Application of convolutional neural networks to the fast simulation of fluid dynamic systems
Proyecto
Trabajo de Fin de Máster
Docente
Ekaitz Zulueta Guerrero
Centro
University of Basque Country (UPV/EHU)
Empresa
Siemens Gamesa
Curso
2022/2023
Image
![Escuela de Ingeniería de Bilbao, Universidad del País Vasco – Euskal Herriko Unibertsitatea](/sites/default/files/styles/large/public/media/upv-ingenieria.png?itok=DBwjy6h-)
Descripción
This project consists of two well differentiated parts. In the first one, a Convolutional Neural Network (CNN) is employed to predict the temporal evolution of the velocity and pressure fields around a circle geometry. Secondly, Deep Learning (DL) techniques are used to predict and measure the most suitable form and size of a Gurney flap (GF) for a specific case.