Simulation of rare and transient events in rarefied gases

• Applicants: Dr. Georgii Oblapenko; Professor Dr. Manuel Torrilhon
• Subject Area: Fluid Mechanics
• Term: since 2026
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 578802155

The present proposal is concerned with the development of new capabilities of the Direct Simulation Monte Carlo (DSMC) method, a stochastic particle-based numerical method widely used for simulating rarefied gas flows and plasmas, where usual continuum mechanics-based simulation approaches are not valid. The method is extensively used in research and industry, however, it has significant deficiencies when simulating unsteady flows, flows with trace species, and flows where low-probability events have a large impact on the flow physics. Such flow scenarios occur for example in unsteady plume expansion and impingement processes in satellite propulsion and landing, radio frequency (RF)-driven plasmas in plasma processing technology, and electrical breakdown processes in circuits, and are relevant for many research and engineering applications. For such flows an efficient, adaptive, and accurate simulation capability is oftentimes lacking. The proposal leverages the variable-weight DSMC framework and is concerned with the development of 1) new computational particle splitting schemes for improved resolution of low-probability processes, 2) new structure-preserving particle merging schemes, and 3) particle-history-tracking methods for solution adaptivity and quantification of the impact of various processes on the quantities of interest. These developments will allow for higher resolution of physical processes in rarefied flows, improved computational performance, and possibility of adaptive process-aware simulations. The developed methods and extensions of DSMC will be implemented in an open-source DSMC code and applied to modelling of plume expansions and RF plasmas to verify them and quantify the impact of the new methods on the simulation accuracy and computational cost. The new DSMC algorithms developed in the present proposal will significantly improve the capability of DSMC to accurately simulate transient flows with low-probability events, leading to improved modelling of a variety of problems relevant to aerospace engineering, semi-conductor manufacturing, and plasma processing technology.

• DFG Programme: Research Grants