Affiliation:
1. MCI - The Entrepreneurial School
Abstract
Abstract
In this study, fluid flow predictions using three different methods were compared: DeepCFD, an artificial intelligence code; computational fluid dynamics (CFD) with Ansys Fluent and openFoam; and two-dimensional, two-component particle image velocimetry (PIV) measurements. The airfoils under investigation were the NACA 0012 with a 10° angle of attack and the NACA 6412 with a 0° angle of attack. To train DeepCFD, 763, 2585, and 6283 openFoam simulations based on primitives were utilized. The investigation was conducted at a free stream velocity of 10 m/s and a Reynolds number of 82000. Results show that once the DeepCFD network is trained, prediction times are negligible, enabling real-time optimization of airfoils. The mean absolute error between CFD and DeepCFD, with 6283 trained primitives, for NACA 0012 predictions resulted in velocity components Ux = 1.77 m/s, Uy = 0.73 m/s, and static pressure p = 8.97 Pa. For NACA 6412, the corresponding MSE are Ux = 0.81 m/s, Uy = 0.59 m/s, and p = 7.5 Pa. Qualitative agreement was observed between PIV measurements, DeepCFD, and CFD. Results are promising that artificial intelligence has the potential for real-time fluid flow optimization of NACA airfoils in the future.
Publisher
Research Square Platform LLC
Reference17 articles.
1. Goodfellow, I., Bengio, Y., Courville, A.: Deep learning. MIT press (2016)
2. Deep learning to replace, improve, or aid CFD analysis in built environment applications: A review;Calzolari G;Build. Environ.,2021
3. Recent advances in convolutional neural networks;Gu J;Pattern Recognit.,2018
4. Applications of computational fluid dynamics (CFD) in the food industry: a review;Xia B;Comput. Electron. Agric.,2002
5. Kundu, P.K., Cohen, I.M., Dowling, D.R.: Fluid mechanics. Academic press (2015)