Unsteady Aerodynamic Prediction of a Variable-Sweep Morphing Wing via Steady–Unsteady Data Fusion

Abstract Variable-sweep wing aircraft demonstrate pronounced unsteady aerodynamic characteristics throughout the wing morphing process. Accurate prediction of unsteady aerodynamics remains a critical challenge due to complex flow physics and high computational cost. Traditional computational fluid dynamics (CFD) methods are precise but time-consuming, and purely data-driven surrogates often lack robustness when applied to unseen flow conditions. To address these limitations, this study proposes a steady–unsteady data fusion framework for rapid and reliable prediction of unsteady pressure distributions and aerodynamic coefficients of a variable-sweep wing undergoing cosine-type sweep motions. The framework integrates a steady prediction module with an unsteady correction module, leveraging the complementary strengths of the data sets. Steady-flow data provide broad parametric coverage over Mach numbers, sweep angles, and angles of attack, whereas unsteady data offer high temporal resolution and capture dynamic response features. The fusion model generates a base pressure field from the steady module and learns unsteady corrections from time-resolved simulations, achieving both accuracy and generalizability. Performance is evaluated across three distinct validation scenarios: same-period validation, cross-period validation, and cross-Mach/angle-of-attack validation. Results showed that the fusion model reduces pressure field errors by more than 50% and decreases lift and drag coefficient errors by nearly an order of magnitude compared with a purely unsteady surrogate. This study demonstrates that combining steady and unsteady data yields a computationally efficient and robust surrogate model, offering a practical tool for rapid and accurate unsteady aerodynamic prediction of morphing wings.