Scale-Bridging Techniques for Disperse Multiphase Flows
Abstract: Disperse multiphase systems—comprising interacting particles, droplets, and bubbles—exhibit rich collective behaviors, nonlinear dynamics, and emergent structures across a wide range of spatial and temporal scales. These systems are central to many scientific and engineering challenges and play an essential role in multiple DOE mission areas, including national security, clean energy production, advanced manufacturing, and predictive modeling of high-speed ejecta and debris in explosive events. Despite their importance, the inherent multiscale nature of particle systems makes predictive modeling remarkably difficult. Microscale particle–fluid and particle–particle interactions strongly influence mesoscale clustering, dispersion, and instabilities, which in turn determine macroscale flow characteristics and material response.
Particle-resolved DNS offers unparalleled detail by resolving all relevant scales, but its computational cost is prohibitive for most engineering-scale applications. As a result, the development of reliable and physically grounded scale-bridging techniques—such as multiscale closure modeling, coarse-grained particle methods, data-driven surrogate models, and hybrid computational frameworks—has become a critical research frontier. This panel will bring together experts in multiphase flow physics, high-fidelity simulation, reduced-order modeling, and machine learning to discuss recent advances and emerging methodologies for bridging micro-, meso-, and macroscales in disperse multiphase flows. The session aims to highlight fundamental challenges, compare promising approaches, and identify opportunities for unifying physics-based and data-driven models to enable next-generation predictive simulations for DOE-relevant applications.