Using genetic algorithms to select inputs for neural networks

Author

Guo, Zhichao ; Uhrig, Robert E.

Author_Institution

Corp. Adv. Syst., Carrier Corp., Syracuse, NY, USA

fYear

1992

fDate

6/6/1992 12:00:00 AM

Firstpage

223

Lastpage

234

Abstract

The application of neural networks to nuclear power plants for fault diagnostics is a very challenging task. How to select proper input variables for neural networks from hundreds of plant processing variables is crucially important to the success. Genetic algorithms are used in this study to guide the search for optimal combination of inputs for the neural networks to reach the criteria of fewer inputs, faster training, and more accurate recall. Data from Tennessee Valley Authority (TVA) Watts Bar Nuclear Power Plant simulator are used to demonstrate the potential applications of genetic algorithms and neural networks to nuclear power plants

Keywords

accidents; digital simulation; fission reactor safety; genetic algorithms; neural nets; nuclear engineering computing; nuclear power stations; PWR; Tennessee Valley Authority; Watts Bar Nuclear Power Plant simulator; accidents; digital simulation; fault diagnostics; genetic algorithms; neural networks; nuclear engineering computing; nuclear power plants; recall; safety; Accidents; Computational modeling; Condition monitoring; Genetic algorithms; Input variables; Neural networks; Pattern classification; Power engineering and energy; Power generation; Sensor systems;

fLanguage

English

Publisher

ieee

Conference_Titel

Combinations of Genetic Algorithms and Neural Networks, 1992., COGANN-92. International Workshop on

Conference_Location

Baltimore, MD

Print_ISBN

0-8186-2787-5

Type

conf

DOI

10.1109/COGANN.1992.273937

Filename

273937