Journal
INTERNATIONAL JOURNAL OF HEAT AND FLUID FLOW
Volume 90, Issue -, Pages -Publisher
ELSEVIER SCIENCE INC
DOI: 10.1016/j.ijheatfluidflow.2021.108822
Keywords
RANS; Machine learning; Turbulence modeling
Categories
Funding
- Newton Fund [ST/R006733/1]
- UK Engineering and Physical Sciences Research Council (EPSRC) through the Computational Science Centre for Research Communities (CoSeC)
- UK Turbulence Consortium [EP/R029326/1]
- EPSRC [EP/N016602/1, EP/P022243/1, EP/N033841/1]
Ask authors/readers for more resources
Machine-learning techniques offer a new perspective for constructing turbulence models for RANS simulations. An iterative ML-RANS computational framework is proposed, ensuring built-in reproducibility. The discussion around closure term and suitable target variables, as well as the study of multi-valued problem for establishing a proper regression system, contribute to the effectiveness of the ML model.
Machine-learning (ML) techniques provide a new and encouraging perspective for constructing turbulence models for Reynolds-averaged Navier-Stokes (RANS) simulations. In this study, an iterative ML-RANS computational framework is proposed that combines an ML algorithm with transport equations of a conventional turbulence model. This framework maintains a consistent procedure for obtaining the input features of an ML model in both the training and predicting stages, ensuring a built-in reproducibility. The effective form of the closure term is discussed to determine suitable target variables for the ML algorithm, and the multi-valued problem of existing constitutive theory is studied to establish a proper regression system for ML algorithms. The developed ML model is trained under a cross-case strategy with data from turbulent channel flows at three Reynolds numbers, and a posteriori simulations of channel flows show that the framework is able to predict both the mean flow field and turbulent variables accurately. Interpolation tests for the channel flow show the proposed framework can reliably predict flow features that lie between the minimum and maximum Reynolds numbers associated with the training data. A further test of the ML model in a flow over periodic hills also demonstrates improved performance compared with the k-omega SST model, even though the model is only trained with planar channel flow data.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available