MAT3706/201/0/2023

Tutorial Letter 201/0/2023

Ordinary Differential Equations

MAT3706

Year module

Department of Mathematical Sciences

This tutorial letter contains Solution

of Assignment 01.

BAR CODE

university
Learn without limits. of south africa
 ASSIGNMENT 01

STUDY GUIDE: CHAPTERS 1, 2 and 3

We marked Q 1, Q2, Q4 and Q5

Question 1 Show that the system:

(D − 2)[x] + 2D[y] = 2 − 4e2t

(2D − 3)[x] + (3D − 1)[y] = 0

is equivalent to both the following triangular systems:

( (
(D2 + D − 2)[x] = 2 + 20e2t (D2 + D − 2)[y] = −6 − 4e2t
and .
D[x] − 2y = 12e2t − 6 x − (D + 1)[y] = 8e2t − 4

Given the three systems above, separately, state (with complete justifications) your strategy in
solving them completely
Solution

(i) The system in operator notation is :

(1) (D − 2)[x] + 2D[y] = 2 − 4e2t

(2) (2D − 3)[x] + (3D − 1)[y] = 0

Multiplying the equation (2) by 2yields :

(3) (4D − 6)[x] + (6D − 2)[y] = 0, XX

which when subtracted from 3× (1) yields :

(4) D[x] − 2y = −6 + 12e2t .XX

from (4) we have y = 12 D[x] + 3 − 6e2t XSubstituting it in the first equation (1) yields :

(D − 2)[x] + 2D 21 D[x] + 3 − 6e2t = 2 − 4e2t ⇒ (D2 + D − 2)[x] = 2 + 20e2t X

(5)

from (4) and (5) we have the required triangular system :

(6) (D2 + D − 2)[x] = 2 + 20e2t

(7) D[x] − 2y = −6 + 12e2t .XX

(8 marks)

2
 MAT3706/201

(ii) The given system is :

(1) (D − 2)[x] + 2D[y] = 2 − 4e2t

(2) (2D − 3)[x] + (3D − 1)[y] = 0

in which (1) is retained Xand the second is replaced by (2) −2(1) X :

(3) x(t) − (D + 1)[y] = −4 + 8e2t , X

which yields :

(4) x(t) = (D + 1)[y] − 4 + 8e2t . X

Substituting (4) into (1) yields :

(5) (D − 2) (D + 1)[y] − 4 + 8e2t + 2D[y] = 2 − 4e2t ⇒ (D2 + D − 2)[y] = −6 − 4e2t ,

XX

so in view of (3) and (5) the required triangular system is :

(6) (D2 + D − 2)[y] = −6 − 4e2t

(7) x − (D + 1)[y] = −4 + 8e2t XX

(8 marks)
D−2 2D
From (1) and (2) we have ∆ = = (− 21 D2 + D)(D − 2 − D) = 2 − D − D2 6= 0X,
2D − 3 −3D − 1
so that the system in not degenerate and hence has a unique solution.XIn same procedure we can see
∆ 6= 0 for both triangular systems.X
For solving the triangular system it is easier than the original system. From we can find x(t) then
substitute it into the we will obtain y(t) (The Calculation has 4 marks).
In same procedure we can find y(t) from then substitute it into the we will obtain x(t)
(The Calculation has 4 marks). (10 marks)
[26 marks]

Question 2 Solve the system:

(D2 + 1)[x1 ] − 2D[x2 ] = 2t,

(2D − 1)[x1 ] − (2 − D)[x2 ] = 5,

by using the elimination method (operator method).

Solution

Given the system:

(1) (D2 + 1)[x1 ] − 2D[x2 ] = 2t

(2) (2D − 1)[x1 ] − (2 − D)[x2 ] = 5

Eliminating [x2 ] first, we calculate (2 − D) × (1) − (2D)[(2)]XXto get:

(3) (−D3 − 2D2 + D + 2)[x1 ] = 4t − 2; XX

3
From (3) we solve for x1 . The auxiliary equation is m3 + 2m2 − m − 2 = (m − 1)(m + 1)(m + 2) =
0XX ⇔ m = ±1, −2, XXyielding the complimentary function:

(4) x1C.F. (t) = c1 e−2t + c2 e−t + c3 et , XXX

where c1 , c2 and c3 are arbitrary constants.

The particular integral is given by:
4t − 2
x1P.I. (t) =
(D − 1)(D + 1)(D + 2)
 
1 2 3
= 1 − D + D − D + . . . (4t − 2)
(D + 2)(D − 1)
1 1
= (4t − 2 − 4) = (4t − 6)
(D + 2)(D − 1) (D + 2)(D − 1)
 
1 2
= 1 + D + D + . . . (4t − 6)
D+2
1 −1 − 21
= (4t − 6 + 4) = (4t − 2) = (4t − 2)
D+2 D+2 1 + D2
D D2
 
1
= 1− + − . . . (4t − 2)
2 2 4
1 1
= (4t − 2 − 2) = (4t − 4) = 2t − 2,
2 2
i.e.,

(5) x1P.I. (t) = 2t − 2.XX

Hence x1 = x1C.F. + x1P.I. yields the general solution for x1 to be:

(6) x1 (t) = c1 e−2t + c2 e−t + c3 et + 2t − 2.XXX

Now we find x2 (t). From (1) we have ;

1
(7) D[x2 ] = (D2 + 1)[x1 ] − tXX
2
Now substitute (6) in the above equation (7) yields
1 1 5
(8) D[x2 ] = (D2 +1)[x1 ]−t = (D2 +1)[c1 e−2t +c2 e−t +c3 et +2t−2]−t = c1 e−2t +c2 e−t +c3 et −1.XX
2 2 2
So by integrating of the above equation we can obtain x2 (t) = − 54 c1 e−2t − c2 e−t + c3 et − tXXXThus
the solution is:

x1 (t) = c1 e−2t + c2 e−t + c3 et + 2t − 2,

and
5
x2 (t) = − c1 e−2t − c2 e−t + c3 et − t.
4
[25 marks]

4
 MAT3706/201

Question 3 Consider the system of linear differential equations:

Ẋ = AX.

(i) Suppose λ is an eigen value of A of multiplicity k > 1. Define a root vector corresponding to the
eigen value λ.

(ii) Show that each eigen value λ of A of multiplicity k has k linearly independent root vectors.

(iii) Suppose that λ is an eigen value of A of multiplicity k and Uλ,1 , Uλ,2 , . . . , Uλ,k are root vectors
corresponding to λ of orders 1, 2, . . . , k, respectively. Further, let:
t2 ti−1
Xi (t) = eλt Uλ,i + tUλ,i−1 + Uλ,i−2 + . . .
 
Uλ,1 , for i = 1, 2, . . . k.
2! (i − 1)!
 
Show that X1 (t), X2 (t), . . . , Xk (t) is a set of linearly independent solutions of the system
of linear differential equations and every solution of the system is a linear combination of the
solutions in this set.
Solution

(i) Definitions 3.1 or 3.2

(ii) Lemma 3.10

(iii) It is a combination of Theorem 3.8 and Lemma 3.11.

[Not marked]

Question 4 Solve the system:

   
0 1 0 1
Ẋ = 0 0 1  X, X(0) = 0 .
2 1 −2 1

Why is there a unique solution to the above system?

Solution

0−λ 1 0
0 0−λ 1 = −λ3 − 2λ2 + λ + 2 = (1 − λ2 )(λ + 2) = 0,i.e., λ = −1 or λ = 1 or
2 1 −2 − λ
λ = −2.XXXX
For λ = −1:       
1 1 0 v1 v1 1
0 1 1  v2  = 0 =⇒ v2  = −1 , XX
2 1 −1 v3 v3 1
 
1
i.e., X1 (t) = c1 −1 e−t are the solutions corresponding to λ = −1.X

1

5
For λ = 1:       
−1 1 0 u1 u1 1
 0 −1 1  u2  = 0 =⇒ u2  = 1 , XX
2 1 −3 u3 u3 1
 
1
i.e., X2 (t) = c2 1 et are the solutions corresponding to λ = 1.X

1
For λ = −2:       
2 1 0 w1 w1 1
0 2 1 w2  = 0 =⇒ w2  = −2 , XX
2 1 0 w3 w3 4
 
1
i.e., X3 (t) = c3 −2 e−2t are the solutions corresponding to λ = −2.X

4
Hence the solutions are:
     
1 1 1
−t
X(t) = X1 (t) + X2 (t) + X3 (t) = c1 −1 e   + c2 1 e + c3 −2 e−2t , XXX
  t 
1 1 4

where c1 , c2 , c3 are arbitrary constants.X

Also in view of the initial condition at t = 0 we have:

c1 + c2 + c3 = 1,
       
1 1 1 1 (
c1 = 21 = c2 ,

X(0) = 0 = c1 −1 + c2 1 + c3 −2 XX ⇒ −c1 + c2 − 2c3 = 0, ⇒ XXX
1 1 1 4
 c3 = 0,
c1 + c2 + 4c3 = 1


yielding:    
1 1
1 1
X(t) = −1 e−t + 1 et XX
2 2
1 1
According to theorem 2.29, We have 3 real different eigenvalues then Ẋ = AX, X(0) = X0 has a
unique solution.X [25 marks]

Question 5 Solve the system:  

4 6 6
Ẋ =  1 3 2 X
−1 −5 −2
completely.
Find the unique solution that passes through the point (11, 3, −7) when t = 0.
Solution

4−λ 6 6
1 3−λ 2 = −λ3 + 5λ2 − 8λ + 4 = (1 − λ)(λ − 2)2 = 0,XXXXi.e., λ = 1 or λ = 2 with
−1 −5 −2 − λ
multiplicity 2.XXX

6
 MAT3706/201

For λ = 1:       
3 6 6 v1 v1 4
1 2 2  v2 = 0 =⇒ v2 = 1  , XX
   
−1 −5 −3 v3 v3 −3
 
4
i.e., X1 (t) = c1  1  et are the solutions corresponding to λ = 1.X
−3
For λ = 2:       
2 6 6 u1 u1 3
1 1 2  u2 = 0 =⇒ u2 = 1  , XX
   
−1 −5 −4 u3 u3 −2
 
3
i.e., X2 (t) = c2 1  e2t are one set of solutions corresponding to λ = 2.X

−2
To find a second solution, we solve:
        
2 6 6 w1 3 w1 3
1 1 2  w2 = 1 =⇒ w2 = 1  , XX
     
−1 −5 −4 w3 −2 w3 − 23
   
 3 3 
i.e., X3 (t) = c3  1 + t 1  e2t are another set of solutions corresponding to λ = 2.XX
 
− 32 −2
Hence the solutions are:
       
4 3  3 3 
t 2t
X(t) = X1 (t) + X2 (t) + X3 (t) = c1 1 e + c2 1 e + c3
     1  + t 1  e2t , X

3
−3 −2 −2 −2

where c1 , c2 , c3 are arbitrary constants.X

Also:

4c1 + 3c2 + 3c3 = 11,
       
11 4 3 3  (
c1 = 2 = c3 ,
X(0) = 3 = c1 1 + c2 1 + c3 1 X ⇒ c1 + c2 + c3 = 3,
        X⇒ XX
−7 −3 −2 − 23
 c2 = −1,
−3c1 − 2c2 − 3 c3 = −7

2

yielding:
         t 
1 0  0 0  8e + (3 + 6t)e2t
X(t) = 3 0 et −  1  e2t + 3 1 + t  1  e2t =  2et + (1 + 2t)e2t  .XXX
0 −1 0 −1 −6et − (1 + 4t)e2t

[24 marks]