The release and uncontrolled explosion of hydrogen as a result of an accident is a central problem in nuclear and process plant safety research. For the structural-mechanical analysis and integrity evaluation of building structures and components, it is necessary to predict explosion loads as accurately as possible. A fully resolved Direct Numerical Simulation (DNS) of the flame propagation in industry scale accident scenarios, which also considers small-scale fluid and thermodynamic effects, is not feasible due to the large spectrum of time and length scales involved. Current calculation tools use low-resolution simulations, enabling efficient calculation of explosion loads. The influence of small-scale effects, which can no longer be calculated directly due to the coarser resolution, must be taken into account using suitable subgrid models. The quality of these models can have a considerable influence on the accuracy of the predicted explosion loads. The goal of this project is the creation of a subgrid model of the Richtmyer-Meshkov Instability (RMI), which can be a significant factor, contributing to flame acceleration and possibly detonation in geometrically confined explosions due to the interaction of shock waves and flames. In the context of combustion the RMI is characterized by a heavy build-up of baroclinic torque across the disturbed flame brush, causing a severe increase in flame wrinkling, greatly increasing the
flame surface and the integral reaction rate.