Feature selection with SVD entropy: Some modification and extension

Banerjee, Mohua; Pal, Nikhil R.

doi:10.1016/j.ins.2013.12.029

Cited by 77 publications

(30 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the first line and the last column elements from the matrix H' are selected as the reconstructed signal [22]:…”

Section: Singular Value Decomposition (Svd) Algorithmmentioning

confidence: 99%

“…Zhao et al [21] provided an algorithm to search for the effective singular values based on the maximum peak of the curvature spectrum, which improves the accuracy of the location regarding bearing damage. The same method was used by Jha et al in [16] to distill the position of demarcation; Banerjee et al in [22] proposed a supervised feature selection algorithm based on SVD-entropy. However, SVD-entropy based methods have a limitation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Development of a Method for Selection of Effective Singular Values in Bearing Fault Signal De-Noising

Gao

Wang

et al. 2016

Applied Sciences

View full text Add to dashboard Cite

Singular value decomposition (SVD) is a widely used and powerful tool for signal extraction under noise. Noise attenuation relies on the selection of the effective singular value because these values are significant features of the useful signal. Traditional methods of selecting effective singular values (or selecting the useful components to rebuild the faulty signal) consist of seeking the maximum peak of the differential spectrum of singular values. However, owing to the small number of selected effective singular values, these methods lead to excessive de-noised effects. In order to get a more appropriate number of effective singular values, which preserves the components of the original signal as much as possible, this paper used a difference curvature spectrum of incremental singular entropy to determine the number of effective singular values. Then the position was found where the difference of two peaks in the spectrum declines in an infinitely large degree for the first time, and this position was regarded as the boundary of singular values between noise and a useful signal. The experimental results showed that the modified methods could accurately extract the non-stationary bearing faulty signal under real background noise.

show abstract

“…Finally, the first line and the last column elements from the matrix H' are selected as the reconstructed signal [22]:…”

Section: Singular Value Decomposition (Svd) Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Development of a Method for Selection of Effective Singular Values in Bearing Fault Signal De-Noising

Gao

Wang

et al. 2016

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Through correlation analysis, SVDS is able to reveal the weak intrinsic pattern buried in a signal and effectively suppress the noise with different distributions. 40,41 At the same time, the noise elimination algorithm based on SVDS is relatively fast and easy to implement. SVDS-based feature extraction method or singular spectrum analysis (SSA) enables a suitable opportunity to monitor the singular values obtained from vibration signals at different operation conditions.…”

Section: Introductionmentioning

confidence: 99%

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

Wang

2018

Journal of Low Frequency Noise, Vibration and Active Control

View full text Add to dashboard Cite

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.

show abstract

“…In this technique, the collected information is divided into various clusters to show the system behavior patterns effectively. In other words, patterns in the same group are similar in some sense and patterns in different groups are dissimilar in the same sense [4,5]. In terms of analysis of variance (ANOVA), the within-variance is low and between-variance is high.…”

Section: Introductionmentioning

confidence: 99%

“…Besides relevant features, there might be derogatory features, indifferent features, and redundant (dependent) ones. Removal of these features not only makes the learning task easier, by reducing computational constraint but also often improves the performance of the classifier [4,5]. Such data reduction is applied to images to achieve image compression.…”

Section: Introductionmentioning

confidence: 99%