CHAPTER 1

Introduction to
Differential Equations
 Chapter Contents
1.1 Definitions and Terminology
1.2 Initial-Value Problems
1.3 Differential Equations as Mathematical Methods

Lecturer: Prof. Hsin-Lung Wu Ch1_2

 1.1 Definitions and Terminology
Introduction: differential equations means that
equations contain derivatives, eg:
dy/dx = 0.2xy (1)
Definition 1.1.1 Differential equation
An equation contains the derivates of one or more
dependent variables with respect to one or more
independent variables (DE).

Ordinary DE: An equation contains only ordinary

derivates of one or more dependent variables of a
single independent variable.
eg: dy/dx + 5y = ex, (dx/dt) + (dy/dt) = 2x + y (2)
Lecturer: Prof. Hsin-Lung Wu Ch1_3
Partial DE: An equation contains partial derivates of
one or more dependent variables of tow or more
independent variables.
 2u  2u  2u  2u u (3)
  0,  2
x 2 y 2 x 2 t 2 t
Notations: Leibniz notation dy/dx, d2y/ dx2
prime notation y’, y”, …..
subscript notation ux, uy, uxx, uyy, uxy , ….
Order: highest order of derivatives
second order first order
 
3
d y  dx 
2
 5   4 y  e x

dx 2  dy 
Lecturer: Prof. Hsin-Lung Wu Ch1_4
General form of n-th order ODE:
F ( x, y, y' , , y ( n ) )  0 (4)
Normal form of (4)
dny ( n1) (5)
n
 f ( x , y , y ' ,  , y )
dx
eg: normal form of 4xy’ + y = x, is
y’ = (x – y)/4x
Linearity: An n-th order ODE is linear if F is linear
in y, y’, y”, …, y(n). It means when (4) is linear, we
have
dny d n1 y dy (6)
a ( x) a    a ( x)  a ( x) y  g ( x)
n1
dx n1
n 1 0
dx n dx
Lecturer: Prof. Hsin-Lung Wu Ch1_5
The following cases are for n = 1 and n = 2
dy
a1 ( x)  a0 ( x) y  g ( x) and
dx
d2y dy
a2 ( x) 2  a1 ( x)  a0 ( x) y  g ( x) (7)
dx dx
Two properties of a linear ODE:
(1) y, y’, y”, … are of the first degree.
(2) Coefficients a0, a1, …, are at most on x
Nonlinear examples:

(1  y ) y ' sin y y2

Lecturer: Prof. Hsin-Lung Wu Ch1_6

Definition 1.1.2 Solution of an ODE
Any function , defined on an interval I, possessing at
least n derivatives that are continuous on I, when
replaced into an n-th order ODE, reduces the equation
into an identity, it said to be a solution of the equation
on the interval.

That is, a solution of (4) is a function  possesses at

least n derivatives and
F(x, (x), ’(x), …, (n)(x)) = 0 for all x in I,
where I is the interval  is defined on.

Lecturer: Prof. Hsin-Lung Wu Ch1_7

 Example 1 Verification of a Solution

Verify the indicated function is a solution of the given

ODE on (－, )
1/2 4
(a) dy/dx = xy ; y = x /16 (b) y 
  2 y  y  0; y  xe x

Solution:
dy x3 x3
(a) Left-hand side:  4 
dx 16 4 1/ 2
 x 4
 x 2
x 3
Right-hand side: xy1/ 2  x     x  
then left = right  16  4 4
(b) Left-hand side:
y  2 y  y  ( xe x  2e x )  2( xe x  e x )  xe x  0
Right-hand side: 0
then left = right
Lecturer: Prof. Hsin-Lung Wu Ch1_8
Note: y = 0 is also the solution of example 1, called
trivial solution

Lecturer: Prof. Hsin-Lung Wu Ch1_9

 Example 2 Function vs. Solution

y = 1/x, is the solution of xy’ + y = 0, however, this

function is not differentiable at x = 0. So, the interval of
definition I is (－, 0), (0, ).

Fig 1.1.1 Ex 2 illustrates the difference between

the function y = 1/x and the solution y = 1/x
Lecturer: Prof. Hsin-Lung Wu Ch1_10
Explicit solution: dependent variable is expressed
solely in terms of independent variable and constants.
Eg: solution is y = (x).

Definition 1.3 Implicit solution of an ODE

G(x, y) = 0 is said to be an implicit solution of (4) on I,
provided there exists at least one function y = (x)
satisfying the relationship as well as the DE on I.

Lecturer: Prof. Hsin-Lung Wu Ch1_11

 Example 3 Verification of an Implicit
Solution
x2 + y2 = 25 is an implicit solution of
dy/dx = −x/y (8)
on the interval -5 < x < 5.
Since
dx2/dx + dy2/dx = (d/dx)(25)
then
2x + 2y(dy/dx) = 0 and dy/dx = -x/y
solution curve is shown in Fig 1.1.2

Lecturer: Prof. Hsin-Lung Wu Ch1_12

 Fig 1.1.2 An Implicit solution and two
explicit solutions in Ex 3

Lecturer: Prof. Hsin-Lung Wu Ch1_13

Families of solutions: A solution containing an
arbitrary constant represents a set G(x, y) = 0 of
solutions is called a one-parameter family of
solutions. A set G(x, y, c1, c2, …, cn) = 0 of solutions
is called a n-parameter family of solutions.
Particular solution: A solution free of arbitrary
parameters. eg: y = cx – x cos x is a solution of xy’ –
y = x2 sin x on (-, ), y = x cos x is a particular
solution according to c = 0. See Fig 1.1.3.

Lecturer: Prof. Hsin-Lung Wu Ch1_14

 Fig 1.1.3 Some solution of xy’-y=x2 sinx

Lecturer: Prof. Hsin-Lung Wu Ch1_15

 Example 4 Using Different Symbols

x = c1cos 4t and x = c2 sin 4t are solutions of

x + 16x = 0
we can easily verify that x = c1cos 4t + c2 sin 4t is also a
solution.

Lecturer: Prof. Hsin-Lung Wu Ch1_16

 Example 5 A piecewise-Defined Solution

We can verify y = cx4 is a solution of xy – 4y = 0 on

(-, ). See Fig 1.1.4(a).
The piecewise-defined function
 x 4 , x  0
y 4
 x , x0
is a particular solution where we choose c = −1 for x <
0 and c = 1 for x  0. See Fig 1.1.4(b).

Lecturer: Prof. Hsin-Lung Wu Ch1_17

 Fig 1.1.4 Some solution of xy’-4y=0 in Ex 5

Lecturer: Prof. Hsin-Lung Wu Ch1_18

Singular solution: A solution can not be obtained by
particularly setting any parameters.
y = (x2/4 + c)2 is the family solution of dy/dx = xy1/2 ,
however, y = 0 is a solution of the above DE.
We cannot set any value of c to obtain the solution y
= 0, so we call y = 0 is a singular solution.

Lecturer: Prof. Hsin-Lung Wu Ch1_19

System of DEs: two or more equations involving of
two or more unknown functions of a single
independent variable.
dx/dt = f(t, x, y)
dy/dt = g(t, x, y) (9)

Lecturer: Prof. Hsin-Lung Wu Ch1_20

 1.2 Initial-value Problems
Introduction
A solution y(x) of a DE satisfies an initial condition.
Example
On some interval
n
I containing xo,
solve: d y ( n1)
n
 f ( x , y , y ' ,  , y )
dx
subject to:
y ( x 0 )  y0 , y ' ( x0 )  y1 , , y ( n1) ( x0 )  yn1 (1)
This is called an Initial-Value Problem (IVP).
y(xo) = yo , y(xo) = y1 , y ( n1) ( x0 )  yn1
are called initial conditions.

Lecturer: Prof. Hsin-Lung Wu Ch1_21

First and Second IVPs
dy
solve :  f ( x, y ) (2)
dx
subject to : y ( x0 )  y0
and
d2y
solve : 2
 f ( x, y , y ' ) (3)
dx
subject to : y ( x0 )  y0 , y ' ( x0 )  y1

are first and second order initial-value problems,

respectively. See Fig 1.2.1 and 1.2.2.

Lecturer: Prof. Hsin-Lung Wu Ch1_22

Lecturer: Prof. Hsin-Lung Wu Ch1_23
 Example 1 First-Order IVPs
We know y = cex is the solutions
of y’ = y on (-, ). If y(0) = 3,
then 3 = ce0 = c. Thus y = 3ex is
a solution of this initial-value
problem
y’ = y, y(0) = 3.
If we want a solution pass
through (1, -2), that is y(1) = -2,
-2 = ce, or c = -2e-1. The
function y = -2ex-1 is a solution
of the initial-value problem
y’ = y, y(1) = -2.

Lecturer: Prof. Hsin-Lung Wu Ch1_24

 Example 2 Interval / of Definition of a
Solution
In Problem 6 of Exercise 2.2, we have the solution of
y’ + 2xy2 = 0 is y = 1/(x2 + c). If we impose y(0) = -1, it
gives c = -1. Consider the following distinctions.
1) As a function, the domain of y = 1/(x2 - 1) is the
set of all real numbers except x = -1 and 1. See Fig
1.2.4(a).
2) As a solution, the intervals of definition are
(-, 1), (-1, 1), (1, )
3) As a initial-value problem, y(0) = -1, the interval
of definition is (-1, 1). See Fig 1.2.4(b).

Lecturer: Prof. Hsin-Lung Wu Ch1_25

 Fig 1.2.4 Graph of the function and solution
of IVP in Ex 2

Lecturer: Prof. Hsin-Lung Wu Ch1_26

 Example 3 Second-Order IVP

In Example 4 of Sec 1.1, we saw x = c1cos 4t + c2sin 4t

is a solution of
x + 16x = 0
Find a solution of the following IVP:
x + 16x = 0, x(/2) = −2, x(/2) = 1 (4)

Solution:
Substitute x(/2) = − 2 into x = c1cos 4t + c2sin 4t, we
find c1 = −2. In the same manner, from x(/2) = 1 we
have c2 = ¼.

Lecturer: Prof. Hsin-Lung Wu Ch1_27

Existence and Uniqueness:

Does a solution of the IVP exist?

If a solution exists, is it unique?

Lecturer: Prof. Hsin-Lung Wu Ch1_28

 Example 4 An IVP Can Have Several
Solution
Since y = x4/16 and y = 0 satisfy the DE
dy/dx = xy1/2 , and also initial-value y(0) = 0, this DE
has at least two solutions, See Fig 1.2.5.

Lecturer: Prof. Hsin-Lung Wu Ch1_29

Theorem 1.2.1 Existence of a Unique Solution
Let R be the region defined by a  x  b, c  y  d that
contains the point (x0, y0) in its interior. If f(x, y) and
f/y are continuous in R, then there exists some
interval I0: (x0- h, x0 + h), h > 0, contained in
[a, b] and a unique function y(x) defined on I0 that
is a solution of the IVP (2).

Lecturer: Prof. Hsin-Lung Wu Ch1_30

 Fig 1.2.6 Rectangular region R

The geometry of Theorem 1.2.1 shows in Fig 1.2.6.

Lecturer: Prof. Hsin-Lung Wu Ch1_31

 Example 5 Example 3 Revisited

For the DE: dy/dx = xy1/2 , inspection of the functions

f x
f ( x, y )  xy1/ 2 and  1/ 2
y 2 y
we find they are continuous in y > 0. From Theorem
1.2.1, we conclude that through any point (x0, y0), y0 > 0,
there is some interval centered at x0 on which this DE
has a unique solution.

Lecturer: Prof. Hsin-Lung Wu Ch1_32

Interval of Existence / Uniqueness
Suppose y(x) is a solution of IVP (2), the following
sets may not be the same:
the domain of y(x),
the interval of definition of y(x) as a solution,
the interval I0 of existence and uniqueness.

Lecturer: Prof. Hsin-Lung Wu Ch1_33

 1.3 DEs as Mathematical Models
Introduction
Mathematical models are mathematical descriptions
of something.

Level of resolution
Make some reasonable assumptions about the system.

The steps of modeling process are as following.

Lecturer: Prof. Hsin-Lung Wu Ch1_34

 Express assumptions in terms Mathematics
Assumptions of differential equations formulation

If necessary,
alter assumptions Solve the DEs
or increase resolution
of the model

Check model Display model predictions, Obtain

Predictions e.g., graphically solution
with know
facts

Lecturer: Prof. Hsin-Lung Wu Ch1_35

Population Dynamics
If P(t) denotes the total population at time t, then
dP/dt  P or dP/dt = kP (1)
where k is a constant of proportionality, and k > 0.
Radioactive Decay
If A(t) denotes the substance remaining at time t, then
dA/dt  A or dA/dt = kA (2)
where k is a constant of proportionality, and k < 0.
A single DE can serve as a mathematical model for
many different phenomena.

Lecturer: Prof. Hsin-Lung Wu Ch1_36

Newton’s Law of Cooling/Warming
If T(t) denotes the temperature of a body at time t, Tm
the temperature of surrounding medium, then

dT/dt  T - Tm or dT/dt = k(T - Tm) (3)

where k is a constant of proportionality.

Lecturer: Prof. Hsin-Lung Wu Ch1_37

Spread of a Disease
If x(t) denotes the number of people who have got the
disease and y(t) the number of people who have not
yet, then
dx/dt = kxy (4)
where k is a constant of proportionality.
From the above description, suppose a small
community has a fixed population on n, If one
inflected person is introduced into this community,
we have x + y = n +1, and
dx/dt = kx(n+1-x) (5)

Lecturer: Prof. Hsin-Lung Wu Ch1_38

Chemical Reactions
Inspect the following equation
CH3Cl + NaOH  CH3OH + NaCl
Assume X is the amount of CH3OH,  and  are the
amount of the first two chemicals, then the rate of
formation is
dx/dt = k( - x)( - x) (6)

Lecturer: Prof. Hsin-Lung Wu Ch1_39

Mixtures
See Fig 1.3.1. If A(t) denotes the amount of salt in the
tank at time t, then

dA/dt = (input rate) – (output rate) = Rin - Rout (7)

We have Rin = 6 lb/min, Rout = A(t)/100 (lb/min), then

dA/dt = 6 – A/100 or dA/dt + A/100 = 6 (8)

Lecturer: Prof. Hsin-Lung Wu Ch1_40

 Fig 1.3.1 Mixing tank

Lecturer: Prof. Hsin-Lung Wu Ch1_41

Draining a Tank
Referring to Fig 1.3.2 and from Torricelli’s Law, if
V(t) denotes the volume of water in the tank at time t,
dV
  Ah 2 gh (9)
dt

From (9), since we have V(t) = Awh, then

dh Ah
 2 gh (10)
dt Aw

Lecturer: Prof. Hsin-Lung Wu Ch1_42

 Fig 1.3.2 Water draining from a tank

Lecturer: Prof. Hsin-Lung Wu Ch1_43

Series Circuits
Look at Fig 1.3.3.
From Kirchhoff’s second law, we have
d 2q dq 1
L 2  R  q  E (t ) (11)
dt dt C

where q(t) is the charge and dq(t)/dt = i(t), which is

the current.

Lecturer: Prof. Hsin-Lung Wu Ch1_44

 Fig 1.3.3 Current i(t) and charge q(t) are
measured in amperes (A) and coulumbs (C)

Lecturer: Prof. Hsin-Lung Wu Ch1_45

Falling Bodies
Look at Fig1.22.
From Newton’s law, we have
2 2
d s d s
m 2
 mg or 2
 g (12)
dt dt
Initial value problem

d 2s (13)
2
 g, s(0)  s0 , s' (0)  v0
dt

Lecturer: Prof. Hsin-Lung Wu Ch1_46

 Fig 1.3.4 Position of rock measured from
ground level

Lecturer: Prof. Hsin-Lung Wu Ch1_47

Falling Bodies and Air Resistance
From Fig 1.3.5.
We have the DE

dv (14)
m  mg  kv
dt
and can be written as
d 2s ds d 2s ds
m 2  mg  k or m 2  k  mg (15)
dt dt dt dt

Lecturer: Prof. Hsin-Lung Wu Ch1_48

 Fig 1.3.5 Falling body of mass m

Lecturer: Prof. Hsin-Lung Wu Ch1_49

A Slipping Chain
From Fig 1.3.6.
We have

L d 2 x d 2 x 64
2
 2 x or 2
 x0 (16)
32 dt dt L

Lecturer: Prof. Hsin-Lung Wu Ch1_50

 Fig 1.3.6 Chain slipping from frictionless peg

Lecturer: Prof. Hsin-Lung Wu Ch1_51

Suspended Cables
From Fig1.25.
We have
dy/dx = W/T1 (17)

Lecturer: Prof. Hsin-Lung Wu Ch1_52

 Fig 1.3.7 Cables suspended between vertical
supports

Lecturer: Prof. Hsin-Lung Wu Ch1_53

 Fig 1.3.8 Element of cable
Fig 1.3.8 explains the Element of cable.

Lecturer: Prof. Hsin-Lung Wu Ch1_54

 Evaluation
One midterm 40%
Final 40%
Quiz & Homework 20%

Lecturer: Prof. Hsin-Lung Wu Ch1_55

 Some information
Textbook: Advanced Engineering Mathematics 4th-ed
by Zill and Wright

Teaching Website: Please link 數位學苑

Lecturer: Prof. Hsin-Lung Wu Ch1_56