Bridge Deflection Sensor Fault Diagnosis Base on ICA

Abstract:

Fault diagnosis accuracy is not high for traditional multivariate statistical methods ,in this paper , the deflection sensor fault diagnosis method based on independent component analysis was proposed .This method not only removed the relationship between process variables , but also make full use of higher order statistical properties of process variables ,establish the independent element model of the process information, finally ,simulation the multivariate process model in the system ，the simulation result shows that this method can quickly and accurately detect the abnomal occurred in the system ,verify the validity of this method and advantage compared with taditional PCA method.

• Periodical: Applied Mechanics and Materials (Volume 344)
• Pages: 103-106
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.344.103
• Online since: July 2013
• Authors: Shun Ren Hu, Rui Ping Li, Li She
• Keywords: Bridge Monitoring, Fault Diagnosis, Independent Component Analysis (ICA)

References

[1] HongLi , Yunlian Sun. The Study and Test of ICA Algorithms. School of Electrical Engineering Wuhan University 7803-9335 (2005 ) 2-3.

[2] Hyv~irinen, Oja E．Independent Component Analysis: Algorithms and Applications[J]．Neural Networks, 2002 108-119.

[3] ShuLiang Hu , Han . Based on ICA process monitoring method research Laboratory science 13 (2) April 2010 2-4.

[4] MouXue jiao, WangJingCheng, XuWenYan, Based on independent component analysis method of condition monitoring control engineering in May 2008, the 15th roll supplement 23-30.

[5] HuShunRen, ChenWeiMin. The bridge structural health monitoring system of deflection adaptive correction technology research . Highway, 10 (2011) 4-9.