Numerical Methods Exam Questions 2025

Q: What is the classification of the partial differential equation 4∂²u/∂x² − 3∂²u/∂x∂y + 2∂²u/∂y² + ∂u/∂x + ∂u/∂y + 5u = 0?

The classification of the provided partial differential equation is elliptic. This classification is determined by examining the discriminant of the equation's second derivatives. Elliptic PDEs have a positive discriminant when written in canonical form .

Q: Using Taylor's series method, what is the value of y(0.1) for the differential equation y''−xy'−y = 0 with initial conditions y(0) = 1, y'(0) = 0?

Using the Taylor series method, the value of y(0.1) for the differential equation is approximately 1.0512. Taylor's series expands the function around a point using its derivatives, directly applying to initial conditions .

Q: What is the order of accuracy for the finite difference expansion with the given truncation error formula −∂²uij/∂x²(∆x)²/6 −∂³uij/∂x³(∆x)³/4?

The order of accuracy for the finite difference equation is 1. This is derived from the highest power of ∆x in the leading term of the truncation error .

Q: What order of accuracy does a finite difference algorithm with f'(x) = af(x) + bf(x−h) + cf(x−2h) generally have, given step size h and non-zero coefficients?

The finite difference algorithm has an accuracy order of O(h²). When including terms at multiple backward steps, the leading truncation error term gives the overall accuracy .

Q: What is the truncation error of the Crank-Nicolson method for a one-dimensional heat equation?

The truncation error of the Crank-Nicolson method for a one-dimensional heat equation is O(∆t²) + O(∆x²). This indicates a second-order accuracy in both time and space .

Q: How are finite difference methods formulated for solving PDEs when discretizing the domain using a finite number of grid points?

Finite difference methods are formulated by replacing derivatives in differential equations with difference equations corresponding to discrete grid points. This converts continuous problems into algebraic equations that can be solved using numerical techniques .

Q: Determine the range of convergence for the explicit method of the equation ∂²u/∂x² = 1/c ∂u/∂t, given r = ck/h².

The range of convergence for the explicit method is 0 ≤ r ≤ 0.5. This constraint ensures stability in computational algorithms for solving PDEs .

Q: Using the Runge-Kutta second-order method, calculate y(0.3) for the differential equation dy/dx = y−x with initial condition y(0) = 2 and step h = 0.1.

The value of y(0.3) computed using the Runge-Kutta second-order method is 2.4210. This approach iteratively updates y based on weighted slopes evaluated at interior points .

Q: Calculate the value of y(1) using the backward Euler’s method for the differential equation dy/dx = 0.25y² with y(0) = 1 and step size h = 1.

The value of y(1) using the backward (implicit) Euler’s method with step size h = 1 is 2.00. The backward Euler method approach involves iterating the function using the equation yn+1 = yn + h*f(xn+1,yn+1).

Q: Which statements are correct regarding truncation errors for first-order one-dimensional PDEs with constant spatial step size: (i) grid discretization, (ii) backward-differencing higher errors than central, (iii) forward equals backward, (iv) forward less than central?

The correct statements are (ii) and (iv). Backward-differencing does have higher truncation errors than central-differencing, and forward-differencing has less truncation error than central-differencing for spatial discretization .

This document is an examination paper for the Numerical Methods course at the Indian Institute of Technology Indore, scheduled for May 2, 2025. It includes multiple-choice questions covering topics such as differential equations, finite difference methods, and truncation errors. The exam has a total of 15 marks and requires students to write their roll number, branch, and signature on the question paper.

Uploaded by

ee230002039

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

0% found this document useful (0 votes)

44 views2 pages

Numerical Methods Exam Questions 2025

Uploaded by

ee230002039

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

A

Roll No.: Branch: Signature:

Indian Institute of Technology Indore

Department of Mathematics
End semester examination, Spring Semester 2025
Numerical Methods (MA 204N and MA 204)
Part-I

Date: 02 May 2025 Time: 2:30 PM–3:30 PM Max. Marks: 15

Instructions
i Write your roll number, branch and signature on the question paper.
ii Attempt all the MCQ questions.
iii Provide your answer by tick mark on the question paper.
iv No partial marks will be awarded.

1. Given the differential equation: y ′′ −xy ′ −y = 0 with the initial conditions: y(0) = 1, y ′ (0) =
0, the value of y(0.1) using Taylor’s series method is [3]

(a) 1.0512
(b) 1.0050
(c) 1.0501
(d) 1.0015
dy
2. Given the differential equation = 0.25 y 2 with y(0) = 1, the value of y(1) using backward
dx
(implicit) Euler’s Method for step size h = 1 is [2]

(a) 2.00
(b) 2.05
(c) 2.01
(d) 2.51
dy
3. Given the differential equation = y − x with y(0) = 2, the value of y(0.3) using the
dx
Runge-Kutta second-order method for the step size h = 0.1 is [2]

(a) 2.4210
(b) 2.2050
(c) 2.6492
(d) 2.8909
∂2u ∂2u ∂ 2 u ∂u ∂u
4. Classification of the PDE 4 − 3 + 2 + + + 5u = 0 is: [1]
∂x2 ∂x∂y ∂y 2 ∂x ∂y
(a) Elliptic
(b) Hyperbolic
(c) Parabolic
(d) Both parabolic and elliptic

1
5. If the finite difference expansion has the truncation error given by the formula,
∂ 2 uij (∆x)2 ∂ 3 uij (∆x)3
− − ,
∂x2 6 ∂x3 4
then the order of accuracy of the finite difference equation is [1]
(a) 1 (b) 2
1 1
(c) (d)
4 6
∂2u 1 ∂u ck
6. The explicit method of solution 2
= is convergent if r = 2 (with step length h in
∂x c ∂t h
space and k in time) must be [1]
(a) 0 ≤ r ≤ 0.5
(b) 0.5 < r ≤ 1
(c) 0≤r≤1
(d) for any positive value r
7. The truncation error of the Crank-Nicolson method for one-dimensional heat equation is: [1]
(a) O(∆t) + O(∆x2 )
(b) O(∆t2 ) + O(∆x)
(c) O(∆t2 ) + O(∆x2 )
(d) O(∆t) + O(∆x)
8. (i) The domain of problem is discretized in finite number of grids and the solution is
obtained at node points.
(ii) The order of truncation error in backward-differencing is more than that of central-
differencing.
(iii) The order of truncation error in forward-differencing is same as that of backward-
differencing.
(iv) The order of truncation error in forward-differencing is less than that of central-
differencing.
which of the above statement(s) is (are) corrects for a first-order one-dimensional partial
differential equation (PDE) with constant spatial step-size? [2]
(a) (i), (ii), (iii)
(b) (i), (iii), (iv)
(c) (ii), (iv)
(d) (ii), (iii), (iv)

9. The finite difference algorithm

f ′ (x) = af (x) + bf (x − h) + cf (x − 2h)
with step size h and non-zero values of a, b, c has accuracy of order: [2]
(a) O(h4 )
(b) O(h3 )
(c) O(h2 )
(d) O(h)
⋆ ⋆ ⋆⋆

Common questions