This paper mainly investigates the projection synchronization of complex chaotic systems with both uncertainty and disturbance. Using the linear feedback method and the uncertainty and disturbance estimation- (UDE-) based control method, the projection synchronization of such systems is realized by two steps. In the first step, a linear feedback controller is designed to control the nominal complex chaotic systems to achieve projection synchronization. An UDE-based controller is proposed to estimate the whole of uncertainty and disturbance in the second step. Finally, numerical simulations verify the feasibility and effectiveness of the control method.
Natural Science Foundation of Shandong ProvinceZR2018MF0161. Introduction
The chaotic synchronization phenomenon that caused a great sensation in academia was firstly proposed by Pecora and Carroll in early 1990 [1]. They achieved chaotic synchronization of two identical systems with different initial conditions in electronic experiments. Until now, many types of chaotic synchronization have been discovered, such as complete synchronization, phase synchronization, lag synchronization, antisynchronization, and projection synchronization, and many other important results have been obtained (see references [2–8]). Especially, projective synchronization has received much attention due to its faster communication and proportionality between the dynamical systems. In case of projective synchronization, the master and the slave system can be synchronized up to a scaling factor and the scaling factor is a constant transformation between the driving and the response variables that can further increase the security of secure communication and the transmission speed of communication. It has potential application prospects in the field of chaotic secure communication.
Many control methods about chaotic projection synchronization have been reported [9–29]. However, most controllers are complicated in structure and difficult in design. Due to the complexity of structure, many control methods are not suitable for projective synchronization control of complex chaotic systems. Among these, the linear feedback controller, because of its simple structure, easy design, and good control effect, was used to realize the projection synchronization of given complex chaotic system. Moreover, in the simulation experiment, it is also proved that the linear feedback controller has a good experimental effect.
We note that most of the literature on solving the control problems of chaotic systems with external perturbations is generally complex and difficult to implement. Moreover, when designing the controller, the method to deal with the external disturbance is just simply to cancel the disturbance term from the formula of the controller, and it is not rigorous in nonlinear system control theory. In fact, in the field of nonlinear system control, the UDE-based controller can deal with many structured and unstructured robust control problems and has been applied to the engineering field in some literatures [30–32]. In the simulation experiment, we have noticed that the UDE control method, which is composed of filters with appropriate bandwidth, has an ideal processing effect on the external disturbance of the system which is finally used by us.
The main contribution of this paper is to design a physical controller, which is simple in form, to realize the projection synchronization of a complex chaotic system. A linear feedback UDE-based control method is proposed by combining the linear feedback controller and the UED-based controller in two steps. A linear feedback control controller is designed for the nominal complex chaotic system in the first step. In the second step, an UDE-based controller is proposed to estimate the whole of uncertainty and disturbance. In the end, two complex chaotic systems with numerical simulations are used to verify the validity and effectiveness of the proposed theoretical results.
2. Preliminary
Consider the following controlled chaotic system:(1)X˙=FX+B∗U∗,where X∈Rn is the state, FX=F1X,⋯,FnXT is a continuous vector function, B∗∈Rn×l, and U∗=U1∗,⋯,Ul∗T is the controller to be designed, l≥1.
Let system (1) be the master system; then, the slave system is given as follows:(2)Y˙=FY,where Y∈Rn is the state and FY=F1Y,⋯,FnYT is a continuous vector function.
Let e=X−αY, where α=Diagα1,⋯,αn, and the error system is shown as follows:(3)e˙=FX−αFX+B∗U∗,where e∈Rn is the state vector.
Definition 1.
Consider the controlled error system (3). If limt⟶∞et=0, then the master system (1) and the slave system (2) are called to achieve projection synchronization.
According to the results in [17], a lemma is introduced as follows.
Remark 1.
The projection synchronization of system (1) is achieved if and only it is divided into the following two subsystems:(4)W˙m=AZWm,(5)Z˙=HZ,Wm,where Wm∈Rs, Z∈Rn−s, s≥1, AZ∈Rs×s is a matrix with constants and variable Z, and HZ,Wm is nonlinear continuous function.
An algorithm was also proposed in [17], by which we can solve the solutions of the projection synchronization and choose the variables Wm and Z.
2.1. Linear Feedback Control-Like Method for Chaos Projection Synchronization
Note that the subsystem W˙m=AZWm is a linear system with respect to variable Wm if the variable Z is considered a constant. Thus, the linear feedback control method is very suitable to be adopted to solve the projective synchronization problem of a given nominal complex chaotic system (i.e., there is no both uncertainty and disturbance). We briefly introduce the linear control method next.
Lemma 1.
Consider the following controlled system:(6)W˙m=AZWm+B1U,where Wm,Z,AZ are given in equations (4) and (5) and B1∈Rs×r; then, the linear feedback controller is designed as follows:(7)U=KZWm,where KZ satisfies the matrix AZ+B1KZ which is Hurwitz no matter what Z is.
2.2. UDE-Based Control Method
It is well known that model uncertainty and external disturbance are inevitable in actual control problem, and the UDE-based control method [32] is an effective tool to deal with that problem.
Consider the following system:(8)x˙=fx+ud+bu,where x∈Rn is the state, ud=Δfx+dt is the whole of model uncertainty and external disturbance, b∈Rn×k is a constant matrix, k≥1, and u∈Rk is the controller to be designed.
The stable linear reference model is given as(9)x˙m=Amxm+BmC,where xm∈Rn is the reference state, Am∈Rn×n is the Hurwitz matrix, Bm∈Rn×k, and C∈Rk×1 is a piecewise continuous and uniformly bounded command to the system.
Lemma 2 (see [32]).
Consider system (8). If the designed filter gft satisfies the following condition:(10)u˜d=u^d−ud⟶0,where u^d=x˙−fx−bu∗gft, then the UDE-based controller u is designed as(11)u=b+−fx+ℓ−111−Gfs∗Amx+BmC−Ke−b+ℓ−1sGfs1−Gfs∗xt,where ℓ−1 denotes the inverse Laplace transform operator, b+=bTb−1bT, ∗ is the convolution operator, and Gfs=ℓgft.
Remark 2.
According to the existing result in [32], the following two filters are often used. One is the first-order low-pass filter:(12)Gfs=1τs+1.
The other is the secondary filter:(13)Gfs=as+c−w02s2+as+c,where w0=4π, a=10w0, and c=100w0.
3. Main Results
In this section, the UDE-based linear feedback control method is proposed in two steps. In the first step, the linear feedback control method is proposed for the nominal system. The UDE-based control method is given in the second step.
3.1. Linear Feedback Control Method for Projection Synchronization
Consider the following nominal system:(14)X˙=FX+BU,where X∈Rn is the state vector, FX=F1X,⋯,FnXT is a continuous function, B∈Rn×r,r≥1 , U=U1,⋯,UrT is the linear feedback controller to be designed, and FX,B is assumed to be controllable.
If the projection synchronization of system (14) exists, then it can be divided into the following two subsystems:(15)W˙m=AZWm+B1U,(16)Z˙=HZ,Wm,where Wm,Z,AZ, HZ,Wm are given in equations (4) and (5), respectively, B1∈Rs×r is given in equation (6), and AZ,B1 is also controllable.
The corresponding slave system is presented as follows:(17)W˙s=AZWs,where HZ,Wm is given in equation (15), Wm∈Rr, Z∈Rn−r, and AZ is a constant matrix.
Let e=Wm−βWS be the error state, where the scalar β≠0,1, and the error system is obtained as follows:(18)e˙=AZe+B1U.
Theorem 1.
Consider error system (18). If AZ,B1 is controllable no matter what Z is, then the linear feedback controller U is designed as follows:(19)U=KZe,where KZ satisfies the matrix AZ+B1KZ which is Hurwitz no matter what Z is; then, error system (18) is globally asymptotically stable. That is, the master system (15) and the slave system (17) achieve the projection synchronization.
Proof.
Since the matrix AZ+B1KZ is Hurwitz no matter what Z is, error system (18) is globally asymptotically stable; therefore, the master system (15) and the slave system (17) achieve the projection synchronization.
3.2. UDE-Based Control Method for Projection Synchronization
In this section, the UDE controller is proposed to cancel the uncertainty and disturbance of the complex chaotic system.
Consider the following controlled master system:(20)W˙m=AZWm+B1V+Ud,where Wm,Z,AZ, HZ,Wm are given in equations (4) and (5), respectively, B1∈Rs×r is given in equation (6), AZ,B1 is controllable, Ud=ΔAZ+Dt, ΔAZ represents the uncertainty and Dt represents the disturbance, and V is the controller to be designed, in which(21)V=U+uude.
The corresponding salve system is(22)W˙s=AZWs.
Let e=Wm−βWs be the error state vector, where β≠0,1; then, the corresponding error system is shown as follows:(23)e˙=AZe+ud+B1V.
The controller V is designed in two steps:
Step one: according to Theorem 1, the linear feedback controller U is designed for the nominal system.
Step two: the controller uude is proposed according to the following theorem.
Theorem 2.
Consider error system (23). If the designed filter gft satisfies the following condition:(24)u˜d=u^d−ud⟶0,where u^d=e˙−AZ,e−B1uude∗gft, then the UDE-based controller u is designed as(25)uude=B1+ℓ−1Gf1−Gfs∗AZ,e−ℓ−1sGfs1−Gfs∗et,where AZ,e=AZ+B1KZe, B1+=B1TB−1B1T, ℓ−1 is the inverse Laplace transform, ∗ is the convolution sign, and Gfs=ℓgft.
Proof.
Substituting V in (21) into system (23) results in(26)e˙=AZe+ud+B1V=AZ+B1KZe+Buude−ud.
According to condition (24), it leads to(27)B1uude=−u^d.
Thus,(28)e˙=Az+BV+ud=AZ,e+u˜dis globally asymptotically stable, which completes the proof.
4. Illustrative Example with Numerical Simulation
In this section, one example with numerical simulations is used to demonstrate the effectiveness and validity of the proposed results.
Consider the following complex Lorenz system:(29)x˙1=10x1−x2,x˙2=110x1−x1x3−x2,x˙3=−2x3+12x¯1x2+x1x¯2,where x1=x1r+jx1i,x2=x2r+jx2i are complex variables, x3 is a real variable, j2=−1 represents imaginary unit, and x¯1 and x¯2 are complex conjugate variables of x1, x2, respectively.
Separating the real and imaginary parts of complex variables x1,x2 in system (29), i.e., setting X1=x1r,X2=x1i,X3=x2r,X4=x2i, and X5=x3 representing x5, a new real-variable system is shown as follows:(30)X˙1=10X3−X1,X˙2=10X4−X2,X˙3=110X1−X1X5−X3,X˙4=110X2−X2X5−X4,X˙5=−2X5+X1X3+X2X4.
4.1. The Existence of Projection Synchronization of the Complex Lorenz System
According to the results in [17], for system (30), the results are obtained as follows:(31)F1αX−α1F1αX=10α3−α1X3≡0,F2αX−α2F2αX=10α4−α2X4≡0,F3αX−α3F3αX=α1−α1α5X1−α3−α1α5X3≡0,F4αX−α4F4αX=α2−α2α5X2−α4−α2α5X4≡0,F5αX−α5F5αX=α5−α1α3X5−α1α3−α2α4X1X3≡0.
It results in(32)α1=α3,(33)α2=α4,(34)α1=α1α5,α3=α1α5,(35)α2=α2α5,α4=α2α5,(36)α5=α1α3,α1α3=α2α4.
It is easy to obtain that α=Diagβ,β,β,β,1 is the one solution of equations (32)–(35), where β≠1 is a nonzero scalar.
Thus, the master system (30) is divided into the following two subsystems:(37)W˙m=AZWm,(38)Z˙=HZ,Wm,where(39)Wm=X1⋮X4,Z=X5,AZ=−1001000−10010110−Z0−100110−Z0−1,(40)HZ,Wm=−2Z+Wm1Wm3+Wm2Wm4.
4.2. The UDE-Based Linear Feedback Controller Design
The UDE-based linear feedback controller is designed by the following two steps.
Step one:
(41)W˙m=AZWm+B1U.
Wm,Z,AZ are given in equations (38) and (39), respectively, and
(42)B1=10010000.
Then, the corresponding slave system is given as follows:
(43)W˙s=AZWs.
AZ,Z are given in equations (38) and (39), respectively.
Let e=Wm−βWs, where β=2, and the uncontrolled error system is given as follows:(44)e˙1=10e3−e1,e˙2=10e4−e2,e˙3=110−X5e1−e3,e˙4=110−X5e2−e4.
Note that if e1=0 and e2=0, the following system:(45)e˙3=−e3,e˙4=−e4,
is globally asymptotically stable.
Thus, AZ,B1 is controllable. According to Theorem 1, the linear feedback controller U is obtained as follows:(46)U=KZe=−10e3−10e4.
Numerical simulation is given, and the initial values of the master-slave systems of given complex Lorenz system are chosen as follows: X10=0.1,X20=0.2,X30=0.3,X40=−0.3,X50=1.1, Y10=−1,Y20=−1,Y30=−1,Y40=−1.
From Figures 1 and 2, we observed that under linear feedback control, the error system between the master system and slave system is globally asymptotically stable. Through the observation of Figures 3–5, it is found that the master system and slave system achieve the projection synchronization. That is, the controlled master system and slave system have the same phase portrait, but the axis is different.
Step two: consider the following master system with both model uncertainty and external disturbance:
(47)W˙m=AZWm+B1V+Ud,
where AZ is given in equation (39), HZ,Wm is presented in equation (40), B1 is given in equation (42), and Ud is the whole of model uncertainty and external disturbance.
e1,e2 are asymptotically stable.
e3,e4 are asymptotically stable.
The phase portrait of master subsystem and the slave subsystem.
The phase portrait of master subsystem and the slave subsystem.
The phase portrait of master subsystem and the slave subsystem.
The salve system is(48)W˙s=AZWs,where AZ is given in equation (39).
Let e=Wm−βWs; then, the error system is shown as follows:(49)e˙=AZe+Ud+B1V,where(50)V=U+uude,where U is given in equation (46).
According to Theorem 2, the UDE-based controller uude is designed as follows:(51)uude=uude1uude2=ℓ−1Gf1−Gfs∗−10e1−sGfs1−Gfs∗e1ℓ−1Gf1−Gfs∗−10e2−sGfs1−Gfs∗e2,where ℓ−1 is the inverse Laplace transform, ∗ is the convolution sign, Gfs=ℓgft, and the design of the filter gft is given in Lemma 2.
Numerical simulation results are given with the following conditions: X10=0.1,X20=0.2,X30=0.3,X40=−0.3,X50=1.1, Y10=−1,Y20=−1,Y30=−1,Y40=−1, β=2.
Case 1:(52)Ud=0.1X1X2+1000.2X3X4+400.
Case 2:(53)Ud=0.1X1X2+0.1sint0.2X3X4+0.3sint.
It can be seen from Figures 6–9 that the error system is asymptotically stable. Through the observation of Figures 10–13, it is found that the master system and slave system achieve the projection synchronization. That is, the controlled master system and slave system have the same phase portrait, but the axis is different. Figure 14 shows that u^d1 tends to ud1, and Figure 15 shows that u^d2 tends to ud2. Similarly, we found that u^d1 tends to ud1 and u^d2 tends to ud2 from Figures 16 and 17.
e1,e2 are asymptotically stable.
e3,e4 are asymptotically stable.
The phase portrait of master subsystem and the slave subsystem.
The phase portrait of master subsystem and the slave subsystem.
The phase portrait of master subsystem and the slave subsystem.
The phase portrait of master subsystem and the slave subsystem.
u^d1 tends to ud1.
u^d2 tends to ud2.
e1,e2 are asymptotically stable.
e3,e4 are asymptotically stable.
u^d1 tends to ud1.
u^d2 tends to ud2.
5. Conclusion
In conclusion, the projective synchronization of a class of complex chaotic systems with both uncertainty and disturbance has been solved. First, the linear feedback control method is proposed for the nominal system (without uncertainty and disturbance), and projection synchronization of such system has been realized. Then, the UDE-based linear feedback control method is presented by two steps, by which the projection synchronization of the complex chaotic systems with both uncertainty and disturbance has been completed. Finally, an experimental simulation example has been used to verify the feasibility and effectiveness of the obtained results.
Data Availability
No data were used to support this study.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This study was supported by the National Natural Science Foundation of Shandong Province (ZR2018MF016).
PecoraL. M.CarrollT. L.Synchronization in chaotic systems1990648–1982182410.1103/physrevlett.64.8212-s2.0-0343689904LiP. Y.DuJ.LiS. L.ZhengY.JiaB.The synchronization of N cascade-coupled chaotic systems2019201910270982010.1155/2019/2709820ZouL. M.PengY.FengY. M.TuZ.Stabilization and synchronization of memristive chaotic circuits by impulsive control2017201710518671410.1155/2017/51867142-s2.0-85042232288RenL.GuoR.A necessary and sufficient condition of anti-synchronization for chaotic systems and its applications20152015743465110.1155/2015/4346512-s2.0-84947571902WangC.ChuR.MaJ.Controlling a chaotic resonator by means of dynamic track control201521137037810.1002/cplx.215722-s2.0-84942292629LiB.ZhouX. B.WangY.Combination synchronization of three different fractional-order delayed chaotic systems201920199518403210.1155/2019/5184032RenL.GuoR.VincentU. E.Coexistence of synchronization and anti-synchronization in chaotic systems2016261697910.1515/acsc-2016-00042-s2.0-84963813274GuoR.-W.Simultaneous synchronizaiton and anti-synchronization of two identical new 4D chaotic systems201128404020504020910.1088/0256-307x/28/4/0402052-s2.0-79955749891Prez-CruzJ. H.Tamayo-MezaP. A.FigueroaM.Exponential synchronization of chaotic Xian system using linear feedback control2019201910470649110.1155/2019/47064912-s2.0-85070477844YangX. B.HeY. G.LiC. L.Dynamics feature and synchronization of a robust fractional-order chaotic system2018201812879731410.1155/2018/87973142-s2.0-85058977442MoysisL.PetavratzisE.MarwanM.VolosC.NistazakisH.AhmadS.Analysis, synchronization, and robotic application of a modified hyperjerk chaotic system2020202015282685010.1155/2020/2826850XuR.ZhangF.ε-Nash mean-field games for general linear-quadratic systems with applications20201141610.1016/j.automatica.2020.108835YiX.GuoR.QiY.Stabilization of chaotic systems with both uncertainty and disturbance by the UDE-based control method202081624716247710.1109/access.2020.2983674BanerjeeS.TheesarS. J.KurthsJ.Generalized variable projective synchronization of time delayed systems2013230131180110.1063/1.47915892-s2.0-84875830843ChenM.ShaoS.-Y.ShiP.ShiY.Disturbance-observer-based robust synchronization control for a class of fractional-order chaotic systems201764441742110.1109/tcsii.2016.25637582-s2.0-85017578208Al-mahbashiG.NooraniM. S. M.BakarS. A.Projective lag synchronization in drive-response dynamical networks with delay coupling via hybrid feedback control20158231569157910.1007/s11071-015-2261-42-s2.0-84944277196GuoR.Projective synchronization of a class of chaotic systems by dynamic feedback control method2017901536410.1007/s11071-017-3645-42-s2.0-85027146804MahmoudE. E.Dynamics and synchronization of new hyperchaotic complex Lorenz system2012557-81951196210.1016/j.mcm.2011.11.0532-s2.0-84857120377MahmoudG. M.AhmedM. E.Modified projective synchronization and control of complex Chen and Lü systems20111781184119410.1177/10775463103788712-s2.0-79960545525MahmoudG. M.AhmedM. E.MahmoudE. E.Analysis of hyperchaotic complex Lorenz systems200819101477149410.1142/s01291831080131512-s2.0-56749091385MahmoudG. M.Al-KashifM. A.AlyS. A.Basic properties and chaotic synchronization of complex Lorenz system200718225326510.1142/s01291831070104252-s2.0-34249024925MahmoudE. E.AboodF. S.A new nonlinear chaotic complex model and its complex antilag synchronization2017201713384895310.1155/2017/38489532-s2.0-85030180914SayedW. S.HeneinM. M. R.Abd-El-HafizS. K.RadwanA. G.Generalized dynamic switched synchronization between combinations of fractional-order chaotic systems2017201717918912010.1155/2017/91891202-s2.0-85023626932HuangD. B.Simple adaptive-feedback controller for identical chaos synchronization200571303720310.1103/physreve.71.0372032-s2.0-41349088998HuangD. B.Adaptive-feedback control algorithm200673606620410.1103/physreve.73.0662042-s2.0-33744802665RenB. B.DaiJ. G.UDE-based robust boundary control for an unstable parabolic PDE with unknown input disturbance201893363368ZhongQ. C.KadirkamV.KupermanA.UDE-based controller equipped with a multi-band-stop filter to improve the voltage quality of inverters20176497433744310.1109/tie.2017.26773092-s2.0-85027050850KupermanA.ZhongQ.-C.UDE-based linear robust control for a class of nonlinear systems with application to wing rock motion stabilization2015811-278979910.1007/s11071-015-2029-x2-s2.0-84930765899GhoshD.BanerjeeS.Projective synchronization of time-varying delayed neural network with adaptive scaling factors2013531910.1016/j.chaos.2013.04.0072-s2.0-84878003424RenB.ZhongQ.-C.DaiJ.Asymptotic reference tracking and disturbance rejection of UDE-based robust control20176443166317610.1109/tie.2016.26334732-s2.0-85015264849RenB.ZhongQ.-C.ChenJ.Robust control for a class of nonaffine nonlinear systems based on the uncertainty and disturbance estimator20156295881588810.1109/tie.2015.24218842-s2.0-84939226697ZhongQ.-C.ReesD.Control of uncertain nonlinear system using an uncertainty and disturbance estimator2004126490591010.1115/1.18505292-s2.0-17044362753