Online Prognosis of Stochastic Discrete Event Systems

DOI

10.2991/ceis-16.2016.88How to use a DOI?

Keywords

stochastic automaton; online prognosis; prognostic probability; state estimation; Markov chain

Abstract

In this paper, we investigate how to calculate the probability of fault occurrence in the future based on the observations online. In a stochastic discrete event system framework, we first calculate the probability of fault occurrence for pre-fault event sequences which do not include fault events. Our idea is to attach label to each state in the stochastic automaton and convert it into a Markov chain. Based on the Markov chain, we determine the probability of fault occurrence by calculating the sum of probabilities of some states in the Morkov Chain. Second, for every pre-fault observable event sequence, we find all the possible pre-fault event sequences and the probability of their occurrences. Our idea is to update the current state estimate for the observable event sequence and then calculate the probability of every state based on the current state estimate which can be done with polynomial complexity. Combining these two results, the online prognosis problem is successfully solved.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Feng-Jiao Lian
AU  - Shao-Long Shu
PY  - 2016/11
DA  - 2016/11
TI  - Online Prognosis of Stochastic Discrete Event Systems
BT  - Proceedings of the 2016 International Conference on Computer Engineering and Information Systems
PB  - Atlantis Press
SP  - 435
EP  - 443
SN  - 2352-538X
UR  - https://doi.org/10.2991/ceis-16.2016.88
DO  - 10.2991/ceis-16.2016.88
ID  - Lian2016/11
ER  -