Using machine learning to detect the turbulent region in flow past a circular cylinder

Li, Binglin; Yang, Zixuan; Zhang, Xing; He, Guowei; Deng, Bing; Shen, Lian

doi:10.1017/jfm.2020.725

Cited by 49 publications

(14 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…With the recent developments in machine learning (ML) methods and their successful application to classical engineering problems, various advances have been made to accelerate numerical methods (Kachrimanis, Karamyan, & Malamataris 2003;Ariana, Vaferi, & Karimi 2015;Benvenuti, Kloss, & Pirker 2016;Chaurasia & Nikkam 2017;Liang et al 2018a,b;Figueiredo et al 2019;Brevis, Muga, & van der Zee 2020;Prieto 2020). This capacity has also been extended to problems related to fluid dynamics and granular flow (Radl & Sundaresan 2014;Kutz 2017;Wan & Sapsis 2018;Fukami, Fukagata & Taira 2019;Li et al 2020a;Park & Choi 2020;Aghaei Jouybari et al 2021) where its applications towards the former has been extensively reviewed (Brenner, Eldredge, & Freund 2019;Brunton, Noack, & Koumoutsakos 2020;Fukami, Fukagata, & Taira 2020a). For example, a ML approach was used for the estimation of gravitational solid flows (Garbaa et al 2014).…”

Section: Introductionmentioning

confidence: 99%

High-resolution fluid–particle interactions: a machine learning approach

Davydzenka

Tahmasebi

2022

J. Fluid Mech.

View full text Add to dashboard Cite

Modelling of fluid–particle interactions is a major area of research in many fields of science and engineering. There are several techniques that allow modelling of such interactions, among which the coupling of computational fluid dynamics (CFD) and the discrete element method (DEM) is one of the most convenient solutions due to the balance between accuracy and computational costs. However, the accuracy of this method is largely dependent upon mesh size, where obtaining realistic results always comes with the necessity of using a small mesh and thereby increasing computational intensity. To compensate for the inaccuracies of using a large mesh in such modelling, and still take advantage of rapid computations, we extended the classical modelling by combining it with a machine learning model. We have conducted seven simulations where the first one is a numerical model with a fine mesh (i.e. ground truth) with a very high computational time and accuracy, the next three models are constructed on coarse meshes with considerably less accuracy and computational burden and the last three models are assisted by machine learning, where we can obtain large improvements in terms of observing fine-scale features yet based on a coarse mesh. The results of this study show that there is a great opportunity in machine learning towards improving classical fluid–particle modelling approaches by producing highly accurate models for large-scale systems in a reasonable time.

show abstract

Section: Introductionmentioning

confidence: 99%

High-resolution fluid–particle interactions: a machine learning approach

Davydzenka

Tahmasebi

2022

J. Fluid Mech.

View full text Add to dashboard Cite

show abstract

“…Recently, studies by Colvert et al [15] and Alsalman et al [16] showed that ANNs and clustering techniques could be used to successfully identify types of turbulent structures based on limited data in airfoil wakes. Li et al [17] used a machine learning technique known as extreme gradient boosting to identify the region of a wake where turbulence is present and showed that these techniques were more generally applicable than ad-hoc determinations. Given such success, this raises the question as to how capably machine learning techniques can distinguish between turbulence produced by different sources, such as wakes, jets, overturning convection, shear, etc.…”

Section: Introductionmentioning

confidence: 99%

Identifying the Origin of Turbulence Using Convolutional Neural Networks

Brown

Zimny

Radko

2022

Fluids

View full text Add to dashboard Cite

Though turbulence is often thought to have universal behavior regardless of origin, it may be possible to distinguish between the types of turbulence generated by different sources. Prior work in turbulence modeling has shown that the fundamental “constants” of turbulence models are often problem-dependent and need to be calibrated to the desired application. This has resulted in the introduction of machine learning techniques to attempt to apply the general body of turbulence simulations to the modeling of turbulence at the subgrid-scale. This suggests that the inverse is likely also possible: that machine learning can use the properties of turbulence at small scales to identify the nature of the original source and potentially distinguish between different classes of turbulence-generating systems, which is a novel pursuit. We perform numerical simulations of three forms of turbulence—convection, wake, and jet—and then train a convolutional neural network to distinguish between these cases using only a narrow field of view of the velocity field. We find that the network is capable of identifying the correct case with 86% accuracy. This work has implications for distinguishing artificial sources of turbulence from natural ones and aiding in identifying the mechanism of turbulence in nature, permitting more accurate mixing models.

show abstract

“…In the fluid prediction field, data-driven methods have been rapidly developed due to the generation of large amounts fluid data and the advancements in computer technology. In the early stage of research on flow field predictions, studies focused on predicting the airfoil and flow field around a cylinder and the application of different flow fields was studied later [26][27][28][29][30][31]. The author used a convolutional neural network (CNN) to predict the flow field around a cylinder and analyzed the difference between the CFD and predicted values from the vortex or velocity vector and vortex length perspectives [32].…”

Section: Introductionmentioning

confidence: 99%

Prediction of Cavitation Performance over the Pump-Jet Propulsor Using Computational Fluid Dynamics and Hybrid Deep Learning Method

Qiu

Huang

Pan

2022

JMSE

View full text Add to dashboard Cite

The cavitation performance of an oblique flow field is different from that under a pure axial flow field. This study analyzed the hydrodynamic performance, bearing force, and tip clearance flow field under different rotating speeds and different cavitation numbers in an oblique flow field. Furthermore, this study proposed a hybrid deep learning model CNN-Bi-LSTM to quickly and accurately predict the bearing force of a pump-jet propulsor (PJP), which will solve the problem of time-consuming calculation and consumption of considerable computing resources in traditional computational fluid dynamics. The Shear–Stress–Transport model and Reynolds-averaged Navier–Stokes equations were utilized to procure the training and testing datasets. The training and testing datasets were reasonably divided in the ratio of 7:3. The results show that the propulsion efficiency decreased more obviously under higher rotating speed conditions, with a maximum decrease of up to 13.59%. The small cavitation numbers 1.4721 and high oblique angle significantly impacted the efficiency reduction; the maximum efficiency loss exceeded 20%. Thus, a small cavitation number 1.4721 is extremely detrimental to the propulsion efficiency of the PJP due to the large cavitation area. Moreover, the intensity of the tip clearance vortex continuously increased with the rotating speed. The CNN-Bi-LSTM deep model successfully predicted the phase difference and trend change of the propulsor bearing force under different conditions. The prediction difference was large at the crest and trough of the bearing force, but it is within the acceptable error range.

show abstract

Using machine learning to detect the turbulent region in flow past a circular cylinder

Abstract: Abstract

Cited by 49 publications

References 45 publications

High-resolution fluid–particle interactions: a machine learning approach

High-resolution fluid–particle interactions: a machine learning approach

Identifying the Origin of Turbulence Using Convolutional Neural Networks

Prediction of Cavitation Performance over the Pump-Jet Propulsor Using Computational Fluid Dynamics and Hybrid Deep Learning Method

Contact Info

Product

Resources

About