scholarly journals Evaluation of the accuracy of the fault diagnosis system based on the signed directed graph. Evaluation by use of the greatest sets of candidates.

1989 ◽  
Vol 15 (2) ◽  
pp. 395-402 ◽  
Author(s):  
BohYoh Shibata ◽  
Hisayoshi Matsuyama
1990 ◽  
Vol 16 (5) ◽  
pp. 899-906 ◽  
Author(s):  
Fumihiko Gujima ◽  
Bohyoh Shibata ◽  
Yoshifumi Tsuge ◽  
Junichi Shiozaki ◽  
Hisayoshi Matsuyama ◽  
...  

1995 ◽  
Vol 31 (1) ◽  
pp. 98-107 ◽  
Author(s):  
Kazuhiro TAKEDA ◽  
BohYoh SHIBATA ◽  
Yoshifumi TSUGE ◽  
Hisayoshi MATSUYAMA

2012 ◽  
Vol 232 ◽  
pp. 359-363
Author(s):  
Zheng Yan ◽  
Xiao Hui Peng ◽  
Yu Qiang Cheng ◽  
Jian Jun Wu

Signed Directed Graph (SDG) has been widely applied to model the cause and effect behavior of process systems in recent years. However, SDG-based diagnosis has poorly discriminatory ability, because of the information loss while going from quantitative to qualitative domain. In this paper, a new method combining SDG with quantitative knowledge is presented to improve the discriminatory ability. In the method, a hybrid reasoning (forward and backward) strategy based on assumption and verification was applied to find all the potential fault sources and corresponding consistent paths in SDG model. Then the SDG-based method was modified by integrating governing equation and temporal information of the system, in order to improve the discriminatory ability. The method has been validated by the artificial telemetry data, and the effectiveness of the method has been confirmed. The method proposed can provide important practical value for the development of on-board fault diagnosis system of spacecraft propulsion systems.


Sign in / Sign up

Export Citation Format

Share Document