A PDE-free, neural network-based eddy viscosity model coupled with RANS equations

Xu, Ran; Zhou, Xu-Hui; Han, Jiequn; Dwight, Richard P.; Xiao, Heng

doi:10.1016/j.ijheatfluidflow.2022.109051

Cited by 4 publications

(3 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The method introduced in this work shares similarities with other publications, which are outlined below. In the present work, we use the local surrounding of a point as input data, comparable to the methods of Xu et al [20] and Zhou et al [21]. We combine this approach with the concept of multi-agent reinforcement learning that was also used by Novati et al [24] and Yousif et al [46].…”

Section: Deep Reinforcement Learning Algorithmmentioning

confidence: 99%

“…Methods that apply machine learning to closure modeling can be categorized into two groups: algorithms that learn from existing turbulence models to mimic their behaviour, and algorithms that learn from high-fidelity data to improve the accuracy of RANS closure models. The former is addressed, for example, by Xu et al [20], who apply a dual neural network architecture to replace the k-ϵ closure model. The inputs of the neural network are composed of local and surrounding information, a concept known as the vector cloud neural network (VCNN) [21].…”

Section: Introductionmentioning

confidence: 99%

“…The objective of the DRL algorithm (reward function) is to match the augmented Spalart-Allmaras RANS results with high-fidelity time-averaged LES velocity fields. Inspired by the studies of Novati et al [24] and Xu et al [20], a multi-agent approach is used, which augments the turbulence model locally and, thus, allows a model trained on one configuration to be applied to other configurations.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Deep Reinforcement Learning-Augmented Spalart–Allmaras Turbulence Model: Application to a Turbulent Round Jet Flow

Fuchs,

von Saldern,

Kaiser

et al. 2024

Fluids

View full text Add to dashboard Cite

The purpose of this work is to explore the potential of deep reinforcement learning (DRL) as a black-box optimizer for turbulence model identification. For this, we consider a Reynolds-averaged Navier–Stokes (RANS) closure model of a round turbulent jet flow at a Reynolds number of 10,000. For this purpose, we augment the widely utilized Spalart–Allmaras turbulence model by introducing a source term that is identified by DRL. The algorithm is trained to maximize the alignment of the augmented RANS model velocity fields and time-averaged large eddy simulation (LES) reference data. It is shown that the alignment between the reference data and the results of the RANS simulation is improved by 48% using the Spalart–Allmaras model augmented with DRL compared to the standard model. The velocity field, jet spreading rate, and axial velocity decay exhibit substantially improved agreement with both the LES reference and literature data. In addition, we applied the trained model to a jet flow with a Reynolds number of 15,000, which improved the mean field alignment by 35%, demonstrating that the framework is applicable to unseen data of the same configuration at a higher Reynolds number. Overall, this work demonstrates that DRL is a promising method for RANS closure model identification. Hurdles and challenges associated with the presented methodology, such as high numerical cost, numerical stability, and sensitivity of hyperparameters are discussed in the study.

show abstract