Evaluation of the mechatronic systems reliability under parametric uncertainties

The aim of this paper is to evaluate the predicted reliability of mechatronic systems, by taking into account the epistemic uncertainties. The work reported here presents a new methodology based on integrating the belief functions in the Petri net (PN) model, in order to create a belief network, and to show how to propagate the parametric uncertainties in reliability models. Some notions of uncertainty related to systems reliability are presented; subsequently a brief definition of the belief function and its application in reliability studies are given and finally its integration in PN is detailed. In order to take into account the interactive aspect of mechatronic systems, we introduce the uncertainties associated to this interaction, by implementing the new method proposed by using belief network. Secondly, we study the propagation of these interaction uncertainties in system reliability. Finally, an industrial example of an "intelligent actuator" is developed, applying the proposed methodology. Keywords— Reliability, predictive reliability, uncertainties, epistemic uncertainties, Belief Function, Petri Net (PN), mechatronics, interactions