Health management for aircraft system using Bayesian probability model

The health management for aircraft system is difficult problem when the system is rife with nonlinear/non-Gaussian time evolution, the model parameters and sensors measurements are subject to uncertainty, and the diagnosis task suffers from some real-time constrains. This paper discusses the most re...

Full description

Saved in:
Bibliographic Details
Published in2016 IEEE International Conference on Aircraft Utility Systems (AUS) pp. 1203 - 1208
Main Authors Wei, W. Feng, Pei, Z. C., Hu, D. D.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.10.2016
Subjects
Aircraft
Atmospheric modeling
Bayes methods
Integrated circuit modeling
Real-time systems
Sensors
Uncertainty
Online AccessGet full text
DOI10.1109/AUS.2016.7748147

Cover

More Information
Summary:The health management for aircraft system is difficult problem when the system is rife with nonlinear/non-Gaussian time evolution, the model parameters and sensors measurements are subject to uncertainty, and the diagnosis task suffers from some real-time constrains. This paper discusses the most relatively recent researches about Bayesian probability model, which focuses on the Bayesian networks (BNs), dynamic Bayesian network (DBN) and arithmetic circuit (AC), and then proposes an novel approach to build a robust dynamic arithmetic circuit (DAC) to successfully address this problem. The experiments results show that the DAC, compared with BN, AC and DBN, not only provides reliable online diagnosis under the presence of uncertainty, but also meets the strict time deadlines of health management.
DOI:10.1109/AUS.2016.7748147