Professional Documents
Culture Documents
on
Population Dynamics
Submitted
by
Amit Das
Student ID: M.Sc-201208
Mathematics Discipline
Science, Engineering & Technology School
Khulna University, Khulna-9208
Submitted
to
Mathematics Discipline
Science, Engineering & Technology School
Khulna University, Khulna-9208
Bangladesh
Basic Definitions
1.1 Population
Populations are the subjects of a study that share at least one common characteristic, which can
be specifically or vaguely defined. A population, in statistics and other areas of mathematics, is a
discrete group of people, animals or things that can be identified by at least one common
characteristic for the purposes of data collection and analysis. To gather information about a
large population, data is usually gathered from a sample.
Population dynamics is the branch of life sciences that studies the size and age composition
of populations as dynamical systems, and the biological and environmental processes driving
them (such as birth and death rates, and by immigration and emigration).
At any given time, a dynamical system has a state given by a tuple of real numbers that can be
represented by a point in an appropriate state space. The evolution rule of the dynamical system
is a function that describes what future states follow from the current state. Often the function
is deterministic, that is, for a given time interval only one future state follows from the current
state. However, some systems are stochastic, in that random events also affect the evolution of
the state variables. A dynamical system is a system in which a function describes
the time dependence of a point in a geometrical space. Examples include the mathematical
2
models that describe the swinging of a clock pendulum, the flow of water in a pipe, and the
number of fish each spring time in a lake.
Consider a dynamical system that are modeled by a finite number of coupled first order ODE as,
x f (t , x, u )
(1)
We call, equation (1) is the state equation. Here x is the state and u is the input.
y h(t , x, u ) (2)
the equation (1) and (2) are called together the state space model or simply state model.
x f (t , x,)
(3)
Where the input specified as a function of time and state i.e u (t , x ) . Then equation (3) is
called unforced state equation.
3
then the function is said to be bounded below by B. The concept should not be confused with
that of a bounded operator.
C 0 , C1 FUNCTION:
1.8
0 1
"The class C consists of all continuous functions. The class C consists of all differentiable
functions whose derivative is continuous; such functions are called continuously differentiable."
Continuous functions with first derivative is called C1. The space C1or precisely C1 (a,b) consists
all continuous function which have continuous first derivatives, defined in the interval [a,b].
1
f ( x) x 2 sin
1
A differentiable function might not be C . The function x for x 0 and f(x) =0
1 1
f ( x ) cos 2 x sin
x x for x 0 and f'(x) =0 for x=0, which is discontinuous at x=0, so it
1
is not C .
A continuous function is a real-valued function whose graph does not have any breaks or holes.
4
A function f is continuous at x=a provided all three of the following are true:
i ) f ( a) exists
ii ) lim f ( x ) exists
xa
iii ) f ( a ) lim f ( x)
xa
In other words, a function f is continuous at a point x=a, when (i) the function f is defined at a,
(ii) the limit off as x approaches a from the right-hand and left-hand limits exist and are equal,
and (iii) the limit of f as x approaches a is equal to f(a).
A autonomous system is a system of ordinary differential equation which does not depend on
independent variable. If the independent variable is time then it is called time invariant systems.
an autonomous system or autonomous differential equation is a system of ordinary differential
equations which does not explicitly depend on the independent variable. When the variable is
time, they are also called time-invariant systems.
x f ( x)
(4)
Then the system (4) is said to be autonomous or time invariant.
5
1.11 Non-autonomous system
If the function f depends explicitly on time t, then the system is called non -autonomus or time
variant.
dy
7ty
dt (5)
A coupled system is formed of two differential equations with two dependent variables and an
independent variable.
Example:
dy
ax by
dt
dx
cx dy
dt
x A(t ) x B (t )u
(6)
y C (t ) x D(t )u
(7)
For example,
6
dx
5t 2 x 6u
dt
y 2tx 5u
x(t ) f (t, x(t )) (1)
x(t0 ) t0
Let g (t ) be a real function defined t in a real interval I . The function g (t ) is called the
solution of the differential equation (1) on I if
i ) f (t , g (t )) is defined t I and
ii) f (t , g (t )) 0 t I
i.e. the substitution of g (t ) and its various derivatives for x and its corresponding derivatives in
equation (1) reduces to an identity on I .
Consider the initial value problem (1), then x(t ) be the solution of (1) if it can be written as,
t
x(t ) x0 f (s, x(s))ds (2)
t0
The unknown function at the different point in the domain of the solution.
Mathematically,
7
x (t ) f (t, x (t ))
x(t0 ) a, x(t1 ) b)
n n
With f : D and (t0 , a), (t1 , a) D
The basic difference between initial value problem and boundary value problem be if the
independent variable is time over the domain [0,1] , a boundary value problem would specify
value for x(t ) at both t 0 and t 1. Where as an initial value problem would specify a value of
x(t ) and x(t ) at time t 0.
For example,
x ( t ) x (t ) 0
with initial condition x(0) 1, x(0) 0
x (t ) x ( t ) 0
x(0) 1, x(1)
In other words a DS is a system whose state evolutes with time over a state space. i.e, it is a time
varying system.
8
Existence and Uniqueness Theorem
Theorem 1:
(Existance theorem)
.
x ( t )=f ( t , x )
x ( t 0 ) =x 0
Suppose that f (t , x) is continuous in some region R={ ( t , x ) :|t−t 0|≤ a ,| x−x 0|≤ b }≤ , ( a , b> 0 ) .
Since f is continuous in a closed and bounded domain,it is necessary bounded in R,i.e there
exists k¿ 0 such that,
| f ( t , x )∨≤ k , ∀(t , x )∈ R .
Then the IVP (1) has at least one solution x=x ( t ) define in the interval |t−t 0∨≤ α
Where,α =min a , { bα ) .
Proof:
Let us consider a solution in the IVP (1) which is equation of the integral equation,
t
x ( t )=x 0 +∫ f (s , x ( s ))ds … … … … … … … … … … … … … … … … ..(2)
t0
x ( t )=t 0
9
t
¿ x 0 ( t )=x 0 +∫ f [ s , x 0 ( s ) ] ds
t0
t
¿ x n ( t )=x 0 +∫ f [ s , x n−1 ( s ) ] ds … … … … … … … … … … … … … ….2( a)
t0
Where
n=1,2,3………………………………………………………………………………………
We shall first show that the seq { x n (t) } are constant and lies in the rectangle |t−t 0|≤ h.
Now,
t
¿ x 1 ( t )=x 0 +∫ f [ s , x ( s ) ] ds
t0
t
¿ x n ( t )=x 0 +∫ f [ s , x n−1 ( s ) ] ds … … … … … … … … … … … … .………………………….. (3)
t0
t
≤∨∫ f ¿ ¿
t0
10
t
≤ k ∫ ¿ ds∨¿ ¿
t0
≤ k∨t−t 0 ∨¿
≤k h
≤b
|t−t 0|≤ h.
|t−t 0|≤ kh .
t
t
≤ L∫ ¿ x 1 ( s ) −x0 ( s)∨ds
t0
t
≤ Lk ∫ ¿ s−s 0∨ds
t0
2
|t−t 0|
≤ Lk
2
11
h2
≤ Lk
2
Similarly,
2 h3
≤L K
3!
∞ ∞ m−1 m
Or, ∑ |x m ( t )−x m−1 ( t )|≤ K ∑ L m!h
m=1 m=1
m
K ( Lh)
≤ ∑ ..............................................................(6)
L m!
From weier strass M-test,we have if { x m (t) }be a seq of function with,
A ∞
|x m ( t )|≤M m [α , β] ∑ Mn ∑ xn ( t ) [α , β]
it in Write m=∞ <0 then m=∞ converges uniformly in
to a unique function x(t).
∞
x 0 ( t ) + ∑ [ x n ( t )−x m−1 ( t ) ]
Applying weierstressismtest in (6) ,if follow that m=t converges
12
Since f(t,x) is = cont. is relation of x and x n ( t ) → x ( t ) as n
t
x 0 + limit n→ ∞∫ f [ s , x n ( s ) ] ds
t0
= [by 1a]
t
x 0 +∫ limit n→∞ f [ s , x n ( s ) ] ds
t0
= [by -7]
t
∴ x ( t ) =x 0 +∫ f ( x , x ( s ) ) ds
t0
Thus x(t) solves i.e (2) imples that x(t) is a solution of (1)
Uniquenemfor proof:
Proof:
Lastly we show that x(t) is the solution of (1) . suppose that Z(t) is also a continuous solution of
(2) for
|t−t 0|≤h
Then ,
t
z ( t )=xz 0 +∫ f [ t , x ( t ) ]
t0
dt
13
Let N=max |x ( t )−z ( t )|
…………………………………………………………………. (8)
Then for
|t−t 0|≤h we obtain
t
|x ( t )−z ( t )|∫ |f [ s, x 0 ( t ) ] −f [ s , z ( t ) ]|ds
t0
t
¿ L∫|x 0 ( s ) −z ( s )|ds
t0
[ by 1(a)and 8]
And
t
|x 2 ( t )−z ( t )|≤∫|f [ s , x 1 ( s ) ]−f [ s , z ( s ) ]|ds
t0
t
L2 Nh2
¿
2!
In General,
NLn h n
xn (t ) z (t )
n!
If n then,
( Lh) n
lim
xn (t ) z (t ) N n n !
(10)
Since the right side of (10) approaches to zero. It follows that x(t ) z (t ) .
14
This implies that x(t ) and z (t ) areidentical.
x t
So, is the only solution of (2)and thus (1).
x x0 a
Hence, the initial value problem.(1) has a unique solution x x(t ) on the interval .
Theorem 2
State and prove the existence and Uniqueness theorem depending on initial condition.
f x, y
Statement: If is continuous and satisfies Lipschitz condition with respect to y ,with
( x0, y0 ) Y y0
is a fixed point in D and 0, h 0 such that for each Y0 satisfying 0 .Then
the IVP
dy
f ( x, y ), y ( x0 ) Y0
dx
x x0 h
Possesses a unique solution y ( x) for Y0 y0 and y ( x) for Y0 y 0 on
y 0 y0 1 y ( x) y0 ( x) 1e kh x x0 h
then on
Where,
Y0 y0 , y 0 y0 1
(2)
From Weierstrass-M.test, we have
y ( x ) lim yn ( x)
n
Where,
15
x
yn ( x) y0 f t , yn 1 (t ) dt , n 1, 2......
x0
y0 ( x ) y0 x x0 h
And on .
y ( x) lim y n ( x)
Also, n ,when
x
y n ( x) y 0 f t , y n 1 (t ) dt , n 1, 2...
x0
x x0 h
and y 0 ( x) y 0 on
Thus we have
x x
y1 ( x) y1 ( x) y 0 f t , y 0 (t ) dt y0 f t , y0 t dt
x0 x0
x x
y 0 y0 f t , y 0 (t ) dt f t. y0 t dt
x0 x0
x
1 k y 0 y0 dt
x0
y1 ( x) y1 x 1 1 k x x0
(3)
Again,
16
x x
y 2 ( x ) y2 ( x ) y 0 f t , y1 (t ) dt y0 f t , y1 (t ) dt
x0 x0
x x
y 0 y0 f t , y1 (t ) dt f t , y1 (t ) dt
x0 x0
x
1 k y1 y1 dt by Lipschitz continuous
x0
x
1 k1 1 k ( x x0 ) dt using (3)
x0
k ( x x0 ) 2
1 k1 ( x x0 )
2
k ( x x0 )
2 2
1 1 k ( x x0 )
2!
2
k j ( x x0 ) j
y 2 ( x ) y2 ( x ) 1
j 0 j!
n
k j ( x x0 ) j n
(kh) j
y n ( x ) yn ( x ) 1 1 x on x x0 h.
j 0 j! j 0 j!
Letting, n , we have
n ( kh) j
lim y n ( x) yn ( x) 1e kh on x x0 h e kh
n
j 0 j !
Thus, if the initial values of the two solutions y ( x) and y ( x) differ by a small amount, then their
x x0 h
values will differ by an arbitrary small amount at every point of .Geometrically, the
x x0 h
corresponding integral curves are sufficiently close to each other x on .
Theorem 3
State and prove the existence and uniqueness theorem depending on the arbitrary function.
17
Statement: If in a domain D , if xy plane
(ii) F ( x, y ) is constant in D
F ( x, y ) f ( x, y ) for ( x, y ) D
(iii)
2.
( x0 , y0 ) is a point if D and
(iii)
x, ( x)
and
x, ( x) D
thus
kh
( x ) ( x) (e 1) x x0 h
k on
Proof:
We have,
dy
f ( x. y ), y ( x0 ) y0
dx (1)
dy
F ( x, y ), y ( x0 ) y0
& dx (2)
F ( x , y ) f ( x, y )
Where, (3)
18
Thus,
( x, y ) F ( x , y ) f ( x , y )
( x, y ) by (3)
Again,
Let, 0 ( x ) ( x) (4a)
x
n ( x ) y0 f t , n 1 (t ) dt , n 1, 2...
When, x0
(5)
( x) lim n ( x)
Then, n is a solution of (1)
x x0 h
By hypothesis 1(i), the IVP (1) has a unique solution on ,
x
( x) y0 F t , (t1 ) dt
x x0 h.
x0
on (6)
Now,
x x
1 ( x) ( x) y0 f t , 0 (t ) dt y0 F t , (t ) dt
x0 x0
x x x
f t , 0 (t ) dt F t , (t ) dt
x0 x0 x0
x x x
f t , (t ) dt F t , (t ) dt
x0 x0 x0
[By (4b)]
1 ( x) ( x) k ( x x0 )
Similarly,
19
x x
2 ( x ) ( x ) y0 f t , 1 (t ) dt y0 F t , (t ) dt
x0 x0
k ( x x0 ) 2
( x x0 )
2!
2
k ( x x0 )
j 1 j
j 0 j!
k j 1 ( x x0 ) j
n
lim n ( x) ( x) lim
n n
j 1 j!
n
(kh) j
lim
k n j 1 j !
kh
( x) ( x)
k
e 1
kh
( x ) ( x)
k
e 1
Thus, .
Lemma 1:
( x ) F ( x, f ( x))
( x 0 ) y0 (1)
x
( x ) y0 F (t , (t )) dt
x0
(2)
20
Proof: Let us first assume that ( x) is differentiable (and ( x) continuous) so that ( x) satisfies
x x0 h.
equation (1). Then we integrate and deduce that for
x x
x0
(t )dt F (t , (t ))dt
x0
However the left hand side is equal to ( x) ( x0 ) by the fundamental theorem of calculus.
Thus, since
( x 0 ) y0 we get
x
( x) y F (t , (t ))dt
x0
Vice versa, assume that ( x) is continuous and satisfies that integral equation (2). Then the
F (t , (t ))dt
calculus the integral x0
is a differentiable function of x with derivative F ( x, ( x)).
Thus the right hand side of equation (2) is a differentiable function with derivative F ( x, ( x))
and by equation (2) ( x) is differentiable with ( x) F ( x, ( x)) which is one part of equation
x0
... dt 0.
x0
Thus we proved the vice versa.
Lemma 2:
g ( x) y0 b x0 h, x0 h Tg ( x) y0 b
If g satisfy for all x in then Tg also satisfies for all x in
21
x0 h, x0 h
Proof:
hb x x0 h min(a, b )
It is here where the crucial M is used. If g ( x) y0 b and m then
x0 x
for
Tg ( x ) y0
F t, g (t ) dt
x0
F t, g (t) dt
x0
x
Mdt M x x
x0
0 Mh b
x x
x x0
....... dt .......
x0
And a similar argument goes for (just write x0
instead of dt in the case)
0 ( x) y0
x
1 ( x) T 0 ( x) y0 F (t ,y0 )dt
x0
x
2 ( x) T1 ( x) y0 F (t ,1 (t ))dt
x0
x
2 ( x) T 1 ( x ) y0 F (t ,1 (t ))dt
x0
x
3 ( x) T2 ( x ) y0 F (t ,2 (t ))dt
x0
22
If
n 1 is already defined, we set
x
n ( x) y0 F (t ,n1 (t ))dt
x0
( x, n1 ( x ))
Lemma 2 says that if belongs to the rectangle R where F was defined and if
t x0 h
. Thus the definition makes indeed sense for all n provided.
0 ( x) y0
and
x x0 h
if
Lemma 3:
M 0 max|t x0 | h | F t , y0 | .
Let
| x x0 |n 1
| n 1 x n x | M 0 K n
The sequence n testifies the estimate n 1 ! for all [ x0 h, x0 h]
The sequence n converges to a limit which is a solution of the initial value problem as well as
the equivalent integral equation.
Proof:
23
We shall prove the inequality for
x0 x x0 h , and after simple modifications the argument
We show the inequality by iteration i.e. “mathematical induction”. That is we first prove the
n x n 1 x
estimate for n=0. Then we show that for all n the estimate for implies the
n 1 n
estimate for .
This means because we have shown it for n=0, it then follows for n=1. Since the estimate for n=1
implies the estimate for n=2 the estimate for n=2 is also true. Since the estimate for n=2 implies
the estimate for n=3 the estimate for n=3 is also true, and so on.
First step, we show the assertion for n=0, that is show that
1 x 0 x M 0 x x0 x x0 h
if
Indeed
1 T 0 and therefore
x
1 x 0 x F t , y dt
0
x0
x
F t , y dt
x0
0
M 0 x x0
n!
n n 1 !
We write
24
n 1 x n x T n x T n 1 x
x
F t , t F t , t dt
x0
n n 1
By the mean value theorem of differential calculus with respect to the y variable as in III
F
F t , n t F t , n 1 t t , n t n 1 t
y
t t
Where is some value between n and n 1 . Therefore
F t , n t F t , n 1 t k n t n 1 t
n!
Thus we get
F t , t F t , t dt
x0
n n 1
x
F t , t F t , t dt
x0
n n 1
t x0
x n
k M 0k n 1
dt
x0
n!
t x0
x n
M 0k n 1
k dt
x0
n!
x x0
n
M 0k n
n 1 !
x x0 a, y y0 b, (a, b 0),
25
Or a strip
x x0 a, y , (a, 0),
f
And that f is a real-valued function defined on S such that y exists, is continuous on S, and
f
( x, y ) K , (( x, y ) in S ),
y
Proof: we have
v1 f
f ( x, y1 ) f ( x, y2 ) ( x, t )dt ,
v2 y
And hence
v1 f
f ( x, y1 ) f ( x, y2 ) v2 y
( x, t ) dt K y1 y2 ,
For all ( x, y1 ) ( x, y2 ) in S.
f ( x, y ) xy 2
On
R: x 1, y 1
Here
f
( x, y ) 2 xy 2
y
For all ( x, y ) on R. This function does not satisfy a Lapschitz condition on the strip
S: x 1, y ,
Since
26
f ( x, y1 ) f ( x, 0)
x y1 ,
y1 0
y1 , x 0.
Which tends to infinity as if
On
R: x 1, y 1.
y1 0,
Indeed, if
2
f ( x, y1 ) f ( x, 0) y1 3 1
1 ,
y1 0 y1 y1 3
y1 0.
Which is unbounded as
dy
x 2 y 2 , y (0) 0
Example 1: Examine the existence and uniqueness of IVP dx . And also find
an interval on which the unique solution exists.
Solution: Here,
f ( x, y ) x02 y02
( x0 , y0 ) (0, 0)
………………….. (1)
f
2y
y
f
Since f and y are defined at (0, 0)
27
: x a, y b
Then f ( x, y ) satisfies the Lip. conditions with respect to y in .
2nd part:
x x0 h
i.e, x h i.e, x a, y b
b
h min(a, )
Where M and M max f ( x, y ) for ( x, y )
Now,
f ( x, y1 ) f ( x, y2 ) x 2 y12 x 2 y22
y1 y2 y1 y2
y12 y22
y1 y2
y1 y2
Which is convergence.
b
h min(a, )
Here, M a b , so a b2
2 2 2
Let,
b
F (b)
a b2
2
(a 2 b 2 ).1 b.2b
F (b)
(a 2 b 2 )2
(a 2 b 2 )
(a 2 b 2 )2
28
(a 2 b 2 )
2 2 2 0
(a b )
a2 b2 0
ba
a 0
Again,
1
F (b)
At b=a, 2a 3 .
a 1
F (b) 2
The maximum value is, 2a 2a .
1
h min(a, )
2a
1 b 1
a
If 2a then, a b 2a
2
b 1
h
a b 2a
2
1
h
i.e, 2a
29
1 1
a h
And if, 2a then, 2a ……………………………………………….(2)
1
h
2a
1
x
2a
1
a
2a
1
a2
2
1
a , a 0
2
1 2
x
1 2
2.
2
1
x
2
1 1
x
i.e, 2 2
Example 2: Examine the existence &uniqueness of y 1 y , y (0) 0 & find the interval of
2
existence.
30
Show that the solution exist over a large interval than the guaranteed by the existence & unique
theorem
Solution:
f ( x, y ) 1 y 2
f
2y
And y
f
We observe that f ( x, y ) & y are defined at (0, 0)
f x, y R : x a, y b
Also satisfies the Lipschitz condition we respect to y in
x x0 h x h x a, y b
i.e in
b
h min a, and M max f ( x, y ) for ( x, y ) R
When M
f ( x, y1 ) f ( x, y2 ) 1 y12 1 y2 2 y12 y2 2
y1 y2
y1 y2 y1 y2 y1 y2
Now,
Here,
b
h min a, 2
M 1 b therefore,
2
1 b
31
b
F (b)
Let, 1 b2
(1 b 2 ).1 b.2b 1 b2
F (b)
(1 b 2 ) 2 (1 b2 ) 2
1 b2
F (b) 0 0 b 2 1 b 1, b 0
For maximum or minimum
1 b
2 2
(1 b 2 )(2b 2 6b) 1
F (b) b 1, F (b)
Again, (1 b 2 ) 4 At 2
1 1
F (b)
The max value is 112
2
1
h min a,
2 .
1 b 1
a
If 2 then 1 b 2
2 b 0
b 1
h
1 b 2
2
1
h
i.e 2
1 1
a h
And if 2 then 2
1
h
Thus in any case 2
1
x
Hence, the interval of existence is 2
32
1 1
x
2 2 (2)
Part-3:
We have
dy
1 y2
dx , y (0) 0
dy
dx
1 y2
tan 1 y x c
Using y (0) 0
c0
tan 1 y x
y tan x
x R (2n 1) , n z
2
x
the interval is 2 2 (3)
For (2) & (3),we observe that the solution exist over a large interval than the guaranteed by the
existence & uniqueness theorem. Example 1
33
f ( x, y ) xy sin y
Define a region
R : ( x, y ) : x 10, y ,
i.e. 2
f
x cos y
y
Which is also continuous.
And here,
f ( x, y ) xy sin y
xy sin y
10. sin
2 2
5 1
k ; where, k 5 1
f
x cos y
y
x cos y
10 0
10
f
x 0 and
So, here f and y are both continuous and bounded in a closed rectangle about 0
y0 2 .
Example 3:
Here given ordinary differential equation are
y ' 1 y 2 ; y (0) 0
So, we can write
34
f ( x, y ) 1 y 2
f
2y
y
Consider the rectangle,
S ( x, y ) : x 100, y 1
f
Clearly, f and y are both continuous in S .
x 100
the original interval
Since, the above equation is separable, we can solve it exactly.
Now,
y ' 1 y2
dy
1 y2
dx
dy
dx
1 y2
35
Integrating both sides,
tan 1 y x c (1)
Using initial condition, we get,
tan 1 0 0 c
c 0
From equation (1) we get,
tan 1 y x
tan(tan 1 y ) tan x
y tan x
, 100,100
This solution is valid only in 2 2 which is also much smaller than but
nevertheless bigger than that predicted by the existence and uniqueness theorem.
Example 4:
The given IVP is
y ' x y , y (1) 0
f ( x, y) x y
Now,
f ( x, y1 ) f ( x, y2 ) x y1 x y2
xy1 xy2
x y1 y2
k y1 y2
36
Example 5:
The given IVP is
1
y ' y 3 x, y (1) 0
So we can write,
1
f ( x, y ) y x
3
Now,
1 1
f ( x, y1 ) f ( x, y2 ) y13 x y2 3 x
1 1
y1 y2 3 3
y1 y2
2 1 1 2
y1 3 y13 y2 3 y2 3
We take,
y2 0 then,
y1 0
f ( x, y1 ) f ( x, 0) 2
y1 3
1
2
y1 3
y1 very closed to zero. Then
Now, we can take becomes unbounded. Hence the relation,
f ( x, y1 ) f ( x, 0) 2 y1 0
does not hold always around a region about (1, 0) .
Since, f does not satisfy Lipschitz condition, we cannot say whether unique solution exists or
does not exists.
1
Example 6:
The given IVP is
37
2y
y' , y ( x0 ) y0
x
So, we can write,
2y
f ( x, y )
x
f 2
And, y x
When, x0 0, there exists no solution with y0 0 . When y0 0, then we have infinite number
2y
y' , y (0) 0
of solutions y x ( any real number) that satisfy the IVP
2
x
Theorem 5
38
Statement:
X (t ) F (t , X )
x(t0 ) x
*
i.e,
x 0 f1 (t , x1 , x2 ,..., xn )
…………………………………
………………………………...
x n f n (t , x1 , x2 ,..., xn )
Suppose that
i) F (t , x ) is constant in,
: t tn a X X * b and J * M 0
S | F (t , x) M (t , x)
F (t , X ) F (t , Y ) L X Y (t , X ), (t , Y )
39
* t t0 a and V * (t ) X * b
iii) X (t ) is constant in thus the system (1) has a unique solution
h
h min 0,
X (t ) defined in the interval J h : t t0 h where m
D : t t0 0, X
i,e
Theorem 6:
Uniqueness theorem
x (t ) f (t , X )
x(t0 ) x0 (1)
f
Suppose that f (t , x ) and x are bounded in
f (t , x) k and
i,e. a)
f
L (t , x)
b) x
Then the initial value problem (1) has at most one solution x x(t ) defined in the interval
b
min a,
t t0 k
where
Combining with existence theorem, the initial value problem (1) has unique solution x x(t )
t t0
defined in the interval .
40
Lipschitz condition:
f (t , x1 ) f (t , x2 ) L x1 x2 ( x1 , x2 )
is called the Lipschitz condition and L is called
Lipschitz constant.
f
If x exists and is bounded then it necessarily satisfies Lipschitz condition.
f
On the other hand, a function f (t , x ) may be Lipschitz constant but x may not exists.
f (t , x)
sup L
D x
Lipschitz condition:
Let us assume that the variation of the function f (t , x ) relative to x remain bounded.
f (t , x1 ) f (t , x2 ) L x1 x2
i.e (1)
condition in any domain D, if the inequality (1) holds holds for all point (t , x1 ),(t , x2 ) in D having
the same t.
Example:
41
2
0 ( x, y ) : x , y c
iii) F ( x, y ) e satisfies Lipschitz condition and
4
with L e .
e
F ( x, y ) y
is not defined in but satisfies Lipschitz condition with L 1 .
2
iv)
Example 7: Examine the existence and uniqueness and show that whether it satisfy the Lipschitz
constant or not, if satisfy, find the Lipschitz constant.
f (t , x) t 2 x , t 1, x 1
i)
ii) x p(t ) x r (t ) , where p (t ) & r (t ) are defined & constant in the interval a t b
f (t , x) t 2 x , t 1, x 1
Solution: (i)
Here,
f (t , x) t 2 x
f t .1 x>0
2
= 2
x -t x<0
t 1 & x 1
The function f is constant at the region
f
But, x does not exist at (t , 0)
f (t , x1 ) t 2 x1 & f (t,x 2 ) t 2 x2
Again,
42
f (t , x1 ) f (t , x2 ) t 2 x1 t 2 x2 t 2 x1 x2
x1 x2
Where, L t 1 t 1
2
Solution: Here,
f (t , x) p (t ) x r (t )
f (t , x1 ) p (t ) x1 r (t )
And f (t , x2 ) p(t ) x2 r (t )
L max p(t )
If a t b then
f (t , x1 ) f (t , x2 ) p(t ) x1 (t ) p(t ) x2 (t )
p (t )( x1 x2 )
L x1 x2
Here, the initial value problem (1) has a unique solution in the interval a t b .
The end of the existence & uniqueness theorem are sufficient but not necessary.
43
Picard’s method of Successive Approximation
Successive approximation technique on iteration method: A method for estimating the value
of an unknown quantity by repeated comparison to a sequence of known quantities is called the
successive approximation technique.
The iterative method is a mathematical way of solving IVP problem which generates a sequence
of approximations. Also by using the known sequence, we can find out the unknown sequence in
this technique.
y=f ( x , y)
y ( x 0 )= y 0 … … … … … … … … … … …( 1)
∫¿
y ¿¿
Or, [ y ] y ¿¿
Or, y ( x ) − y 0=∫ f ( x , y ) dx
x0
Or, y ( x ) = y 0 +∫ f ( x , y ) dx
x0
Or, y ( x ) = y 0 +∫ f ( t , y (t ) ) dt … … … … … … … … … … .. … ...(3)
x0
44
We also assume that y 0 is the zeroth approximation of y(x).
The 1st approximation y 1 ( x ) is obtained by putting y 0 in place of y(t) in the right hand side of
equation (3).
x
y n ( x ) = y 0 +∫ f ( t , y n −1 (t ) ) dt
x0
Example 8: Apply Picard’s iteration for the IVP ẏ=2 x (1− y ) ; y ( 0 ) =2.
dy
Solution: Given that, =2 x ( 1− y ) ; y ( 0 )=2 … … … … … … … … …(1)
dx
dy
Comparing (1) with IVP =f ( x , y) ; y ( x 0 )= y 0we get,
dx
x
y 1 ( x )= y 0+∫ f ( t , y 0 (t ) ) dt
x0
¿y x
0+ ¿∫ f ( t , 2) dt ¿
0
45
x
¿ 2+∫ −2 t dt
0
¿ 2−x 2
∴ y 1 ( t )=2−t 2∧f ¿
x
y 2 ( x ) = y 0+∫ f ( t , y 1 (t) ) dt
x0
x
¿ 2+∫ 2 t(t 2−1) dt
0
x
¿ 2+∫ (2 t 3−2 t) dt
0
2 x4
¿ 2−x +
2
4
2 t 2 t4
∴ y 2 ( t )=2−t + ∧f ( t , y 2 ( t ) ) =2 t(t − −1)
2 2
x
y 3 ( x ) = y 0+∫ f ( t , y 2 (t )) dt
x0
x
t4
¿ 2+∫ 2 t (t 2− −1)dt
0 2
x
¿ 2+∫ (2 t 3−t 5−2 t) dt
0
2 x 4 x6
¿ 2−x + −
2 6
46
2 x 4 x6
¿ 2−x + −
2 3!
4
2 t t6 t6 t4 2
∴ y 3 ( t )=2−t + − ∧f ( t , y 3 ( t ) ) =2 t( − +t −1)
2 3! 3! 2
x
y 4 ( x )= y 0 +∫ f ( t , y 3 (t) ) dt
x0
x
t6 t4 2
¿ 2+∫ 2 t( − +t −1) dt
0 3! 2
2 x 4 x6 x8
¿ 2−x + − +
2 3! 4!
x 4 x 6 x8 2n
2 n x
y n ( x ) =2−x + − + +…+(−1)
2 3! 4! n!
Example-9: Compute the 1st three iteration by Picard’s method for the IVP
dy
Comparing (1) with IVP =f ( x , y) ; y ( x 0 )= y 0we get,
dx
47
x
y 1 ( x )= y 0+∫ f ( t , y 0 (t ) ) dt
x0
x
¿ 0+∫ f ( t , t ) dt
0
x
¿ ∫ (t 2−t 2−1) dt
0
x
¿ ∫ −1 dt
0
¿−x
x
y 2 ( x ) = y 0+∫ f ( t , y 1 (t) ) dt
x0
x
¿ ∫ −1 dt
0
¿−x
x
y 3 ( x ) = y 0+∫ f ( t , y 2 (t )) dt
x0
x
¿ ∫ −1 dt
0
¿−x
48
Which is the required solution.
dy
Example 10: Solve by Picard’s Method =sinx + y 2 ; y ( 0 )=0.
dx
dy
Solution: Given that, =sinx + y 2 ; y ( 0 )=0 … … … … … … … … . (1 )
dx
dy
Comparing (1) with IVP =f ( x , y) ; y ( x 0 )= y 0we get,
dx
2
y 0=0 , x 0=0 , f ( x , y )=sinx+ y … … … … … … … …(2)
x
y 1 ( x )= y 0+∫ f ( t , y 0 (t ) ) dt
x0
x
¿ 0+∫ f ( t , 0 ) dt
0
x
¿ ∫ sint dt
0
¿ 1−cos x
∴ y 1 ( t )=1−cos t
And f ¿
x
y 2 ( x ) = y 0+∫ f ( t , y 1 (t) ) dt
x0
x
1
¿ 0+∫ (sin t+1−2cos t+ .2 cos2 t) dt
0 2
49
x
¿∫ ¿ ¿
0
x sin 2 x
¿−cos x +cos 0+ x−2sin x +2 sin 0+ +
2 4
sin 2 x 3 x
¿ 1−cos x−2sin x + +
4 2
sin2 t 3 t
∴ y 2 ( t )=1−cos t−2sin t + +
4 2
y ' xy, y 0 1
Example 10: Solve the following equation using Picard iteration.
y0 x 1
Solution: Here,
Now,
x
y1 x y0 f t , y0 t dt
x0
x
y1 x 1 tdt
0
x
t 2
1
2
0
2
x
1
2
50
x
t2
y2 x 1 t 1 dt
0
2
x
t3
1 t dt
0
2
x
t2 t4
1
2 8 0
2 4
x x
1
2 8
x
t2 t4
y3 x 1 t 1 dt
0
2 8
x
t3 t5
1 t dt
0
2 8
x
t t t
2 4 6
1
2 8 48 0
2 4 6
x x x
1
2 8 48
.
.
.
x 2 x 4 x6
yn x 1 ....................
2 8 48
x2
e 2
Picard iteration is giving us the power series of this solution. Each iteration gives us an
additional term.
Theorem 8:
51
x x0 a y y0 b
If (i) f(x) is continuous in the closed rectangles , and hence a M 0 such
that f ( x, y ) M
x x0 a y y0 b
(iii) y0 ( x) is constant in and
b
J n : x x0 h min a,
m
( Lh) m
y ( x) ym ( x) Ne Lh min 1, ( A) ( m 0,1, 2,3...)
m!
max xyh y ( x) y0 ( x) N
Where
y f ( x, y )
y ( x0 ) y0 (1)
x
ym1 ( x) y0 f (t , ym (t )) dt , where m 0,1, 2,3...
x0
(2)
52
ym ( x) exist as constant function in J n
At first we shall show that, the successive approximation
since
y0 ( x ) is constant x : x x0
x0
Also
y1 ( x) y0
x
| f (t , y0 (t )) dt M | x x0 | Mh b
( L | x x0 |) m 1
ym ( x) ym 1 ( x) N where m 1, 2,3...
( m 1)! (3)
x
yk 1 ( x) yk ( x ) x0
| f (t , yk (t )) f (t , yk 1 (t )) | dt
x
L x0
| yk (t ) yk 1 (t ) | dt
( L | t x0 |) k 1
x
L N
x0
( k 1)!
dt
53
( L | t x0 |) k 1
x
yk 1 ( x) yk ( x) L N dt
x0
( k 1)!
( L | x x0 |) k
N
k!
Next since
( L | x x0 |) m1
( Lh) m
N N Ne Lh
m 0 (m 1)! m0 m! (4)
y0 ( x) ( ym ( x) ym 1 ( x))
Applying Weierstrass’ M test in (4) it follows that the series m 1
y ( x) Lim ym ( x)
i.e m
x
y ( x) y0 f (t , y (t ))dt
x0
(5)
54
( x, z ( x)) s x J h
x
y ( x ) z ( x)
x0
| f (t , y (t )) f (t , z (t )) | dt
x
L | y (t ) z (t ) | dt
x0
| y ( x) z ( x) | 0 x J h
y ( x ) z ( x)
Theorem 9:
T : x x0 a, | y |
(i) f(x, y) is constant in the strip
y0 ( x) is constant in x x0 a
(iii)
Or, If the function f(x, y) satisfy the existence and uniqueness theorem for IVP (1) then the
successive approximation yn ( x) converges to the unique solution y(x) of the IVP (1).
55
y f ( x, y )
y ( x0 ) y0 (1)
x
ym1 ( x) y0 f (t , ym (t )) dt , where m 0,1, 2,3...
x0
(2)
At first we shall show that, the successive approximation ym ( x) exist in the strip
T :| x x0 | a, | y |
Since
y0 ( x) is constant x : x x0 <
( L | x x0 |) m 1
ym ( x) ym 1 ( x) N where m 1, 2,3...
( m 1)! (3)
56
x
yk 1 ( x) yk ( x ) x0
| f (t , yk (t )) f (t , yk 1 (t )) | dt
x
L x0
| yk (t ) yk 1 (t ) | dt
( L | t x0 |) k 1
x
L N
x0
( k 1)!
dt
( L | t x0 |) k 1
x
yk 1 ( x) yk ( x) L N
x0
( k 1)!
dt
( L | x x0 |) k
N
k!
Next since
( L | x x0 |) m1
( Lh) m
N N Ne Lh
m 0 (m 1)! m0 m! (4)
y0 ( x) ( ym ( x) ym 1 ( x))
Applying Weierstrass’ M test the series m 1 converges absolutely and
y ( x) Lim ym ( x)
i.e m
57
x
y ( x) z ( x )
x0
f (t , y (t )) f (t , z (t )) dt
x
L y (t ) z (t ) dt
x0
y ( x) z ( x) 0
Let,
y ( x) z ( x )
Theorem 10.1:
Let, f (t , x) be continuous in the domain D , then any solution of the IVP(1) is also a solution of
t
x(t ) x0 f ( s, x( s))ds
the integral equation t0
and conversely.
Proof:
x(t ) f ( y , x)
x(t 0 ) x0 (1)
Any solution x(t ) of the differential equation x(t ) f (t , x) converts it into an identity in t
58
t
x(t ) x(t 0 ) f ( s, x( s )) ds
t0
t
x(t ) x 0 f ( s, x ( s ))ds
t0
(3)
x(t0 ) x 0
Conversely suppose that, x(t ) is the solution of the IE (3), then
x(t ) 0 f (t , x(t ))
x(t ) f (t , x(t ))
Theorem 10.2:
Let the domain D be convex and the function f (t , x ) be differentiable with respect to x in D .
Then for the Lipchitz condition (Preliminaries to existence and uniqueness of solutions)
f (t , x1 ) f (t , x2 ) L x1 x2
to be satisfied,
Sup f (t , x)
L
D x
59
Proof:
(t , x1 ) (t , x2 ) D
Now, for all and
f (t , x* )
f (t , x1 ) f (t , x2 ) ( x1 x2 )
x (2)
Sup f (t , x)
L
Given, D x (3)
Using (3) in (2) we obtain (taking absolute value on both sides of (2))
f (t , x1 ) f (t , x2 ) L x1 x2
(4)
f (t , x1 ) f (t , x2 ) L x1 x2
f (t , x1 ) f (t , x2 )
L
x1 x2
Lt f (t , x1 ) f (t, x 2 )
L
x2 x4 x1 x2
60
f (x, y1 )
L
y1
Sup f ( x, y)
L
D y
[Proved]
61