Dynamic reliability is an extension of classical dependability theory. The term dynamic refers to environmental and operational changes of the nominal design conditions that can affect the performance and the failure be-havior of a system. Traditional techniques of dependability are not suitable for modelling such class of prob-lems and, at the state of the art, no user-friendly software tools able to solve dynamic dependability analyses of complex system exist. Encouraged by the results of earlier works, in this paper, the authors present DY-RAATSS, an extension of the Matlab toolbox RAATSS (Reliability Availability Adaptive Transition System Solver), that couples Repairable Dynamic Fault Trees with deterministic physical models developed in Sim-ulink. Following the concept of “separation of concerns”, the authors will show how to breakdown a dynamic reliability problem into two mutually dependent processes, the deterministic and the stochastic one and how to use DY-RAATSS to build and solve the two models in combination, under a single simulation environ-ment. The final assembly describes exhaustively a dynamic reliability problem, providing a software model that can be easily simulated within the framework. A case study of a non-linearly aging temperature control system in presence of a failure detector is used to show the approach.
|Titolo:||Hybrid-pair modelling in dynamic reliability: Concepts, tool implementation and applications|
|Data di pubblicazione:||2015|
|Appare nelle tipologie:||4.1 Contributo in Atti di convegno|