Modelling, identification and fault diagnosis of a simulated model of an industrial gas turbine

Author

Yousefi, I. ; Khaloozadeh, Hamid ; Ashraf-Modarres, Ali

Author_Institution

MAPNA Electr. & Control Eng. & Manuf. Co. - MECO, Karaj, Iran

fYear

2011

fDate

18-19 Oct. 2011

Firstpage

1

Lastpage

6

Abstract

The objective of this paper is to model, identify, and detect and isolate faults to an industrial gas turbine. The detection scheme is based on the generation of so-called “residuals” that are errors between estimated and measured variables of the process. An ARX model is used for residual generation, while for residual evaluation a neural network classifier for MLP is used. The proposed fault detection and isolation tool has been tested on a single-shaft industrial gas turbine model.

Keywords

fault diagnosis; gas turbines; multilayer perceptrons; pattern classification; power generation faults; power system identification; power system measurement; power system simulation; shafts; ARX model; MLP; fault detection tool; fault diagnosis; fault isolation tool; multilayer perceptron; neural network classifier; residual generation; simulation model; single-shaft industrial gas turbine model; Computational modeling; Equations; Fault detection; Fault diagnosis; Mathematical model; Nonlinear dynamical systems; Turbines; Fault Detection; Fault Diagnosis; Gas Turbine; Identification Methods; Neural Networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Thermal Power Plants (CTPP), 2011 Proceedings of the 3rd Conference on

Conference_Location

Tehran

Print_ISBN

978-1-4799-0591-1

Type

conf

Filename

6576982