Chapter 10

Qualitative Theory of Differential

Equations

by

Dwijendra Narain Pandey

Department of Mathematics
Indian Institute of Technology Roorkee
Roorkee 247667, Uttarakhand.
E-mail: [Link]@[Link]
Module 1: Introduction.

1 Introduction
In this chapter we will discuss the qualitative properties of solutions of differ-
ential system
   
x1 f1 (t, x1 , · · · , xn )
x0 = f (t, x), where x =  ...  , and f (t, x) =  ..
 (1)
   
.
xn fn (t, x1 , · · · , xn )
is a nonlinear function of xi , i = 1, · · · , n.
There are several methods available to find a solution of linear system of differ-
ential equation but the same is not true for nonlinear equations, in-fact there is
no (general) method available to solve nonlinear equation (1).
But do we really need to have a explicit solution all the time that is there
are many instances where we are not interested in finding a solution rather than
we are interested in qualitative behavior of the solution.

For example, consider a population model in which two species lives together to
compete among themselves for the available living space and limited food. Also
let x1 (t) and x2 (t) be the rates of growth in population and are connected by
the system of differential equation (1) for n = 2. So for the study of such model
we are not much interested in knowing the exact population of each species
rather we want to study their respective behavior patterns. In particular, we
may think about the following three possibilities.
• Is there any possibility that their battle ends in a draw i.e., the numbers
x1 (t) ≡ α and x2 (t) ≡ β forms a set of constant solutions of (1). For such
a constant solutions, there respective growth rates are stationary so we
call these constant solutions as stationary solutions of (1). Other popu-
lar names of stationary solutions are steady state solutions, equilibrium
solutions, or critical solutions. We prefer to use the word equilibrium
solution.
• Suppose we have a set of equilibrium solution that is both the species
are living in a draw/equilibrium state, and we facilitate one species by
adding a small number of members of one species say species one, then
what is going to happen with the perturbed situation. Is there any effect
on the equilibrium state or facilitating species one will result in winning
the battle by species one or otherwise?
• Starting from a set of initial values at a time t = t0 , what is there asymp-
totic behavior i.e., there behavior as time t → ∞. Is there any possibility
of a draw or battle will end in annihilation of a species.

1
 These fundamental queries have a very important role in study of qualitative
behavior of a solution of nonlinear differential equations. The first one is related
to the existence of equilibrium solutions, second one is about the stability of an
equilibrium solution and the last one is about the asymptotic behavior of the
solution. Finding the answer of the first query
 is very easy as at the equilibrium
x01
state solution x0 (t) x00 (t) = 0 =  ...  , and it is possible if and only if
 

x0n
f (t, x0 ) ≡ 0.
Example 1. For the differential system
x01 = a11 x1 + a12 x2
x02 = a21 x1 + a22 x2 , a11 a22 − a21 a12 6= 0 (2)
2
there is only one critical point, namely (0, 0) in D = R .
Example 2. For the simple undamped pendulum system
x01 = x2 ,
g
x02 = −( ) sin x1
L
there is an infinite number of critical points (nπ, 0), n = 0, ±1, . . . in D = R2 .

2 Critical Points:
Here in this section we will consider the following few definition (taken from the
book titled Differential Equation by Shepley L. Ross) regarding types of critical
points and stability of critical points.

Definition 1. Given the autonomous system

dx
= P (x, y)
dt
dy
= Q (x, y) . (3)
dt
A point (x0 , y0 ) at which both
P (x0 , y0 ) = 0 and Q (x0 , y0 ) = 0
is called a critical point (equilibrium or singular point) of (3).

Definition 2. Isolated Critical Point: A critical point (x0 , y0 ) of the sys-

tem (3) is called isolated if there exist a constant r > 0 such that a circular
neighborhood
2 2
Br (x0 , y0 ) := {(x, y) ∈ R2 : (x − x0 ) + (y − y0 ) < r2 }
about the point (x0 , y0 ) does not contain any other critical points of (3).

2
 Figure 1: Approaching Path

Figure 2: Entering Path

Definition 3. Approaching path: Let C be a path of the system (3), and let
x = f (t), y = g(t) be a solution of (3) which represents C parametrically. Let
(0, 0) be a critical point of (3). We shall say that the path C approaches the
critical point (0, 0) as t → +∞ if

lim f (t) = 0 and lim g(t) = 0

t→+∞ t→+∞

Definition 4. Entering path: Let C be a path of the system (3) which ap-
proaches the critical point (0, 0) of (3) as t → +∞, and let x = f (t), y = g(t)
be a solution of (3) which represents C parametrically. We say that C enters
the critical point (0, 0) as t → +∞ if
g(t)
lim (4)
t→+∞ f (t)
exists or if the quotient in (3) become either positively or negatively infinite as
t → +∞.
Definition 5. Center: The isolated critical point (0, 0) of (3) is called a center
if there exist a neighborhood of (0, 0) which contains a countably infinite number
of closed paths Pn (n = 1, 2, 3, ...), each of which contains (0, 0) in its interior,
and which are such that the diameters of the paths approach 0 as n → ∞ [but
(0, 0) is not approached by any path either as t → +∞ or t → −∞ ].

3
 Figure 3: Center

Figure 4: Saddle Point

Definition 6. Saddle point: The isolated critical point (0, 0) of (3) is called
a saddle point if there exist a neighborhood of (0, 0) in which the following two
conditions hold:
1. There exists two paths which approach and enter (0, 0) from a pair of op-
posite directions as t → +∞, and there exists two paths which approach
and enter (0, 0) from a different pair of opposite directions as t → −∞.

2. In each of the four domains between any two of the four directions in
3 there are infinitely many paths which are arbitrarily close to (0, 0) but
which do not approach (0, 0) either as t → +∞ or t → −∞.

Definition 7. Spiral point: The isolated critical point (0, 0) of (3) is called a
spiral point (or focal point) if there exist a neighborhood of (0, 0) such that every
path P in this neighborhood has the following properties:
1. P is defined for all t > t0 (or for all t < t0 ) for some number t0 .

2. P approaches (0, 0) as t → +∞ (or as t → −∞).

4
 Figure 5: Stable Spiral point

Figure 6: Stable Node

3. P approaches (0, 0) in a spiral-like manner, winding around (0, 0) an infi-

nite number of times as t → +∞ (or as t → −∞).

Definition 8. Node: The isolated critical point (0, 0) of (3) is called a node if
there exist a neighborhood of (0, 0) such that every path P in this neighborhood
has the following properties:
1. P is defined for all t > t0 (or for all t < t0 ) for some number t0 .

2. P approaches (0, 0) as t → +∞ (or as t → −∞).

3. P enters (0, 0) as t → +∞ (or as t → −∞).

Definition 9. Stable: Assume that (0, 0) is an isolated critical point of the

system (3). Let C be a path of (3); let x = f (t), y = g(t) be a solution of (3)
defining C parametrically. Let
p
D(t) = [f (t)]2 + [g(t)]2 (5)

denote the distance between the critical point (0, 0) and the point R:[f (t), g(t)]
on C. The critical point (0, 0) is called stable if for every number  > 0, there
exists a number δ > 0 such that the following is true: Every path C for which

D(t0 ) < δ f or some value t0 (6)

5
 Figure 7: Stability

is defined for all t ≥ t0 and is such that

D(t) <  f or t0 ≤ t < ∞. (7)

If (0,0) is stable, then every path C which is inside the circle K1 of radius δ
at t = t0 , will remain inside the circle of radius K2 for t ≥ t0 .
Definition 10. Asymptotically stable: Assume that (0, 0) is an isolated crit-
ical point of the system (3). Let C be a path of (3); let x = f (t), y = g(t) be a
solution of (3) representing C parametrically. Let
p
D(t) = [f (t)]2 + [g(t)]2 (8)

denote the distance between the critical point (0, 0) and the point R:[f (t), g(t)]
on C.
The critical point (0, 0) is called asymptotically stable if (1) it is stable and (2)
there exists a number δ0 > 0 such that if

D(t0 ) < δ0 (9)

for some value t0 , then

lim f (t) = 0, lim g(t) = 0 (10)

t→+∞ t→+∞

Every path that gets sufficiently close to (0,0) ultimately approaches (0,0) as
t → +∞.
Definition 11. Unstable: A critical point is called unstable if it is not stable.