The simulation research and neural network modeling of superheated steam temperature characteristics for ultra-supercritical unit

Author

Yongguang Ma ; Shiru Yang ; Shuo Cai ; Yufang Wang

Author_Institution

Dept. of Autom., North China Electr. Power Univ., Baoding, China

fYear

2015

fDate

23-25 May 2015

Firstpage

2474

Lastpage

2478

Abstract

In this paper, with a 1000MW ultra-supercritical (USC) unit as the object investigated, its superheated steam temperature characteristic was studied. Based on the operating data which were extracted from a full-scope simulator of the given USC power unit, the BP RBF and Elman neural networks were respectively used to establish the model of superheated steam temperature characteristic. There were ten input variables, including main steam pressure, fuel flow, total air flow, feedwater flow, etc. The output was superheated steam temperature. The simulation test results verified the validity of the three models.

Keywords

backpropagation; boilers; heat transfer; neurocontrollers; pressure control; radial basis function networks; temperature control; BP RBF neural networks; Elman neural networks; USC boiler; USC power unit; feedwater flow; fuel flow; main steam pressure; neural network modeling; superheated steam temperature characteristics; total air flow; ultra-supercritical unit; Boilers; Data models; Mathematical model; Neural networks; Predictive models; Temperature; Training; Modeling; Neural Network; Superheated Steam Temperature; Ultra-supercritical Unit;

fLanguage

English

Publisher

ieee

Conference_Titel

Control and Decision Conference (CCDC), 2015 27th Chinese

Conference_Location

Qingdao

Print_ISBN

978-1-4799-7016-2

Type

conf

DOI

10.1109/CCDC.2015.7162337

Filename

7162337