Fault Diagnosis of Bearing Based on Selective Ensemble of Multiple Fuzzy ARTMAP Neural Networks

Abstract:

A novel selective ensemble of multiple fuzzy ARTMAP (FAM) classifiers based on the correlation measure method and Bayesian belief method is proposed to apply to the fault diagnosis of rolling element bearings in this paper. The test results show that the selective ensemble of four optimal FAM classifiers can identify the different fault categories accurately and has a better classification performance compared to the single FAM and ensemble of all FAM classifiers.

• Periodical: Applied Mechanics and Materials (Volumes 423-426)
• Pages: 2480-2485
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.423-426.2480
• Online since: September 2013
• Authors: Zeng Bing Xu, Hong Wu, Tie Lin Shi
• Keywords: Fault Diagnosis, Fuzzy ARTMAP, Selective Ensemble of Multiple Classifiers

References

[1] B. Samanta K.R. Al-Balushi S.A. Al-Araimi: Engineering Applications of Artificial Intelligence. 16(2003), p.657.

DOI: 10.1016/j.engappai.2003.09.006

[2] R. Javadpour, G.M. Knapp: Computers & Industrial Engineering 45(2003). p.323.

[3] G.A. Garperter, S. Grossberg, N. Markuzon: IEEE Transactions on Neural Networks 3(1992), p.698.

[4] M. Georgiopoulos, H. Fernlund, G. Bebis, et al. Heileman: Neural Networks, 9 (9) (1996), p.1541.

[5] G. Fumera,F. Roli: IEEE Transactions on Pattern Analysis and Machine Intelligence 27(2005) p.942.

[6] Bearings vibration data set, http: /www. eecs. cwru. edu/laboratory/ bearing/download. htm. (2003).

[7] Y.P. Zhang, S.H. Huang, et al. Mechanical Systems and Signal Processing 20 (5), (2006)p.1202.

[8] L. Chen and H.L. Tang,: Electronics letters. 40(4), (2004). p.238.