2026-11-10 九州大学
九州大学などの研究グループは、航空機の翼やジェットエンジンの圧縮機羽根などについて、形状と流体力学的性能を高精度に予測するAI「PI-GAE（Physics-Informed Geometry-Embedded Graph Autoencoder）」を開発した。複雑な三次元形状を少ない情報に圧縮し、その情報から元の形状を再構成すると同時に、空気抵抗や圧縮性能などを予測できる。2次元・3次元の翼や圧縮機羽根を含む4種類の対象で性能を検証し、翼の空気抵抗や羽根の圧縮性能を高い精度で推定した。形状の再構成では、元の寸法に対する平均誤差が0.3％未満となった例も示された。また、学習データを従来の10分の1に減らした条件でも、性能だけを予測する手法より安定した予測を実現した。多数の設計候補を効率よく比較できるため、航空機や産業機械の設計期間短縮、燃料・電力消費の削減への貢献が期待される。ただし、学習時の形状に近い範囲での利用が前提であり、大きく異なる形状や条件には追加検証が必要である。
参考図：羽根の形と性能（圧縮比）の予測の例
ここに示す3種類に限らず、100種類の新たな設計候補について、PI-GAEは形と性能を高い精度で予測できます。
＜関連情報＞
- https://www.kyushu-u.ac.jp/ja/researches/view/1621
- https://www.sciencedirect.com/science/article/pii/S0045782526008054
形状縮小、再構築、および代理モデル化のための、物理法則に基づいた幾何学埋め込み型グラフオートエンコーダー A physics-informed geometry-embedded graph autoencoder for shape reduction, reconstruction, and surrogate modeling
Thanasak Wanglomklang,Frédéric Gillot,Sébastien Besset,Koji Shimoyama
Computer Methods in Applied Mechanics and Engineering Available online: 28 October 2026
DOI:https://doi.org/10.1016/j.cma.2026.119534
Highlights
- A GNN-based latent code jointly reconstructs geometry and predicts response.
- Controlled manifold studies reveal when nonlinear encoding outperforms POD.
- Aerodynamic benchmarks confirm accurate reconstruction and response prediction.
Abstract
High-dimensional shapes need reduced representations that are compact, reconstructable, mesh-aware, and predictive of physical performance. This is challenging because low-dimensional shape families may form nonlinear manifolds in nodal-coordinate space, so linear bases can require more modes than the number of independent deformation coordinates. This paper presents a physics-informed geometry-embedded graph autoencoder (PI-GAE). It encodes mesh or profile deformations into a shared latent vector, reconstructs coordinates through a singular-value-decomposition-assisted decoder, and predicts scalar or aligned field responses with condition-aware heads. Controlled translation-manifold studies compare the nonlinear latent requirement with proper orthogonal decomposition (POD) at matched reconstruction accuracy. The main transition follows the prescribed intrinsic dimension, although finite model capacity limits accuracy as intrinsic dimension or translation range grows. The method is evaluated on AirfRANS, OptiWing3D, SuperWing, and Rotor37, spanning two-dimensional airfoils, three-dimensional wings, and a turbomachinery blade surface with structured and neighborhood graphs. Held-out scalar prediction reaches coefficient-of-determination values from 0.946 to 0.992, and AirfRANS also demonstrates node-wise pressure-coefficient prediction. A parameter-matched comparison shows graph-based encoding matches simpler encoders on most cases but yields the clearest advantage on the highest-resolution SuperWing geometry. Mean normalized reconstruction error remains below 0.3% for AirfRANS, OptiWing3D, and Rotor37; the upper-deformation SuperWing quartile reaches 1.84%. The shared latent representation provides a compact, mesh-aware coordinate for both geometry reconstruction and physical-response prediction within a single model.
