Deep reinforcement learning-based active control for drag reduction of three equilateral-triangular circular cylinders

Chen, Ning; Zhang, Ruigang; Liu, Quansheng; Ding, Zhaodong

doi:10.1016/j.euromechflu.2023.12.001

Cited by 2 publications

(1 citation statement)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nowadays, there have been abundant achievements concerning flows around the cylinders from multiple aspects, such as the different Reynolds numbers [9,14], conditions of incoming flow [15], section shapes [5,6,14] and geometric arrangements [16][17][18]. Comprehensive research achievements can refer to many specialized topic reviews [19][20][21][22][23][24][25][26].…”

Section: Introductionmentioning

confidence: 99%

Wake Structures and Hydrodynamic Characteristics of Flows around Two Near-Wall Cylinders in Tandem and Parallel Arrangements

Chang,

Yin,

Xin

et al. 2024

JMSE

View full text Add to dashboard Cite

To clarify the hydrodynamic interference characteristics of flows around multiple cylinders under the wall effect, the two-dimensional (2D) flows around the near-wall single, two tandem and parallel cylinders are simulated under different gap ratios (0.15 ≤ G/D ≤ 3.0) and spacing ratios (1.5 ≤ T/D ≤ 4.0) at a Reynolds number of Re = 6300. We also examine the wake patterns, the force coefficients, and the vortex-shedding frequency with emphases on the wall effect and effects of the two-cylinder interference. A critical wall gap of G/D = 0.6 is identified in the single-cylinder case where the wall can exert significant influences. The two near-wall tandem cylinders exhibit three wake states: stretching mode, attachment mode, and impinging mode. The force coefficients on the upstream cylinder are significantly affected by the wall for G/D ≤ 0.6. The downstream cylinder is mainly influenced by the upstream cylinder. For G/D > 0.6, the force coefficients on the two cylinders exhibit a similar variation trend. In the parallel arrangement, the two cylinders exhibit four wake states in different G/D and T/D ranges: double stretching mode, hetero-vortex scale mode, unilateral vortex mode, and free vortex mode. Moreover, the two parallel cylinders in the hetero-vortex scale or free vortex mode have two states: synchronous in-phase state and synchronous out-of-phase state. The mean drag coefficients on the two cylinders decrease, while the mean lift coefficients exhibit opposite variation trends, as the T/D grows.

show abstract

Section: Introductionmentioning

confidence: 99%

Wake Structures and Hydrodynamic Characteristics of Flows around Two Near-Wall Cylinders in Tandem and Parallel Arrangements

Chang,

Yin,

Xin

et al. 2024

JMSE

View full text Add to dashboard Cite

show abstract

An interactive platform of deep reinforcement learning and wind tunnel testing

Dong,

Wang,

Lin

et al. 2024

Physics of Fluids

View full text Add to dashboard Cite

Flow around bluff bodies is a classic problem in fluid mechanics, and flow control is a critical approach for manipulating the aerodynamic characteristics of bluff bodies. Recently, deep reinforcement learning (DRL) has emerged as a highly potential method of flow control. However, the application of DRL to wind tunnel testing involves significant obstacles, which can be classified into software, hardware, and interaction challenges. These challenges make the application of DRL-based wind tunnel testing particularly complex and challenging for many researchers. To address these challenges, this paper proposes a novel DRL-based wind tunnel testing platform, named DRLinWT. DRLinWT introduces a universal adapter capable of managing interactive communications across multiple mainstream communication protocols and integrates commonly used reinforcement learning libraries, thereby significantly reducing the interaction cost between DRL algorithms and wind tunnel tests. Using this platform, a DRL-based flow control experiment for a square cylinder in three flow fields of varying complexity was conducted.

show abstract

Deep reinforcement learning-based active control for drag reduction of three equilateral-triangular circular cylinders

Cited by 2 publications

References 36 publications

Wake Structures and Hydrodynamic Characteristics of Flows around Two Near-Wall Cylinders in Tandem and Parallel Arrangements

Wake Structures and Hydrodynamic Characteristics of Flows around Two Near-Wall Cylinders in Tandem and Parallel Arrangements

An interactive platform of deep reinforcement learning and wind tunnel testing

Contact Info

Product

Resources

About