Multi-Fidelity Moment Method for Kinetic Equations with Uncertainties


Yanli Wang, Beijing Computational science Research Ceter


2021.01.14 15:00-16:00


Room 338, No 5 Science Building


The numerical simulation of the kinetic equations with stochastic parameters is full of challenges due to high dimensionality. Recently, we have introduced a globally hyperbolic moment method to solve the deterministic kinetic equations, which has been successfully applied in the simulation of micro-flow, plasma, and radiative transfer equations. In this work, the moment method is extended to the stochastic kinetic equation. Moreover, the moment models with different fidelities are adopted to accelerate the slow convergence of the standard Monte Carlo method for the stochastic parameters. Several numerical examples validate the high efficiency of this multi-fidelity moment method.


Dr. Yanli Wang is a research assistant professor in Beijing Computational science Research Ceter. She obtained her Ph.D degree from Peking University in 2014, under the supervision of Prof. Ruo Li. Dr. Wang’s research interests include model reduction in kinetic theory, moment method to solve Boltzmann and related equations. She has published several papers on JFM, SINUM, JCP, etc.