Learning Interface Breakup: A Geometry-Conditioned Latent Surrogate for Spray Formation

2026-06-15

AI Summary

This work addresses the challenge of efficiently predicting how nozzle geometry influences transient gas–liquid interface breakup, a task hindered by the prohibitive computational cost of high-fidelity Volume-of-Fluid (VOF) simulations and the inability of conventional surrogate models to handle adaptive mesh refinement (AMR) and dynamic interface evolution. To overcome this, the authors propose a geometry-conditioned latent-space surrogate model that, for the first time, treats the AMR grid density field as a learnable compact representation. Their two-stage architecture accurately reconstructs both transient density evolution and full flow fields. Trained on 797 two-phase nozzle simulations, the model achieves key interfacial dynamics fidelity while requiring only 0.045 seconds per inference—over 60,000× faster than Basilisk CFD—and substantially reduces reliance on extensive multi-channel flow field data.

Abstract

Designing spray nozzles requires predicting how geometry shapes transient two-phase breakup, but high-fidelity volume-of-fluid (VOF) simulations with adaptive mesh refinement (AMR) are too expensive for iterative design exploration. Standard surrogate models are also challenged by this setting because both the liquid--gas interface and the underlying adaptive discretization evolve across time and geometries. We introduce a geometry-conditioned latent surrogate trained on 797 two-phase nozzle simulations that addresses this by encoding the AMR cell-density field, rather than the full multi-channel flow state, as a compact proxy for where the solver concentrates resolution. From this representation, the model reconstructs transient density evolution and nozzle geometry, and a lightweight second stage recovers the remaining flow variables. On held-out simulations, the method accurately captures key interface dynamics while reducing inference time to 0.045 seconds per trajectory, corresponding to a speed-up of more than $6\times10^4$ relative to Basilisk CFD. These results suggest that AMR refinement structure can serve as a compact and learnable representation for geometry-conditioned surrogate modeling of transient two-phase flows.

Problem

• interface breakup
• two-phase flow
• adaptive mesh refinement
• surrogate modeling
• spray formation

Innovation

• geometry-conditioned surrogate
• adaptive mesh refinement (AMR)
• two-phase flow
• interface breakup
• latent representation