Wear trend forecast of aero-engine based on improved RBF neural network

Author

Jiang, Liying ; Wang, Lei ; Xi, Jianhui ; Li, Yibo ; Zhang, Yan

Author_Institution

Coll. of Autom., Shenyang Inst. of Aeronaut. Eng., Shenyang, China

fYear

2010

fDate

26-28 May 2010

Firstpage

2234

Lastpage

2237

Abstract

An improved RBF neural network is proposed in this paper, which is to solve the problem of wear trend prediction accuracy for aero-engine. The number of neurons in the input layer of this improved model is determined by the ideology of equal dimensionality vectors, obtain the optimal model, then the content of iron and silicon element in the spectral can be predicted by the trained model, finally wear trend of aero-engine is determined. The simulation results show that, comparing with other models, the improved RBF neural network has great practicability and satisfied prediction accuracy in the field of wear trend.

Keywords

aerospace engines; condition monitoring; fault diagnosis; mechanical engineering computing; radial basis function networks; vectors; aeroengine trend prediction accuracy; aeroengine wear trend forecast; equal dimensionality vector; improved RBF neural network; satisfied prediction accuracy; Accuracy; Engines; Information analysis; Lubrication; Neural networks; Neurons; Petroleum; Predictive models; Radial basis function networks; Support vector machines; Aero-Engine; Equal Dimensionality Vectors; RBF Neural Network; Spectra; Wear Trend forecast;

fLanguage

English

Publisher

ieee

Conference_Titel

Control and Decision Conference (CCDC), 2010 Chinese

Conference_Location

Xuzhou

Print_ISBN

978-1-4244-5181-4

Electronic_ISBN

978-1-4244-5182-1

Type

conf

DOI

10.1109/CCDC.2010.5498830

Filename

5498830