Machine learning errors in simulating viscous fluid behavior explained
Structured Extrema Errors in Classical Surrogates for Viscous Burgers: A Physics-Consistent Interpretation
Machine Learning
Summary
Machine learning models are often used to predict how fluids behave over time. This paper looks at the errors these models make when simulating a specific fluid flow described by the viscous Burgers equation. The authors find that errors near peaks and troughs in the fluid flow are linked to how the models handle the fluid's curvature and smoothing. They show that some models tend to keep too much small wiggle detail that should be smoothed out, and they propose a correction to reduce these mistakes.
What this means in practice
- •For computational fluid dynamics engineers: Improve surrogate models to reduce error when predicting viscous fluid flows in simulation tasks.
Authors
Youssef Oubari
Abstract
We study the local errors of classical machine-learning surrogate models, which approximate the time evolution of the one-dimensional viscous Burgers equation. Four models are compared on the same prediction task, using the spatial grid values directly: radial basis function (RBF) kernel ridge regression (KRR), linear Ridge, ExtraTrees, and Random Forests. Across all four models, the one-step residual, defined here as the true value minus the predicted value at each grid point, forms clear curved branches near predicted maxima and minima. A more detailed analysis of KRR shows that these errors are much more strongly related to the second spatial derivative, which measures local curvature, than to the first spatial derivative. Near a smooth extremum, predicted value and curvature form a local two-branch fold. Under our local curvature-based model of the residual, this fold predicts a leading-order near-parabolic relation between predicted value and residual. This geometric result motivates a direct test of the Burgers advection (transport) and diffusion (smoothing) terms. For KRR and Ridge, regression tests on held-out trajectories, a control that breaks the spatial alignment of the diffusion term, and a spectral test of high-frequency content are consistent with insufficient viscous smoothing at moderate and high viscosity. In this case, the surrogate retains more small-scale structure than the true future state. The same physical explanation is much weaker for the tree models. Finally, a correction that uses only predicted quantities reduces both one-step error and error during recursive rollout, where each prediction is used as the next input.