Abstract
We present a completely automatic biasing technique where the parameters of the biased simulation are deduced from the solution of the adjoint transport equation calculated by collision probabilities. In this study we shall estimate the importance function through collision probabilities method and we shall evaluate its possibilities thanks to a Monte Carlo calculation. We have run simulations with this new biasing method for one-group transport problems with isotropic shocks (one dimension geometry and X-Y geometry) and for multigroup problems with anisotropic shocks (one dimension geometry). For the anisotropic problems we solve the adjoint equation with anisotropic collision probabilities. The results show that for the one-group and homogeneous geometry transport problems the method is quite optimal without Splitting and Russian Roulette technique but for the multigroup and heterogeneous X-Y geometry ones the figures of merit are higher if we add Splitting and Russian Roulette technique.
Citation Formats
Morillon, B, Both, J P, and Nimal, J C.
Importance function by collision probabilities for Monte Carlo code Tripoli.
France: N. p.,
1993.
Web.
Morillon, B, Both, J P, & Nimal, J C.
Importance function by collision probabilities for Monte Carlo code Tripoli.
France.
Morillon, B, Both, J P, and Nimal, J C.
1993.
"Importance function by collision probabilities for Monte Carlo code Tripoli."
France.
@misc{etde_10154490,
title = {Importance function by collision probabilities for Monte Carlo code Tripoli}
author = {Morillon, B, Both, J P, and Nimal, J C}
abstractNote = {We present a completely automatic biasing technique where the parameters of the biased simulation are deduced from the solution of the adjoint transport equation calculated by collision probabilities. In this study we shall estimate the importance function through collision probabilities method and we shall evaluate its possibilities thanks to a Monte Carlo calculation. We have run simulations with this new biasing method for one-group transport problems with isotropic shocks (one dimension geometry and X-Y geometry) and for multigroup problems with anisotropic shocks (one dimension geometry). For the anisotropic problems we solve the adjoint equation with anisotropic collision probabilities. The results show that for the one-group and homogeneous geometry transport problems the method is quite optimal without Splitting and Russian Roulette technique but for the multigroup and heterogeneous X-Y geometry ones the figures of merit are higher if we add Splitting and Russian Roulette technique.}
place = {France}
year = {1993}
month = {Jun}
}
title = {Importance function by collision probabilities for Monte Carlo code Tripoli}
author = {Morillon, B, Both, J P, and Nimal, J C}
abstractNote = {We present a completely automatic biasing technique where the parameters of the biased simulation are deduced from the solution of the adjoint transport equation calculated by collision probabilities. In this study we shall estimate the importance function through collision probabilities method and we shall evaluate its possibilities thanks to a Monte Carlo calculation. We have run simulations with this new biasing method for one-group transport problems with isotropic shocks (one dimension geometry and X-Y geometry) and for multigroup problems with anisotropic shocks (one dimension geometry). For the anisotropic problems we solve the adjoint equation with anisotropic collision probabilities. The results show that for the one-group and homogeneous geometry transport problems the method is quite optimal without Splitting and Russian Roulette technique but for the multigroup and heterogeneous X-Y geometry ones the figures of merit are higher if we add Splitting and Russian Roulette technique.}
place = {France}
year = {1993}
month = {Jun}
}