Title :
Method research of noise diagnosis based on fuzzy neural network
Author :
Wu, Chang-Zhi ; Hua, Wu ; Ma, Yan Jian-Fen
Author_Institution :
Comput. Center, Taiyuan Univ. of Technol., China
Abstract :
In this paper, a new noise diagnosis method is presented, and its algorithm deduced, based on a fuzzy neural network. As a practical example, the noise field intensity is diagnosed and the coupling noise is identified for the noise from a steam turbine unit. The results show that the fuzzy neural network has a good capacity for pattern recognition, fast convergence and good stability in dealing with uncertain or ambiguous data at the classification boundary. The noise radiated from the steam turbines unit is diagnosed accurately and the noise node is suppressed effectively and is restructured
Keywords :
convergence; fault diagnosis; fuzzy neural nets; noise; pattern recognition; stability; steam turbines; classification boundary; coupling noise; fast convergence; fuzzy neural network; noise diagnosis; noise field intensity; nonlinear pattern classification; pattern recognition; stability; steam turbine unit; Active noise reduction; Computer networks; Fault diagnosis; Frequency; Fuzzy neural networks; Monitoring; Signal to noise ratio; Stability; Turbines; Working environment noise;
Conference_Titel :
Signal Processing Proceedings, 1998. ICSP '98. 1998 Fourth International Conference on
Conference_Location :
Beijing
Print_ISBN :
0-7803-4325-5
DOI :
10.1109/ICOSP.1998.770875
