Transfer learning-enhanced deep reinforcement learning for aerodynamic airfoil optimisation subject to structural constraints

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Ramos, David, Lacasa, Lucas, Valero, Eusebio, Rubio, Gonzalo
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913969112875008
author Ramos, David
Lacasa, Lucas
Valero, Eusebio
Rubio, Gonzalo
author_facet Ramos, David
Lacasa, Lucas
Valero, Eusebio
Rubio, Gonzalo
contents The main objective of this paper is to introduce a transfer learning-enhanced deep reinforcement learning (DRL) methodology that is able to optimise the geometry of any airfoil based on concomitant aerodynamic and structural integrity criteria. To showcase the method, we aim to maximise the lift-to-drag ratio $C_L/C_D$ while preserving the structural integrity of the airfoil -- as modelled by its maximum thickness -- and train the DRL agent using a list of different transfer learning (TL) strategies. The performance of the DRL agent is compared with Particle Swarm Optimisation (PSO), a traditional gradient-free optimisation method. Results indicate that DRL agents are able to perform purely aerodynamic and hybrid aerodynamic/structural shape optimisation, that the DRL approach outperforms PSO in terms of computational efficiency and aerodynamic improvement, and that the TL-enhanced DRL agent achieves performance comparable to the DRL one, while further saving substantial computational resources.
format Preprint
id arxiv_https___arxiv_org_abs_2505_02634
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Transfer learning-enhanced deep reinforcement learning for aerodynamic airfoil optimisation subject to structural constraints
Ramos, David
Lacasa, Lucas
Valero, Eusebio
Rubio, Gonzalo
Machine Learning
Computational Physics
The main objective of this paper is to introduce a transfer learning-enhanced deep reinforcement learning (DRL) methodology that is able to optimise the geometry of any airfoil based on concomitant aerodynamic and structural integrity criteria. To showcase the method, we aim to maximise the lift-to-drag ratio $C_L/C_D$ while preserving the structural integrity of the airfoil -- as modelled by its maximum thickness -- and train the DRL agent using a list of different transfer learning (TL) strategies. The performance of the DRL agent is compared with Particle Swarm Optimisation (PSO), a traditional gradient-free optimisation method. Results indicate that DRL agents are able to perform purely aerodynamic and hybrid aerodynamic/structural shape optimisation, that the DRL approach outperforms PSO in terms of computational efficiency and aerodynamic improvement, and that the TL-enhanced DRL agent achieves performance comparable to the DRL one, while further saving substantial computational resources.
title Transfer learning-enhanced deep reinforcement learning for aerodynamic airfoil optimisation subject to structural constraints
topic Machine Learning
Computational Physics
url https://arxiv.org/abs/2505.02634