MAS 201 Spring 2024 (CD)

Differential Equations and Applications

§2.2

Gyo Taek Jin

March 4, 2024

1 / 13
 Chapter 2
First-Order Differential Equations
§2.1 Solution Curves Without a Solution
§2.2 Separable Equations
§2.3 Linear Equations
§2.4 Exact Equations
§2.5 Solutions by Substitutions
..
.

2 / 13
Introduction
If a first-order differential equation is of the form
dy
= g(x ), (1)
dx
one can easily find a solution by integration.
If g(x ) is a continuous function, then integrating both sides of (1),
we get Z
y = g(x ) dx = G(x ) + c

where G(x ) is an antiderivative of g(x ). (i.e., G 0 (x ) = g(x ))

For example, if dy /dx = 1 + e 2x , then
Z
y= (1 + e 2x )dx = x + 12 e 2x + c.

3 / 13
Separable Equation
Definition (Separable Equation)
A first-order differential equation of the form
dy
= g(x )h(y )
dx
is said to be separable or to have separable variables.
For example, the equation
dy
= x 2 y 4 e 5x −3y
dx
is separable because x 2 y 4 e 5x −3y = (x 2 e 5x )(y 4 e −3y ).
But the equation
dy
= y + cos x
dx
is nonseparable because y + cos x is not of the form g(x )h(y ).
4 / 13
Solving a Separable Equation
If y = φ(x ) is a solution of
dy
= g(x )h(y ),
dx
then
1 dy 1
= g(x ) ⇒ φ0 (x ) = g(x ).
h(y ) dx h(φ(x ))
Integrating both sides by x , we obtain
1
Z Z
φ0 (x )dx = g(x )dx .
h(φ(x ))
By substituting y = φ(x ), we have
1
Z Z
dy = g(x )dx ⇒ H(y ) = G(x ) + c
h(y )
where H(y ) and G(x ) are antiderivatives of 1/h(y ) and g(x ),
respectively.
5 / 13
Example (1) (Solving a Separable DE)
Solve
(1 + x ) dy − y dx = 0

Solution
Dividing the equation by (1 + x )y , we obtain

dy dx
=
y 1+x
Integration gives
dy dx
Z Z
=
y 1+x
ln |y | = ln |1 + x | + c1 = ln(e c1 |1 + x |)
y = ±e c1 (1 + x )

Replacing ±e c1 by a new constant c, we finally have

y = c(1 + x ) 6 / 13
Example (2) (Solution Curve)
Solve the initial-value problem
dy x
=− , y (4) = −3.
dx y
Solution
The equation can be written as y dy = −x dx . We get
Z Z
y dy = − x dx , ⇒ 12 y 2 = − 21 x 2 + c1 ⇒ x 2 + y 2 = 2c1

Substituting the initial condition

(x , y ) = (4, −3), we obtain the implicit so-
lution x 2 + y 2 = 25. √
Solving for y , we have y = ± 25 − x 2 .
Using the initial value
√ again, we finally get
the solution y = − 25 − x 2 , −5 < x < 5.

7 / 13
Example (3) (Initial-Value Problem)
dy
Solve the initial-value problem = 1 + y 2 , y (0) = 0.
dx
Solution
The equation can be written as dy /(1 + y 2 ) = dx . Then we can
commpute
dy
Z Z
= dx ⇒ tan−1 y = x + C ⇒ y = tan(x + C )
1 + y2
By the initial condition, we have 0 = tan C .Therefore C = 0 and
we have the solution y = tan x , −π/2 < x < π/2.

8 / 13
Losing a Solution
Example (4) (Losing a Solution)
Solve dy /dx = y 2 − 4.
Solution
We change the DE into a differential form and continue as follows:
h i
dy 1 1 1
y 2 −4
= dx ⇒ 4 y −2 − y +2 dy = dx ⇒
1 y −2
4 (ln |y − 2| − ln |y + 2|) = x + c1 ⇒ ln y +2 = 4x + c2 ⇒
!
y −2 1 + ce 4x
y +2 = ±e 4x +c2 = ce 4x ⇒ y = 2
1 − ce 4x

Going back to the DE, the zeros of y 2 − 4 give two constant

(equilibrium) solutions y = 2 and y = −2. The zero y = 2
corresponds to c = 0 in the family of solutions but y = −2 is a
singular solution. In order not to miss any solution, we must
consider constant solutions at the beginning.
9 / 13
Example (5) (An Initial-Value Problem)
Solve the initial-value problem
dy
cos x (e 2y − y ) = e y sin 2x , y (0) = 0
dx

Solution
We move dx to the other side and divide both sides by e y cos x .
Then
e 2y − y sin 2x
Z Z
y
dy = dx ⇒ (e y − ye −y )dy = 2 sin x dx
e cos x
⇒ e y − (−ye −y − e −y ) = −2 cos x + c ⇒ 2 = −2 + c

Finally we obtain the solution

e y + ye −y + e −y = 4 − 2 cos x

10 / 13
This figure shows graphs of the one-parameter family of solutions

e y + ye −y + e −y = −2 cos x + c.

The red curve corresponds to the solution for the initial value
y (0) = 0 with c = 4 and the blue curve corresponds to the
solution for the initial value y (π/2) = 0 with c = 2

11 / 13
An IVP with Infinitely Many Solutions
In Example (4) of §1.2, we saw the IVP
dy
= xy 1/2 , y (0) = 0.
dx
1 4
We had two solutions y = 16 x and y = 0.
We divide the equation by y 1/2 and then integrate y −1/2 dy = x dx .
 2
2y 1/2 = 21 x 2 + c1 ⇒ y = 1 2
4x +c
The trivial solution y = 0 cannot be obtained
from this one-parameter family of solutions, be-
cause the divison by y 1/2 eliminated the possi-
bility. In fact there are infinitely many solutions:
(
0, x <a
y=
(x 2 − a2 )2 /16, x ≥a

for any a ≥ 0.
12 / 13
Solutions Defined by Integrals
If g(x ) is continuous on an open interval containing a, the
d Rx
Fundamental Theorem of Calculus says dx a g(t)dt = g(x ). The
solution of an initial-value problem dy /dx = g(x ), y (x0 ) = y0 can
be given by Z x
y = y0 + g(t)dt
x0

Example (5) (An Initial-Value Problem)

2
Solve dy /dx = e −x , y (2) = 6.
Solution
2
The function g(x ) = e −x is continuous on (−∞, ∞) but its
antiderivative is not an elementary function. We integrate the
equation:
Z x Z x Z x
dy 2 2
dt = e −t dt ⇒ y (x ) − y (2) = e −t dt
2 dt 2
Z x 2
2
⇒ y (x ) = 6 + e −t dt, −∞ < x < ∞
2 13 / 13