Title:
Generative Modeling for Flap Kinematics: Enhancing Design Pipelines with Variational Autoencoders
Authors:
A. Schreiber, R.Y. Yanev, I. Staack, U. Scholz
Abstract:
The preliminary design of high-lift configurations for fixed-wing aircraft is a complex, multi-disciplinary task usually relying on extensive Design of Experiment (DoE) studies. This work presents a Variational Autoencoder (VAE)-based pipeline for generating Adaptive Dropped-Hinge Flap (ADHF) concepts that satisfy engineering constraints on cruise drag, separation behavior, and actuation efficiency. Trained on a curated set of desirable designs, the VAE produces novel configurations with characteristics similar to proven solutions, enabling efficient exploration of the design space. A post-processing framework evaluates these candidates and relates them to the original user input. The ADHF-VAE pipeline delivers a broader and higher-quality range of designs in significantly fewer iterations than standard DoE, providing improved starting points for downstream aerodynamic, kinematic, and integration analyses. This framework represents a first step towards automated, data-driven synthesis of high-lift systems for future aircraft design.
Event:
Deutscher Luft- und Raumfahrtkongress 2025, Augsburg
Location, Year:
Deutsche Gesellschaft für Luft- und Raumfahrt - Lilienthal-Oberth e.V., Bonn, 2025
Format:
21,0 x 29,7 cm, 12 Pages
Descriptors:
EN: data-driven design, kinematics, generative modeling, design space exploration, variational autoencoder
Citation:
Schreiber, A.; Yanev, R.Y.; et al. (2025): Generative Modeling for Flap Kinematics: Enhancing Design Pipelines with Variational Autoencoders. DGLR Published Conference Papers. Volume 4. (2026). ISSN: 2941-3451. Deutsche Gesellschaft für Luft- und Raumfahrt - Lilienthal-Oberth e.V.. (Text). https://doi.org/10.25967/650353. urn:nbn:de:101:1-2510241301071.164089603828.
Publishing Date:
24.10.2025