Aerodynamic Parameter Identification Method Based on Physics-Informed Radial Basis Function-Deep Neural Networks

doi:10.1016/j.isatra.2025.08.039

Aerodynamic Parameter Identification Method Based on Physics-Informed Radial Basis Function-Deep Neural Networks

2025-08-22

0

PRE

AI

DOI：10.1016/j.isatra.2025.08.039

Original

Original request for help

Save

Abstract

En 中文

• Identify the strong nonlinear perturbations between real and nominal aerodynamic parameters. • The physics-informed loss function is designed to train network without real perturbation data. • Radial basis function layer is embedded in deep neural network to improve network identification accuracy and reduce time cost.

Keywords:

nonlinear perturbations

physics-informed loss function

radial basis function

deep neural network

aerodynamic parameters

Aerodynamic Parameter Identification Method Based on Physics-Informed Radial Basis Function-Deep Neural Networks

Abstract

Journal

Organization