ISpliter:an intelligent and automatic surface mesh generator using neural networks and splitting lines

In this paper,we present a novel surface mesh generation approach that splits B-rep geometry models into isotropic triangular meshes based on neural networks and splitting lines.In the first stage,a recursive method is designed to generate plentiful data to train the neural network model offline.In the second stage,the implemented mesh generator,ISpliter,maps each surface patch into the parameter plane,and then the trained neural network model is applied to select the optimal splitting line to divide the patch into subdomains continuously until they are all triangles.In the third stage,ISpliter remaps the 2D mesh back to the physical space and further optimizes it.Several typical cases are evaluated to compare the mesh quality generated by ISpliter and two baselines,Gmsh and NNW-GridStar.The results show that ISpliter can generate isotropic triangular meshes with high average quality,and the generated meshes are comparable to those generated by the other two software under the same configuration.

FlowDNN:a physics-informed deep neural network for fast and accurate flow prediction

For flow-related design optimization problems,e.g.,aircraft and automobile aerodynamic design,computational fluid dynamics(CFD)simulations are commonly used to predict flow fields and analyze performance.While important,CFD simulations are a resource-demanding and time-consuming iterative process.The expensive simulation overhead limits the opportunities for large design space exploration and prevents interactive design.In this paper,we propose Flow DNN,a novel deep neural network(DNN)to efficiently learn flow representations from CFD results.Flow DNN saves computational time by directly predicting the expected flow fields based on given flow conditions and geometry shapes.Flow DNN is the first DNN that incorporates the underlying physical conservation laws of fluid dynamics with a carefully designed attention mechanism for steady flow prediction.This approach not only improves the prediction accuracy,but also preserves the physical consistency of the predicted flow fields,which is essential for CFD.Various metrics are derived to evaluate Flow DNN with respect to the whole flow fields or regions of interest(RoIs)(e.g.,boundary layers where flow quantities change rapidly).Experiments show that Flow DNN significantly outperforms alternative methods with faster inference and more accurate results.It speeds up a graphics processing unit(GPU)accelerated CFD solver by more than 14000×,while keeping the prediction error under 5%.