Fault diagnosis in hydraulic turbine governor based on BP neural network

Author

Xiaohui, Yu ; Ruijin, Liao ; Chenguo, Yao

Author_Institution

Ge Zhou Ba Hydroelectric Power Station, YiChang, China

Volume

1

fYear

2001

fDate

2001

Firstpage

335

Abstract

This paper describes a new fault diagnosis model of the hydraulic turbine governing system with the advanced BPNN (backpropagation neural network), which consists of three layers: i.e. input layer (17 neurons), hidden layer, output layer (13 neurons). It is proved that the system can rind the faults correctly in GeZhouBa hydroelectric power station, and it can conduct the faults examination and repair of governing systems. So this diagnosis system should be applied widely in practice

Keywords

backpropagation; diagnostic expert systems; fault diagnosis; hydraulic turbines; hydroelectric power stations; machine testing; maintenance engineering; neural nets; turbogenerators; China; Gezhouba hydroelectric power station; backpropagation neural network; fault diagnosis model; hidden layer; hvdraulic turbine governing system; input laver; output layer; Artificial intelligence; Fault diagnosis; Hydraulic turbines; Intelligent networks; Neural networks; Neurons; Power generation; Power system faults; Power system modeling; Power system reliability;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical Machines and Systems, 2001. ICEMS 2001. Proceedings of the Fifth International Conference on

Conference_Location

Shenyang

Print_ISBN

7-5062-5115-9

Type

conf

DOI

10.1109/ICEMS.2001.970680

Filename

970680