An iterative data-driven turbulence modeling framework based on Reynolds stress representation. Issue 5 (September 2022)

• Record Type: Journal Article
• Authors: Yin, Yuhui; Shen, Zhi; Zhang, Yufei; Chen, Haixin; Fu, Song
• Is Part Of: Theoretical & applied mechanics letters. Volume 12:Issue 5(2022)
• Publication Date: 2022-09
• Subjects: Turbulence modeling -- Reynolds-averaged Navier-Stokes equations -- Reynolds stress representation -- Machine learning; Mechanics, Applied -- Periodicals; Mechanics, Analytic -- Periodicals; Mechanics, Analytic; Mechanics, Applied; Periodicals; 620.1
• DOI: 10.1016/j.taml.2022.100381
• Languages: English
• ISSNs: 2095-0349

Abstract

Highlights: Extended the relevant tensor arguments of Reynolds stress by performing the theoretical derivation. Proposed an adaptive regularization term to enhance the Reynolds stress representation performance. Constructed an iterative coupling framework of the ML model and CFD solver with consistent convergence. Abstract: Data-driven turbulence modeling studies have reached such a stage that the basic framework is settled, but several essential issues remain that strongly affect the performance. Two problems are studied in the current research: (1) the processing of the Reynolds stress tensor and (2) the coupling method between the machine learning model and flow solver. For the Reynolds stress processing issue, we perform the theoretical derivation to extend the relevant tensor arguments of Reynolds stress. Then, the tensor representation theorem is employed to give the complete irreducible invariants and integrity basis. An adaptive regularization term is employed to enhance the representation performance. For the coupling issue, an iterative coupling framework with consistent convergence is proposed and then applied to a canonical separated flow. The results have high consistency with the direct numerical simulation true values, which proves the validity of the current approach.