The research of optimal selection method for wavelet packet basis in compressing the vibration signal of a rolling bearing in fans and pumps

Modern machinery becomes more precious with the advance of science, and fault diagnosis is vital for avoiding economical losses or casualties. Among massive diagnosis methods, deep learning algorithms stand out to open an era of intelligent fault diagnosis. Deep residual networks are the state-of-the-art deep learning models which can continuously improve performance by deepening the network structures. However, in vibration-based fault diagnosis, the transient property instability of vibration signal usually calls for time–frequency analysis methods, and the characters of time–frequency matrices are distinct from standard images, which brings some natural limitations for the diagnosis performance of deep learning algorithms. To handle this issue, an enhanced deep residual network named the multilevel correlation stack-deep residual network is proposed in this article. Wavelet packet transform is used to preprocess the sensor signal, and then the proposed multilevel correlation stack-deep residual network uses kernels with different shapes to fully dig various kinds of useful information from any local regions of the processed input. Experiments on two rolling bearing datasets are carried out. Test results show that the multilevel correlation stack-deep residual network exhibits a more satisfactory classification performance than original deep residual networks and other similar methods, revealing significant potentials for realistic fault diagnosis applications.

Download Full-text

Rolling bearing fault detection approach based on improved dispersion entropy and AFSA optimized SVM

International Journal of Electrical Engineering Education ◽

10.1177/0020720920940584 ◽

2020 ◽

pp. 002072092094058

Author(s):

Wuqiang Liu ◽

Jinxing Shen ◽

Xiaoqiang Yang

Keyword(s):

Fault Detection ◽

Wavelet Packet ◽

Vibration Signal ◽

Rolling Bearing ◽

Classification Model ◽

Coarse Grained ◽

Fault Classification ◽

Support Vector ◽

Artificial Fish Swarm Algorithm ◽

Artificial Fish Swarm

The support vector machine (SVM) does not have a fixed parameter selection method and the manual selection of parameters is difficult to determine the validity, which affects the accuracy of recognition. simultaneously, The existing coarse-grained approach cannot effectively analyze the high-frequency components of time series. In view of the shortcomings of the above method, we put forward a new technique of rolling bearings for fault detection, which combines wavelet packet dispersion entropy (WPDE) and artificial fish swarm algorithm (AFSA) optimize support vector machines (AFSA-SVM). First of all, wavelet packet is devoted to decompose the original vibration signal into components of different frequency bands. Secondly, the dispersion entropy (DE) are calculated for each of the obtained frequency band components to acquire more comprehensive and complete fault information. Afterward, Input feature samples into the SVM model for training, and AFSA is used to optimize the parameters of SVM to obtain the optimal value so as to establish the best classification model. Finally, the prepared test set is input into AFSA-SVM for fault classification. The achievement of bearing detection experiments show that this approach can accurately and quickly identify fault types.

Download Full-text

Rolling Bearing Fault Detection Method Based on Wavelet Packet Energy Spectrum - PLS

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.971-973.697 ◽

2014 ◽

Vol 971-973 ◽

pp. 697-700 ◽

Cited By ~ 1

Author(s):

Fang Qian Li ◽

Yu Gang Fan ◽

Ya Xiong Zhang

Keyword(s):

Energy Spectrum ◽

Fault Detection ◽

Detection Method ◽

Feature Vector ◽

Wavelet Packet ◽

Vibration Signal ◽

Rolling Bearing ◽

Frequency Bands ◽

Detection Model ◽

Spectrum Feature

For the problem of rolling bearing mechanical fault detection, this paper proposed a fault detection method based on wavelet packet energy spectrum and partial least squares (PLS). Firstly, the collected vibration signal was decomposed by wavelet packet into independent frequency bands. The different frequencies energy spectrum could be collected from the independent frequency bands. That means the energy spectrum feature vectors whose changes reflect the running state could be constructed. Then analyze the energy spectrum feature vector by PLS, and establish the fault detection model. The fault could be detected by the T2 and SPE statistics. Finally, the experimental results showed that the method is feasible and effective.

Download Full-text

Automated Bearing Fault Diagnosis using Packet Features of Vibration Signal and Gaussian Support Vector Machine

International Journal of Recent Technology and Engineering - 2 ◽

10.35940/ijrte.f1014.0386s20 ◽

2020 ◽

Vol 8 (6S) ◽

pp. 72-81

Keyword(s):

Support Vector Machine ◽

Wavelet Packet ◽

Vibration Signal ◽

Rolling Bearing ◽

True Positive Rate ◽

Support Vector ◽

True Negative ◽

Bearing Fault ◽

Dimensional Reduction Method ◽

Positive Rate

Effective detection of the bearing fault and, specifically performance dilapidation assessment of a bearing is the topic of intensive analysis that may scale back prices and therefore the nonscheduled down time. This article presents an adaptive approach that is based on Bhattacharya space ranking method and dimensional reduction method as general discriminate analysis (GDA) with Gaussian support vector machine (GSVM) to accurately detect the defects of rolling bearing. For this investigation, first, vibration signal generated by rolling bearing was disintegrated to five levels employing wavelet packet (WP) method. Sixty three logarithmic wavelet packet features (LWPFs) were taken out from five level disintegrated vibration signals. After this, sixty three features were ranked by Bhattacharya space and top ten LWPFs were chosen. The top ten features were reduced to a new feature using GDA for effective detection and then applied to GSVM for detection of bearing fault. The experimental results show that new automated diagnosing approach attained classifier performance parameters as sensitivity (SE) or true positive rate, specificity (SP) or true negative rate, accuracy (AC) and positive prediction value (PPV) of 100, 98.50, 100 and 99.67 % for inner raceway (IR) and, AC: 99.49, SE: 100, SP: 98.78 and PPV: 99.87 for ball bearing (BB) at 0.18 mm diameter faults.

Download Full-text

Varying Rotation Speed Bearing Fault Features and Extraction Method Based on Wavelet Packet

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.644-650.286 ◽

2014 ◽

Vol 644-650 ◽

pp. 286-289

Author(s):

Bin Wu ◽

Shan Ping Yu ◽

Yue Gang Luo ◽

Chang Jian Feng

Keyword(s):

Rotation Speed ◽

Hydrodynamic Lubrication ◽

Contact Stiffness ◽

Wavelet Packet ◽

Vibration Signal ◽

Rolling Bearing ◽

Bearing Fault ◽

Resonance Frequencies ◽

Bearing Vibration ◽

The Impact

When bearing rotates, it comes with elastic hydrodynamic lubrication effect. Interaction between the effect and the bearing vibration leads to the change of lubricant film thickness, thus, contact stiffness of contact pair changes along with the rotation speed of the bearing, and then the resonance frequencies of the bearing system changes according to the rotation speed. In addition, the impact signal of varying speed bearing damage point no longer has the periodic characteristics. Based on the analysis of the bearing failure mechanism, this paper proposed a varying speed bearing vibration signal fault model, and also utilizes wavelet packet to extract the bearing fault signal by means of a variable speed rolling bearing vibration experiment table.

Download Full-text

Experimental Investigation for Fault Diagnosis Based on a Hybrid Approach Using Wavelet Packet and Support Vector Classification

The Scientific World JOURNAL ◽

10.1155/2014/145807 ◽

2014 ◽

Vol 2014 ◽

pp. 1-10 ◽

Cited By ~ 6

Author(s):

Pengfei Li ◽

Yongying Jiang ◽

Jiawei Xiang

Keyword(s):

Fault Diagnosis ◽

Wavelet Packet ◽

Hybrid Approach ◽

Vibration Signal ◽

Rolling Bearing ◽

Support Vector ◽

Gear Fault ◽

Mechanical Faults ◽

Energy Ratios ◽

Robust To Noise

To deal with the difficulty to obtain a large number of fault samples under the practical condition for mechanical fault diagnosis, a hybrid method that combined wavelet packet decomposition and support vector classification (SVC) is proposed. The wavelet packet is employed to decompose the vibration signal to obtain the energy ratio in each frequency band. Taking energy ratios as feature vectors, the pattern recognition results are obtained by the SVC. The rolling bearing and gear fault diagnostic results of the typical experimental platform show that the present approach is robust to noise and has higher classification accuracy and, thus, provides a better way to diagnose mechanical faults under the condition of small fault samples.

Download Full-text

Composite Fault Diagnosis of Rolling Bearing Based on Optimized Wavelet Packet AR Spectrum Energy Entropy Combined with Adaptive No Velocity Term PSO-SOM-BPNN

Journal of Sensors ◽

10.1155/2021/4138652 ◽

2021 ◽

Vol 2021 ◽

pp. 1-15

Author(s):

Hongwei Fan ◽

Yang Yan ◽

Xuhui Zhang ◽

Xiangang Cao ◽

Jiateng Ma

Keyword(s):

Fault Diagnosis ◽

Wavelet Packet ◽

Back Propagation ◽

Vibration Signal ◽

Rolling Bearing ◽

Back Propagation Neural Network ◽

Spectral Energy ◽

Self Organizing Map ◽

Learning Factor ◽

Energy Entropy

Aiming at the problem of low diagnosis efficiency and accuracy, due to noise and cross aliasing among various faults when diagnosing composite faults of rolling bearing under actual working conditions, a composite fault diagnosis method of rolling bearing based on optimized wavelet packet autoregressive (AR) spectral energy entropy and adaptive no velocity term particle swarm optimization-self organizing map-back propagation neural network (ANVTPSO-SOM-BPNN) is proposed. The energy entropy feature is extracted from the bearing vibration signal through wavelet packet AR spectrum, and SOM and BPNN are combined to form a series network. For PSO, the velocity term is discarded and the inertia weight and learning factor are adaptively adjusted. Finally, the Dempster-Shafer (D-S) evidence fusion diagnosis is carried out. To get closer to the application condition, the data are collected near and far away from the fault point for the composite fault diagnosis, which verifies the effectiveness of the proposed method.

Download Full-text

Machine Condition Classification by Using Wavelet Packet Decomposition and Multi-Scale Entropy

Advanced Engineering Forum ◽

10.4028/www.scientific.net/aef.2-3.743 ◽

2011 ◽

Vol 2-3 ◽

pp. 743-748

Author(s):

Hong Kun Li ◽

Shu Ai Zhou ◽

Yu Zhen Chen

Keyword(s):

Frequency Band ◽

Working Condition ◽

Feature Vector ◽

Wavelet Packet ◽

Vibration Signal ◽

Rolling Bearing ◽

Operating Conditions ◽

Wavelet Packet Decomposition ◽

Multi Scale ◽

Series Of Experiments

A new condition classification method is put forward based on the analysis of vibration signals. Machine working condition can be recognized by the combination of wavelet packet decomposition (WPD) and multi-scale entropy (MSE). Firstly, vibration signal of machine is decomposed by wavelet packet with the appropriate decomposition layer. Then, each sub-signal in different frequency band is analyzed with the multi-scale entropy. Through analyzing the multi-scale entropy distribution curves of sub-signals for different operating conditions in each frequency band, entropy of certain frequency bands and scales will be chosen as the feature vector, which is used to distinguish different machine conditions. This method presents a novel perspective for rolling bearing default diagnosis and is tested to be very effective to classify different bearing operating conditions through series of experiments.

Download Full-text

The Feature Extraction of Rolling Bearing Fault Based on Wavelet Packet-EMD Energy Distribution

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.233.234 ◽

2012 ◽

Vol 233 ◽

pp. 234-238 ◽

Cited By ~ 1

Author(s):

Cheng Wen ◽

Chuan De Zhou

Keyword(s):

Energy Distribution ◽

Wavelet Packet ◽

Vibration Signal ◽

Rolling Bearing ◽

Resonance Signal ◽

Outer Race ◽

Bearing Fault ◽

Noise Characteristics ◽

Mode Decomposition ◽

Simulation Signal

The new signal analysis method based on the combination of wavelet packet and empirical mode decomposition (EMD) energy distribution was proposed for rolling bearing vibration signal presenting modulating characteristic, non-stationary characteristics and containing a lot of noise characteristics. In this method, initial vibration signal was decomposed first by wavelet packet to extract the resonance signal with obvious modulating characteristics. Then the resonance signal was decomposed by EMD method and energy distribution of each Intrinsic Mode Function (IMF) was obtained. Finally the IMF component, which can reflect the vibration condition, was processed by Hilbert envelope demodulation to extract rolling bearing fault characteristics information. The application analysis of the simulation signal and fault signal of inner race, outer race and rolling element of rolling bearing shows that this method can effectively analyze rolling bearing fault information and realize the fault diagnosis.

Download Full-text

Optimal Selection Method of Gyroscope Groups Based on Dedicated Observer Scheme

2019 Chinese Automation Congress (CAC) ◽

10.1109/cac48633.2019.8997061 ◽

2019 ◽

Author(s):

Liliang Li ◽

Yanzhao Zhang ◽

Yi Shen ◽

Shang Sun ◽

Kaicheng Zhang

Keyword(s):

Selection Method ◽

Optimal Selection

Download Full-text