Direct numerical simulation of two-dimensional and non-compressible boundary flow using compressed finite difference method

Abstract

The dimensionless Navier-Stokes equation is solved in a rotational form for the flow of a two-dimensional boundary layer of plates by a direct numerical method. Considering the speed profile at the input of the computational domain, the thickness of the boundary layer has been used as the characteristic length and the uniform velocity of the environment has been used as the characteristic velocity for dimensionlessness. The governing differential equations are broken down using the method of finite compression difference in the main directions of the flow and perpendicular to the flow. A forced mapping has been used to convert the physical domain to the computational domain. In order to develop the calculations in the time domain, the third-order compact Ranj Kota method has been used. The output boundary condition is determined using the transfer model. The simulation results of this type of flow are compared with the resolution of Blasius, which shows the accuracy of the code. In this study, the flow characteristics of the quiet boundary layer to evaluate the accuracy of the code, test and by dividing the lengths and velocities by the thickness of the boundary layer and the uniform velocity of the environment, profiles and contours of velocity and vortex in the device of dimensionless coordinates and self-similarity have seen.

Keywords

• direct numerical simulation
• boundary layer flow
• compact finite differences
• navier-stokes equations
• self-similarity