Application of FastICA Algorithm in Wind Turbine Condition Monitoring

Abstract:

In wind turbine condition monitoring, the sensors often can not be installed to the ideal position. Compare the common signal processing method comprehensively and give the advantage of the fastICA algorithm in the wind turbine condition monitoring. Give the basic principle and mathematical model of the fastICA algorithm, while monitor and analysis the wind turbine state data based on the fastICA algorithm. The results show that this algorithm can separate the vibration characteristics of the tested compenent of the wind turbine from the vibration signals quickly and accurately.

• Periodical: Applied Mechanics and Materials (Volumes 217-219)
• Pages: 2750-2753
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.217-219.2750
• Online since: November 2012
• Authors: Guang Kun Shan, Hai Long Zhang, Xiao Dong Wang, Ying Ming Liu
• Keywords: Condition Monitoring, Fast ICA, Signal Processing, Wind Turbine (WT)

References

[1] L. S. Qu: Mechanical Fault Diagnosis Theories and Methods(Xi'an Jiaotong University Press, China 2009), p.220.(In Chinese)

[2] Y. Bi: Research and Application of Blind Source Separation Algorithm Based on Independent Component Analysis (MS., Xi'an University of Technology, China 2007), p.11.(In Chinese)

[3] F. S. Yang and B. Hong: Principle and Application of Independent Component Analysis(Tsinghua University Press, China 2006), p.97.(In Chinese)

[4] A. Hyvarinen: Neural Network, Vol. 10 (1999) No.3, p.626.

[5] W. D. Jiao: Rotating Machinery Fault Diagnosis Method Based on Independent Component Analysis (Ph.D., Zhejiang University, China 2003), p.111.(In Chinese)