Stability Analysis and Synchronization of Fractional

Order Complex Lorenz System

Ajit K. Singh
Department of Mathematics
Amity University Maharashtra
Mumbai - 410206, India
[email protected]

Abstract—Chaotic nature of the fractional-order system and II. F RACTIONAL D ERIVATIVE AND I TS D EFINITIONS
synchronization have been two very significant problems in
the nonlinear sciences, even in the complex fractional-order The integral-differential operator a Dtα is defined by
system. In the present manuscript, a study of the fractional-order
 α
complex Lorenz system which exhibits a chaotic behavior would d
be analyzed by numerical simulation results. Stability analysis of

 when R (α) > 0,
 dtα


fractional-order system has been studied and obtains some useful α
results for the researcher working in the nonlinear dynamics. a Dt = 1 when R (α) = 0, (1)
 Z t
Synchronization of the proposed systems is also studied and it  −α
(dτ ) when R (α) < 0,


has been achieved by active control method. 
a
Index Terms—synchronization, chaotic system, Lorenz system,
fractional derivative, active control method where R (α) is the real part of α, order of the fractional-order
complex chaotic system. Values a and t are the limits of the
operator. A large number of fractional derivatives are observed
I. I NTRODUCTION
in the literature survey, author discusses three commonly used
Since the progress of the clasical calculus, thoughts of definitions by first considering the Grunwald-Letnikov (GL)
fractional calculus have been known. In 1695, the meaning of definition of the fractional-order α is stated as
half order derivative was examined by Leibniz and L’Hospital [ t−a
h ]  
[1]. Applications of fractional-order system to engineering GL α 1 X j α
a Dt f (t) = lim α (−1) f (t − jh) (2)
and physical scincss are just recent pay particular attention h→0 h j
j=0
to of interest even though it is more than 325 years old. It
was observed that many systems in multidisciplinary fields where [·] symbol represents the integral part.
can be fashionable described with the aid of fractional-order The Riemann-Liouville (RL) definition is stated as
derivatives. The majority of fractional-order complex systems
Z t
are known to show dynamical behavior, such as electrode- GL α 1 dn f (τ )
electrolyte polarization [2], viscoelastic systems [3] , dielec- a D t f (t) = dτ (3)
Γ (n − α) dt a (t − τ )α−n+1
n
tric polarization [4], quantitative finance [5], electromagnetic
waves [6], and quantum evolution of complex systems [7] [8]. where n − 1 < α < n and Γ (·) is the gamma function.
Chaotic areas are originate in a large scope of parametric The Caputo (C) definition is defined in the following way.
space. Different action along the trajectory to chaos are
t
f (n) (τ )
Z
observed and depicted through graph and phase portrait. C α 1
For stability analysis of the fractional-order complex chaotic a Dt f (t) = α−n+1 dτ (4)
Γ (n − α) a (t − τ )
system, an effective way is studied to found the lowest order
of the fractional-order. where n − 1 < α < n and Γ (·) is the gamma function.
In the year 1990, synchronization of the chaotic systems The Caputo definition for the fractional derivatives is used
was seen by Pecora and Carroll [9], experimental along with thoughout the article and a convenient notation Dα , instead
the theoretical research on synchronization has been executed of C α
a Dt , is used.
in the plenty of nonlinear chaotic systems, as it could be
applied in immense areas of engineering, physical sciences III. FRACTIONAL O RDER C OMPLEX L ORENZ S YSTEM
and particularly in secure communication [10]. Different types
of phenomenon have been put forward for the synchroniza- Fowler et al. was studied the complex Lorenz system [22]
tion, such as complete synchronization [11], [12], projective firt time in the year 1982, and in 2007, its dynamical bahavior
synchronization [13]–[15], anti-synchronization [16]–[18], and was introduced by Mahmoud et al. [23]. This system is useful
generalized synchronization [19]–[21], of chaotic systems. to simulate the physics of liquid flows. The integer-order

Electronic copy available at: https://ssrn.com/abstract=3703649

complex Lorenz system is described as As per the RouthHurwitz conditions [24] of the fractional-
dx01 order system, when
= a1 (x02 − x01 )
dt (1 + a1 + a3 ) (a1 a3 + a2 a3 ) > 2a1 a3 (a2 − 1) , (9)
dx02
= a2 x01 − x02 − x01 x03 (5)
Eθ will be stable.
dt
dx03 1 0 0 
= x1 x2 + x01 x02 − a3 x03 E. Chaotic Attractors
dt 2
T When we take the parameters value a1 = 10, a2 = 180,
where x0 = [x01 , x02 , x03 ] is state vector variable. Variables T
a3 = 1 and initial condition x0 (0) = [4 + 3i, 4 + 6i, 7] , the
x01 = x1 + ix2 and x02 = x3 + ix4 2 are taken as complex and system (7) exhibits the chaotic attractors which are shown by
x03 = x5 is real variable. The fractional-order complex Lorenz Fig. 1 and Fig. 2 at fractional-order α = 0.95.
system (FOCLS) is described by
Dα x01 = a1 (x02 − x01 ) 20

Dα x02 a2 x01 x02 x01 x03

x1(t)
= − − (6)
0

1  −20
Dα x03 = x0 x0 + x01 x02 − a3 x03 . 50
0 50 100 150
2 1 2

x2(t)
0
IV. S TABILITY A NALYSIS
−50
0 50 100 150
Dynamical analysis of the FOCLS will be investigated, 50
including real version, symmetry and invariance, equilibrium

x3(t)
points, stability and chaotic attractors of system (6) in this 0
0 50 100 150
section. t

A. Real Version Fig. 1. State trajectories of the chaotic system.

The Caputo fractional derivative (4) is linear, the system (6)
can be split in real version as
V. S YNCHRONIZATION
Dα x1 = a1 (x3 − x1 )
In this section, chaos synchronization of the FOCLS dis-
Dα x2 = a1 (x4 − x2 ) cussed. Let us consider the system (7) as a drive sys-
Dα x3 = a2 x1 − x3 − x1 x5 (7) tem. Then response system with control function u (t) =
T
α
D x4 = a2 x2 − x4 − x2 x5 [u1 (t) , u2 (t) , u3 (t) , u4 (t) , u5 (t)] is given by
Dα x5 = x1 x3 + x2 x4 − a3 x5 . Dα y1 = b1 (y3 − y1 ) + u1
B. Symmetry and Invariance Dα y2 = b1 (y4 − y2 ) + u2
Symmetry about x5 -axis, due to the invariance of Dα y3 = b2 y1 − y3 − y1 y5 + u3 (10)
equations under the change (x1 , x2 , x3 , x4 , x5 ) → Dα y4 = b2 y2 − y4 − y2 y5 + u4
(−x1 , −x2 , −x3 , −x4 , x5 ) . Hence, if (x1 , x2 , x3 , x4 , x5 ) is a Dα y5 = y1 y3 + y2 y4 − b3 y5 + u5 .
solution of chaotic system (7), then (−x1 , −x2 , −x3 , −x4 , x5 )
is also a solution of the same system (7). In order to estimate the control functions u1 (t) , u2 (t) ,
u3 (t) , u4 (t) and u5 (t) , let us consider the error functions
C. Equilibrium Points
between the system (10) and the system (7) as
The equilibrium points of system (7) is calculated by
dealing the equations Dα xj = 0, j = 1, 2, 3, 4, 5 and e1 = y1 − x1 , e2 = y2 − x2 , e3 = y3 − x3 ,
system (7) have an isolated equilibrium point E0 = e4 = y4 − x4 , e5 = y5 − x5 (11)
(0, 0, 0, 0, 0) and non-trivial equilibrium points Eθ =
√
(r cos θ, r sin θ, r cos θ, r sin θ, x5 ) where r = a3 x5 , x5 = One would find the error system as follows
a2 − 1, θ ∈ [0, 2π] . It is understood that the non-trivial
Dα e1 =b1 (e3 − e1 ) + (b1 − a1 ) (x3 − x1 ) + u1
equilibrium point exist when a2 > 1.
Dα e2 =b1 (e4 − e2 ) + (b1 − a1 ) (x4 − x2 ) + u2
D. Stability Dα e3 =b2 e1 − e3 + (b2 − a2 ) x1 − e1 e5 − e1 x5
Since equilibrium point E0 is stable when a2 < 1, and − e5 x1 + u3
unstable when a2 > 1. For Eθ , the characteristic polynomial (12)
of Jacobian matrix for a2 > 1 is Dα e4 =b2 e2 − e4 + (b2 − a2 ) x2 − e2 e5 − e2 x5
− e5 x2 + u4
z (z + a1 + 1) z3 + (1 + a1 + a3 ) z 2 α
D e5 = − b3 e5 + (a3 − b3 ) x5 + e1 e3 + e2 e4
+ (a1 a3 + a2 a3 ) z + 2a1 a3 (a2 − 1) = 0. (8) + e1 x3 + e3 x1 + e2 x4 + e4 x2 + u5 .

Electronic copy available at: https://ssrn.com/abstract=3703649

 30 where the terms Vj (t) are the error terms ej (t) which is linear
20 functions. Now, the system (12) is changed to
10 Dα e1 =b1 (e3 − e1 ) + V1
Dα e2 =b1 (e4 − e2 ) + V2
x2(t)

−10 Dα e3 =b2 e1 − e3 + V3 (14)

−20 Dα e4 =b2 e2 − e4 + V4
−30
−20 −15 −10 −5 0 5 10 15 20
Dα e5 = − b3 e5 + V5 .
x1(t)
An appropriate feedback control to design for which sta-
bilizes the used system, so that ej (t) , j = 1, 2, 3, 4, 5
60 approach to zero as time t tends to infinity. Due
to many possible options available to the control in-
40
puts Vj (t) , consider V (t) = A e (t) . where V (t) =
x3(t)

T
20 [V1 (t) , V2 (t) , V3 (t) , V4 (t) , V5 (t)] , A is 5 × 5 constant
T
0 matrix and e (t) = [e1 (t) , e2 (t) , e3 (t) , e4 (t) , e5 (t)] . To
40
20
10
20 stabilize the closed loop system, one must chose the matrix in
0 0
−20 −10 such a way that the eigenvalues λj of A should be satisfying
−40 −20
x2(t) x1(t) the condition |arg (λj )| > απ/2, j = 1, 2, 3, 4, 5. Matrix A is
taken in the following form as there is no unique option.
 
−1 + b1 0 −b1 0 0
50
 0 −1 + b1 0 −b1 0 
x (t)

 
A= −b 0 0 0 0 
5

 2 
0 20
 0 −b2 0 0 0 
40
20
0 0 0 0 0 0 −1 + b3
−20
−40 −20
x4(t) x3(t) In this particular choice, the error system (14) becomes
Dα ej = −ej , j = 1, 2, 3, 4, 5 (15)
The eigenvalues −1, −1, −1, −1 and −1 were found for
50
closed loop system (15). For this choice, error states ej (t) ,
j = 1, 2, 3, 4, 5 converge to zero as time becomes large and
x4(t)

therefore, the synchronization of the FOCLS achieved.

0
40 VI. S IMULATION R ESULTS
20
In the current section, feasibility of verification and
0
demonstration of the FOCLS and synchronization, au-
−20
5 10 15 20 thor obtains the simulation results of the considered sys-
x3(t) −40 −15 −10 −5 0
−20
x2(t)
tems in complex space at the fractional-order α =
0.95. During synchronization, the values of the parame-
Fig. 2. Phase portraits of the chaotic system. ters remain same. The initial conditions are consider as
T T
[−3 + 2i, −3 − 2i, 10] and [10 + 10i, 8 + 8i, 10] hence the
T
initial error is [13 + 8i, 11 + 10i, 20] . Fig. 3 display the time
response of the states xj (t) and yj (t) of the drive system
(7) and the response system (10). Figure 4 shows that the
Defining the active control functions uj (t) in the following
error vectors asymptotically approach to zero as time goes to
way
infinity which means that chaos synchronization of the FOCLS
achieved.

u1 =V1 − (b1 − a1 ) (x3 − x1 ) VII. C ONCLUDING R EMARKS

u2 =V2 − (b1 − a1 ) (x4 − x2 ) Active control method is successfully used to achieve per-
u3 =V3 − (b2 − a2 ) x1 + e1 e5 + e1 x5 + e5 x1 fect control of a pair of FOCLS along the required trajectory,
(13) which sincerely shows the potential and reliability of the
u4 =V4 − (b2 − a2 ) x2 + e2 e5 + e2 x5 + e5 x2
method for considered systems to be synchronized. Dynamical
u5 =V5 − (a3 − b3 ) x5 − e1 e3 − e2 e4 − e1 x3 behavior of the FOCLS is investigated which is the most
− e3 x1 − e2 x4 − e4 x2 . important part of the article.

Electronic copy available at: https://ssrn.com/abstract=3703649

 Fig. 4. Error plot of drive and response systems.

R EFERENCES
[1] I. Podlubny, Fractional differential equations: an introduction to frac-
tional derivatives, fractional differential equations, to methods of their
solution and some of their applications. Elsevier, 1998.
[2] M. Ichise, Y. Nagayanagi, and T. Kojima, “An analog simulation of
non-integer order transfer functions for analysis of electrode processes,”
Journal of Electroanalytical Chemistry and Interfacial Electrochemistry,
vol. 33, no. 2, pp. 253–265, 1971.
[3] R. L. Bagley and R. Calico, “Fractional order state equations for
the control of viscoelasticallydamped structures,” Journal of Guidance,
Control, and Dynamics, vol. 14, no. 2, pp. 304–311, 1991.
[4] H. Sun, A. Abdelwahab, and B. Onaral, “Linear approximation of
transfer function with a pole of fractional power,” IEEE Transactions
on Automatic Control, vol. 29, no. 5, pp. 441–444, 1984.
[5] N. Laskin, “Fractional market dynamics,” Physica A: Statistical Me-
chanics and its Applications, vol. 287, no. 3-4, pp. 482–492, 2000.
[6] O. Heaviside, “Electromagnetic theory, chelsea pub,” Co., New York,
1971.
[7] D. Kusnezov, A. Bulgac, and G. Do Dang, “Quantum levy processes
and fractional kinetics,” Physical review letters, vol. 82, no. 6, p. 1136,
1999.
[8] A. K. Singh, V. K. Yadav, and S. Das, “Synchronization of time-
delay chaotic systems with uncertainties and external disturbances,”
Discontinuity Nonlinearity and Complexity, vol. 8, no. 1, pp. 13–21,
2019.
[9] L. M. Pecora and T. L. Carroll, “Synchronization in chaotic systems,”
Physical review letters, vol. 64, no. 8, p. 821, 1990.
[10] A. K. Singh, V. K. Yadav, and S. Das, “Dual combination synchro-
nization of the fractional order complex chaotic systems,” Journal of
Computational and Nonlinear Dynamics, vol. 12, no. 1, p. 0110171,
2017.
[11] L. Kocarev and U. Parlitz, “General approach for chaotic synchro-
nization with applications to communication,” Physical review letters,
vol. 74, no. 25, p. 5028, 1995.
[12] A. K. Singh, V. K. Yadav, and S. Das, “Comparative study of syn-
chronization methods of fractional order chaotic systems,” Nonlinear
Engineering, vol. 5, no. 3, pp. 185–192, 2016.
[13] E.-H. Park, M. A. Zaks, and J. Kurths, “Phase synchronization in the
forced lorenz system,” Physical Review E, vol. 60, no. 6, p. 6627, 1999.
[14] A. K. Singh, V. K. Yadav, and S. Das, “Synchronization between
fractional order complex chaotic systems,” International Journal of
Dynamics and Control, vol. 5, no. 3, pp. 756–770, 2017.
[15] Y.-J. Liu and Y.-Q. Zheng, “Adaptive robust fuzzy control for a class
of uncertain chaotic systems,” Nonlinear dynamics, vol. 57, no. 3, pp.
431–439, 2009.
[16] M. M. Al-Sawalha and M. S. M. Noorani, “Chaos anti-synchronization
between two novel different hyperchaotic systems,” Chinese Physics
Letters, vol. 25, no. 8, p. 2743, 2008.
[17] A. K. Singh, V. K. Yadav, and S. Das, “Synchronization between
Fig. 3. State trajectories of the drive and response systems. fractional order complex chaotic systems with uncertainty,” Optik, vol.
133, pp. 98–107, 2017.
[18] Y.-J. Liu, W. Wang, S.-C. Tong, and Y.-S. Liu, “Robust adaptive tracking
control for nonlinear systems based on bounds of fuzzy approximation
parameters,” IEEE Transactions on Systems, Man, and Cybernetics-Part
A: Systems and Humans, vol. 40, no. 1, pp. 170–184, 2009.

Electronic copy available at: https://ssrn.com/abstract=3703649

[19] A. C. Luo, “Corrigendum to a theory for synchronization of dynamical
systems[commun nonlinear sci numer simulat 14 (2009) 1901-1951],”
CNSNS, vol. 14, no. 7, pp. 3214–3214, 2009.
[20] A. K. Singh, V. K. Yadav, and S. Das, “Nonlinear control technique for
dual combination synchronization of complex chaotic systems,” Journal
of Applied Nonlinear Dynamics, vol. 8, no. 2, pp. 261–277, 2019.
[21] H. Zhang, T. Ma, G.-B. Huang, and Z. Wang, “Robust global exponential
synchronization of uncertain chaotic delayed neural networks via dual-
stage impulsive control,” IEEE Transactions on Systems, Man, and
Cybernetics, Part B (Cybernetics), vol. 40, no. 3, pp. 831–844, 2009.
[22] A. Fowler, J. Gibbon, and M. McGuinness, “The complex lorenz
equations,” Physica D: Nonlinear Phenomena, vol. 4, no. 2, pp. 139–
163, 1982.
[23] G. M. Mahmoud, M. Al-Kashif, and S. A. Aly, “Basic properties
and chaotic synchronization of complex lorenz system,” International
Journal of Modern Physics C, vol. 18, no. 02, pp. 253–265, 2007.
[24] X.-y. Wang, Y.-j. He, and M.-j. Wang, “Chaos control of a fractional
order modified coupled dynamos system,” Nonlinear Analysis: Theory,
Methods & Applications, vol. 71, no. 12, pp. 6126–6134, 2009.

Electronic copy available at: https://ssrn.com/abstract=3703649