Compressible fluid simulations get new improvements

Numerical simulations can be invaluable for studying reacting flows, like those taking place in engines or other chemical engineering processes. These numerical simulations help predict the efficiency of engines and other devices. Modeling molecular diffusion and thermal conduction is important for numerical simulations of these reacting flows because chemical reactions are usually controlled by the temperature and mixing of chemicals at the molecular level.

In Physics of Fluids, researchers propose a new model for molecular diffusion and thermal conduction in compressible turbulence, expanding on what’s called the mixing volume model (MVM). MVM is based on interactions among spatially distributed particles in a finite volume — and can be easily applied to Lagrangian (classical) numerical simulations of reacting flows.

MVM was developed for incompressible flows with constant diffusivity coefficient and density. The researchers extend MVM to compressible turbulence, in which the density, viscosity and diffusivity coefficients are not constants.

MVM does a good job of predicting molecular diffusion and thermal conduction for a wide range of mixing volume sizes and numerous mixing particles. Co-author Tomoaki Watanabe said that the model works well for passive scalar and temperature fields represented by Lagrangian particles for a wide range in Mach number, the measure used to approximate flow compressibility.

This work helps improve numerical methods used for predicting reacting flows at low computational cost. Next, the authors plan to use MVM in Lagrangian simulations of high-speed reacting flows.

Source: “Modeling of molecular diffusion and thermal conduction with multi-particle interaction in compressible turbulence,” by Y. Tai, T. Watanabe, and K. Nagata, Physics of Fluids (2018). The article can be accessed at https://doi.org/10.1063/1.5018248 .