10 May 2020 in International Journal for Numerical Methods in Fluids
1 August 2022 in Physics of Fluids
Ice accretion on wind turbine blades and wings changes the effective shape of the airfoil and considerably deteriorates the aerodynamic performance. However, the unsteady performance of iced airfoil is often difficult to predict. In this study, the unsteady aerodynamic performance of iced airfoil is simulated under different pitching amplitudes and reduced frequencies. In order to efficiently predict aerodynamic performance under icing conditions, a multi-fidelity reduced-order model based on multi-task learning is proposed. The model is implemented using lift and moment coefficient of clean airfoil as low-fidelity data. Through using few aerodynamic data from iced airfoils as high-fidelity data, the model can achieve aerodynamic prediction for different ice shapes and pitching motions. The results indicate that, compared with single-fidelity and single-task modeling, the proposed model can achieve better accuracy and generalization capability. At the same time, the model can be generalized to different ice shapes, which can effectively improve the unsteady prediction efficiency.
22 November 2023 in Chinese Journal of Aeronautics
1 July 2022 in AIAA Journal
Unsteady aerodynamic prediction is an important part of modern aircraft safety and control law design. To improve the accuracy and efficiency for unsteady aerodynamic prediction of aircraft at high...
19 October 2024 in Aerospace Science and Technology
1 December 2021 in Physics of Fluids
Dynamic stall requires both accurate and efficient predictions. To model the unsteady aerodynamics of dynamic stall, a symbiosis method for dynamic stall prediction is proposed through fusing experimental data and numerical simulations based on computational fluid dynamics. With only a fraction of wind tunnel test data of the National Advisory Committee for Aeronautics 0012 airfoil, the proposed framework is able to predict the lift and moment coefficients of dynamic stall under different balanced angles of attacks, amplitudes, and reduced frequencies. Results indicate that compared with the Unsteady Reynolds-Averaged Navier–Stokes simulation, the proposed model reduces the prediction error about two to five times. In addition, a posteriori analysis shows that with efficient hyperparameter optimization, the framework can separate the dynamics for attached and separated flows adaptively. The proposed data fusion model provides a way to combine the physics of the dynamic stall phenomenon to prediction models for the aerodynamic loading at high angles of attack.