← back

project

in progress

Flow-map learning for turbulent fluids

Neural surrogate models for Nek5000-generated CFD trajectories. Curriculum rollout with Sobolev loss and noise injection for long-horizon stability.

April 2026

  • ai
  • cfd
  • research
  • pytorch

Overview

A neural flow-map for incompressible turbulent flows, trained on Nek5000 simulations. The architecture learns a one-step push-forward operator and is composed at inference time to produce long-horizon predictions.

Method

The training objective combines a primary L2 step loss with a Sobolev penalty on velocity gradients and a noise-injection term that improves stability under autoregressive rollout. We adopt a curriculum that grows the rollout horizon as accuracy improves.

Status

In progress as of May 2026. Master thesis at the MIT van Rees Lab.