Fixed Wing UAV Winglet Optimization Through AI-Integrated Aerodynamic Simulations With Multi-Physics Approach

Abstract This study focuses on optimizing winglet design parameters for fixed-wing UAVs for enhancing the efficiency of a fixed-wing UAV by exploring the efficient aerodynamic design of winglets along with mass and structural trade off study for the whole wing. Winglets are the devices that are implemented to reduce the induced drag and improve the lift-to-drag ratio therefore increasing the aerodynamic performance of the UAV. The CFD parametric analysis on a NACA-4412 airfoil wing with winglets of varied heights, taper ratios and cant angles were performed using the Altair® FlightStream™, an aerodynamics tool that merges panel methods with modern computational techniques. It was found that the implementation of winglets can increase the L/D ratios by 14% as compared to the regular wing. The structural analysis on the wing geometry with winglets was conducted using a FEM based structural solver Altair® Optistruct and a design of experiments study was performed using Altair® Design Explorer to obtain optimal structure-mass tradeoff. The simulation data obtained in DoE is trained with the help of Altair® PhysicsAI™, a geometric deep learning AI driven software for quick results.