Aerodynamic Modeling and Simulation of a Reduced-Scale Generic Future Fighter Using Neuro-Fuzzy with Differential Evolution

Abstract

The pursuit of accurate aerodynamic modeling in aeronautical engineering has driven the exploration of advanced computational techniques. This research applies Neuro-Fuzzy Hybridized with Differential Evolution (NF-DE) to develop a high-fidelity aerodynamic model and simulation for the Generic Future Fighter (GFF), a reduced-scale aircraft. The primary objective is to create a system identification methodology through flight testing for detailed and unsteady aerodynamic models. The methodology combines the interpretability of Fuzzy Inference Systems (FIS) with the adaptability of Artificial Neural Networks (ANN). The study includes a comparison between NF-DE and alternative optimization and fuzzy inference methods. After developing unsteady aerodynamic models based on first Taylor series equations for each force and moment coefficient, a 6-degrees-of-freedom (DOF) simulation is designed in Matlab’s Simulink environment.