Professional Documents
Culture Documents
Contents
1 First-Order Dierential Equations
1.1
1.2
1.3
Existence-Uniqueness Theorem
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.4
First-Order: Separable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.5
First-Order: Linear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.6
Substitution Methods
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.6.1
Bernoulli Equations
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.6.2
Homogeneous Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.7
. . . . . . . . . . . . . . . . . . . . . . . . . .
1.8
10
1.9
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
1.9.1
Problems
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
1.9.2
Solutions
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11
13
15
16
16
17
The reason for this is that most anything interesting involves change of some kind, and the derivative
expresses the rate of change.
Thus, anything that can be measured numerically with values that change in some known manner will
give rise to a dierential equation.
Example: The populations of several species in an ecosystem which aect one another in some way.
Example: The position, velocity, and acceleration of a physical object which has external forces acting
on it.
Here are some examples of single dierential equations and systems of dierential equations, with and without
additional conditions.
Example:
Answer:
Example:
y(x) = Cex
00
y + 2y + y = 3x
Answer:
Example:
y 0 + y = 0.
, with
C.
y(0) = y 0 (0) = 1
f 00 f = (f 0 )2 ,
Answer:
with
f (1) = f 0 (1) = 1.
f (x) = ex1 .
Example:
f 0 = 2f g
and
f (s, t) = st
and
f (s, t) =
1 2 1 2
s + t .
2
2
Most dierential equations (very much unlike the carefully chosen ones above) are dicult if not impossible
to nd exact solutions to, in the same way that most random integrals or innite series are hard to evaluate
exactly.
0
Prototypical example:
However, it is generally possible to nd very accurate approximate solutions using numerical techniques
or by using Taylor series.
In this course we will only cover how to solve some of the more basic types of equations and systems:
rst-order separable, linear, and exact equations; higher-order linear equations with constant coecients;
and systems of rst-order linear equations with constant coecients.
is a function only of
x)
then it
We will only talk about ODEs in this course. But for completeness, dierential equations involving functions of several variables are called partial dierential equations, or PDEs. (Recall that the derivatives
of functions of more than one variable are called partial derivatives, hence the name.)
PDEs, obviously, arise when functions depend on more than one variable. They occur often in physics
(with functions that depend on space and time) and economics (with functions that depend on time and
other parameters).
An
nth
y 0 + xy = 3x2
00
0000
y +y +y =0
e =y
and
y0 y = 2
y2 ,
or
(y 0 )3 ,
or
derivative.
are rst-order.
is second-order.
is fourth-order.
nth
y y0 ,
or
ln(y),
or
y 0 + xy = 3x2
y 0 y = 3x2 , x2 + (y 0 )2 = 1,
and
ey .
y 00 + y 0 + y = 0
and
are linear.
y 00 = sin(y)
1.2
dP
= k P,
dt
where
P (t)
and
It's not hard to see that one population model that works is
P (t) = ekt
More complicated example: The Happy Sunshine Valley is home to Cute Mice and Adorable Kittens. The
Cute Mice grow at a rate proportional to their population, minus the number of Mice that are eaten by their
predators, the Kittens. The population of Adorable Kittens grows proportional to the number of mice (since
they have to catch Mice to survive and reproduce).
dK
dM
= k1 M k2 K , and
= k3 M , where M (t) and K(t) are the populations
dt
dt
and k1 , k2 , k3 are some constants.
This is a system of two linear dierential equations; we will learn how to solve a system like this later in
the course.
The conditions here are fairly natural for a simple predator-prey system.
be non-linear terms too perhaps when two Kittens meet, they ght with each other and cause injury,
which might change the equation to
dK
= k3 M k4 K 2 .
dt
This system would get even more dicult if we added additional species each of which interacts in some
way with the others.
Non-linear example: A simple pendulum consists of a weight suspended on a string, with gravity the only
force acting on the weight. If
is the angle the pendulum's string makes with a vertical line, then horizontal
d2
= k sin().
dt2
1.3
sin().
d2
= k ,
dt2
Existence-Uniqueness Theorem
Before we discuss how to solve various classes of rst-order dierential equations, we would like to know when
(in general) we can say that a rst-order dierential solution has an equation, and when we can say that
solution is unique.
Theorem (Existence-Uniqueness):
a)
if the function
with
y(a) = b
y 0 = f (x, y)
a)
f
y
(a, b).
The
is continuous on a
(a, b).
The proof of the theorem is fairly dicult. The general idea of one proof of the theorem is to construct
a sequence of functions (dened on some small interval around
a),
The continuity of
of
ensures that the sequence will converge; one way of doing this is to use the continuity
to show that the functions far out in the sequence eventually become very close together.
f
y
f
y
Example:
a.
y(a) = b
y 0 = ey + xy
is
In fact,
f (x, y) = ey + xy
is continuous everywhere.
all initial conditions lead to a unique solution , because the partial derivative
fy (x, y) = ey + x
It is, in fact, not possible to solve this equation explicitly using any of the techniques we will learn.
Nonetheless, the theorem guarantees that it has a unique solution!
is guaranteed
2 1/3
y
3
(a, 0) for
fy =
zero at
x = 0,
y=0
is continuous everywhere.
a,
but
1 3
y=
x .
27
y = 0,
unique otherwise .
In fact, we can even write down two dierent solutions to the IVP
constant function
f (x, y) = y 2/3
y 0 = y 1/3
with
y(0) = 0:
namely, the
y(a) = b
y0 =
yx
is guar-
f (x, y) =
1
fy (x, y) =
2 yx
(a, b)
with
is not dened if
b>a
xy
Note: If we decide that we are not afraid of complex numbers, then the function
dened and continuous on the entire plane (although it will take non-real values
this convention, the solution is guaranteed to exist for all
(a, b)
1.4
(a, b)
with
b>a.
f (x, y) = y x is
if x > y ). So with
unique for
by zero is still
First-Order: Separable
A separable equation is of the form
Step 1:
Replace
(including
dy
)
dx
y0
with
dy
,
dx
x-stu
on the other.
y -stu
on one side
x).
only.
+C
dy
dx
dx
to get
dy
and an integral
x-terms.
Step 3: If given, plug in the initial condition to solve for the constant
C.
it is.)
1.5
Example: Solve
as a function of
y0 = k y,
where
x,
if possible.
is some constant.
1 dy
= k.
y dx
1 dy
to get
dx = k dx,
y dx
Step 1: Rewrite as
Step 2: Integrate
Evaluate to get
ln(y) = kx + C .
dy
= ex .
dx
y dy
get
e
dx = ex dx.
dx
ey
Step 2: Integrate to
dy = k dx.
y
y 0 = exy .
Step 1: Rewrite as
Example: Find
or
y = ekx+C = C ekx
given that
y 0 = x + xy 2
and
y = ln(ex + C)
ey = ex + C .
y(0) = 1.
dy
= x dx.
1 + y2
dy
1
dx = x dx,
to get
1 + y 2 dx
Step 1: Rewrite as
Step 2: Integrate
tan1 (y) =
1 2
x + C.
2
First-Order: Linear
The general form for a rst-order linear dierential equation is (upon dividing by the coecient of
y + P (x) y = Q(x),
where
P (x)
and
Q(x)
y0 )
given
x.
We would really like it if we could just integrate both sides to solve the equation. However, in general, we
cannot: the
by
y0
P (x) y
I(x)
When written this way, this sum looks sort of like the output of the product rule. If we can nd
that the derivative of
What we want is
and the solution
I(x)
is
I(x)P (x),
0
I(x)P (x) =
I (x). This
P (x) dx
is I(x) = e
.
I(x)
so
d
[I(x) y].
dx
Motivated by the above logic, here is the method for solving rst-order linear equations:
I(x),
y 0 + P (x) y = Q(x).
Step
2: Multiply both sides by the integrating factor
P (x) dx
P (x) dx
to get
P (x) dx 0
y +e
P (x) dx
P (x) y =
Q(x).
C.
d h P (x) dx i
e
y ,
dx
Step 4: If given, plug in the initial condition to solve for the constant
C.
it is.)
Example: Find
as a function of
given that
x.
y 0 + 2xy = x
Step 1: We have
Step 3:
Step 4:
Step 5:
P (x) = 2x
y(0) = 1.
and
Q(x) = x.
and
ex y 0 + ex 2x y = x ex
1 x2
x2
e + C.
Taking the antiderivative on both sides yields e y =
2
1
1 0
0
e + C hence C = .
Plugging in yields e 1 =
2
2
1 1 x2
Solving for y gives y =
.
+ e
2 2
P (x) dx
for which
Step 1: We have
Step 2:
Step 3:
Step 5:
Example: Find
y0 +
4
y = x3 ,
x
= ex
to get
xy 0 = x4 4y .
4
3
and Q(x) = x .
x
P (x) dx
Multiply both sides by e
= e4 ln(x) = x4 to get x4 y 0 + 4x3 y = x7 ,
1 8
4
x + C.
Taking the antiderivative on both sides yields x y =
8
1 4
Solving for y gives y =
x + C x4 .
8
1.6
given that
Step 1: We have
so
P (x) =
y 0 cot(x) = y + 2 cos(x)
y 0 y tan(x) = 2 sin(x),
P (x) dx
with
gives
=e
1
1
= 1+C
2
2
y=
cos(2x)
2 cos(x)
y(0) =
1
.
2
P (x) = tan(x)
Q(x) = 2 sin(x).
and
ln(cos(x))
and
hence
C = 0.
Substitution Methods
Just like with integration, sometimes we come across dierential equations which we cannot obviously solve,
but which, if we change variables, will turn into a form we know how to solve.
Determining what substitutions to try is a matter of practice, in much the same way as in integral calculus.
In general, there are two kinds of substitutions: obvious ones that arise from the form of the dierential
equation, and formulaic ones which are standard substitutions to use if a dierential equation has a particular
form.
Step 2: Find
dv
dx
in terms of
y0 , y,
in terms of
and
v.
and
x.
as a dierential equation in
v.
(The hope is, after making the substitution, the new equation is in
y.
y 0 = (x + y)2 .
This equation is not linear, nor is it separable as written. The obstruction is that the term
both
and
Step 2:
Step 3:
x+y
involves
y.
v = x + y.
dv
dy
0
0
=1+
, so y = v 1.
Dierentiating yields
dx
dx
0
2
0
2
The new equation is v 1 = v , or v = v + 1.
v
dv
= 1 dx,
+1
v2
so that
tan1 (v) = x + C ,
v = tan(x + C).
y = tan(x + C) x
As with rst-order linear equations, sometimes Bernoulli equations can be hidden in a slightly dierent
form.
So we have
For
Substituting in to the original equation then yields the rst-order linear equation
v = y 1n
we have
for
v 0 = (1 n)y n y 0 .
y 0 + 2xy = xy 3 .
P (x) = 2x, Q(x) = x,
v 0 4xv = 2x.
v = y 2
I(x) = e
4x dx
2x2 0
2x2
gives
v 0 + (1 n)P (x) v =
v.
(1 n) Q(x)
(1 n) y
v = y 1n .
y=
n = 3.
.
2
v = 2xe2x .
1
2
2
e2x v = e2x + C ,
2
v 4xe
2
1
+ Ce2x
2
= e2x
and
so that
v=
1/2
.
y 2 y 0 = ex y 3 .
y 3 and then divide
P (x) = 1, Q(x) = ex , and
The equation as written is not of Bernoulli type. However, if to both sides we add
by y , we
n = 2.
1
2
+ Ce2x .
2
x 2
y +y = e y
v = y3
v 0 + 3v = 3ex .
I(x) = e
3 dx
= e3x ,
y0 = f
y
x
v=
y = vx.
Then dierentiating
y = vx
shows
y 0 = v + xv 0 , hence
1
v0
= .
f (v) v
x
y
,
x
or equivalently to set
v + xv 0 = f (v),
which is
2x2 y 0 = x2 + y 2 .
This equation is not separable nor linear, and it is not a Bernoulli equation. If we divide both sides by
2x2
then we obtain
y0 =
1 1 y 2
,
+
2 2 x
which is homogeneous.
1 2
1
2v 0
1
v v + , and rearranging gives
= .
2
2
(v 1)2
x
2dv
1
2
2
Then integrating yields
=
dx, so
= ln(x)+C . Solving for v gives v = 1
,
2
(v 1)
x
v1
ln(x) + C
2x
.
so y = x
ln(x) + C
Setting
v = y/x
xv 0 =
y0 =
x2 + y 2
.
xy
x2 ,
we obtain
y0 =
1 + (y/x)2
,
(y/x)
which is
homogeneous.
Setting
1.7
v = y/x
y=x
yields
1 + v2
1
v = .
v
v
1
yields
v dv =
dx,
x
xv 0 =
2 ln(x) + C
so that
1 2
v = ln(x) + C ,
2
so
v=
p
2 ln(x) + C
and
My
with respect to
y,
namely
M
,
y
functions.
The equation
M (x, y) + N (x, y) y 0 = 0
and
y.
M (x, y) dx + N (x, y) dy = 0.
In this
that
M y = Nx .
we are not necessarily out of luck like with rst-order linear equations, there may exist an integrating factor
I(x, y)
which we can multiply the equation by, in order to make the equation exact.
Unfortunately, we don't really get much for free: trying to solve for the integrating factor is often as hard
as solving the original equation. Finding
I (My Nx ) = 0,
I(x, y),
which is just as tricky to solve as the original equation. Only in a few special cases
I
I
M
N +
y
x
I(x, y).
Case 1: Suppose we want to see if there exists an integrating factor that depends only on
on
y ).
I
y
Then
I0
M y Nx
=
. This can
I
N
P (x) dx
then I(x) = e
.
M y Nx
N
(and not
and so
I(x)
is a function
P (x)
only of
y,
(and not
y );
The form of this integrating factor should look familiar it is the same as the one from a rst-order
linear equation. There is a very good reason for this; namely, a rst-order linear equation is a special
case of this form of equation.
Case 2: We could also look to see if there is an integrating factor that depends only on
We can do the same calculation, this time using
the ratio
Nx M y
M
is a function
Q(y)
only of
I
= 0,
x
(and not
and not on
x.
x);
then
I(y) = e
Q(y) dy
Remark: There is no really good reason only to consider these cases, aside from the fact that they're the
easiest. We could just as well try to look for integrating factors that are a function of the variable
Or of
v = x/y .
Or of
w = y + ln(x).
t = xy .
In each case we'd end up with some other kind of condition. But
we won't think about those things we really just care about the two kinds of integrating factors above.
y(x),
if
There is no obvious substitution to make, and it is not separable, linear, homogeneous, or Bernoulli. So
we must check for exactness.
Therefore we have
So we want to nd
My = 8y
and
Nx = 8y .
Therefore,
M = 4y 2 + 2x
and
N = 8xy .
y(x),
if
4xy 2 + x2 = C
There is no obvious substitution to make, and it is not separable, linear, homogeneous, or Bernoulli. So
we must check for exactness.
N = 3x y 8x.
Therefore,
M = 2xy 2 4y
and
Therefore we have
My = 4xy 4
and
Nx = 6xy 8.
M y Nx
2xy + 4
= 2
is not a function of x only.
N
3x y 8x
2xy 4
1
Nx M y
=
= is a function of y only.
Second, we have
M
2xy2 4y
y
(1/y) dy
by the integrating factor I(y) = e
= y.
First, we have
with
f 0 (y) = 0.
1.8
x2 y 3 4xy 2 = C
Note: It is not really necessary to check ahead of time whether the equation is separable or a rst-order
linear equation. Separable equations are exact, and rst-order linear equations can be made exact after
multiplying by an integrating factor, which will be detected using the
M y Nx
N
two special types at the beginning only because it's faster to solve it using the usual methods.
y 0 = f (x, y)
and
M (x, y) + N (x, y) y 0 = 0
and
I(x) = e
P (x) dx
y 0 = f (x) g(y)
y -terms and x-terms on opposite sides of the equation and then take the antiderivative of both sides.
y 0 = f (x, y)
form).
Step 2a: Check to see if there is any 'obvious' substitution that would simplify the equation.
Step 2b: Check to see if the equation is of Bernoulli type namely, of the form
(1 n) y
and then make the
dv
+ (1 n)P (x) v = (1 n) Q(x).
dx
substitution
y 0 +P (x)y = Q(x)y n .
v = y 1n to obtain a
y
y0 = F
for some
x
y
dv
function F . If so, make the substitution v =
to obtain a separable equation x
= F (v) v .
x
dx
Step 3: If the equation is not of a special type, use the M (x, y) + N (x, y) y 0 = 0 form to nd the partial
derivatives My and Nx .
Step 2c: Check to see if the equation is homogeneous namely, of the form
Step 4: If
factor
My = Nx ,
Step 3a:
I(x) = e
My 6= Nx ,
Step 3b:
I(y) = e
Compute
P (x) dx
If it is a function
P (x)
only of
x,
Nx My
.
M
If it is a function
Q(y)
only of
y,
Compute
Q(y) dy
M y Nx
.
N
If neither of these methods works, you're out of luck unless you can nd an integrating factor some
other way.
Step 5:
solutions as
1.9
F (x, y)
with
Fx = M
and
Fy = N ,
F (x, y) = C .
1.9.1 Problems
xy 0 = y +
y0 =
y 2xy 2
.
3x2 y 2x
xy 0 = y + x.
y 0 1 = y 2 + x3 + x3 y 2 .
y0 =
y 0 = xy 3 6xy .
y0 =
xy .
4x3 y 2 + y
.
2x4 y + x
2xy + 2x
.
x2 + 1
1.9.2 Solutions
xy 0 = y +
xy .
y
y = +
x
0
y
x
(y
and
xy) + xy 0 = 0.
Setting
Solving for
v = y/x
(with
gives
y = vx
and
y 0 = xv 0 + v )
y
y = +
x
0
yields
dv
1
=
dx
x
v
2
ln(x)
y=x
+C
.
2
y
is a
x
xv 0 = v .
hence
homogeneous equation.
2v 1/2 = ln(x) + C ,
so
v=
ln(x)
+C
2
2
.
Note: The equation is also of Bernoulli type, and could be solved that way too. Of course, it will give
the same answer.
y0 =
y 2xy 2
.
3x2 y 2x
Step 3: We have
M = 2xy 2 y
and
N = 3x2 y 2x
so
My = 4xy 1
and
Nx = 6xy 2.
My 6= Nx .
We have
2xy 1
1
Nx M y
=
= ,
2
M
2xy y
y
M y Nx
2xy + 1
= 2
,
N
3x y 2x
1
dy
y
= y.
I(y) = e
with respect to
gives
x y xy 2 = C
2 3
solutions are
xy 0 = y +
x.
y0 =
y
1
+
x
x
and
(y
x) + xy 0 = 0.
linear.
y 0 (x1 )y = x1/2 .
1
I(x) = e
= e ln(x) = eln(x
dx
= x1 .
x1 y 0 x2 y = x3/2 .
1
x1 y = x1/2 + C ,
2
so
1
y = x1/2 + Cx
2
y 0 1 = y 2 + x3 + x3 y 2 .
y 0 = (y 2 + 1)(x3 + 1).
This equation is
separable.
Separating it gives
Integrating yields
y0
= x3 + 1 .
y2 + 1
3
x4 x2
dy
1
=
(x
+
1)
dx
, so tan
(y)
=
+ + C.
y2 + 1
4
2
y0 =
Then
y = tan
x4
x2
+
+C
4
2
.
(4x3 y 2 + y)
.
(2x4 y + x)
(4x3 y 2 + y) + (2x4 y + x) y 0 = 0.
N = 2x4 y + x
Step 3: We have
M = 4x3 y 2 + y
f (y) = 0.
and
so
My = 8x3 y + 1
and
Nx = 8x3 y + 1.
M y = Nx .
with respect to
gives
4 2
x y + xy = C
F (x, y) = x4 y + xy + f (y),
and checking
y 0 = xy 3 6xy .
v = y 2
2y 3
to get
2y 3 y 0
with
(12/x) dx
y 0 + 6xy = xy 3 .
12 2
y = 2x.
x
v 0 = 2y 3 y 0
v0
12
v = 2x.
x
1 10
1
x +C , so v = x2 +Cx12 and y =
5
5
1 2
x + Cx12
5
1/2
.
y0 =
2xy + 2x
.
x2 + 1
(method #1)
Exponentiating yields
2x
dy
=
dx,
y+1
x2 + 1
y + 1 = e ln(x
+1)+C
C
,
x2 + 1
so that
y=
so
y0 =
2x
(y + 1).
+1
x2
ln(y + 1) = ln(x2 + 1) + C .
C
1
x2 + 1
(method #2)
Step 3: We have
M = 2xy + 2x
Step 4: We have
M y = Nx ,
f (y) = 1
so
and
N = x2 + 1
so
My = 2x
and
Nx = 2x.
f (y) = y .
with respect to
gives
F (x, y) = x2 y + x2 + f (y),
x y+x +y =C
and checking
other solution.
1.10
dy
= f (y)
dt
f.
f (y)
equation explicitly.
Nonetheless, would like to be able to say something about what the solutions look like, without actually
solving the equation. Happily, this is possible.
An equilibrium solution, also called a steady state solution or a critical point, is a solution of the form
for some constant
Clearly, if
c.
y(t)
y(t) = c,
is constant, then
to an autonomous equation
solve
f (y) = 0.
hard.)
A solution
A solution
An equilibrium solution
y=c
y = c is stable from below if when we solve y 0 = f (y) with the initial condition y(0) = c for
some small but positive , the solution y(t) moves toward c as t increases. This statement is equivalent
to f (c ) > 0.
y = c is unstable
A solution
,
the solution
y(t)
moves
f (c + ) > 0.
y = c is unstable
f (c ) < 0.
,
the solution
y(t)
moves
from above and from below. Otherwise (if it is stable from one side and unstable from the other) we say it is
semistable.
f,
y 0 = f (y):
Step 2: Mark all the equilibrium values on a number line, and then in each interval between two critical
for which
f (c) = 0,
to determine whether
is negative,
draw left-arrows.
Step 4: Using the arrows, classify each critical point: if the arrows point toward it from both sides, it
is stable. If the arrows point away, it is unstable. If the arrows both point left or both point right, it is
semistable.
Step 5 (optional): Draw some solution curves, either by solving the equation or by using the stability
information.
y0 = y
Step 1: We have
Step 2: We draw the line and plug in 2 test points (or just think for a second) to see that the sign
f (y) = y ,
y = 0.
| .
0
| .
0
Step 4: So we can see from the diagram that the only equilibrium point
Step 5: We can of course solve the equation to see that the solutions are of the form
indeed, the equilibrium solution
y=0
Step 1: We have
y = 1,
and
is unstable .
y(t) = C et ,
and
is unstable:
y 0 = y 2 (y 1)(y 2)
y = 0,
y = 2.
Step 2: We draw the line and plug in 4 test points (or just think for a second) to see that the sign
diagram looks like
| | | .
0
| | | .
0
Step 5: In this case, it is possible to obtain an implicit solution by integration; however, an explicit
is semistable ,
is stable , and
is unstable .
solution does not exist. However, we can graph some solution curves to see, indeed, our classication is
accurate:
1.11
Euler's Method
If we have exhausted all of our techniques trying to solve a rst-order initial value problem
y(a) = y0 , we
interval [a, b].
y 0 = f (x, y)
with
are sad. However, perhaps we would like to be able to nd an approximate solution on some
One method we can use to nd an approximate solution is Euler's Method, named after the Swiss mathematician Leonhard Euler (pronounced oiler).
The general idea behind Euler's Method, which should bring back memories of basic calculus, is to break
up the interval
[a, b]
into many small pieces, and then to use a linear approximation to the function
y(x)
ba
be the width of the subintervals.
n
x0 = a, x1 = x0 + h, x2 = x1 + h, ... , xn = xn1 + h = b.
Step 3: Take y0 to be the given initial value. Then compute, iteratively, the values y1 = y0 + h f (x0 , y0 ),
y2 = y1 + h f (x1 , y1 ), ... , yn = yn1 + f (xn1 , yn1 ). It is easiest to organize this information in a
x-values
n,
and let
h=
table.
y 0 = ln(x + y)
with
y(1) = 1.
h = 0.1.
x-values.
y -value and the
Steps 2+3: We organize our information in the table below. We ll out the rst row with the
Then we ll in the empty columns one at a time: to start the next column, we add the
h f (x, y)
y -value.
x
y
f (x, y)
h f (x, y)
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
1.0693
1.1467
1.2320
1.3249
1.4251
1.5324
1.6466
1.7674
1.8946
2.0280
0.693
0.774
0.853
0.929
1.002
1.073
1.1418
1.208
1.272
1.334
0.0693
0.0774
0.0853
0.0929
0.1002
0.1073
0.1142
0.1208
0.1272
0.1334
Step 4: Finally, we can plot the points, and (for comparison) the actual solution curve obtained using a
computer. As can be seen from the graph, the approximation is very good:
1.12
Step 1: Rewrite as
Step 2:
P 0 = kP (M P ),
where
and
M dP
= kM dt.
P (M P )
M
dP = kM dt. To evaluate the P -integral, use
P (M P )
M
1
1
=
+
. Evaluating the integrals therefore yields
P (M P )
P
M P
ln(P ) ln(M P ) = kM t + C .
CekM t .
P
= kM t + C ;
M P
M
P (t) =
.
1 + CekM t
Solving for
yields, nally,
ln
P (0) = P0 ,
P (t) =
M P0
P0 + (M P0 )ekM t
C=
P
=
M P
M
1.
P0
Then
here, or using the properties of stable and unstable equilibria, we can observe some properties of the solution
curves.
Since
and
P =0
and
P =M
| | .
0
So
is a
stable equilibrium.
Thus, as
t ,
carrying capacity of
M.
P0
P (t)
k<0
P0
rather than
k > 0,
M,
M,
Setup:
We have some reservoir (pool, lake, ocean, planet, room) of liquid (water, gas) which has some substance
(pollution, solute) dissolved in it.
V0
k(t)
y0
in the reservoir.
We assume that the substance is uniformly and perfectly mixed in the reservoir, and want to know the
amount
y(t)
t.
Note that this is the general setup. In more specic examples, the amount of liquid owing in or out may be
constants (i.e., not depending on time), and similarly the concentration of the liquid owing in could also be
a constant. The solution is the same, of course.
Solution:
Let
V (t)
Then if
k In(t)
and the total amount of substance moving out is the concentration of substance
Thus we have
V 0 (t) =
In(t)
y(t)
Out(t).
V (t)
Out(t),
and
y(t)
Out(t).
V (t)
diagram:
V (t)
y0 +
explicitly.
Out(t)
V (t)
y = k(t) In(t),
it is rst-order linear.