Berhan International Research Journal of Science and Humanities

(BIRJSH), 2021, 5(1), 155 -165

Journal homepage: www.journals.dbu.et ISSN: 2414-2794

A review on phase plane and non-linear system of differential equations

with application

Asnake Tadese Abtew 1, Tesfaye Tefra Mamo 2*

1
Amhara Region, North Shoa Zone, Department of mathematics, Deneba high school, Deneba, Ethiopia
2
College of Natural and Computational Science, Department of Mathematics, Debre Berhan University,
Ethiopia

Abstract
In this paper we review on nonlinear phenomena and properties, particularly those with
physical relevance. Often, mathematical models of real-world phenomena are formulated in
terms of systems of nonlinear differential equations, which may be difficult to solve explicitly.
Finding a solution to a differential equation may not be so important if that solution never
appears in the physical model represented by the system, or is only realized in exceptional
circumstances. Thus, equilibrium solutions, which correspond to configurations in which the
physical system does not move, only occur in everyday situations if they are stable. An unstable
equilibrium will not appear in practice, since slight perturbations in the system or its physical
surroundings will immediately dislodge the system far away from equilibrium. We take a
qualitative approach to the analysis of solutions to nonlinear systems by making phase portraits
and using stability analysis.

Keywords: Phase portrait, Phase trajectory, Closed orbit, Separatrix, Eigenvalues,

Linearization
*
Corresponding author email: testef21@gmail.com
Article information: Received 05 February 2021; Revised 10 October 2021; Accepted 20 December 2021
© 2022 Debre Berhan University. All rights reserved.

Introduction
Differential equations first came into functions generally represent physical
existence with the invention of calculus by quantities, the derivatives represent their
physicist Sir Isaac Newton (1642 – 1727) rates of change, and the differential
and the mathematician Gottfried Wilhelm equation defines a relationship between the
Leibniz (1646 – 1716) (Newton, 1744). In two. Such relations are common; therefore,
mathematics, a differential equation is differential equations play a prominent role
an equation that relates one or more in many disciplines
unknown functions and their derivatives including engineering, physics, economics,
(Zill, 2012). In applications, the and biology. Differential equations can be
divided into several types. Commonly used

The International Research Journal of Debre Berhan University, 2021

 Abtew and Mamo BIRJSH 2021, 5(1), 155- 165

distinctions include whether the equation is nonlinear equations. These approximations

ordinary or partial, linear or non-linear, and are only valid under restricted conditions.
homogeneous or heterogeneous. An For example, the harmonic oscillator
ordinary differential equation (ODE) is an equation is an approximation to the
equation containing an unknown function nonlinear pendulum equation that is valid
of one real or complex variable y, its for small amplitude oscillations.
derivatives, and some given functions of x. Differential equations play an important
The unknown function is generally role in modeling virtually every physical,
represented by a variable (often denoted y), technical, or biological process; from
which, therefore, depends on x. Thus x is celestial motion to bridge design, to
often called the independent variable of the interactions between neurons (Johnson, E.,
equation. The term "ordinary" is used in and Johnson, J. R. 1965). Another
contrast with the term partial. A partial characteristic of differential equation in the
differential equation (PDE) is a differential 20th century was the creation of geometrical
equation that contains unknown or topological methods, especially for non-
multivariable functions and their partial Strogatz, S. H. (2018). Linear DEs
derivatives. PDEs can be used to describe a (Strogatz, S. H. 2018). The goal is to
wide variety of phenomena in nature such understand at least the qualitative behavior
as sound, heat, electrostatics, of solutions from a geometrical, as well as
electrodynamics, fluid flow, elasticity, or from an analytical point of view.
quantum mechanics. Linear differential Natural systems are highly non-linear.
equations are the differential equations that Interactions between individuals, species,
are linear in the unknown function and its or populations lead to relationships that
derivatives. A non-linear differential depend on variables in a way more
equation is a differential equation that is not complicated than that of simple
a linear equation in the unknown function proportionality. Among other things, this
and its derivatives. Non-linear differential means that models describing such
equations can exhibit very complicated phenomena contain non-linear equations
behavior over extended time intervals, that are often difficult, if not impossible, to
characteristic of chaos (complete disorder solve explicitly in closed form. Our aim is
and confusion) (Byatt-Smith, 1979 and to solve the limitation for calculating
156
Strogatz, 2018). explicitly solutions to non-linear ODE's.
Linear differential equations We would be focusing on determining
frequently appear as approximations to qualitative features of these solutions. The

The International Research Journal of Debre Berhan University, 2021

 Abtew and Mamo BIRJSH 2021, 5(1), 155- 165

approach is graphical and geometric; and (Bender, Orszag, S., and Orszag, S. A.
provides a description of the solutions 1999). All of these disciplines are
behavior, which allow us to understand the concerned with the properties of
phenomena captured in the modeling in a differential equations of various types. Pure
pictorial form. mathematics focuses on the existence and
Non-linear ordinary differential uniqueness of solutions, while applied
equation play an important role in many mathematics emphasizes the rigorous
branches of applied and pure mathematics justification of the methods for
and their applications in engineering approximating solutions. Many
,applied mechanics ,quantum physics fundamental laws
,analytical chemistry ,astronomy and of physics and chemistry can be
biology (Strogatz, 2018). From last formulated as differential equations.
decades, researchers pay attentions towards In biology and economics, differential
analytical and numerical solutions of non- equations are used to model the behavior of
linear ordinary differential equations. complex systems.
Therefore, it becomes increasingly
Important preliminaries
important to be familiar with all traditional
Before analyzing systems of ordinary
and recently developed methods for solving
differential equations, we better first
non- linear ordinary differential equations.
establish the existence of solution of an
Realizing that non-linear differential is
ordinary differential equation which is
rarely solvable analytically and not yet
fundamental for further analyze. This is the
having the benefit of computers generate
purpose of the existence and uniqueness
solution numerically.The theory and
theory for systems of differential
applications of differential equations and
equations.
their system play an important role in
Theorem: (Existence and uniqueness of
modern dynamics. Such equations are
solution)
mathematical models of various real-life
Let R be a rectangular region in the 𝑥, 𝑦 −
physical phenomena.
plane defined by 𝑎 ≤ 𝑥 ≤ 𝑏 𝑎𝑛𝑑 𝑐 ≤ 𝑦 ≤
𝑑 that contains the point (𝑥0 , 𝑦0 ) in its
Materials and Methods
𝜕𝑓
interior. If 𝑓(𝑥, 𝑦) 𝑎𝑛𝑑 are continuous
𝜕𝑦 157
The study of differential equations is a wide
on R, then there exists some
field in pure and applied
interval 𝐼0 : (𝑥0 − ℎ, 𝑥0 + ℎ) , h>0,
mathematics, physics, and engineering

The International Research Journal of Debre Berhan University, 2021

 Abtew and Mamo BIRJSH 2021, 5(1), 155- 165

contained in [𝑎, 𝑏], and a unique function From a dynamical point of view a singular
𝑦(𝑥), defined in 𝐼0 , that is a solution of point is an equilibrium point, or fixed point,
initial value problem. because if we start at such appoint then we
𝑑𝑥
Definition: In mathematics, an autonomous remain there for all time 𝑡 because =
𝑑𝑡
system or autonomous differential 𝑑𝑦
= 0 . Singular points are of great
𝑑𝑡
equation is a system of ordinary
importance in phase plane analysis.
differential equations which does not
To study a singular point of a non-linear
explicitly depend on the independent
system, we focused attention on its
variable. When the variable is time, they are
immediate neighborhood by expanding
also called time-invariant systems. It is the
𝑑𝑦 𝑓(𝑥, 𝑦) and 𝑔(𝑥, 𝑦) in Taylor series about
form of = 𝑓(𝑦).
𝑑𝑡 that point and linearizing; that is, cutting
Definitions:- In applied mathematics, in
them off after the first –order terms. Thus,
particular the context of nonlinear system
using Taylor’s expansion series, 𝑓(𝑥, 𝑦)
analysis, a phase plane is a visual display of
and 𝑔(𝑥, 𝑦) are approximated at
certain characteristics of certain kinds
equilibrium point (𝑥0 , 𝑦0 ) as:
of differential equations; a coordinate plane
𝜕𝑓
𝑓(𝑥, 𝑦) ≈ 𝑓(𝑥0 , 𝑦0 ) + (𝑥 , 𝑦 )[𝑥 − 𝑥0 ]
with axes being the values of the two state 𝜕𝑥 0 0
variables, say (x, y), or (any pair of 𝜕𝑓
+ (𝑥 , 𝑦 )[𝑦 − 𝑦0 ]
variables). It is a two-dimensional case of 𝜕𝑦 0 0

the general n-dimensional phase space. 𝜕𝑔

𝑔(𝑥, 𝑦) ≈ 𝑔(𝑥0 , 𝑦0 ) + (𝑥 , 𝑦 )[𝑥 − 𝑥0 ]
𝜕𝑥 0 0
Definition:- A phase portrait or integral
𝜕𝑔
curve is a geometric representation of the + (𝑥 , 𝑦 )[𝑦 − 𝑦0 ]
𝜕𝑦 0 0
trajectories of a dynamical system in But, at critical point 𝑓(𝑥0 , 𝑦0 ) =
the phase plane. Each set of initial 𝑔(𝑥0 , 𝑦0 ) = 0 . Hence, the non- linear
conditions is represented by a different
autonomous system is reduced to:
curve, or point. 𝑑𝑥 𝜕𝑓
= 𝑓(𝑥, 𝑦) = 𝜕𝑥 (𝑥0 , 𝑦0 )[𝑥 − 𝑥0 ]
𝑑𝑥 𝑑𝑡
= 𝑓(𝑥, 𝑦) 𝜕𝑓
Definition:- Let {𝑑𝑦𝑑𝑡 be a + 𝜕𝑦 (𝑥0 , 𝑦0 )[𝑦 − 𝑦0 ]
= 𝑔(𝑥, 𝑦 ) 𝑑𝑦 𝜕𝑔
𝑑𝑡 = 𝑔(𝑥, 𝑦 ) = (𝑥0 , 𝑦0 )[𝑥 − 𝑥0 ]
𝑑𝑡 𝜕𝑥
system of autonomous ordinary differential 𝜕𝑔
+ 𝜕𝑦 (𝑥0 , 𝑦0 )[𝑦 − 𝑦0 ]
equation. We say that any point
…..… … … … .. (𝜃) 158
(𝑥0 , 𝑦𝑜 ) which both f (𝑥, 𝑦) and g (𝑥, 𝑦)
This is the linearized system of non- linear
vanish that is; 𝑓(𝑥0 , 𝑦0 ) = 𝑔(𝑥0 , 𝑦0 ) = 0
system with coefficient matrix
is called a singular point (or critical point).

The International Research Journal of Debre Berhan University, 2021

 Abtew and Mamo BIRJSH 2021, 5(1), 155- 165

𝜕𝑓 𝜕𝑓 Case 2. If 𝜆1,2 are both real, district and

(𝑥0 , 𝑦0 ) (𝑥0 , 𝑦0 )
𝜕𝑥 𝜕𝑦 𝑎 𝑏
[𝜕𝑔 𝜕𝑔
]=[ ]. positive in sign, then the fixed point is
(𝑥0 , 𝑦0 ) (𝑥0 , 𝑦0 ) 𝑐 𝑑
𝜕𝑥 𝜕𝑦
unstable node. This means that all solution
This matrix is called Jacobean matrix.
curves of the two dimensional dynamical
Therefore, (𝜃) is expressed by standard
system go away from the equilibrium point
𝑑𝑋
form as: = 𝐴𝑋, where 𝑋= as shown in Figure 1.
𝑑𝑡
𝑥 𝑎 𝑏
(𝑦) 𝑎𝑛𝑑 𝐴 = ( ).
𝑐 𝑑
Stability and behavior of solutions of such
the system can be determined using
eigenvalues of A. The characteristic
equation is |𝐴 − 𝜆𝐼| = 0
⟹ 𝜆2 − (𝑎 + 𝑑)𝜆 + 𝑎𝑑 − 𝑏𝑐 = 0
Figure 2. The graph of unstable node to two
⟹ 𝜆2 − (𝑡𝑟𝐴)𝜆 + 𝑑𝑒𝑡𝐴 = 0
dimensional dynamical system with (0,0)
−𝑡𝑟𝐴±√(𝑡𝑟𝐴)2 −4𝑑𝑒𝑡𝐴
The solution are 𝜆1,2 = .
2 fixed point.
Based on the Eigen value of the solution of
characteristic equation, we get the Case 3. If 𝜆1,2 are both real, district and
following cases: opposite in sign, then the equilibrium point
Case 1. If 𝜆1,2 are both real, district and is saddle node. This means that all solution
negative in sign, then the equilibrium point curves of the two dimensional dynamical
is stable node. This means that all system asymptotic to the axes as shown in
solution curves of the two dimensional Figure 3.
dynamical system attract towards the
equilibrium point as shown Figure 2.

Figure 3. The graph of saddle node to two

dimensional dynamical system with (0,0)
Figure 1. The graph of stable node to two 159
fixed point.
dimensional dynamical system with (0,0)
fixed point.

The International Research Journal of Debre Berhan University, 2021

 Abtew and Mamo BIRJSH 2021, 5(1), 155- 165

Case 4. If 𝜆1,2 are complex, with real part Case 6. If 𝜆1,2 are pure imaginary, then
negative then, the fixed point is stable spiral fixed point is center. This means that all
focus. This means that all solution curves of solution curves of the two dimensional
the two dimensional dynamical system dynamical system makes a circle around the
spirally move towards the equilibrium point fixed point as shown in Figure 6.
as shown in Figure 4.

Figure 6. The graph of center to two

Figure 4. The graph of stable spiral focus
dimensional dynamical system with (0,0)
to two dimensional dynamical system with
fixed point
(0,0) fixed point
Some physical application

Case 5. If 𝜆1,2 are complex, with real part As indicated in Figure 7, let us consider the
positive then, the critical point is unstable pendulum that is released from the rest for
spiral focus. This means that all solution a rigid body undergoing pure rotation about
curves of the two dimensional dynamical a pivot axis. The inertia about the pivot O
system spirally go away from the fixed times the angular acceleration is equal to
point as indicated in Figure 5. the applied torque.

Figure 5. The graph of unstable spiral focus 160

to two dimensional dynamical system with
Figure 7. Pendulum
(0,0) fixed point

The International Research Journal of Debre Berhan University, 2021

 Abtew and Mamo BIRJSH 2021, 5(1), 155- 165

For the pendulum shown in Figure 7, the To study larger motions, suppose that we
inertia about O is 𝑚𝑙 2 , the angle from the approximate 𝑠𝑖𝑛 𝑥 by the first two terms of
vertical is 𝛼 (𝑡 ), the angular acceleration is its Taylor series, that is: 𝑠𝑖𝑛𝑥 ≈ 𝑥 −
𝛼 "(𝑡), and the down ward gravitational 𝑥3
.Then we have the non-linear, but still
6
force 𝑚𝑔 gives a torque of −𝑚𝑔𝑙𝑠𝑖𝑛𝑥. If
approximate, equation of motion " +
the air resistance is proportional to the 1
𝑥 − 𝑥3 = 0
6
velocity 𝑙 𝑥′ , say 𝑐𝑙 𝑥′ , then it gives an
The latter is of the same form as the
additional torque – 𝑐𝑙2𝑥′, where 𝛼 = 𝑥; so,
equation governing the rectilinear motion
the equation of motion is:
of a mass restrained by “a soft spring” .The
𝑚𝑙2𝑥" = −𝑚𝑔𝑙 𝑠𝑖𝑛 𝑥 − 𝑐𝑙2𝑥′𝑜𝑟 𝑥" +
system
𝑔
𝑟𝑥′ + 𝑠𝑖𝑛 𝑥 = 0 ….. (𝛽) 𝑥′ = 𝑦
𝑙
𝑐 { 𝑥3 …… (𝛾)
Where 𝑟 ≡ 𝑚 . The 𝑟𝑥′ term is damping 𝑦 ′ = −𝑥 +
6
𝑔
term. For definiteness, let 𝑙 =1 and Consider (𝛾) Has a center at (0, 0) and saddles at

the following cases. (±√6, 0) in the (𝑥, 𝑦) phase plane and its

Case 1. The undamped case where 𝑟 = 0 phase portrait is shown in Figure 9.

For small motions we can approximate

𝑠𝑖𝑛𝑥 by the first term of its Taylor
series, 𝑠𝑖𝑛 𝑥 ≈ 𝑥 , so that we have the
simple harmonic oscillator equation:
𝑥′ = 𝑦
𝑥" + 𝑥 = 0 Or { …….(𝜖).
𝑦 ′ = −𝑥
The equilibrium point of (𝜖) is (𝑥, 𝑦) = Figure 9. Phase portrait for the improved

(0,0) and its phase portrait is given in the model (large motion)

Figure 8.
Finally, if we keep 𝑠𝑖𝑛𝑥 intact, then we
have the full non- linear system:
𝑥′ = 𝑦
{ ……(𝜏)
y′ = − sinx
With singular points at (𝑥, 𝑦) = (𝑛𝜋, 0)
for = 0, ±1, ±2, …. . To classify these 161
singularities, let us linearize (𝜏) about the
Figure 8. Stable center for damped case singular point (𝑥, 𝑦) = (𝑛 𝜋 , 0) using

The International Research Journal of Debre Berhan University, 2021

 Abtew and Mamo BIRJSH 2021, 5(1), 155- 165

linearization techniques. Doing so, deviates somewhat from simple harmonic.

knowing that sin n 𝜋 =0 and 𝑐𝑜𝑠𝑛𝜋 = If we start at C, then we approach the saddle
(−1)𝑛 and setting 𝑋 = 𝑥 − 𝑛𝜋 and at 𝑥 = 𝜋 as 𝑡 → ∞; that is, the pendulum
𝑌 = 𝑦 − 0 = 𝑦, the linearized version of approaches the inverted position as 𝑡 → ∞.
(𝜏) is: If we impart more energy by starting at D,
𝑋′ = 𝑌 then, even though it slows down as it
{ .
𝑌 ′ = (−1)𝑛+1 𝑋
approaches the inverted position, it has
Here, we have centers at 𝑥=
enough energy to pass through that position
0, ±2𝜋, ±4𝜋, …. and saddles at 𝑥 =
and to keep going round and round
±𝜋, ±3𝜋, … on the x axis. indefinitely.
we have centers at x=0,±2p,±4p ,… and 𝑐
Case 2. Damping case where 𝑟 = 𝑚 ≠ 0
saddles at x=±p,±3p,… on the x axis.
With damping, the system is
𝑔
𝑥 ′′ + 𝑟𝑥 ′ + 𝑠𝑖𝑛𝑥 = 0,
𝑙
𝑐
where 𝑟 = 𝑚 , and with small angle

approximation, 𝑠𝑖𝑛𝑥 ≈ 𝑥 and the system is

𝑐 𝑔
𝑥 ′′ + 𝑚 𝑥 ′ + 𝑙 𝑥 = 0. Let 𝑥 ′ = 𝑦.Then, we

have:
Figure 10. phase portrait of the full 𝑥′ = 𝑦
{ ′ 𝑔 c ….. (𝜙).
nonlinear system 𝑥" + sin 𝑥 = 0. 𝑦 = − 𝑙 𝑥 − m𝑦

The equilibrium point of the system is

To understand the phase portrait as (𝑥, 𝑦) = (0,0) .
indicated in Figure 10, suppose that the Linearized equation of the system in the:
pendulum is hanging straight down (𝑥 = 𝑥′ = 𝑦
down position is { ′ 𝑔 𝑐 and up
0) initially, and that we impart an initial 𝑦 = − 𝑙 𝑥 −𝑚𝑦
angular velocity 𝑦 (0), so that the initial 𝑥′ = 𝑦
position is { ′ 𝑔 𝑐 . The
point is A, B, C, or D in Figure 11. If we 𝑦 = 𝑙 𝑥 −𝑚𝑦
start at A, then we follow a closed Orbits corresponding Eigen values are λ1, 2
that is very close to elliptical, and the 𝑐 𝑐 𝑔 𝑐 𝑐 𝑔
− ±√( )2 −4( ) − ±√( )2 +4( )
𝑚 𝑚 𝑙 𝑚 𝑚 𝑙
motion is very close to simple harmonic = and λ 1, 2 =
2 2

motion at frequency 𝜔 = 1. If we start at respectively. 162

B, the orbit is not so elliptical, there is an Now in the down position, we focus on
increase in the period and the motion character of fixed point (0,0) in

The International Research Journal of Debre Berhan University, 2021

 Abtew and Mamo BIRJSH 2021, 5(1), 155- 165

dependence of the values of 𝑙, 𝑐, 𝑔. For their negative and different eigenvalues, trivial
different values, phase portraits of system solution is asymptotically stable,
will be displayed and stability of trivial equilibrium point is stable node and the
solution will be determined. There are three phase portrait is indicated in Figure 12.
cases depending on discriminant part.
I. Critical damped pendulum

Critical damping in this case is occur when

the discriminant part is zero. That is,
𝑔
𝑐 = 𝑐𝑐𝑟 =2𝑚√ 𝑙 . Hence value of damping

coefficient is equal to value of critical

damping coefficient. Since all coefficients Figure 12. phase portrait of stable node
must be positive, there exists singe real for over damped pendulum
negative eigenvalues III. Under damped pendulum
𝑐
λ1 = - 𝑚 trivial solution is asymptotically
In the underdamped pendulum, the
stable, equilibrium point is stable node and discriminant part is negative. That is, 𝑐 <
the phase portrait is shown in Figure 11. 𝑔
𝑐𝑐𝑟 = 2𝑚√ 𝑙

Hence the value of damping coefficient is

lower than the value of critical damping
coefficient. Since all coefficients must be
positive, there exists two eigenvalues which
are complex conjugate, their real part is
negative, trivial solution is asymptotically
Figure 11. phase portrait of stable node for
stable, equilibrium is stable spiral focus and
critical damped pendulum
the phase portrait is displayed in Figure 13.
II. Over damped pendulum

Over damping occurs when the

discriminant part is positive. That is 𝑐 >
𝑔
𝑐𝑐𝑟 = 2𝑚√ 𝑙 . Hence, value of damping
163
coefficient is higher than value of critical
damping coefficient. Since all coefficients
must be positive, there exists two real

The International Research Journal of Debre Berhan University, 2021

 Abtew and Mamo BIRJSH 2021, 5(1), 155- 165

Figure 13. Phase portrait of stable spiral in equations. Due to the difficulty of solving
focus for underdamped pendulum. non-linear differential equations,
Finally, if we retain the entire Taylor series approximation methods have been
(i.e., if we keep 𝑠𝑖𝑛𝑥 intact), then we have developed. Qualitative method has been
𝑐
the full non- linear system, with 𝑟 = 𝑚, or discussed along with several linear and

𝑥′ = 𝑦 non-linear techniques to approximate or

{ ′ 𝑔 …… (𝜑) solve the non-linear problems. Singular
𝑦 = − 𝑙 𝑠𝑖𝑛𝑥 − 𝑟𝑦
points, stability, relationship between
With singular points at (𝑥, 𝑦) = (𝑛𝜋, 0)
Portrait diagram and solution of non-linear
for = 0, ±1, ±2, … . To classify these
differential equations were our particular
singularities about the singular point (𝑥, 𝑦)
focus with examples of each technique
= (𝑛 𝜋 , 0), knowing that 𝑠𝑖𝑛 𝑛𝜋 = 0 and
shown. A brief history Taylor’s expansion
𝑐𝑜𝑠𝑛𝜋 = (−1)𝑛 , then the linearized
and Jacobean method was discussed and
version has the phase portrait as indicated
highlighted the recent developments and
in Figure 14.
applications that these expansions provide
2. Visuals were used to demonstrate the
accuracy, or lack thereof, of each
technique. These techniques are integral in
applied mathematics and the correct project
allow us to see the behavior of a differential
equation when the exact solution may not
be attainable. Over all, this paper has
demonstrated a few techniques that can be
used to approximate non-linear differential
Figure 14. phase portrait for the entire equations that can be implemented when
𝑔
damped pendulum 𝑥 ′′ + 𝑟𝑥 + 𝑙 𝑠𝑖𝑛𝑥 = 0 the exact solution may not be achievable.

Conclusion References
Non-linear differential equations are Bender, C. M., Orszag, S., & Orszag, S. A.
(1999). Advanced mathematical methods for
incorporated in mathematical models in scientists and engineers I: Asymptotic methods
numerous instances. We have discussed and perturbation theory (Vol. 1). Springer 164
Science & Business Media.
various methods of solving and
approximating non-linear differential

The International Research Journal of Debre Berhan University, 2021

 Abtew and Mamo BIRJSH 2021, 5(1), 155- 165

Strogatz, S. H. (2018). Nonlinear dynamics and Equations (Oxford Applied Mathematics and
chaos: with applications to physics, biology, Computing Series). Proceedings of the
chemistry, and engineering. CRC press. Edinburgh Mathematical Society, 22(1), 71-71.

Johnson, D. E., & Johnson, J. R. (1965). Zill, D. G. (2012). A first course in differential
Mathematical Methods in Engineering and equations with modeling applications. Cengage
Physics: Special Functions and Boundary-value Learning.
Problems. Ronald Press Company.
Newton, I. (1744). Methodus fluxionum et
Byatt-Smith, J. G. (1979). DW Jordan and P. serierum infinitarum (the method of fluxions
Smith, Nonlinear Ordinary Differential and infinite series). Opuscula, 1(1744), 66.

165

The International Research Journal of Debre Berhan University, 2021