Chapter 1
Differential Equations Introductions
1.1. Some Basic Mathematical Models
1.2. Numerical Approximations
1.3. Classification of Differential
Equations
10/28/15
NGUYN TH HOI
1.1 Some basic mathematical models
Many of the principles, or laws of the natural world are
statements or relations involving rates at which things happen.
In mathematical terms, the relations are equations and the rate
are derivatives.
Equations containing derivatives are differential equations.
A differential equation that describes some physical process is
often called a mathematical model of the process.
10/28/15
NGUYN TH HOI
1.1 Some basic mathematical models
Problem 1 (A Falling Object) Suppose that an object is falling in the
atmosphere near sea level. Formulate a differential equation that
describes the motion.
Solution. Let us use t to denote time, v to represent the velocity of
the falling object. The velocity changes with time, so v is a function of t,
here t is the independent variable and v is the dependent variable. Let
us measure time t in second (s) and velocity in meters/second(m/s). We
will assume that v is positive in the downward direction.
10/28/15
NGUYN TH HOI
1.1 Some basic mathematical models
The physical law that describes the motion of everyday
objects is Newtons second law
F = ma = m (dv/dt),
here m is the mass of the object measuring in kilograms (kg),
a is its acceleration which is measured in meters/second 2
(m/s2) and F is the net force exerted on the object which is
measured in Newton (N).
10/28/15
NGUYN TH HOI
1.1 Some basic mathematical models
Consider the two forces that act on the object as it falls. The
first one is gravity force, that equals to mg, where g is the
acceleration due to gravity and the second one is drag force
that has the value v, where is a constant called the drag
coefficient (kg/s). Thus gravity acts in the downward direction
and drag acts in the upward direction, so the net force F exerted
on the falling object is
F = mg v m(dv/dt) = mg v .
10/28/15
NGUYN TH HOI
1.1 Some basic mathematical models
The last equation is a mathematical model of an falling
object in the atmosphere near sea level. This model contains
the three constants m, g, and . The constants m and
depend on the particular object, while g is the same for all
objects.
v
m
mg
10/28/15
NGUYN TH HOI
1.1 Some basic mathematical models
Problem 2 (Heating and Cooling) Consider a building as a partly
insulated box that is subject to external temperature fluctuations.
Construct a model that describes the temperature fluctuations inside
the building.
Solution Let u(t) and T(t) be the internal and external
temperatures, respectively, at time t. From the Newtons law of
cooling, which states that the rate of change of u(t) is proportional to
the temperature difference u(t) T(t), we obtain the differential
equation
10/28/15
NGUYN TH HOI
1.1 Some basic mathematical models
du/dt = k(u(t) T(t)),
where k is a positive constant, the minus sign in the righthand side indicated that du/dt must be negative if u(t) > T(t).
Let us suppose that u and T are measured in degrees
Fahrenheit and that t is measured in days. Then k must have
the units of 1/day.
10/28/15
NGUYN TH HOI
1.1 Some basic mathematical models
Constructing Mathematical Models
1. Identify the independent and dependent variables and
assign letters to represent them.
2. Choose the units of measurement for each variable.
3. Relate the basic principle with investigating problems.
10/28/15
NGUYN TH HOI
1.1 Some basic mathematical models
4. Express the principle or law in step 3 in terms of the variable
chosen in step 1. It may require the introduction of physical
constants or parameters (such as drag coefficient in P. 1) and
the determination of appropriate values for them.
5. Make sure that each term in the equation has the same
physical units.
6. The result of step 4 is either a single differential equation or a
system of several differential equations.
10/28/15
NGUYN TH HOI
10
1.1 Some basic mathematical models
Definition An equation of the form
F (x, y, y, y, y(n)) = 0,
(1)
where x as an independent variable; y = y(x) is an unknown
function and y, y, , y(n) are derivatives of this function, is
called a differential equation.
The order of the highest derivative of y in (1) is called the
order of differential equation.
10/28/15
NGUYN TH HOI
11
1.1 Some basic mathematical models
Example yy = exyy2sinx - first order differential equation,
y + 2(y)3 = 1 - second order differential equation,
(1) - n-th order differential equation.
10/28/15
NGUYN TH HOI
12
1.1 Some basic mathematical models
The equation (1) is called a linear differential equation if F
is a linear function of the variables y, y, y, , y(n) . The
general linear differential equation of order n is
a0(x) y(n) + a1(x) y(n-1) ++ an(x)y = g(x),
where a0(x), a1(x), an(x), g(x) are given functions.
A solution of differential equation (1) is a function (x) such
that , , (n) exist and satisfy
F(x, , , , (n)) = 0.
10/28/15
NGUYN TH HOI
13
1.1 Some basic mathematical models
The geometrical representation of a solution of (1) is called
the integral curve. Each integral curve is defined by the
formula y = (x) or (x,y) = 0 or by parameterization
equation x = x(t), y = y(t).
10/28/15
NGUYN TH HOI
14
1.1 Some basic mathematical models
Direction Fields
A slope field (or direction field) is a graphical
representation of the solutions of a first-order differential
equation. It is useful because it can be created without solving
the differential equation analytically. The representation may
be used to qualitatively visualize solutions, or to numerically
approximate them.
10/28/15
NGUYN TH HOI
15
1.1 Some basic mathematical models
How to plot the direction fields
Consider the first order general differential equations of the form
dy
f (t , y )
dt
The direction field of this equation is an infinite set of line
segments, passing through a point P(t,y) in ty plane and whose
slope is the value of f at that point (thus each line segment in the
direction field is tangent to the graph of the solution passing
through that point).
10/28/15
NGUYN TH HOI
16
1.1 Some basic mathematical models
From the direction field for a differential equation, we can
Sketch of solutions. Since the arrows in the direction
fields are tangents to the solutions to the DEs we can use
these as guides to sketch the graphs of solutions to the
differential equation.
Long Term Behavior. Direction fields can be used to find
information about this long term behavior of the solution (how
the solutions behave as t increases).
10/28/15
NGUYN TH HOI
17
1.1 Some basic mathematical models
Example 1. Sketch the direction field for the following
differential equation. Sketch the set of integral curves for this
differential equation.
y' y x
10/28/15
NGUYN TH HOI
18
1.2 Numerical Approximations
Many DEs do not have solutions that can be expressed in
term of simple functions. Just as there is no formula in terms
of a and b for the integral
b
e
a
t 2
dt
there is generally no way to solve DEs exactly. For this reason,
we must rely on numerical methods that produce
approximations to the desired solutions.
10/28/15
NGUYN TH HOI
19
1.2 Numerical Approximations
We will describe four families of methods for numerically
solving some DEs:
1) The Taylor series (TS) method,
2) Linear multistep methods (LMMs),
3) The Runge Kutta (RK) methods and
4) Finite difference methods for boundary value
problems.
10/28/15
NGUYN TH HOI
20
1.2 Numerical Approximations
We aim to solve all initial value problems (IVPs) of the form
dy
dt f (t , y ),
y (t ) y .
0
0
(1)
that posses a unique solution on some specified interval.
The first three methods can all be interpreted as
generalizations of the worlds simplest method: Eulers method.
In this chapter we will start with Eulers method.
10/28/15
NGUYN TH HOI
21
1.2 Numerical Approximations
Eulers method
We use h to refer to a small positive number called the
step size: we will seek approximations to the solution of the
IVP (1) at particular time t1 = t0 + h, t2 = t0 + 2h, , tn = t0 +
nh, i.e., approximations to the sequence of numbers (t 1),
(t2), , (tn), , rather than an approximation to the
solution y = (t) of the IVP (1).
10/28/15
NGUYN TH HOI
22
1.2 Numerical Approximations
Step 1. Start with the point P0(t0, y0), the equation of the
tangent line to the graph of the solution y = (t) of the IVP (1)
at the initial point (t0, y0) is
y y0 f (t 0 , y 0 )(t t 0 )
the approximation value of (t1) = (t0 + h) is
y1 y 0 f (t 0 , y 0 )(t1 t 0 ) y 0 f (t 0 , y 0 )h
10/28/15
NGUYN TH HOI
23
1.2 Numerical Approximations
Step 2. Continue with the point P1(t1, y1), the equation of
the tangent line to the graph of the solution of the DE in (1)
that passes through (t1, y1) is
y y1 f (t1 , y1 )(t t1 )
the approximation value of (t2) = (t0 + 2h) is
y 2 y1 f (t1 , y1 )(t 2 t1 ) y1 f (t1 , y1 )h
10/28/15
NGUYN TH HOI
24
1.2 Numerical Approximations
Continue this process we can obtain the approximation value
of the solution (t) of the IVP (1) at the moment t = tn+1
according to the formula
(t n1 ) y n 1 y n f (t n , y n )(t n 1 t n ) y n f (t n , y n )h
y y n f (t n , y n )(t t n )
(t n 1 ) y n1 y n f (t n , y n )(t n 1 t n )
10/28/15
NGUYN TH HOI
25
1.2 Numerical Approximations
Example 1. Use a step size h = 0.3 to develop an
approximate solution to the IVP
y ' (t ) (1 2t ) y (t ),
y ( 0) 1
t 0
over the interval 0 t 0.9.
Solution The exact solution of this IVP problem is
1 1
y (t ) exp t
4 2
10/28/15
NGUYN TH HOI
26
1.2 Numerical Approximations
At t = 0, the equation of the tangent line to the solution curve
passing through (0, 1) is y = 1 + t, hence (0.3) 1.3
At t = 0.3, the equation of the tangent line to the solution
curve passing through (0.3, 1.3) is y = 0.52t + 1.144, hence
(0.6) 1.456
At t = 0.6, the equation of the tangent line to the solution
curve passing through (0.6, 1.456) is y = 0.2912t + 1.63072,
hence (0.9) 1.36864
10/28/15
NGUYN TH HOI
27
1.2 Numerical Approximations
The results of computing the numerical solutions over the
interval [0,3] with h = 0.3, 0.15, and 0.075
10/28/15
NGUYN TH HOI
28
1.2 Numerical Approximations
h
yn
Global errors (GEs)
GE/h
0.3
y3 = 1.3686
(0.9) y3 = 0.2745
0.91
0.15
y6 = 1.2267
(0.9) y6 = 0.1325
0.89
0.075
y12 = 1.1591
(0.9) y12 = 0.0649 0.86
Exact
(0.9) =
1.0942
Numerical
results at t = 0.9 with h = 0.3, 0.15 and 0.075
Here global error (GE) at t = tn is given by
en (t n ) y n
10/28/15
NGUYN TH HOI
29
1.2 Numerical Approximations
Analysing the Numbers
1. Twice as many steps are needed to find the solution at t = 3.
2. The computed points lie closer to the exact solution curve.
This illustrates the notion that the numerical solution
converges to the exact solution as h 0.
3. It is seen from the table that halving h results in the error
being approximately halved. This suggests that the GE is
proportional to h.
10/28/15
NGUYN TH HOI
30
1.2 Numerical Approximations
The final column in the table suggests that the constant of
proportionality in this case is about 0.9 : en 0.9h, when
nh = 0.9.
If we require that |en| < 0.0005 then h < 0.0005/0.9
0.00055. Consequently, the integration to t = 0.9 requires
about n = 0.9/h 1620 steps.
10/28/15
NGUYN TH HOI
31
1.3 Classification of DEs
Classification of Differential Equations
1. From the number of independent variables in unknown
function, we have an ordinary differential equation
(ODE) and a partial differential equation (PDE).
2. From the number of unknown functions, we need a single
equation or a system of equations.
10/28/15
NGUYN TH HOI
32
1.3 Classification of DEs
3. The order of a differential equation is the order of the
highest derivative that appears in the equation. We have the
first order differential equation, the second order
differential equation, etc.
4. If F is a linear function of y, y, y, y(n) , then the ordinary
differential equation (1) is said to be linear, otherwise (1) is a
nonlinear equation.
10/28/15
NGUYN TH HOI
33
1.3 Classification of DEs
Some Important Questions
1. The existence of a solution?
2. The uniqueness of solution?
3. How can we determine a solution?
10/28/15
NGUYN TH HOI
34
1.3 Classification of DEs
Consider the first order differential equation of the form
dy
f ( x, y ),
dx
there are some important subclasses of this equation that can
be solve analytically
1. Linear equation;
2. Separable equation;
3. Homogeneous equation;
4. Autonomous equation;
5. Exact equation;
10/28/15
NGUYN TH HOI
35
Summary
S1. Differential Equations, Direction Fields, Some
Mathematical Models.
S2. Eulers Method for Approximating the Values of
Exact Solutions at Some Moment.
S3. Classification of DEs: from the number of
independent variables, from the number of dependent
variables, from the order of highest derivative, from
the linearity of given equations .
10/28/15
NGUYN TH HOI
36
