Research on grid‐dependence of neural network turbulence model

Song, Shangxiao; Zhang, Weiwei; Sun, Xuxiang; Zhu, Linyang; Liu, Yilang

doi:10.1002/fld.5125

Cited by 3 publications

(2 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The studies that focused on RANS model augmentation can be further subdivided into three categories: (i) direct estimation of the turbulent eddy viscosity ( 𝑇 ) for linear models [47][48][49][50][51][52][53][54][55], (ii) correction terms for the linear models [56][57][58][59][60][61][62][63], and (iii) enhancement of the accuracy of the turbulence transport equations used in linear models [64][65][66][67][68]. Studies in category (i) have been applied for both incompressible [47][48][49][50] and compressible flows [51][52][53][54]. They have either used neural networks with two to six hidden layers with 20 to 40 neurons per layer or random forest models with three to four hidden layers with 64 to 128 trees.…”

Section: Kaandorp and Dwightmentioning

confidence: 99%

“…Some studies identified additional advantages of the ML approach, e.g., Maulik et al [49] reported that ML  𝑇 estimation sped up simulation time by 100% compared to the two-equation RANS model. Song et al [54] reported that ML turbulence model had more lenient near-wall y + requirements compared physics-based models, and thus required smaller grid sizes and lower computational costs.…”

Section: Kaandorp and Dwightmentioning

confidence: 99%

See 1 more Smart Citation

Development and Validation of a Machine Learned Turbulence Model

Bhushan

Burgreen

Brewer³

et al. 2021

Energies

View full text Add to dashboard Cite

A stand-alone machine learned turbulence model is developed and applied for the solution of steady and unsteady boundary layer equations, and issues and constraints associated with the model are investigated. The results demonstrate that an accurately trained machine learned model can provide grid convergent, smooth solutions, work in extrapolation mode, and converge to a correct solution from ill-posed flow conditions. The accuracy of the machine learned response surface depends on the choice of flow variables, and training approach to minimize the overlap in the datasets. For the former, grouping flow variables into a problem relevant parameter for input features is desirable. For the latter, incorporation of physics-based constraints during training is helpful. Data clustering is also identified to be a useful tool as it avoids skewness of the model towards a dominant flow feature.

show abstract

Section: Kaandorp and Dwightmentioning

confidence: 99%

Section: Kaandorp and Dwightmentioning

confidence: 99%

Development and Validation of a Machine Learned Turbulence Model

Bhushan

Burgreen

Brewer³

et al. 2021

Energies

View full text Add to dashboard Cite

show abstract

Assessment of neural network augmented Reynolds averaged Navier Stokes turbulence model in extrapolation modes

et al. 2023

View full text Add to dashboard Cite

This study proposes and validates a novel machine-learned (ML) augmented linear Reynolds averaged Navier Stokes (RANS) model, and the applicability of model assessed in both interpolation and extrapolation modes for periodic hill (Hill) test case, which involves complex flow regimes, such as attached boundary layer, shear-layer, and separation and reattachment. For this purpose, the ML model is trained using direct numerical simulation (DNS)/LES datasets for nine different cases with different flow separation and attachment regimes, and by including various percentages of the Hill DNS dataset during the training, ranging from no data (extrapolation mode) to all data (interpolation mode). The predictive capability of the ML model is then assessed using a priori and a posteriori tests. Tests reveal that the ML model's predictability improves significantly as the Hill dataset is partially added during training, e.g., with the addition of only 5% of the hill data increases correlation with DNS to 80%. Such models also provide better turbulent kinetic energy (TKE) and shear stress predictions than RANS in a posteriori tests. Overall, the ML model for TKE production is identified to be a reliable approach to enhance the predictive capability of RANS models. The study also performs (1) parametric investigation to evaluate the effect of training and neural network hyperparameters, and data scaling and clustering on the ML model accuracy to provide best practice guidelines for ML training; (2) feature importance analysis using SHapley Additive exPlanations (SHAP) function to evaluate the potential of such analysis in understanding turbulent flow physics; and (3) a priori tests to provide guidelines to determine the applicability of the ML model for a case for which reference DNS/LES datasets are not available.

show abstract

Artificial neural network-substituted transition model for crossflow instability: Modeling strategy and application prospect

Wu,

Cui,

Wang

et al. 2024

Physics of Fluids

View full text Add to dashboard Cite

Data-driven approaches have made preliminary inroads into the area of transition–turbulence modeling, but are still in their infancy with regard to widespread industrial adoption. This paper establishes an artificial neural network (ANN)-based transition model to enhance the capacity of capturing the crossflow (CF) transition phenomena, which are frequently identified over a wide range of aerodynamic problems. By taking a new CF-extended shear stress transport (SST) transition-predictive (SST-γ) model as the baseline, a mapping from mean flow variables to transition intermittency factor (γ) is constructed by ANN algorithm at various Mach and Reynolds numbers of an infinite swept wing. Generalizability of the resulting ANN-based (SST-γANN) model is fully validated in the same infinite swept wing, an inclined 6:1 prolate spheroid, and a finite swept wing in extensive experiment regimes, together with two effective a priori analysis strategies. Furthermore, the calculation efficiency, grid dependence, and performance of the present model in non-typical transitional flow are also assessed to inspect its industrial feasibility, followed by the elucidation of rationality behind the preliminary success and transferability of present framework. The results manifest that the SST-γANN model aligns well with the benchmark SST-γ model, and both can capture the CF transition accurately compared with their experiment counterpart, completely breaking through the disability of original SST-γ model without CF correction. In addition, good properties of efficiency, robustness, and generalizability are achieved for the ANN-alternative transition model, together with the usability of present framework across various transitional flows.

show abstract

Research on grid‐dependence of neural network turbulence model

Cited by 3 publications

References 43 publications

Development and Validation of a Machine Learned Turbulence Model

Development and Validation of a Machine Learned Turbulence Model

Assessment of neural network augmented Reynolds averaged Navier Stokes turbulence model in extrapolation modes

Artificial neural network-substituted transition model for crossflow instability: Modeling strategy and application prospect

Contact Info

Product

Resources

About