Unsteady aerodynamic prediction using limited samples based on transfer learning

TL;DR

This paper introduces a transfer learning approach with LSTM networks to predict unsteady aerodynamic forces under varying initial conditions, significantly reducing the required sample size while improving accuracy and efficiency.

Contribution

It proposes a transfer learning-based method that leverages pre-trained LSTM models to predict aerodynamic forces with limited new samples, avoiding extensive high-fidelity simulations.

Findings

Predicts aerodynamic forces with only 1/3 of the original sample size.

Improves prediction accuracy compared to direct LSTM modeling.

Demonstrates effectiveness on high-spinning projectile flight data.

Abstract

In this study, a method for predicting unsteady aerodynamic forces under different initial conditions using a limited number of samples based on transfer learning is proposed, aiming to avoid the need for large-scale high-fidelity aerodynamic simulations. First, a large number of training samples are acquired through high-fidelity simulation under the initial condition for the baseline, followed by the establishment of a pre-trained network as the source model using a long short-term memory (LSTM) network. When unsteady aerodynamic forces are predicted under the new initial conditions, a limited number of training samples are collected by high-fidelity simulations. Then, the parameters of the source model are transferred to the new prediction model, which is further fine-tuned and trained with limited samples. The new prediction model can be used to predict the unsteady aerodynamic forces of the entire process under the new initial conditions. The proposed method is validated by predicting the aerodynamic forces of free flight of a high-spinning projectile with a large extension of initial angular velocity and pitch angle. The results indicatethat the proposed method can predict unsteady aerodynamic forces under different initial conditions using 1/3 of the sample size of the source model. Compared with direct modeling using the LSTM networks, the proposed method shows improved accuracy and efficiency.