Analysis of multiple dynamic simulation methods applied in the study of flow field in subway tunnels

Abstract

The computational fluid dynamics method has been widely used in the field of subway tunnel research. The simulation of the train's motion is always implemented through dynamic mesh models. However, the computational cost of dynamic mesh methods is high, and the methods adopted by researchers are different. Comprehensive simulation performance analysis of various dynamic mesh models in tunnel flow field applications has not been thoroughly conducted. In this study, based on four different dynamic mesh models (remeshing, smoothing and remeshing (S & R), layering and overset mesh) provided by ANSYS Fluent, comparing them in reduced-size, full-size and real train models. The mesh generation complexity, computational efficiency, application feasibility and accuracy in tunnel flow simulations were analysed. The results show the layering method has the highest calculation accuracy and simulation efficiency, the error was only 4.9% and the simulation time was only 25% of the other three methods. The study has provided new ideas for the simulation of complex models and has established a key reference basis for selecting the optimal dynamic mesh models. In addition, the study has significantly improved the simulation accuracy of tunnel flow mechanics and provides practical technical specifications for the application of metro tunnel engineering applications.