An accuracy-enhanced transonic flow prediction method fusing deep learning and reduced-order model

TL;DR

This paper introduces a novel method combining deep learning and reduced-order modeling to accurately and efficiently predict transonic flow fields with shock waves over aircraft surfaces.

Contribution

It develops a fused CNN-POD approach for improved flow prediction accuracy, especially in nonlinear shock regions, outperforming existing methods.

Findings

Prediction error reduced by up to 46% in shock regions

Enhanced robustness and efficiency demonstrated

Applicable to various aerodynamic shapes

Abstract

It's difficult to accurately predict the flow with shock waves over an aircraft due to the flow's strongly nonlinear characteristics. In this study, we propose an accuracy-enhanced flow prediction method that fuses deep learning and reduced-order model to achieve fast flow field prediction for various aerodynamic shapes. First, we establish the convolutional neural network-proper orthogonal decomposition (CNN-POD) model for mapping geometries to the entire flow field. Next, local flow regions containing nonlinear flow structures are identified through POD reconstruction for enhanced modeling. Then, a new CNN model is employed to map geometries to the local flow field. The proposed method is finally applied in predicting transonic flow over airfoils. The results indicate that the proposed enhanced DNN method can reduce the prediction error of flow properties, particularly in the regions with shock waves (up to 13%-46%). Additionally, the better efficiency and robustness of the proposed methods have been validated in comparison to existing methods.