A Novel Method for Fault Diagnosis Based on PCA Enhanced by Wavelet Denoising

Feng Qiao; Hao Ming  Zhao; Feng  Zhang; Qing  Ma

doi:10.4028/www.scientific.net/AMR.756-759.3450

Paper Titles

A Statistical Local Binary Fitting Model for Blood Vessel Segmentation
p.3430

A Study of Negative Association Rules Mining Algorithm Based on Multi-Database
p.3435

A Revised Distributed Interpretation for Ontology Integration
p.3440

The Design for Automatic Target Reporting System Based on Image Processing
p.3446

A Novel Method for Fault Diagnosis Based on PCA Enhanced by Wavelet Denoising
p.3450

Improvements on Motion Estimation Algorithms in Video Coding for H.264/AVC
p.3455

The Calibration of Key Parameters for Sugarcane Harvester Basecutter
p.3461

Trajectory Optimization Applied to Space Rendezvous
p.3466

Research on K-Anonymity Algorithm in Privacy Protection
p.3471

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 756-759A Novel Method for Fault Diagnosis Based on PCA...

A Novel Method for Fault Diagnosis Based on PCA Enhanced by Wavelet Denoising

Abstract:

There are some disadvantages for fault detection and diagnosis with traditional Principal Component Analysis (PCA) method because of its shortcomings. It is, in this paper, presented a novel fault diagnosis method based on conventional PCA enhanced by wavelet denoising. The proposed method employs wavelet denoising to deal with the signals, which can reserve enough information of original data, and then establishes PCA model. Based on SPE and T² statistics, abnormal situation can be detected. And the location of the fault can be recognized via contribution plots. At last, the simulation studies with Matlab are carried out to verify the correctness and effectiveness of the proposed method, the advantages of the proposed method over the conventional PCA also are shown in the simulation.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 756-759)

Pages:

3450-3454

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.756-759.3450

Citation:

Cite this paper

Online since:

September 2013

Authors:

Feng Qiao, Hao Ming Zhao, Feng Zhang, Qing Ma

Keywords:

Fault Diagnosis, Principal Component Analysis (PCA), Wavelet Denoising

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] J. J. Li, Z. C. Fang and L. L. Sun: J. of SJZU (Natural Sci. ), Vol. 25, No. 4 (2009), p.792.

Google Scholar

[2] K. Pearson: Philosophical Magazine Vol. 2, No. 11 (1901), p.559.

Google Scholar

[3] Raich and A. Cinar: Computers &Chemical Engineering, Vol. 21, No. 6 (1997), p.661.

Google Scholar

[4] M. A. Kramer: AIChE JOURNAL, Vol. 37, No. 2 (1991), p.233.

Google Scholar

[5] D. Dong and T. J. McAvoy: Computers &Chemical Engineering, Vol. 2 (1996), p.65.

Google Scholar

[6] Y. W. Xiao and B. G. Xu: Computers and Applied Chemistry, Vol. 24, No. 2 (2007), p.163.

Google Scholar

[7] W. Ren, W. D. Tian and T. Z. Du: Cntrl and Inst. in Ch. Ind., Vol. 37, No. 6 (2010), p.20.

Google Scholar

[8] J. Qian: Engineering Application, Vol. 4, No. 2 (2007), p.24.

Google Scholar

[9] M. A. Shams, H. M. Budman and T. A. Duever: Ch. Eng. Sci., Vol. 66, No. 20 (2011), p.488.

Google Scholar

[10] J. Westerhuis, S. Gurden and A. Smilde: Chemom. Intel. Lab. Syst, Vol. 51 (2000), p.95.

Google Scholar

[11] J. E. Jackson and G. S. Mudholkar: Technometrics, Vol. 21, No. 3 (1979), p.341.

Google Scholar