EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Aerodynamics-constrained generative design of exterior form for a high-speed maglev experimental vehicle

Peng Ji, Chaoyi Zhu, Xu Zhang, Jie Zhang, Bosen Qian, Jingming Ma, Danhua Zhao

Transportation Safety and Environment · 10.10.2026 · 8 (3)

DOI: 10.1093/tse/tdag036 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Transportation Safety and Environment, online erschienen am 08.07.2026. Verfasst von Peng Ji und 6 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hunan Provincial Natural Science Foundation, Hunan Provincial Research Project on Degree and Postgraduate Teaching Reform.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Under high-speed operating conditions, the form–performance matching between aerodynamic performance and formal aesthetics becomes a core challenge in the design of high-speed maglev experimental vehicles, while conventional experience-driven approaches struggle to achieve coordinated optimization. Taking a single-unit high-speed maglev experimental vehicle as the research object, this paper proposes a generative intelligent exterior-form design method integrating aerodynamics with Stable Diffusion. Parametric modelling of two-dimensional primary profiles and numerical simulations of three-dimensional topological forms are first conducted to extract the mapping relationships between morphological features and aerodynamic performance under representative operating conditions, thereby determining optimal parameter combinations in both two- and three-dimensional domains. The correlation mechanisms between profile geometry and the perceptual semantics of lightweight, elongation and sharpness are then clarified, and a quantitative aesthetic evaluation framework is established. Based on the optimal aerodynamic contour, a performance-constrained generative design workflow is constructed through ControlNet-guided conditional generation and customized Low-Rank Adaptation (LoRA) model training, enabling the coupled generation of styling features, aerodynamic performance and formal aesthetics, followed by multi-scheme exploration and secondary numerical evaluation. Engineering validation indicates that the proposed scheme achieves a drag coefficient of 0.042, with pitching-moment fluctuations controlled within ±175 N, and full-scale tests at 300 km/h on the China Railway Rolling Stock Corporation Limited Zhuzhou test line confirm that all key performance indicators meet the required standards. By integrating simulation-based analysis, generative learning and aesthetic evaluation, this study establishes a form–performance collaborative design framework, providing a technically feasible and theoretically grounded pathway for the innovative design of next-generation high-speed maglev experimental vehicles.

Förderung

  • Hunan Provincial Natural Science Foundation
  • Hunan Provincial Research Project on Degree and Postgraduate Teaching Reform

Zitieren

APA

Ji, P., Zhu, C., Zhang, X., Zhang, J., Qian, B., Ma, J., & Zhao, D. (2026). Aerodynamics-constrained generative design of exterior form for a high-speed maglev experimental vehicle. Transportation Safety and Environment, 8(3). https://doi.org/10.1093/tse/tdag036

Vancouver

Ji P, Zhu C, Zhang X, Zhang J, Qian B, Ma J, et al. Aerodynamics-constrained generative design of exterior form for a high-speed maglev experimental vehicle. Transportation Safety and Environment. 2026;8(3). doi:10.1093/tse/tdag036

BibTeX
@article{Ji2026,
  author = {Ji, Peng and Zhu, Chaoyi and Zhang, Xu and Zhang, Jie and Qian, Bosen and Ma, Jingming and Zhao, Danhua},
  title = {Aerodynamics-constrained generative design of exterior form for a high-speed maglev experimental vehicle},
  journal = {Transportation Safety and Environment},
  year = {2026},
  volume = {8},
  number = {3},
  doi = {10.1093/tse/tdag036}
}
Kostenlos starten →