Formulating turbulence closures using sparse regression with embedded form invariance

Abstract

A data-driven framework for formulation of closures of the Reynolds-Average Navier--Stokes (RANS) equations is presented. In recent years, the scientific community has turned to machine learning techniques to distill a wealth of highly resolved data into improved RANS closures. While the body of work in this area has primarily leveraged Neural Networks (NNs), we alternately leverage a sparse regression framework. This methodology has two important properties: (1) The resultant model is in a closed, algebraic form, allowing for direct physical inferences to be drawn and naive integration into existing computational fluid dynamics solvers, and (2) Galilean invariance can be guaranteed by thoughtful tailoring of the feature space. Our approach is demonstrated for two classes of flows: homogeneous free shear turbulence and turbulent flow over a wavy wall. This work demonstrates equivalent performance to that of modern NNs but with the added benefits of interpretability, increased ease-of-use and dissemination, and robustness to sparse training datasets.

References in corpus (5)

• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Turbulence Modeling in the Age of Data
• RANS Turbulence Model Development using CFD-Driven Machine Learning
• Discovery of Algebraic Reynolds-Stress Models Using Sparse Symbolic Regression
• RANS Equations with Explicit Data-Driven Reynolds Stress Closure Can Be Ill-Conditioned

Cited by in corpus (24)

• Enhancing Computational Fluid Dynamics with Machine Learning
• Ensemble-SINDy: Robust sparse model discovery in the low-data, high-noise limit, with active learning and control
• Perspectives on Machine Learning-augmented Reynolds-averaged and Large Eddy Simulation Models of Turbulence
• On closures for reduced order models $-$ A spectrum of first-principle to machine-learned avenues
• Physics-guided deep learning framework for predictive modeling of the Reynolds stress anisotropy
• Sparse identification of nonlinear dynamics with low-dimensionalized flow representations
• Physics-constrained, low-dimensional models for MHD: First-principles and data-driven approaches
• Experimental velocity data estimation for imperfect particle images using machine learning
• Nonlinear stochastic modeling with Langevin regression
• Sparse nonlinear models of chaotic electroconvection
• CFD-driven Symbolic Identification of Algebraic Reynolds-Stress Models
• PySINDy: A Python package for the Sparse Identification of Nonlinear Dynamics from Data
• Discovering explicit Reynolds-averaged turbulence closures for turbulent separated flows through deep learning-based symbolic regression with non-linear corrections
• Machine-learning accelerated turbulence modelling of transient flashing jets
• Space-dependent Aggregation of Stochastic Data-driven Turbulence Models
• Assessment of Neural Network Augmented Reynolds Averaged Navier Stokes Turbulence Model in Extrapolation Modes
• Neural stochastic differential equations for particle dispersion in large-eddy simulations of homogeneous isotropic turbulence
• Promoting global stability in data-driven models of quadratic nonlinear dynamics
• A Parallel Integrated Computational-Statistical Platform for Turbulent Transport Phenomena
• Artificial neural network approach for turbulence models: A local framework
• Dynamics of tidal spiral arms: Machine learning-assisted identification of equations and application to the Milky Way
• Sparse identification of multiphase turbulence closures for coupled fluid--particle flows
• Modeling compressed turbulent plasma with rapid viscosity variations
• Turbulence closure modeling with data-driven techniques: Investigation of generalizable deep neural networks