本文主要是介绍【CFD理论】对流项-03,希望对大家解决编程问题提供一定的参考价值,需要的开发者们随着小编来一起学习吧!
【CFD理论】对流项-03
- first order upwind
- high resolution schemes
- 一维定常对流扩散方程
- example
- TVD schemes
first order upwind
Divergence schemes
一维稳态定常(稳态)对流扩散方程
F e ϕ e − F w ϕ w = D e ( ϕ E − ϕ P ) − D w ( ϕ P − ϕ W ) F_e\phi_e-F_w\phi_w=D_e(\phi_E-\phi_P)-D_w(\phi_P-\phi_W) Feϕe−Fwϕw=De(ϕE−ϕP)−Dw(ϕP−ϕW)
对流项(面心)=梯度(体心)
high resolution schemes
QUICK
- Quadratic Upstream Interpolation for Convective Kinematics
- Second order
- Unbounded
- Leonard
- 在网格规则的情况下是三阶
ϕ f a c e = 6 8 ϕ i − 1 + 3 8 ϕ i − 1 8 ϕ i − 2 \phi_{face}=\frac{6}{8}\phi_{i-1}+\frac{3}{8}\phi_i-\frac{1}{8}\phi_{i-2} ϕface=86ϕi−1+83ϕi−81ϕi−2
ϕ e = 6 8 ϕ P + 3 8 ϕ E − 1 8 ϕ W \phi_e=\frac{6}{8}\phi_{P}+\frac{3}{8}\phi_E-\frac{1}{8}\phi_{W} ϕe=86ϕP+83ϕE−81ϕW
一维定常对流扩散方程
F e ϕ e − F w ϕ w = D e ( ϕ E − ϕ P ) − D w ( ϕ P − ϕ W ) F_e\phi_e-F_w\phi_w=D_e(\phi_E-\phi_P)-D_w(\phi_P-\phi_W) Feϕe−Fwϕw=De(ϕE−ϕP)−Dw(ϕP−ϕW)
F e ( 6 8 ϕ P + 3 8 ϕ E − 1 8 ϕ W ) − F w ( 6 8 ϕ P + 3 8 ϕ W − 1 8 ϕ W W ) = D e ( ϕ E − ϕ P ) − D w ( ϕ P − ϕ W ) F_e(\frac{6}{8}\phi_{P}+\frac{3}{8}\phi_E-\frac{1}{8}\phi_{W})-F_w(\frac{6}{8}\phi_{P}+\frac{3}{8}\phi_W-\frac{1}{8}\phi_{WW})=D_e(\phi_E-\phi_P)-D_w(\phi_P-\phi_W) Fe(86ϕP+83ϕE−81ϕW)−Fw(86ϕP+83ϕW−81ϕWW)=De(ϕE−ϕP)−Dw(ϕP−ϕW)
⇒ \Rightarrow ⇒
[ D w − 3 8 F w + D e + 6 8 F e ] ϕ P = [ D w + 6 8 F w + 1 8 F e ] ϕ W + [ D e − 3 8 F e ] ϕ E − 1 8 F w ϕ W W [D_w-\frac{3}{8}F_w+D_e+\frac{6}{8}F_e]\phi_P=[D_w+\frac{6}{8}F_w+\frac{1}{8}F_e]\phi_W+[D_e-\frac{3}{8}F_e]\phi_E-\frac{1}{8}F_w\phi_{WW} [Dw−83Fw+De+86Fe]ϕP=[Dw+86Fw+81Fe]ϕW+[De−83Fe]ϕE−81FwϕWW
非对称矩阵
- 对流项存在
- 采样点3个
- 不稳定容易发散
example
TVD schemes
- TVD (total variation diminishing),总变差变小
- upwind scheme 无条件稳定,有界。但是会带来 false diffusion
- QUICK等高阶格式,当Pe较大时,可能spurious oscillation or wiggles。当计算一些物理量时候,诱发计算不稳定,TVD格式就是用来处理该问题的。
- 增加人工扩散或者增加上游的权重,基于此的想法,叫做flux corrected transport (FCT)
- OpenFOAM TVD参考: HIGH RESOLUTION NVD DIFFERENCING SCHEME FOR
ARBITRARILY UNSTRUCTURED MESHES
- upwind differencing (UD)
ϕ e = ϕ p \phi_e=\phi_p ϕe=ϕp - linear upwind differencing (LUD)
ϕ e = ϕ P + 1 2 ( ϕ P − ϕ W ) \phi_e=\phi_P+\frac{1}{2}(\phi_P-\phi_W) ϕe=ϕP+21(ϕP−ϕW)
ψ = ϕ P − ϕ W ϕ E − ϕ P \psi=\frac{\phi_P-\phi_W}{\phi_E-\phi_P} ψ=ϕE−ϕPϕP−ϕW - CD格式
ϕ e = ϕ P + 1 2 ( ϕ P − ϕ W ) \phi_e=\phi_P+\frac{1}{2}(\phi_P-\phi_W) ϕe=ϕP+21(ϕP−ϕW) - QUICK格式
ϕ e = ϕ P + 1 8 [ 3 ϕ E − 2 ϕ P − ϕ W ] \phi_e=\phi_P+\frac{1}{8}[3\phi_E-2\phi_P-\phi_W] ϕe=ϕP+81[3ϕE−2ϕP−ϕW]
ψ = ( 3 + ϕ P − ϕ W ϕ E − ϕ P ) 1 4 \psi=(3+\frac{\phi_P-\phi_W}{\phi_E-\phi_P})\frac{1}{4} ψ=(3+ϕE−ϕPϕP−ϕW)41
广义高阶格式
ϕ e = ϕ P + 1 2 ψ ( r ) ( ϕ E − ϕ P ) \phi_e=\phi_P+\frac{1}{2}\psi(r)(\phi_E-\phi_P) ϕe=ϕP+21ψ(r)(ϕE−ϕP)
这篇关于【CFD理论】对流项-03的文章就介绍到这儿,希望我们推荐的文章对编程师们有所帮助!