![]() Formally, airfoil inverse design is described as the prediction of airfoil shapes based on given desired performance metrics 4, making our problem an inverse design problem. Specifically, the problem at hand involves generating airfoils that achieve a desired performance. Relying on fluid mechanics, aerodynamic related problems exhibit non-linearity and are complex in nature 1, 3. Here, we focus on the optimization of airfoil shapes. An optimization process leading to an increase in the aerodynamic efficiency of aircraft components is needed. As such, its reduction could represent a 20 to 25% decrease in fuel burn 2 and lead to fewer emissions. For aircraft, aerodynamic drag represents the main source of energy losses 1. Similar content being viewed by othersĪs demand for air travel continues to grow, so are concerns regarding the environmental impacts of aviation. Overall, the presented approach demonstrates the relevance of DRL to airfoil shape optimization and brings forward a successful application of DRL to a physics-based aerodynamics problem. The strong resemblance between the artificially produced shapes and those found in the literature highlights the rationality of the decision-making policy learned by the agent. Results show that the DRL agent is able to generate high performing airfoils within a limited number of learning iterations. ![]() The learning abilities of the DRL agent are demonstrated through various experiments in which the agent’s objective-maximizing L/ D, maximizing C l or minimizing C d-as well as the initial airfoil shape are varied. A custom RL environment is developed allowing the agent to successively modify the shape of an initially provided 2D airfoil and to observe the associated changes in aerodynamic metrics such as lift-to-drag ( L/ D), lift coefficient ( C l) and drag coefficient ( C d). We formulate the airfoil design as a Markov decision process (MDP) and investigate a Deep Reinforcement Learning (DRL) approach to airfoil shape optimization. Reinforcement learning (RL) provides a data-driven approach bearing generative capabilities. Supervised learning approaches have addressed these limitations but are constrained by user-provided data. Current approaches relying on gradient-based or gradient-free optimizers are data-inefficient in that they do not leverage accumulated knowledge, and are computationally expensive when integrating Computational Fluid Dynamics (CFD) simulation tools. However, the inherent complexity and non-linearity associated with fluid mechanics as well as the high-dimensional design space intrinsic to such problems make airfoil shape optimization a challenging task. Shape optimization is an indispensable step in any aerodynamic design.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |