A Particle-based Multiscale Solver for Compressible Liquid-Vapor Flow
This work provides a novel multiscale framework for simulating two-phase flows with sharp interfaces, addressing a known bottleneck in computational fluid dynamics for researchers modeling liquid-vapor systems.
The authors developed a multiscale solver for compressible liquid-vapor flow that uses a particle-based microscale model to replace ad-hoc constitutive relations, enabling more accurate simulation of sharp interfaces without reporting specific numerical gains.
To describe complex flow systems accurately, it is in many cases important to account for the properties of fluid flows on a microscopic scale. In this work, we focus on the description of liquid-vapor flow with a sharp interface between the phases. The local phase dynamics at the interface can be interpreted as a Riemann problem for which we develop a multiscale solver in the spirit of the heterogeneous multiscale method, using a particle-based microscale model to augment the macroscopic two-phase flow system. The application of a microscale model makes it possible to use the intrinsic properties of the fluid at the microscale, instead of formulating (ad-hoc) constitutive relations.