Randomized Quasi Monte Carlo Sampling Techniques in Nuclear Reactor Uncertainty Assessment, invited

摘要

Sampling techniques are an essential technique in uncertainty assessment in nuclear reactors. Unfortunately, the rate of convergence of N~(-0.5) (where N is the number of samples) of Monte Carlo (MC) methods may result in the need of using too many samples to achieve a prescribed level of error. In this paper, we analyze ways of reducing the variance and therefore the number of samples need for a given level of confidence by using randomized quasi Monte Carlo (RQMC) methods.