COMPUTATIONAL FLUID DYNAMICS

Presented by
Dr. Hesham M. Khalil
•
•
•
•
•
•
•
•
•

•
•
•
•
Introduction

•
Introduction

•
Introduction

•
•
The TDMA

•
∅1 = 𝐶1
−𝛽2 ∅1 + 𝐷2 ∅2 − 𝛼2 ∅3 = 𝐶2
−𝛽3 ∅2 + 𝐷3 ∅3 − 𝛼3 ∅4 = 𝐶3
−𝛽4 ∅3 + 𝐷4 ∅4 − 𝛼4 ∅5 = 𝐶4
…………………………..= ⋯
−𝛽𝑛 ∅𝑛−1 + 𝐷𝑛 ∅𝑛 − 𝛼𝑛 ∅𝑛+1 = 𝐶𝑛+1
∅𝑛+1 = 𝐶𝑛+1
• 𝜙1 𝜙𝑛+1

−𝛽𝑗 ∅𝑗−1 + 𝐷𝑗 ∅𝑗 − 𝛼𝑗 ∅𝑗+1 = 𝐶𝑗

𝛼2 𝛽2 𝐶2
∅2 = ∅ + ∅ +
𝐷2 3 𝐷2 1 𝐷2
𝛼3 𝛽3 𝐶3
∅3 = ∅4 + ∅2 +
𝐷3 𝐷3 𝐷3
𝛼4 𝛽4 𝐶4
∅4 = ∅ + ∅ +
𝐷4 5 𝐷4 3 𝐷4
𝛼𝑛 𝛽𝑛 𝐶𝑛
∅𝑛 = ∅𝑛+1 + ∅𝑛−1 +
𝐷𝑛 𝐷𝑛 𝐷𝑛
The TDMA

•
• 𝜙2

𝛽2 𝐶
𝛼3 𝛽3 ∅1 + 2 + 𝐶3
𝐷2 𝐷2
∅3 = 𝛼2 ∅4 + 𝛼
𝐷3 − 𝛽3 𝐷3 − 𝛽3 2
𝐷2 𝐷2
The TDMA
𝛼2 𝛽2 𝐶2
• 𝐴2 = 𝐶2′ = 𝜙1 +
𝐷2 𝐷2 𝐷2

𝛽2 𝐶
𝛼3 𝛽3 ∅1 + 2 + 𝐶3 𝛼3 𝛽3 𝐶2′ + 𝐶3
𝐷2 𝐷2
∅3 = 𝛼2 ∅4 + 𝛼2 ∅3 = ∅ +
𝐷3 − 𝛽3 𝐷3 − 𝛽3 𝐷3 − 𝛽3 𝐴2 4 𝐷3 − 𝛽3 𝐴2
𝐷2 𝐷2
The TDMA
𝛼3 ′
′
𝛽3 2 + 𝐶3
𝐶
• 𝐴3 = 𝐶3 =
𝐷3 − 𝛽3 𝐴2 𝐷3 − 𝛽3 𝐴2

𝛼3 𝛽3 𝐶2′ + 𝐶3
∅3 = ∅4 + ∅3 = 𝐴3 ∅4 + 𝐶3′
𝐷3 − 𝛽3 𝐴2 𝐷3 − 𝛽3 𝐴2

• 𝜙3
The TDMA

∅𝑗 = 𝐴𝑗 ∅𝑗+1 + 𝐶𝑗′

𝛼𝑗 ′
𝛽𝑗 𝐶𝑗−1 + 𝐶𝑗
𝐴𝑗 = 𝐶𝑗′ =
𝐷𝑗 − 𝛽𝑗 𝐴𝑗−1 𝐷𝑗 − 𝛽𝑗 𝐴𝑗−1

• 𝑗=1 𝑗 =𝑛+1
𝐴𝑗 𝐶𝑗′ 𝐴1 = 0 𝐶1′ = ∅1
′
𝐴𝑛+1 = 0 𝐶𝑛+1 = ∅𝑛+1
The TDMA

•
𝛼𝑗 𝛽𝑗 𝐷𝑗 𝐶𝑗

−𝛽𝑗 ∅𝑗−1 + 𝐷𝑗 ∅𝑗 − 𝛼𝑗 ∅𝑗+1 = 𝐶𝑗

• 𝐴𝑗 𝐶𝑗′ 𝑗=2 𝑗=𝑛

𝛼𝑗 ′
𝛽𝑗 𝐶𝑗−1 + 𝐶𝑗
𝐴𝑗 = 𝐶𝑗′ =
𝐷𝑗 − 𝛽𝑗 𝐴𝑗−1 𝐷𝑗 − 𝛽𝑗 𝐴𝑗−1
The TDMA

• 𝜙 𝜙𝑗
𝜙𝑛 𝜙𝑛−1 𝜙𝑛−2 𝜙2

∅𝑗 = 𝐴𝑗 ∅𝑗+1 + 𝐶𝑗′

•
Application of the TDMA to Two Dimensional Problems

𝑎𝑃 ∅𝑃 = 𝑎𝑊 ∅𝑊 + 𝑎𝐸 ∅𝐸 +𝑎𝑆 ∅𝑆 +𝑎𝑁 ∅𝑁 + 𝑏

−𝑎𝑆 ∅𝑆 + 𝑎𝑃 ∅𝑃 − 𝑎𝑁 ∅𝑁 = 𝑎𝑊 ∅𝑊 + 𝑎𝐸 ∅𝐸 + 𝑏
Application of the TDMA to Two Dimensional Problems

•
• 𝛼𝑗 = 𝑎𝑁 𝛽𝑗 = 𝑎𝑆 𝐷𝑗 = 𝑎𝑃
𝐶𝑗 = 𝑎𝑊 𝜙𝑊 + 𝑎𝐸 𝜙𝐸 + 𝑏
• 𝑗 𝑛

• 𝜙𝑊 𝑃

• 𝜙𝐸
Application of the TDMA to Two Dimensional Problems

• 𝜙𝐸

•
Point-Iterative Methods

•
•

2𝑥1 + 𝑥2 + 𝑥3 = 7
−𝑥1 + 3𝑥2 − 𝑥3 = 2
𝑥1 − 𝑥2 + 2𝑥3 = 5

• 𝑥1
𝑥2

𝑥1 = 7 − 𝑥2 − 𝑥3 /2
𝑥2 = 2 + 𝑥1 + 𝑥3 /3
𝑥3 = 5 − 𝑥1 + 𝑥2 /2
Point-Iterative Methods

• 𝑥1 𝑥2 𝑥3
•
𝑥1 𝑥2 𝑥3
•

•
•
Jacobi Iteration Method
(𝑘) (𝑘)
• 𝑥1 𝑥2 𝑘

(𝑘−1) (𝑘−1)
𝑥1 𝑥2 𝑘−1
(0) (0) (0)
• 𝑥1 𝑥2 𝑥3
•

(1) (1) (1)

•
Jacobi Iteration Method

• 𝑥1 = 1.0000 𝑥2 = 2.0000 𝑥3 = 3.0000

•
Jacobi Iteration Method

• 𝑛 𝑛
𝑨. 𝒙=𝒃 𝑨

𝑎𝑖𝑗 𝑥𝑗 = 𝑏𝑖
𝑖=1

• 𝑥𝑖
𝑖 𝑡ℎ
𝑛

𝑎𝑖𝑖 𝑥𝑖 = 𝑏𝑖 − 𝑎𝑖𝑗 𝑥𝑗 𝑖 = 1, 2, … , 𝑛
𝑗=1
𝑗≠𝑖
Jacobi Iteration Method

• 𝑎𝑖𝑗
𝑘 𝑥𝑗
𝑘−1

𝑛
(𝑘) −𝑎𝑖𝑗 (𝑘−1) 𝑏𝑖
𝑥𝑖 = 𝑥𝑗 + 𝑖 = 1, 2, … , 𝑛
𝑎𝑖𝑖 𝑎𝑖𝑖
𝑗=1
𝑗≠𝑖

•
Gauss-Seidel Iteration Method

𝑥1 = 7 − 𝑥2 − 𝑥3 /2
𝑥2 = 2 + 𝑥1 + 𝑥3 /3
𝑥3 = 5 − 𝑥1 + 𝑥2 /2
•

•
Gauss-Seidel Iteration Method

•
(0) (0)
𝑥2 = 0 𝑥3 = 0
0 0
(1)
7 − 𝑥2 − 𝑥3 7−0−0
𝑥1 = = = 3.5
2 2
• 𝑥2 = 2 + 𝑥1 + 𝑥3 /3
(0)
• 𝑥1 𝑥3 𝑥1 =0
(0)
𝑥3 = 0
(1)
𝑥1 𝑥1 = 3.5
Gauss-Seidel Iteration Method

1 0
(1)
2 + 𝑥1 + 𝑥3 2 + 3.5 + 0
𝑥2 = = = 1.833
3 3

• 𝑥3 = 5 − 𝑥1 + 𝑥2 /2
1
𝑥1 = 3.5
(1)
𝑥2 = 1.833
1 1
(1)
5 − 𝑥1 + 𝑥2 5 − 3.5 + 1.833
𝑥3 = = = 1.6667
2 2
Gauss-Seidel Iteration Method

•
𝑖−1 𝑛
(𝑘) −𝑎𝑖𝑗 (𝑘) −𝑎𝑖𝑗 (𝑘−1) 𝑏𝑖
𝑥𝑖 = 𝑥𝑗 + 𝑥𝑗 + 𝑖 = 1, 2, … , 𝑛
𝑎𝑖𝑖 𝑎𝑖𝑖 𝑎𝑖𝑖
𝑗=1 𝑗=𝑖+1
Relaxation Methods

𝛼
•
𝑛
(𝑘) −𝑎𝑖𝑗 (𝑘−1) 𝑏𝑖
𝑥𝑖 = 𝑥𝑗 + 𝑖 = 1, 2, … , 𝑛
𝑎𝑖𝑖 𝑎𝑖𝑖
𝑗=1
𝑗≠𝑖

•
𝑛
(𝑘) (𝑘−1) −𝑎𝑖𝑗 (𝑘−1) 𝑏𝑖
𝑥𝑖 = 𝑥𝑖 + 𝑥𝑗 + 𝑖 = 1, 2, … , 𝑛
𝑎𝑖𝑖 𝑎𝑖𝑖
𝑗=1
Relaxation Methods

•
𝛼

𝑛
(𝑘) (𝑘−1) −𝑎𝑖𝑗 (𝑘−1) 𝑏𝑖
𝑥𝑖 = 𝑥𝑖 +𝛼 𝑥𝑗 + 𝑖 = 1, 2, … , 𝑛
𝑎𝑖𝑖 𝑎𝑖𝑖
𝑗=1

• 𝛼 𝛼

• 𝛼 𝛼
ASSIGNMENT NO. 2

•
•

•
•