Differential Equations 2020/21

MA 209

Solutions to homework for week 1

Lecture Notes – Section 1.1

1 A system x0 (t) = f (t, x(t)) is autonomous if f does not depend on t. And an n-dimensional
system is linear if f (t, x(t)) = Ax for some n × n matrix A.
Note that a nonautonomous system is never linear, since there is no way the expression Ax
can depend on t.
(a) This ODE is not autonomous and hence also not linear.
(b) This ODE is autonomous, but not linear.
(c) This system is not autonomous and hence also not linear.
(d) This system is autonomous, but not linear (the expression xy is not a linear expression).
" #
0 1
(e) This system is autonomous and linear. For the matrix we can take A = .
1 1

2 (a) If x(t) ≡ π, then x0 (t) = 0 for all t. For the expression on the right side of the ODE we
get esin(x(t)) + cos(x(t)) = esin π + cos(π) = e0 + (−1) = 1 − 1 = 0.
We see that x(t) ≡ π satisfies x0 (t) = esin(x(t)) + cos(x(t)) for all t, and hence this function
is a solution to the ODE.
(b) If x(t) = eta x0 , then x0 (t) = aeta x0 . For the right hand side of the ODE we get ax(t) =
aeta x0 . So the function satisfies the ODE.
For the initial value we have x(0) = e0a x0 = e0 x0 = 1 · x0 = x0 , as required.
So x(t) = eta x0 satisfies x0 (t) = ax(t) for all t, and x(0) = x0 , hence is a solution to the
ODE with initial value.

c London School of Economics, 2019
MA 209 Differential Equations Solutions to homework for week 1 — Page 2

(c) If x(t) = Ceλ0 t , then x0 (t) = λ0 · Ceλ0 t , x00 (t) = λ20 · Ceλ0 t , x000 (t) = λ30 · Ceλ0 t , etc. In
general we have that the i-th derivative is x(i) (t) = λi0 · Ceλ0 t . If we fill that in in the
expression

an x(n) (t) + an−1 x(n−1) (t) + · · · + a1 x0 (t) + a0 x(t),

we get
an λn0 · Ceλ0 t + an−1 λn−1
0 · Ceλ0 t + · · · + a1 λ0 · Ceλ0 t + a0 · Ceλ0 t
= (an λn0 + an−1 λn−1
0 + · · · + a1 λ0 + a0 ) · Ceλ0 t
= p(λ0 ) · Ceλ0 t .

But we are given that p(λ0 ) = 0. So we get that

an x(n) (t) + an−1 x(n−1) (t) + · · · + a1 x0 (t) + a0 x(t) = p(λ0 ) · Ceλ0 t = 0 · Ceλ0 t = 0

for all t. Hence x(t) = Ceλ0 t is indeed a solution to the ODE.

(1 − t2 ) · 0 − 1 · (−2t) 2t  1 2
x01 (t) = = = 2t = 2t(x1 (t))2 .
(1 − t2 )2 (1 − t2 )2 1 − t2
1 1
So x1 (t) = 2
is a solution for the ODE x0 (t) = 2tx(t). Since is not defined if
1−t 1 − t2
1−t2 = 0, hence if t = −1 or t = +1, we cannot consider it as a solution on the whole real
1
line (−∞, +∞). But in the interval (−1, 1) there are no problems, hence x1 (t) =
1 − t2
0 2
is indeed a solution to the ODE x (t) = 2t(x(t)) for t ∈ (−1, 1).

For x2 (t) ≡ 0 we have x02 (t) = 0 = 2t · 02 = 2t(x2 (t))2 . So the function x2 (t) ≡ 0 is also
a solution to x0 (t) = 2t(x(t))2 . Note that x2 is a solution for all t ∈ R.

(Note that we have x1 (0) = 1 and x2 (0) = 0, so the solutions don’t give the same initial
values. This example merely shows that different initial values for the same ODE can
give very different looking solutions valid for very different values of t.)
(f) If x1 (t) ≡ 0, then x01 (t) = 0 = 3 · 02/3 = 3(x1 (t))2/3 . So x1 (t) ≡ 0 is indeed a solution to
the ODE.
For x2 (t) = t3 we get x02 (t) = 3t2 = 3(t3 )2/3 = 3(x2 (t))2/3 . So also x2 (t) = t3 is a solution
to the ODE.
Notice that the functions x1 an x2 satisfy not only the same ODE, but that they also
give the same initial value: x(0) = 0. So this is an example of an ODE where even if we
add an initial value condition, more than one solution is still possible.
 MA 209 Differential Equations Solutions to homework for week 1 — Page 3

(g) The candidate solution should have been written as x(s, t) = Cetτ0 +sσ0 (and not x(t) =
∂ ∂
. . .). For this function we have x(s, t) = Cetτ0 +sσ0 = τ0 Cetτ0 +sσ0 and
∂t ∂t
∂2 ∂ ∂ ∂
Cetτ0 +sσ0 = σ0 Cetτ0 +sσ0 = σ02 Cetτ0 +sσ0 .


2
x(s, t) =
∂s ∂s ∂s ∂s
∂ ∂2
So we have x(s, t) − 2 x(s, t) = τ0 Cetτ0 +sσ0 − σ02 Cetτ0 +sσ0 = (τ0 − σ02 ) · Cetτ0 +sσ0 = 0
∂t ∂s
(since τ0 − σ02 = 0). This shows that x(s, t) is a solution to the given partial differential
equation.

d m
3 If x(t) = tm is a solution to 2tx0 (t) = x(t), then, since t = mtm−1 , we must have
dt
2t(mtm−1 ) = tm . This is equivalent to (2m − 1)tm = 0. In order for this to be valid for all
t ≥ 1, we must have 2m − 1 = 0, and so m = 1/2.
√
So the only possible solution of this type is x(t) = t1/2 = t. But we still need to check that
this is really a solution. For that, we confirm that 2tx0 (t) = 2t· 21 t−1/2 = t·t−1/2 = t1/2 = x(t).
Also, x(t) = t1/2 is properly defined and differentiable for all t ≥ 1 (in fact for all t > 0), so
it is indeed a solution to the given ODE.

4 We notice that for any value of x(t) we have (x(t))2 ≥ 0, hence (x(t))2 + 1 ≥ 1 > 0. So
if some function x(t) is the solution of the differential equation x0 (t) = (x(t))2 + 1, then its
derivative x0 (t) is always greater than 0. From elementary calculus we know that a function
f : R → R with a positive derivative is an increasing function.
If we want to prove this last statement a bit more formal, we can argue as follows: Suppose x(t)
is a solution defined on some interval I ⊆ R. Take t1 , t2 ∈ I with t1 < t2 . Then by the
Fundamental Theorem of Calculus we have that
Z t2 Z t2 Z t2
0 2
x(t2 ) − x(t1 ) = x (t) dt = [(x(t)) + 1] dt ≥ 1 dt = t2 − t1 > 0.
t1 t1 t1

Thus we have that x(t2 ) > x(t1 ) for all t1 , t2 ∈ I with t1 < t2 . In other words, x(t) is an
increasing function on any interval I on which it is defined.

Lecture Notes – Section 1.3

1 If we take x1 (t) := x(t) and x2 (t) := x0 (t), then we find

x01 = x0 = x2 ,
0 00
x2 = x = −ω x = −ω 2 x1 .
2

x01 = x2 ,

So the state equation is the system
x02 = −ω 2 x1 .
 MA 209 Differential Equations Solutions to homework for week 1 — Page 4

2 This time we need four state variables : w1 (t) := x(t), w2 (t) := (t)x0 , w3 (t) := y(t) and
w4 (t) := y 0 (t). Then we find

w10 = x0 = w2 ,
w20 = x00 = −x = −w1 ,
w30 = y0 = w4 ,
w40 = y 00 = −y 0 − y = −w4 − w3 .


 w10 = w2 ,
 w0 = −w ,

2 1
And hence the state equation is the system 0
 w3 = w4 ,

 0

w4 = −w3 − w4 .

3 Introduce the state variables x1 (t) := x and x2 (t) := x0 . This leads to

x01 = x0 = x2 ,
x02 = x00 = −t sin x = −t sin(x1 ).

x01 = x2 ,

So the state equation is the system
x02 = −t sin(x1 ).