ICA and BP Neutral Network Based Fault Detecting Algorithm for Large Engine

Abstract:

With the rapid development of information technology, some experience intensive work like fault detecting has been greatly improved by information technology. Traditional fault detecting method depends on the knowledge of expert to some extent which can not satisfy the requirement of large engine, which is with poor testability. Regarding this problem, this paper apply the technology of Independent Components Analysis (ICA) and BP neutral network to the process of fault detecting, the accuracy and efficiency are greatly improved with these methods, and the experiment result has proven the validity of the method described.

• Periodical: Applied Mechanics and Materials (Volumes 110-116)
• Pages: 4549-4553
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.110-116.4549
• Online since: October 2011
• Authors: Yong Mei Jiang
• Keywords: BP Neural Network (BPNN), Fault Detecting, Independent Component Analysis (ICA), Large Engine

References

[1] Gelso, E.R., E. Frisk, and J. Armengol Llobet. Robust fault detection using consistency techniques with application to an automotive engine[C]. 2008. Seoul, Korea, Republic of: Elsevier.

DOI: 10.3182/20080706-5-kr-1001.00929

[2] Vinsonneau, J.A.F., et al. Improved SI engine modelling techniques with application to fault detection[C]. 2002. Glasgow, United kingdom: Institute of Electrical and Electronics Engineers Inc.

[3] Shang, L. and G. Liu. Sensor and actuator fault detection and isolation for a high performance aircraft engine bleed air temperature control system[C]. 2009. Shanghai, China: Institute of Electrical and Electronics Engineers Inc.

DOI: 10.1109/cdc.2009.5400042

[4] Xue, W., Y. -Q. Guo, and R. Li, Algorithm and experimental validation for condition monitoring, fault detection for gas turbine engine[J]. Tuijin Jishu/Journal of Propulsion Technology. 32(2): pp.271-275.

[5] Wang, X., et al. Semi-physical neural network model in detecting engine transient faults using the local approach[C]. 2008. Seoul, Korea, Republic of: Elsevier.

[6] Lu, F., et al., Research on sensor fault diagnosis of aero-engine based on data fusion of SPSO-SVR[J]. Hangkong Dongli Xuebao/Journal of Aerospace Power, 2009. 24(8): pp.1856-1865.

[7] Alkan, Y., B.B. Biswal, and T.L. Alvarez. Visual cortical circuits revealed using fMRI and ICA[C]. New York, NY, United states: IEEE Computer Society.

DOI: 10.1109/nebc.2010.5458272

[8] Lv, J.C., et al., Stability and chaos of a class of learning algorithms for ICA neural networks[J]. Neural Processing Letters, 2008. 28(1): pp.35-47.

DOI: 10.1007/s11063-008-9080-2

[9] Wang, Z. and S. Li. Research and implement for vehicle license plate recognition based on improved BP network[C]. Chengdu, China: IEEE Computer Society.

DOI: 10.1109/cctae.2010.5543425

[10] Zhou, H., H. Chen, and J. Li. Design of multi-parameter fusion coulometer based on self-feedback BP network[C]. Wuhan, China: IEEE Computer Society.

DOI: 10.1109/icece.2010.196