Title :
Neural Network Integration Fusion Model and Application
Author :
Zhang, Xiaodan ; Niu, Zhendong
Author_Institution :
Sch. of Comput. Sienstist & Technol., Beijing Inst. of Technol., Beijing
Abstract :
A new fusion model is proposed, which is the combination of BP neural networks and rough set algorithm, to solve the problems of low precision rate in aircraft engine fault diagnosis by traditional methods. The method realizes feature level fusion of all subjective data and expert experiments on different parts of engine, and the predominance compensation of different models. In simulation experiment, the method proposed in the paper can improve diagnosis precision 5.0% more than expert system.
Keywords :
aerospace computing; aerospace engines; backpropagation; fault diagnosis; neural nets; rough set theory; BP neural networks; aircraft engine fault diagnosis; feature level fusion; neural network integration fusion model; rough set algorithm; Aircraft propulsion; Algorithm design and analysis; Application software; Computer networks; Diagnostic expert systems; Fault diagnosis; Intelligent networks; Intelligent systems; Neural networks; Uncertainty;
Conference_Titel :
Intelligent Systems Design and Applications, 2008. ISDA '08. Eighth International Conference on
Conference_Location :
Kaohsiung
Print_ISBN :
978-0-7695-3382-7
DOI :
10.1109/ISDA.2008.331
