Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Acta Armamentarii》 2007-11
Add to Favorite Get Latest Update

Research on Intelligent Built-in Test Fault Diagnosis of More-electric Aircraft Electrical Power System

LIU Zhen,LIN Hui(College of Automation,Northwestern Polytechnical University,Xi'an 710072,Shaanxi,China)  
An intelligent built-in test(BIT) technology based on generalized learning vector quantization(GLVQ) neural network was proposed and applied to the BIT system of more-electric aircraft electrical power system(MEAEPS).A power spectrum analysis method was employed to get the characteristics of BIT signals in frequency domain.In order to reduce the dimension of eigenvectors,the spectrum characteristics of BIT signals were compressed by using the wavelet packet decomposition,and the energy of each frequency-band was computed to form the final eigenvectors,which were used as learning samples to train the GLVQ neural network.Since the original GLVQ algorithm suffered from several major problems,some modifications were made and a supervised LVQ layer was added to the GLVQ network,which made the boundaries among the fault classes more discriminative than using the GLVQ network alone.The proposed method was applied to the BIT system of MEAEPS,and the results showed that the proposed method is promising to improve BIT performance of MEAEPS.
【Fund】: 航空科学基金资助项目(04F53036)
【CateGory Index】: V242
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved