Certainty improvement in diagnosis of multiple faults by using versatile membership functions for fuzzy neural networks

Authors:
Yuan Kang;Chun-Chieh Wang;Yeon-Pun Chang;Chien-Ching Hsueh;Ming-Chang Chang
Affiliations:
Department of Mechanical Engineering, Chung Yuan Christian University, Chung Li, Taiwan;Department of Mechanical Engineering, Chung Yuan Christian University, Chung Li, Taiwan;Department of Mechanical Engineering, Chung Yuan Christian University, Chung Li, Taiwan;Power System Dept., Industrial Technology Research Institute, Hsinchu, Taiwan;Magnate Aeronautical Industry Co.,Ltd, Kaohsiung, Taiwan
Venue:
ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part III
Year:
2006

Citing 1
Cited 3

Neural networks: algorithms, applications, and programming techniques

Neural networks: algorithms, applications, and programming techniques

Applications of fuzzy neural networks for grading

AIAP'07 Proceedings of the 25th conference on Proceedings of the 25th IASTED International Multi-Conference: artificial intelligence and applications
Gear Fault Diagnosis by Using Wavelet Neural Networks

ISNN '07 Proceedings of the 4th international symposium on Neural Networks: Advances in Neural Networks, Part III
Applications of fault diagnosis in rotating machinery by using time series analysis with neural network

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

Because the relationship between frequency symptoms and fault causes are different, this study uses fuzzy neural network (FNN) with versatile membership functions to diagnose multiple faults in rotary machinery. According to the frequency symptom values for each fault causes, three kinds of membership functions are used. Besides, the structure of the FNN is large which spend much training time. Thus, when the matrix between frequency symptoms and fault causes can decoupled, the relational matrix decomposed into several sub-matrixes and the structure of the FNN can also divided into several sub-networks. In this study, two above-mention approaches are combined to diagnose multiple faults and compared with neural network (NN), FNN with single/versatile membership functions in two actual cases.