A New Hybrid Intelligent Fault Diagnosis Model for Steamer

Considering the ability of rough sets theory on reduction of decision system and that of neural networks for clustering and nonlinear mapping, a new hybrid intelligent model of rough sets and neural networks for fault diagnosis is proposed. Meanwhile, a novel attribute reduction approach of rough set based on immune clonal selection is proposed, in order to find the minimal feature set of decision table. Then, RBF neural network was designed to diagnose the faults occurred in steamer axes vibration, in which the results of attribute reduction are regarded as the input nodes and the decision attributes are regarded as the output nodes correspondingly. The experimental results showed that the model can reduce the cost of diagnosis and increase the efficiency of diagnosis. There will be well application prospect in practice.

[1]  Zdzislaw Pawlak,et al.  Rough sets and intelligent data analysis , 2002, Inf. Sci..

[2]  Leandro Nunes de Castro,et al.  The Clonal Selection Algorithm with Engineering Applications 1 , 2000 .

[3]  R. Gershon,et al.  "Clonal selection and after," and after. , 1979, The New England journal of medicine.

[4]  Rolf Isermann,et al.  Supervision, fault-detection and fault-diagnosis methods — An introduction , 1997 .

[5]  Licheng Jiao,et al.  Clonal operator and antibody clone algorithms , 2002, Proceedings. International Conference on Machine Learning and Cybernetics.

[6]  Shang-Liang Chen,et al.  Orthogonal least squares learning algorithm for radial basis function networks , 1991, IEEE Trans. Neural Networks.

[7]  Byeong Seok Ahn,et al.  The integrated methodology of rough set theory and artificial neural network for business failure prediction , 2000 .

[8]  Li Pheng Khoo,et al.  Feature extraction using rough set theory and genetic algorithms--an application for the simplification of product quality evaluation , 2002 .

[9]  Peter J. Bentley,et al.  Towards an artificial immune system for network intrusion detection: an investigation of clonal selection with a negative selection operator , 2001, Proceedings of the 2001 Congress on Evolutionary Computation (IEEE Cat. No.01TH8546).

[10]  Wang Hong.  An Improved Learning Algorithm for RBF Neural Networks , 2002 .

[11]  Jerzy W. Grzymala-Busse,et al.  Rough Sets , 1995, Commun. ACM.

[12]  Rolf Isermann,et al.  Preface to the special section of papers on supervision, fault detection and diagnosis of technical systems , 1997 .

[13]  B. Wen.  COMBINING ROUGH SET METHODOLOGY AND FUZZY CLUSTERING FOR KNOWLEDGE DISCOVERY FROM QUANTITATIVE DATA , 2004 .