Identification of aircraft gas-turbine dynamics using B-splines neural networks

Lopes, J. C.; Ruano, Antonio; Fleming, P. J.

http://hdl.handle.net/10400.1/2209

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
38.pdf		135.07 KB	Adobe PDF	Download

Send Feedback

Authors

Lopes, J. C.

Ruano, Antonio

Fleming, P. J.

Abstract(s)

This paper describes preliminary results regarding the development of a B-splines neural network model of the fuel feed to shaft-speed dynamics of a twin-shaft gas turbine engine. Data recorded from practical testing of the turbine to a multisine input were employed, and models were identified at different points along the turbine operating curve. B-splines neural networks have been found to be good models of the system, delivering good results especially for long-range predictions.

Keywords

Gas Turbines Systems Identification Neural Networks

URI

http://hdl.handle.net/10400.1/2209

Citation

Lopes, J. C.; Ruano, A. E.; Fleming, P. J. Identification of aircraft gas-turbine dynamics using B-splines neural networks, Trabalho apresentado em 6th IFAC Workshop on Algorithms and Architectures for Real-Time Control (AARTC' 00), In 6th IFAC Workshop on Algorithms and Architectures for Real-Time Control (AARTC' 00), Palma de Mallorca, 2000.