Computational Fluid Dynamics Multiple Choice Questions on “High Resolution Schemes – Downwind and Normalized Weighing Factor”.
1. The Downwind Weighing Factor in the normalized form is given by __________
a) (frac{tilde{phi_f}-tilde{phi_c}}{1-tilde{phi_c}})
b) (frac{tilde{phi_c}-tilde{phi_f}}{1-tilde{phi_c}})
c) (frac{tilde{phi_f}-tilde{phi_c}}{1-tilde{phi_f}})
d) (frac{tilde{phi_c}-tilde{phi_f}}{1-tilde{phi_f}})
Answer: a
Clarification: The Downwind Weighing Factor is given by
DWFf=(frac{phi_f-phi_c}{phi_d-phi_c})
Normalizing this, we get
DWFf=(frac{tilde{phi_f}-tilde{phi_c}}{tilde{phi_d}-tilde{phi_c}})
But, the value of (tilde{phi_d}) is 1. So,
DWFf=(frac{tilde{phi_f}-tilde{phi_c}}{1-tilde{phi_c}})
2. The value of the Downwind Weighing Factor (DWF) lies between ___________
a) 0≤DWF≤∞
b) DWF≥0
c) 0≤DWF≤1
d) DWF≤1
Answer: c
Clarification: By using DWF, the high-resolution estimate of (tilde{phi_f}, or, phi_f) is redistributed between the upwind and the downwind nodes. As the value of Φf computed using Φc and Φc. The value of DWF always lies between 0 and 1.
3. The value of DWF for the downwind scheme is __________
a) 0
b) 1
c) 2
d) 3
Answer: b
Clarification: The relation between the DWF formulation and the TVD formulation is given by
DWFf=(frac{1}{2}) ψ(rf)
The ψ(rf) value for downwind scheme is 2. Therefore, the DWFf value is 1.
4. DWFf for the FROMM scheme is ___________
a) (frac{1}{2(1-tilde{phi_c})})
b) (frac{1}{4(1-tilde{phi_c})})
c) (frac{1}{2})
d) (frac{1}{4})
Answer: b
Clarification: For FROMM scheme,
(tilde{phi_f}=tilde{phi_c}+frac{1}{4})
Therefore,
DWFf=(frac{(tilde{phi_c}+1/4)-tilde{phi_c}}{1-tilde{phi_c}})
DWFf=(frac{1}{4(1-tilde{phi_c})}.)
5. For a scheme modelled using the DWF method, the diagonal coefficient becomes zero when ___________
a) DWFf > 0
b) DWFf > 1
c) DWFf > 0.5
d) DWFf > 2
Answer: c
Clarification: For values of DWFf larger than 0.5, results in a system with negative diagonal coefficients. So, the system becomes unsolvable by iterative methods. This happens whenever Φf > 0.5(Φc+Φd).
6. The value of DWFf for the central difference scheme is __________
a) 1
b) (frac{1}{3})
c) (frac{1}{4})
d) (frac{1}{2})
Answer: d
Clarification: For the central difference scheme,
ψ(rf)=1
So, the value of DWFf for this scheme is ½.
7. The deferred correction source term of the NWF method using he normalized interpolation profile (tilde{phi_f}=ltilde{phi_c}+k) is _________
a) (1-l-k)Φu
b) (k)Φu
c) (-l)Φu
d) (l-k)Φu
Answer: a
Clarification: We have the equation
(tilde{phi_f}=ltilde{phi_c}+k)
This can be expanded as
(frac{phi_f-phi_u}{phi_d-phi_u}=l frac{phi_c-phi_u}{phi_d-phi_u}+k)
(frac{phi_f-phi_u}{phi_d-phi_u}=l frac{phi_c-phi_u}{phi_d-phi_u}+kfrac{phi_d-phi_u}{phi_d-phi_u})
Φf=l(Φc-Φu)+k(Φd-Φu)+Φu
Φf=l(Φc))+k(Φd))+(1-l-k)Φu
The term (1-l-k)Φu in this equation is the deferred correction source term.
8. The high-resolution schemes formulated using the NWF method with the equation (tilde{phi_f} = ltilde{phi_c}+k) are stable without any alteration when __________
a) k>2
b) l>2
c) k>l
d) l>k
Answer: d
Clarification: The NWF formulation of the high-resolution schemes, when the value of l is greater than the value of k, the diagonal coefficients are all positive and hence the solution is highly stable. This is the case everywhere except a narrow region in NVD.
9. What is DWFf for the second-order upwind scheme?
a) (frac{tilde{phi_c}}{2(1-tilde{phi_c})})
b) (frac{1}{2(1-tilde{phi_c})})
c) (frac{tilde{phi_c}}{4(1-tilde{phi_c})})
d) (frac{1}{4(1-tilde{phi_c})})
Answer: a
Clarification: For the second order upwind scheme,
(tilde{phi_f}=frac{3}{2} tilde{phi_c})
Therefore,
DWFf=(frac{frac{3}{2}tilde{phi_c}-tilde{phi_c}}{1-tilde{phi_c}})
DWFf=(frac{tilde{phi_c}}{2(1-tilde{phi_c})}).
10. Along the downwind line of the NVD, the values of _____________ are changed to make the system stable.
a) ac
b) (l,k)
c) Φc
d) Φf
Answer: b
Clarification: Along the downwind line of NVD, the values of (l,k)=(0,1), a value of zero is obtained for the diagonal coefficient and the system becomes unstable. To overcome this problem, the values of (l,k) are set equal to (L,1-LΦf). The value of L can be chosen which is usually set to l in the previous interval.
