Scribd
Upload
11 views4 pages

Assignment M.S. 2022

The document outlines an assignment for a course on Applied Numerical Methods, detailing various problems related to finite element analysis (FEA), boundary value problems (BVP), and numerical methods. Students are required to submit their assignments by May 12, 2024, and the tasks include deriving methods, applying numerical techniques, and analyzing stability conditions. The assignment covers a range of topics, including Galerkin methods, Ritz method, shape functions, and heat transfer equations.

Uploaded by

priontimahi76
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
11 views4 pages

Assignment M.S. 2022

The document outlines an assignment for a course on Applied Numerical Methods, detailing various problems related to finite element analysis (FEA), boundary value problems (BVP), and numerical methods. Students are required to submit their assignments by May 12, 2024, and the tasks include deriving methods, applying numerical techniques, and analyzing stability conditions. The assignment covers a range of topics, including Galerkin methods, Ritz method, shape functions, and heat transfer equations.

Uploaded by

priontimahi76
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Assignment on Applied Numerical Methods

M.S. (Mathematics)
Course No: Math-531: Total Marks: 30

[N.B: Students are required to submit their Assignment on or before 12 May, 2024.
No assignment will receive after the assigned date]

1. What are nodes and elements in FEA. Explain the basic differences
between FDM and FEM.

2. Derive Galerkin and Modified Galerkin finite element method.


3. Derive Modified Galerkin Method for matrix formulation and use this
formula for cubic approximation.

4. Use the Ritz method to obtain a one-parameter and a two-parameter


𝑑2𝑢
approximation to the BVP: + 𝑢 = −𝑥 , 0 < 𝑥 < 1, with 𝑢(0) =
𝑑𝑥 2
𝑢(1) = 0.

5. Using a two-parameter trial solution:


𝑢̃(𝑥) = 𝑎1 (1 − 𝑥 2 ) + 𝑎2 (1 − 𝑥 4 )
determine an approximate solution to the boundary value problem:
𝑑2𝑢
+ (1 + 𝑥 2 )𝑢 + 1 = 0 , −1 ≤ 𝑥 ≤ 1 with
𝑑𝑥 2
𝑢(−1) = 0, 𝑢(1) = 0.

6. Consider the boundary value problem


𝑑2𝑢
=2; 1≤𝑥≤3
𝑑𝑥 2
𝑢(1) = 1; 𝑢′ (3) = 6
(i) Determine modified Galaerkin equations appropriate to this
problem, distinguishing between essential and suppressible
boundary conditions.

(ii) Consider the trial solution:

Page 1 of 4
𝑢̃(𝑥) = 𝜃0 (𝑥) + 𝑎1 𝜃1 (𝑥) + 𝑎2 𝜃2 (𝑥) in which
2−𝑥 , 1≤𝑥≤2
𝜃0 (𝑥) = {
0, 2≤𝑥≤3
𝑥−1 , 1≤𝑥≤2
𝜃1 (𝑥) = {
3−𝑥 , 2≤𝑥≤3
0, 1≤𝑥≤2
𝜃2 (𝑥) = {
𝑥−2 , 2≤𝑥 ≤3
In this case the given coordinate functions are piecewise linear
functions. Sketch 𝜃0 (𝑥), 𝜃1 (𝑥), 𝜃2 (𝑥) and 𝑢̃(𝑥) and give
interpretations to the parameters 𝑎1 and 𝑎2 .

7. Apply the matrix procedure, obtain the approximate solution to the


boundary value problem:
𝑑2𝑢
+ 𝑢 = 𝑒𝑥 ; 0 ≤ 𝑥 ≤ 1
𝑑𝑥 2
𝑢(0) = 0; 𝑢(1) = 0
use the trial solution of the form 𝑢̃(𝑥) = 𝛼0 + 𝛼1 𝑥 + 𝛼2 𝑥 2 + 𝛼3 𝑥 3 .

8. Determine one or two parameter approximations to the boundary value


problem:
𝑑2𝑢
+ 𝑢 = −𝑥 , 0 ≤ 𝑥 ≤ 1
𝑑𝑥 2
𝑢(0) = 𝑢(1) = 0
using standard Galerkin method. As trial solutions use (i) 𝑢̃(𝑥) =
𝑎1 𝑥(1 − 𝑥) and (ii) 𝑢̃(𝑥) = 𝑥(1 − 𝑥)(𝑎1 + 𝑎2 𝑥). Compare your
answers with the exact solution.

9. Obtain a two-term Galerkin solution for the differential


𝑑2𝑢
− 10𝑥 2 = 5 , 0 ≤ 𝑥 ≤ 1; 𝑢(0) = 𝑢(1) = 0. Using the trial
𝑑𝑥 2
functions 𝑁1 (𝑥) = 𝑥(𝑥 − 1), 𝑁2 (𝑥) = 𝑥 2 (𝑥 − 1). [D. Hutton-135]

10. Develop the linear shape functions 𝐿1 (𝜉) and 𝐿2 (𝜉) directly by
assuming an approximation solution 𝑢̃(𝜉) = 𝛼1 + 𝜉𝛼2 in the standard
element −1 ≤ 𝜉 ≤ 1 and solving for 𝛼1 , 𝛼2 .

Page 2 of 4
11. Briefly describe four approximation methods(i.e. Point collocation,
Sub-domain, Least square and Galerkin). Use 4-approximation
𝑑2𝑦
methods to solve the DEs: + 400𝑥 2 = 0, 0 ≤ 𝑥 ≤ 1; B.Cs:
𝑑𝑥 2
𝑦(0) = 𝑦(1) = 0. Compare the solution from Galerkin’s and the exact
solution.

12. Using Lagrange polynomial, find shape function for (i) three noded
bar element (ii) Five noded bar element. Plot the variation of shape
functions.

13. Solve the following differential equation using


𝑑2𝜑
− 𝜑 = 𝑥; 𝜑(𝑥 = 0) = 0 and 𝜑(𝑥 = 1) = 0.
𝑑𝑥 2
using Least square method, point collocation method and sub-domain
method. Find the value of 𝜑 at 𝑥 = 0.25, 0.5 & 0.75.
14. Derive one dimensional heat transfer equation for finite element
method.

15. A wall of 0.6m thickness having thermal conductivity of 1.2 W/mk. The
wall is to be insulated with a material of thickness 0.06m having an average
thermal conductivity of 0.3 W/mk. The inner surface temperature is 1000oC
and outside of the insulation is exposed to atmospheric air at 30oC with heat
transfer coefficient of 35 W/m2k. Calculate the nodal temperature.
16. For the spring assemblage shown in the figure below, obtain (i) the
global stiffness matrix (ii) the displacements of nodes 2-4 (iii) the
global nodal forces and (iv) the local element forces. Node 1 is fixed
while node 5 is given a fixed, know displacement 𝛿 = 20 mm. The
spring constant are all equal to 𝑘 = 200 𝑘𝑁/𝑚.

17. (i) Explain convergence, consistence and stability of numerical


scheme.
(ii) Define well-posed and illposed problems.
(iii) Explain CFL (Courant-Friedrichs-Lewy) condition.

Page 3 of 4
18. Derive finite difference scheme for the 1-D wave equation. Hence
discuss the Von-Neumann stability condition.

20. Derive Lax-Wendroff (Lx-W) scheme from the advection equation


𝑣𝑡 + 𝑎𝑣𝑥 = 0 . What is the order of Lx- W scheme? Show that Lx-W
scheme is stable and consistent.

21. Define pointwise convergence of a scheme. Show that the solution of


the difference scheme
𝑢𝑘𝑛+1 = (1 − 2𝑟)𝑢𝑘𝑛 + 𝑟(𝑢𝑘+1𝑛 𝑛
+ 𝑢𝑘−1 ); 𝑢𝑘0 = 𝑓(𝑘𝛥𝑥)
𝛥𝑡
where 𝑟 = 𝛼 2, 0 ≤ 𝑟 ≤ 1/2, converges pointwise to the solution
𝛥𝑥
of the initial-value problem 𝑣𝑡 = 𝛼𝑣𝑥𝑥 , 𝑥 ∈ ℝ, 𝑡 > 0; 𝑣(𝑥, 0) = 𝑓(𝑥),
𝑥 ∈ ℝ.

22. Analyze the stability of the difference scheme


𝑢𝑘𝑛+1 = 𝑟𝑢𝑘−1
𝑛
+ (1 − 2𝑟)𝑢𝑘𝑛 + 𝑟𝑢𝑘+1
𝑛
, −∞ < 𝑘 < ∞;
𝛥𝑡
where 𝑟 = 𝜈 2.
𝛥𝑥

23. Discuss the stability of Dufort and Frankel scheme for one-
dimensional diffusion equation 𝑇𝑡 = 𝛼𝑇𝑥𝑥 given by
(1 + 2𝑟)𝑇𝑘𝑛+1 = 2𝑟[𝑇𝑘−1 𝑛 𝑛
+ +𝑇𝑘+1 ] + (1 − 2𝑟)𝑢𝑘𝑛−1 ,
𝛥𝑡
where 𝑟 = 𝛼 2, using Von-Neumann condition.
𝛥𝑥

24. Consider the 1D wave equation 𝜕2 𝑢 = 𝛼 2 𝜕2𝑢 , 0 ≤ 𝑥 ≤ 𝐿, 𝑡 ≥ 0


𝜕𝑡 2 𝜕𝑥 2
(i) Derive finite difference approximation using CTCS scheme.
(ii) What order approximation is this in time and in space?
(iii) Use Von Neumann stability criterion to find the mode of
amplitude.

Page 4 of 4

You might also like