Deep Learning-Based Surrogate Model for Flight Load Analysis

Flight load computations (FLC) are generally expensive and time-consuming. This paper studies deep learning (DL)-based surrogate models of FLC to provide a reliable basis for the strength design aircraft structures. We mainly analyze influence Mach number, overload, angle attack, elevator deflection, altitude, other factors on loads key monitoring components, based which input output variables set. The data used train validate DL derived using flight simulation results wind tunnel test data. According features, neural network (DNN) random forest (RF) proposed establish models. DNN meets accuracy requirement rich sources in FLC; RF can alleviate overfitting evaluate importance parameters. Numerical experiments show that both DNN-and RF-based achieve high accuracy. analysis demonstrates vertical overload deflection have significant FLC. believe synthetic applications these DL-based methods great promise field