Title :
A steam turbine-generator vibration fault diagnosis method based on rough set
Author :
Jian, Ou ; Cai-Xin, Sun ; Weimin, Bi ; Bide, Zhang ; Ruijin, Liao
Author_Institution :
Lab. of High Voltage Eng. & Electr. New Technol. of Minist. of Educ., Chongqing Univ., China
Abstract :
According to turbine-generator vibration characteristic spectrum, a discretized generator fault attribute decision table and condition. attribute set reduction method based on rough set theory are presented in this paper, though the key character which influences classifying is picked up. BP network input dimension is reduced and training time is saved. Experiment shows that the result is effective.
Keywords :
backpropagation; fault diagnosis; neural nets; power engineering computing; rough set theory; steam turbines; turbogenerators; vibration measurement; BP network input dimension reduction; condition attribute set reduction method; discretized generator fault attribute decision table; neural networks; rough set theory; steam turbine-generator; training time reduction; vibration fault diagnosis method; Bismuth; Character generation; Fault diagnosis; Information systems; Machine learning; Neural networks; Rough sets; Set theory; Sun; Turbines;
Conference_Titel :
Power System Technology, 2002. Proceedings. PowerCon 2002. International Conference on
Print_ISBN :
0-7803-7459-2
DOI :
10.1109/ICPST.2002.1067789
