Math 292 Homework 5

Pranav Tikkawar
April 12, 2024

Problem 1
Solve x′′ (t) + 4x(t) = 3cos(2t) with x(0) = x0 and x′ (0) = y0
Creating a driven first order System:
We can rewrite the equation as a first order system by letting y(t) = x′ (t)
Then we have y ′ (t) + 4x(t) = 3cos(2t)      
d x(t) x(t) 0
Thus we get the matrix A in the equation dt =A +
y(t) y(t) 3cos(2t)
 
0 1
Where A =
−4 0
Solving the Homogeneous System:
The characteristic equation of A is det(A − µI) = 0
This gives us µ2 + 4 = 0
Thus we have µ = 2i and µ = −2i    
1 1
The eigenvectors of A are v1 = and v2 =
2i −2i
We can then split one of the eigenvectors and imaginary exponential into real
and imaginary parts to get e2it = cos(2t) + isin(2t)
 
1
cos(2t) + isin(2t)
2i
Solve x′′ (t) + 4x′ (t) = 3cos(2t) with x(0) = x0 and x′ (0) = y0
Creating a driven first order System:
We can rewrite the equation as a first order system by letting y(t) = x′ (t)
Then we have y ′ (t) + 4y(t) = 3cos(2t)      
d x(t) x(t) 0
Thus we get the matrix A in the equation dt =A +
  y(t) y(t) 3cos(2t)
0 1
Where A =
0 −4
Solving the Homogeneous System:
The characteristic equation of A is det(A − µI) = 0
This gives us µ2 + 4µ = 0
Thus we have µ = 0 and µ = −4   
1 −1
The eigenvectors of A are v1 = and v2 =
0 4

1
 −1
−1 e0t
  
At 1 0 1 −1
Thus the matrix exponential of A is e =
0 4 0 e−4t 0 4
 −4t 
e
At e − e4
e = 4
0 e−4t
Solving the Inhomogeneous System:
Given the matrix exponential of A, Rwe can solve the inhomogeneous system by
At t A(t−s)
using the formula
 x(t) = e x(0) + 0 e g(s)ds
0
Where g(s) =
3cos(2s) " #
−4t  
R t e e − e−4(t−s)
  
e 4e − e 4 x0 0
Thus we have x(t) = + 0 4 4 ds
0 e−4t y0 0 e−4(t−s) 3cos(2s)
" #
R t 3cos(2s)( e − e−4(t−s) )
The integral can be simplified to 0 4 4 ds
3cos(2s)(e−4(t−s) )
Thus the integral evalutes to
3 −4t
3 
40 ((5e − 1)sin(2t) − 2cos(2t)) + 20 e
3 3 −4t
10 (sin(2t) + 2cos(2t)) − 5 e

Thus the solution to the system’s x component is

e e−4t 3 3
x(t) = ex0 + y0 ( − ) + ((5e − 1)sin(2t) − 2cos(2t)) + e−4t
4 4 40 20

Problem 2
 
−(2 + y)(x + y)
Consider the vector field v(x, t) =
−y(1 − x)

a
Finding the Equilibrium Points:
−(2 + y)(x + y) = 0, −y(1 − x) = 0
Thus we have (x, y) = (0, 0), (1, −1), (1, −2) Linearizing the System:
The Jacobian of the system is
 
−2 − y −x − 2y − 2
J=
y x−1

Evaluating the Jacobian at the equilibrium points gives us

 
−2 −2
J(0, 0) =
0 −1
 
−1 −1
J(1, −1) =
−1 0

2
  
0 1
J(1, −2) =
−2 0
Stability of the Equilibrium Points:
The Trace and Determinant of the Jacobian at (0, 0) are −3 and 2 respectively.
Thus near the equilibrium point (0, 0) the system is a sink, and therefore is a
stable equilibrium point.
The Trace and Determinant of the Jacobian at (1, −1) are −1 and −1 respec-
tively.
Thus near the equilibrium point (1, −1) the system is a saddle, and therefore is
an unstable equilibrium point.
The Trace and Determinant of the Jacobian at (1, −2) are 0 and 2 respectively.
Thus near the equilibrium point (1, −2) the system is a periodic orbit, and
therefore is a not a stable equilibrium point, but is Lyanupov Stable.

b
 
(2 + y)(x + y)
Consider v(x, t) = Finding the Equilibrium Points:
−y(1 − x)
(2 + y)(x + y) = 0, −y(1 − x) = 0
Thus we have (x, y) = (0, 0), (1, −1), (1, −2) Linearizing the System:
The Jacobian of the system is
 
2 + y x + 2y + 2
J=
y x−1

Evaluating the Jacobian at the equilibrium points gives us

 
2 2
J(0, 0) =
0 −1
 
1 1
J(1, −1) =
−1 0
 
0 −1
J(1, −2) =
−2 0
Stability of the Equilibrium Points:
The Trace and Determinant of the Jacobian at (0, 0) are 1 and −2 respectively.
Thus near the equilibrium point (0, 0) the system is a saddle, and therefore is
an unstable equilibrium point.
The Trace and Determinant of the Jacobian at (1, −1) are 1 and 1 respectively.
Thus near the equilibrium point (1, −1) the system is a source, and therefore is
an unstable equilibrium point.
The Trace and Determinant of the Jacobian at (1, −2) are 0 and 2 respectively.
Thus near the equilibrium point (1, −2) the system is a periodic orbit, and
therefore is a not a stable equilibrium point, but is Lyanupov Stable.

3
Problem 3
Find exact solution of x′ (t) = v(x(t), t) and x(0) = 0 for v(x, t) = 2t(1 + x)
Starting from X0 = 0, then compute X1 , X2 , X3 , X4
Picard Iteration:
We can solve the equation by using Picard Iteration.

X0 = 0
Z t Z t
X1 = 2sds = 2sds = t2
0 0

t t
t4
Z Z
X2 = 2s(1 + X1 )ds = 2s(1 + s2 )ds = t2 +
0 0 2

t t
s4 t4 t6
Z Z
X3 = 2s(1 + X2 )ds = 2s(1 + s2 + )ds = t2 + +
0 0 2 2 6

t t
s4 s6 t4 t6 t8
Z Z
X4 = 2s(1 + X3 )ds = 2s(1 + s2 + + )ds = t2 + + +
0 0 2 6 2 6 24
t2n
P∞
Thus the exact solution is X(t) = n=1 n!
2
X(t) = et − 1