A Bearing Intelligent Fault Diagnosis Method Based on Cluster Analysis

Abstract:

In recent years, machine learning techniques have been widely used in intelligent fault diagnosis field. As a major unsupervised learning technology, cluster analysis plays an important role in fault intelligent diagnosis based on machine learning. In rolling bearing fault diagnosis, the traditional spectrum analysis method usually adopts the resonant demodulation technology, but when the inner circle, rolling body or multi-point faults produce composite modulation, it is difficulty to identify the fault type from demodulation spectral lines. According to this, a novel rolling bearing fault diagnosis method based on KFCM (Kernel-based Fuzzy C-Means) cluster analysis is proposed. Through clustering on test data and the known samples, the memberships of test data are obtained. From these, the rolling bearing fault type can be determined. Experimental results show that this method is effective.

• Periodical: Applied Mechanics and Materials (Volumes 152-154)
• Pages: 1628-1633
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.152-154.1628
• Online since: January 2012
• Authors: Su Qun Cao, Xiao Ming Zuo, Ai Xiang Tao, Jun Min Wang, Xiang Zhi Chen
• Keywords: Cluster Analysis, Fault Diagnosis, Rolling Bearing, Unsupervised Pattern

References

[1] Jingyi Du, Yuanlin Hou: Fault Diagnosis Theory based on Kernel Method and its Method Study. Beijing University Press, Beijing (2010).

[2] Jianjun Zhang, Li Zhang, Haifang Mu, Zhengyu Liu and Juan Xu: Improved BP-neural network of the particle swarm optimization in the research on engine fault diagnosis. Journal of agricultural machinery 42(2011)198-203.

[3] Xuewen He, Haiming Zhao: Support vector machine and its application in the mechanical fault diagnosis. Journal of Zhongnan University 36(2005)97-101.

[4] Gelan Yang, Xue Xu, Gang Yang and Jianming Zhang: Semi-supervised Classification by Local Coordination. Lecture Notes in Computer Science 6444(2010)517-524.

DOI: 10.1007/978-3-642-17534-3_64

[5] Wentao Sui, Changhou Lu and Dan Zhang: Bearing Fault Diagnosis Based on Feature Weighted FCM Cluster Analysis. International Conference on Computer Science and Software Engineering (2008)518-521.

DOI: 10.1109/csse.2008.289

[6] G. Yu, C.N. Li and Sagar. Kamarthi: A Cluster-based Wavelet Feature Extraction Method for Machine Fault Diagnosis. Applied Mechanics and Materials 10-12(2008)548-552.

DOI: 10.4028/www.scientific.net/amm.10-12.548

[7] Qing Yang, Jingran Guo, Dongxu Zhang and Chang Liu: Fault Diagnosis Based on Fuzzy C-means Algorithm of the Optimal Number of Clusters and Probabilistic Neural Network. International Journal of Intelligent Engineering & Systems 4(2011)51-59.

DOI: 10.22266/ijies2011.0630.06

[8] Qing Lu, Jinfu Feng, Guangshu Nie and Qian Li: Fault diagnosis based on semi-fuzzy clustering algorithm. Journal of Test and Measurement Technology 21(2007)551-556.

[9] J.C. Bezdek: Pattern Recognition with Fuzzy Objective Function Algorithms. Plenum Press, New York (1981).

[10] Dao-Qiang Zhang, Song-Can Chen: A Novel Kernelized Fuzzy C-means Algorithm with Application in Medical Image Segmentation. Artificial Intelligence in Medicine 32(2004)37-50.

DOI: 10.1016/j.artmed.2004.01.012

[11] Guoxin Zhang, Zuoshi Liu: The method of fuzzy clustering for bearing fault diagnosis. Journal of Jiangxi University of Science and Technology 27(2006)15-17.