is proposed with the advantage of SR using noise to enhance weak signal features. Firstly, the frequency shift modulation algorithm is adopted to move the target signal to the designated low frequency domain. Then the obtained modulated signal is processed by the multiscale noise tuning algorithm based on discrete wavelet transform (DWT). The resulting output is used as the input of SR system. Finally, the tuning parameter of multiscale noise is determined by maximizing the modified signal-to-noise ratio (SNR) of system output. The proposed method can realize the feature enhancement and extraction of weak signal with arbitrary frequency. Experiments and fault diagnosis case of generator bearing in wind turbine validate the effectiveness of the proposed method.

Feature extraction for rolling element bearing weak fault based on MOMEDA and ICEEMDAN

There are always the nonlinear and non-stationary characteristics and periodic pulse in vibration signals of rolling element bearings when there are partial faults in those bearings. Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) overcomes the presence of spurious modes and residual noise in Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), but it cannot clearly and accurately extract the weak fault feature of rolling element bearings under the strong background noise. Here, Multipoint Optimal Minimum Entropy Deconvolution Adjusted (MOMEDA) was proposed. A bearing simulator was used to collect vibration signals of bearing inner and outer race, which was enhanced by MOMEDA, decomposed into several Intrinsic Mode Functions(IMFs) by ICEEMDAN, and analyzed by the envelope demodulation, finally gaining the frequency of shaft speed, BPFI (ball pass frequency, inner race) and harmonics, sidebands spaced, BPFO (ball pass frequency, outer race) and harmonics. The results show that this method can be used to accurately extract different frequency components of bearing fault vibration signals and diagnose bearing different fault location.

Planetary gearbox fault feature learning using conditional variational neural networks under noise environment

The features signals of early fault collected from planetary gearbox are usually weak. It is difficult to extract effective fault features from the collected vibration signals under noise environment. In this paper, a new feature learning method for fault diagnosis of planetary gearbox based on deep conditional variational neural networks (CVNN) is proposed. First, the new method utilizes multi-layer perceptron (MLP) to model the normal distribution features of frequency spectra from noisy vibration signals. Second, the new features are obtained by resampling normal distribution features in order to eliminate the effect of noise. Then the denoised features are compressed and reduced dimensionally by MLP. Third, the effective denoised features are input to classifier. Finally, the trained CVNN is applied for intelligent fault diagnosis of planetary gearbox. The experimental results confirm that CVNN method can extract effective fault features from noisy vibration signals, and it has higher accuracy of fault diagnosis than other methods in the case of low signal to noise ratio (SNR) values.