Research on Fault Diagnosis Method of Steam Turbine Unit Based on the Theory of Super Ball

Abstract:

A fault diagnosis method of steam turbine based on the theory of super ball is proposed, which combines density clustering with hierarchical clustering. The correlation of vibration and thermal parameters is introduced as the clustering factors. The efficiency of diagnosis,the sensitivity of noise and the accuracy of diagnosis are improved. Experiments show that the method and the selection of clustering factor are feasible.

• Periodical: Applied Mechanics and Materials (Volumes 496-500)
• Pages: 953-957
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.496-500.953
• Online since: January 2014
• Authors: Xiao Wen Deng, Yan Peng Han*, Cheng Cheng Wang, Yu Jiong Gu
• Keywords: Cluster, Correlation of Vibration, Theory of Super Ball, Thermal Parameters

References

[1] Wang-Guobiao, He-zhengjia and Chen-Xuefeng, in Where to Go of Basic Research of Mechanical Fault Diagnosis, edited by Chinese Journal of Mechanical Engineering (2013), pp.63-72.

[2] Shi-Weixin and Shi-Jingbo, in Steam turbine generator set vibration and accidents, Beijing, edited by China Electric Power Press (2008).

[3] Li-luping and Lu-Xuxiang, Steam turbine generator set vibration and processing, Beijing, edited by China Electric Power Press (2007).

[4] Lu-Bingjie and Su-he, The experimental research of correlation between blowing converter process parameters and vibration of converter, edited by Steelmaking (1994), pp.21-26.

[5] Gan-wenyan and Li-Deyi, Hierarchical clustering algorithm based on kernel density estimation, edited by Journal of System Simulation (2004), pp.302-309.

[6] Zhou-Haiyan. Research of Spatial data mining, edited by The PLA Information Engineering University (2003).

[7] Lv-Zonglei and Wang-jiandong, A fast clustering algorithm based on multidimensional space super ball, edited by Journal of Nanjing University of Aeronautics and Astronautics (2006).