scholarly journals A supervised neural network for drag prediction of arbitrary 2D shapes in laminar flows at low Reynolds number

2020 ◽  
Vol 210 ◽  
pp. 104645 ◽  
Author(s):  
Jonathan Viquerat ◽  
Elie Hachem
Aerospace ◽  
2022 ◽  
Vol 9 (1) ◽  
pp. 35
Author(s):  
Abu Bakar ◽  
Ke Li ◽  
Haobo Liu ◽  
Ziqi Xu ◽  
Marco Alessandrini ◽  
...  

The airfoil is the prime component of flying vehicles. For low-speed flights, low Reynolds number airfoils are used. The characteristic of low Reynolds number airfoils is a laminar separation bubble and an associated drag rise. This paper presents a framework for the design of a low Reynolds number airfoil. The contributions of the proposed research are twofold. First, a convolutional neural network (CNN) is designed for the aerodynamic coefficient prediction of low Reynolds number airfoils. Data generation is discussed in detail and XFOIL is selected to obtain aerodynamic coefficients. The performance of the CNN is evaluated using different learning rate schedulers and adaptive learning rate optimizers. The trained model can predict the aerodynamic coefficients with high accuracy. Second, the trained model is used with a non-dominated sorting genetic algorithm (NSGA-II) for multi-objective optimization of the low Reynolds number airfoil at a specific angle of attack. A similar optimization is performed using NSGA-II directly calling XFOIL, to obtain the aerodynamic coefficients. The Pareto fronts of both optimizations are compared, and it is concluded that the proposed CNN can replicate the actual Pareto in considerably less time.


Author(s):  
O. Marfaing ◽  
J. Laviéville

Abstract In recent work, we investigated analytically low Reynolds number bubbly flows in pipes. We showed that the distribution of bubbles results from a balance between lift, dispersion and wall forces, and exhibited an analytical expression for this void fraction profile. We then performed a comparison of this analytical Bubble Force Balance Formula (BFBF) with an experiment from the literature. Antal’s model was used for the wall force. The objective of the present work is to compare and assess the three main wall force models in the literature: Antal’s, Tomiyama’s and Frank’s models. We begin by deriving two new BFBF, respectively with Tomiyama’s and Frank’s forces. We can see that the choice of the model impacts the velocity with which the analytical void fraction profile goes to zero at the wall. We then compare our three analytical profiles with experimental measurements and DNS simulations of laminar flows from the literature. We restrict ourselves to the near-wall region. The choice of Antal’s wall force model yields the best agreement. The data is also used to estimate the dispersion coefficient at the wall. Interestingly, we obtain the same order of magnitude with the three wall force models.


2018 ◽  
Vol 12 (3) ◽  
pp. 255
Author(s):  
Muhammad Zal Aminullah Daman Huri ◽  
Shabudin Bin Mat ◽  
Mazuriah Said ◽  
Shuhaimi Mansor ◽  
Md. Nizam Dahalan ◽  
...  

Author(s):  
Vadim V. Lemanov ◽  
Viktor I. Terekhov ◽  
Vladimir V. Terekhov

Sign in / Sign up

Export Citation Format

Share Document