Turbulence control for drag reduction through deep reinforcement learning

Lee, Taehyuk; Kim, Junhyuk; Lee, Changhoon

doi:10.1103/physrevfluids.8.024604

Cited by 10 publications

(3 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DRA are usually long‐chain polymers that are added to the fluid and that, through the so‐called visco‐elastic effect , are able to alter the flow characteristics creating a more streamlined flow profile and reducing the energy dissipation caused by turbulence along the walls and the boundaries (L'vov et al., 2004). By reducing the energy required to overcome friction, DRA can enhance the efficiency of fluid transport and potentially lead to cost savings (Lee et al., 2023). The use of DRA has been successful in various industries, including oil and gas, where the transport of crude oil and petroleum products over long distances is crucial.…”

Section: Open Challenges For Future Directionsmentioning

confidence: 99%

Why (Still) Studying Turbulence in Fluids and Plasmas?

Alberti,

Benzi,

Carbone

2023

Perspectives of Earth and Spac

View full text Add to dashboard Cite

Turbulence, a captivating and intricate phenomenon, continues to attract researchers across diverse scientific disciplines. Despite considerable efforts, turbulence remains a fascinating challenge and stands as one of the unsolved enigmas in classical physics. Researchers strive to unravel the underlying physical mechanisms and refine mathematical models to unlock a comprehensive understanding of this complex phenomenon. This paper delves into the reasons why the study of turbulence still persists for a long time, highlighting its history and fundamentals, wide‐ranging applications, significance in environmental and climate sciences, and outstanding open challenges. Through these endeavors, the quest for unraveling the mysteries of turbulence promises to yield profound scientific insights and practical applications in the years to come.

show abstract

Section: Open Challenges For Future Directionsmentioning

confidence: 99%

Why (Still) Studying Turbulence in Fluids and Plasmas?

Alberti,

Benzi,

Carbone

2023

Perspectives of Earth and Spac

View full text Add to dashboard Cite

show abstract

“…(1997), who used a neural network for turbulence control to reduce drag in a turbulent channel flow. Recently, various studies that applied ML to flow control for drag reduction in turbulent channel flow (Han & Huang 2020; Park & Choi 2020; Lee, Kim & Lee 2023), drag reduction of flow around a cylinder (Rabault et al. 2019; Rabault & Kuhnle 2019; Tang et al.…”

Section: Introductionmentioning

confidence: 99%

Prediction and control of two-dimensional decaying turbulence using generative adversarial networks

Kim,

Lee

2024

J. Fluid Mech.

View full text Add to dashboard Cite

An accurate prediction of turbulence has been very costly since it requires an infinitesimally small time step for advancing the governing equations to resolve the fast-evolving small-scale motions. With the recent development of various machine learning (ML) algorithms, the finite-time prediction of turbulence became one of promising options to relieve the computational burden. Yet, a reliable prediction of the small-scale motions is challenging. In this study, PredictionNet, a data-driven ML framework based on generative adversarial networks (GANs), was developed for fast prediction of turbulence with high accuracy down to the smallest scale using a relatively small number of parameters. In particular, we conducted learning of two-dimensional (2-D) decaying turbulence at finite lead times using direct numerical simulation data. The developed prediction model accurately predicted turbulent fields at a finite lead time of up to half the Eulerian integral time scale over which the large-scale motions remain fairly correlated. Scale decomposition was used to interpret the predictability depending on the spatial scale, and the role of latent variables in the discriminator network was investigated. The good performance of the GAN in predicting small-scale turbulence is attributed to the scale-selection and scale-interaction capability of the latent variable. Furthermore, by utilising PredictionNet as a surrogate model, a control model named ControlNet was developed to identify disturbance fields that drive the time evolution of the flow field in the direction that optimises the specified objective function.

show abstract

“…Most studies on reinforcement learning in fluid mechanics are concerned with flow control, with both numerical [34][35][36][37][38][39][40][41][42] and experimental studies [43,44]. The study by Novati et al [24] applies reinforcement learning to improve the subgrid-scale (SGS) turbulence modeling of a large-eddy simulation (LES) by learning from DNS data.…”

Section: Introductionmentioning

confidence: 99%

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

Turbulence control for drag reduction through deep reinforcement learning

Cited by 10 publications

References 37 publications

Why (Still) Studying Turbulence in Fluids and Plasmas?

Why (Still) Studying Turbulence in Fluids and Plasmas?

Prediction and control of two-dimensional decaying turbulence using generative adversarial networks

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

Contact Info

Product

Resources

About