Ensemble Kalman method for learning turbulence models from indirect observation data

• Published in: Journal of fluid mechanics Vol. 949
• Main Authors: Zhang, Xin-Lei, Xiao, Heng, Luo, Xiaodong, He, Guowei
• Format: Journal Article
• Language: English
• Published: Cambridge, UK Cambridge University Press 25.10.2022
• Subjects: Approximation; Deep learning; Design optimization; Eddy viscosity; Equilibrium; Flow control; Flow separation; JFM Papers; Methods; Neural networks; Reynolds averaged Navier-Stokes method; Reynolds stress; Reynolds stresses; Scale models; Simulation; Solvers; Tensors; Training; Turbulence; Turbulence models; Turbulent flow; Velocity; Viscosity; machine learning; turbulence modelling

In this work, we propose using an ensemble Kalman method to learn a nonlinear eddy viscosity model, represented as a tensor basis neural network, from velocity data. Data-driven turbulence models have emerged as a promising alternative to traditional models for providing closure mapping from the mea...