DocumentCode :
3005307
Title :
A novel method for identifying large steam turbine-generator parameters by load rejection test
Author :
Erliang, Kang ; Li, Sun ; Bo, Li ; Changsheng, Yu
Author_Institution :
Dept. of Electr. Eng., Harbin Inst. of Technol., China
Volume :
1
fYear :
2005
fDate :
27-29 Sept. 2005
Firstpage :
119
Abstract :
This paper employs a novel method, which is first decomposing the load rejection model and then deriving optimal combination models from the characteristics of the decaying speeds of the subtransient component and transient component of one phase voltage, to directly identify the transient parameters in different regions employing nonlinear least squares regression theory respectively. Employing the load-rejection model constructed with the design values of turbo-generator QFSN-600-2YH to calculate the quantities by the different methods indicates that the parameters-identified errors of applying the optimal models are far less than the errors of directly estimating all the parameters from the model. By the models one can obtain the direct parameters and quadrature parameters through load rejections under special reference, i.e. direct axis and arbitrary reference. The direct parameters of a 600 MW turbine-generator made by Harbin Large Electric Machine Plant China are determined by the method.
Keywords :
least squares approximations; load (electric); parameter estimation; regression analysis; steam turbines; turbogenerators; 600 MW; Harbin Large Electric Machine Plant China; QFSN-600-2YH; load rejection model; load rejection test; nonlinear least squares regression theory; parameter identification; parameters estimation; steam turbine-generator; transient parameters; Circuit testing; Electronic equipment testing; Equations; IEC standards; Load modeling; Power system transients; Steady-state; Synchronous machines; System testing; Voltage;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electrical Machines and Systems, 2005. ICEMS 2005. Proceedings of the Eighth International Conference on
Print_ISBN :
7-5062-7407-8
Type :
conf
DOI :
10.1109/ICEMS.2005.202497
Filename :
1574730
Link To Document :
بازگشت