Causal analysis for troubleshooting and decision support system

Author

Byoung Uk Kim ; Vohnout, Sonia ; Mikkola, Esko ; Mingyang Li ; Jian Liu

Author_Institution

Ridgetop Group, Inc., Tucson, AZ, USA

fYear

2011

fDate

20-23 June 2011

Firstpage

1

Lastpage

7

Abstract

Troubleshooting and decision support system with reasoning is an active research topic, and causal analysis for complex component interactions in complex systems has remained a critical challenge to be overcome. We developed an innovative, constraint-based causal analysis to better detect, isolate, and troubleshoot complex systems. The feasibility of the causal Bayesian network (CBN) approach has been proven with implementation using test data acquired from electromechanical actuator (EMA) systems. The validation step is facilitated by comparing the trained CBN with original structure and shows the flexibility and extensibility of our solutions. This causal analysis processing in integrated system health management (ISHM) will enable enhancements in flight safety and condition based maintenance (CBM) by increasing availability and mission effectiveness while reducing maintenance costs.

Keywords

aerospace engineering; aerospace safety; aircraft maintenance; belief networks; condition monitoring; decision support systems; inference mechanisms; causal Bayesian network; condition based maintenance; constraint-based causal analysis; decision support system; electromechanical actuator; flight safety; integrated system health management; reasoning; troubleshooting; Actuators; Bayesian methods; Cognition; Mathematical model; Skeleton; Switched-mode power supply; Training; Bayesian network; causal analysis; decision support; fault; health management; troubleshooting;

fLanguage

English

Publisher

ieee

Conference_Titel

Prognostics and Health Management (PHM), 2011 IEEE Conference on

Conference_Location

Montreal, QC

Print_ISBN

978-1-4244-9828-4

Type

conf

DOI

10.1109/ICPHM.2011.6024344

Filename

6024344