Assignment #2 Axisymmetric Stagnation Point Flow and Tensor Algebra (Due Monday, Nov. 30 On Google Class)
Assignment #2 Axisymmetric Stagnation Point Flow and Tensor Algebra (Due Monday, Nov. 30 On Google Class)
Fall 2020
Assignment #2
Axisymmetric Stagnation Point Flow
and Tensor Algebra
(Due Monday, Nov. 30th on google class)
1. We derived the differential equation of planar stagnation point flow. You are required to
derive the differential equation of the axisymmetric flow and write down the boundary
conditions. Do not solve the differential equation.
2. Which of the following expressions are allowed in index notation (a, b, c, d, and e
are arbitrary quantities)?
3. Write the following formula in Gibbs notation using the symbol ∇, then use the index
notation to prove that the equation is correct.
div(𝜙𝐯) = 𝜙 div 𝐯 + 𝐯 ∙ grad 𝜙
div(𝐮 × 𝐯) = 𝐯 ∙ curl 𝐮 − 𝐮 ∙ curl 𝐯
4. The total fluid acceleration is conventionally written in terms of local and convective
accelerations as follows:
𝜕𝐯
𝐚= + 𝐯 ∙ ∇𝐯 …..(C)
𝜕𝑡
Derive the following other forms of the acceleration:
𝜕𝐯
𝐚= + ∇ ∙ 𝐯𝐯….. (D)
𝜕𝑡
𝜕𝐯 1
𝐚= + (∇ × 𝐯) × 𝐯 + ∇ ( |𝐯|2 )…..(R)
𝜕𝑡 2
𝜕𝐯 1 1
𝐚= + 𝐯 ∙ ∇𝐯 + ∇ ∙ (𝐯𝐯)…..(S)
𝜕𝑡 2 2
Here (C), (D), (R) and (S) correspond to convection, divergence, rotational and skew-symmetric
forms of acceleration. In principle, there is no difference between these forms. But, in terms of
numerical computation it makes a difference which form is used. For example, when properly
discretized (R) and (S) conserve both momentum and kinetic energy, (D) conserves momentum
only and (C) conserves neither. Computational efficiency depends on how many derivatives are
needed in evaluating each scheme. Six derivatives are needed for (R), 18 derivatines are needed
for (S) and 9 derivatives for the other two. Although (R) is computationally efficienct, (S)
produces the best results with least error.
5. For an incompressible flow, show that the curl of the vorticity ∇ × 𝛚 = −∇2 𝐯 using index
notation.
6. Consider the tensor Tij defined below. Compute T(ij) and T[ij]
a) 𝑣𝑥 = 𝑏𝑦 𝑣𝑦 = 0 𝑣𝑧 = 0
b) 𝑣𝑥 = 𝑏𝑦 𝑣𝑦 = 𝑏𝑥 𝑣𝑧 = 0
c) 𝑣𝑥 = −𝑏𝑦 𝑣𝑦 = 𝑏𝑥 𝑣𝑧 = 0