Research article

Synchronization control of time-delay neural networks via event-triggered non-fragile cost-guaranteed control


  • Received: 10 August 2022 Revised: 13 September 2022 Accepted: 22 September 2022 Published: 29 September 2022
  • This paper is devoted to event-triggered non-fragile cost-guaranteed synchronization control for time-delay neural networks. The switched event-triggered mechanism, which combines periodic sampling and continuous event triggering, is used in the feedback channel. A piecewise functional is first applied to fully utilize the information of the state and activation function. By employing the functional, various integral inequalities, and the free-weight matrix technique, a sufficient condition is established for exponential synchronization and cost-related performance. Then, a joint design of the needed non-fragile feedback gain and trigger matrix is derived by decoupling several nonlinear coupling terms. On the foundation of the joint design, an optimization scheme is given to acquire the minimum cost value while ensuring exponential stability of the synchronization-error system. Finally, a numerical example is used to illustrate the applicability of the present design scheme.

    Citation: Wenjing Wang, Jingjing Dong, Dong Xu, Zhilian Yan, Jianping Zhou. Synchronization control of time-delay neural networks via event-triggered non-fragile cost-guaranteed control[J]. Mathematical Biosciences and Engineering, 2023, 20(1): 52-75. doi: 10.3934/mbe.2023004

    Related Papers:

    [1] 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
    [2] Xinggui Zhao, Bo Meng, Zhen Wang . Event-triggered integral sliding mode control for uncertain networked linear control systems with quantization. Mathematical Biosciences and Engineering, 2023, 20(9): 16705-16724. doi: 10.3934/mbe.2023744
    [3] Na Zhang, Jianwei Xia, Tianjiao Liu, Chengyuan Yan, Xiao Wang . Dynamic event-triggered adaptive finite-time consensus control for multi-agent systems with time-varying actuator faults. Mathematical Biosciences and Engineering, 2023, 20(5): 7761-7783. doi: 10.3934/mbe.2023335
    [4] Guodong Zhao, Haitao Li, Ting Hou . Survey of semi-tensor product method in robustness analysis on finite systems. Mathematical Biosciences and Engineering, 2023, 20(6): 11464-11481. doi: 10.3934/mbe.2023508
    [5] 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
    [6] 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
    [7] Haitao Zhu, Xinrui Ji, Jianquan Lu . Impulsive strategies in nonlinear dynamical systems: A brief overview. Mathematical Biosciences and Engineering, 2023, 20(2): 4274-4321. doi: 10.3934/mbe.2023200
    [8] Siyu Li, Shu Li, Lei Liu . Fuzzy adaptive event-triggered distributed control for a class of nonlinear multi-agent systems. Mathematical Biosciences and Engineering, 2024, 21(1): 474-493. doi: 10.3934/mbe.2024021
    [9] Yilin Tu, Jin-E Zhang . Event-triggered impulsive control for input-to-state stability of nonlinear time-delay system with delayed impulse. Mathematical Biosciences and Engineering, 2025, 22(4): 876-896. doi: 10.3934/mbe.2025031
    [10] Bingrui Zhang, Jin-E Zhang . Fixed-deviation stabilization and synchronization for delayed fractional-order complex-valued neural networks. Mathematical Biosciences and Engineering, 2023, 20(6): 10244-10263. doi: 10.3934/mbe.2023449
  • This paper is devoted to event-triggered non-fragile cost-guaranteed synchronization control for time-delay neural networks. The switched event-triggered mechanism, which combines periodic sampling and continuous event triggering, is used in the feedback channel. A piecewise functional is first applied to fully utilize the information of the state and activation function. By employing the functional, various integral inequalities, and the free-weight matrix technique, a sufficient condition is established for exponential synchronization and cost-related performance. Then, a joint design of the needed non-fragile feedback gain and trigger matrix is derived by decoupling several nonlinear coupling terms. On the foundation of the joint design, an optimization scheme is given to acquire the minimum cost value while ensuring exponential stability of the synchronization-error system. Finally, a numerical example is used to illustrate the applicability of the present design scheme.



    Neural networks (NNs) are a kind of intricate nonlinear information processing systems with functional features such as self-learning, self-organizing, noise-resistance and distortion resistance, and thus can be utilized to solve problems that are difficult to handle by using traditional information processing approaches. Not surprisingly, NNs have been successfully applied across a variety of fields including speech synthesis [1], grammar learning [2], pattern classification [3] and time series prediction [4]. It is well known that time delays are almost unavoidable in the circuit implementation of NNs, and that are capable of causing oscillation of the NNs and even generating more complicated chaotic attractors. In view of this, the study of time-delay NNs (TDNNs) has drawn a great deal of attention from the academic communities of automation, mathematics and physics [5,6,7,8,9,10,11,12,13]. {Particularly, chaos synchronization of TDNNs has become one of the research hotspots due to its importance in nonlinear theory and strong application potential in many fields, including parameter identification, image encryption and secure communication [14,15,16].} The objective of chaos synchronization is to ensure that the state {trajectories} of the drive and response system {tend} to be consistent [17]. In order to achieve this objective, various control methodologies, including nonlinear feedback control [18], observer-based control [19], adaptive control [20] and sampled-data control [21,22], have been proposed in the literature.

    In the automation community, event-triggered control (ETC) has been increasingly recognized as an ideal control strategy as it can not only ensure the desired control performance but it can also mitigate the over-occupancy of communication channels in digital communication networks [23,24,25,26]. Within such a control strategy, the measurement signal is transmitted to the input of the controller only when pre-specified events occur. In this way, ETC can successfully reduce the update frequency of the controller, thereby saving computational and network resources. At present, there are several meaningful ETC mechanisms that have been put forward, including the continuous ETC mechanism [27], discrete ETC mechanism [28], self-triggered ETC mechanism [29], dynamic ETC mechanism [30], switched event-triggered control (SETC) mechanism [31] and so on. As shown in [31,32,33,34,35], the SETC mechanism ensures a positive minimum inter-event interval between any two adjacent events, thus avoiding the so-called Zeno phenomenon. In addition, it can markedly reduce the data transmission frequency while maintaining performance.

    Very recently, the chaos synchronization of TDNNs based on the SETC mechanism has been investigated by many researchers, and a few meaningful results have been proposed [36,37,38]. An implicit assumption in these reports is that there are no uncertainties of the controller gain. However, fluctuations of control gains may occur inevitably owing to component aging, round-off errors in calculations, and conversion between digital and analog [39,40,41,42]. Such fluctuations can result in performance degradation or even instability of the closed-loop system [43]. {In addition}, the control cost has not been considered in the above literature. However, while achieving the purpose of control, there may be certain requirements for the cost to be paid. Based on the above discussion, the non-fragile cost-guaranteed synchronization control (CGSC) for TDNNs under the SETC mechanism is a significant issue that deserves thorough investigation. However, to our knowledge, there are no relevant results so far and the topic remains open and challenging.

    Motivated by the above observations, the study was designed to explore the SETC mechanism-based non-fragile CGSC for TDNNs. The following are the main contributions of our work:

    1) A piecewise functional was developed by fully utilizing the information of the state and activation function of the switched synchronization-error system;

    2) A joint design of the needed non-fragile feedback gain and trigger matrix has been derived by eliminating nonlinear coupling terms;

    3) An optimization method was developed to acquire the minimum cost value while ensuring exponential stability by using the properties of matrix trace operations.

    The rest of this paper is organized as follows. The models of drive and response TDNNs and {some} necessary preliminaries are provided in Section 2. In Section 3, the non-fragile exponential synchronization and cost-related performance analysis under the SETC mechanism is discussed and the results of the joint design and its optimization are presented. In Section 4, a numerical simulation is described to verify the validity of the developed results. In the last section, we give the conclusion.

    Notations: Rl×n represents l×n-dimensional real matrices, Rl stands for the l-dimensional Euclidean space and represents the corresponding vector norm. For a square matrix W, W>0 means that W is positive semi-definite; I and 0 represent the identity matrix and the zero matrix with the proper dimensions, respectively; λmin(W) is used to represent the smallest eigenvalue of W. Furthermore, we use to denote a symmetry term in a matrix, and we denote a diagonal matrix by diag{}.

    Consider a TDNN described by:

    ˙x(t)=Ax(t)+Bf(x(t))+Bτf(x(tτ(t)))+I, (2.1a)
    y(t)=Cx(t), (2.1b)

    in which x(t)=[x1(t),,xp(t)]TRp denotes the neuron state, y(t)=[y1(t),,yq(t)]TRq represents the output vector, A Rp×p stands for the self-feedback matrix, BRp×p and BτRp×p are the connection weight matrices, CRq×p is a constant real matrix, IRp represents a constant input and τ(t) refers to the time-varying delays satisfying 0τ(t)τ and ˙τ(t)μ<1. f(x(t))=[f1(t),,fp(t)]TRp denotes the neuron activation function, satisfying the following hypothesis:

    Assumption 1. For any s1, s2 R,s1s2, there exists a positive matrix L=diag{L1,,Lp} such that

    0fi(s2)fi(s1)s2s1Li,i=1,,p. (2.2)

    Remark 1. A number of assumptions about the activation function have been proposed over the last several decades. Among these, Assumption 1 has been widely used in existing studies [22,44,45]. Many common functions satisfy such an assumption, such as the sigmoid function g(s)=(1+es)1, the piecewise linear function g(s)=0.5(|s+1||s1|), and tanh function g(s)=tanh(s).

    Remark 2. The TDNN in (2.1) covers many famous NN models, such as cellular NNs, BAM NNs and Hopfield NNs, as special cases. In addition, as shown in [46], the network model is capable of generating complex chaotic attractors, thus affording it with strong application potential in secure communication and image encryption.

    TDNN (2.1) is considered a drive system. We set the response system as

    ˙ˆx(t)=Aˆx(t)+Bf(ˆx(t))+Bτf(ˆx(tτ(t)))+u(t)+I, (2.3a)
    ˆy(t)=Cˆx(t), (2.3b)

    in which ˆx(t)=[ˆx1(t),,ˆxp(t)]TRp and ˆy(t)=[ˆy1(t),,ˆyq(t)]TRq are the state and the output of the response system, respectively. u(t)Rp denotes the system control input, which has the form

    u(t)=ˉK(ˆy(t)y(t)),

    where ˉK=K+ ΔK. KRp×q refers to the non-fragile controller gain to be designed and ΔK denotes the possible gain perturbation which is assumed to be of the form

    ΔK=GT(t)M, (2.4)

    where G and M are known constant matrices and T(t) is an uncertain parameter matrix satisfying TT(t)T(t)I. {Here, the gain perturbation is assumed to have the additive norm-bounded nonlinearity form. For multiplicative norm-bounded nonlinearity, one may refer to [11]}.

    Then, by defining synchronization errors ε(t)=ˆx(t)x(t) and ˉy(t)=ˆy(t)y(t), we can establish the error system as follows:

    ˙ε(t)=Aε(t)+Bˉf(ε(t))+Bτˉf(ε(tτ(t))+u(t), (2.5a)
    ˉy(t)=Cε(t), (2.5b)

    in which ˉf(ε(t))=f(ˆx(t))f(x(t)).

    In recent years, sampled-data control methods have been widely used in the literature [47,48,49,50,51]. Compared with sampled-data control methods, ETC methods can significantly reduce the update frequency of the controller, thereby saving computational and network resources. In the paper, the synchronization problem between drive-response TDNNs (2.1) and (2.3) is considered in the frame of SETC. The SETC mechanism adopted takes the form of

    αk+1=min{tαk+β(ˉy(t)ˉy(αk))TΓ(ˉy(t)ˉy(αk))ϵˉyT(t)Γˉy(t)}, (2.6)

    where αk represents the kth triggering instant, Γ0 is called the trigger matrix and ϵ0 and β>0 are given scalars with β denoting the sensor waiting interval. That is, β means the shortest time between two adjacent event-triggering instances. According to the SETC mechanism (2.6) a controller can be recast as a control input subject to sampling for t[αk,αk+β) and as a control input subject to continuous event triggering for t[αk+β,αk+1). Then, the control input takes the form of

    u(t)={ˉKCε(tϱ(t)),t[αk,αk+β),ˉK[e(t)+Cε(t)],t[αk+β,αk+1), (2.7)

    where

    ϱ(t)=tαkβ,t[αk,αk+β), (2.8)
    e(t)=ˉy(αk)ˉy(t),t[αk+β,αk+1). (2.9)

    Remark 3. The working mechanism of SETC is as follows. The sensor will wait β seconds after the k-th event is triggered at the instant αk. When the time comes to the instant αk+β, it will begin to monitor the event-trigger condition continuously to determine the next triggering instant. Namely, once the judgment condition is true, the new measurement will be sent to update the controller at the instant αk+1. Note that, in the case of ϵ=0, the SETC mechanism in (2.6) degenerates to the periodic sampled-data control mechanism.

    Hence, the error system can be constructed by using the system (2.5) and controller (2.7) as follows:

    ˙ε(t)=Aε(t)+Bˉf(ε(t))+Bτˉf(ε(tτ(t))ˉKCε(tϱ(t)),t[αk,αk+β), (2.10a)
    ˙ε(t)=(AˉKC)ε(t)ˉKe(t)+Bˉf(ε(t))+Bτˉf(ε(tτ(t)),t[αk+β,αk+1). (2.10b)

    To discuss the issue regarding the guaranteed cost index of the system in this paper, let us define

    J(t)=t0[εT(s)Qε(s)+uT(s)Ru(s)]ds, (2.11)

    as a quadratic cost function, where Q >0 and R >0.

    Next, we give a definition and some lemmas, which are required to gain our main results.

    Definition 1. Consider the error system (2.10); if there exist a positive number J and an event-triggered control law u(t) such that the error system is exponentially stable and the quadratic cost function (2.11) meets J()J, then the control law u(t) is called the cost-guaranteed controller and the upper-bound value J refers to the cost-related performance index.

    Lemma 1. [52] Let M>0 be an appropriate dimension matrix. Then, the inequality

    1μ2μ1[μ2μ1ϑ(s)ds]TM[μ2μ1ϑ(s)ds]μ2μ1ϑT(s)Mϑ(s)ds

    holds, where the scalars μ1 and μ2 satisfy μ2>μ1, and a vector function ϑ:[μ1,μ2]Rn.

    Lemma 2. [53] For any matrices M1Rn×m,M2Rn×m,M3=MT3>0,M3Rn×n, one can write

    MT1M2+MT2M1MT1M3M1+MT2M13M2.

    Lemma 3. [54] For any real matrices M1, M2, M3 with appropriate dimensions,

    [M1M2M3]<0

    holds if and only if

    M3<0  and  M1M2M13MT2<0.

    Before ending this section, let us clarify the purpose of this work, which was to design a SETC mechanism-based non-fragile cost-guaranteed controller to make sure that the error system in (2.10) is exponentially stable and the cost function in (2.11) satisfies J()J.

    In the following, a criterion for the non-fragile exponential stability and cost-related performance under the conditions of the SETC mechanism is given.

    Theorem 1. For the given scalars γ>0, β>0, ϵ0, assume that there exist p×p matrices P>0, N>0, S>0,W, W1, F1, F2, E1, E2, E3, Y, p×p diagonal matrices U>0, Λ1>0, Λ2>0, and a q×q matrix Γ0 such that

    ϝ>0,Θ0<0,Θ1<0,Φ<0, (3.1)

    in which

    ϝ=[P+βW+WT2βW+βW1βW1βWT1+βW+WT2], (3.2)
    Θ0=[Θ11WγΘ12+βW+WT2Θ13+ˆW1γΨ10FT1BτΘ22+βNΘ23β(WW1)Ψ20FT2BτΘ33ˆW2γ0002Λ100Ψ3LΛ22Λ2], (3.3)
    Θ1=[Θ11W+WT2Θ12Θ13+ˇW1γβET1Ψ10FT1BτΘ22Θ23βET2Ψ20FT2BτΘ33ˇW2γβET3000βe2γβN0002Λ100Ψ3LΛ22Λ2], (3.4)
    Φ=[Φ11Φ12FT1ˉK+CTˉKTRˉKΨ10FT1BτFT2F2FT2ˉKΨ20FT2BτΓ+ˉKTRˉK0002Λ100Ψ3LΛ22Λ2], (3.5)

    with

    Θ11=FT1AATF1ET1E1+2γP+2γUL+S+Q,Θ12=PFT1ATF2E2,Θ13=E3FT1KC+ET1,Θ22=F2FT2,Θ23=FT2ˉKC+ET2,Θ33=E3+ET3+CTˉKTRˉKC,Ψ1=FT1B+LΛ1,  Ψ2=FT2B+  U,  Ψ3=(μ1)e2γτS,U=diag{u1,u2,,up},Wγ=(1/2γβ)(W+WT),ˆW1γ=(12γβ)(WW1),ˇW1γ=WW1,ˆW2γ=(1/2γβ)(W+WT2W12WT1),ˇW2γ=(1/2)(W+WT2W12WT1),Φ11=FT1(AˉKC)+(AˉKC)TF1+ϵCTΓC+2γP+CTˉKTRˉKC+2γUL+S+Q,Φ12=P+(AˉKC)TF2FT1  .

    Then, the system (2.10) remains exponentially stable and the cost function in (2.11) satisfies J()εT(0)(P+UL)ε(0)+0τεT(s)Sε(s)ds.

    Proof. Let us choose a time-dependent Lyapunov functional as

    V(t)={V1(t),t[αk,αk+β),V2(t),t[αk+β,αk+1), (3.6)

    in which V1(t)=VP(t)+VU(t)+VS(t)+VN(t)+VW(t), V2(t)=VP(t)+VU(t) +VS(t) with

    VP(t)=εT(t)Pε(t),VU(t)=2ni=1uiεi(t)0ˉfi(s)ds,VS(t)=ttτ(t)e2γ(st)εT(s)Sε(s)ds,VN(t)=(βϱ(t))ttϱ(t)e2γ(st)˙εT(s)N˙ε(s)ds,VW(t)=(βϱ(t))ψT(t)[W+WT2W+W1W1WT1+W+WT2]ψ(t),

    where ψ(t)=col{ε(t),ε(tϱ(t))}. Clearly, V(t) is continuous on [0,+).

    Note that the linear matrix inequality (LMI) (3.2) can ensure that VP(t)+VW(t) is positive definite, as follows:

    VP(t)+VW(t)=εT(t)Pε(t)+(βϱ(t))ψT(t)[W+WT2W+W1W1WT1+W+WT2]ψ(t)=ψT(t)[P+(βϱ(t))W+WT2(βϱ(t))(W+W1)(βϱ(t))(W1WT1+W+WT2)]ψ(t)=(βϱ(t)β)ψT(t)ϝψ(t)+(ϱ(t)β)ψT(t)[P00]ψ(t). (3.7)

    Owing to ϝ>0 and P>0, it follows from (3.6) and (3.7) that

    min{λmin(P),λmin(ϝ)}ε(t)2V(t). (3.8)

    Taking the time derivative of the above functions along the trajectories of the system (2.10), it is not hard to derive that

    ˙VP(t)=2εT(t)P˙ε(t), (3.9)
    ˙VU(t)=2ˉfT(ε(t))U˙ε(t), (3.10)
    ˙VS(t)=2γVS(t)+εT(t)Sε(t)(1˙τ(t))×e2γτ(t)εT(tτ(t))Sε(tτ(t))2γVS(t)+εT(t)Sε(t)(1μ)×e2γτεT(tτ(t))Sε(tτ(t)) (3.11)
    ˙VN(t)=ttϱ(t)e2γ(st)˙εT(s)N˙ε(s)ds2γ(βϱ(t))ttϱ(t)e2γ(st)˙εT(s)N˙ε(s)ds+(βϱ(t))˙εT(t)N˙ε(t)ttϱ(t)e2γβ˙εT(s)N˙ε(s)ds2γ(βϱ(t))ttϱ(t)e2γ(st)˙εT(s)N˙ε(s)ds+(βϱ(t))˙εT(t)N˙ε(t), (3.12)
    ˙VW(t)=ψT(t)[W+WT2W+W1W1WT1+W+WT2]ψ(t)+(βϱ(t))[˙εT(t)(W+WT)ε(t)+2˙εT(t)(W+W1)ε(tϱ(t))]. (3.13)

    The proof can be divided into two cases in the light of the segmented time periods as follows.

    Case 1: With respect to t[αk,αk+β), the Lyapunov functional V1(t) can be adopted for the system (2.10a).

    Denote

    z1=1ϱ(t)ttϱ(t)˙ε(s)ds.

    Applying Lemma 1 to the first term of (3.12), it can be acquired that

    e2γβttϱ(t)˙εT(s)N˙ε(s)dsϱ(t)e2γβzT1Nz1. (3.14)

    For free-weighting matrices E1,E2,E3,F1 and F2 with proper dimensions, the following expressions hold:

    0=2[εT(t)ET1+˙εT(t)ET2+εT(tϱ(t))ET3][ϱ(t)z1+ε(tϱ(t))ε(t)], (3.15)
    0=2[εT(t)FT1+˙εT(t)FT2][Aε(t)ˉKCε(tϱ(t))+Bˉf(ε(t))+Bτˉf(ε(tτ(t))˙ε(t)]. (3.16)

    In view of Assumption 1, for any diagonal matrix Λ1>0 and Λ2>0, it follows that

    02ˉfT(ε(t))Λ1ˉf(ε(t))+2εT(t)LΛ1ˉf(ε(t)), (3.17)
    02ˉfT(ε(tτ(t)))Λ2ˉf(ε(tτ(t)))+2εT(tτ(t))LΛ2ˉf(ε(tτ(t))). (3.18)

    According to Assumption 1, we can also obtain

    0εi(t)0ˉfi(s)ds12ε2i(t)Li.

    Then, we have

    2ni=1ωiεi(t)0ˉfi(s)ds2(ω112ε21(t)L1++ωn12ε2n(t)Ln)=εT(t)ULε(t). (3.19)

    By summing up (3.9)–(3.12) and (3.15)–(3.19) and using (3.14), we find that

    ˙V1(t)+2γV1(t)+εT(t)Qε(t)+εT(tϱ(t))CTˉKTRˉKCε(tϱ(t))2εT(t)P˙ε(t)+2γεT(t)Pε(t)+2ˉfT(ε(t))U˙ε(t)+2γεT(t)ULε(t)+εT(t)Sε(t)(1μ)×e2γτεT(tτ(t))Sε(tτ(t))ϱ(t)e2γβzT1Nz1+(βϱ(t))˙εT(t)N˙ε(t)εT(t)(W+WT2)ε(t)εT(tϱ(t))(W+W1)Tε(t)εT(t)(W+W1)ε(tϱ(t))εT(tϱ(t))(W1WT1+W+WT2)ε(tϱ(t))+˙εT(t)(βϱ(t))(W+WT)ε(t)+2˙εT(t)(βϱ(t))(W+W1)ε(tϱ(t))+2γ(βϱ(t))ψT(t)[W+WT2W+W1W1WT1+W+WT2]ψ(t)+2(εT(t)ET1ϱ(t)z1+˙εT(t)ET2ϱ(t)z1+εT(tϱ(t))ET3ϱ(t)z1+εT(t)ET1ε(tϱ(t))+˙εT(t)ET2ε(tϱ(t))+εT(tϱ(t))ET3ε(tϱ(t))εT(t)ET1ε(t)˙εT(t)ET2ε(t)εT(tϱ(t))ET3ε(t))+2(εT(t)FT1Aε(t)˙εT(t)FT2Aε(t)εT(t)FT1ˉKCε(tϱ(t))˙εT(t)FT2ˉKCε(tϱ(t))+εT(t)FT1Bˉf(ε(t))+˙εT(t)FT2Bˉf(ε(t))+εT(t)FT1Bτˉf(ε(tτ(t))+˙εT(t)FT2Bτˉf(ε(tτ(t))εT(t)FT1˙ε(t)˙εT(t)FT2˙ε(t))2ˉfT(ε(t))Λ1ˉf(ε(t))+2εT(t)LΛ1ˉf(ε(t))2ˉfT(ε(tτ(t)))Λ2ˉf(ε(tτ(t)))+2εT(tτ(t))LΛ2ˉf(ε(tτ(t)))+εT(t)Qε(t)+εT(tϱ(t))CTˉKTRˉKCε(tϱ(t))=βϱ(t)βηT1(t)Θ0η1(t)+ϱ(t)βηT2(t)Θ1η2(t),

    where ηT1(t)={εT(t),˙εT(t),εT(tϱ(t)),ˉfT(ε(t)),εT(tτ(t)),ˉfT(ε(tτ(t)))} and ηT2(t)={εT(t),˙εT(t),εT(tϱ(t)),zT1,ˉfT(ε(t)),εT(tτ(t)),ˉfT(ε(tτ(t)))}. Then, the conditions Θ0<0 and Θ1<0 can ensure that

    ˙V1(t)+2γV1(t)εT(t)Qε(t)εT(tϱ(t))CTˉKTRˉKCε(tϱ(t))  0,t[αk,αk+β). (3.20)

    Case 2: With respect to t[αk+β,αk+1), we employ the Lyapunov functional V2(t) for the system (2.10b). Similar to (3.16), applying the free-weighting matrix approach, one can gain

    0=2[εT(t)FT1+˙εT(t)FT2][(AˉKC)ε(t)ˉKe(t)+Bˉf(ε(t))+Bτˉf(ε(tτ(t))˙ε(t)]. (3.21)

    In view of the event-triggering condition (2.6), one easily obtains that

    0eT(t)Γe(t)+ϵ[Cε(t)]TΓ[Cε(t)]. (3.22)

    In view of Assumption 1, for any diagonal matrix Λ1>0 and Λ2>0, it follows that

    02ˉfT(ε(t))Λ1ˉf(ε(t))+2εT(t)LΛ1ˉf(ε(t)) (3.23)
    02ˉfT(ε(tτ(t)))Λ2ˉf(ε(tτ(t)))+2εT(tτ(t))LΛ2ˉf(ε(tτ(t))) (3.24)

    By summing up (3.9), (3.10), (3.19) and (3.21)–(3.24), we find that

    ˙V2(t)+2γV2(t)+εT(t)Qε(t)+[ˉK(e(t)+Cε(t))]TR[ˉK(e(t)+Cε(t))]2εT(t)P˙ε(t)+2γεT(t)Pε(t)+2ˉfT(ε(t))U˙ε(t)+2γεT(t)ULε(t)+εT(t)Sε(t)(1μ)×e2γτεT(tτ(t))Sε(tτ(t))+2(εT(t)FT1(AˉKC)ε(t)+˙εT(t)FT2(AˉKC)ε(t)εT(t)FT1ˉKe(t)˙εT(t)FT2ˉKe(t)+εT(t)FT1Bˉf(ε(t))+˙εT(t)FT2Bˉf(ε(t))+εT(t)FT1Bτˉf(ε(tτ(t))+˙εT(t)FT2Bτˉf(ε(tτ(t))εT(t)FT1˙ε(t)˙εT(t)FT2˙ε(t))+ϵεT(t)CTΓCε(t)eT(t)Γe(t)2ˉfT(ε(t))Λ1ˉf(ε(t))+2εT(t)LΛ1ˉf(ε(t))2ˉfT(ε(tτ(t)))Λ2ˉf(ε(tτ(t)))+2εT(tτ(t))LΛ2ˉf(ε(tτ(t)))+εT(t)Qε(t)+[ˉK(e(t)+Cε(t))]TR[ˉK(e(t)+Cε(t))]=ηT3(t)Φη3(t),

    where ηT3(t)={εT(t),˙εT(t),eT(t),ˉfT(ε(t)),εT(tτ(t)),ˉfT(ε(tτ(t)))}. Similarly, the condition Φ <0 implies

    ˙V2(t)+2γV2(t)εT(t)Qε(t)[ˉK(e(t)+Cε(t))]TR[ˉK(e(t)+Cε(t))]  0, t[αk+β,αk+1). (3.25)

    From (3.20) and (3.25), it can be concluded that

    ˙V(t)+2γV(t)0

    for any t[αk,αk+1). Especially, for t[αk+β,αk+1), integrating over (3.25) with respect to t, one can get

    V(t)V(αk+β)e2γ(tαkβ)V(αk)e2γ(tαk)V(αk1+β)e2γ(tαk1β)V(αk1)e2γ(tαk1)V(0)e2γt.

    For t[αk,αk+β), by a similar procedure as above, we can derive the same result from the inequality (3.20). Therefore, we have

    V(t)V(0)e2γt. (3.26)

    From (3.8) and (3.26) we get

    ε(t)∥≤V(0)min{λmin(P),λmin(ϝ)}eγt.

    Thus, the system in Eq (2.10) is exponentially stable. In addition, since V(t)0 and γ>0, combining (3.20) with (3.25), we can write

    ˙V(t)εT(t)Qε(t)uT(t)Ru(t).

    It follows immediately that

    J()0˙V(t)dtV(0)εT(0)(P+UL)ε(0)+0τεT(s)Sε(s)ds. (3.27)

    This finishes the proof.

    Remark 4. The synchronization-error system composed of (2.10) and (2.7) is actually a switched system over the sampling interval under SETC mechanism (2.6). To match the switched system, a piecewise functional V(t) is constructed; it switches between the functions V1(t) and V2(t). Specifically, for the waiting time interval, the time-dependent functional V1(t) is applied, while for the continuous event detection interval, the time-independent functional V2(t) is employed. The function constructed not only fully utilizes the available state information, but also the activation function information on the interval [αk,αk+1].

    This subsection gives a joint design of the event-trigger matrix Γ in (2.6) and the desired non-fragile controller gain K for the system (2.10).

    Theorem 2. For the given scalars κ, γ>0, β>0, ϵ0, suppose that there exist p×p matrices P>0, N>0, S>0, W, W1, F1, E1, E2, E3, Y, p×p diagonal matrices U>0, Λ1>0, Λ2>0, a q×q matrix Γ0, and scalars α0>0 and α1>0 such that

    ϝ>0,Θ0<0,Θ1<0,Φ<0, (3.28)

    where ϝ is given in Theorem 1, and

    Θ0=[Ξ11Ξ12Ξ13Ψ10FT1Bτ0FT1GΞ22Ξ23˜Ψ20κFT1Bτ0κFT1GΞ33000CTYT02Λ10000Ψ3LΛ2002Λ200Ψ4Gα0I], (3.29)
    Θ1=[Π11Π12Π13βET1Ψ10FT1Bτ0FT1GΠ22Π23βET2˜Ψ20κFT1Bτ0κFT1GΠ33βET3000CTYT0βe2γβN000002Λ10000Ψ3LΛ2002Λ200Ψ4Gα0I], (3.30)
    Φ=[Σ11Σ12Σ13Ψ10FT1BτCTYTFT1GΣ22κY˜Ψ20κFT1Bτ0κFT1GΣ33000YT02Λ10000Ψ3LΛ2002Λ200Ψ4Gα1I], (3.31)

    with

    Ξ11=(1/2γβ)(W+WT)FT1AATF1ET1E1+2γP+2γUL+S+Q,Ξ12=βW+WT2+PFT1κATF1E2,Ξ13=(12γβ)(WW1)+ET1YCE3,Ξ22=κFT1κF1+βN,Ξ23=ET2κYCβ(WW1),Ξ33=E3+ET3+α0CTMTMC(1/2γβ)(W+WT2W12WT1),Π11=W+WT2FT1AATF1+2γUL+S+QET1E1+2γP,Π12=PFT1κATF1E2,Π13=ET1YCE3+WW1,Π22=κFT1κF1,Π23=ET2κYC,Π33=E3+ET3+α0CTMTMC(1/2)(W+WT2W12WT1),Σ11=FT1AATF1YCCTYT+ϵCTΓC+2γP+2γUL+S+Q+α1CTMTMC,Σ12=PκATF1κCTYTFT1,Σ13=Y+α1CTMTM,Σ22=κFT1κF1,Σ33=Γ+α1MTM,˜Ψ2=κFT1B+  U,Ψ4=ϑF1ϑFT1+ϑ2R,

    and the other notations are the same as those in Theorem 1. Then, the proposed controller (2.7) with the gain matrix

    K=(FT1)1Y (3.32)

    exponentially stabilizes the error system (2.10) under the conditions of the SETC mechanism, and the cost function in (2.11) satisfies J()εT(0)(P+UL)ε(0)+0τεT(s)Sε(s)ds.

    Proof. According to (2.4) and Lemma 3, (3.5) is equivalent to

    Φ=˘Φ+He(Υ1T(t)Υ2)

    where

    ˘Φ=[˘Σ11˘Σ12FT1KΨ10FT1BτCTKTFT2F2FT2KΨ20FT2Bτ0Γ000KT2Λ1000Ψ3LΛ202Λ20R1],

    with

    ˘Σ11=FT1(AKC)+(AKC)TF1+ϵCTΓC+2γP+2γUL+S+Q,˘Σ12=P+(AKC)TF2FT1,Υ1=[GTF1GTF20000GT]T,Υ2=[MC0M0000].

    By using Lemma 2 and (2.4), we have

    He(Υ1T(t)Υ2)1α1Υ1ΥT1+α1ΥT2Υ2.

    Hence, Φ<0 if the following inequality holds:

    ˘Φ+1α1Υ1ΥT1+α1ΥT2Υ2<0.

    Then, by premultiplying and postmultiplying diag{I,I,I,I,I,I,FT1} and its transpose on both sides of ˘Φ, and using Lemma 3 again, it follows that

    ˉΦ=[ˉΣ11˘Σ12ˉΣ13Ψ10FT1BτCTKTF1FT1GFT2F2FT2KΨ20FT2Bτ0FT2GΣ33000KTF102Λ10000Ψ3LΛ2002Λ200FT1R1F1Gα1I]<0, (3.33)

    where

    ˉΣ11=FT1(AKC)+(AKC)TF1+ϵCTΓC+2γP+2γUL+S+Q+α1CTMTMC,ˉΣ13=FT1K+α1CTMTM.

    For the item FT1R1F1 of the matrix ˉΦ, according to Lemma 2, the following inequality holds true:

    ϑF1+ϑFT1ϑ2R+FT1R1F1.

    Note that

    FT1R1F1ϑF1ϑFT1+ϑ2R,

    so we can re-express (3.33) as (3.31) by setting

    FT1K=Y,F2=κF1.

    Then, along similar lines as those for the above proof, (3.3) {and} (3.4) can be guaranteed by (3.29) and (3.30), respectively. The proof is completed.

    Remark 5. Through the use of matrix congruence transformation and a few inequality techniques, a method for the design of the non-fragile controller is developed in Theorem 2. It is shown that the needed gain matrix can be obtained by solving multiple LMIs that can be readily verified by utilizing the MATLAB software.

    When there is no gain perturbation, we can get the following corollary:

    Corollary 1. For the given scalars κ, γ>0, β>0, ϵ0, suppose that there exist p×p matrices P>0, N>0, S>0, W, W1, F1, E1, E2, E3, Y, p×p diagonal matrices U>0, Λ1>0 and Λ2>0, and a q×q matrix Γ0 such that

    ϝ>0,˜Θ0<0,˜Θ1<0,˜Φ<0, (3.34)

    where ϝ is given in Theorem 1, and

    ˜Θ0=[Ξ11Ξ12Ξ13Ψ10FT1Bτ0Ξ22Ξ23˜Ψ20κFT1Bτ0˜Ξ33000CTYT2Λ1000Ψ3LΛ202Λ20Ψ4],˜Θ1=[Π11Π12Π13βET1Ψ10FT1Bτ0Π22Π23βET2˜Ψ20κFT1Bτ0˜Π33βET3000CTYTβe2γβN00002Λ1000Ψ3LΛ202Λ20Ψ4],˜Φ=[˜Σ11Σ12YΨ10FT1BτCTYTΣ22κY˜Ψ20κFT1Bτ0Γ000YT2Λ1000Ψ3LΛ202Λ20Ψ4],

    with

    ˜Ξ33=E3+ET3(1/2γβ)(W+WT2W12WT1),˜Π33=E3+ET3    (1/2)(W+WT2W12WT1),˜Σ11=FT1AATF1YCCTYT+ϵCTΓC+2γP+2γUL+S+Q,

    and the other notations are the same as those in Theorem 2. Then, the proposed controller (2.7) with the gain matrix

    K=(FT1)1Y

    exponentially stabilizes the error system (2.10) under the conditions of the SETC mechanism, and the cost function in (2.11) satisfies J()εT(0)(P+UL)ε(0)+0τεT(s)Sε(s)ds.

    Theorem 2 presents an approach to design the non-fragile cost-guaranteed controller and the event-trigger matrix. Based on Theorem 2, we can give an optimization scheme to estimate the minimum cost value of J. That is to say, the following theorem will present an approach to select a controller that can ensure the minimum upper bound of the guaranteed cost control performance index.

    Before stating the Theorem 3, we define

    HHT=0τε(s)εT(s)ds. (3.35)

    Theorem 3. Consider the error system (2.10) with the quadratic cost function in (2.11) if the optimization issue

    minλ1εT(0)ε(0)+tr(Π)subject  to(i)  LMIs  in  (3.28),(ii)  P+UL<λ1I, (3.36)
    (iii)[ΠHTHD]<0 (3.37)

    is solvable, where tr(Π) denotes the trace of Π and D=S1. Then the proposed control law u(t) in (2.7) with the gain matrix (3.32) is an optimal event-triggered cost-guaranteed control law, and the quadratic cost function in (2.11) satisfies J()<λ1εT(0)ε(0)+tr(Π).

    Proof. By (3.36), it is not difficult to see that there exists λ1>0, such that

    εT(0)(P+UL)ε(0)λ1εT(0)ε(0).

    In addition, from (3.35) and (3.37), we can write

    0τεT(s)Sε(s)ds=0τtr(εT(s)Sε(s))ds=tr(HHTS)=tr(HTD1H)<tr(Π).

    Therefore, it follows from (3.27) that J()<λ1εT(0)ε(0)+tr(Π). Thus, the minimization of λ1εT(0)ε(0)+tr(Π) implies the minimization of the cost-related performance index in Eq (3.27). The proof is complete.

    When there is no gain perturbation, we can readily write the following result:

    Corollary 2. Consider the error system (2.10) with the quadratic cost function in (2.11) if the optimization issue

    minλ2εT(0)ε(0)+tr(Π)subject  to(i)  LMIs  in  (3.34),(ii)  P+UL<λ2I,(iii)[ΠHTHD]<0

    is solvable, where tr(Π) denotes the trace of Π and D=S1. Then the proposed control law u(t) in (2.7) with gain matrix (3.32) is an optimal event-triggered cost-guaranteed control law, and the quadratic cost function in (2.11) satisfies J()<λ2εT(0)ε(0)+tr(Π).

    Consider TDNNs (2.1) and (2.3) with

    A=[1001],B=[20.153],Bτ=[1.50.10.22.5],C=[1100.5].

    The activation function is chosen as f(x(t))=[tanh(x1(t),tanh(x2(t)]T. Note that f(x(t)) satisfies Assumption 1 with L=diag{1,1}. And the time delay is taken as τ(t)=1+0.2sin(2t), which implies μ=0.4 and τ=1.2.

    In the simulation, the initial conditions of the drive and response systems were set to be x(s)=[0.27 0.24]T and ˆx(s)=[0.42 0.24]T, respectively, where s[τ,0]. The chaotic attractor of the drive TDNN is depicted in Figure 1. Simultaneously, from (3.35), we have

    H=[0.76000].
    Figure 1.  Chaotic behaviors of the drive TDNN.

    The two parameter matrices in the quadratic cost function (2.11) were chosen as

    Q=[1001],R=[0.1000.1].

    Given β=0.02 and ϵ=0.2, the following SETC mechanism can be obtained:

    αk+1=min{tαk+0.02(ˉy(t)ˉy(αk))TΓ(ˉy(t)ˉy(αk))0.2ˉyT(t)Γˉy(t)}.

    Next, let us show the applicability of the present non-fragile control approaches. The parameter matrices for the gain perturbation ΔK in (2.4) were set to be

    G=g[0.5000.5],M=[1001],T(t)=0.8|sin(t)|.

    We choose γ=0.01, ϑ=19.2, and κ=0.1. Then, for different values of the parameter g, by using Theorem 3 and Corollary 2, the controller gain K, triggered matrix Γ, and optimal cost-related performance index J were obtained as shown Table 1. From the table we can find that J will increase as g increases. Based on the above parameters, by utilizing MATLAB, the triggered moments and the corresponding sampled intervals were obtained as shown in Figure 2. The trajectories of the error system (2.10) and input signals are shown in Figures 3 and 4, respectively. Obviously, the trajectories of the state and input quickly converged to 0. Figure 5 shows the trajectory of J(t). It was found that J(t) converged to 0.5497 (i.e., J()=0.5497), which is less than J=1.6237. Thus, the simulations verify the present theoretical results.

    Table 1.  Optimal cost-related performance J for different values of g.
    Conditions Theorem 3 Corollary 2
    g 0.5 0.3 0.1 0
    K [12.092820.34182.965320.7362] [11.917620.10742.934720.5661] [11.751719.88432.906720.3973] [11.672119.77692.893620.3135]
    Γ [48.183249.920849.9208224.3081] [45.491447.578847.5788216.3803] [43.050445.446945.4469208.9249] [41.914344.451144.4511205.3583]
    J 1.7812 1.6990 1.6237 1.5883

     | Show Table
    DownLoad: CSV
    Figure 2.  Triggering moments and sampling intervals.
    Figure 3.  State response of the error system.
    Figure 4.  Control input signals.
    Figure 5.  J(t) along the system (2.10).

    The non-fragile event-triggered CGSC of TDNNs has been investigated. From the point of view of saving computational and network resources, a SETC mechanism has been introduced to determine the event-trigging moments. A piecewise functional V(t) has been constructed to make efficient use of the sampling intervals [αk,αk+1) and activation function ˉf(ε(t)). With the help of the constructed functional and several inequalities, a criterion (see Theorem 1) has been derived to ensure the exponential stability and the cost-related performance of the synchronization-error system. Based on the criterion, a novel joint design of the designed control gain and trigger matrix has been developed (i.e., Theorem 2), and an optimization scheme for the minimum cost-related performance index has been given (i.e., Theorem 3). The validity and practicability of the obtained results have been illustrated through a numerical example. Future research will be focused on the event-triggered synchronization control of TDNNs under cyber attack.

    This work was supported by the Key Research and Development Project of Anhui Province (Grant No. 202004a07020028).

    The authors declare that there is no conflict of interest.



    [1] W. Ren, R. W. Beard, Speech synthesis from neural decoding of spoken sentences, Nature, 568 (2019), 493–498. https://doi.org/10.1038/s41586-019-1119-1 doi: 10.1038/s41586-019-1119-1
    [2] F. A. Gers, E. Schmidhuber, LSTM recurrent networks learn simple context-free and context-sensitive languages, {IEEE Trans. Neural Netw., 12 (2001), 1333–1340. https://doi.org/10.1109/72.963769
    [3] R. K. Brouwer, Growing of a fuzzy recurrent artificial neural network (FRANN) for pattern classification, Int. J. Neural Syst., 9 (1999), 335–350. https://doi.org/10.1142/S0129065799000320 doi: 10.1142/S0129065799000320
    [4] H. Hewamalage, C. Bergmeir, K. Bandara, Recurrent neural networks for time series forecasting: Current status and future directions, Int. J. Forecasting, 37 (2021), 388–427. https://doi.org/10.1016/j.ijforecast.2020.06.008 doi: 10.1016/j.ijforecast.2020.06.008
    [5] Y. Cao, R. Sriraman, N. Shyamsundarraj, R. Samidurai, Robust stability of uncertain stochastic complex-valued neural networks with additive time-varying delays, Math. Comput. Simul., 171 (2020), 207–220. https://doi.org/10.1016/j.matcom.2019.05.011 doi: 10.1016/j.matcom.2019.05.011
    [6] T. H. Lee, H. M. Trinh, J. H. Park, Stability analysis of neural networks with time-varying delay by constructing novel Lyapunov functionals, IEEE Trans. Neural Netw. Learn. Syst., 29 (2018), 4238–4247. https://doi.org/10.1109/TNNLS.2017.2760979 doi: 10.1109/TNNLS.2017.2760979
    [7] S. Arik, New criteria for stability of neutral-type neural networks with multiple time delays, IEEE Trans. Neural Netw. Learn. Syst., 31 (2020), 1504–1513. https://doi.org/10.1109/TNNLS.2019.2920672
    [8] N. M. Thoiyab, P. Muruganantham, Q. Zhu, N. Gunasekaran, Novel results on global stability analysis for multiple time-delayed BAM neural networks under parameter uncertainties, Chaos Solitons Fractals, 152 (2021), 111441. https://doi.org/10.1016/j.chaos.2021.111441 doi: 10.1016/j.chaos.2021.111441
    [9] L. Fan, Q. Zhu, Mean square exponential stability of discrete-time Markov switched stochastic neural networks with partially unstable subsystems and mixed delays, Inf. Sci., 580 (2021), 243–259. https://doi.org/10.1016/j.ins.2021.08.068 doi: 10.1016/j.ins.2021.08.068
    [10] Q. Wu, Z. Yao, Z. Yin, H. Zhang, Fin-TS and Fix-TS on fractional quaternion delayed neural networks with uncertainty via establishing a new Caputo derivative inequality approach, Math. Biosci. Eng., 19 (2022), 9220–9243. https://doi.org/10.3934/mbe.2022428 doi: 10.3934/mbe.2022428
    [11] W. Tai, D. Zuo, Z. Xuan, J. Zhou, Z. Wang, Non-fragile L2L filtering for a class of switched neural networks, Math. Comput. Simul., 185 (2021), 629–645. https://doi.org/10.1016/j.matcom.2021.01.014 doi: 10.1016/j.matcom.2021.01.014
    [12] J. Cheng, L. Liang, H. Yan, J. Cao, S. Tang, K. Shi, Proportional-integral observer-based state estimation for Markov memristive neural networks with sensor saturations, IEEE Trans. Neural Netw. Learn. Syst., 2022. https://doi.org/10.1109/TNNLS.2022.3174880
    [13] N. Gunasekaran, M. S. Ali, Design of stochastic passivity and passification for delayed BAM neural networks with Markov jump parameters via non-uniform sampled-data control, Neural Process. Lett., 53 (2021), 391–404. https://doi.org/10.1007/s11063-020-10394-6 doi: 10.1007/s11063-020-10394-6
    [14] A. Abdurahman, H. Jiang, C. Hu, Z. Teng, Parameter identification based on finite-time synchronization for Cohen–Grossberg neural networks with time-varying delays, Nonlinear Anal., Model. Control, 20 (2015), 348–366. https://doi.org/10.15388/NA.2015.3.3 doi: 10.15388/NA.2015.3.3
    [15] M. Kalpana, K. Ratnavelu, P. Balasubramaniam, M. Z. M. Kamali, Synchronization of chaotic-type delayed neural networks and its application, Nonlinear Dyn., 93 (2018), 543–555. https://doi.org/10.1007/s11071-018-4208-z doi: 10.1007/s11071-018-4208-z
    [16] A. M. Alimi, C. Aouiti, E. A. Assali, M. Z. M. Kamali, Finite-time and fixed-time synchronization of a class of inertial neural networks with multi-proportional delays and its application to secure communication, Neurocomputing, 332 (2019), 29–43. https://doi.org/10.1016/j.neucom.2018.11.020 doi: 10.1016/j.neucom.2018.11.020
    [17] T. H. Lee, J. H. Park, Improved criteria for sampled-data synchronization of chaotic {Lur'e} systems using two new approaches, Nonlin. Anal. Hybrid Syst., 24 (2017), 132–145. https://doi.org/10.1016/j.nahs.2016.11.006 doi: 10.1016/j.nahs.2016.11.006
    [18] L. Wang, Y. Shen, G. Zhang, Synchronization of a class of switched neural networks with time-varying delays via nonlinear feedback control, IEEE Trans. Cybern., 46 (2016), 2300–2310. https://doi.org/10.1109/TCYB.2015.2475277
    [19] Y. Zhao, X. Li, P. Duan, Observer-based sliding mode control for synchronization of delayed chaotic neural networks with unknown disturbance, Neural Networks, 117 (2019), 268–273. https://doi.org/10.1016/j.neunet.2019.05.013 doi: 10.1016/j.neunet.2019.05.013
    [20] D. Tong, L. Zhang, W. Zhou, J. Zhou, Y. Xu, Asymptotical synchronization for delayed stochastic neural networks with uncertainty via adaptive control, Int. J. Control Autom. Syst., 14 (2016), 706–712. https://doi.org/10.1007/s12555-015-0077-0 doi: 10.1007/s12555-015-0077-0
    [21] H. Li, C. Li, D. Ouyang, S. K. Nguang, Impulsive synchronization of unbounded delayed inertial neural networks with actuator saturation and sampled-data control and its application to image encryption, IEEE Trans. Neural Networks Learn. Syst., 32 (2021), 1460–1473. https://doi.org/10.1109/TNNLS.2020.2984770 doi: 10.1109/TNNLS.2020.2984770
    [22] G. Chen, J. Xia, J. H. Park, H. Shen, G. Zhuang, Sampled-data synchronization of stochastic Markovian jump neural networks with time-varying delay, IEEE Trans. Neural Networks Learn. Syst., 33 (2022), 3829–3841. https://doi.org/10.1109/TNNLS.2021.3054615 doi: 10.1109/TNNLS.2021.3054615
    [23] M. Dlala and S. O. Alrashidi, Rapid exponential stabilization of Lotka-McKendrick's equation via event-triggered impulsive control, Math. Biosci. Eng., 18 (2021), 9121–9131. https://doi.org/10.3934/mbe.2021449 doi: 10.3934/mbe.2021449
    [24] Z. Gu, S. Yan, J. H. Park, X. Xie, Event-triggered synchronization of chaotic Lur'e systems via memory-based triggering approach, IEEE Trans. Circuits Syst. II Express Briefs, 69 (2022), 1427–1431. https://doi.org/10.1109/TCSII.2021.3113955 doi: 10.1109/TCSII.2021.3113955
    [25] J. Wu, S. Qiu, M. Liu, H. Li, Y. Liu, Finite-time velocity-free relative position coordinated control of spacecraft formation with dynamic event triggered transmission, Math. Biosci. Eng., 19 (2022), 6883–6906. https://doi.org/10.3934/mbe.2022324 doi: 10.3934/mbe.2022324
    [26] J. Cheng, J. H. Park, Z. Wu, Observer-based asynchronous control of nonlinear systems with dynamic event-based try-once-discard protocol, IEEE Trans. Cybern., 2021. https://doi.org/10.1109/TCYB.2021.3104806
    [27] J. Lunze, D. Lehmann, A state-feedback approach to event-based control, Automatica, 46 (2010), 211–215. https://doi.org/10.1016/j.automatica.2009.10.035 doi: 10.1016/j.automatica.2009.10.035
    [28] D. Yue, E. Tian, Q. Han, A delay system method for designing event-triggered controllers of networked control systems, IEEE Trans. Autom. Control, 58 (2012), 475–481. https://doi.org/10.1109/TAC.2012.2206694
    [29] M. Xing, F. Deng, X. Zhao, Synchronization of stochastic complex dynamical networks under self-triggered control, Int. J. Robust Nonlinear Control, 27 (2017), 2861–2878. https://doi.org/10.1002/rnc.3716 doi: 10.1002/rnc.3716
    [30] J. Zhang, E. Fridman, Dynamic event-triggered control of networked stochastic systems with scheduling protocols, IEEE Trans. Autom. Control, 66 (2021), 6139–6147. https://doi.org/10.1109/TAC.2021.3061668 doi: 10.1109/TAC.2021.3061668
    [31] A. Selivanov, E. Fridman, Event-triggered H control: A switching approach, IEEE Trans. Autom. Control, 61 (2015), 3221–3226. https://doi.org/10.1109/TAC.2015.2508286 doi: 10.1109/TAC.2015.2508286
    [32] Z. Yan, X. Huang, J. Cao, Variable-sampling-period dependent global stabilization of delayed memristive neural networks based on refined switching event-triggered control, Sci. China Inf. Sci., 63 (2020), 212201. https://doi.org/10.1007/s11432-019-2664-7 doi: 10.1007/s11432-019-2664-7
    [33] S. Ding, X. Xie, Y. Liu, Event-triggered static/dynamic feedback control for discrete-time linear systems, Inf. Sci., 524 (2020), 33–45. https://doi.org/10.1016/j.ins.2020.03.044 doi: 10.1016/j.ins.2020.03.044
    [34] W. Wu, L. He, J. Zhou, Z. Xuan, S. Arik, Disturbance-term-based switching event-triggered synchronization control of chaotic Lurie systems subject to a joint performance guarantee, Commun. Nonlinear Sci. Numer. Simul., 115 (2022), 106774. https://doi.org/10.1016/j.cnsns.2022.106774 doi: 10.1016/j.cnsns.2022.106774
    [35] W. Wang, D. Xu, J. Zhou, Z. Yan, Cost-guaranteed exponential stabilization of Lurie systems via switched event-triggered control, Discrete Cont. Dyn. B, 2022. https://doi.org/10.3934/dcdsb.2022194
    [36] Y. Zhou, H. Zhang, Z. Zeng, Quasisynchronization of memristive neural networks with communication delays via event-triggered impulsive control, IEEE Trans. Cybern., 52 (2022), 7682–7693. https://doi.org/10.1109/TCYB.2020.3035358 doi: 10.1109/TCYB.2020.3035358
    [37] Z. Yan, X. Huang, Y. Fan, J. Xia, H. Shen, Threshold-function-dependent quasi-synchronization of delayed memristive neural networks via hybrid event-triggered control, IEEE Trans. Syst., Man, Cybern., 51 (2021), 6712–6722. https://doi.org/10.1109/TSMC.2020.2964605 doi: 10.1109/TSMC.2020.2964605
    [38] S. Yan, S. K. Nguang, Z. Gu, H weighted integral event-triggered synchronization of neural networks with mixed delays, IEEE Trans. Ind. Informat., 17 (2021), 2365–2375. https://doi.org/10.1109/TII.2020.3004461 doi: 10.1109/TII.2020.3004461
    [39] X. Chang, Y. Liu, M. Shen, Resilient control design for lateral motion regulation of intelligent vehicle, IEEE ASME Trans. Mechatronics, 24 (2019), 2488–2497. https://doi.org/10.1109/TMECH.2019.2946895 doi: 10.1109/TMECH.2019.2946895
    [40] J. Zhou, Y. Liu, J. Xia, Z. Wang, S. Arik, Resilient fault-tolerant anti-synchronization for stochastic delayed reaction–diffusion neural networks with semi-Markov jump parameters, Neural Networks, 125 (2020), 194–204. https://doi.org/10.1016/j.neunet.2020.02.015 doi: 10.1016/j.neunet.2020.02.015
    [41] R. Sakthivel, C. Wang, S. Santra, B. Kaviarasan, Non-fragile reliable sampled-data controller for nonlinear switched time-varying systems, Nonlin. Anal. Hybrid Syst., 27 (2018), 62–76. https://doi.org/10.1016/j.nahs.2017.08.005 doi: 10.1016/j.nahs.2017.08.005
    [42] J. H. Park, H. Shen, X. Chang, T. H. Lee, Fuzzy resilient energy-to-peak filter design for continuous-time nonlinear systems, in Recent Advances in Control and Filtering of Dynamic Systems with Constrained Signals, Cham: Springer, (2019), 119–139. https://doi.org/10.1007/978-3-319-96202-3
    [43] X. Chang, Takagi-Sugeno Fuzzy Systems Non-fragile H-infinity Filtering, Berlin: Springer-Verlag, 2012. https://doi.org/10.1007/978-3-642-28632-2
    [44] L. Chen, Y. Chen, N. Zhang, Synchronization control for chaotic neural networks with mixed delays under input saturations, Neural Process. Lett., 53 (2021), 3735–3755. https://doi.org/10.1007/s11063-021-10577-9 doi: 10.1007/s11063-021-10577-9
    [45] L. He, W. Wu, Q. Zhu, G. Yao, J. Zhou, Input-to-state stabilization of delayed semi-Markovian jump neural networks via sampled-data control, Neural Process. Lett., 2022. https://doi.org/10.1007/s11063-022-11008-z
    [46] H. Lu, Chaotic attractors in delayed neural networks, Phys. Lett. A, 298 (2002), 109–116. https://doi.org/10.1016/S0375-9601(02)00538-8 doi: 10.1016/S0375-9601(02)00538-8
    [47] S. Santra, R. Sakthivel, K. Mathiyalagan, A. S. Marshal, Exponential passivity results for singular networked cascade control systems via sampled-data control, J. Dyn. Syst., Means. Control, 139 (2017). https://doi.org/10.1115/1.4034781
    [48] M. S. Ali, N. Gunasekaran, J. Cao, Sampled-data state estimation for neural networks with additive time-varying delays, Acta Math. Sci., 39 (2019), 195–213. https://doi.org/10.1007/s10473-019-0116-7 doi: 10.1007/s10473-019-0116-7
    [49] N. Gunasekaran, G. Zhai, Q. Yu, Sampled-data synchronization of delayed multi-agent networks and its application to coupled circuit, Neurocomputing, 413 (2020), 499–511. https://doi.org/10.1016/j.neucom.2020.05.060 doi: 10.1016/j.neucom.2020.05.060
    [50] S. Santra, M. Joby, M. Sathishkumar, S. M. Anthoni, LMI approach-based sampled-data control for uncertain systems with actuator saturation: application to multi-machine power system, Nonlinear Dyn., 107 (2022), 967–982. https://doi.org/10.1007/s11071-021-06995-y doi: 10.1007/s11071-021-06995-y
    [51] N. Gunasekaran, M. S. Ali, S. Arik, H. A. Ghaffar, A. A. Z. Diab, Finite-time and sampled-data synchronization of complex dynamical networks subject to average dwell-time switching signal, Neural Networks, 149 (2022), 137–145. https://doi.org/10.1016/j.neunet.2022.02.013 doi: 10.1016/j.neunet.2022.02.013
    [52] K. Gu, J. Chen, V. L. Kharitonov, Stability of Time-Delay Systems, Boston, MA: Birkhauser, 2003. https://doi.org/10.1007/978-1-4612-0039-0
    [53] K. Zhou and P. P. Khargonekar, Robust stabilization of linear systems with norm-bounded time-varying uncertainty, Syst. Contr. Lett., 10 (1988), 17–20. https://doi.org/10.1016/0167-6911(88)90034-5 doi: 10.1016/0167-6911(88)90034-5
    [54] S. Boyd, L. E. Ghaoui, E. Feron, V. Balakrishnan, Linear Matrix Inequalities in System and Control Theory, Philadelphia, PA: SIAM, 1994. https://doi.org/10.1137/1.9781611970777
  • This article has been cited by:

    1. Tianqi Yu, Lei Liu, Yan-Jun Liu, Observer-based adaptive fuzzy output feedback control for functional constraint systems with dead-zone input, 2022, 20, 1551-0018, 2628, 10.3934/mbe.2023123
    2. Yuxiang Ji, Yu Zhang, Xiaofeng Ma, Xinling Li, Taiping Jiang, Resilient Energy-to-peak Event-triggered Control of NCSs, 2023, 2638, 1742-6588, 012006, 10.1088/1742-6596/2638/1/012006
    3. Huiyu Wang, Shutang Liu, Xiang Wu, Jie Sun, Wei Qiao, Synchronization of Fractional Delayed Memristive Neural Networks with Jump Mismatches via Event-Based Hybrid Impulsive Controller, 2024, 8, 2504-3110, 297, 10.3390/fractalfract8050297
  • 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(2135) PDF downloads(161) Cited by(3)

Figures and Tables

Figures(5)  /  Tables(1)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog