Section: New Results

Hidden Markov Model for the detection of a degraded state in an optronic equipment

Participants : Camille Baysse, Anne Gégout-Petit, Jérôme Saracco.

As part of optimizing the reliability, Thales Optronics now includes systems that examine the state of its equipment. This function is performed by HUMS (Health & Usage Monitoring System). We hope to implement a program based on these observations that can determine the lifetime of this optronic equipment. Our study focuses on a simple example of HUMS. As part of our research, we are interested in a variable called "time-to cold" noted TMF, which reflects the state of system. Using this information about this variable, we seek to detect as soon as possible a degraded state and propose maintenance before failure. For this we use a hidden Markov model. The state of our system at time $t$ is then modeled by a Markov chain $X_t$. However we do not observe directly this chain but indirectly through the TMF, a noisy function of this chain. Thanks to filtering equations, we obtained results on the probability that an equipment breaking down at time $t$, knowing the history of the TMF until this moment. We have subsequently studied this methodology with simulated data. Then finally we applied these results on the analysis of our real data and we have checked that the results are consistent with the reality. So using this method could allow the company to recall equipments which are estimated in deteriorated state and do not control those estimated in stable state. Thales Optronics could improve its maintenance system and reduce its cost function.

This work is a part of the CIFRE PhD of Camille Baysse also supervised for the Thales part by Didier Bihannic and Michel Prenat. It was presented in an national conference [38] and is submitted for publication in an national per-reviewed journal [58] .