Scribd
Upload
119 views3 pages

Ma 214 HW4

This document contains 6 problems related to numerical analysis of differential equations. Problem 1 asks to show a differential equation has infinitely many solutions. Problem 2 defines order of accuracy for numerical methods and asks to determine the order of two methods. Problem 3 asks to show convergence of one of the methods to the exact solution. Problem 4 applies Euler's method to an example and estimates error. Problem 5 checks if two methods can approximate a certain solution. Problem 6 applies Euler's method to another example and shows convergence of the approximations.

Uploaded by

arvind murali
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
119 views3 pages

Ma 214 HW4

This document contains 6 problems related to numerical analysis of differential equations. Problem 1 asks to show a differential equation has infinitely many solutions. Problem 2 defines order of accuracy for numerical methods and asks to determine the order of two methods. Problem 3 asks to show convergence of one of the methods to the exact solution. Problem 4 applies Euler's method to an example and estimates error. Problem 5 checks if two methods can approximate a certain solution. Problem 6 applies Euler's method to another example and shows convergence of the approximations.

Uploaded by

arvind murali
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/ 3

MA 214, Spring 2021 IIT Bombay

Introduction to Numerical Analysis


Problem Set 4

1. (Non-uniqueness) Let k ∈ Z+ be fixed. Consider the following differential equation for


the unknown y : [0, ∞) → R

 dy = y 2k+1
2k
for t > 0,
dt (1)
y(0) = 0.

Show that (1) has infinitely many continuously differentiable solutions.

2. (Order of a numerical method) Consider the initial value problem:



 dy = f (t, y(t)) for t > 0,
dt (2)
y(0) = y0 ,

for the unknown y : [0, ∞) → R. Here the function f : [0, ∞) × R → R and y0 ∈ R are
given.

We will be further assuming, throughout this problem sheet, that the function f (t, y) is
Lipschitz continuous with respect to the y variable, i.e. there exists a constant Λ > 0
such that

|f (t, x) − f (t, y)| ≤ Λ|x − y| for all x, y ∈ R and for all t ≥ 0.

Take 0 < h  1 to be the time step and let ti = ih be the mesh points with i = 0, 1, 2, . . .
We shall denote the approximation to y(t) at ti by yi ∈ R, i.e.

yi ≈ y(ti ) for i = 1, 2, . . .

Taking the parameter θ ∈ [0, 1], let us consider the following time-stepping methods:

yi+1 = yi + h [θf (ti , yi ) + (1 − θ)f (ti+1 , yi+1 )] for i = 0, 1, 2, . . . (3)

and

yi+1 = yi + hf (ti + (1 − θ)h, θyi + (1 − θ)yn+1 ) for n = 0, 1, 2, . . . (4)

(i) Determine the order (local truncation error) of the method given by (3).
(ii) Determine the order (local truncation error) of the method given by (4).
(iii) In both the methods (3) and (4), comment on their orders when the parameter
θ = 21 .

Page 1 of 3
MA 214, Spring 2021 IIT Bombay

3. (Convergence of a numerical method) The goal of this exercise is to show that the ap-
proximation obtained by the theta-method (3) converges to the exact solution y(t) of
the differential equation (2) on an interval [0, T ] with T < ∞ fixed.

More precisely, let us denote the error committed at the ith time step by:

ei,h := yi,h − y(ti ) (5)

where, by a slight abuse of notation, we have denoted the approximation yi by yi,h to


make the dependance on the step size h explicit.

For a fixed h > 0, the global error committed is denoted by

Eh := max |ei,h | (6)


i=0,1,··· ,b Th c

where we have used the notation bβc ∈ Z to denote the integer part of β.

Show that, for the theta-method (3),

lim Eh = 0.
h→0

Also, determine the order of convergence.

4. Consider the initial value problem:


 0
y (t) = −2y, 0 ≤ t ≤ 1,
y(0) = 1.

(i) Find an upper bound on the error in approximating the value of y(1) computed
using the (forward) Euler’s method in terms of the step size h.
(ii) For the choices of h = 0.5, 0.25, 0.2, obtain approximate values of y(1) using (for-
ward) Euler’s method.
(iii) Find the error involved in the approximate value of y(1) obtained in (ii) above by
comparing with the exact solution.

5. Let 0 < α < 1. Check whether or not the Euler’s and the Trapezoidal methods deter-
mines an approximation to the non-trivial solution of the initial value problem:
 0
y (t) = y α ,
y(0) = 0.

Note that the ODE is in separable form and hence a non-trivial solution to initial value
problem can be found analytically.

Page 2 of 3
MA 214, Spring 2021 IIT Bombay

6. Using the Euler’s method , write the formula for approximating the solution to the initial
value problem:
 0
y (t) = te−5t − 5y,
y(0) = 0.

on the interval [0, 1] with step size h = 1/N .

Denoting by yN the resulting approximation to y(1), show that yN → y(1) as N → +∞.

Page 3 of 3

You might also like