Sloshing reduced-order model based on neural networks for aeroelastic analyses

A thorough understanding of the effects sloshing on aircraft dynamic loads is great relevance for future design flexible to be able reduce their structural mass and environmental impact. Indeed, high vertical accelerations caused by vibrations structure can lead fragmentation fuel free surface. Fluid impacts tank ceiling are potentially a new source damping that has hardly been considered before when computing wings. This work aims at applying recently developed reduced-order models case research wing investigate aeroelastic response under pre-critical post-critical conditions. The dynamics using neural networks trained with experimental data from scaled then integrated into system following suitable scaling procedure. results concern gust input (flutter) conditions as well highlighting onset limit cycle oscillations sloshing, only nonlinear phenomenon modelled in present simulation framework. Moreover, load alleviation performances will assessed typical landing input.