2015 QPRC

When importance sampling meets stochastic simulations.

Abstract: As simulation models become more realistic, reliability evaluation remains challenging due to the high computational cost of each simulation replication. We provide computationally efficient methods for reliability evaluation using stochastic simulations that generate random outputs given a fixed input. A new sampling method has been devised and validated using aeroelastic simulators developed by the U.S. National Renewable Energy Laboratory.  This is joint work with Eunshin Byon (University of Michigan) and Nan Chen (National University of Singapore).