Vibration Acceleration Signals Research Articles

Investigation on the characterisation of axle box resonance characteristics to wheel excitation

A growing number of studies have demonstrated the effectiveness and superiority of reverse engineering in using axle box vibration acceleration signals to deduce the operation state of wheelsets. However, there is a lack of discussion and summary on the overall mapping relationship between wheel excitation and axle box response, and there has been no detailed investigation on resonance characteristics of axle box. To address the deficiency, a novel resonant excitation loading method is proposed, which is applied to the vehicle model with flexible wheelset and axle box. Amplitude frequency response (AFR) curve of the resonance characteristics of axle box is obtained. AFR is used to characterise the characteristics of various excitation states of wheel, including wheel polygon, wheel tread flat, and wheel-rail separation state, and the corresponding evolution law is revealed. Finally, through two case studies, the guiding role of axle box resonance characteristics on wheel operation status detection is discussed in detail. Analyses show that mastering the resonance characteristics of axle box is an effective way to deduce the operation status of wheel based on the complex axle box vibration response signal, and it will provide a basis for the quantitative evaluation of wheel local damage.

Rolling Bearing Fault Diagnosis Based on MFDFA-SPS and ELM

Rolling bearings, as important parts on supporting rotating shafts, frequently suffer from fatigue failures. If these rolling bearing failures are not found in time, it will have a huge impact on the whole mechanical system’s operating safety and operating life. To improve the diagnosis of different faults as well as different degrees of faults, a fault diagnosis method based on the multifractal detrended fluctuation analysis (MFDFA) method-singularity power spectrum (SPS) with extreme learning machine (ELM) is proposed. First, MFDFA and SPS analyses are performed on vibration acceleration signals with different faults and different degrees of damage under the same operating conditions, the spectral parameters of stability and quantitative description of differentiation are selected for feature extraction, and then the selected six feature parameters are put into the extreme learning machine for fault classification. The effectiveness of the MFDFA-SPS feature extraction method is demonstrated by analyzing and testing the measured bearing signals. The fault diagnosis accuracy of the bearing fault signals can reach 99.2% based on the MFDFA-SPS with ELM method by using the Case Western Reserve database. The improvements are 6.79% and 18.42% compared to the fault diagnosis methods based on MFDFA with ELM and SPS with ELM. Compared with the methods based on MFDFA-SPS with LSSVM classifier and SVM classifier, the accuracy improvements are 3.54% and 4.25%, respectively. The results show that the method proposed in this paper can achieve the diagnosis of bearing faults and the method based on MFDFA-SPS with ELM is more efficient than the methods based on MFDFA-SPS with LSSVM and SVM classifiers, which is suitable for practical engineering problem-solving.

Machine Learning Approaches for Monitoring of Tool Wear during Grey Cast-Iron Turning.

The dynamic development of new technologies enables the optimal computer technique choice to improve the required quality in today’s manufacturing industries. One of the methods of improving the determining process is machine learning. This paper compares different intelligent system methods to identify the tool wear during the turning of gray cast-iron EN-GJL-250 using carbide cutting inserts. During these studies, the experimental investigation was conducted with three various cutting speeds vc (216, 314, and 433 m/min) and the exact value of depth of cut ap and federate f. Furthermore, based on the vibration acceleration signals, appropriate measures were developed that were correlated with the tool condition. In this work, machine learning methods were used to predict tool condition; therefore, two tool classes were proposed, namely usable and unsuitable, and tool corner wear VBc = 0.3 mm was assumed as a wear criterium. The diagnostic measures based on acceleration vibration signals were selected as input to the models. Additionally, the assessment of significant features in the division into usable and unsuitable class was caried out. Finally, this study evaluated chosen methods (classification and regression tree, induced fuzzy rules, and artificial neural network) and selected the most effective model.

An Accurate Prediction Algorithm of RUL for Bearings: Time-Frequency Analysis Based on MRCNN

Remaining useful life (RUL) prediction of mechanical components is of high research value in the field of prognostics and health management (PHM). However, RUL prediction problems are completely challenging due to the complicacy of bearings’ operating environment. In this paper, we transform the vibration acceleration signal collected by sensors into a time-frequency domain matrix through continuous wavelet transform (CWT) and then extract the features of the time-frequency domain matrix through the proposed multiscale residual convolutional neural network (MRCNN), which enables the model to extract more local and global features while constructing more accurate health indicators (HI). In order to highlight the degradation trend of mechanical components, the obtained health indicators are smoothed by exponential moving average (EMA). Finally, linear regression is exploited to predict the RUL of the bearing. Performance evaluations based on the public dataset PRONOSTIA demonstrate the effectiveness of our proposed algorithm, which is superior to existing data-driven algorithms in terms of prediction accuracy.

Vibration analysis during AZ31 magnesium alloy milling with the use of different toolholder types

Machining vibrations are an important issue as they occur in all types of machining processes. Due to its negative impact on machining results, this phenomenon is undesirable, and so there have been continuous efforts to find solutions that will minimise it, and thus improve the stability and safety of the machining process. The paper attempts to determine the impact of toolholder type and cutting condition on the vibrations generated while milling an AZ31 magnesium alloy. The tests were performed using the three most common types of toolholders: ER, Shrink Fit and hydraulic. The vibration displacement and acceleration signals were analysed based on parameters such as Peak-to-Peak, Peak, and Root Mean Square. Composite Multiscale Entropy was also applied to check the stability of cutting processes and define the level of signal irregularity. To determine the frequencies of vibrations and to detect chatter vibrations Fast Fourier Transform was performed. This provides information on the stability and enables vibrations to be minimized by avoiding unfavourable cutting conditions.

Comparative study of stator current-based and vibration-based methods for railway traction motor bearing cage fault diagnosis at high-speed condition

Traction motor’s rolling element bearings are one of the most critical components for the structural health monitoring of high-speed train vehicles. Most structural health monitoring techniques for railway rolling element bearings are using vibration signal. However, the vibration characteristic of cage defect is difficult to be transmitted to external monitoring point owing to the location of cage in bearing. Hence, the cage defect of Railway Traction Motor Bearing (RTMB) is usually difficult to be recognized with motor housing vibration acceleration signal. Considering Induction Motor (IM) itself a sensor, the mechanical imbalance of bearing cage defect can reflect in stator current through the motor gap magnetic field rather than a solid transmission path. A novel strategy with stator current for monitoring the traction motor bearing cage health status in High-Speed Train (HST) is proposed in this paper. This method requires no additional sensor in the practical application. To prove the strategy is feasible and more effective than vibration-based method for the cage fault diagnosis of RTMB, a comprehensive experiment investigation with a real railway traction motor of a HST is completed. Artificial cage defects are processed on bearings at both ends of the induction motor, respectively. Three high-speed (≥250 km/h) conditions are observed. The data analysis results show that vibration-based method is almost invalid at high-speed condition and the proposed current-based method is feasible and effective for the fault diagnosis of high-speed railway traction motor bearing cage.

Fault Diagnosis of Axle Box Bearing with Acoustic Signal Based on Chirplet Transform and Support Vector Machine

Acoustic fault diagnosis technology equipment is non-contact, and the acoustic signal is easy to access. However, it is difficult to extract the feature information of the acoustic signal with low signal-to-noise ratio (SNR). In this paper, a fault diagnosis model (FDM) of axle box bearing based on Chirplet transform (CT) and support vector machine (SVM) is established to diagnose bearing fault based on acoustic signal. The availability of the model is verified by comparing with the vibration acceleration signal bearing fault diagnosis results, and the correctness of the model is verified by utilizing the open database of Western Reserve University. The acoustic-vibration comprehensive bearing fault diagnosis experiment platform (AVEP) is established to investigate the acoustic signal and acceleration signal diagnosis accuracy. The results suggest that, based on the FDM, the diagnosis accuracy and stability of acoustic signal are not as good as acceleration signal when the number of samples is small under the single condition; the diagnosis accuracy of acoustic signal is similar to that of acceleration signal when the number of samples is enough under the multiple condition, which provides a reference for the application of acoustic fault diagnosis technology in engineering in the future.

Deep subdomain generalisation network for health monitoring of high-speed train brake pads

Uneven wear at different locations in high-speed train brake pads creates inconsistent vibrations, which make the criteria learned by the intelligent model hard to standardise to determine friction block failure. Variable friction-induced vibration makes brake pad health monitoring a cross-domain diagnostic problem. This study proposes a deep subdomain generalisation network for online monitoring of the health status of train brake pads. In contrast to existing transfer learning methods, the proposed approach does not assume the availability of test data during training. The model is generalised using unsupervised learning to new scenarios with unknown specific locations of uneven wear. The network reduces the discrepancy explicitly among the distributions of relevant source subdomains using local maximum mean discrepancy, which helps subdomains with identical labels more accurately find an appropriate common subspace. Furthermore, Bayesian optimisation is used to unify feature extraction, domain generalisation and hyper-parameter optimisation into a framework. The vibration–acceleration signals of different friction blocks are collected, and generalisation experiments are performed on both braking-friction and rotating-machinery datasets. The results indicate that the proposed model outperforms the other domain generalisation approaches and accurately identifies the state of a target friction block.

Investigation of a wheel-rail roller rig for wheel-track defect detection

The wheel-rail defect detection method based on the dynamic response of vehicle and track has attracted much attention, and the rig-based experiment is of great significance to the detection algorithm. Aiming at the common defects of wheel and track, a 1:4 scaled wheel-rail roller rig for wheel-track defect detection with one wheel and one roller is investigated. The wheel has two profiles of the railway vehicle wheel, and the roller also has two profiles of rail. This special profile makes the contact between wheel and roller is more stable in the lateral direction and can simulate the wheel-rail contact. The novel wheel-rail roller rig includes vehicle module, track wheel module, vertical loading device, irregularity input device, corrugation excitation device, stiffness and damping simulation device. Meanwhile, the multifunctional control system and multi-channel vibration acceleration signal data acquisition system are developed. The proposed scaled wheel-rail roller rig has the function of simulating various wheel-rail defects. Finally, the simulation experiment results of track irregularity, wheel flat defects, rail corrugation and structure defects under the rail are analysed. The designed scaled wheel-rail roller rig can simulate various wheel-rail defects well, which is conducive to the research of wheel-rail defect detection algorithm.

Fault Diagnosis of Induction Motors with Imbalanced Data Using Deep Convolutional Generative Adversarial Network

A homemade defective model of an induction motor was created by the laboratory team to acquire the vibration acceleration signals of five operating states of an induction motor under different loads. Two major learning models, namely a deep convolutional generative adversarial network (DCGAN) and a convolutional neural network, were applied for fault diagnosis of the induction motor to the problem of an imbalanced training dataset. Two datasets were studied and analyzed: a sufficient and balanced training dataset and insufficient and imbalanced training data. When the training datasets were adequate and balanced, time–frequency analysis was advantageous for fault diagnosis at different loads, with the diagnostic accuracy achieving 95.06% and 96.38%. For the insufficient and imbalanced training dataset, regardless of the signal preprocessing method, the more imbalanced the training dataset, the lower the diagnostic accuracy was for the testing dataset. Samples generated by DCGAN were found to exhibit 80% similarity with the actual data through comparison. By oversampling the imbalanced dataset, DCGAN achieved a 90% diagnostic accuracy, close to the accuracy achieved using a balanced dataset. Among all oversampling techniques, the pro-balanced method yielded the optimal result. The diagnostic accuracy reached 85% in the cross-load test, indicating that the generated data had successfully learned the different fault features that validate the DCGAN’s ability to learn parts of input signals.

Evaluation of the Acceleration Vibration Signal for Aggregates of the Horizontal Drilling Stand

During the operation of each machine, there are dynamic effects causing vibrations. Such a device is also an experimental horizontal drilling stand with aggregates, i.e., a direct current motor (DC), a pump, and a hydro-generator. During their operation, unwanted vibration acceleration signals are generated. It is clear that the accompanying vibration signal carries integrating information about the current state of the drilling rig. Vibration signal processing methods for the time and frequency domains were used. The results of time-domain processing showed significant differences in time waveforms, statistical characteristics, and auto-correlation functions. The auto-correlation function pointed to the periodicity and dependence of the vibrational signal samples. Based on the acquired knowledge, the signals were classified, and a strong source of vibration was determined. Noise is superimposed on the harmonic components of the signals. Amplitude and power spectra were constructed in the frequency domain. Dominant frequencies were identified for each investigated mode in the operating mode. Power spectra removed less significant frequencies and focused on the dominant ones. Time-frequency spectrograms revealed significantly higher frequency bands. The proposed methods can be implemented in diagnosing the operation of the machine and aggregates, determining the source of the greatest vibrations, wear of parts of the equipment such as the drill bit, and recognition of the overall condition of the equipment.

Quantitative detection of locomotive wheel polygonization under non-stationary conditions by adaptive chirp mode decomposition

Wheel polygonal wear is a common and severe defect, which seriously threatens the running safety and reliability of a railway vehicle especially a locomotive. Due to non-stationary running conditions (e.g., traction and braking) of the locomotive, the passing frequencies of a polygonal wheel will exhibit time-varying behaviors, which makes it too difficult to effectively detect the wheel defect. Moreover, most existing methods only achieve qualitative fault diagnosis and they cannot accurately identify defect levels. To address these issues, this paper reports a novel quantitative method for fault detection of wheel polygonization under non-stationary conditions based on a recently proposed adaptive chirp mode decomposition (ACMD) approach. Firstly, a coarse-to-fine method based on the time–frequency ridge detection and ACMD is developed to accurately estimate a time-varying gear meshing frequency and thus obtain a wheel rotating frequency from a vibration acceleration signal of a motor. After the rotating frequency is obtained, signal resampling and order analysis techniques are applied to an acceleration signal of an axle box to identify harmonic orders related to polygonal wear. Finally, the ACMD is combined with an inertial algorithm to estimate polygonal wear amplitudes. Not only a dynamics simulation but a field test was carried out to show that the proposed method can effectively detect both harmonic orders and their amplitudes of the wheel polygonization under non-stationary conditions.

Unmanned aerial vehicle rotor fault diagnosis based on interval sampling reconstruction of vibration signals and a one-dimensional convolutional neural network deep learning method

With the aim of identifying possible mechanical faults in unmanned aerial vehicle (UAV) rotors during operation, this paper proposes a method based on interval sampling reconstruction of vibration signals and one-dimensional convolutional neural network (1D-CNN) deep learning. Firstly, experiments were designed to collect the vibration acceleration signals of a UAV working at high speed under three states (normal, rotor damage by varying degrees, and rotor crack by different degrees). Then, considering the powerful feature extraction and complex data analysis abilities of 1D-CNN, an effective deep learning model for fault identification is established utilizing 1D-CNN. During analysis, it is found that the recognition rate for minor faults is not ideal, with all weak states being identified as normal, reducing the overall identification accuracy, when using conventional sequential sampling to construct learning sample sets. To this end, in order to make the sample data cover the whole process of data collection as much as possible, a learning sample processing method based on interval sampling reconstruction of the vibration signal is proposed. And it is also verified that the reconstructed sample set can easily reflect the global information of mechanical operation. Finally, according to the comparison of analysis results, the recognition rate of the deep learning model for different degrees of faults is greatly improved, and minor faults could also be accurately identified through this method. The results show that the 1D-CNN deep learning model could diagnose and identify UAV rotor damage faults accurately, by incorporating the proposed method of interval sampling reconstruction.

Non-linear dynamic response of misaligned spline coupling: Theoretical modeling and experimental investigation

Premature wear, fatigue and other failures caused by non-linear vibration are one of the key problems of aerodynamic power transmission in spline coupling which maybe suffer from abnormal meshing effects due to misalignment between internal spline and external spline. A non-linear dynamic model of the misaligned spline coupling is established, which involves the calculation of dynamic meshing behavior of misaligned tooth pairs. The movement state of spline coupling can change from a periodic state to a chaotic state with the increase of speed. Under light load, the main resonance speed of system may be modified due to changes in the load. The existence of misalignment promotes the vibration intensity, advances the main resonance speed and increases the instability of spline coupling. In addition, a test bench for the misalignment spline coupling was developed to measure the vibration acceleration signal, and the effect of the misalignment on the dynamic behavior of spline coupling was discussed.

A novel structural damage identification method based on the acceleration responses under ambient vibration and an optimized deep residual algorithm

Recently, structural health monitoring (SHM) methods for civil structures have been investigated widely, especially Deep Learning (DL)-based methods. However, it is usually difficult to fully train a deep neural network, and thus, typical DL-based SHM methods are limited in terms of performance. While addressing these issues, in this paper, a novel methodology is proposed for smart damage identification of frame structures. The newly proposed SHM method is based on raw time-domain structural response signals and deep residual network (DRN). The introduced DRN algorithm has been designed and tested in an effective way for extracting and learning the optimum features of the 1D raw ambient vibration acceleration signals, without any need for engineered features. Also, the network’s performance has been optimized using Bayesian optimization, which clearly enhances the network’s accuracy and information flow across it. Next, the outputs of DRNs are further utilized through new methods for damage size estimation and damage localization. The proposed methodology has been evaluated using the datasets of numerical and experimental frames of the SHM benchmark problem and the dataset of a real-world full-scale truss bridge. The results show that the proposed method is capable of detecting, localizing, and quantifying structural damage accurately for all of the simulated cases of the two examples. Furthermore, conducted comparison studies have approved that the new approach is more efficient than other machine learning-based methods, and it can overcome the major limitations of Artificial intelligence-based SHM models.

ADOPTION OF VIBRATION OF FUZZY PARTIAL FRACTIONAL-ORDER EQUATION IN MONITORING OF NOISE BIG DATA UNDER INFINITE IMPULSE RESPONSE DIGITAL FILTER ALGORITHM

In order to study the application of fuzzy partial fractional-order equation based on infinite impulse response (IIR) digital filter in noise big data monitoring, in this study, first, the noise level is calculated through IIR filter to obtain the sound signal sequence after C weight and A weight; the weight equivalent continuous sound level of C, A, and Z is calculated. Then, the vibration condition of fuzzy partial fractional-order equation is studied; the vibration acceleration signal is converted into the vibration velocity and displacement signal; the corresponding effective values are calculated according to the root mean square (RMS), and the noise monitoring platform is designed from hardware and software aspects. Finally, the calibrated sound level value is compared with the standard signal value generated by the digital signal generator; the sequence of sound signal after weight of the original signal, the actual acceleration signal, velocity signal, and displacement signal are calculated; the acceleration signal, velocity signal, and theoretical signal values are compared at different frequencies. The results show that the error between the calculated and calibrated sound level and the standard signal value is very small, and the results of weight counting of C and A are good and meet the requirement of calculation; the error between the measured value of actual acceleration, velocity, and displacement signal and the theoretical value is small, which can realize the accurate acquisition of vibration signal; the error between the actual measured value and the theoretical value of acceleration and velocity at different frequencies is small, about 1% and 10%, respectively. Therefore, the fuzzy partial fractional-order equation based on IIR digital filter can effectively monitor the big noise data.

THE SUPPORT VECTOR MACHINE PARAMETER OPTIMIZATION METHOD BASED ON ADAPTIVE ELITIST DIFFERENTIAL EVOLUTION ALGORITHM AND ITS APPLICATION TO ROLLER BEARING FAULT DIAGNOSIS

SVM parameters have serious effects on the accuracy rate of classification result. Tuning SVM parameters is always a challenge for scientists. In this paper, a SVM parameter optimization method based on Adaptive Elitist Differential Evolution (AeDE-SVM) is proposed. Furthermore, AeDE-SVM is applied to diagnose roller bearing fault by using complementary ensemble empirical mode decomposition (CEEMD) and singular value decomposition (SVD) techniques. First, original acceleration vibration signals are decomposed into Intrinsic Mode Function (IMFs) by using CEEMD method. Second, initial feature matrices are extracted from (IMFs) by singular value decomposition (SVD) techniques to obtain single values. Third, these values serve as input vector for AeDE-SVM classifier. The results show that the combination of AeDE-SVM classifiers and the CEEMD-SVD method obtains higher classification accuracy and lower cost time compared to other methods. In this paper, the roller bearing vibration signals were used to evaluate the proposed method. The experimental results showed that the superior performance compared to other SVM parameter optimization techniques and successfully recognized different fault types of roller bearing during its operation.

Research on compound fault diagnosis of rolling bearing based on intrinsic time scale decomposition and information fusion

In view of the difficulty to effectively extract compound faults of rolling bearing from aero-engine and precisely identify their types, the paper has proposed a method integrating signal separation algorithm and information fusion. Firstly, the method decomposes the vibration acceleration signals collected by sensors from different positions at the same moment based on intrinsic time scale decomposition algorithm. Secondly, cross correlation analysis is given to the proper rotation component (PRC) of the same layer, which are obtained after decomposition and correspond to the sensors from different positions and cross-correlation function is introduced to embody information fusion. Thirdly, signals are reconstructed according to cross-correlation function of each PRC. Finally, based on the frequency spectrum of reconstructed signal, extract the characteristics of rolling bearing and identify the type of faults under different sensor combinations and multiple compound fault types. The result shows, the proposed method can effectively extract the characteristics of compound faults of bearing and precisely identify the type of faults under different sensor combinations and multiple compound fault types of rolling bearing.

Intelligent fault diagnosis of rolling bearings under imbalanced data conditions using attention-based deep learning method

Data-driven intelligent method has been widely used in fault diagnostics. However, it is observed that previous research studies focusing on imbalanced datasets for fault diagnosis have a limitation, that is, the number of normal and fault samples is assumed to be same or similar in the diagnosis process. This hypothesis decreases the accuracy and stability of fault diagnosis model for imbalanced datasets in practical working conditions. In this paper, a rolling bearing fault diagnosis model which combines Dual-stage Attention-based Recurrent Neural Network (DA-RNN) and Convolutional Block Attention Module (CBAM) is proposed. Firstly, the DA-RNN model is used to extend imbalanced datasets in real fault diagnosis cases. Secondly, an image processing method is designed to convert vibration signal into image by using vibration acceleration signal and its corresponding integrated velocity and displacement signals. Finally, the Convolution Neural Network (CNN) model with embedded CBAM structure is used for fault classification. Two datasets of vibration data from rolling bearings are used to evaluate the performance of the proposed methodology for fault diagnosis. Results show that the proposed DARNN-CBAM-CNN method improved fault diagnosis accuracy in 10.90%, 7.56%, 2.73% and 1.90% the performance metrics compared to a neural network based method, a machine learning based method, a deep learning based method, and a DARNN-CNN based method without using CBAM when the imbalance ratio of the dataset is 100:50. Diagnosis accuracy results of datasets with four different imbalance ratios show that the proposed method has the best performance compared to other six intelligent fault diagnosis methods, indicating that the proposed method is a promising potential for rolling bearings under imbalanced data conditions.

Evaluation of Wear of Disc Brake Friction Linings and the Variability of the Friction Coefficient on the Basis of Vibroacoustic Signals

The article presents the results of friction and vibroacoustic tests of a railway disc brake carried out on a brake stand. The vibration signal generated by the friction linings provides information on their wear and offers evaluation of the braking process, i.e., changes in the average friction coefficient. The algorithm presents simple regression linear and non-linear models for the thickness of the friction linings and the average coefficient of friction based on the effective value of vibration acceleration. The vibration acceleration signals were analyzed in the amplitude and frequency domains. In both cases, satisfactory values of the dynamics of changes above 6 dB were obtained. In the case of spectral analysis using a mid-band filter, more accurate models of the friction lining thickness and the average coefficient of friction were obtained. However, the spectral analysis does not allow the estimation of the lining thickness and the friction coefficient at low braking speeds, i.e., 50 and 80 km/h. The analysis of amplitudes leads to the determination of models in the entire braking speed range from 50 to 200 km/h, despite the lower accuracy compared to the model, based on the spectral analysis. The vibroacoustic literature presents methods of diagnosis of the wear of various machine elements such as bearings or friction linings, based on amplitude or frequency analysis of vibrations. These signal analysis methods have their limitations with regard to their scope of use and the accuracy of diagnosis. There are no cases of simultaneous use of different methods of analysis. This article presents the simultaneous application of the amplitude and frequency methods in the analysis of vibroacoustic signals generated by brake linings. Moreover, algorithms for assessing the wear of friction linings and the average coefficient of friction were presented. The algorithm enables determination of the time at which the friction linings should be replaced with new ones. The final algorithm analyzes the vibration acceleration signals using both amplitude analysis for low braking speeds, as well as spectral analysis for medium and high braking speeds.