Multi-Objective Optimization of Aircraft Wing Beam Cross-Sections Using Reinforcement Learning

The aircraft wing beam serves as the primary load-bearing structure of an aircraft, and the quality of its cross-section design significantly impacts the aircraft’s performance, safety, and economic efficiency. Traditional design methods for wing beams often rely on experience and trial-and-error processes, which are not only time-consuming and labor-intensive but also result in low design efficiency and difficulty in achieving optimal results. Furthermore, conventional optimization algorithms lack transferability, requiring re-optimization for each design iteration. In contrast, reinforcement learning enhances the design process through trial-and-error learning and feedback mechanisms, improving both efficiency and quality. This study proposes a reinforcement learning-based multi-objective optimization design method for aircraft wing beam structures. The method facilitates multi-objective optimization, enables the storage of design knowledge, and allows for the future expansion of the design space.