Fault Diagnosis Method of Rolling Bearing Based on Ensemble Local Mean Decomposition and Neural Network

2013 ◽  
Vol 300-301 ◽  
pp. 714-720 ◽  
Author(s):  
Xing Zhi Liao ◽  
Zhou Wan ◽  
Yi Yang Li ◽  
Li Cheng
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Wavelet Packet ◽  
Recognition Rate ◽  
Rolling Bearing ◽  
Vibrational Signal ◽  
Local Mean Decomposition ◽  
Outer Race ◽  
Local Mean ◽  
Working Status

For a problem of mode mixing occurs in implementation process of local mean decomposition (LMD) method, an analytical method based on ensemble local mean decomposition (ELMD) and neural network is proposed to apply to fault diagnosis of rolling bearing, the vibrational signal of rolling bearing is decomposed into a series of product functions(PF) by ELMD method. The PF components which contain main fault information are selected to perform a further analysis. The kurtosis coefficient and energy characteristic parameters extracted from these PF components can be used as the input parameters of the neural network to identify the working status and fault types of rolling bearing. Through the analysis of rolling bearing with fault-free, inner-race fault and outer-race fault, the results indicate that the method based on ELMD and neural network has a higher failure recognition rate than the method based on wavelet packet analysis and neural network, and the working status and fault types of rolling bearing can be identified accurately and effectively.

The Application of Local Mean Decomposition and Variable Predictive Model-Based Class Discriminate in Gear Fault Diagnosis

2014 ◽  
Vol 1014 ◽  
pp. 510-515 ◽  
Author(s):  
You Cai Xu ◽  
Xin Shi Li ◽  
Ran Tao ◽  
Shu Guo ◽  
Min Gou ◽  
...  
Keyword(s):  
Fault Diagnosis ◽  
Predictive Model ◽  
Feature Vector ◽  
Local Mean Decomposition ◽  
Outer Race ◽  
Model Based ◽  
Gear Fault ◽  
Local Mean ◽  
Gear Fault Diagnosis ◽  
Non Stationary Signal

The time-domain energy message conveyed by vibration signals of different gear fault are different, so a method based on local mean decomposition (LMD) and variable predictive model-based class discriminate (VPMCD) is proposed to diagnose gear fault model. The vibration signal of gear which is the research object in this paper is decomposed into a series of product functions (PF) by LMD method. Then a further analysis is to select the PF components which contain main fault information of gear, the energy feature parameters of the selected PF components are used to form a fault feature vector. The variable predictive model-based class discriminate is a new multivariate classification approach for pattern recognition, through taking fully advantages of the fault feature vector. Finally, gear fault diagnosis is distinguished into normal state, inner race fault and outer race fault. The results show that LMD method can decompose a complex non-stationary signal into a number of PF components whose frequency is from high to low. And the method based on LMD and VPMCD has a high fault recognition function by analyzing the fault feature vector of PF.

scholarly journals Rolling Bearing Fault Diagnosis Based on Wavelet Packet Transform and Convolutional Neural Network

2020 ◽  
Vol 10 (3) ◽  
pp. 770 ◽  
Author(s):  
Guoqiang Li ◽  
Chao Deng ◽  
Jun Wu ◽  
Zuoyi Chen ◽  
Xuebing Xu
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Fault Diagnosis ◽  
Convolutional Neural Network ◽  
Wavelet Packet ◽  
Rolling Bearing ◽  
Wavelet Packet Transform ◽  
Fault Classification ◽  
Bearing Fault ◽  
Bearing Fault Diagnosis

Timely sensing the abnormal condition of the bearings plays a crucial role in ensuring the normal and safe operation of the rotating machine. Most traditional bearing fault diagnosis methods are developed from machine learning, which might rely on the manual design features and prior knowledge of the faults. In this paper, based on the advantages of CNN model, a two-step fault diagnosis method developed from wavelet packet transform (WPT) and convolutional neural network (CNN) is proposed for fault diagnosis of bearings without any manual work. In the first step, the WPT is designed to obtain the wavelet packet coefficients from raw signals, which then are converted into the gray scale images by a designed data-to-image conversion method. In the second step, a CNN model is built to automatically extract the representative features from gray images and implement the fault classification. The performance of the proposed method is evaluated by a real rolling-bearing dataset. From the experimental study, it can be seen the proposed method presents a more superior fault diagnosis capability than other machine-learning-based methods.

scholarly journals A hybrid fault diagnosis method based on singular value difference spectrum denoising and local mean decomposition for rolling bearing

2018 ◽  
Vol 37 (4) ◽  
pp. 928-954 ◽  
Author(s):  
Jun Ma ◽  
Jiande Wu ◽  
Xiaodong Wang
Keyword(s):  
Fault Diagnosis ◽  
Decomposition Method ◽  
Energy Operator ◽  
Rolling Bearing ◽  
Difference Spectrum ◽  
Singular Value ◽  
Local Mean Decomposition ◽  
Diagnosis Method ◽  
Local Mean ◽  
Teager Energy

Rolling bearing is one of the most crucial components in rotating machinery and due to their critical role, it is of great importance to monitor their operation conditions. However, due to the background noise in acquired signals, it is not always possible to identify probable faults. Therefore, signal denoising preprocessing has become an essential part of condition monitoring and fault diagnosis. In the present study, a hybrid fault diagnosis method based on singular value difference spectrum denoising and local mean decomposition for rolling bearing is proposed. First, as a denoising preprocessing method, singular value difference spectrum denoising is applied to reduce the noise of the bearing vibration signal and improve the signal-to-noise ratio. Then, local mean decomposition method is used to decompose the denoised signals into several product functions. And product functions corresponding to the fault feature are selected according to the correlation coefficient criterion. Finally, Teager energy spectrum is analyzed by applying the Teager energy operator to the constructed amplitude modulation component. The proposed method is successfully applied to analyze the vibration signals collected from an experimental motive rolling bearing and rolling bearing of the self-made rotor experimental platform. The experimental results demonstrate that the proposed singular value difference spectrum denoising and local mean decomposition method can achieve fairly or slightly better performance than the normal local mean decomposition-Teager energy operator method, fast kurtogram, and the wavelet denoising and local mean decomposition method.

Rolling Bearing Fault Diagnosis Based on Wavelet Packet- Neural Network Characteristic Entropy

2010 ◽  
Vol 108-111 ◽  
pp. 1075-1079 ◽  
Author(s):  
Li Ying Wang ◽  
Wei Guo Zhao ◽  
Ying Liu
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Bp Neural Network ◽  
Wavelet Packet ◽  
Rolling Bearing ◽  
Wavelet Packet Decomposition ◽  
Intelligent Fault Diagnosis ◽  
Network Characteristic ◽  
Bearing Fault Diagnosis ◽  
Diagnosis Method

On the basis of neural network based on wavelet packet-characteristic entropy(WP-CE) the author proposes a new fault diagnosis method of vibrating of hearings, in which three layers wavelet packet decomposition of the acquired vibrating signals of hearings is performed and the wavelet packet-characteristic entropy is extracted, the eigenvector of wavelet packet of the vibrating signals is constructed，and taking this eigenvector as fault sample the three layers BP neural network is trained to implement the intelligent fault diagnosis. The simulation result from the proposed method is effective and feasible.

Fault Diagnosis Method of Gear Based on Local Mean Decomposition and Least Squares Support Vector Machine

2013 ◽  
Vol 415 ◽  
pp. 548-554
Author(s):  
Zhou Wan ◽  
Xing Zhi Liao ◽  
Xin Xiong ◽  
Zhi Rong Li
Keyword(s):  
Support Vector Machine ◽  
Fault Diagnosis ◽  
Least Squares ◽  
Vibration Signal ◽  
Support Vector ◽  
Vibrational Signal ◽  
Local Mean Decomposition ◽  
Gear Fault ◽  
Input Sample ◽  
Local Mean

For differences of time-domain energy distribution of different gear fault vibration signal, an analytical method based on local mean decomposition (LMD) and least squares support vector machine (LS-SVM) is proposed to apply to gear fault diagnosis. First vibrational signal of gear is decomposed into a series of product functions (PF) by LMD method. Then extracting energy characteristic parameters of PF components which contain main fault information to constitute a fault feature vectors, which is considered as input sample of well-trained LS-SVM, and then identifying working state and fault type of different gear can be identified accurately and effectively by diagnostic method based on LMD and LS-SVM.

scholarly journals Application of Multi-Dimension Input Convolutional Neural Network in Fault Diagnosis of Rolling Bearings

2019 ◽  
Vol 9 (13) ◽  
pp. 2690 ◽  
Author(s):  
Tao Zan ◽  
Hui Wang ◽  
Min Wang ◽  
Zhihao Liu ◽  
Xiangsheng Gao
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Convolutional Neural Network ◽  
Recognition Rate ◽  
Rolling Bearing ◽  
Diagnostic Model ◽  
Original Signal ◽  
Proposed Model ◽  
Convolution Model ◽  
The Stability

Aiming at the problem of poor robustness of the intelligent diagnostic model, a fault diagnosis model of rolling bearing based on a multi-dimension input convolutional neural network (MDI-CNN) is proposed. Compared with the traditional convolution neural network, the proposed model has multiple input layers. Therefore, it can fuse the original signal and processed signal—making full use of advantages of the convolutional neural networks to learn the original signal characteristics automatically, and also improving recognition accuracy and anti-jamming ability. The feasibility and validity of the proposed MDI-CNN are verified, and its advantages are proved by comparison with the other related models. Moreover, the robustness of the model is tested by adding the noise to the test set. Finally, the stability of the model is verified by two experiments. The experimental results show that the proposed model improves the recognition rate, robustness and convergence performance of the traditional convolution model and has good generalization ability.

Sign in / Sign up

Export Citation Format

Share Document