Fault Diagnosis Based on Rough Set and BP Neural Network (RS-BP) for Gas Turbine Engine

Abstract:

This paper presents a new fusion diagnosis based on rough set and BP neural network for the fault diagnosis of gas turbine. The frame is designed to fusion fault diagnosis, which is composed by three parts: the rough set data pre-processor, rough set diagnosis model and BP neural network diagnosis model. Aiming at the difficulty in getting adequate fault samples in fault diagnosis, rough set theory is first used to process the original data, establish the decision table and generate rules, which can eliminate the redundant information and build the rough set diagnosis model. After that, according to the optimal decision attribute pre-treated by rough set, BP neural network is designed for fault diagnosis, which can reduce the scale of neural network, improve the identification rate, and improve the efficiency of the whole fusion diagnosis system. Finally, an example of gas turbine generator sets fuel system is taken as a case study to demonstrate the feasibility and validity of the proposed method in this paper.

• Periodical: Advanced Materials Research (Volumes 732-733)
• Pages: 397-401
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.732-733.397
• Online since: August 2013
• Authors: Ning Bo Zhao*, Shu Ying Li, Shuang Yi, Yun Peng Cao, Zhi Tao Wang
• Keywords: BP Neural Network (BPNN), Fusion Fault Diagnosis, Gas Turbine Engine, Rough Set (RS)

References

[1] Y.G. Li, Performance-analysis-based gas turbine diagnostics: a review, P I Mech Eng A-J Pow 216 (2002) 363-377.

[2] Y.K. Lee, D.N. Mavris, V.V. Volovoi, M. Yuan, T. Fisher, A fault diagnosis method for industrial gas turbines using Bayesian data analysis, J Eng Gas Turb Power 132 (2010) 82-89.

DOI: 10.1115/1.3204508

[3] S. Simani, C. Fantuzzi, Dynamic system identification and model-based fault diagnosis of an industrial gas turbine prototype, Mechatronics 16 (2006) 341-363.

DOI: 10.1016/j.mechatronics.2006.01.002

[4] K. Mathioudakis, Ph. Kamboukos, A. Stamasis, Turbofan performance deterioration tracking using non-linear models and optimization techniques, J Turbomach 124 (2002) 580-587.

DOI: 10.1115/1.1512678

[5] Z. Pawlak, Rough sets, International Journal of Computer and Information Sciences 11 (1982) 341-356.

[6] L.X. Shen, Francis E.H. Tay, L. S Qu, Y. D, Shen, Fault diagnosis using rough sets theory, Comput Ind 43 (2000) 61-72.

DOI: 10.1016/s0166-3615(00)00050-6

[7] S. Tsumoto, Automated extraction of medical expert system rules from clinical databases based on rough set theory, Inform Sciences 112 (1998) 67-84.

DOI: 10.1016/s0020-0255(98)10021-x

[8] R. Nowicki, R. Słowiński, J. Stefanowski, Evaluation of vibroacoustic diagnostic symptoms by means of the rough sets theory, Comput Ind 20 (1992) 141-152.

DOI: 10.1016/0166-3615(92)90048-r

[9] B. Zheng, X.Y. Zhu, Investigation of fault diagnosis technology for aeroengine, Aeroengine 36 (2010) 22-25.

[10] T. Kohonen, Self-Organizing Maps, Springer, New York, USA, (1995).