Vibration Signal Processing Research Articles

Fault diagnosis of a planetary gearbox based on a local bi-spectrum and a convolutional neural network

The transmission paths of vibration signals in the planetary gearboxes are complex. The signals have the characteristics of strong background noise, instability and non-Gaussian. Bi-spectrums can suppress Gaussian colored noise and are suitable for vibration signal processing of planetary gearboxes. In the traditional fault diagnosis methods based on bi-spectrums, the amplitudes of fault characteristic frequency, or the other further quantitative calculations values, are generally used as the basis of fault diagnosis processes. It has been found that bi-spectrum images can directly characterize the faults of the planetary gearboxes. Convolutional neural networks (CNNs) have been used in mechanical fault diagnoses in recent years. One-dimensional original signals are converted into two-dimensional images as CNN input, which is an effective method for mechanical fault diagnoses. At the present time, there has not been any relevant research conducted using bi-spectral images as CNN input. In this study, a fault diagnosis method based on local bi-spectrum and CNN was proposed. A bi-spectral analysis of the vibration signals of the planetary gearbox was first carried out in order to reveal the fault information while retaining the non-Gaussian information. Then, according to the bi-spectrum symmetry, local images containing the main information were taken as the input of the CNN, which reduced the redundancy of the fault information. Then, in order to improve the diagnostic accuracy of the CNN, the key parameters of CNN architecture were optimized. Finally, a CNN diagnosis model was built to realize the classification diagnoses of different fault positions and different fault degrees of planetary gearboxes. This study’s comparison of the diagnosis results of the full bi-spectrum + CNN, original vibration signal + CNN, local bi-spectrum + (support vector machines), and local bi-spectrum + (stacked auto-encoder) showed that the proposed method in this study had achieved both accuracy and rapidity in the fault diagnoses of planetary gearboxes.

Research on fault diagnosis of automobile engines based on the deep learning 1D-CNN method

Traditional engine fault diagnosis methods usually need to extract the features manually before classifying them by the pattern recognition method, which makes it difficult to solve the end-to-end fault diagnosis problem. In recent years, deep learning has been applied in different fields, bringing considerable convenience to technological change, and its application in the automotive field also has many applications, such as image recognition, language processing, and assisted driving. In this paper, a one-dimensional convolutional neural network (1D-CNN) in deep learning is used to process vibration signals to achieve fault diagnosis and classification. By collecting the vibration signal data of different engine working conditions, the collected data are organized into several sets of data in a working cycle, which are divided into a training sample set and a test sample set. Then, a one-dimensional convolutional neural network model is built in Python to allow the feature filter (convolution kernel) to learn the data from the training set and these convolution checks process the input data of the test set. Convolution and pooling extract features to output to a new space, which is characterized by learning features directly from the original vibration signals and completing fault diagnosis. The experimental results show that the pattern recognition method based on a one-dimensional convolutional neural network can be effectively applied to engine fault diagnosis and has higher diagnostic accuracy than traditional methods.

Second-order Synchrosqueezing Modified S Transform for wind turbine fault diagnosis

Under the working condition of variable speed, vibration signals of wind turbine gearboxes are characterized by instability coupled with noise interference, which has made it difficult to identify early fault characteristics through common Time-Frequency (TF) Analysis (TFA) methods. Stockwell Transform (ST) has such advantages as undergoing inverse transformation without loss and being free from cross-term interference when processing signals. Therefore, this paper has put forward Modified Stockwell Transform (MST), which enables the window width to change proportionally with the varying frequency. Moreover, the change caused by parameter adjustment is much more purposeful. It ensures that TF analysis windows are more flexible and changeable at different positions in the TF plane. Furthermore, to obtain much more accurate Instantaneous Frequency (IF) estimation of multi-component vibration signals, this paper uses the second-order partial derivative of phase function to modify it. Eventually, in order to achieve a concentrated Time-Frequency Representations (TFR) of wind turbines vibration signals, this paper proposes Second-order Synchrosqueezing Modified S Transform, namely SSMST2, through making use of the theoretical framework of Synchrosqueezing Transform (SST) and second-order IF estimations to extend MST. By reassigning the TF transform coefficients of MST, the proposed SSMST2 can better show the time-varying features. To verify this method is feasible, this paper will apply it to fault diagnosis of bearing test rig and to vibration signal processing of rolling bearing in 2 MW wind turbine. The result demonstrates that SSMST2 has yielded great outcome in TFR performance of variable speed vibration signals.

Sound-aided fault feature extraction method for rolling bearings based on stochastic resonance and time-domain index fusion

The vibration signals of rolling bearings contain a large amount of running information, and have been widely applied in fault diagnosis and status monitoring. However, the fault features decay significantly in the transmission paths, which reduces the detection efficiency. This paper fuses the time-domain vibration and sound signals, and proposes a fault feature extraction method based on the fusion signal. A new indicator is set to show the fusion performance, and a sliding Hanning window is put on the time axis for the optimal fusion signal. The ant colony algorithm is used for the optimization of parameters in stochastic resonance system, and the final output signal is obtained at the highest signal-to-noise ratio. Experimental results show that the fault feature extraction performance gets greatly increased through the proposed method, and the feature enhancement is closely related with the parameters of the Hanning window. The phase difference between the vibration and sound signals is eliminated in the time-domain signal fusion, making the method capable of processing vibration and sound signals collected with time difference. This study proposes a new approach for fault diagnosis, and helps in the status monitoring of bearing systems.

Evaluation of the mechanism damage using model of vibration signal as a periodically correlated random process

The results of the processing of the wind turbine gearbox vibration signals are presented. The least square estimators of the periods of the deterministic part of the vibration signal and time variations power of stochastic part are obtained. The amplitude spectra of the deterministic and stochastic oscillations for different degrees of the gearbox damage are analyzed. The covariance structure of stochastic part of vibration was analyzed. The efficient indicator for practical evaluation of gear teeth damage, formed on the basis of amplitude spectra was proposed.

A New Diagnostic Method Applied to Gearbox Missing Gear Faults ——LOD-ICA

With the increasingly stringent requirements for carbon emissions, countries have increased the scale of clean energy use in recent years. As an important new clean energy source, the ratio of wind power in energy utilization has been increasing. The horizontal axis wind turbine is the main form of wind power generation, which is subject to random wind loads during operation and is prone to various failures after a long period of operation, resulting in reduced power generation efficiency or even shutdown. In order to ensure stable external power transmission, it is necessary to perform fault diagnosis for wind turbines. However, the traditional time-frequency analysis method is defective. This paper proposes a new LOD-ICA method to realize the resolution of the vibration signals mode mixing problem incorporated the merits of both methods. The LOD-ICA method and the LOD method based on noise-assisted analysis decompose the same signal to produce different signal components. The feasibility of the LOD-ICA method was verified by comparing the correlation coefficients between each of the signal components generated by the two methods and the original signal. In the field of wind turbine fault diagnosis, the LOD-ICA method is employed to the fault characteristics of gearboxes to extract the fault signs of vibration signals, further demonstrated the superiority of the LOD-ICA method in processing vibration signals of rotating machinery.

Life Prediction Method of Remanufactured Machinery Equipment Based on Vibration Signal Feature Extraction

Mechanical equipment is a key component of mechanical equipment, and its working condition is directly related to the overall performance of mechanical equipment. Accurate evaluation and prediction of the performance degradation trend of mechanical equipment is of great significance to ensure the reliability and safety of the mechanical equipment system. Based on the data of typical faulty equipment, this paper analyzes the energy characteristic parameters of mechanical equipment under different types and degrees of failure in the time domain. Using amplitude spectrum analysis, Hilbert envelope demodulation and wavelet packet decomposition method, and other vibration signal processing methods, preliminary extraction of multiple statistical feature parameters are given. Secondly, in view of the irrelevant and redundant components of multiple statistical parameters, a new method for extracting fault features of mechanical equipment based on variance value and principal component analysis is proposed. This method can effectively classify the fault status of mechanical equipment. The effectiveness of the method is verified by actual equipment signals. After that, the value extracted from the vibration signal of the double-row roller equipment is used as the degradation feature. In order to reduce the influence of irregular characteristics in the vibration signal and simplify the complexity of the vibration signal, the wavelet transform and the support vector machine model are combined, according to the degradation after decomposition. The 95% confidence interval of the predicted value is also given. The SVM model is established based on data characteristics, and single-step and multistep prediction of equipment degradation trends are carried out. The prediction result shows that, according to the mapping position formula, the distribution of equipment degradation prediction points is obtained, and a 95% confidence interval based on the distribution of the prediction points is given. Finally, on the basis of completing feature extraction, this paper applies an unsupervised feature selection method. The sensitive characteristics of life prediction and the prediction results of a single SVM model and a neural network model are compared and analyzed at the same time.

Fully interpretable neural network for locating resonance frequency bands for machine condition monitoring

In recent years, various neural networks have been developed to process vibration signals for machine condition monitoring. Nevertheless, the physical interpretation of neural networks is still on-going and not fully explored. This paper aims to design a fully interpretable neural network for machine condition monitoring from the aspects of signal processing and physical feature extraction. The main idea of the fully interpretable neural network is to extend the uninterpretable structure of extreme learning machine (ELM) to an interpretable structure for machine condition monitoring. From the aspect of signal processing, wavelet transform, square envelope and Fourier transform are incorporated into the input layer of the original ELM to extract repetitive transients, localize informative frequency bands for an enhancement of a signal-to-noise ratio, and realize square envelope spectra for exhibiting cyclo-stationarity of repetitive transients. Hence, the first to four layers of the proposed network are physically interpretable. From the aspect of physical feature extraction, popular sparsity measures are innovatively incorporated into all random nodes in the single-hidden layer of the original ELM to interpret the use of all hidden nodes in the fifth layer of the proposed network to characterize cyclo-stationarity of repetitive transients. The significance of this paper is to show that signal processing algorithms and physical feature extraction can be reformulated as the architecture of an interpretable neural network to automatically localize informative frequency bands for machine condition monitoring. This paper attempts to inspire researchers in the field of signal processing and machine learning to think about the design of more advanced interpretable neural networks for machine condition monitoring.

Exploration of Vibration Signal Measurement Method of Automobile Parts Utilizing FPGA/Photoelectric Sensor

According to the current situation of the sensor, a vibration signal monitoring system is proposed of automobile parts based on Field Programmable Gate Array (FPGA) and photoelectric sensor. FPGA controls the peripheral circuit and sets instructions to process the acquired vibration signal. The Serial Port Interface (SPI) is utilized for the data transmission and communication between the instruction set and FPGA. FPGA transmits the acquired data to the Digital Signal Processing (DSP) through the SPI and then backs up to the Secure Digital (SD) card for in-depth analysis. The Fiber Bragg Grating (FBG) is selected for the monitoring system. The sensor presents many characteristics: anti-interference, high sensitivity, lightweight, and small volume. Afterward, the FBG is used for vibration signal acquisition, and the corresponding sensor module circuit is designed. Subsequently, an Analog to Digital (A/D) conversion circuit is designed on the acquisition circuit together with the vibration conditioning circuit and FPGA peripheral circuit. In the experiment, the vibration signal is monitored in real-time through the designed monitoring system, and the changes in the Center Wavelength (CWL) of the FBG from vibration are recorded. The data are screened, and more reasonable data are produced for vibration signal processing. The vibration signal is processed through Time-Domain (TD) linear fitting and time-frequency transformation, to assess the FBG vibration sensor’s vibration response in the monitoring system.

COMBINED SIGNAL PROCESSING METHOD FOR DIAGNOSIS AND MONITORING OF THE INDUCTION MOTORS OPTIMIZED FOR EMBEDDED SYSTEMS

Induction Motors (IM) play a key role in modern industry, so the condition monitoring systems are becoming increasingly relevant. Commercial monitoring systems are usually based on the measurement of IM’s vibrations and the further processing of the measured vibration signals. For those purposes the embedded systems (such as microcontrollers and inexpensive processors) are used. Embedded systems have limited resources, so data processing algorithms should have low computational complexity and require little memory. In this paper, the wellknown methods of processing vibration signals for fault diagnosis of the IM are considered and their main advantages and disadvantages for the implementation in embedded systems are highlighted. The previously proposed method based on a combination of the fast Fourier transform and the statistics of the fractional moments is optimized for vibration signal processing and implementation in embedded systems. The efficiency of diagnosis of such faults as eccentricity and a broke rotor bar, using the proposed method, is verified on the radial vertical vibrations measurements of the real motors under different constant load levels: no load, 50 % of the rated load, 75% of the rated load. The results show that this approach allows accurately diagnose the considered faults independently from the load level.

Evolutionary Optimization of Machining Parameters Based on Surface Roughness in End Milling of Hot Rolled Steel.

Surface roughness measurements of machined parts are usually performed off-line after the completion of the machining operation. The objective of this work is to develop a surface roughness prediction method based on the processing of vibration signals during steel end milling operation performed on a vertical CNC machining center. The milling cuts were run under varying conditions (such as the spindle speed, feed rate, and depth of cut). This is a first step in the attempt to develop an online milling process monitoring system. The study presented here involves the analysis of vibration signals using statistical time parameters, frequency spectrum, and time-frequency wavelet decomposition. The analysis resulted in the extraction of 245 features that were used in the evolutionary optimization study to determine optimal cutting conditions based on the measured surface roughness of the milled specimen. Three feature selection methods were used to reduce the extracted feature set to smaller subsets, followed by binarization using two binarization methods. Three evolutionary algorithms—a genetic algorithm, particle swarm optimization and two variants, differential evolution and one of its variants, have been used to identify features that relate to the “best” surface finish measurements. These optimal features can then be related to cutting conditions (cutting speed, feed rate, and axial depth of cut). It is shown that the differential evolution and its variant performed better than the particle swarm optimization and its variants, and both differential evolution and particle swarm optimization perform better than the canonical genetic algorithm. Significant differences are found in the feature selection methods too, but no difference in performance was found between the two binarization methods.

Methods of Hidden Periodicity Discovering for Gearbox Fault Detection.

It is shown that the models of gear pair vibration, proposed in literature, are particular cases of the bi-periodically correlated random processes (BPCRPs), which describe its stochastic recurrence with two periods. The possibility of vibration and analysis within the framework of BPCRP approximation, in the form of periodically correlated random processes (PCRPs), is grounded and the implementation of vibration processing procedures using PCRP techniques, which are worked out by the authors, is given. Searching for hidden periodicities of the first and the second orders was considered as the main issue of this approach. The estimation of the non-stationary period (basic frequency) allowed us to carry out a detailed analysis of the deterministic part, the covariance structure of the stochastic part, and to form, using their parameters, the sensitive indicators for fault detection. The results of the processing of the wind turbine gearbox vibration signals are presented. The amplitude spectra of the deterministic oscillations and the time changes of the stochastic part power for different fault stages are analyzed. The most efficient indicators, which are formed using the amplitude spectra for practical applications, are proposed. The presented approach was compared with known in literature cyclostationary analysis and envelope techniques, and its advantages are shown.

Bearing fault diagnosis based on kernel independent component analysis and antlion optimization

Fault diagnosis of gearboxes based on vibration signal processing is challenging, as vibration signals collected by acceleration sensors are typically a nonlinear mixture of unknown signals. Furthermore, the number of source signals is usually larger than that of sensors because of the practical limitation on sensor positions. Hence, the fault characterization is actually a nonlinear underdetermined blind source separation (NUBSS) problem. In this paper, a novel NUBSS algorithm based on kernel independent component analysis (KICA) and antlion optimization (ALO) is proposed to address the technical challenge. The mathematical model demonstrates the nonlinear mixing of source signals in the underdetermined cases. Ensemble empirical mode decomposition is used as a preprocessing tool to decompose the observed signals into a set of intrinsic mode functions that suffers from the problem of redundant components. The correlation coefficient is utilized to eliminate the redundant components. An adaptive threshold singular value decomposition method is proposed to estimate the number of source signals. Then a whitening process is carried out to transform the overdetermined blind source separation (BSS) into determined BSS, which can be solved by the KICA method. However, the reasonable selection of parameters in KICA limits its application to some extent. Therefore, ALO and Fisher’s linear discriminant analysis are adopted to further enhance the accuracy of the KICA method. The separation performance of the proposed method is assessed through simulation. The numerical results show that the proposed method can accurately estimate the number of source signals and attains a higher separation quality in tackling nonlinear mixed signals when compared with the existing methods. Finally, the inner ring fault experiment is conducted to preliminarily validate the practicability of the proposed method in bearing fault diagnosis.

Analysis and Application of Vibration Signals of Aerospace Servo Actuator

This paper analyzes the vibration signals of aerospace servo actuator and studies their applications. In the signal pre-processing, an adaptive filter based on genetic algorithm is added to improve the signal-to-noise ratio. Time domain analysis is used to monitor the vibration signals and determine the main vibration directions. Frequency domain analysis is used to extract the main vibration frequencies and perform preliminary fault analysis. The WT method with optional window function is used for time-frequency analysis, which can describe the relationship between vibration signal frequency and time, and decompose the signal characteristics required for the analysis. In the research, LabVIEW and MATLAB are used for data processing and analysis. The research lays the foundation of aerospace servo actuator vibration signal analysis and processing system.

Determination of the state and functioning of the equipment of rod well pumping units in the operation process by the parameters of the wattmetrogram

TARGET.The purpose ofthis work is to develop a method for diagnostics of equipment of sucker rod pumping units (SRPU) during operation according to the parameters of the wattmetrogram. A wattmetering module is used as a measuring element, the developed software in the MatLab software package serves as a tool for processing vibration signals. The technical condition of the sucker rod pumping unit is analyzed by the parameters of the wattmetrogram and the amplitude spectra of the object under study, formed using the fast Fourier transform procedure in the MatLab software environment.METHODS.Thewattmeteringmethodallowsyouto control the sucker rod-pumping unit during its operation under voltage. This control method is highly accurate, sensitive to developing defects, and allows the use of computer technologies for signal processing and analysis.RESULTS.The wattmetering method is fully disclosed. The construction of the spectrum is considered. The relationship between the wattmetrogram and the dynamogram has been established. The removal and evaluation of experimental data from the operating sucker-rod pumping unit during its operation was carried out. A wattmetrogram and a dynamogram were built, a spectrum was also built, which showed the presence of defects in this installation.CONCLUSION.ThismethodofwattmeteringwiththeMatLabsoftwareallowsyoutomonitorthetechnicalconditionofthesuckerrod pumping unit according to the amplitude-frequency characteristics of the wattmetrogram parameters.

Contactless laser control and measuring complex for determining the level of winding pressure and magnetic conductor of power transformer

TARGET. The purpose of this work is to develop a non-contact laser control and measuring complex (LCMC) for vibration control of the level of pressing of windings and magnetic circuit of a power transformer. A laser vibrometer is used as a measuring element, the developed software in the LabVIEW graphical programming environment serves as a tool for processing vibration signals. The technical condition of the power transformer is analyzed by the amplitude spectra of the vibration of the tank of the transformer under study, formed using the fast Fourier transform algorithm in the LabVIEW software.METHODS. The vibration control method makes it possible to control a power transformer during its operation under voltage, which makes it possible to move from a planned system of transformer repairs to a system for taking out for repair according to the current technical condition.RESULTS. The developed LCMC allows to carry out non-contact measurements of vibration parameters of a power transformer under voltage and to establish the current level of pressing of windings and magnetic circuit.CONCLUSION. The developed LCMC with software allows contactless control of the technical condition of the magnetic circuit and the windings of the power transformer according to the amplitude-frequency characteristics of the vibration parameters, as well as the use of statistical methods for processing and analyzing signals received from the transformer.

Contactless laser control and measuring complex for determining the level of winding pressure and magnetic conductor of power transformer

TARGET. The purpose of this work is to develop a non-contact laser control and measuring complex (LCMC) for vibration control of the level of pressing of windings and magnetic circuit of a power transformer. A laser vibrometer is used as a measuring element, the developed software in the LabVIEW graphical programming environment serves as a tool for processing vibration signals. The technical condition of the power transformer is analyzed by the amplitude spectra of the vibration of the tank of the transformer under study, formed using the fast Fourier transform algorithm in the LabVIEW software.METHODS. The vibration control method makes it possible to control a power transformer during its operation under voltage, which makes it possible to move from a planned system of transformer repairs to a system for taking out for repair according to the current technical condition.RESULTS. The developed LCMC allows to carry out non-contact measurements of vibration parameters of a power transformer under voltage and to establish the current level of pressing of windings and magnetic circuit.CONCLUSION. The developed LCMC with software allows contactless control of the technical condition of the magnetic circuit and the windings of the power transformer according to the amplitude-frequency characteristics of the vibration parameters, as well as the use of statistical methods for processing and analyzing signals received from the transformer.

Determination of the state and functioning of the equipment of rod well pumping units in the operation process by the parameters of the wattmetrogram

TARGET.The purpose ofthis work is to develop a method for diagnostics of equipment of sucker rod pumping units (SRPU) during operation according to the parameters of the wattmetrogram. A wattmetering module is used as a measuring element, the developed software in the MatLab software package serves as a tool for processing vibration signals. The technical condition of the sucker rod pumping unit is analyzed by the parameters of the wattmetrogram and the amplitude spectra of the object under study, formed using the fast Fourier transform procedure in the MatLab software environment.METHODS.Thewattmeteringmethodallowsyouto control the sucker rod-pumping unit during its operation under voltage. This control method is highly accurate, sensitive to developing defects, and allows the use of computer technologies for signal processing and analysis.RESULTS.The wattmetering method is fully disclosed. The construction of the spectrum is considered. The relationship between the wattmetrogram and the dynamogram has been established. The removal and evaluation of experimental data from the operating sucker-rod pumping unit during its operation was carried out. A wattmetrogram and a dynamogram were built, a spectrum was also built, which showed the presence of defects in this installation.CONCLUSION.ThismethodofwattmeteringwiththeMatLabsoftwareallowsyoutomonitorthetechnicalconditionofthesuckerrod pumping unit according to the amplitude-frequency characteristics of the wattmetrogram parameters.

THE USE OF A WAVELET TRANSFORMATION FOR REMOVAL OF THE NOISE COMPONENT FROM THE VIBRO SIGNAL

The article investigates the use of wavelets to remove noise from the measuring vibration signal. It is determined that wavelets are well adapted for signal analysis, for which the principle of causality is important: wavelets preserve the direction of time and do not create parasitic interference between the past and the future. Criteria for selecting an analytical wavelet have been developed, depending on what information should be extracted from the signal and the need to more fully identify and emphasize certain properties of the analyzed signal. It is proposed to use Daubechies wavelets to process the vibration signal data. The simulation of vibration signal filtering from noise with the normal distribution law is performed in the MATCAD package. It is proved that the method of wavelet transform allows to solve the problem of filtering the vibration signal from noise when processing vibration signals obtained by autonomous recording devices in conditions of increased interference from the environment. The obtained results evidence to the prospects of the developed method and its advantages in comparison with the hardware solution of the filtering problem.

HHT-Based Detection Method of Cutter Abnormal Vibration in Spiral Surface Machining

Background:: Cutter abnormal vibration occurs frequently during the spiral surface machining process, and it results in low quality of the finished surface. In order to suppress cutter abnormal vibration effectively, it is necessary to detect abnormal vibration as soon as possible, but the analysis and processing of the cutter abnormal vibration signal in spiral surface machining are difficult because of its complicated components and non-linear non-stationary characteristics. In this paper, a detection method of cutter abnormal vibration signal based on Empirical Mode Decomposition (EMD) and Hilbert–Huang Transform (HHT) is proposed to be applied in spiral surface machining. Method:: First of all, EMD of the cutter vibration signal in the spiral surface machining is performed to obtain a series of Intrinsic Mode Function (IMF) components in different frequency bands. Secondly, the variation in the energy of each IMF component in the frequency domain and the correlation with the original signal are analyzed to obtain the IMF component with the largest amount of information on abnormal vibration symptom. Finally, the Hilbert transform is conducted on the IMF component to extract the symptom features of abnormal vibration. Results:: The Hilbert-Huang spectrogram obtained by Hilbert transform is a two-variable function of time and frequency, from which the frequency information at any time can be obtained, including the magnitude and amplitude of the frequency and the corresponding moments appearing, which can describe the time-frequency characteristics of the non-stationary non-linear signal in detail. Experimental results show that the HHT based method to analyze the cutter vibration signal in the spiral surface machining can extract the symptom of abnormal vibration quickly and effectively, and can detect cutter abnormal vibration rapidly. Conclusion:: The proposed method based on HHT in this paper is fundamentally different from the traditional signal time-frequency analysis methods, and has achieved good results in practical applications. This method could be successfully used in abnormal vibration detection, which could also provide basis and guarantee for the subsequent suppression of abnormal vibration.