Open YinLiu-91 opened 11 months ago
For wall boundaries, we need to impose a no-normal-flux condition on the momentum variable, whereas we use a Neumann condition for the density and energy with and . To achieve the no-normal flux condition, we set the exterior values to the interior values and subtract two times the velocity in wall-normal direction, i.e., in the direction of the normal vector.对于壁面边界,我们需要对动量变量施加非正规通量条件,而对于密度和能量,我们使用Neumann条件。为了达到法向通量为0的条件,我们将外部值设置为内部值,并在壁面法线方向(即法向量方向)减去两倍的速度。
这里是在处理密度与能量时认为不变,而速度认为切向与内部相等,而法向为0
这里认为边界的左右值都是给定的Dirichlet边界值,所以要给定如上图的w+值
理解就是:边界处的左右值应该连续的,并且是确定的,这样能够改定一个确定的Dirichlet边界
如 #5 中例子中写的step-67
刘黎论文添加dg边界的本质
出流和入流边界的添加
无反射边界条件的添加
PML边界的添加
固壁边界添加
具体推导过程如下(以欧拉方程而非线化欧拉方程为例): 通量为:
$$ \begin{matrix} \rho u_1 + \rho u_2 \ \rho u_1^2 + p + \rho u_1 u_2 \ \rho u_2 u_1 + \rho u_2^2 +p \ u_1(E+p) + u_2(E+p)
\end{matrix} \tag{1} $$
可以整理为
$$ \begin{matrix} \rho (u_1 + u_2 ) \ \rho u_1(u_1 + u_2) +p \ \rho u_2( u_1 + u_2) +p \ (u_1 + u_2)(E+p)
\end{matrix} \tag{2} $$
通量乘以法向量 $F\cdot n$ 为
$$ \begin{matrix} (\rho (u_1 + u_2 ))\cdot n \ (\rho u_1(u_1 + u_2) +p) \cdot n \ (\rho u_2( u_1 + u_2) +p) \cdot n \ ((u_1 + u_2)(E+p)) \cdot n \end{matrix} \tag{2} $$
而边界条件为 $F\cdot n$中 $(u_1+u_2)\cdot n=0$ , 将此条件代入上式中有:
$$ \begin{matrix} (\rho (u_1 + u_2 ))\cdot n =0 \ (\rho u_1(u_1 + u_2) +p) \cdot n = p\cdot n \ (\rho u_2( u_1 + u_2) +p) \cdot n =p\cdot n\ ((u_1 + u_2)(E+p)) \cdot n = 0 \end{matrix} \tag{2} $$