Stochastic discrete velocity averaging lemmas and Rosseland approximation
Author(s) -
Nathalie Ayi
Publication year - 2017
Publication title -
séminaire laurent schwartz — edp et applications
Language(s) - English
Resource type - Journals
ISSN - 2266-0607
DOI - 10.5802/slsedp.100
Subject(s) - class (philosophy) , mathematics , set (abstract data type) , variable (mathematics) , pure mathematics , combinatorics , mathematical analysis , computer science , artificial intelligence , programming language
In this note, we investigate some questions around velocity averaging lemmas, a class of results which ensure the regularity of the “velocity average” ∫ f(x, v)ψ(v) dμ(v) when f and v · ∇xf both belong to L, p ∈ [1,∞) and the measured set of velocities (V , dμ) satisfy a nondegeneracy assumption. We are interested in the case when the variable v lies in a discrete subset of R. We present results obtained in collaboration with T. Goudon in [2]. First of all, we provide a rate, depending on the number of velocities, to the defect of H regularity which is reached when v ranges over a continuous set. Second of all, we show that the H regularity holds in expectation when the set of velocities is chosen randomly. We apply this statement to obtain a consistency result for the diffusion limit in the case of the Rosseland approximation.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom