Aerodynamic and structural airfoil shape optimisation via Transfer Learning-enhanced Deep Reinforcement Learning

2025-05-05

AI Summary

This paper addresses the aerodynamic–structural co-optimization of airfoil geometries, aiming to maximize the lift-to-drag ratio ($C_L/C_D$) while satisfying a maximum-thickness constraint to ensure structural integrity. We propose a transfer learning–enhanced multi-objective deep reinforcement learning (DRL) method built upon the Proximal Policy Optimization (PPO) framework. Crucially, transfer learning is integrated at both policy and feature levels to enable cross-task knowledge reuse. Aerodynamic evaluations are accelerated via a computationally efficient CFD surrogate model. Particle swarm optimization (PSO) serves as the baseline for comparison. Experimental results demonstrate that the proposed method improves optimization efficiency by over 40% relative to PSO and achieves a significantly higher $C_L/C_D$. Moreover, it attains near-optimal performance while reducing computational resource consumption by more than 35%. To the best of our knowledge, this work represents the first successful integration of transfer learning with DRL for multi-objective airfoil geometric optimization.

Technology Category

• Machine Learning: Optimization
• Search and Optimization: Learning to Search
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The main objective of this paper is to introduce a transfer learning-enhanced, multi-objective, deep reinforcement learning (DRL) methodology that is able to optimise the geometry of any airfoil based on concomitant aerodynamic and structural 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 multi-objective shape optimisation, that the DRL approach outperforms PSO in terms of computational efficiency and shape optimisation performance, and that the TL-enhanced DRL agent achieves performance comparable to the DRL one, while further saving substantial computational resources.

Problem

• Optimize airfoil shape for aerodynamic and structural performance
• Compare DRL with PSO in computational efficiency
• Enhance DRL with transfer learning to save resources

Innovation

• Transfer Learning-enhanced Deep Reinforcement Learning
• Multi-objective airfoil shape optimization
• Outperforms Particle Swarm Optimization efficiency