Loading [MathJax]/jax/output/SVG/jax.js
Research article Special Issues

Fixed-deviation stabilization and synchronization for delayed fractional-order complex-valued neural networks

  • Academic editor: Xiaodi Li
  • In this paper, we study fixed-deviation stabilization and synchronization for fractional-order complex-valued neural networks with delays. By applying fractional calculus and fixed-deviation stability theory, sufficient conditions are given to ensure the fixed-deviation stabilization and synchronization for fractional-order complex-valued neural networks under the linear discontinuous controller. Finally, two simulation examples are presented to show the validity of theoretical results.

    Citation: Bingrui Zhang, Jin-E Zhang. Fixed-deviation stabilization and synchronization for delayed fractional-order complex-valued neural networks[J]. Mathematical Biosciences and Engineering, 2023, 20(6): 10244-10263. doi: 10.3934/mbe.2023449

    Related Papers:

    [1] Qiong Wu, Zhimin Yao, Zhouping Yin, Hai Zhang . Fin-TS and Fix-TS on fractional quaternion delayed neural networks with uncertainty via establishing a new Caputo derivative inequality approach. Mathematical Biosciences and Engineering, 2022, 19(9): 9220-9243. doi: 10.3934/mbe.2022428
    [2] Pan Wang, Xuechen Li, Qianqian Zheng . Synchronization of inertial complex-valued memristor-based neural networks with time-varying delays. Mathematical Biosciences and Engineering, 2024, 21(2): 3319-3334. doi: 10.3934/mbe.2024147
    [3] Biwen Li, Xuan Cheng . Synchronization analysis of coupled fractional-order neural networks with time-varying delays. Mathematical Biosciences and Engineering, 2023, 20(8): 14846-14865. doi: 10.3934/mbe.2023665
    [4] Yuan Ma, Yunxian Dai . Stability and Hopf bifurcation analysis of a fractional-order ring-hub structure neural network with delays under parameters delay feedback control. Mathematical Biosciences and Engineering, 2023, 20(11): 20093-20115. doi: 10.3934/mbe.2023890
    [5] Shuai Zhang, Yongqing Yang, Xin Sui, Yanna Zhang . Synchronization of fractional-order memristive recurrent neural networks via aperiodically intermittent control. Mathematical Biosciences and Engineering, 2022, 19(11): 11717-11734. doi: 10.3934/mbe.2022545
    [6] Ganesh Kumar Thakur, Sudesh Kumar Garg, Tej Singh, M. Syed Ali, Tarun Kumar Arora . Non-fragile synchronization of BAM neural networks with randomly occurring controller gain fluctuation. Mathematical Biosciences and Engineering, 2023, 20(4): 7302-7315. doi: 10.3934/mbe.2023317
    [7] Haydar Akca, Chaouki Aouiti, Farid Touati, Changjin Xu . Finite-time passivity of neutral-type complex-valued neural networks with time-varying delays. Mathematical Biosciences and Engineering, 2024, 21(5): 6097-6122. doi: 10.3934/mbe.2024268
    [8] Ruoyu Wei, Jinde Cao . Prespecified-time bipartite synchronization of coupled reaction-diffusion memristive neural networks with competitive interactions. Mathematical Biosciences and Engineering, 2022, 19(12): 12814-12832. doi: 10.3934/mbe.2022598
    [9] Gayathri Vivekanandhan, Hamid Reza Abdolmohammadi, Hayder Natiq, Karthikeyan Rajagopal, Sajad Jafari, Hamidreza Namazi . Dynamic analysis of the discrete fractional-order Rulkov neuron map. Mathematical Biosciences and Engineering, 2023, 20(3): 4760-4781. doi: 10.3934/mbe.2023220
    [10] Zhen Yang, Zhengqiu Zhang, Xiaoli Wang . New finite-time synchronization conditions of delayed multinonidentical coupled complex dynamical networks. Mathematical Biosciences and Engineering, 2023, 20(2): 3047-3069. doi: 10.3934/mbe.2023144
  • In this paper, we study fixed-deviation stabilization and synchronization for fractional-order complex-valued neural networks with delays. By applying fractional calculus and fixed-deviation stability theory, sufficient conditions are given to ensure the fixed-deviation stabilization and synchronization for fractional-order complex-valued neural networks under the linear discontinuous controller. Finally, two simulation examples are presented to show the validity of theoretical results.



    Fractional calculus is a theory of differentiation and integration of arbitrary order, which is an extension of integer order calculus. Initially, the study of fractional calculus theory was mainly conducted in the field of pure number theory, but as it developed further, fractional calculus was widely used in fluid mechanics [1], mechanical systems [2], signal processing [3,4], system identification [5], and many other fields. Fractional calculus has become an essential theory in many fields. Many scholars have applied fractional-order derivatives to neural networks and have built fractional-order neural networks (FONNs). So far, the study of FONNs has yielded some interesting results [6,7,8,9,10,11,12,13,14,15,16,17]. Zhang and Zeng [18] showed asymptotic stability of nonlinear FONNs with unbounded time-varying delays and asymptotic synchronization of FONNs under a linear controller. Ding et al. [19] investigated the robust finite-time stability of FONNs.

    Complex-valued neural networks (CVNNs), whose input/output signals, connection weights, and activation functions are derived from the complex domain. Unlike real-valued neural networks, functions that are both bounded and analytic in the complex domain must be constant according to Liouville's theorem [20]. Therefore, the study of the dynamics of CVNNs is essential. In recent years, the dynamic behavior of fractional-order CVNNs (FOCVNNs) has been reported in many kinds of literatures, including finite-time stability [21,22], impulse stability and synchronization [23], and Mittag-Leffler stability and synchronization [24,25].

    In neural networks, time delays are prevalent. Failure to take into account time delays will cause stable systems to be unstable and lead to a reduction in the capabilities of the neural network [26,27]. Therefore it is relevant to study FOCVNNs with time delays in practical applications. Bao et al. [28] obtained sufficient conditions to guarantee the synchronization of FOCVNNs with time delays using linear delay feedback control and fractional-order inequalities. Liu and Yu [29] derived several conditions for quasi-projective synchronization and complete synchronization of FOCVNNs with time delays based on generalized discrete fractional Halanay inequality and Lyapunov generalized function methods without dividing the complex-valued neural network into two real-valued systems.

    Deviation dynamics is particularly important for the evolutionary characterization of control systems. Fixed-deviation stabilization and synchronization are very important dynamical behaviors of discontinuous neural network systems. There have been some important findings about fixed-deviation dynamics [30,31]. Chen et al. [30] initially proposed the concept of fixed-deviation stability to describe the stability properties of discontinuous systems, and sufficient conditions to ensure globally uniform asymptotic fixed-deviation stability of delayed fractional-order memristive neural networks were given. Based on the theory of fixed-deviations in [30], Zhang [31] used linear-type discontinuous control and fractional-order calculus methods to address fixed-deviation stability and synchronization problems of FONNs. Clearly, the investigation of fixed-deviation dynamics for FONNs is an important topic. But so far, there are few results on the fixed-deviation dynamics of FOCVNNs.

    In the above view, we present the problems of fixed-deviation stability and synchronization of FOCVNNs. Continuous FONNs are difficult to achieve fixed-deviation stability and synchronization, and a special control method needs to be imposed to make the continuous system generate fixed-deviation dynamics behavior. A natural idea is to add a discontinuous controller so that continuous FOCVNNs turn into the discontinuous system under the discontinuous controller, and then impose complex-valued conditions to make the FOCVNNs achieve fixed-deviation stability and synchronization. Also based on the theory of fixed-deviations in [30], fractional-order calculus and Lyapunov method, sufficient conditions for the formation of fixed-deviation stability and synchronization of FOCVNNs under linear discontinuous controllers are obtained.

    In this section, necessary definitions and lemmas will be provided for the proof of the theorem in Section 3.

    The Caputo's fractional derivative of a function H(t)Cλ+1([t0,+),R) with order α>0 is defined by

    CDαt0H(t)=1Γ(λα)tt0H(λ)(s)(ts)αλ+1  ds,

    where tt0, λ1<α<λ, λ is positive integer, α is a positive constant and Γ() is Gamma function, that is

    Γ(α)=0tα1etdt.

    The Riemann-Liouville fractional derivative of order α>0 for a function H(t)Cλ+1([t0,+),R) is defined by

    RLDαt0H(t)=1Γ(λα)dλdtλtt0H(s)(ts)αλ+1  ds,

    where λ1<α<λ, λ>0.

    By the above definition, the following relation holds:

    CDαt0H(t)=RLDαt0H(t)H(t0)Γ(1α)(tt0)α.

    Now, we introduce delayed FOCVNNs as follows:

    CDαt0zk(t)=akzk(t)+n=1bkf(z(t))+n=1dklg(z(tϖkl(t)))+Uk(t), (2.1)

    where 0<α<1, zk(t)C denotes the state variable; ak>0 is the self-feedback connective weight of the kth neuron; bk and dk are the connective weights matrix without and with time delay respectively; f(z(t)), g(z(tϖk(t))) represent the complex-valued state activation functions at time t and tϖk(t); ϖk(t) is the time-varying delay satisfying 0ϖk(t)ϖ; Uk(t) stands for the external input.

    Let Cϖ=C([ϖ,0],Rn) be the Banach space of continuous functions mapping [ϖ,0] into Rn. For ψCϖ,ψc=supϖs0ψ(s).

    Note the initial conditions of delayed FOCVNNs (2.1) as

    zk(t0+s)=ψRk(s)+ψIk(s),ϖs0,k=1,,n. (2.2)

    Let z=zR+izIC. For any , f(z) and g(z(tϖ)) can be shown by dividing into its real and imaginary parts as

    f(z)=fR(zR,zI)+ifI(zR,zI),g(z(tϖ))=gR(zR(tϖ),zI(tϖ))+igI(zR(tϖ),zI(tϖ)). (2.3)

    Let zk(t)=zRk(t)+izIk(t). Delayed FOCVNNs (2.1) can be described as the following equation:

    CDαt0zRk(t)=akzRk(t)+n=1bRkfR(z(t))n=1bIkfI(z(t))+n=1dRkgR(z(tϖk(t)))n=1dIkgI(z(tϖk(t)))+URk(t),
    CDαt0zIk(t)=akzIk(t)+n=1bRkfI(z(t))+n=1bIkfR(z(t))+n=1dRkgI(z(tϖk(t)))+n=1dIkgR(z(tϖk(t)))+UIk(t). (2.4)

    Definition 1 ([30]): FOCVNNs (2.1) is called globally uniformly β-stable if for any ξ>0 and any initial values ϕ,φCϖ, ϕφCξ, there is a constant T(ξ)0, such that

    z(t,t0,ϕ)z(t,t0,φ)β

    for all tt0+T(ξ), where β>0.

    Remark 1: β-stability, also known as fixed-deviation stability, specifically, when the difference between two different initial values of the described neural network are kept in a certain range, the difference among final values of the system trajectories starting from these two initial values will be maintained in a fixed-deviation degree.

    Definition 2: The zero solution of delayed FOCVNNs (2.1) is called globally uniformly β-stable if for any ψCϖ, ξ>0, ψCξ, there is a constant T(ξ)0, such that

    z(t,t0,ψ)β

    for all tt0+T(ξ), where β>0 is a constant.

    In this paper, we propose the below assumptions:

    (i) The activation functions f() and g() satisfy f(0)=g(0)=0.

    (ii) For functions fR(,), fI(,), gR(,), gI(,), there exist positive constants FRR, FRI, FIR, FII, GRR, GRI, GIR, GII, such that

    {|fR(˜zR,˜zI)fR(zR,zI)|FRR|˜zRzR|+FRI|˜zIzI||fI(˜zR,˜zI)fI(zR,zI)|FIR|˜zRzR|+FII|˜zIzI||gR(˜zR,˜zI)gR(zR,zI)|GRR|˜zRzR|+GRI|˜zIzI||gI(˜zR,˜zI)gI(zR,zI)|GIR|˜zRzR|+GII|˜zIzI|. (2.5)

    Remark 2: Condition (i) holds if and only if both its real and imaginary parts are 0, i.e., fR(0,0)=fI(0,0)=0 and gR(0,0)=gI(0,0)=0 for any R.

    Next, we present two necessary lemmas.

    Lemma 1 ([30]): If functions f(t) and g(t) together with their derivatives are continuous in [t0,t], then fractional differentiation of the Leibniz rule is in the form

    RLDαt0(p(t)q(t))=nm=0(αm)dmp(t)dtmRLDαmt0q(t)Iαn(t),

    where nα+1,

    (αm)=Γ(α+1)m!Γ(αm+1),

    and

    Iαn(t)=(1)n(tα)nα+1n!Γ(α)1010ϝα(t,ζ,)dζd,ϝα(t,ζ,)=q(t0+(tt0))p(n+1)(t0+(tt0)(ζ+ζ)).

    Lemma 2: For a continuous differentiable function P(t):[t0,+)[0,+) and Q(t)=(tt0+σ)αP(t), then

    CDαt0Q(t)(tt0+σ)αCDαt0P(t)+1α+α2σαΓ(2α)¯Q(t),

    where tt0, σ>0 and ¯Q(t)=supt0stQ(s).

    Proof: From Lemma 1, we know

    CDαt0Q(t)=RLDαt0Q(t)Q(t0)Γ(1α)(tt0)α=(tt0+σ)αRLDαt0P(t)+α2(tt0+σ)α1RLDα1t0P(t)Rα2(t)σαP(t0)Γ(1α)(tt0)α(tt0+σ)α(CDαt0P(t)+P(t0)Γ(1α)(tt0)α)σαP(t0)Γ(1α)(tt0)α+α2(tt0+σ)α1RLDα1t0P(t)(tt0+σ)αCDαt0P(t)+P(t0)Γ(1α)+α2(tt0+σ)α1RLDα1t0P(t)(tt0+σ)αCDαt0P(t)+¯Q(t)σαΓ(1α)+α2(tt0+σ)α1RLDα1t0P(t).

    Also by the definition of Riemann-Liouville fractional derivative,

    α2(tt0+σ)α1RLDα1t0P(t)=α2Γ(1α)(tt0+σ)α1tt0(ts)αP(s)dsα2σαΓ(2α)¯Q(t).

    Therefore,

    CDαt0Q(t)(tt0+σ)αCDαt0P(t)+¯Q(t)σαΓ(1α)+α2σαΓ(2α)¯Q(t)=(tt0+σ)αCDαt0P(t)+1α+α2σαΓ(2α)¯Q(t)

    for tt0. Proof of Lemma 2 is finished.

    In this section, we will provide some sufficient conditions to guarantee fixed-deviation stability and synchronization of delayed FOCVNNs (2.1).

    We design linear discontinuous control for system (2.1):

    Uk(t)=Mkzk(t)+Nk[sgn(zRk(t))+isgn(zIk(t))], (3.1)

    where k=1,,n.

    Thus by controller (3.1), system (2.4) is converted as

    CDαt0zRk(t)=akzRk(t)+n=1bRkfR(z(t))n=1bIkfI(z(t))+n=1dRkgR(z(tϖk(t)))n=1dIkgI(z(tϖk(t)))+MkzRk(t)+Nksgn(zRk(t)),
    CDαt0zIk(t)=akzIk(t)+n=1bRkfI(z(t))+n=1bIkfR(z(t))+n=1dRkgI(z(tϖk(t)))+n=1dIkgR(z(tϖk(t)))+MkzIk(t)+Nksgn(zIk(t)). (3.2)

    Theorem 1: If there are positive constants σ>ϖ0 and μr>0,vr>0(r=1,,n) such that the following conditions

    ar|Mr|1α+α2σαΓ(2α)1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)+(|dRr|(GRR+GRI)+|dIr|(GIR+GII))(σϖ+σ)α]μ>0 (3.3)

    and

    ar|Mr|1α+α2σαΓ(2α)1vrn=1[|bRr|(FIR+FII)+|bIr|(FRR+FRI)+(|dRr|(GIR+GII)+|dIr|(GRR+GRI))(σϖ+σ)α]v>0 (3.4)

    hold, then delayed FOCVNNs (3.2) is globally uniformly β-stable, that is, delayed FOCVNNs (2.1) is globally uniformly β-stable via control rule (3.1).

    Proof: Construct an auxiliary function as follows

    P(t)=max1knmax{|zRk(t)|μk,|zIk(t)|vk}.

    Let

    Q(t)=(tt0+σ)αP(t),¯Q(t)=supt0σstQ(s).

    There exists r{1,,n} for given tt0 having

    P(t)=max{|zRr(t)|μr,|zIr(t)|vr}.

    Then we get P(t)=|zRr(t)|μr, P(t)=|zIr(t)|vr. Now, we let P(t)=|zRr(t)|μr, and another case is similar. By (2.5) and (2.7) it follows that

    CDαt0P(t)=1μrCDαt0|zRr(t)|sgn(zRr(t))μrCDαt0zRr(t)(ar|Mr|)μr|zRr(t)|+|Nr|μr+1μrn=1|bRr|(FRR|zR(t)|+FRI|zI(t)|)+1μrn=1|bIr|(FIR|zR(t)|+FII|zI(t)|)+1μrn=1|dRr|(GRR|zR(tϖr(t))|+GRI|zI(tϖr(t))|)+1μrn=1|dIr|(GIR|zR(tϖr(t))|+GII|zI(tϖr(t))|)(ar|Mr|)P(t)+|Nr|μr+1μrn=1|bRr|(FRR+FRI)μP(t)+1μrn=1|bIr|(FIR+FII)μP(t)+1μrn=1|dRr|(GRR+GRI)μP(tϖr(t))+1μrn=1|dIr|(GIR+GII)μP(tϖr(t))={(ar|Mr|)+1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)]μ}P(t)+1μrn=1[|dRr|(GRR+GRI)+|dIr|(GIR+GII)]μP(tϖr(t))+|Nr|μr. (3.5)

    By using Lemma 2 and (3.5), then

    CDαt0Q(t)(tt0+σ)αCDαt0P(t)+1α+α2σαΓ(2α)¯Q(t){(ar|Mr|)+1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)]μ}Q(t)+1μrn=1[|dRr|(GRR+GRI)+|dIr|(GIR+GII)]μ(tt0+σtϖr(t)t0+σ)αQ(tϖr(t))+(tt0+σ)α|Nr|μr+1α+α2σαΓ(2α)¯Q(t){(ar|Mr|)+1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)]μ}Q(t)+1μrn=1[|dRr|(GRR+GRI)+|dIr|(GIR+GII)]μ(tt0+σtϖr(t)t0+σ)α¯Q(t)+(tt0+σ)α|Nr|μr+1α+α2σαΓ(2α)¯Q(t).

    It is known that σ+EEϖr(t)+σ is monotone non-increasing for E0, and thus

    tt0+σtϖr(t)t0+σσϖrl(t)+σσϖ+σ,

    therefore,

    CDαt0Q(t){(ar|Mr|)+1α+α2σαΓ(2α)+1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)+(|dRr|(GRR+GRI)+|dIr|(GIR+GII))(σϖ+σ)α]μ}Q(t)+(tt0+σ)α|Nr|μrAQ(t)+(tt0+σ)αB (3.6)

    when Q(t)=¯Q(t), for tt0, where

    Amin1rn{ar|Mr|1α+α2σαΓ(2α)1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)+(|dRr|(GRR+GRI)+|dIr|(GIR+GII))(σϖ+σ)α]μ},Bmax1rn(|Nr|μr).

    Next, from the definition ¯Q(t)=supt0σstQ(s), we will divide into three cases to prove fixed-deviation stable.

    Case 1: ¯Q(s)>Q(s) for any t0<st. Now, we consider ¯Q(t) is the maximum value of Q(s) at moment t0, that is

    ¯Q(t)=¯Q(t0),tt0.

    Hence,

    z(t)μP(t)=μ(tt0+σ)αQ(t)μ(tt0+σ)α¯Q(t)=μ(tt0+σ)α¯Q(t0)μσα(tt0+σ)μminψCμσαξ(tt0+σ)μmin,

    when ψCξ, where μmin=min1rn{μr}.

    Case 2: ¯Q(t)=Q(t). We obtain

    CDαt0¯Q(t)CDαt0Q(t),tt0. (3.7)

    From divisional integration method, we have

    tt0¯Q(s)Q(s)(ts)αds=limst¯Q(s)Q(s)(ts)α¯Q(t0)Q(t0)(tt0)ααtt0¯Q(s)Q(s)(ts)α+1ds=limst1α[¯Q(s)Q(s)](ts)1α¯Q(t0)Q(t0)(tt0)ααtt0¯Q(s)Q(s)(ts)α+1ds=¯Q(t0)Q(t0)(tt0)ααtt0¯Q(s)Q(s)(ts)α+1ds0,

    thus, (3.7) holds.

    Next, we demand

    P(t)BA,tt0. (3.8)

    Otherwise, from (3.6) and (3.7) we have

    CDαt0¯Q(t)CDαt0Q(t)AQ(t)+(tt0+σ)αBA(tt0+σ)αP(t)+(tt0+σ)αB<0.

    It is known that ¯Q(t) is monotonically increasing, so ¯Q(t)0, then

    CDαt0¯Q(t)=1Γ(1α)tt0¯Q(s)(ts)αds0,

    which is a contradiction. Hence, (3.8) is true.

    Therefore,

    z(t)μP(t)μBA

    for tt0.

    Case 3: ¯Q(ˆt)=Q(ˆt), t0ˆt<t, and ¯Q(s)>Q(s), for s(ˆt,t].

    Combining Cases 1 and 2, we get

    P(ˆt)BA

    and

    Q(t)<¯Q(t)=¯Q(ˆt)=Q(ˆt)=(ˆtt0+σ)αP(ˆt)(ˆtt0+σ)αBA.

    Therefore, for tt0

    z(t)μP(t)=μQ(t)(tt0+σ)αμBA.

    In conclusion, let

    T(ξ)=max{[(AξBμmin)1α1],0},

    then

    z(t)μBAβ

    for all tt0+T(ξ), when ψCξ. So, it can be inferred that then delayed FOCVNNs (2.1) is globally uniformly β-stable via control rule (3.1).

    Corollary 1: If there are n positive constants μr,vr such that

    ar|Mr|1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)+(|dRr|(GRR+GRI)+|dIr|(GIR+GII))]μ>0 (3.9)

    and

    ar|Mr|1vrn=1[|bRr|(FIR+FII)+|bIr|(FRR+FRI)+(|dRr|(GIR+GII)+|dIr|(GRR+GRI))]v>0 (3.10)

    hold, then delayed FOCVNNs (2.1) is globally uniformly fixed-deviation stable via control rule (3.1).

    Proof: Let

    L(ϑ)=ar|Mr|1α+α2ϑαΓ(2α)1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)+(|dRr|(GRR+GRI)+|dIr|(GIR+GII))(ϑϖ+ϑ)α]μ,
    X(ϑ)=ar|Mr|1α+α2ϑαΓ(2α)1vrn=1[|bRr|(FIR+FII)+|bIr|(FRR+FRI)+(|dRr|(GIR+GII)+|dIr|(GRR+GRI))(ϑϖ+ϑ)α]v,

    where ϑ>ϖ, then from conditions (3.9), (3.10),

    limϑ+L(ϑ)=ar|Mr|1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)+(|dRr|(GRR+GRI)+|dIr|(GIR+GII))]μ>0,
    limϑ+X(ϑ)=ar|Mr|1vrn=1[|bRr|(FIR+FII)+|bIr|(FRR+FRI)+(|dRr|(GIR+GII)+|dIr|(GRR+GRI))]v>0.

    By the property of the limit, there is a constant σ>ϖ such that L(σ)>0 and X(σ)>0. So (3.3) and (3.4) hold. The proof is completed.

    Regard the following system (3.11) as the drive system,

    CDαt0zk(t)=akzk(t)+n=1bkf(z(t))+n=1dkg(z(tϖk(t))), (3.11)

    and the response system is defined by the following:

    CDαt0˜zk(t)=ak˜zk(t)+n=1bkf(˜z(t))+n=1dkg(˜z(tϖk(t)))+Uk(t). (3.12)

    where ˜zk(t)=˜zRk(t)+i˜zIk(t).

    The initial values of system (3.12) is given by

    ˜zk(t0+s)=˜ψRk(s)+˜ψIk(s),ϖs0.

    Define Λk(t)=˜zk(t)zk(t),

    ΛRk(t)=˜zRk(t)zRk(t),ΛIk(t)=˜zIk(t)zIk(t),

    then we consider the following error system

    CDαt0Λk(t)=akΛk(t)+n=1bkf(Λ(t))+n=1dkg(Λ(tϖk(t)))+Uk(t), (3.13)

    where

    f(Λ(t))=f(˜z(t))f(z(t)),g(Λ(tϖk(t)))=g(˜z(tϖk(t)))g(z(tϖkl(t))).

    The initial value of the error system (3.13) is noted in the following form:

    Λk(t0+s)=˜ψk(s)ψk(s)=Ψk(s),ϖs0.

    For error system (3.13), we construct the following controller:

    Uk(t)=MkΛk(t)+Nk[sgn(ΛRk(t))+isgn(ΛIk(t))]. (3.14)

    Thus by controller (3.14), system (3.13) is converted as

    CDαt0ΛRk(t)=akΛRk(t)+n=1bRkfR(Λ(t))n=1bIkfI(Λ(t))+n=1dRkgR(Λ(tϖk(t)))n=1dIkgI(Λ(tϖk(t)))+MkΛRk(t)+Nksgn(ΛRk(t)),CDαt0ΛIk(t)=akΛIk(t)+n=1bRkfI(Λ(t))+n=1bIkfR(Λ(t))+n=1dRkgI(Λ(tϖk(t)))+n=1dIkgR(Λ(tϖk(t)))+MkΛIk(t)+Nksgn(ΛIk(t)). (3.15)

    Under assumption (ii), the following inequality holds:

    {|fR(Λ(t))|FRR|ΛR(t)|+FRI|ΛI(t)||fI(Λ(t))|FIR|ΛR(t)|+FII|ΛI(t)||gR(Λ(tϖk(t)))|GRR|ΛR(tϖk(t))|+GRI|ΛI(tϖk(t))||gI(Λ(tϖk(t)))|GIR|ΛR(tϖk(t))|+GII|ΛI(tϖk(t))|. (3.16)

    Theorem 2: If there are positive constants σ>ϖ0 and μr>0,vr>0(r=1,,n) such that the following conditions

    ar|Mr|1α+α2σαΓ(2α)1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)+(|dRr|(GRR+GRI)+|dIr|(GIR+GII))(σϖ+σ)α]μ>0 (3.17)

    and

    ar|Mr|1α+α2σαΓ(2α)1vrn=1[|bRr|(FIR+FII)+|bIr|(FRR+FRI)+(|dRr|(GIR+GII)+|dIr|(GRR+GRI))(σϖ+σ)α]v>0 (3.18)

    hold, then delayed FOCVNNs (3.15) is globally uniformly β-stable. In other words, fixed-deviation synchronization between the drive system (3.11) and the response system (3.12) can be achieved.

    Proof: Construct an auxiliary function as follows

    P(t)=max1knmax{|ΛRk(t)|μk,|ΛIk(t)|vk}.

    Let

    Q(t)=(tt0+σ)αP(t),¯Q(t)=supt0σstQ(s).

    There exists r{1,,n} for given tt0 having

    P(t)=max{|ΛRr(t)|μr,|ΛIr(t)|vr}.

    Then we get P(t)=|ΛRr(t)|μr, P(t)=|ΛIr(t)|vr. Now, we let P(t)=|ΛRr(t)|μr, another case is similar. By (3.15) and (3.16) it follows that

    CDαt0P(t)=1μrCDαt0|ΛRr(t)|sgn(ΛRr(t))μrCDαt0ΛRr(t)(ar|Mr|)μr|ΛRr(t)|+|Nr|μr+1μrn=1|bRr|(FRR|ΛR(t)|+FRI|ΛI(t)|)+1μrn=1|bIr|(FIR|ΛR(t)|+FII|ΛI(t)|)+1μrn=1|dRr|(GRR|ΛR(tϖr(t))|+GRI|ΛI(tϖr(t))|)+1μrn=1|dIr|(GIR|ΛR(tϖr(t))|+GII|ΛI(tϖr(t))|)(ar|Mr|)P(t)+|Nr|μr+1μrn=1|bRr|(FRR+FRI)μP(t)+1μrn=1|bIr|(FIR+FII)μP(t)+1μrn=1|dRr|(GRR+GRI)μP(tϖr(t))+1μrn=1|dIr|(GIR+GII)μP(tϖr(t))={(ar|Mr|)+1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)]μ}P(t)+1μrn=1[|dRr|(GRR+GRI)+|dIr|(GIR+GII)]μP(tϖr(t))+|Nr|μr. (3.19)

    By applying Lemma 2 and (3.19), we have

    CDαt0Q(t)(tt0+σ)αCDαt0P(t)+1α+α2σαΓ(2α)¯Q(t){(ar|Mr|)+1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)]μ}Q(t)+1μrn=1[|dRr|(GRR+GRI)+|dIr|(GIR+GII)]μ(tt0+σtϖr(t)t0+σ)αQ(tϖr(t))+(tt0+σ)α|Nr|μr+1α+α2σαΓ(2α)¯Q(t){(ar|Mr|)+1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)]μ}Q(t)+1μrn=1[|dRr|(GRR+GRI)+|dIr|(GIR+GII)]μ(tt0+σtϖr(t)t0+σ)α¯Q(t)+(tt0+σ)α|Nr|μr+1α+α2σαΓ(2α)¯Q(t).

    It is known that σ+EEϖr(t)+σ is monotone non-increasing for E0, and thus

    tt0+σtϖr(t)t0+σσϖr(t)+σσϖ+σ,

    therefore,

    CDαt0Q(t){(ar|Mr|)+1α+α2σαΓ(2α)+1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)+(|dRr|(GRR+GRI)+|dIr|(GIR+GII))(σϖ+σ)α]μ}Q(t)+(tt0+σ)α|Nr|μrAQ(t)+(tt0+σ)αB, (3.20)

    when Q(t)=¯Q(t), for tt0, where

    Amin1rn{ar|Mr|1α+α2σαΓ(2α)1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)+(|dRr|(GRR+GRI)+|dIr|(GIR+GII))(σϖ+σ)α]μ},Bmax1rn(|Nr|μr).

    Similar to cases 1–3 in Theorem 1, we finally obtain

    z(t)μBAβ

    for all tt0+T(ξ), when ΨCξ, where

    T(ξ)=max{[(AξBμmin)1α1],0}.

    So, fixed-deviation synchronization between the drive system (3.11) and the response system (3.12) can be achieved.

    Corollary 2: If there are n positive constants μr,vr such that

    ar|Mr|1μrn=1[|bRr|(FRR+FRI)+|bIr|(FIR+FII)+(|dRr|(GRR+GRI)+|dIr|(GIR+GII))]μ>0 (3.21)

    and

    ar|Mr|1vrn=1[|bRr|(FIR+FII)+|bIr|(FRR+FRI)+(|dRr|(GIR+GII)+|dIr|(GRR+GRI))]v>0 (3.22)

    hold, then delayed FOCVNNs (3.15) is globally uniformly β-stable. In other words, fixed-deviation synchronization between the drive system (3.11) and the response system (3.12) can be achieved.

    Proof: The proof of Corollary 2 is similar to the proof of Corollary 1.

    Example 1: We consider the following delayed FOCVNNs:

    CD0.9t0zk(t)=2z1(t)+b11f1(z1(t))+b12f2(z2(t))+d11g1(z1(t1))+d12g2(z2(t1))+U1(t),
    CD0.9t0zk(t)=4z2(t)+b21f1(z1(t))+b22f2(z2(t))+d21g1(z1(t1))+d22g2(z2(t1))+U2(t), (4.1)

    where t0=0, zk(t)=zRk(t)+izIk(t), f(z)=g(z)=tanh(zR)+tanh(zI)i, =1,2,

    B=(bk)2×2=(0.03+0.05i0.05+0.04i0.020.01i0.03+0.02i),D=(dk)2×2=(0.070.02i0.03+0.01i0.05+0.03i0.01+0.05i).

    It's not hard to choose that FRR+FRI=FIR+FII=2, GRR+GRI=GIR+GII=2.

    Let M1=0.06, M2=0.04, σ=10, μ1=μ2=1, v1=v2=1, then (3.3) and (3.4) hold. Hence, system (4.1) is globally uniformly β-stable from Theorem 1. Figure 1 shows the numerical simulation of FOCVNNs (4.1) under discontinuous control rules U1(t)=0.06z1(t)+1.19[sgn(zR1(t))+isgn(zI1(t))] and U2(t)=0.04z2(t)+1.19[sgn(zR2(t))+isgn(zI2(t))].

    Figure 1.  The fixed-deviation stabilization of system (4.1), where β=N(β1+β2),N=10.

    Example 2: Regard the following FOCVNNs (4.2) as the drive system:

    CD0.95t0zk(t)=5z1(t)+b11f1(z1(t))+b12f2(z2(t))+d11g1(z1(t1))+d12g2(z2(t1)),
    CD0.95t0zk(t)=3z2(t)+b21f1(z1(t))+b22f2(z2(t))+d21g1(z1(t1))+d22g2(z2(t1)), (4.2)

    where t0=0, f(z)=g(z)=cos(zR)+cos(zI)i, =1,2,

    B=(bk)2×2=(0.02+0.05i0.04+0.06i0.020.01i0.07+0.09i),D=(dk)2×2=(0.030.02i0.08+0.09i0.05+0.03i0.01+0.05i).

    The response system is given by

    CD0.95t0˜zk(t)=5˜z1(t)+b11f1(˜z1(t))+b12f2(˜z2(t))+d11g1(˜z1(t1))+d12g2(˜z2(t1))+U1(t),
    CD0.95t0˜zk(t)=3z2(t)+b21f1(˜z1(t))+b22f2(˜z2(t))+d21g1(˜z1(t1))+d22g2(˜z2(t1))+U2(t), (4.3)

    where the parameters bk, dk, f(), g() are all the same as in FOCVNNs (4.2).

    As above choose that FRR+FRI=FIR+FII=2, GRR+GRI=GIR+GII=2.

    Let M1=0.02, M2=0.04, σ=20, μ1=μ2=1, v1=v2=1, then (3.17) and (3.18) hold. Hence, it can be seen that drive system (4.2) and response system (4.3) are fixed-deviation synchronization from Theorem 2. Figure 2 shows the numerical simulation of error system under discontinuous control rules U1(t)=0.02Λ1(t)+2.22[sgn(ΛR1(t))+isgn(ΛI1(t))] and U2(t)=0.04Λ2(t)+2.22[sgn(ΛR2(t))+isgn(ΛI2(t))].

    Figure 2.  The fixed-deviation synchronization of drive system (4.2) and response system (4.3), where β=N(β1+β2+β3+β4),β1+β2=β3+β4=0.996,N=10.

    This paper discusses fixed-deviation stability and synchronization of FOCVNNs. The system investigated in this paper is a continuous neural network, and a discontinuous controller is introduced to address this problem. Under the discontinuous controller, fixed-deviation stability theory and fractional calculus method are used to observe the fixed-deviation dynamical behavior of delayed FOCVNNs. In this paper, a continuous system is transformed into a discontinuous system by imposing a discontinuous controller to achieve fixed-deviation dynamics, this technique can be extended to other more complex systems, which would be a future direction of research.

    This work is supported by the Natural Science Foundation of China under Grant 61976084, the Natural Science Foundation of Hubei Province of China under Grant 2021CFA080, the Young Top-Notch Talent Cultivation Program of Hubei Province of China.

    The authors declare there is no conflict of interest.



    [1] V. V. Kulish, J. L. Lage, Application of fractional calculus to fluid mechanics, J. Fluids Eng., 124 (2002), 803–806. https://doi.org/10.1115/1.1478062 doi: 10.1115/1.1478062
    [2] J. M. Balthazar, P. B. Goncalves, S. Lenci, Y. V. Mikhlin, Models, methods, and applications of dynamics and control in engineering sciences: state of the art, Math. Probl. Eng., 2010 (2010), 487684. https://doi.org/10.1155/2010/487684 doi: 10.1155/2010/487684
    [3] P. Panda, M. Dash, Fractional generalized splines and signal processing, Signal Process., 86 (2006), 2340–2350. https://doi.org/10.1016/j.sigpro.2005.10.017 doi: 10.1016/j.sigpro.2005.10.017
    [4] M. S. Aslam, M. A. Z. Raja, A new adaptive strategy to improve online secondary path modeling in active noise control systems using fractional signal processing approach, Signal Process., 107 (2015), 433–443. https://doi.org/10.1016/j.sigpro.2014.04.012 doi: 10.1016/j.sigpro.2014.04.012
    [5] C. J. Z. Aguilar, J. F. Gmez-Aguilar, V. M. Alvarado-Martnez, H. M. Romero-Ugalde, Fractional order neural networks for system identification, Chaos, Solitons Fractals, 130 (2020), 109444. https://doi.org/10.1016/j.chaos.2019.109444 doi: 10.1016/j.chaos.2019.109444
    [6] S. Fazzino, R. Caponetto, L. Patanè, A new model of Hopfield network with fractional-order neurons for parameter estimation, Nonlinear Dyn., 104 (2021), 2671–2685. https://doi.org/10.1007/s11071-021-06398-z doi: 10.1007/s11071-021-06398-z
    [7] Y. Liu, Y. Sun, L. Liu, Stability analysis and synchronization control of fractional-order inertial neural networks with time-varying delay, IEEE Access, 10 (2022), 56081–56093. https://doi.org/10.1109/ACCESS.2022.3178123 doi: 10.1109/ACCESS.2022.3178123
    [8] E. Kaslik, S. Sivasundaram, Nonlinear dynamics and chaos in fractional-order neural networks, Neural Networks, 32 (2012), 245–256. https://doi.org/10.1016/j.neunet.2012.02.030 doi: 10.1016/j.neunet.2012.02.030
    [9] H. Wang, Y. Yu, G. Wen, S. Zhan, J. Yu, Global stability analysis of fractional-order Hopfield neural networks with time delay, Neurocomputing, 154 (2015), 15–23. https://doi.org/10.1016/j.neucom.2014.12.031 doi: 10.1016/j.neucom.2014.12.031
    [10] C. Huang, J. Wang, X. Chen, J. Cao, Bifurcations in a fractional-order BAM neural network with four different delays, Neural Networks, 141 (2021), 344–354. https://doi.org/10.1016/j.neunet.2021.04.005 doi: 10.1016/j.neunet.2021.04.005
    [11] C. Xu, D. Mu, Z. Liu, Y. Pang, M. Liao, C. Aouiti, New insight into bifurcation of fractional-order 4D neural networks incorporating two different time delays, Commun. Nonlinear Sci. Numer. Simul., 118 (2023), 107043. https://doi.org/10.1016/j.cnsns.2022.107043 doi: 10.1016/j.cnsns.2022.107043
    [12] C. Huang, H. Liu, X. Shi, X. Chen, M. Xiao, Z. Wang, et al., Bifurcations in a fractional-order neural network with multiple leakage delays, Neural Networks, 131 (2020), 115–126. https://doi.org/10.1016/j.neunet.2020.07.015 doi: 10.1016/j.neunet.2020.07.015
    [13] C. Xu, W. Zhang, C. Aouiti, Z. Liu, L. Yao, Bifurcation insight for a fractional-order stage-structured predator-prey system incorporating mixed time delays, Math. Methods Appl. Sci., 2023. https://doi.org/10.1002/mma.9041
    [14] C. Xu, D. Mu, Z. Liu, Y. Pang, M. Liao, P. Li, et al., Comparative exploration on bifurcation behavior for integer-order and fractional-order delayed BAM neural networks, Nonlinear Anal. Modell. Control, 27 (2022), 1030–1053. https://doi.org/10.15388/namc.2022.27.28491 doi: 10.15388/namc.2022.27.28491
    [15] C. Xu, Z. Liu, Y. Pang, S. Saifullah, J. Khan, Torus and fixed point attractors of a new hyperchaotic 4D system, J. Comput. Sci., 67 (2023), 101974. https://doi.org/10.1016/j.jocs.2023.101974 doi: 10.1016/j.jocs.2023.101974
    [16] C. Xu, M. Rahman, D. Baleanu, On fractional-order symmetric oscillator with offset-boosting control, Nonlinear Anal. Modell. Control, 27 (2022), 1–15. https://doi.org/10.15388/namc.2022.27.28279 doi: 10.15388/namc.2022.27.28279
    [17] C. Xu, W. Alhejaili, S. Saifullah, A. Khan, J. Khan, M. A. El-Shorbagy, Analysis of Huanglongbing disease model with a novel fractional piecewise approach, Chaos Solitons Fractals, 161 (2022), 112316. https://doi.org/10.1016/j.chaos.2022.112316 doi: 10.1016/j.chaos.2022.112316
    [18] F. Zhang, Z. Zeng, Asymptotic stability and synchronization of fractional-order neural networks with unbounded time-varying delays, IEEE Trans. Syst. Man Cybern. Syst., 51 (2021), 5547–5556. https://doi.org/10.1109/TSMC.2019.2956320 doi: 10.1109/TSMC.2019.2956320
    [19] Z. Ding, Z. Zeng, L. Wang, Robust finite-time stabilization of fractional-order neural networks with discontinuous and continuous activation functions under uncertainty, IEEE Trans. Neural Networks Learn. Syst., 29 (2018), 1477–1490. https://doi.org/10.1109/TNNLS.2017.2675442 doi: 10.1109/TNNLS.2017.2675442
    [20] W. Rudin, Real and Complex Analysis, Mcgraw-Hill, New York, 1987.
    [21] X. Ding, J. Cao, X. Zhao, F. E. Alsaadi, Finite-time stability of fractional-order complex-valued neural networks with time delays, Neural Process. Lett., 46 (2017), 561–580. https://doi.org/10.1007/s11063-017-9604-8 doi: 10.1007/s11063-017-9604-8
    [22] T. Hu, Z. He, X. Zhang, S. Zhong, Finite-time stability for fractional-order complex-valued neural networks with time delay, Appl. Math. Comput., 365 (2020), 124715. https://doi.org/10.1016/j.amc.2019.124715 doi: 10.1016/j.amc.2019.124715
    [23] P. Wan, J. Jian, Impulsive stabilization and synchronization of fractional-order complex-valued neural networks, Neural Process. Lett., 50 (2019), 2201–2218. https://doi.org/10.1007/s11063-019-10002-2 doi: 10.1007/s11063-019-10002-2
    [24] X. You, Q. Song, Z. Zhao, Global Mittag-Leffler stability and synchronization of discrete-time fractional-order complex-valued neural networks with time delay, Neural Networks, 122 (2020), 382–394. https://doi.org/10.1016/j.neunet.2019.11.004 doi: 10.1016/j.neunet.2019.11.004
    [25] J. Chen, B. Chen, Z. Zeng, Global asymptotic stability and adaptive ultimate Mittag-Leffler synchronization for a fractional-order complex-valued memristive neural networks with delays, IEEE Trans. Syst. Man Cybern. Syst., 49 (2019), 2519–2535. https://doi.org/10.1109/TSMC.2018.2836952 doi: 10.1109/TSMC.2018.2836952
    [26] X. Li, J. Wu, Stability of nonlinear differential systems with state-dependent delayed impulses, Automatica, 64 (2016), 63–69. https://doi.org/10.1016/j.automatica.2015.10.002 doi: 10.1016/j.automatica.2015.10.002
    [27] X. Li, S. Song, Stabilization of delay systems: delay-dependent impulsive control, IEEE Trans. Autom. Control, 62 (2017), 406–411. https://doi.org/10.1109/TAC.2016.2530041 doi: 10.1109/TAC.2016.2530041
    [28] H. Bao, J. H. Park, J. Cao, Synchronization of fractional-order complex-valued neural networks with time delay, Neural Networks, 81 (2016), 16–28. https://doi.org/10.1016/j.neunet.2016.05.003 doi: 10.1016/j.neunet.2016.05.003
    [29] X. Liu, Y. Yu, Synchronization analysis for discrete fractional-order complex-valued neural networks with time delays, Neural Comput. Appl., 33 (2021), 10503–10514. https://doi.org/10.1007/s00521-021-05808-y doi: 10.1007/s00521-021-05808-y
    [30] J. Chen, B. Chen, Z. Zeng, Global uniform asymptotic fixed-deviation stability and stability for delayed fractional-order memristive neural networks with generic memductance, Neural Networks, 98 (2018), 65–75. https://doi.org/10.1016/j.neunet.2017.11.004 doi: 10.1016/j.neunet.2017.11.004
    [31] J. Zhang, Linear-type discontinuous control of fixed-deviation stabilization and synchronization for fractional-order neurodynamic systems with communication delays, IEEE Access, 6 (2018), 52570–52581. https://doi.org/10.1109/ACCESS.2018.2870979 doi: 10.1109/ACCESS.2018.2870979
  • This article has been cited by:

    1. Zichen Yao, Zhanwen Yang, Yongqiang Fu, Long time decay analysis of complex-valued fractional abstract evolution equations with delay, 2024, 460, 00963003, 128292, 10.1016/j.amc.2023.128292
    2. Yali Cheng, Wanying Yang, Wenbo Xu, Shouming Zhong, Impulsive effects on delayed fractional-order neural networks: sliding mode control-based fixed-time synchronization analysis, 2025, 0924-090X, 10.1007/s11071-025-10955-1
  • Reader Comments
  • © 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(1615) PDF downloads(96) Cited by(2)

Figures and Tables

Figures(2)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog