Learning Stable Galerkin Models of Turbulence with Differentiable Programming

TL;DR

This paper introduces a differentiable programming method that combines neural networks with Galerkin projection to create stable, interpretable reduced-order models for turbulent flow, improving prediction accuracy and stability.

Contribution

It presents Neural Galerkin projection, a novel approach embedding neural networks into Galerkin ODEs, enhancing stability, interpretability, and prediction horizon in turbulence modeling.

Findings

Neural Galerkin learns stable ODE coefficients from POD data.

It achieves longer, more accurate temporal predictions than classical methods.

The approach offers low computational costs and improved interpretability.

Abstract

Turbulent flow control has numerous applications and building reduced-order models (ROMs) of the flow and the associated feedback control laws is extremely challenging. Despite the complexity of building data-driven ROMs for turbulence, the superior representational capacity of deep neural networks has demonstrated considerable success in learning ROMs. Nevertheless, these strategies are typically devoid of physical foundations and often lack interpretability. Conversely, the Proper Orthogonal Decomposition (POD) based Galerkin projection (GP) approach for ROM has been popular in many problems owing to its theoretically consistent and explainable physical foundations. However, a key limitation is that the ordinary differential equations (ODEs) arising from GP ROMs are highly susceptible to instabilities due to truncation of POD modes and lead to deterioration in temporal predictions. In this work, we propose a \textit{differentiable programming} approach that blends the strengths of both these strategies, by embedding neural networks explicitly into the GP ODE structure, termed Neural Galerkin projection. We demonstrate this approach on the isentropic Navier-Stokes equations for compressible flow over a cavity at a moderate Mach number. When provided the structure of the projected equations, we show that the Neural Galerkin approach implicitly learns stable ODE coefficients from POD coefficients and demonstrates significantly longer and accurate time horizon predictions, when compared to the classical POD-GP assisted by calibration. We observe that the key benefits of this differentiable programming-based approach include increased flexibility in physics-based learning, very low computational costs, and a significant increase in interpretability, when compared to purely data-driven neural networks.