A Multi-Fault Diagnosis Method of Rolling Bearing Based on Wavelet-PCA and Fuzzy K-Nearest Neighbor

2010 ◽  
Vol 29-32 ◽  
pp. 1602-1607 ◽  
Author(s):  
Xiang Shun Chen ◽  
Hu Biao Zeng ◽  
Zhi Xiong Li
Keyword(s):  
Fault Diagnosis ◽  
Nearest Neighbor ◽  
Rolling Bearing ◽  
Fault Detection And Diagnosis ◽  
Operating Conditions ◽  
K Nearest Neighbor ◽  
Rolling Bearings ◽  
Classification Rate ◽  
Wavelet Pca ◽  
Original Feature

Rolling bearings are widely used in various areas including aircraft, mining, manufacturing, and agriculture, etc. The breakdowns of the rotational machinery resulted from the rolling bearing failures account for 30%. It is therefore imperative to monitor the rolling bearing conditions in time in order to prevent the malfunctions of the plants. In the present paper is described a fault detection and diagnosis technique for rolling bearing multi-faults based on wavelet-principle component analysis (PCA) and fuzzy k-nearest neighbor (FKNN). In the diagnosis process, the wavelet analysis was firstly employed to decompose the vibration data of the rolling bearings under eight different operating conditions, and for each sample its energy of each sub-band was calculated to obtain the original feature space. Then, the PCA was used to reduce the dimensionality of the original feature vector and hence the most important features could be gotten. Lastly, the FKNN algorithm was employed in the pattern recognition to identify the conditions of the bearings of interest. The experimental results suggest that the sensitive fault features can be extracted efficiently after the wavelet-PCA processing, and the proposed diagnostic system is effective for the rolling bearing multi-fault diagnosis. In addition, the proposed method can achieve higher performance than that without PCA with respect to the classification rate.

scholarly journals Fault Diagnosis Method for Rolling Bearings Based on Composite Multiscale Fluctuation Dispersion Entropy

Entropy ◽  
2019 ◽  
Vol 21 (3) ◽  
pp. 290 ◽  
Author(s):  
Xiong Gan ◽  
Hong Lu ◽  
Guangyou Yang
Keyword(s):  
Fault Diagnosis ◽  
Nearest Neighbor ◽  
Coarse Grained ◽  
K Nearest Neighbor ◽  
Rolling Bearings ◽  
Entropy Estimation ◽  
Fault Recognition ◽  
Knn Classifier ◽  
Diagnosis Method ◽  
The Stability

This paper proposes a new method named composite multiscale fluctuation dispersion entropy (CMFDE), which measures the complexity of time series under different scale factors and synthesizes the information of multiple coarse-grained sequences. A simulation validates that CMFDE could improve the stability of entropy estimation. Meanwhile, a fault recognition method for rolling bearings based on CMFDE, the minimum redundancy maximum relevancy (mRMR) method, and the k nearest neighbor (kNN) classifier (CMFDE-mRMR-kNN) is developed. For the CMFDE-mRMR-kNN method, the CMFDE method is introduced to extract the fault characteristics of the rolling bearings. Then, the sensitive features are obtained by utilizing the mRMR method. Finally, the kNN classifier is used to recognize the different conditions of the rolling bearings. The effectiveness of the proposed CMFDE-mRMR-kNN method is verified by analyzing the standard experimental dataset. The experimental results show that the proposed fault diagnosis method can effectively classify the conditions of rolling bearings.

Hybrid data fusion approach for fault diagnosis of fixed-axis gearbox

2017 ◽  
Vol 17 (4) ◽  
pp. 936-945 ◽  
Author(s):  
Vanraj ◽  
SS Dhami ◽  
BS Pabla
Keyword(s):  
Fault Diagnosis ◽  
Empirical Mode Decomposition ◽  
Nearest Neighbor ◽  
Energy Operator ◽  
Operating Conditions ◽  
Maintenance Cost ◽  
K Nearest Neighbor ◽  
Forward Selection ◽  
Statistical Measures ◽  
Mode Decomposition

Intelligent fault diagnosis system based on condition monitoring is expected to assist in the prevention of machine failures and enhance the reliability with lower maintenance cost. Most machine breakdowns related to gears are a result of improper operating conditions and loading, hence leads to failure of the whole mechanism. With advancement in technology, various gear fault diagnosis techniques have been reported which primarily focus on vibration analysis with statistical measures. However, acoustic signals posses a huge potential to monitor the status of the machine but a few studies have been carried out till now. This article describes the implementation of Teager–Kaiser energy operator and empirical mode decomposition methods for fault diagnosis of the gears using acoustic and vibration signals by extracting statistical features. A cross-correlation-based fault index that assists the automatic selection of the sensitive Intrinsic Mode Function (IMF) containing fault information has also been described. The features extracted by all combinations of signal processing techniques are sorted by order of relevance using floating forward selection method. The effectiveness is demonstrated using the results obtained from the experiments. The fault diagnosis is performed with k-nearest neighbor classifier. The results show that the hybrid of empirical mode decomposition–Teager–Kaiser energy operator techniques employs the advantages traits of one or the other technique to generate overall improvement in diagnosing severity of local faults.

A Fault Diagnosis Method of Rolling Bearing through Wear Particle and Vibration Analyses

2010 ◽  
Vol 26-28 ◽  
pp. 676-681 ◽  
Author(s):  
Zhong Yu Huang ◽  
Zhi Qiang Yu ◽  
Zhi Xiong Li ◽  
Yuan Cheng Geng
Keyword(s):  
Fault Diagnosis ◽  
Vibration Analysis ◽  
Wear Particle ◽  
Rolling Bearing ◽  
Particle Analysis ◽  
Complex Nature ◽  
Chemical Compositions ◽  
Rolling Bearings ◽  
Classification Rate ◽  
Wear Particle Analysis

Wear particle and vibration analysis are the two main condition monitoring techniques for machinery maintenance and fault diagnosis in industry. Due to the complex nature of machinery, these two techniques can only diagnose about 30% to 40% of faults when used independently. Therefore, it is critical to integrate vibration analysis and wear particle analysis to provide a more effective maintenance program. This paper presents a new fault diagnosis approach of rolling bearings via the combination of vibration analysis and wear particle analysis. Both the tribological and vibrant information of the rolling bearings with typical faults were collected by an experimental test rig. Wear particle analysis was applied to the oil samples to obtain the wear particle number and size distribution, the particle texture and the chemical compositions, etc. Vibration analysis was used to get the time and frequency characteristics of the vibration data. Then, an intelligent data fusion method based on the genetic algorithm based fuzzy neural network was employed to identify the rolling bearing conditions. The analysis results suggest that the proposed method is more feasible and effective for the rolling bearing fault diagnosis than separated use of the two techniques with respect to the classification rate, and thus has application importance.

scholarly journals Deep transfer learning for rolling bearing fault diagnosis under variable operating conditions

2019 ◽  
Vol 11 (12) ◽  
pp. 168781401989721 ◽  
Author(s):  
Changchang Che ◽  
Huawei Wang ◽  
Qiang Fu ◽  
Xiaomei Ni
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Convolutional Neural Network ◽  
Transfer Learning ◽  
Rolling Bearing ◽  
Operating Conditions ◽  
Feature Representation ◽  
Target Domain ◽  
Rolling Bearings ◽  
Variable Operating Conditions

Rolling bearings are the vital components of rotary machines. The collected data of rolling bearing have strong noise interference, massive unlabeled samples, and different fault features. Thus, a deep transfer learning method is proposed for rolling bearings fault diagnosis under variable operating conditions. To obtain robust feature representation, the denoising autoencoder is used to denoise and reduce dimension of unlabeled rolling bearing signals. For those unlabeled target domain signals, a feature matching method based on multi-kernel maximum mean discrepancies between source domain and target domain is adopted to get enough labeled target domain samples. Then, these rolling bearing signals are converted to multi-dimensional graph samples and fed into a convolutional neural network model for fault diagnosis. To improve the generalization of convolutional neural network under variable operating conditions, we combine model-based transfer learning with feature-based transfer learning to initialize and optimize the convolutional neural network parameters. The effectiveness of the proposed method is validated through several comparative experiments of Case Western Reserve University data. The results demonstrate that the proposed method can learn features adaptively from noisy data and increase the accuracy rate by 2%–8% comparing with other models.

scholarly journals Fault Diagnosis of Rolling Bearing Based on Improved VMD and KNN

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Quanbo Lu ◽  
Xinqi Shen ◽  
Xiujun Wang ◽  
Mei Li ◽  
Jia Li ◽  
...  
Keyword(s):  
Fault Diagnosis ◽  
Nearest Neighbor ◽  
Recognition Rate ◽  
Real Life ◽  
Rolling Bearing ◽  
Fault Classification ◽  
End Effect ◽  
K Nearest Neighbor ◽  
Basic Algorithm ◽  
Mirror Extension

Variational modal decomposition (VMD) has the end effect, which makes it difficult to efficiently obtain fault eigenvalues from rolling bearing fault signals. Inspired by the mirror extension, an improved VMD is proposed. This method combines VMD and mirror extension. The mirror extension is a basic algorithm to inhibit the end effect. A comparison is made with empirical mode decomposition (EMD) for fault diagnosis. Experiments show that the improved VMD outperforms EMD in extracting the fault eigenvalues. The performance of the new algorithm is proven to be effective in real-life mechanical fault diagnosis. Furthermore, in this article, combining with singular value decomposition (SVD), fault eigenvalues are extracted. In this way, fault classification is realized by K-nearest neighbor (KNN). Compared with EMD, the proposed approach has advantages in the recognition rate, which can accurately identify fault types.

Fault Diagnosis of Rolling Bearings Based on Multi-Scale Entropy and Ensembled Artificial Neural Network

2014 ◽  
Author(s):  
Fen Chen ◽  
Quan Liu ◽  
Qin Wei ◽  
Deng Ting ◽  
Yan Ting ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Fault Diagnosis ◽  
Mechanical System ◽  
Back Propagation ◽  
Rolling Bearing ◽  
Fault Detection And Diagnosis ◽  
Rolling Bearings ◽  
Multi Scale ◽  
Artificial Neural

Rolling bearing is widely used in rotating mechanical system, and its operating state has great effects on availability, reliability and the life cycle of whole mechanical system. Therefore, fault diagnosis of rolling bearing is indispensable for the health monitoring in rotating machinery system. In this paper, a method based on multi-scale entropy (MSE) and ensembled artificial neural network (EANN) is proposed for feature extraction and fault recognition in rolling bearings respectively. MSE is mainly in charge for quantizing the complexity of the nonlinear time series in different scales. Then, EANN is employed to identify various faults of rolling bearing after overcoming the two disadvantages like local minimization and slow convergence speed in back propagation neural network (BPNN). The experimental results indicate that the method based on MSE and EANN is feasible and effective to classify different categories of faults and to identify the severity level of fault in the rolling bearings. Therefore, it is available for fault detection and diagnosis in rolling bearings with good performance.

An intelligent fault diagnosis method based on domain adaptation for rolling bearings under variable load conditions

2021 ◽  
pp. 095440622110329
Author(s):  
Jianqun Zhang ◽  
Qing Zhang ◽  
Xianrong Qin ◽  
Yuantao Sun
Keyword(s):  
Feature Extraction ◽  
Fault Diagnosis ◽  
Domain Adaptation ◽  
Rolling Bearing ◽  
Training Data ◽  
Variable Load ◽  
K Nearest Neighbor ◽  
Target Domain ◽  
Bearing Faults ◽  
Load Conditions

To identify rolling bearing faults under variable load conditions, a method named DISA-KNN is proposed in this paper, which is based on the strategy of feature extraction-domain adaptation-classification. To be specific, the time-domain and frequency-domain indicators are used for feature extraction. Discriminative and domain invariant subspace alignment (DISA) is used to minimize the data distributions’ discrepancies between the training data (source domain) and testing data (target domain). K-nearest neighbor (KNN) is applied to identify rolling bearing faults. DISA-KNN’s validation is proved by the experimental signal collected under different load conditions. The identification accuracies obtained by the DISA-KNN method are more than 90% on four datasets, including one dataset with 99.5% accuracy. The strength of the proposed method is further highlighted by comparisons with the other 8 methods. These results reveal that the proposed method is promising for the rolling bearing fault diagnosis in real rotating machinery.

scholarly journals Fault Diagnosis of Rolling Bearing Based on a Novel Adaptive High-Order Local Projection Denoising Method

Complexity ◽  
2018 ◽  
Vol 2018 ◽  
pp. 1-15 ◽  
Author(s):  
Rui Yuan ◽  
Yong Lv ◽  
Gangbing Song
Keyword(s):  
Fault Diagnosis ◽  
Numerical Simulations ◽  
Rolling Bearing ◽  
High Order ◽  
Rolling Bearings ◽  
Theoretical Derivation ◽  
Denoising Method ◽  
Vibration Signals ◽  
Local Projection ◽  
Ring Fault

Rolling bearings are vital components in rotary machinery, and their operating condition affects the entire mechanical systems. As one of the most important denoising methods for nonlinear systems, local projection (LP) denoising method can be used to reduce noise effectively. Afterwards, high-order polynomials are utilized to estimate the centroid of the neighborhood to better preserve complete geometry of attractors; thus, high-order local projection (HLP) can improve noise reduction performance. This paper proposed an adaptive high-order local projection (AHLP) denoising method in the field of fault diagnosis of rolling bearings to deal with different kinds of vibration signals of faulty rolling bearings. Optimal orders can be selected corresponding to vibration signals of outer ring fault (ORF) and inner ring fault (IRF) rolling bearings, because they have different nonlinear geometric structures. The vibration signal model of faulty rolling bearing is adopted in numerical simulations, and the characteristic frequencies of simulated signals can be well extracted by the proposed method. Furthermore, two kinds of experimental data have been processed in application researches, and fault frequencies of ORF and IRF rolling bearings can be both clearly extracted by the proposed method. The theoretical derivation, numerical simulations, and application research can indicate that the proposed novel approach is promising in the field of fault diagnosis of rolling bearing.

scholarly journals A Denoising Autoencoder-Based Bearing Fault Diagnosis System for Time-Domain Vibration Signals

2021 ◽  
Vol 2021 ◽  
pp. 1-7
Author(s):  
Yi Gu ◽  
Jiawei Cao ◽  
Xin Song ◽  
Jian Yao
Keyword(s):  
Fault Diagnosis ◽  
Time Domain ◽  
Rolling Bearing ◽  
Rolling Bearings ◽  
Bp Network ◽  
Bearing Fault ◽  
Bearing Fault Diagnosis ◽  
Novel Approach ◽  
Fault Diagnosis System ◽  
Original Time

The condition monitoring of rotating machinery is always a focus of intelligent fault diagnosis. In view of the traditional methods’ excessive dependence on prior knowledge to manually extract features, their limited capacity to learn complex nonlinear relations in fault signals and the mixing of the collected signals with environmental noise in the course of the work of rotating machines, this article proposes a novel approach for detecting the bearing fault, which is based on deep learning. To effectively detect, locate, and identify faults in rolling bearings, a stacked noise reduction autoencoder is utilized for abstracting characteristic from the original vibration of signals, and then, the characteristic is provided as input for backpropagation (BP) network classifier. The results output by this classifier represent different fault categories. Experimental results obtained on rolling bearing datasets show that this method can be used to effectively diagnose bearing faults based on original time-domain signals.

Sign in / Sign up

Export Citation Format

Share Document