航空学报 > 2025, Vol. 46 Issue (11): 531282-531282   doi: 10.7527/S1000-6893.2024.31282

eVTOL适坠性分析及优化

丁梦龙1, 李道春1,2(), 周尧明1,2, 冯传宴1, 邵浩原2, 向锦武1,2   

  1. 1.天目山实验室 绿色民机智能设计研究中心,杭州 311115
    2.北京航空航天大学 航空科学与工程学院,北京 100191
  • 收稿日期:2024-09-30 修回日期:2024-10-29 接受日期:2024-11-18 出版日期:2024-11-26 发布日期:2024-11-25
  • 通讯作者: 李道春 E-mail:lidc@buaa.edu.cn
  • 基金资助:
    国家自然科学基金(T2288101);国家自然科学基金(52272382);国家自然科学基金(52402507);国家重点研发计划(2020YFC1512500);天目山实验室关键领域项目(TK-2024-D-010)

Crashworthiness analysis and optimization for eVTOL vehicles

Menglong DING1, Daochun LI1,2(), Yaoming ZHOU1,2, Chuanyan FENG1, Haoyuan SHAO2, Jinwu XIANG1,2   

  1. 1.Research Center for Intelligent Design of Green Civil Aircraft,Tianmushan Laboratory,Hangzhou 311115
    2.School of Aeronautic Science and Engineering,Beihang University,Beijing 100191,China
  • Received:2024-09-30 Revised:2024-10-29 Accepted:2024-11-18 Online:2024-11-26 Published:2024-11-25
  • Contact: Daochun LI E-mail:lidc@buaa.edu.cn
  • Supported by:
    National Natural Science Foundation of China(T2288101);National Key Research and Development Project(2020YFC1512500);Reward Funds for Research Project of TianmuMountain Laboratony(TK-2024-D-010)

摘要:

电动垂直起降(eVTOL)飞行器可以在城市内提供便捷的点对点飞行,有望革命性地改变城市居民的出行方式。然而与常规飞机相比,它在耐撞性设计方面存在诸多新挑战。对此,对滑橇式起落架和吸能元件进行了适坠性优化设计,并将其应用于整机适坠性分析。同时,开展了多角度、多速度的离轴坠撞仿真研究。鉴于坠撞仿真耗时多、不利于优化设计的问题,提出了使用机器学习技术预测滑橇式起落架应力、吸能元件吸能效率的方法,并且开发了与遗传算法相结合的优化方法。本研究提高了滑橇式起落架、吸能元件、以及eVTOL整机的适坠性能,初步掌握了各离轴坠撞参数对乘员安全的威胁程度,并能在误差可接受范围内实时预测滑橇式起落架应力和吸能元件的吸能效率。

关键词: eVTOL, 能量吸收, 适坠性, 机器学习, 离轴坠撞

Abstract:

Electric Vertical Take-Off and Landing (eVTOL) vehicles could transform urban transportation by enabling convenient point-to-point flights. However, their crashworthiness design presents unique challenges compared to conventional aircraft. This study optimized the crashworthiness of the skid landing gear and energy-absorbing components, followed by comprehensive analysis, including multi-angle, multi-speed off-axis crash simulations. Considering the time-consuming nature of crash simulations, which limits optimization, we introduce a machine learning method to predict stress on the skid landing gear and the energy absorption efficiency of the energy absorber. An optimization method incorporating genetic algorithms is developed. The results show a significant enhancement in the crashworthiness of the skid landing gear, energy absorber, and the whole eVTOL. Moreover, the preliminary threat level of various off-axis crash parameters to passenger safety is identified and real-time prediction tools for stress and energy absorption efficiency are introduced, maintaining accuracy within acceptable margins.

Key words: eVTOL, energy absorption, crashworthiness, machine learning, off-axis crash

中图分类号: