03.1 - Aerodynamics – CFD Methods and Validation

MACHINE LEARNING-DRIVEN AERODYNAMIC SHAPE OPTIMIZATION OF NACA AIRFOILS UNDER GEOMETRIC DESIGN CONSTRAINTS

M.S. Abulela¹, M.I. Alam¹; ¹King Fahd University of Petroleum and Minerals, Saudi Arabia

This automated framework couples Kriging surrogates with high-fidelity CFD for NACA airfoil optimization. Uniquely incorporating practical wing-box constraints, it achieves a lift-to-drag ratio of 60.21 with <1% deviation from CFD. Reducing manual preprocessing by 98%, this tool enables rapid, structure-aware aerodynamic design.

View Paper

DOI: 10.71945/icas2026_0530