An automatic isotropic/anisotropic hybrid grid generation technique for viscous flow simulations based on an artificial neural network  

作　　者：Peng LU Nianhua WANG Xinghua CHANG Laiping ZHANG Yadong WU 

机构地区：[1]School of Information Engineering,Southwest University of Science and Technology,Mianyang 621010,China [2]School of Intelligent Manufacturing Engineering,Chongqing University of Arts and Science,Chongqing 402160,China [3]Stake Key Laboratory of Aerodynamics,China Aerodynamics Research and Development Center,Mianyang 621000,China [4]Unmanned Systems Research Center,National Innovation Institute of Defense Technology,Beijing 100071,China [5]School of Computer Science and Technology,Sichuan University of Science&Engineering,Yibin 644005,China

出　　处：《Chinese Journal of Aeronautics》2022年第4期102-117,共16页中国航空学报（英文版）

基　　金：supported by the National Key Re-search and Development Program of China(No.2016YFB0200701);the National Natural Science Foundation of China(Nos.11532016 and 11672324);the National Key Project(No.GJXM92579)。

摘　　要：Based on the author’s previous research, a novel hybrid grid generation technique is developed by introducing an Artificial Neural Network(ANN) approach for realistic viscous flow simulations. An initial hybrid grid over a typical geometry with anisotropic quadrilaterals in the boundary layer and isotropic triangles in the off-body region is generated by the classical mesh generation method to train two ANNs on how to predict the advancing direction of the new point and to control the grid size. After inputting the initial discretized fronts, the ANN-based Advancing Layer Method(ALM) is adopted to generate the anisotropic quadrilaterals in boundary layers. When the high aspect ratio of the anisotropic grid reaches a specified value, the ANN-based Advancing Front Method(AFM) is adopted to generate isotropic triangles in the off-body computational domain.The initial isotropic triangles are smoothed to further improve the grid quality. Three typical cases are tested and compared with experimental data to validate the effectiveness of grids generated by the ANN-based hybrid grid generation method. The experimental results show that the two ANNs can predict the advancing direction and the grid size very well, and improve the adaptability of the isotropic/anisotropic hybrid grid generation for viscous flow simulations.

关 键 词：Advancing front method Advancing layer method Anisotropic quadrilateral grid generation Artificial neural network Isotropic triangular grid generation Machine learning 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] O35[自动化与计算机技术—控制科学与工程]