Incipient Fault Diagnosis of Roller Bearing Using Optimized Wavelet Transform Based Multi-Speed Vibration Signatures

Huo, Zhiqiang; Zhang, Yu; Francq, Pierre; Shu, Lei; Huang, Jianfeng

doi:10.1109/access.2017.2661967

Cited by 167 publications

(95 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The weighted sum passed through a nonlinear function such as a rectified linear unit (ReLU). This is shown in (1). ReLU is a half-wave rectifier, ( ) = max( , 0), and is like the Softplus activation function; that is, Softplus( ) = ln(1 + ).…”

Section: Deep Learning and Cnn Backgroundmentioning

confidence: 99%

“…This has had wide ranging economic benefits for the owners of the assets and has opened new possibilities of revenue by allowing original equipment manufacturers (OEMs) to contract in maintainability and availability value. However, the state of current diagnostics involves a laborious process of creating a feature vector from the raw signal via feature extraction [1][2][3]. For example, Seera proposes a FuzzyMin-Max Classification and Regression Tree (FMM-CART) model for diagnostics on Case Western's bearing data [4].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning Enabled Fault Diagnosis Using Time-Frequency Image Analysis of Rolling Element Bearings

Verstraete

Ferrada

Droguett

et al. 2017

Shock and Vibration

294

198

View full text Add to dashboard Cite

Traditional feature extraction and selection is a labor-intensive process requiring expert knowledge of the relevant features pertinent to the system. This knowledge is sometimes a luxury and could introduce added uncertainty and bias to the results. To address this problem a deep learning enabled featureless methodology is proposed to automatically learn the features of the data. Timefrequency representations of the raw data are used to generate image representations of the raw signal, which are then fed into a deep convolutional neural network (CNN) architecture for classification and fault diagnosis. This methodology was applied to two public data sets of rolling element bearing vibration signals. Three time-frequency analysis methods (short-time Fourier transform, wavelet transform, and Hilbert-Huang transform) were explored for their representation effectiveness. The proposed CNN architecture achieves better results with less learnable parameters than similar architectures used for fault detection, including cases with experimental noise.

show abstract

Section: Deep Learning and Cnn Backgroundmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Deep Learning Enabled Fault Diagnosis Using Time-Frequency Image Analysis of Rolling Element Bearings

Verstraete

Ferrada

Droguett

et al. 2017

Shock and Vibration

294

198

View full text Add to dashboard Cite

show abstract

“…Herein, WPD is first applied to generate F2C signals by reconstructing selected wavelet coefficients. WPD is essentially a multi-resolution analysis that can provide the good time and frequency resolution as one of powerful wavelet analysis methods [13]. Similar to discrete wavelet decomposition, WPD is also a filtering operation that decomposes a signal into approximation parts (low-pass filter) and detail parts (high-pass filter) by iteratively using wavelet filtering operation until desired frequency resolution is achieved [14].…”

Section: Fine-to-coarse Multiscale Permutation Entropy(f2cmpe)mentioning

confidence: 99%

Fine-to-Coarse Multiscale Permutation Entropy for Rolling Bearing Fault Diagnosis

Huo

Shu

2018

2018 14th International Wireless Communications &Amp; Mobile Computing Conference (IWCMC)

Self Cite

View full text Add to dashboard Cite

Abstract-Multiscale Permutation Entropy (MPE) has been applied as a non-linear measure for estimating the complexity of time series. Nevertheless, the coarse-grained procedure in MPE only takes low-frequency information into account. To overcome this shortcoming, in this paper, a new entropy measure, named Fine-to-Coarse Multiscale Permutation Entropy (F2CMPE), is proposed to provide stable and reliable results by offering both low-frequency and high-frequency information. Firstly, the F2C signals are created based on the reconstruction of selected wavelet coefficients using wavelet packet decomposition. Then, permutation entropy is used to estimate the complexity and dynamic change of the F2C signals. Experimental analysis is carried out to investigate and compare the performance of the proposed F2CMPE with that of the MPE. Results indicate that the proposed method can give consistent and stable entropy measure for rolling bearing fault diagnosis.

show abstract

“…Fast Fourier transform and short time Fourier transform are two classical approaches for signal decomposition; however their performances are limited by the finite window size which is not suitable to analyze non-stationary and non-linear vibration signals. Instead, wavelet transform and empirical mode decomposition (EMD) have shown their effectiveness in providing high resolution in both time and frequency domain, which have been successfully applied in the field of fault diagnosis of shafts, such as continuous wavelet transform coefficients were used in works [1]- [5]. Apart from that, wavelet packets Corresponding author: Lei Shu (email: lshu@lincoln.ac.uk) decomposition (WPD) benefits from effectively decomposing frequency bands into detail and approximate coefficients with multi-levels [6], which has been applied in [7], [8].…”

Section: Introductionmentioning

confidence: 99%

A comparative study of WPD and EMD for shaft fault diagnosis

Huo

Shu

2017

IECON 2017 - 43rd Annual Conference of the IEEE Industrial Electronics Society

Self Cite

View full text Add to dashboard Cite

Abstract-Fault diagnosis of incipient crack failure in rotating shafts allows the detection and identification of performance degradation as early as possible in industrial plants, such as downtime and potential injury to personnel. The present work studies the performance and effectiveness of crack fault detection by means of applying wavelet packet decomposition (WPD) and empirical mode decomposition (EMD) on fault diagnosis of rotating shafts using multiscale entropy (MSE). After WPD and EMD, the most sensitive reconstruction vectors and intrinsic mode functions (IMFs) are selected using Shannon entropy. Then, these feature vectors are fed into support vector machine (SVM) for fault classification, where the entropy features represent the complexity of vibration signals with different scales. Experimental results have demonstrated that WPD combined with MSE can achieve an accuracy of 97.3% for crack fault detection in rotating shafts, whilst EMD combined with MSE has shown a higher detection rate of 98.5%.

show abstract

Incipient Fault Diagnosis of Roller Bearing Using Optimized Wavelet Transform Based Multi-Speed Vibration Signatures

Cited by 167 publications

References 24 publications

Deep Learning Enabled Fault Diagnosis Using Time-Frequency Image Analysis of Rolling Element Bearings

Deep Learning Enabled Fault Diagnosis Using Time-Frequency Image Analysis of Rolling Element Bearings

Fine-to-Coarse Multiscale Permutation Entropy for Rolling Bearing Fault Diagnosis

A comparative study of WPD and EMD for shaft fault diagnosis

Contact Info

Product

Resources

About