Beruflich Dokumente
Kultur Dokumente
Basic course
Legal notes:
- This offering is not approved or endorsed by OpenCFD Limited, the producer of the OpenFOAM software and owner of the OpenFOAM® and OpenCFD® trade marks. OpenFOAM® is a registered trade mark of
OpenCFD Limited, a wholly owned subsidiary of the ESI Group.
- This content was made in 2014 and may contain incorrect or outdated information. The reader is solely responsible for his or her use of this information and AirShaper cannot be held liable for any damages.
www.airshaper.com
Content
• Introduction
• Literature
• Modeling
• Discretization
• Iteration
𝑣∙𝐿∙𝜌 𝑣∙𝐿
𝑅𝑒 = =
𝜇 ν
with
v = characteristic speed
L = characteristic length
µ = dynamic viscosity
ν = kinematic viscosity
d d
𝜌𝜙 d𝛺 = 𝜌𝜙 d𝛺 + 𝜌𝜙 𝑣 − 𝑣𝑏 ∙ n d𝑆
d𝑡 𝛺CM d𝑡 𝛺CV 𝑆CV
𝜕𝜌 𝜕(𝜌𝑢𝑖 )
+
𝜕𝑡 𝜕𝑥𝑖
𝜕𝜌 𝜕(𝜌𝑢𝑥 ) 𝜕(𝜌𝑢𝑦 ) 𝜕(𝜌𝑢𝑧 )
= + + + =0
𝜕𝑡 𝜕𝑥 𝜕𝑦 𝜕𝑧
𝜕
𝜌𝑣 d𝛺 + 𝜌𝑣𝑣 ∙ 𝑛 d𝑆 = 𝑓
𝜕𝑡 Ω 𝑆
𝜕
𝜌𝑣 d𝛺 + 𝜌𝑣𝑣 ∙ 𝑛 d𝑆 = T ∙ 𝑛 d𝑆 + 𝜌b dΩ
𝜕𝑡 Ω 𝑆 𝑆 Ω
𝜕(𝜌𝑣)
+ div 𝜌𝑣𝑣 = div T + 𝜌𝑏
𝜕𝑡
with stress tensor T – (surface forces):
2
𝑇 = − 𝑝 + 𝜇 𝑑𝑖𝑣 𝑣 𝐼 + 2𝜇𝐷
3
with rate of strain (deformation) tensor D:
1
𝐷= 𝑔𝑟𝑎𝑑 𝑣 + (𝑔𝑟𝑎𝑑 𝑣)𝑇
2
Basic course www.airshaper.com
Modeling
• Navier stokes equations:
– Momentum conservation: Ф = v
• Differential form – cartesian coordinate system:
𝜕(𝜌𝑢𝑖 )
+ div 𝜌𝑢𝑖 𝑣 = div 𝑡𝑖 + 𝜌𝑏𝑖
𝜕𝑡
with
𝜕𝑢𝑖 𝜕𝑢𝑗 2 𝜕𝑢𝑗
𝑡𝑖 = 𝜇 + 𝑖𝑗 − 𝑝 + 𝜇 𝑖𝑖
𝜕𝑥𝑗 𝜕𝑥𝑖 3 𝜕𝑥𝑗
𝑆
𝑓d𝑆 = 𝑘 𝑆𝑘
𝑓d𝑆
𝜙𝑒 = 𝜙𝐸 λ𝑒 + 𝜙𝑃 (1 − λ𝑒 )
AP ФP + Al Фl = Q P
l
with P the node being examined and “l” the involved neighbouring nodes.
𝐴Ф = 𝑄
Basic course www.airshaper.com
Discretization
• Calculating the cell value:
– Because each node only involves a few neighbouring nodes,
the matrix both sparse and diagonal:
□ □ □ □ □
□ □ □ □ Ф𝑊 □
□ □ □ □ □ □
□ □ □ □ □ Ф𝑆 □
𝐴𝑊 𝐴𝑆 𝐴𝑃 𝐴𝑁 𝐴𝐸 ∗ Ф𝑃 = 𝑄𝑃
□ □ □ □ □ Ф𝑁 □
□ □ □ □ □ □
□ □ □ □ Ф𝐸 □
□ □ □ □ □
Basic course www.airshaper.com
Discretization
• Discretization error (Ferziger & Peric):
“The difference between the exact solution of the
governing equations and the exact solution of the discrete
approximation”
– Refine the grid locally where variable changes are large
– Approximations become more accurate with a finer grid
– Higher order approximations can be more accurate, but
require more programming effort & computing power
– Higher order approximations are only accurate on a grid
that is fine enough and can cause oscillations
Back substitution
– https://www.youtube.com/watch?v=2j5Ic2V7wq4