文献类型：期刊文献

英文题名：Study on Flow and Heat Characteristics of Compressible Gas in a Supersonic Nozzle Based on PINNs with Sparse Data

作者：Shen, Yida Dong, Bin Ma, Quan Dang, Chao Li, Congjian Ren, Guojian Wang, Shaozhan Sun, Xiaozhe Ding, Yong

第一作者：Shen, Yida

通信作者：Dang, C[1];Li, CJ[2]

机构：[1]Beijing Jiaotong Univ, Sch Mech Elect & Control Engn, Beijing Key Lab Flow & Heat Transfer Phase Changin, Beijing 100044, Peoples R China;[2]China Aerodynam Res & Dev Ctr, High Speed Aerodynam Inst, Mianyang 621000, Peoples R China;[3]Beijing Jiaotong Univ, Sch Math & Stat, Beijing 100044, Peoples R China;[4]Guizhou Inst Technol, Sch Aerosp Engn, Guiyang 550025, Peoples R China

第一机构：Beijing Jiaotong Univ, Sch Mech Elect & Control Engn, Beijing Key Lab Flow & Heat Transfer Phase Changin, Beijing 100044, Peoples R China

通信机构：corresponding author), Beijing Jiaotong Univ, Sch Mech Elect & Control Engn, Beijing Key Lab Flow & Heat Transfer Phase Changin, Beijing 100044, Peoples R China;corresponding author), China Aerodynam Res & Dev Ctr, High Speed Aerodynam Inst, Mianyang 621000, Peoples R China.

年份：2026

卷号：24

期号：2

外文期刊名：FRONTIERS IN HEAT AND MASS TRANSFER

收录：EI(收录号：20261820637300);WOS:【ESCI(收录号:WOS:001764638100001)】；

基金：Funding Statement: This research was supported by the Guizhou Provincial Major Scientific and Technological Program (XKBF (2025) 031) . This research was supported by the Fundamental Research Funds for the Central Universities (No. 2024JBMC016) .

语种：英文

外文关键词：Physics-informed neural networks; sparse data; supersonic flow; specific physical knowledge; shortcut connections

摘要：This article explores the application of Physics-Informed Neural Networks (PINNs) in solving supersonic flow problems within a Laval nozzle, proposing innovative methods by integrating physical constraints and neural network optimization techniques. The main innovations of this study include the construction of a novel neural network architecture with shortcut connections to enhance the prediction of overall flow trends and local fluctuations, thereby improving convergence speed, reducing computational costs, and increasing the accuracy of flow field reconstruction. Additionally, this study designs a PINNs framework that incorporates specific physical knowledge (SPK) to improve model stability, generalization, and accuracy, even with sparse training data. A dynamic loss weighting strategy is employed to optimize training convergence, and velocity components are reformulated as magnitude and angle to simplify boundary conditions and reduce the dimensionality of the solution space. The results demonstrate that the proposed methods achieve satisfactory accuracy and robustness in solving supersonic problems, highlighting their potential application value.