QMC Working Group I: Parallel Monte Carlo Neutronics Simulation

Group Leaders:

Weekly Meetings: Meeting Times to be determined

Description:

We simulate neutron flux intensity far from the source or behind shielding. Applications include nuclear power plants where we model radiation in machine rooms near the core, but behind shielding. Monte Carlo simulation randomly generates particles using information on sources and scattering properties of the problem domain and then follows those particles until they either leave the domain or are absorbed. This is not a PDE-based approach and has no discretization error.

Fluxes are very small and standard MC would resolve the radiation field inefficiently. We address this problem with CADIS (Consistent Adjoint-Driven Importance Sampling) which eliminates particles that move away from a region of interest. The computation must be done in parallel and the research focus of this project is to optimize the domain decomposition that is central to parallelization.

News and Updates: Coming soon…

SAMSI Directorate Liaison: Ilse Ipsen

Questions: email qmc@samsi.info

Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applied Mathematics (QMC)