BlendedNet++ provides a new dataset of 12,492 BWB geometries with RANS-derived Cp and Cf fields and benchmarks geometric deep learning for field prediction plus conditional diffusion models for inverse design achieving R^2 > 0.99 on lift-to-drag targets verified by CFD.
Deep learning in aircraft design, dynamics, and control: Review and prospects.IEEE Transactions on Aerospace and Electronic Systems, 57(4):2346–2368
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
BlendedNet++: A dataset and benchmark for field-resolved aerodynamics and inverse design of blended wing body aircraft
BlendedNet++ provides a new dataset of 12,492 BWB geometries with RANS-derived Cp and Cf fields and benchmarks geometric deep learning for field prediction plus conditional diffusion models for inverse design achieving R^2 > 0.99 on lift-to-drag targets verified by CFD.