Volume 55, Number 3, May-June 2021
|Page(s)||1039 - 1065|
|Published online||31 May 2021|
Multilevel Monte Carlo finite volume methods for random conservation laws with discontinuous flux
Department of Mathematics, University of Würzburg, Würzburg, Germany
2 Department of Mathematics, University of Oslo, Oslo, Norway
3 Seminar for Applied Mathematics, ETH Zürich, Switzerland
* Corresponding author: email@example.com
Accepted: 9 March 2021
We consider conservation laws with discontinuous flux where the initial datum, the flux function, and the discontinuous spatial dependency coefficient are subject to randomness. We establish a notion of random adapted entropy solutions to these equations and prove well-posedness provided that the spatial dependency coefficient is piecewise constant with finitely many discontinuities. In particular, the setting under consideration allows the flux to change across finitely many points in space whose positions are uncertain. We propose a single- and multilevel Monte Carlo method based on a finite volume approximation for each sample. Our analysis includes convergence rate estimates of the resulting Monte Carlo and multilevel Monte Carlo finite volume methods as well as error versus work rates showing that the multilevel variant outperforms the single-level method in terms of efficiency. We present numerical experiments motivated by two-phase reservoir simulations for reservoirs with varying geological properties.
Mathematics Subject Classification: 35L65 / 35R05 / 65C05 / 65M12
Key words: Uncertainty quantification / conservation laws / discontinuous flux / numerical methods
© EDP Sciences, SMAI 2021
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.