By Topic

Migrating Fault Trees To Decision Trees For Real Time Fault Detection On International Space Station

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
C. Lee ; SAIC NASA Ames Research Center Moffet Field, CA. 94035 650-604-6054 clee@mail.arc.nasa.gov ; R. L. Alena ; P. Robinson

Fault Tree Analysis shows the possible causes of a system malfunction by enumerating the suspect components and their respective failure modes that may have induced the problem. Complex systems often use fault trees to analyze the faults. Fault diagnosis, when error occurs, is performed by engineers and analysts performing extensive examination of all data gathered during the mission. International Space Station (ISS) control center operates on the data feedback from the system and decisions are made based on threshold values by using fault trees. Since those decision-making tasks are safety critical and must be done promptly, the engineers who manually analyze the data are facing time challenge. To automate this process, this paper present an approach that uses decision trees to capture the contents of fault trees and detect faults by running the telemetry data through the decision trees in real time. Decision trees (also called classification trees) are the binary trees built from data samples and can classify the objects into different classes. In our case, the decision trees can classify different fault events or normal events. Given a set of data samples, decision trees can be built and trained, and then by running the new data through the trees, classification and prediction can be made. In this way, diagnostic knowledge for fault detection and isolation can be represented as diagnostic rules; we call this tree the diagnostic decision tree (DDT). By showing the fault path in decision trees, we also can point out the root cause when a fault occurs. Since all the procedures and algorithms are available to build decision trees, the trees built are cost effective and time effective. Because the diagnostic decision trees are based on available data and previous knowledge of subsystem logic, the DDT can also be trained to predict faults and detect unknown faults. Based on this, the needs for on-board real time diagnostics can readily be met. Diagnostic Decision Tree- s are built based on the fault trees as static trees that serve as the fundamental diagnostic trees, and the dynamic DDTs are built over time from vehicle telemetry data. The dynamic DDT will add the functionalities of prediction, and will be able to detect unknown faults. Crew or maintenance engineers can use the decision tree system without having previous knowledge or experience about the diagnosed system. To our knowledge, this is the first paper to propose a solution to build diagnostics decision trees from fault trees, which convert the reliability analysis models to diagnostic models. We show through mapping and ISS examples that the approach is feasible and effective. We also present future work and development

Published in:

2005 IEEE Aerospace Conference

Date of Conference:

5-12 March 2005