DocumentCode :
2088299
Title :
Integrated Modeling and Simulating of the Three-axis Turbine Power Generation based on the Neural Network Identification
Author :
Zhang, Xiaoyun ; Li, Shuying ; Wang, Jianqing ; Fan, Huanran
Author_Institution :
Coll. of Power & Energy Eng., Harbin Eng. Univ. (HEU), Harbin, China
fYear :
2010
fDate :
28-31 March 2010
Firstpage :
1
Lastpage :
4
Abstract :
Gas turbine units are used in industrial applications more and more widely, and many people have researched widely in the fault diagnosis technology and the control technology for the gas turbine units. Simulation of Gas Turbine technology is the basis and the mathematical model of real-time is the basis of all relevant digital simulation. Using the neural network recognition technology of the radial basis function and the curve-fitting technology of part-characteristics, the mathematical model of three-axis gas turbine is established. At the same time the simulation model is established based on Matlab/simulink software. The dynamic simulation of the gas turbine for the burden loading and reducing has been researched and the response of for the output speed and fuel capacity is obtained. It is shown that the model of the gas turbine has the fast learning speed, high sampling rate and high accuracy.
Keywords :
curve fitting; fault diagnosis; gas turbine power stations; power system simulation; radial basis function networks; curve-fitting technology; digital simulation; dynamic simulation; fault diagnosis technology; gas turbine units; neural network identification; radial basis function; three-axis turbine power generation; Curve fitting; Digital simulation; Fault diagnosis; Fuels; Gas industry; Industrial control; Mathematical model; Neural networks; Power generation; Turbines;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Power and Energy Engineering Conference (APPEEC), 2010 Asia-Pacific
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-4812-8
Electronic_ISBN :
978-1-4244-4813-5
Type :
conf
DOI :
10.1109/APPEEC.2010.5448235
Filename :
5448235
Link To Document :
بازگشت