Usage profile in physical systems modelized with stochastic hybrid automata

This paper focuses on a way to represent complex systems by using a modification in the Stochastic Hybrid Automata (SHA) first designed by the CRAN by expending finite-state automaton. SHA are used to represent continuous physical phenomenon (aging for example) that can induce discrete event (failur...

Full description

Saved in:
Bibliographic Details
Main Authors: Hequet, Gael, Brinzei, Nicolae, Petin, Jean-Francois
Format: Conference Proceeding
Language:English
Subjects:
Aging
Automata
Automobiles
Degradation
hybrid system
Reliability
Stochastic Hybrid Automaton
Stochastic processes
Tires
Usage profile
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper focuses on a way to represent complex systems by using a modification in the Stochastic Hybrid Automata (SHA) first designed by the CRAN by expending finite-state automaton. SHA are used to represent continuous physical phenomenon (aging for example) that can induce discrete event (failures). With this tool, it is possible to represent the lifecycle of a component, its aging, and failures. To have a more precise model, a new addition to the SHA can be done. A system can have many usage profiles and each of these profiles can change the aging and probability of failures of the system. This work aims to add this view to the SHA by creating a virtual racing car. The driver of this racing car can drive in different ways. Depending on his driving profile, the components of the car (body, engine, tyres, electronics, clutch, transmission, brakes, and gearbox) can suffer from different types of degradations. The level of degradation of a component has an impact on its probability of failure. In addition to this impact given by the degradation, another impact is considered, the profile of driving itself. Depending on the driving profile, the failure rate can vary. In this idea, different failure rates are linked to each driving profile. The impacts of these profile variation will be shown in this paper.
ISSN:2575-677X
DOI:10.1109/IDT52577.2021.9497617