Other areas in Physics Revista Mexicana de Fı́sica 67 051401 1–18 SEPTEMBER-OCTOBER 2021

Numerical solution of a fractal-fractional order chaotic circuit system

Muhammad Altaf Khana,∗ , A. Atangana a,b , Taseer Muhammad c , and Ebraheem Alzahranid
a
Institute for Groundwater Studies, University of the Free State, Bloemfontein 9301, South Africa.
∗
e-mail: [email protected]
b
Department of Medical Research China Medical University Hospital,
China Medical University, Taichung, Taiwan.
c
Department of Mathematics, College of Sciences,
King Khalid University, Abha 61413, Saudi Arabia.
d
Department of Mathematics, Faculty of Science, King Abdulaziz University,
P.O. Box 80203, Jeddah 21589, Saudi Arabia.
Received 19 March 2021; accepted 22 April 2021

The dynamical system has an important research area and, due to its wide applications, many researchers and scientists are working to
develop new models and techniques for their solution. In this work, we present in this work the dynamics of a chaotic model in the presence
of newly introduced fractal-fractional operators. The model is formulated initially in ordinary differential equations, and then we utilize
the fractal-fractional (FF) in power law, exponential, and Mittag-Leffler to generalize the model. For each fractal-fractional order model,
we briefly study its numerical solution via the numerical algorithm. We present some graphical results with arbitrary order of fractal and
fractional orders, and present that these operators provide different chaotic attractors for different fractal and fractional order values. The
graphical results demonstrate the effectiveness of the fractal-fractional operators.

Keywords: Fractal-fractional operators; chaotic model; numerical results.

PACS: 05.45.-a; 05.40.-a; 02.60.Cb

DOI: https://doi.org/10.31349/RevMexFis.67.051401

1. Introduction of fractional derivative to zoonotic disease [10], application

to dengue epidemic, see [11], tuberculosis, see [12–14], and
the Chua model [15] has been investigated in arbitrary deriva-
The fractional differential equation (FDE) and the fractional tive.
calculus (FC) are now the most growing research areas and
receiving good feedback from scientists and authors work- Mostly the problems formulated in scientific and engi-
ing in modeling real-life problems in the scientific and en- neering areas are of high nonlinear nature, and its exact solu-
gineering fields. It is well-known that the fractional differ- tion is not possible. The solution of a fractional differentiable
ential equation has many applications in scientific and en- equation is also very complicated due to its order and difficult
gineering sectors. In particular, the dynamical systems and to solve analytically, and thus the researchers always look-
the chaos theory are also getting much attention from the re- ing to have their solution numerically. Fractional differential
searchers due to their wide applications in many fields such equation and its application to nonlinear problems of differ-
as autonomous mobile robots, chaotic communication, image ent fields with novel numerical investigations have been stud-
watermarking, and circuits, etcetera [1–3]. Many researchers ied in Refs. [10,16,17] and the references therein. Nowadays
studied the chaotic and other models in fractional operators the researchers from different areas developing mathematical
see for example [4-15]. These operators in power law, ex- models in fractional calculus due to its applications. One of
ponential decay, and Mittag-Leffler kernel were utilized suc- the interesting applications is the best fitting to the real cases
cessfully to present the various models of practical purpose. for an epidemic disease and another real data setting that
For example, the authors considered in Ref. [4] the chaotic fi- come from different areas of engineering and sciences. The
nancial model and its control strategy. The time delay chaotic memory and the heredity properties are attached to the frac-
model for the circuit and its synchronization has been in- tional operators and can best determine the dynamics of dy-
vestigated in Ref. [5]. The fractal-fractional (FF) operators namical systems and other related areas problems efficiently
introduced recently has been getting an impressive response than the classical orders. It is the advantage of fractional cal-
from researchers. Using the idea of FF and its application culus that can determine well the analysis between any two
to the existing chaotic models in the literature are studied in suggested points. Some recent work related to fractional op-
Ref. [6]. The new chaotic system in fractional derivative and erators and their applications can be seen in Refs. [18–20].
its comparative study is suggested in Ref. [7], while the ap- In this work, the authors focused on studying the circuit sys-
plication of fractional calculus to African tortoise and image tem, fractional-order capacitor, and sound encryption system
processing has been discussed in Ref. [8, 9]. The application using arbitrary order derivatives. The fractal order deriva-
2 MUHAMMAD ALTAF KHAN, A. ATANGANA, TASEER MUHAMMAD, AND EBRAHEEM ALZAHRANI

tive and the fractional-order, as well as their combinations 2. Fundamental of FF operators

for the solution of physical problems have been suggested
in Refs. [21–26]. The chaotic Shinriki’s oscillator fractional The present section determines the necessary results associ-
model has been investigated by the authors in Ref. [21]. In ated with fractal-fractional differential and their correspond-
Ref. [22], the authors studied the periodicity and multiple at- ing integral [6, 36].
tractors for Villis model. The idea of FF has been considered Definition 1. A function g(t) with fractal order β and
to study influenza and malaria model in Refs. [23, 24]. The fractional order α is continuous and differentiable over
fractal-fractional operators have been applied effectively for (m1 , m2 ), then, in a power-law sense, the following is pre-
the related problems, see [25–27], where the authors obtain sented:
results for new chaotic attractors, the study of the Hepatitis C Zt
³ ´ 1 d
model, and for system identification. Some more interesting F F P α,β
D0,t g(t) = (t − δ)
m−α−1
g(δ)dδ,
work, where the authors proposed the fractional derivatives Γ(k − α) dtβ
0
as an application to the scientific problems, see [28–35]. For
example, the author in Ref. [28] used the Laplace transform with k − 1 < α, β ≤ k ∈ N and dg(t)/dtβ =
for getting the solution of a fractional telegraph equation. limt1 →t (g(t1 ) − g(t)/tβ1 − tβ ).
The formulation of the tumor fractional model using non- Definition 2. A function g(t) with fractal order β and
singular kernel has been studied in Ref. [29]. In Ref. [30], fractional order α is continuous and differentiable over
the authors studied the solution of the fifth-order evolution (m1 , m2 ), then, in exponential decay kernel, the following
equation in a fractional derivative. Using the Haar wavelet is shown:
method, the authors in Ref. [31] obtained the solution of ³ ´ M (α) d
F F E α,β
the Lotka-Volterra model related to population in a fractional D0,t g(t) =
derivative. Using Yang-Abdel-Aty-Cattani fractional opera- 1 − α dtβ
tor, the authors discussed the solution of heat equations in Zt ³ ´
α
Ref. [32]. The classical optical solution for the propaga- × exp − (t − δ) g(δ)dδ, (1)
tion model has been derived in Ref. [33]. The two interact- 1−α
0
ing species model using numerical method is discussed in
Ref. [34]. A fractional-order SEIR model using Genocchi with M (0) = M (1) = 1 and for positive α with β ≤ m ∈ N.
has been studied in Ref. [35]. Definition 3. A function g(t) with fractal and fractional or-
der β and α respectively, is continuous and differentiable
In this investigation, we considered the FF designed op- over (m1 , m2 ), then, the Mittag-Leffler kernel follows the for-
erators suggested in the sense of power law, exponential de- mulate given by:
cay, and the Mittag-Leffler kernel. These suggested opera-
tors were found interesting from a researchers point of view FFM α,β AB(α) d
D0,t (g(t)) =
and have been found reliable for chaotic order systems. Us- 1 − α dtβ
ing these operators, many hidden attractors have been found Zt ³ ´
α
which were not possible for ordinary fractional operators. It × Eα − (t − δ)α g(δ)dδ, (2)
is clear that these operators attract various attractors by con- 1−α
0
sidering their different orders. Recently, fractal-fractional or-
ders problems are formulated in Ref. [6, 36], etc. It is well- with α > 0, β ≤ 1 ∈ N and AB(α) = 1 − α + (α/Γ(α)).
known that the results shown in Ref. [6, 36] are the applica- Definition 4. A function g(t) continuous over (m1 , m2 ), with
tions for a physical systems. Further, these operators were fractal order β and fractional order α then, for the power-law
found best for data fitting; one can see [36]. Moreover, the kernel, we have the following integral:
results suggested in Refs. [6, 36–38] demonstrate the com- Zt
binations of these FF operators provide many insights about ³ ´ β
FFP α
J0,t g(t) = (t − δ)α−1 δ β−1 g(δ)dδ. (3)
the dynamics, which can only be observed for the FF orders. Γ(α)
With the above many useful applications of the FF operators, 0

we give a chaotic model and present its analysis and provide Definition 5. A function g(t) continuous over (m1 , m2 ), with
a useful and effective scheme for its solution. We will find fractal order β and fractional order α then, the FF integral
many useful numerical results by considering the combina- exponentially decaying kernel for f (t) is described by the fol-
tions of the FF orders. lowing:
The work shown in this paper has been arranged section- Zt
FFE α αβ
wise is as follows. We provide the details of the FF operators J0,t (g(t)) = δ α−1 g(δ)dδ
and model description in Secs. 2 and 3. Section 4 presents the M (α)
0
novel solution technique for the FF model in Caputo, CF, and
AB operators. We discussed the simulation results in Secs. 5 β(1 − α)tβ−1 g(t)
+ . (4)
and 6, we summarized the finding. M (α)

Rev. Mex. Fis. 67 051401

 NUMERICAL SOLUTION OF A FRACTAL-FRACTIONAL ORDER CHAOTIC CIRCUIT SYSTEM 3

Definition 6. A function z(t) continuous on (a, b), with frac- is differentiable, so, we can use
tal order β and fractional order α then, for the Mittag-Leffler
Zt
kernel, the following integral is suggested:
F (t) = (t − δ)−α g(δ)dδ.
Zt 0
FFM α,β αβ
J0,t (g(t)) = δ β−1 (t − δ)α−1 g(δ)dδ
AB(α) So,
0
1 d
β(1 − α)tβ−1 g(t) F F P α,β
0 Dt g(t) = F (t),
+ . (5) Γ(1 − α) dtβ
AB(α)
1 F (t1 ) − F (t)
Lemma 1. The following result holds for fractional and frac- = lim ,
tal order α, and β, respectively,
Γ(1 − α) t1 →t tβ1 − tβ
1 F (t1 ) − F (t) t1 − t
Zt = lim ,
F F α,β β Γ(1 − α) t1 →t t1 − t tβ1 − tβ
0 Jt g(t) = tβ−1 (t − δ)α−1 g(δ)dδ.
Γ(α) 1 1
0 = F 0 (t) β−1 .
Γ(1 − α) βt
Proof. Consider
C α,β
0 Dt g(t) = u(t),
C α
0 Dt g(t) = βtβ−1 u(t). 3. The chaotic model
Riemann-Liouville integral gives the following, We consider a model which chaotic and has been presented in
Ref. [39]. The model has three equations, and its differential
Zt equations form considered in Ref. [39] is shown below:
1
g(t) = βtβ−1 (t − δ)α−1 u(δ)dδ,  dx(t)
Γ(α)  = y,


0
 dt
dy(t)
Zt dt = yz − x,
(7)
β 

g(t) = tβ−1 (t − δ)α−1 u(δ)dδ. 
 dz(t) 2
Γ(α) dt = 1 − y ,
0

Therefore, where x(t), y(t) and z(t) represent the state variables and be-
Z come chaotic for the initial values of the variables x(0) = 0,
F F α,β β y(0) = 1 and z(0) = 0. The fractal-fractional representation
0 Jt g(t) = t0tβ−1 (t − δ)α−1 g(δ)dδ.
Γ(α) of the model (7) is given by:
 ¡ ¢
 F F P α, β
 D0,t x(t) = y,


F F P α, β
¡ ¢
Lemma 2. The following result holds for fractional and frac- D0,t y(t) = yz − x, (8)


tal order α and β, respectively,  F F P α, β ¡
 ¢
D0,t z(t) = 1 − y 2 ,
Zt
F F P α,β 1 d where α represents the fractional order while β is fractal or-
0 Dt g(t) = (t − δ)−α g(δ)dδ. (6)
Γ(1 − α) dtβ der. No equilibria exist for the model (8). Further, in the
0
following, we are going to formulate the problem above us-
Proof. In order to show this result given by ing FF operators and present the numerical scheme for each
case.
Zt
F F P α,β 1 d
0 Dt g(t) = (t − δ)−α g(δ)dδ.
Γ(1 − α) dtβ 4. Numerical procedure for the chaotic model
0

Since This section presents briefly the procedure to handle the

Zt model formulated in FF with fractional operators sense, Ca-
(t − δ)−α g(δ)dδ puto, CF, and AB. One by one, we apply the fractal operator
0
to the model described above (8).

Rev. Mex. Fis. 67 051401

4 MUHAMMAD ALTAF KHAN, A. ATANGANA, TASEER MUHAMMAD, AND EBRAHEEM ALZAHRANI

4.1. Numerical scheme for FF-Caputo model

We start the model to provide a numerical algorithm by considering the model in power-law case. To start the algorithm,
we need first the write the model (8) in Volterra representations, because the fractional integral is differentiable, so in the
Riemann-Liouville (RL) sense, we have,
Zt
FFP α,β 1 d 1
D0,t f (t) = (t − δ)−α f (δ)dδ , (9)
Γ(1 − α) dt βtβ−1
0

we present the results in the following,

RL α
¡ ¢ £ ¤
D0,t x(t) = βtβ−1 y ,
RL α
¡ ¢ £ ¤
D0,t y(t) = βtβ−1 yz − x ,
RL α
¡ ¢ £ ¤
D0,t z(t) = βtβ−1 1 − y 2 . (10)
Consider now replacing the operator shown by Riemann-Liouville(RL) in the above system with the Caputo operator to get the
initial conditions in the integer case. Further, we get the result by using RL fractional integral:
Zt
β
x(t) = x(0) + δ β−1 (t − δ)α−1 f1 (x, y, z, δ)dδ,
Γ(α)
0

Zt
β
y(t) = y(0) + δ β−1 (t − δ)α−1 f2 (x, y, z, δ)dδ,
Γ(α)
0

Zt
β
z(t) = z(0) + δ β−1 (t − δ)α−1 f3 (x, y, z, δ)dδ, (11)
Γ(α)
0
where
f1 (x, y, z, δ) = y, f2 (x, y, z, δ) = yz − x, f3 (x, y, z, δ) = 1 − y 2 .
For t = tn+1 , we have the following,
tZ
n+1

n+1 β
0
x =x + δ β−1 (tn+1 − δ)α−1 f1 (x, y, z, δ)dδ,
Γ(α)
0
tZ
n+1
β
y n+1 = y 0 + δ β−1 (tn+1 − δ)α−1 f2 (x, y, z, δ)dδ,
Γ(α)
0
tZ
n+1
β
z n+1 = z 0 + δ β−1 (tn+1 − δ)α−1 f3 (x, y, z, δ)dδ. (12)
Γ(α)
0

Simplifying the above results further, we get

tZj+1
n
n+1 β X
0
x =x + δ β−1 (tn+1 − δ)α−1 f1 (x, y, z, δ)dδ,
Γ(α) j=0
tj

tZj+1
n
n+1 β X
0
y =y + δ β−1 (tn+1 − δ)α−1 f2 (x, y, z, δ)dδ,
Γ(α) j=0
tj

tZj+1
n
n+1 β X
0
z =z + δ β−1 (tn+1 − δ)α−1 f3 (x, y, z, δ)dδ, (13)
Γ(α) j=0
tj

Rev. Mex. Fis. 67 051401

 NUMERICAL SOLUTION OF A FRACTAL-FRACTIONAL ORDER CHAOTIC CIRCUIT SYSTEM 5

where t(0) = 0. The functions δ β−1 fi (x, y, z, δ) for i = 1, 2, 3 shown in Eq. (13) are approximated in the interval [tj , tj+1 ]
using the Lagrangian interpolation concept, and getting the following,

δ − tj−1 β−1 δ − tj β−1

Lj (δ) = t f1 (xj , y j , z j , tj ) − t f1 (xj−1 , y j−1 , z j−1 , tj−1 ),
tj − tj−1 j tj − tj−1 j−1
δ − tj−1 β−1 δ − tj β−1
Mj (δ) = tj f2 (xj , y j , z j , tj ) − t f2 (xj−1 , y j−1 , z j−1 , tj−1 ),
tj − tj−1 tj − tj−1 j−1
δ − tj−1 β−1 δ − tj β−1
Nj (δ) = t f3 (xj , y j , z j , tj ) − t f3 (xj−1 , y j−1 , z j−1 , tj−1 ). (14)
tj − tj−1 j tj − tj−1 j−1

Using the results shown in (14), we have the below system,

tZj+1
n
n+1 0 β X
x =x + δ β−1 (tn+1 − δ)α−1 Lj (δ)dδ,
Γ(α) j=0
tj

tZj+1
n
n+1 0 β X
y =y + δ β−1 (tn+1 − δ)ρ−1 Mj (δ)dδ,
Γ(α) j=0
tj

tZj+1
n
β X
z n+1 = z0 + δ β−1 (tn+1 − δ)ρ−1 Nj (δ)dδ. (15)
Γ(α) j=0
tj

Further solution of system (15) leads to the following,

βhα X h β−1 i
n
xn+1 0
=x + tj f1 (xj , y j , z j , tj )an,j − tβ−1
j−1 f1 (x
j−1 j−1 j−1
, y , z , tj−1 )bn,j ,
Γ(α + 2) j=1

βhα X h β−1 i
n
y n+1 = y 0 + tj f2 (xj , y j , z j , tj )an,j − tβ−1
j−1 f2 (xj−1 j−1 j−1
, y , z , t j−1 )b n,j ,
Γ(α + 2) j=1

βhα X h β−1 i
n
z n+1 = z 0 + tj f3 (xj , y j , z j , tj )an,j − tβ−1
j−1 f3 (x
j−1 j−1 j−1
, y , z , tj−1 )bn,j , (16)
Γ(α + 2) j=1

where
³ ´
an,j = (n + 1 − j)α (n − j + 2 + α) − (n − j)α (n − j + 2 + 2α) ,
³ ´
bn,j = (n − j + 1)α+1 − (n − j)α (n − j + 1 + α) ,

where n = 0 to N and j = 1 to n.

4.2. Numerical solution of FF-CF model

Now, we consider the model in the FF-CF case and present in detail the numerical scheme. For this, we need to write the model
in FF exponential decay form using the following structure:

CF α
¡ ¢
D0,t x(t) = βtβ−1 f1 (x, y, z, t),
CF α
¡ ¢
D0,t y(t) = βtβ−1 f2 (x, y, z, t),
CF α
¡ ¢
D0,t z(t) = βtβ−1 f3 (x, y, z, t). (17)

Rev. Mex. Fis. 67 051401

6 MUHAMMAD ALTAF KHAN, A. ATANGANA, TASEER MUHAMMAD, AND EBRAHEEM ALZAHRANI

F IGURE 1. FF-Caputo model results, when α = 1, β = 1, where (a-d) show different phase planes.

F IGURE 2. FF-Caputo model results, when α = β = 0.98, where (a-d) show different phase planes.

Rev. Mex. Fis. 67 051401

 NUMERICAL SOLUTION OF A FRACTAL-FRACTIONAL ORDER CHAOTIC CIRCUIT SYSTEM 7

F IGURE 3. FF-Caputo model results, when α = 0.98, β = 0.9, where (a-d) show different phase planes.

F IGURE 4. FF-Caputo model results, when α = 1, β = 0.5, where (a-d) show different phase planes.

Rev. Mex. Fis. 67 051401

8 MUHAMMAD ALTAF KHAN, A. ATANGANA, TASEER MUHAMMAD, AND EBRAHEEM ALZAHRANI

We obtain the following by using the Caputo-Fabrizio integral,

Zt
βtβ−1 (1 − α)
0 αβ
x(t) = x + f1 (x, y, z, t) + δ β−1 f1 (x, y, z, δ)dδ,
M (α) M (α)
0

Zt
βtβ−1 (1 − α)
0 αβ
y(t) = y + f2 (x, y, z, t) + δ β−1 f2 (x, y, z, δ)dδ,
M (α) M (α)
0

Zt
βtβ−1 (1 − α)
0 αβ
z(t) = z + f3 (x, y, z, t) + δ β−1 f3 (x, y, z, δ)dδ. (18)
M (α) M (α)
0

The following is presented at tn+1 :

tZ
n+1

n+1 0βtβ−1 (1 − α) αβ
x =x + f1 (xn , y n , z n , tn ) + δ β−1 f1 (x, y, z, δ)dδ,
M (α) M (α)
0
tZ
n+1

n+1 0βtβ−1 (1 − α) αβ
y =y + f2 (xn , y n , z n , tn ) + δ β−1 f2 (x, y, z, δ)dδ,
M (α) M (α)
0
tZ
n+1
βtβ−1 (1 − α) αβ
z n+1 = z 0 + f3 (xn , y n , z n , tn ) + δ β−1 f3 (x, y, z, δ)dδ. (19)
M (α) M (α)
0

Further, we have the following:

βtβ−1 (1 − α) βtβ−1 (1 − α)
xn+1 = x0 + f1 (xn , y n , z n , tn ) − n−1 f1 (xn−1 , y n−1 , z n−1 , tn−1 )
M (α) M (α)
tZ
n+1
αβ
+ δ β−1 f1 (x, y, z, δ)dδ,
M (α)
tn

βtβ−1 (1 − α) βtβ−1 (1 − α)
y n+1 = y 0 + f2 (xn , y n , z n , tn ) − n−1 f2 (xn−1 , y n−1 , z n−1 , tn−1 )
M (α) M (α)
tZ
n+1
αβ
+ δ β−1 f2 (x, y, z, δ)dδ,
M (α)
tn

βtβ−1 (1 − α) βtβ−1 (1 − α)
z n+1 = z 0 + f3 (xn , y n , z n , tn ) − n−1 f3 (xn−1 , y n−1 , z n−1 , tn−1 )
M (α) M (α)
tZ
n+1
αβ
+ δ β−1 f3 (x, y, z, δ)dδ. (20)
M (α)
tn

Rev. Mex. Fis. 67 051401

 NUMERICAL SOLUTION OF A FRACTAL-FRACTIONAL ORDER CHAOTIC CIRCUIT SYSTEM 9

F IGURE 5. FF-CF model results, when α = 1, β = 1, where (a-d) show different phase planes.

F IGURE 6. FF-CF model results, when α = 0.99, β = 1, where (a-d) show different phase planes.

Rev. Mex. Fis. 67 051401

10 MUHAMMAD ALTAF KHAN, A. ATANGANA, TASEER MUHAMMAD, AND EBRAHEEM ALZAHRANI

F IGURE 7. FF-CF model results, when α = 0.97, β = 1, where (a-d) show different phase planes.

F IGURE 8. FF-CF model results, when α = 1, β = 0.98, where (a-d) show different phase planes.

Rev. Mex. Fis. 67 051401

 NUMERICAL SOLUTION OF A FRACTAL-FRACTIONAL ORDER CHAOTIC CIRCUIT SYSTEM 11

F IGURE 9. FF-CF model results, when α = 0.98, β = 0.98, where (a-d) show different phase planes.

F IGURE 10. FF-CF model results, when α = 1, β = 0.5, where (a-d) show different phase planes.

Rev. Mex. Fis. 67 051401

12 MUHAMMAD ALTAF KHAN, A. ATANGANA, TASEER MUHAMMAD, AND EBRAHEEM ALZAHRANI

The idea of Lagrange polynomial is used to obtain the following result:

βtβ−1 (1 − α) βtβ−1 (1 − α)
xn+1 = x0 + n
f1 (xn , y n , z n , tn ) − n−1 f1 (xn−1 , y n−1 , z n−1 , tn−1 )
M (α) M (α)
· ¸
αβ 3h β−1 n n n h β−1 n−1 n−1 n−1
+ t f1 (x , y , z2 , tn ) − tn−1 f1 (x ,y ,z , tn−1 ) ,
M (α) 2 n 2
βtβ−1 (1 − α) βtβ−1 (1 − α)
y n+1 = y 0 + n
f2 (xn , y n , z n , tn ) − n−1 f2 (xn−1 , y n−1 , z n−1 , tn−1 )
M (α) M (α)
· ¸
αβ 3h β−1 n n n h β−1 n−1 n−1 n−1
+ t f1 (x , y , z2 , tn ) − tn−1 f2 (x ,y ,z , tn−1 ) ,
M (α) 2 n 2
βtβ−1 (1 − α) βtβ−1 (1 − α)
z n+1 = z 0 + n
f3 (xn , y n , z n , tn ) − n−1 f1 (xn−1 , y n−1 , z n−1 , tn−1 )
M (α) M (α)
· ¸
αβ 3h β−1 n n n h β−1 n−1 n−1 n−1
+ t f3 (x , y , z2 , tn ) − tn−1 f3 (x ,y ,z , tn−1 ) . (21)
M (α) 2 n 2

Further, we obtain the following,

³1−α 3αh ´ ³1−α αh ´
xn+1 = x0 + βtβ−1
n + f1 (xn , y n , z n , tn ) − βtβ−1
n−1 + f1 (xn−1 , y n−1 , z n−1 , tn−1 ),
M (α) 2M (α) M (α) 2M (α)
³1−α 3αh ´ ³ αh ´
β−1 1 − α
y n+1 = y 0 + βtβ−1
n + f2 (x n n n
, y , z , t n ) − βt n−1 + f2 (xn−1 , y n−1 , z n−1 , tn−1 ),
M (α) 2M (α) M (α) 2M (α)
³1−α 3αh ´ ³1−α αh ´
z n+1 = z 0 + βtβ−1
n + f3 (xn , y n , z n , tn ) − βtβ−1
n−1 + f3 (xn−1 , y n−1 , z n−1 , tn−1 ). (22)
M (α) 2M (α) M (α) 2M (α)

Further in next subsection, we use the FF model and present a numerical procedure in order to obtain their solution numer-
ically. We provide the details of the scheme in the below subsection.

4.3. Numerical solution of FF-AB model

In order to have a numerical procedure to handle the FF model in Atangana-Baleanu (AB) operator sense, the model shown in
(8) can be re-written using the following way:
³ ´
ABR α
D0,t x(t) = βtβ−1 f1 (x, y, z, t),
ABR α
¡ ¢
D0,t y(t) = βtβ−1 f2 (x, y, z, t),
³ ´
ABR α
D0,t z(t) = βtβ−1 f3 (x, y, z, t). (23)

The following is presented based on Atangana-Baleanu (AB) integral,

Zt
βtβ−1 (1 − α) αβ
x(t) = x(0) + f1 (x, y, z, t) + δ β−1 (t − δ)α−1 f1 (x, y, z, δ)dδ,
AB(α) AB(α)Γ(α)
0

Zt
βtβ−1 (1 − α) αβ
y(t) = y(0) + f2 (x, y, z, t) + δ β−1 (t − δ)α−1 f2 (x, y, z, δ)dδ,
AB(α) AB(α)Γ(α)
0

Zt
βtβ−1 (1 − α) αβ
z(t) = z(0) + f3 (x, y, z, t) + δ β−1 (t − δ)α−1 f3 (x, y, z, δ)dδ. (24)
AB(α) AB(α)Γ(α)
0

Rev. Mex. Fis. 67 051401

 NUMERICAL SOLUTION OF A FRACTAL-FRACTIONAL ORDER CHAOTIC CIRCUIT SYSTEM 13

Further, at tn+1 , we get,

tZ
n+1
βtβ−1 (1 − α αβ
xn+1
=x + n0
f1 (xn , y n , z n , tn ) + δ β−1 (tn+1 − δ)α−1 f1 (x, y, z, δ)dδ,
AB(α) AB(α)Γ(α)
0
tZ
n+1
βtβ−1
n (1 − α αβ
y n+1 = y 0 + f2 (xn , y n , z n , tn ) + δ β−1 (tn+1 − δ)α−1 f2 (x, y, z, δ)dδ,
AB(α) AB(α)Γ(α)
0
tZ
n+1
βtβ−1
n (1 − α αβ
z n+1 = z 0 + f3 (xn , y n , z n , tn ) + δ β−1 (tn+1 − δ)α−1 f3 (x, y, z, δ)dδ. (25)
AB(α) AB(α)Γ(α)
0

We obtain the following after simplifying the integral in the above equations:

tZj+1
X n
βtβ−1 (1 − α) αβ
x n+1
=x + n
0
f1 (xn , y n , z n , tn ) + δ β−1 (tn+1 − δ)α−1 f1 (x, y, z, δ)dδ,
AB(α) AB(α)Γ(α) j=0
tj

tZj+1
X n
βtβ−1 (1 − α) αβ
y n+1
=y + n
0
f2 (xn , y n , z n , tn ) + δ β−1 (tn+1 − δ)α−1 f2 (x, y, z, δ)dδ,
AB(α) AB(α)Γ(α) j=0
tj

tZj+1
X n
βtβ−1 (1 − α) αβ
z n+1
=z + n
0
f3 (xn , y n , z n , tn ) + δ β−1 (tn+1 − δ)α−1 f3 (x, y, z, δ)dδ. (26)
AB(α) AB(α)Γ(α) j=0
tj

Now, approximating the expressions in (26) given by δ β−1 f1 (x, y, z, δ), δ β−1 f2 (x, y, z, δ) and δ β−1 f3 (x, y, z, δ) in the given
interval [tj , tj+1 ], the following numerical scheme is presented,

βtβ−1
n (1 − α) β(∆t)α
xn+1 = x0 + f1 (xn , y n , z n , tn ) +
AB(α) AB(α)Γ(α + 2)
n h
X i
× tβ−1
j f1 (xj , y j , z j , tj )an,j − tβ−1
j−1 f1 (x
j−1 j−1 j−1
, y , z , tj−1 )bn,j ,
j=1

βtβ−1
n (1 − α) β(∆t)α
y n+1 = y 0 + f2 (xn , y n , z n , tn ) +
AB(α) AB(α)Γ(α + 2)
n h
X i
× tβ−1
j f2 (xj , y j , z j , tj )an,j − tβ−1
j−1 f2 (x
j−1 j−1 j−1
, y , z , tj−1 )bn,j ,
j=1

βtβ−1
n (1 − α) β(∆t)α
z n+1 = z 0 + f3 (xn , y n , z n , tn ) +
AB(α) AB(α)Γ(α + 2)
n h
X i
× tβ−1
j f 3 (xj
, y j j
, z , t j )an,j − t β−1
j−1 f3 (x j−1 j−1 j−1
, y , z , t j−1 )b n,j , (27)
j=1

where
³ ´
an,j = (n + 1 − j)α (n − j + 2 + α) − (n − j)α (n − j + 2 + 2α) ,
³ ´
bn,j = (n − j + 1)α+1 − (n − j)α (n − j + 1 + α) ,

where n = 0 to N and j = 1 to n.

Rev. Mex. Fis. 67 051401

14 MUHAMMAD ALTAF KHAN, A. ATANGANA, TASEER MUHAMMAD, AND EBRAHEEM ALZAHRANI

F IGURE 11. FF-AB model results, when α = 1, β = 1, where (a-d) show different phase planes.

F IGURE 12. FF-AB model results, when α = 0.99, β = 1, where (a-d) show different phase planes.

Rev. Mex. Fis. 67 051401

 NUMERICAL SOLUTION OF A FRACTAL-FRACTIONAL ORDER CHAOTIC CIRCUIT SYSTEM 15

F IGURE 13. FF-AB model results, when α = 0.99, β = 0.98, where (a-d) show different phase planes.

F IGURE 14. FF-AB model results, when α = 1, β = 0.5, where (a-d) show different phase planes.

Rev. Mex. Fis. 67 051401

16 MUHAMMAD ALTAF KHAN, A. ATANGANA, TASEER MUHAMMAD, AND EBRAHEEM ALZAHRANI

5. Discussion sense. We presented briefly the simulation of the chaotic

model by using novel numerical approaches. We obtained
We studied above the detailed algorithms for the solution of the numerical results presented for the solution of the fractal-
FF models numerically using the power law, exponential de- fractional model in the sense of three different fractal-
cay, and the Mittag-Leffler cases. For each operator, the fractional operators. We used the initial value for t = 0.1
scheme has been presented in a detailed manner. We use and the step size h = 0.01. We presented the graphical re-
the numerical algorithm shown above in order to obtain the sults for each operator by selecting some suitable values for
solution of the chaotic model in graphical form. Using the the fractal and fractional orders. We see that for each value
scheme presented in (16) for the FF system in Caputo sense, of the fractal and fractional order, different chaotic behav-
we used the scheme, and the corresponding graphical results iors of the model are observed. It is observed that by chang-
are shown in Figs. 1-4 with different fractal and fractional ing the values of the fractal and fractional orders, some new
orders. We see, for different orders of fractal and fractional, chaotic results are obtained graphically. This new analysis of
the model behaves chaotically. We also utilized a different the FF model in different operators bring new doors of inves-
set of fractal and fractional orders and obtained interesting tigations for scientists and researchers working in modeling
chaotic results. For the model formulated in FF-CF deriva- electrical circuits theory. The results reveal that the fractal-
tive, we considered the suggested scheme (22), and the solu- fractional operators and their application to a chaotic model
tion of the model has been shown graphically in Figs. 5-10, provide useful results that cannot be seen for an ordinary op-
with arbitrary orders of fractal and fractional orders. One can erator. These operators can be beneficial in modeling other
observe that different combinations of the suggested opera- engineering fields and sciences related areas.
tors provide useful numerical results. Further, the procedure
given in (27) is used for the numerical results of the fractal-
fractional Atangana-Baleanu model with arbitrary order of Conflict of interest
fractal and fractional order parameters values and obtain the
Figs. 11-14. From all these results, it can be concluded that No conflict of interest exists regarding the publications of this
the chaotic model provides different behaviors for the sug- work.
gested values of the fractal and fractional orders. Using these
fractal-fractional operators, many chaotic results and hidden
attractors can be studied effectively. These operators can be Acknowledgment
applied to other nonlinear problems arising from physical sci-
ences, social sciences, and engineering areas where one can The authors extend their appreciation to the Deanship of Sci-
find the best possible results for their proposed problems. entific Research at King Khalid University, Abha, Saudi Ara-
bia for funding this work through general research groups
6. Conclusion program under grant number G.R.P-150/42.

We presented a chaotic model in FF operators where the

fractional operators were considered in Caputo, CF, and AB

1. B. Wang, S. Zhou, X. Zheng, Image watermarking using 5. O. Atan, Synchronisation and circuit model of fractional-
chaotic map and dna coding, Optik. 126 (2015) 4846–4851. order chaotic systems with time-delay, IFAC-PapersOnLine,
10.1016/j.ijleo.2015.09.217 49(29) (2016) 68–72. https://doi.org/10.1177/
2. S. Çiçek, A. Ferikoğlu, and I. Pehlivan, A new 3d chaotic 1687814019866540
system: dynamical analysis, electronic circuit design, active 6. A. Atangana and S. Qureshi, Modeling attractors of chaotic dy-
control synchronization and chaotic masking communication namical systems with fractal–fractional operators, Chaos, Soli-
application Optik-International Journal for Light and Elec- tons & Fractals. 123 (2019) 320–337. DOI:10.1016/j.
tron Optics. 127(8) (2016) 4024–4030. 10.1016/j.ijleo. chaos.2019.04.020
2016.01.069
3. C. K Volos, I.M. Kyprianidis, and I. N. Stouboulos, A chaotic 7. M. A. Khan, The dynamics of a new chaotic system
path planning generator for autonomous mobile robots, through the caputo–fabrizio and atanagan–baleanu frac-
Robotics and Autonomous Systems. 60(4) (2012) 651–656. tional operators, Advances in Mechanical Engineering,
https://doi.org/10.1016/j.robot.2012.01. 11(7):1687814019866540, 2019. https://doi.org/10.
001 1177/1687814019866540

4. C. Xu, M. Liao, P. Li, Q. Xiao, and S. Yuan, Control strat- 8. A. Atangana, Fractional discretization: The african’s tortoise
egy for a fractional-order chaotic financial model, Complexity. walk, Chaos, Solitons & Fractals, 130:109399, 2020. https:
2019, 2019. 10.1016/j.chaos.2019.04.020 //doi.org/10.1016/j.chaos.2019.109399

Rev. Mex. Fis. 67 051401

 NUMERICAL SOLUTION OF A FRACTAL-FRACTIONAL ORDER CHAOTIC CIRCUIT SYSTEM 17

9. B. Ghanbari and A. Atangana. A new application of fractional model, Journal of Atmospheric and Solar-Terrestrial Physics.
atangana-baleanu derivatives: Designing abc-fractional masks 197 (2020) 105172. https://doi.org/10.1016/j.
in image processing, Physica A: Statistical Mechanics and its jastp.2019.105172
Applications, 2019.
23. B. Ghanbari, J. F. Gómez-Aguilar, Analysis of two avian in-
10. M. A. Khan, O. Kolebaje, A. Yildirim, S. Ullah, P. Kumam, and fluenza epidemic models involving fractal-fractional deriva-
P. Thounthong, Fractional investigations of zoonotic visceral tives with power and Mittag-Leffler memories. Chaos: An In-
leishmaniasis disease with singular and non-singular kernel, terdisciplinary Journal of Nonlinear Science. 29 (2019) 123113.
The European Physical Journal Plus. 134 (2019) 481. https: https://doi.org/10.1063/1.5117285
//doi.org/10.1140/epjp/i2019-12861-1
24. J. F. Gómez-Aguilar, T. Cordova-Fraga, T. Abdeljawad, A.
11. R. Jan, M. A. Khan, P. Kumam, and P. Thounthong, Mod- Khan, H. Khan. Analysis of fractal-fractional malaria trans-
eling the transmission of dengue infection through fractional mission model, Fractals. (2020) 7:2040041. https://doi.
derivatives, Chaos, Solitons & Fractals, 127 (2016) 189–216. org/10.1142/S0218348X20400411
10.1016/j.chaos.2019.07.002
25. J. F. Gomez-Aguilar, New chaotic attractors: Application of
12. M. A. Khan and F. Gómez-Aguilar, Tuberculosis model with fractal-fractional differentiation and integration, Mathemati-
relapse via fractional conformable derivative with power law, cal Methods in the Applied Sciences, 44 ( 2021) 3036-65.
Mathematical Methods in the Applied Sciences. 2019. https://doi.org/10.1002/mma.6432
13. S. Ullah, M. A. Khan, M. Farooq, Z. Hammouch, and D.
26. K. M. Saad, M. Alqhtani, J. F. Gómez-Aguilar, Fractal-
Baleanu, A fractional model for the dynamics of tuberculosis
fractional study of the hepatitis C virus infection model. Results
infection using caputo-fabrizio derivative, Discrete & Continu-
in Physics. 2020 Dec 1;19:103555. https://doi.org/
ous Dynamical Systems-S, 29 (2019) 11–27. doi:10.3934/
10.1016/j.rinp.2020.103555
dcdss.2020057
27. C. J. Zúñiga-Aguilar, J. F. Gómez-Aguilar, H. M. Romero-
14. S. Ullah, M. A. Khan, and M. Farooq, A fractional model for the
Ugalde, H. Jahanshahi, F. E. Alsaadi, Fractal-fractional neuro-
dynamics of tb virus, Chaos, Solitons & Fractals. 116 (2018)
adaptive method for system identification. Engineering with
63–71.
Computers. 22 (2021) 1-24.
15. A. Atangana and S. İğret Araz. Fractional stochastic modelling
illustration with modified chua attractor, The European Phys- 28. S. Kumar, A new analytical modelling for fractional tele-
ical Journal Plus. 134(4) (2019) 160. https://doi.org/ graph equation via Laplace transform, Applied Mathematical
10.1140/epjp/i2019-12565-6 Modelling. 38 ( 2014) 3154–63. https://doi.org/10.
1016/j.apm.2013.11.035
16. S. Ullah, M. A. Khan, and M. Farooq, Modeling and
analysis of the fractional hbv model with atangana- 29. B. Ghanbari, S. Kumar, R. Kumar, A study of behaviour for
baleanu derivative, The European Physical Journal Plus. immune and tumor cells in immunogenetic tumour model with
133(8) (2018) 313. https://doi.org/10.1140/epjp/ non-singular fractional derivative, Chaos, Solitons & Frac-
i2018-12120-1 tals. 133 (2020) 109619. DOI:10.1016/j.chaos.2020.
109619
17. A. Atangana and M. A. Khan, Validity of fractal derivative to
capturing chaotic attractors, Chaos, Solitons & Fractals. 126 30. E. F. Goufo, S. Kumar, S. B. Mugisha, Similarities in a
(2019 )50–59. 10.1016/j.chaos.2019.06.002 fifth-order evolution equation with and with no singular kerne
Chaos, Solitons & Fractals. 130 (2020) 109467. DOI:10.
18. W. Sayed, AG. Radwan, Two-dimensional rotation of
1016/j.chaos.2019.109467
chaotic attractors: Demonstrative examples and fpga real-
ization, Circuits, Systems, and Signal Processing. 38(10) 31. S. Kumar, R. Kumar, R.P. Agarwal, B. Samet, A study of
(2019) 4890–4903. https://doi.org/10.1007/ fractional Lotka-Volterra population model using Haar wavelet
s00034-019-01096-z and Adams-Bashforth Moulton methods, Mathematical Meth-
19. M. S. Semary, A. Radwan, Realization of fractional-order ca- ods in the Applied Sciences. 43 (2020) 5564–5578. https:
pacitor based on passive symmetric network, Journal of ad- //doi.org/10.1002/mma.6297
vanced research. 18 (2019) 147–159. https://doi.org/ 32. S. Kumar, S. Ghosh, B. Samet, E. F. Goufo, An analysis
10.1016/j.jare.2019.02.004 for heat equations arises in diffusion process using new Yang-
20. A. J. Abd El-Maksoud, A. A Abd El-Kader, Fpga implemen- Abdel-Aty-Cattani fractional operator, Mathematical Methods
tation of sound encryption system based on fractional-order in Applied Sciences, 43 (2020) 6062–6080. https://doi.
chaotic systems, Microelectronics Journal. 90 (2019) 323–335. org/10.1002/mma.6347
https://doi.org/10.1016/j.mejo.2019.05. 33. P. Veeresha, D. G. Prakasha, S. Kumar, A fractional model
005 for propagation of classical optical solitons by using nonsin-
21. J. F. Gomez-Aguilar, Chaos and multiple attractors gular derivative, Mathematical Methods in Applied Sciences,
in a fractal-fractional Shinriki’s oscillator model, Phys- https://doi.org/10.1002/mma.6335 (2020)
ica A: Statistical Mechanics and its Applications. 539 ( 34. S. Kumar, S. Ghosh, R. Kumar, M. Jleli, A fractional model for
2020) 122918. https://doi.org/10.1016/j.physa. population dynamics of two interacting species by using spec-
2019.122918 tral and Hermite wavelets methods, Numerical Methods for
22. J. F. Gomez-Aguilar, Multiple attractors and periodicity on Partial Differential Equations, 37 (2021) 1652–1672. https:
the Vallis model for El Niño/La Niña-Southern oscillation //doi.org/10.1002/num.22602

Rev. Mex. Fis. 67 051401

18 MUHAMMAD ALTAF KHAN, A. ATANGANA, TASEER MUHAMMAD, AND EBRAHEEM ALZAHRANI

35. S. Kumar, R. Kumar, M. S. Osman, B. Samet, A wavelet Applied Mathematics. 369 (2020) 112–646. https://doi.
based numerical scheme for fractional order SEIR epidemic org/10.1016/j.cam.2019.112646
of measles by using Genocchi polynomials, Numerical Meth-
ods for Partial Differential Equations, 37 (2021) 1250–1268. 38. A. Atangana, M. A. Khan, et al. Modeling and analysis of
https://doi.org/10.1002/num.22577 competition model of bank data with fractal-fractional caputo-
fabrizio operator, Alexandria Engineering Journal, 2020.
36. Z. Li, Z. Liu, and M. A. Khan, Fractional investigation of bank
https://doi.org/10.1016/j.aej.2019.12.032
data with fractal-fractional caputo derivative, Chaos, Solitons
& Fractals. page 109528, 2019. 39. T. Mekkaoui, Z. Hammouch, F. BM Belgacem, and A. El. Ab-
37. W. Wang and M. A. Khan, Analysis and numerical simula- bassi, Fractional-order nonlinear systems: Chaotic dynamics,
tion of fractional model of bank data with fractal–fractional numerical simulation and circuits design, Fractional Dynamics.
atangana–baleanu derivative, Journal of Computational and 343–356. Sciendo Migration, 2015.

Rev. Mex. Fis. 67 051401