A finite-volume scheme for dynamic reliability models |
| |
Authors: | Cocozza-Thivent C; Eymard R; Mercier S |
| |
Institution: |
Laboratoire d'Analyse et de Mathématiques Appliquées (CNRS UMR 8050), Université de Marne-la-Vallée, 5, boulevard Descartes, Champs-sur-Marne, 77454 Marne-la-Vallée Cedex 2, France
|
| |
Abstract: | ** Email: christiane.cocozza{at}univ-mlv.fr*** Email: robert.eymard{at}univ-mlv.fr**** Email: sophie.mercier{at}univ-mlv.fr In a model arising in the dynamic reliability study of a system,the probability of the state of the system is completely describedby the ChapmanKolmogorov equations, which are scalarlinear hyperbolic partial differential equations coupled bytheir right-hand side, the solution of which are probabilitymeasures. We propose in this paper a finite-volume scheme toapproximate these measures. We show, thanks to the proof ofthe tightness of the approximate solution, that the conservationof the probability mass leads to a compactness property. Theconvergence of the scheme is then obtained in the space of continuousfunctions with respect to the time variable, valued in the setof probability measures on graphic: see PDF] . We finally show on a numerical example the accuracy and efficiencyof the approximation method. |
| |
Keywords: | dynamic reliability Markov processes finite-volume method weak convergence |
本文献已被 Oxford 等数据库收录! |
|