
This paper investigates the finite time synchronization (Fin-TS) and fixed time synchronization (Fix-TS) issues on Caputo quaternion delayed neural networks (QDNNs) with uncertainty. A new Caputo fractional differential inequality is constructed, then Fix-TS settling time of the positive definite function is estimated, which is very convenient to derive Fix-TS condition to Caputo QDNNs. By designing the appropriate self feedback and adaptive controllers, the algebraic discriminant conditions to achieve Fin-TS and Fix-TS on Caputo QDNNs are proposed based on quaternion direct method, Lyapunov stability theory, extended Cauchy Schwartz inequality, Jensen inequality. Finally, the correctness and validity of the presented results under the different orders are verified by two numerical examples.
Citation: 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[J]. Mathematical Biosciences and Engineering, 2022, 19(9): 9220-9243. doi: 10.3934/mbe.2022428
[1] | Haiyan Song, Cuihong Liu, Shengnan Li, Peixiao Zhang . TS-GCN: A novel tumor segmentation method integrating transformer and GCN. Mathematical Biosciences and Engineering, 2023, 20(10): 18173-18190. doi: 10.3934/mbe.2023807 |
[2] | Zia Ud Din, Amir Ali, Zareen A. Khan, Gul Zaman . Heat transfer analysis: convective-radiative moving exponential porous fins with internal heat generation. Mathematical Biosciences and Engineering, 2022, 19(11): 11491-11511. doi: 10.3934/mbe.2022535 |
[3] | Debao Yan . Existence results of fractional differential equations with nonlocal double-integral boundary conditions. Mathematical Biosciences and Engineering, 2023, 20(3): 4437-4454. doi: 10.3934/mbe.2023206 |
[4] | Noura Laksaci, Ahmed Boudaoui, Seham Mahyoub Al-Mekhlafi, Abdon Atangana . Mathematical analysis and numerical simulation for fractal-fractional cancer model. Mathematical Biosciences and Engineering, 2023, 20(10): 18083-18103. doi: 10.3934/mbe.2023803 |
[5] | 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 |
[6] | 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 |
[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] | Adnan Sami, Amir Ali, Ramsha Shafqat, Nuttapol Pakkaranang, Mati ur Rahmamn . Analysis of food chain mathematical model under fractal fractional Caputo derivative. Mathematical Biosciences and Engineering, 2023, 20(2): 2094-2109. doi: 10.3934/mbe.2023097 |
[9] | Luís P. Castro, Anabela S. Silva . On the solution and Ulam-Hyers-Rassias stability of a Caputo fractional boundary value problem. Mathematical Biosciences and Engineering, 2022, 19(11): 10809-10825. doi: 10.3934/mbe.2022505 |
[10] | M. Botros, E. A. A. Ziada, I. L. EL-Kalla . Semi-analytic solutions of nonlinear multidimensional fractional differential equations. Mathematical Biosciences and Engineering, 2022, 19(12): 13306-13320. doi: 10.3934/mbe.2022623 |
This paper investigates the finite time synchronization (Fin-TS) and fixed time synchronization (Fix-TS) issues on Caputo quaternion delayed neural networks (QDNNs) with uncertainty. A new Caputo fractional differential inequality is constructed, then Fix-TS settling time of the positive definite function is estimated, which is very convenient to derive Fix-TS condition to Caputo QDNNs. By designing the appropriate self feedback and adaptive controllers, the algebraic discriminant conditions to achieve Fin-TS and Fix-TS on Caputo QDNNs are proposed based on quaternion direct method, Lyapunov stability theory, extended Cauchy Schwartz inequality, Jensen inequality. Finally, the correctness and validity of the presented results under the different orders are verified by two numerical examples.
Neural network dynamics have attracted much attention owing to the wide applications in associative memory [1], signal processing [2], intelligent control [3], and so on. The mainly dynamical behaviours in neural network systems include stability [4,5,6,7,8,9,10], synchronization [11,12], bifurcation [13], chaos [14] and reaction-diffusion[15], which reflect the characteristics of network systems. Fractional calculus has good free space and infinite storage space in many fields, such as artificial neural networks [16], genetic algorithm [17] and image enhancement [18].
Time delay is an interesting phenomenon in network systems, resulting from differences in the efficiency of signal transmission from the point of generation to the point of reception during network transmission, as well as interference from other external factors. Recently, the results with regard to time delay phenomena are extensively reported including leakage time delay[19], discrete time delay[20], distributed time delay[21] and so on. In actual industrial operation, uncertainties in the parameters associated with the model coefficients can arise due to operational errors, incomplete considerations and model construction. Therefore, the models with uncertain parameters will further improve the robustness of systems in [22,23,24].
Synchronization, as one of the typical dynamical behaviours, such as collective recitations, military parades and resonance phenomena, and has become a hot topic [15,22,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43] including Fin-TS [22,26,27,28,29], Fix-TS [22,25,26,30], exponential synchronization[31,32,33,34,35], Mittag-Leffler synchronization[36,37,38], projective synchronization[36,39], quasi-synchronization[40,41], global synchronization[42]. Lin et al. [15] explored the synchronization of directly coupled QDNNs based on pinning control. Li et al. [22] investigated Fin-TS and Fix-TS of Caputo QVNNs by employing feedback and adaptive controllers. Wei et al. [25] studied Fix-TS of QNNs by a pure power-law control design. Kashkynbayev et al. [26] discussed Fin-TS and Fix-TS of DNNs by designing proper control functions. Hu et al.[43] used an adaptive distributed approach at the edge to study the synchronization of spatio-temporal networks. In fact, Fin-TS and Fix-TS are more relevant to realistic control requirements. However, very few research results on Fin-TS and Fix-TS for fractional order NNs are reported except [44,45], where the controller designs for Fix-TS are denoted in fractional order form, then this will undoubtedly increase the difficulty to explore the dynamics of systems.
Inspired by these works, this paper focuses on Fin-TS and Fix-TS issues on Caputo QDNNs with uncertainty. Different from the decomposition method [23,25,35,46], Laplace transform [40,47,52], comparison principle [23,48,49] and linear matrix inequality (LMI) [30,41,50], we apply the quaternion direct method, the Lyapunov stability method and the quaternion inequality method to establish the synchronization criterion for Caputo QDNNs. It is worth noting that a new Caputo differential inequality is constructed, then the fixed time of the positive definite function is estimated. The propoesd method can greatly simplify the complexity of the calculation and solve the complexity of controller design caused by the immaturity of the fractional fixed time stability theory, which is very convenient to derive Fix-TS condition to Caputo QDNNs.
The main innovations and highlights of this paper are presented below.
∙ The model includes the discrete and distributed delays, uncertain term and Caputo derivative operator.
∙ A new Caputo differential inequality is constructed, then the fixed time of the positive definite function is estimated.
∙ Applying quaternion direct method rather than the decomposition method, the algebraic discriminant conditions to achieve Fin-TS and Fix-TS on Caputo QDNNs are proposed.
∙ The control strategies to design the appropriate self feedback and adaptive controllers can effectively reduce the consumption of control costs caused by the time delay term.
In this section, the basic definitions, related lemmas and model introduction are described.
Definition 2.1. Fractional integral of the order μ of k(t) is defined as [51]
t0D−μtk(t)=1Γ(μ)∫tt0(t−l)μ−1k(l)dl. |
Definition 2.2. The Caputo-type fractional-order derivative of the function k(t) is defined as [51]
Ct0Dμtk(t)=1Γ(n−μ)∫tt0(t−l)n−μ−1k(n)(l)dl, |
where t>t0, n=[μ]+1.
Lemma 2.1. If 0<μ<1, α∈R, then [49]
Ct0Dμtkα(t)=Γ(1+α)Γ(1+α−μ)kα−μ(t)Ct0Dμtk(t). |
Lemma 2.2. Let k(t) be a differentiable function over the quaternion field Q, then [22]
Ct0Dμt[¯k(t)k(t)]⩽¯k(t)Ct0Dμtk(t)+[Ct0Dμt¯k(t)]k(t), 0<μ<1. |
Lemma 2.3. Let χ and φ be any constant in the quaternion field Q, for any positive number θ in the real number field R, then [44]
¯χφ+¯φχ⩽θ¯χχ+θ−1¯φφ. |
Lemma 2.4. If ξi⩾0, 0<θ⩽1, γ>1, for i=1,2,⋯,s, then [25]
s∑i=1ξθi⩾(s∑i=1ξ)θ, s∑i=1ξγi⩾s1−γ(s∑i=1ξi)γ. |
Lemma 2.5. Suppose the positive definite function V(t) is a continuously differentiable function on [0,+∞), if there are positive numbers 0<μ<1, λ>0, v⩾1, γ>0, such that [22]
Ct0DμtV(t)⩽−λV−v(t)−γ, |
then when t→¯T, limt→¯TV(t)=0, and for any t⩾¯T, V(t)=0 holds, where
¯T=t0+{Γ(1+μ)λ(1+v)[(V(t0)+(λγ)1v)1+v−(λγ)1+vv]}1μ. |
Lemma 2.6. Suppose k(t) is a continuously differentiable function on [t0,ξ), for any constant ζ on interval [t0,ξ), then [22]
Ct0Dμt[k(t)−ζ]2⩽2[k(t)−ζ]Ct0Dμtk(t), 0<μ<1. |
In this paper, a class of Caputo QDNNs with delays and uncertain coefficients is considered as the drive system
Ct0Dμtkp(t)=−apkp(t)+n∑w=1(bpw+Δbpw(t))gw(kw(t))+n∑w=1(cpw+Δcpw(t))gw(kw(t−τ1(t)))+n∑w=1(dpw+Δdpw(t))∫tt−τ2(t)gw(kw(s))ds+Ip(t), | (2.1) |
where 0<μ<1, kp(t) is the state vector representing the p-th neuron, ap is the self feedback regulation coefficient, gw(⋅) is the activation function, bpw+Δbpw(t), cpw+Δcpw(t) and dpw+Δdpw(t) are connection weights, Δbpw(t), Δcpw(t), Δdpw(t) are the uncertain coefficients, Ip(t) is an external input, τ1(t), τ2(t) are time-varying delays, ap∈R, the other coefficients belong to quaternion field Q.
The response system associated with System (2.1) is
Ct0Dμtop(t)=−apop(t)+n∑w=1(bpw+Δbpw(t))gw(ow(t))+n∑w=1(cpw+Δcpw(t))gw(ow(t−τ1(t)))+n∑w=1(dpw+Δdpw(t))∫tt−τ2(t)gw(oq(s))ds+Ip(t)+Up(t), | (2.2) |
where Up(t)∈Q stands for controller, kp(t0)∈Q and ow(t0)∈Q are the initial conditions of derive-response Systems (2.1) and (2.2).
Let ~ℑp(t)=op(t)−kp(t), then error system of derive-response Systems (2.1) and (2.2) is:
Ct0Dμt~ℑp(t)=−ap~ℑp(t)+n∑w=1(bpw+Δbpw(t))(gw(ow(t))−gw(kw(t)))+n∑w=1(cpw+Δcpw(t))(gw(ow(t−τ1(t)))−gw(kw(t−τ1(t))))+n∑w=1(dpw+Δdpw(t))∫tt−τ2(t)(gw(ow(s))−gw(kw(s)))ds+Up(t), | (2.3) |
then the initial condition to System (2.3) is ~ℑp(t0)=op(t0)−kp(t0).
Throughout this paper, the following assumptions and synchronization definitions are given:
Assumption 2.1. For any ki,oi, there exists a positive number li, such that
|gi(ki)−gi(oi)|⩽li|ki−oi|, i=1,2,...,n, |
where g(⋅)=(g1(⋅),g2(⋅),...,gn(⋅))T∈Qn is the activation function in System (2.1).
Assumption 2.2. For uncertain coefficients Δ(t), Δcpw(t), Δdpw(t), there exist constants ^bpw, ^cpw, ^dpw, such thatΔbpw(t)=ˆbpwαpw(t), Δcpw(t)=ˆcpwβpw(t),Δdpw(t)=ˆdpwωpw(t), where αpw(t),βpw(t), ωpw(t) are uncertain quantities under ¯αpw(t)αpw(t)⩽1, ¯βpw(t)βpw(t)⩽1, ¯ωpw(t)ωpw(t)⩽1.
Definition 2.3. For Systems (2.1) and (2.2), if there exists a positive constant ¯T related to the initial condition, such that limt→¯T||~ℑp(t)||=0, ||~ℑp(t)||=0 holds for any t⩾¯T, then Systems (2.1) and (2.2) are said to reach Fin-TS [22].
Definition 2.4. For Systems (2.1) and (2.2), if there exists a positive constant T independent of the initial condition, such that limt→T||~ℑp(t)||=0, for any t⩾T, ||~ℑp(t)||=0 holds, then Systems (2.1) and (2.2) are said to reach Fix-TS [44].
Remark 2.1. The models with uncertain parameters can further improve the robustness of systems in [22,23,24]. In this paper, the model (2.1) includes the discrete and distributed delays, uncertain term and Caputo derivative operator. Different from the constant time delays in [23,34,47], the time delays in model (2.1) are variable, which can enhance the applicability of the model and enrich the results of Fin-TS and Fix-TS.
In this section, we mainly establish Fin-TS and Fix-TS criteria between Systems (2.1) and (2.2).
Firstly, we construct a Caputo derivative differential inequality to estimate Fix-TS time for the positive definite function V(t).
Theorem 3.1. Let V(t)≜V(t,φ(t)) be a continuously positive definite and radially unbounded function, and V(t,φ(t))=0 if and only if φ(t)=0. If
Ct0DαtV(t)⩽−ξVδ(t)−ηVε(t), | (3.1) |
where 0<ε<α<1, 1<δ<1+α, ξ>0, η>0, then for any t⩾T=t0+T1+T2, the equality V(t)=0 holds, where
T1=[Γ(1−δ)Γ(1+α)−ξΓ(1+α−δ)]1α , T2=[Γ(1−ε)Γ(1+α)ηΓ(1+α−ε)]1α. | (3.2) |
Proof. We first prove that when t⩾t0, there exists a t∗, such that V(t∗)⩽1. Otherwise, for any t⩾t0, V(t)>1. From inequality (3.1), we can get that the following two differential inequalities are simultaneously true
Ct0DαtV(t)⩽−ξVδ(t), Ct0DαtV(t)⩽−ηVε(t). | (3.3) |
Taking the Caputo derivative of order α on t0→t for Vα−δ(t), from Lemma 2.1, then we get
Ct0DαtVα−δ(t)=Γ(1+α−δ)Γ(1−δ)V−δ(t)Ct0DαtV(t)⩾Γ(1+α−δ)Γ(1−δ)V−δ(t)(−ξVδ(t))=−ξΓ(1+α−δ)Γ(1−δ). | (3.4) |
For both sides of inequality (3.4), we take α-order fractional integral on t0→t below
t0D−αtCt0DαtVα−δ(t)⩾t0Iαt(−ξΓ(1+α−δ)Γ(1−δ)). | (3.5) |
Then, we have
Vα−δ(t)−Vα−δ(t0)⩾−ξΓ(1+α−δ)Γ(1−δ)1Γ(α+1)(t−t0)α, | (3.6) |
(1V(t))δ−α⩾Vα−δ(t0)+−ξΓ(1+α−δ)Γ(1−δ)Γ(1+α)(t−t0)α, | (3.7) |
V(t)⩽[1Vα−δ(t0)+−ξΓ(1+α−δ)Γ(1−δ)Γ(1+α)(t−t0)α]1δ−α. | (3.8) |
From inequality (3.8), we know that V(t) is monotonically decreasing. Take
t∗=t0+[Vδ−α(t0)−1−ξΓ(1+α−δ)Γ(1−δ)Γ(1+α)Vδ−α(t0)]1α, |
then the inequality V(t∗)⩽1 holds. Let
T1=t0+[Γ(1−δ)Γ(1+α)−ξΓ(1+α−δ)]1α, |
thus we have T1>t∗. Similarly, we can also get
Ct0DαtVα−ε(t)=Γ(1+α−ε)Γ(1−ε)V−ε(t)Ct0DαtV(t)⩽−ηΓ(1+α−ε)Γ(1−ε), | (3.9) |
V(t)⩽[Vα−ε(t∗)−ηΓ(1+α−ε)Γ(1−ε)Γ(1+α)(t−t∗)α]1α−ε⩽[1−ηΓ(1+α−ε)Γ(1−ε)Γ(1+α)(t−t∗)α]1α−ε. | (3.10) |
Let
T2=[Γ(1−ε)Γ(1+α)ηΓ(1+α−ε)]1α, |
when t=t∗+T2, then the right hand side of inequality (3.10) is equal to 0. Furthermore, we can get that V(t)⩽0 when t=t∗+T2. Since V(t) is a positive definite function, when t=t∗+T2, V(t)≡0. Let t=t0+T1+T2, then we can get V(t)=0 holds. When t⩾t0+T1+T2, we have φ(t)=0, and the settling time is estimated as T.
Remark 3.1. Recently, many synchronization results have been reported in [15,22,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43] including Fin-TS [22,26], Fix-TS [22,25,26], exponential synchronization[31,32,33,34,35], Mittag-Leffler synchronization[36,37,38], projective synchronization[36,39] and quasi-synchronization[40,41]. In Theorem 3.1, a new Caputo differential inequality is constructed, then the fixed time of the positive definite function is estimated, which is very convenient to derive Fix-TS conditions between Systems (2.1) and (2.2).
To achieve Fin-TS between Systems (2.1) and (2.2), the self feedback controller is designed
{Up(t)=ψp(t)+ϕp(t),ψp(t)=−αp~ℑp(t),ϕp(t)=−δ~ℑp(t)|~ℑp(t)|2−σ~ℑp(t)|~ℑp(t)|2ε, | (3.11) |
where ε⩾2,αp,δ and σ are positive constants.
Theorem 3.2. Under Assumptions 2.1, 2.2 and controller (3.11), if there exist positive constants h>1 and r>1, such that
2nlp2+n∑w=1(Ω+Π+Ψ)−2ap−2αp+2nl′p2h+2nlp″ | (3.12) |
then Systems (2.1) and (2.2) can achieve Fin-TS, and the settling time of Systems (2.1) and (2.2) is
\begin{equation} \overline T = {t_0}{\rm{ + }}{\left\{ {\frac{{\Gamma (1 + \mu )}}{{2\sigma {n^\varepsilon }\varepsilon }}\left[ {{{\left( {V({t_0}) + n{{(\frac{\sigma }{\delta })}^{\frac{1}{{\varepsilon - 1}}}}} \right)}^\varepsilon } - {n^\varepsilon }{{(\frac{\sigma }{\delta })}^{\frac{\varepsilon }{{\varepsilon - 1}}}}} \right]} \right\}^{\frac{1}{\mu }}}, \end{equation} | (3.13) |
where \Omega = |{b_{pw}}|^2 + \hat{|{b_{pw}}|}^2, \ \Pi = |{c_{pw}}|^2 + \hat{|{c_{pw}}|}^2, \ \Psi = |{d_{pw}}|^2 + \hat{|{d_{pw}}|}^2 , l_p, {l'_p} and {l''_p} are Lipschitz coefficients.
Proof. Choosing the following Lyapunov function
\begin{equation} {V_1}(t) = \sum\limits_{p = 1}^n {\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}}. \end{equation} | (3.14) |
From Lemma 2.2, we have
\begin{align} {}_{{t_0}}^CD_t^\mu {V_1}(t)\leqslant& \sum\limits_{p = 1}^n {\Bigg[\overline {\widetilde{\Im_p(t)}} {}_{{t_0}}^CD_t^\mu \widetilde{\Im_p(t)} + ({}_{{t_0}}^CD_t^\mu \overline {\widetilde{\Im_p(t)}} )\widetilde{\Im_p(t)}\Bigg]} \\ = & \sum\limits_{p = 1}^n {\overline {\widetilde{\Im_p(t)}} } \left[ { - {a_p}\widetilde{\Im_p(t)} + \sum\limits_{w = 1}^n {\Big({b_{pw}} + \Delta {b_{pw}}(t)\Big) \Big({g_w}} ({o_w}(t)) - {g_w}({k_w}(t))\Big)} \right]\\ &+ \sum\limits_{p = 1}^n {\overline {\widetilde{\Im_p(t)}} } \left[ {\sum\limits_{w = 1}^n \Big({c_{pw}} + \Delta {c_{pw}}(t)\Big) \Big({g_w}({o_w}(t - {\tau _1}(t))) - {g_w}({k_w}(t - {\tau _1}(t)))\Big)} \right]\\ &+ \sum\limits_{p = 1}^n {\overline {\widetilde{\Im_p(t)}} } \left[ {\sum\limits_{w = 1}^n {\Big({d_{pw}} + \Delta {d_{pw}}(t)\Big)} \int_{t - {\tau _2}(t)}^t {\Big({g_w}({o_w}(s))} - {g_w}({k_w}(s))\Big)ds + {U_p}(t)} \right]\\ &+ \sum\limits_{p = 1}^n {\left[ { - {a_p}\overline {\widetilde{\Im_p(t)}} + \sum\limits_{w = 1}^n {\overline {\Big({g_w}({o_w}(t)) - {g_w}({k_w}(t))\Big)} } \Big(\overline {{b_{pw}}} + \overline {\Delta {b_{pw}}(t)} \Big)} \right]\widetilde{\Im_p(t)}} \\ &+ \sum\limits_{p = 1}^n {\left[ {\sum\limits_{w = 1}^n {\overline {\Big({g_w}({o_w}(t - {\tau _1}(t))) - {g_w}({k_w}(t - {\tau _1}(t)))\Big)} } \Big(\overline {{c_{pw}}} + \overline {\Delta {c_{pw}}(t)} \Big)} \right]} \widetilde{\Im_p(t)}\\ &+ \sum\limits_{p = 1}^n {\left[ {\sum\limits_{w = 1}^n {\int_{t - {\tau _2}(t)}^t {\overline {\Big({g_w}({o_w}(s)) - {g_w}({k_w}(s))\Big)} } } ds \Big(\overline {{d_{pw}}} + \overline {\Delta {d_{pw}}(t)} \Big) + \overline {{U_p}(t)} } \right]} \widetilde{\Im_p(t)}. \end{align} | (3.15) |
By computation, we further get
\begin{align} \label{eq:3.16} {}_{{t_0}}^CD_t^\mu {V_1}(t)\leqslant& \sum\limits_{p = 1}^n {\left[ { - 2{a_p}\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)} + \overline {\widetilde{\Im_p(t)}} {U_p}(t) + \overline {{U_p}(t)} \widetilde{\Im_p(t)}} \right]} \\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} } } {b_{pw}}\Big({g_w}({o_w}(t)) - {g_w}({k_w}(t))\Big)\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\Big({g_w}({o_w}(t)) - {g_w}({k_w}(t))\Big)} } } \overline {{b_{pw}}} \widetilde{\Im_p(t)}\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} } } \Delta {b_{pw}}(t)\Big({g_w}({o_w}(t)) - {g_w}({k_w}(t))\Big)\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\Big({g_w}({o_w}(t)) - {g_w}({k_w}(t))\Big)} } } \overline {\Delta {b_{pw}}(t)} \widetilde{\Im_p(t)}\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} } } {c_{pw}}\Big({g_w}({o_w}(t - {\tau _1}(t))) - {g_w}({k_w}(t - {\tau _1}(t)))\Big)\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\Big({g_w}({o_w}(t - {\tau _1}(t))) - {g_w}({k_w}(t - {\tau _1}(t)))\Big)} } \overline {{c_{pw}}} \widetilde{\Im_p(t)}} \\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} } } \Delta {c_{pw}}(t)\Big({g_w}({o_w}(t - {\tau_1}(t))) - {g_w}({k_w}(t - {\tau _1}(t)))\Big)\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\Big({g_w}({o_w}(t - {\tau _1}(t))) - {g_w}({k_w}(t - {\tau _1}(t)))\Big)} \overline {\Delta {c_{pw}}(t)}\widetilde{\Im_p(t)}} } \\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} {d_{pw}}\int_{t - {\tau _2}(t)}^t {\Big({g_w}({o_w}(s)) - {g_w}({k_w}(s))\Big)} } } ds\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\int_{t - {\tau _2}(t)}^t {\overline {\Big({g_w}({o_w}(s)) - {g_w}({k_q}(s))\Big)} } ds\overline {{d_{pw}}} \widetilde{\Im_p(t)}} } \\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} } } \Delta {d_{pw}}(t)\int_{t - {\tau _2}(t)}^t {\Big({g_w}({o_w}(s)) - {g_w}({k_w}(s))\Big)} ds\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\int_{t - {\tau _2}(t)}^t {\overline {\Big({g_w}({o_w}(s)) - {g_w}({k_w}(s))\Big)} ds\overline {\Delta {d_{pw}}(t)}\widetilde{\Im_p(t)}}}}. \end{align} | (3.16) |
According to controller (3.11), we have
\begin{align} \sum\limits_{p = 1}^n {\Bigg[\overline {\widetilde{\Im_p(t)}} {U_p}(t) + \overline {{U_p}(t)} \widetilde{\Im_p(t)}\Bigg]} = & \sum\limits_{p = 1}^n {\overline {\widetilde{\Im_p(t)}} } \left[ { - {\alpha _p}\widetilde{\Im_p(t)}- \frac{{\delta \widetilde{\Im_p(t)}}}{{|\widetilde{\Im_p(t)}{|^2}}} - \frac{{\sigma \widetilde{\Im_p(t)}}}{{|\widetilde{\Im_p(t)}{|^{2\varepsilon }}}}} \right]\\ &+ \sum\limits_{p = 1}^n {\left[ { - {\alpha _p}\overline {\widetilde{\Im_p(t)}} - \frac{{\delta \overline {\widetilde{\Im_p(t)}} }}{{|\widetilde{\Im_p(t)}{|^2}}} - \frac{{\sigma \overline {\widetilde{\Im_p(t)}} }}{{|\widetilde{\Im_p(t)}{|^{2\varepsilon }}}}} \right]} \widetilde{\Im_p(t)} \\ \leqslant & \sum\limits_{p = 1}^n {\left[-2\alpha_p \overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)} - 2\sigma {{\Big(\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}\Big)}^{1 - \varepsilon } - 2\delta } \right]}. \end{align} | (3.17) |
In terms of Lemma 2.3, Assumptions 2.1 and 2.2, we have
\begin{align} & \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} } } {b_{pw}}\Big({g_w}({o_w}(t)) - {g_w}({k_w}(t))\Big)+\sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\Big({g_w}({o_w}(t)) - {g_w}({k_w}(t))\Big)} } } \overline {{b_{pw}}} \widetilde{\Im_p(t)}\\ & \leqslant \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {|{b_{pw}}{|^2}\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}} } + \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {l_w^2} } \overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}\\ &\leqslant \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {|{b_{pw}}{|^2}\overline {\widetilde{\Im_p(t)}}\widetilde{\Im_p(t)}} } + n\sum\limits_{p = 1}^n {l_w^2} \overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}. \end{align} | (3.18) |
Similarly, we can get
\begin{align} \label{eq:3.19} & \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} } } \Delta {b_{pw}}(t)\Big({g_w}({o_w}(t)) - {g_w}({k_w}(t))\Big)\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\Big({g_w}({o_w}(t)) - {g_w}({k_w}(t))\Big)} } } \overline {\Delta {b_{pw}}(t)}\widetilde{\Im_p(t)}\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} } } {c_{pw}}\Big({g_w}({o_w}(t - {\tau _1}(t))) - {g_w}({k_w}(t - {\tau _1}(t)))\Big)\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\Big({g_w}({o_w}(t - {\tau _1}(t))) - {g_w}({k_w}(t - {\tau _1}(t)))\Big)} } \overline {{c_{pw}}} \widetilde{\Im_p(t)}} \\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} } } \Delta {c_{pw}}(t)\Big({g_w}({o_w}(t - {\tau _1}(t))) - {g_w}({k_w}(t - {\tau _1}(t)))\Big)\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\Big({g_w}({o_w}(t - {\tau _1}(t))) - {g_w}({k_w}(t - {\tau _1}(t)))\Big)} \overline {\Delta {c_{pw}}(t)}\widetilde{\Im_p(t)}} } \\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} {d_{pw}}\int_{t - {\tau _2}(t)}^t {\Big({g_w}({o_w}(s)) - {g_w}({k_w}(s))\Big)} } } ds\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\int_{t - {\tau _2}(t)}^t {\overline {\Big({g_w}({o_w}(s)) - {g_w}({k_w}(s))\Big)} } ds\overline {{d_{pw}}} \widetilde{\Im_p(t)}} } \\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\overline {\widetilde{\Im_p(t)}} } } \Delta {d_{pw}}(t)\int_{t - {\tau _2}(t)}^t {\Big({g_w}({o_w}(s)) - {g_w}({k_w}(s))\Big)} ds\\ &+ \sum\limits_{p = 1}^n {\sum\limits_{w = 1}^n {\int_{t - {\tau _2}(t)}^t {\overline {\Big({g_w}({o_w}(s)) - {g_w}({o_w}(s))\Big)} ds\overline {\Delta {d_{pw}}(t)} \widetilde{\Im_p(t)}} } }\\ \leqslant& \sum\limits_{p = 1}^n \left[ nl_p^2+ \sum\limits_{w = 1}^n{\hat{|b_{pw}|}}^2 + \sum\limits_{w = 1}^n({{|c_{pw}|}}^2+{\hat{|c_{pw}|}}^2)+ \sum\limits_{w = 1}^n({{|d_{pw}|}}^2+{\hat{|d_{pw}|}}^2) \right]\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)} \\ &+ 2n\sum\limits_{p = 1}^n {{l'_p}^2\overline {\widetilde{\Im_p\Big(t-\tau_1(t)\Big)}} } \widetilde{\Im_p\Big(t-\tau_1(t)\Big)}+ 2n\tau _2^2(t)\sum\limits_{p = 1}^n {{l''_p}^2} \mathop {\sup }\limits_{t - {\tau _2}(t) \leqslant l \leqslant t} |\widetilde{\Im_p(l)}{|^2}. \end{align} | (3.19) |
Substituting (3.17) – (3.19) to (3.16) , we can get
\begin{align} \label{eq:3.20} {}_{{t_0}}^CD_t^\mu {V_1}(t)\leqslant&\sum\limits_{p = 1}^n {\left[ 2nl_p^2 + \sum\limits_{w = 1}^n (\Omega+\Pi+\Psi)-2a_p-2\alpha_p \right]}\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}\\ &+ \sum\limits_{p = 1}^n {2n{l'_p}^2 \overline {\widetilde{\Im_p\Big(t-\tau_1(t)\Big)}} } \widetilde{\Im_p\Big(t-\tau_1(t)\Big)}+ \sum\limits_{p = 1}^n {2n{l''_p}^2 \tau _2^2(t)\mathop {\sup }\limits_{t - {\tau _2}(t) \leqslant l \leqslant t} } |\widetilde{\Im_p(l)}{|^2}\\ &- 2\sigma \sum\limits_{p = 1}^n {{{\Big(\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}\Big)}^{1 - \varepsilon }}} - 2\delta n, \end{align} | (3.20) |
where \Omega = |{b_{pw}}|^2 + \hat{|{b_{pw}}|}^2, \ \Pi = |{c_{pw}}|^2 + \hat{|{c_{pw}}|}^2, \ \Psi = |{d_{pw}}|^2 + \hat{|{d_{pw}}|}^2 . From (3.12) , Lemma 2.4, fractional-order Razumikhin theorem and Jensen Inequality, there exist h > 1 and r > 1 , such that
\begin{equation} \begin{aligned} {}_{{t_0}}^CD_t^\mu {V_1}(t)\leqslant& \sum\limits_{p = 1}^n {\left[ 2nl_p^2 + \sum\limits_{w = 1}^n (\Omega+\Pi+\Psi) -2a_p-2\alpha_p \right]}\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}\\ &+\sum\limits_{p = 1}^n {2n{l'_p}^2 h \overline {\widetilde{\Im_p(t)}} {\widetilde{\Im_p(t)}}} + \sum\limits_{p = 1}^n {2n{l''_p}^2 r \overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}}- 2\sigma \sum\limits_{p = 1}^n {{{\Big(\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}\Big)}^{1 - \varepsilon }}} - 2\delta n \\ \leqslant& \sum\limits_{p = 1}^n {\left[ 2nl_p^2 + \sum\limits_{w = 1}^n (\Omega+\Pi+\Psi) -2a_p-2\alpha_p +2n{l'_p}^2 h +2n{l''_p}^2 r \right]}\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)} \\ & -2\sigma n^{\varepsilon} \Big(\sum\limits_{p = 1}^n \overline{\widetilde{\Im_p(t)}}\widetilde{\Im_p(t)}\Big)^{1-\varepsilon}-2\delta n \\ \leqslant& -2\sigma n^{\varepsilon}{V_1^{^{-(\varepsilon-1)}}(t)} -2\delta n. \end{aligned} \end{equation} | (3.21) |
According to Lemma 2.5, Systems (2.1) and (2.2) can realize Fin-TS under controller (3.11) , and the settling time of Systems (2.1) and (2.2) is estimated as
\begin{equation} \begin{aligned} \overline T = {t_0}{\rm{ + }}{\left\{ {\frac{{\Gamma (1 + \mu )}}{{2\sigma {n^\varepsilon }\varepsilon }}\left[ {{{\left( {V({t_0}) + n{{(\frac{\sigma }{\delta })}^{\frac{1}{{\varepsilon - 1}}}}} \right)}^\varepsilon } - {n^\varepsilon }{{(\frac{\sigma }{\delta })}^{\frac{\varepsilon }{{\varepsilon - 1}}}}} \right]} \right\}^{\frac{1}{\mu }}}. \end{aligned} \end{equation} | (3.22) |
Therefore, when t \geqslant \overline{T} , we have ||\widetilde{\Im_p(t)}|| = 0 , then Systems (2.1) and (2.2) can realize Fin-TS under controller (3.11) .
In order to realize Fix-TS of Systems (2.1) and (2.2) , the following adaptive controller is designed
\begin{equation} \left\{ \begin{aligned} & {{U_p}(t) = {\Xi _p}(t) + {{\rm H}_p}(t)}, \\ & {{\Xi _p}(t) = - {\sigma _p}(t)\widetilde{\Im_p(t)}}, \\ & {{{\rm H}_p}(t) = - {k_{1p}}\frac{{\widetilde{\Im_p(t)}}}{{{{\Big(\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}\Big)}^\delta }}} - {k_{2p}}\frac{{\widetilde{\Im_p(t)}}}{{{{\Big(\overline {\widetilde{\Im_p(t)}} \widetilde{\Im_p(t)}\Big)}^\varepsilon }}}}, \end{aligned} \right. \end{equation} | (3.23) |
where - \mu < \delta < 0, \ 1-\mu < \varepsilon < 1, k_{1p} > 0, k_{2p} > 0, and \sigma_p(t) is the adaptive control gain satisfying
\begin{align} {}_{{t_0}}^CD_t^\mu \sigma_p(t) = &\overline {\widetilde{\Im_p(t)}}\widetilde{\Im_p(t)} -S_p\Big[sign(\sigma_p(t)-\sigma^*_p)\Big]|\sigma_p(t)-\sigma^*_p|^{1-2\delta}\\ &-R_p\Big[sign(\sigma_p(t)-\sigma^*_p)\Big]|\sigma_p(t)-\sigma^*_p|^{1-2\varepsilon}. \end{align} | (3.24) |
Theorem 3.3. Under Assumptions 2.1, 2.2 and controller (3.23) , if there exist positive constants h > 1 and r > 1 , such that
\begin{equation} \begin{aligned} n{l_p}^2 + \frac{1}{2}\sum\limits_{w = 1}^n {(\Omega + \Pi + \Psi )} - {a_p} + n{l'_p}^2 h + n{l''_p}^2 r\leqslant {\sigma _p^*}, \end{aligned} \end{equation} | (3.25) |
then Systems (2.1) and (2.2) can realize Fix-TS, and the settling time of Systems (2.1) and (2.2) is
\begin{equation} \begin{aligned} T = {t_0} + {\left[ {\frac{{\Gamma (\delta )\Gamma (1 + \mu )}}{{ - {{(2n)}^\delta }{H_1}\Gamma (\mu + \delta )}}} \right]^{\frac{1}{\mu }}} + {\left[ {\frac{{\Gamma (\varepsilon )\Gamma (1 + \mu )}}{{{H_2}\Gamma (\mu + \varepsilon )}}} \right]^{\frac{1}{\mu }}}, \end{aligned} \end{equation} | (3.26) |
where {l_p}, {l'_p} and {l''_p} are Lipschitz coefficients.
Proof. Take into account the following Lyapunov function
\begin{equation} \begin{aligned} {V_2}(t) = \sum\limits_{p = 1}^n {{\lambda _p}\overline {\widetilde {{\Im_p}(t)}} } \widetilde {{\Im_p}(t)} + {\sum\limits_{p = 1}^n {{\lambda _p}\Big[{\sigma _p}(t) - \sigma _p^*\Big]} ^2}. \end{aligned} \end{equation} | (3.27) |
Similar to the proof Theorem 3.2, by Lemma 2.6 we have
\begin{equation} \begin{aligned} {}_{{t_0}}^CD_t^\mu {V_2}(t)\leqslant &\sum\limits_{p = 1}^n {{\lambda _p}\left[ {2nl_p^2 + \sum\limits_{w = 1}^n {(\Omega + \Pi + \Psi )} - 2{a_p} - 2{\sigma _p}(t)} \right]} \overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}\\ & + \sum\limits_{p = 1}^n {2n \lambda_p {l'_p}^2 \overline {\widetilde {{\Im_p}\Big(t - {\tau _1}(t)\Big)}} {\widetilde {{\Im_p}\Big(t - {\tau _1}(t)\Big)}} } + \sum\limits_{p = 1}^n {2n{\lambda _p} {l''_p}^2{\tau _2}(t)\mathop {\sup }\limits_{t - {\tau _2}(t) \leqslant l \leqslant t} } |\widetilde {{\Im_p}(l)}{|^2}\\ &- 2{k_{1p}}{\sum\limits_{p = 1}^n {{\lambda _p}\Big(\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}\Big)} ^{1 - \delta }} - 2{k_{2p}}{\sum\limits_{p = 1}^n {{\lambda _p}\Big(\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}\Big)} ^{1 - \varepsilon }}\\ &+ \sum\limits_{p = 1}^n {{\lambda _p}2\Big({\sigma _p}(t) - \sigma _p^*\Big)} {}_{{t_0}}^CD_t^\mu {\sigma _p}(t). \end{aligned} \end{equation} | (3.28) |
Under adaptive controller (3.23), we have
\begin{equation} \begin{aligned} {}_{{t_0}}^CD_t^\mu {V_2}(t)\leqslant& \sum\limits_{p = 1}^n {{\lambda _p}\left[ {2nl_p^2 + \sum\limits_{w = 1}^n {(\Omega + \Pi + \Psi )} - 2{a_p} - 2\sigma _p^*} \right]} \overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}\\ & + \sum\limits_{p = 1}^n {2n \lambda_p {l'_p}^2 \overline {\widetilde {{\Im_p}\Big(t - {\tau _1}(t)\Big)}} {\widetilde {{\Im_p}\Big(t - {\tau _1}(t)\Big)}} } + \sum\limits_{p = 1}^n {2n{\lambda _p} {l''_p}^2{\tau _2}(t)\mathop {\sup }\limits_{t - {\tau _2}(t) \leqslant l \leqslant t} } |\widetilde {{\Im_p}(l)}{|^2}\\ &- 2{k_{1p}}{\sum\limits_{p = 1}^n {{\lambda _p}\Big(\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}\Big)} ^{1 - \delta }} - \sum\limits_{p = 1}^n {2{\lambda _p}{S_p}{{\left( {{{({\sigma _p}(t) - \sigma _p^*)}^2}} \right)}^{1 - \delta }}} \\ &- 2{k_{2p}}{\sum\limits_{p = 1}^n {{\lambda _p}\Big(\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}\Big)} ^{1 - \varepsilon }} - \sum\limits_{p = 1}^n {2{\lambda _p}{R_p}{{\left( {{{({\sigma _p}(t) - \sigma _p^*)}^2}} \right)}^{1 - \varepsilon }}}. \end{aligned} \end{equation} | (3.29) |
By fractional-order Razumikhin theorem, there are h > 1, r > 1 , The following formula holds:
\begin{equation} \begin{aligned} {}_{{t_0}}^CD_t^\mu {V_2}(t) &\leqslant \sum\limits_{p = 1}^n {{\lambda _p}\left[ {2nl_p^2 + \sum\limits_{w = 1}^n {(\Omega + \Pi + \Psi )} - 2{a_p} - 2\sigma _p^* + 2n {l'_p}^2 h + 2n{l''_p}^2 r} \right]} \overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}\\ &- 2{k_{1p}}{\sum\limits_{p = 1}^n {{\lambda _p}\Big(\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}\Big)} ^{1 - \delta }} - \sum\limits_{p = 1}^n {2{\lambda _p}{S_p}{{\left( {{{({\sigma _p}(t) - \sigma _p^*)}^2}} \right)}^{1 - \delta }}} \\ &- 2{k_{2p}}{\sum\limits_{p = 1}^n {{\lambda _p}\Big(\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}\Big)} ^{1 - \varepsilon }} - \sum\limits_{p = 1}^n {2{\lambda _p}{R_p}{{\left( {{{({\sigma _p}(t) - \sigma _p^*)}^2}} \right)}^{1 - \varepsilon }}}. \end{aligned} \end{equation} | (3.30) |
An application of Lemma 2.4 yields that
\begin{equation} \begin{aligned} &\sum\limits_{p = 1}^n {2{k_{1p}}} {\lambda _p}{\left( {\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}} \right)^{1 - \delta }} + \sum\limits_{p = 1}^n {2{S_p}{\lambda _p}{{\left( {{{({\sigma _p}(t) - \sigma _p^*)}^2}} \right)}^{1 - \delta }}} \\ & = \sum\limits_{p = 1}^n {2{k_{1p}}\lambda _p^\delta } {\left( {{\lambda _p}\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}} \right)^{1 - \delta }} + \sum\limits_{p = 1}^n {2{S_p}\lambda _p^\delta } {\left( {{\lambda _p}{{({\sigma _p}(t) - \sigma _p^*)}^2}} \right)^{1 - \delta }}\\ &\geqslant {H_1}\left[ {\sum\limits_{p = 1}^n {{{\left( {{\lambda _p}\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}} \right)}^{1 - \delta }}} + \sum\limits_{p = 1}^n {{{\left( {{\lambda _p}{{({\sigma _p}(t) - \sigma _p^*)}^2}} \right)}^{1 - \delta }}} } \right]\\ &\geqslant {2^\delta }{H_1}\left[ {\sum\limits_{p = 1}^n {{{\left[ {{\lambda _p}\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)} + {\lambda _p}{{\Big({\sigma _p}(t) - \sigma _p^*\Big)}^2}} \right]}^{1 - \delta }}} } \right]\\ &\geqslant {(2n)^\delta }{H_1}{\left[ {\sum\limits_{p = 1}^n {\left[ {{\lambda _p}\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)} + {\lambda _p}{{\Big({\sigma _p}(t) - \sigma _p^*\Big)}^2}} \right]} } \right]^{1 - \delta }}\\ & = {(2n)^\delta }{H_1}V_2^{1 - \delta }(t). \end{aligned} \end{equation} | (3.31) |
Similarly, one can get
\begin{equation} \begin{aligned} &{\sum\limits_{p = 1}^n {2{k_{2p}}{\lambda _p}\left( {\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}} \right)} ^{1 - \varepsilon }} + \sum\limits_{p = 1}^n {2{R_p}{\lambda _p}{{\left( {{{\left( {{\sigma _p}(t) - \sigma _p^*} \right)}^2}} \right)}^{1 - \varepsilon }}} \\ &\geqslant {H_2}\left[ {\sum\limits_{p = 1}^n {{{\left( {{\lambda _p}\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)}} \right)}^{1 - \varepsilon }} + \sum\limits_{p = 1}^n {\left( {{\lambda _p}{{\left( {{\sigma _p}(t) - \sigma _p^*} \right)}^2}} \right)} } } \right]\\ &\geqslant {H_2}\sum\limits_{p = 1}^n {{{\left[ {{\lambda _p}\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)} + {\lambda _p}{{\left( {{\sigma _p}(t) - \sigma _p^*} \right)}^2}} \right]}^{1 - \varepsilon }}} \\ &\geqslant {H_2}{\left[ {\sum\limits_{p = 1}^n {\left[ {{\lambda _p}\overline {\widetilde {{\Im_p}(t)}} \widetilde {{\Im_p}(t)} + {\lambda _p}{{\left( {{\sigma _p}(t) - \sigma _p^*} \right)}^2}} \right]} } \right]^{1{\rm{ - }}\varepsilon }}\\ &{\rm{ = }}{H_2}V_2^{1{\rm{ - }}\varepsilon }(t), \end{aligned} \end{equation} | (3.32) |
where {H_1} = \mathop {\min }\limits_{1 \leqslant p \leqslant n} \left\{ {2{k_{1p}}\lambda _p^\delta, 2{S_p}\lambda _p^\delta } \right\}, {H_2} = \mathop {\min }\limits_{1 \leqslant p \leqslant n} \left\{ {2{k_{2p}}\lambda _p^\varepsilon, 2{R_p}\lambda _p^\varepsilon } \right\} . Substituting (3.31) and (3.32) to (3.30) , from (3.25) , we have
\begin{equation} \begin{aligned} {}_{{t_0}}^CD_t^\mu {V_2}(t) \leqslant - {(2n)^\delta }{H_1}V_2^{1 - \delta }(t) - {H_2}V_2^{1 - \varepsilon }(t). \end{aligned} \end{equation} | (3.33) |
According to Theorem 3.1 , Systems (2.1) and (2.2) can achieve Fix-TS under controller (3.23) , and the settling time is estimated as
\begin{equation} \begin{aligned} T = {t_0} + {\left[ {\frac{{\Gamma (\delta )\Gamma (1 + \mu )}}{{ - {{(2n)}^\delta }{H_1}\Gamma (\mu + \delta )}}} \right]^{\frac{1}{\mu }}} + {\left[ {\frac{{\Gamma (\varepsilon )\Gamma (1 + \mu )}}{{{H_2}\Gamma (\mu + \varepsilon )}}} \right]^{\frac{1}{\mu }}}. \end{aligned} \end{equation} | (3.34) |
Therefore, when t\geqslant T , we have \mathop {\lim }\limits_{t \to T} ||\widetilde {{\Im_p}(t)}|| = 0 . Thus, Systems (2.1) and (2.2) can realize Fix-TS under controller (3.23) .
Remark 3.2. Different from other synchronization controller designs such as exponential synchronization [31,32,33,34,35], quasi-synchronization [40,41] and global synchronization [42], the Fin-TS and Fix-TS controllers (3.11) and (3.23) are designed in two steps including the synchronization control term and stability control term, which can greatly improve the performance of the controller, enhance the operability and system stability, and also allow the synchronization term gain and stability term gain to be adjusted.
Remark 3.3. In contrast to the decomposition method [46], Laplace transform [47], comparison principle [48,49] and linear matrix inequality (LMI) [50], the algebraic condition (3.12) in Theorem 3.2 is proposed to realize Fin-TS based on the quaternion direct method, Lyapunov stability theory, extended Cauchy Schwartz inequality and Jensen inequality.
In this section, two numerical examples are given to verify the correctness and validity of Theorems 3.1–3.3 under the different derivative orders.
Consider the following Caputo QDNN model:
{}_{{t_0}}^CD_t^\mu {k_p}(t) = - {a_p}{k_p}(t) + \sum\limits_{w = 1}^2 {\Big({b_{pw}} + \Delta {b_{pw}}(t)\Big)} {g_w}\Big({k_w}(t)\Big) + \sum\limits_{w = 1}^2 {\Big({c_{pw}} + \Delta {c_{pw}}(t)\Big){g_w}\\ \Big({k_w}(t - {\tau _1}(t))\Big)}\\ + \sum\limits_{w = 1}^2 {\Big({d_{pw}} + \Delta {d_{pw}}(t)\Big)\int_{t - {\tau _2}(t)}^t {{g_w}\Big({k_w}(s)\Big)} ds} + {I_p}(t), \ p = 1, 2, | (4.1) |
where
\begin{align*} \mu = 0.73, \ \tau_1(t) = 0.5+0.5|sin(t)|, \ \tau_2(t) = 0.5+0.5|cos(t)|, \ I_1(t) = I_2(t) = 0, \end{align*} |
\begin{align*} {g_w}\Big({k_w}(t)\Big) = \tanh \Big(k_w^R(t)\Big) + i\tanh \Big(k_w^I(t)\Big) + j\tanh \Big(k_w^J(t)\Big) + k\tanh \Big(k_w^K(t)\Big), \end{align*} |
\begin{align*} {k_p}(t) = k_p^R(t) + ik_p^I(t) + jk_p^J(t) + kk_p^K(t) \in \mathbb{Q}, \end{align*} |
\begin{align*} \begin{array}{l} A{\rm{ = }}\left( {\begin{array}{*{20}{c}} 1&0\\ 0&1 \end{array}} \right), \ B = \left( {\begin{array}{*{20}{c}} {0.1 + 0.5i + 0.8j + 0.1k}&{0.2 - 0.2i + 0.2j - 0.2k}\\ {0.1 + 0.1i - 0.1j + 0.1k}&{0.1 + 0.3i - 0.1j + 0.2k} \end{array}} \right)\\ + \left( {\begin{array}{*{20}{c}} {0.4\Big(\sin t + i\sin t - j\sin t - k\sin t\Big)}&{0.4\Big(\cos t + i\cos t + j\cos t + k\cos t\Big)}\\ {0.4\Big(\cos t + i\cos t + j\cos t + k\sin t\Big)}&{0.4\Big(\sin t + i\sin t + j\sin t + k\sin t\Big)} \end{array}} \right), \\\\ C = \left( {\begin{array}{*{20}{c}} {0.3 + 0.3i + 0.2j + 0.2k}&{0.1 - 0.2i + 0.1j + 0.2k}\\ {0.1 + 0.1i + 0.1k}&{0.6 + 0.6j + 0.2k} \end{array}} \right)\\ + \left( {\begin{array}{*{20}{c}} {0.4\Big(\cos t + i\cos t - j\sin t - k\sin t\Big)}&{0.4\Big(\sin t + i\cos t + j\cos t - k\sin t\Big)}\\ {0.4\Big(\cos t + i\sin t + j\sin t + k\sin t\Big)}&{0.4\Big(\cos t - i\cos t - j\cos t + k\sin t\Big)} \end{array}} \right), \\\\ D = \left( {\begin{array}{*{20}{c}} {0.1 + 0.1i + 0.1j + 0.1k}&{0.3 + 0.1i + 0.2j + 0.2k}\\ {0.1 + 0.1i + 0.1j + 0.1k}&{0.3 + 0.1i + 0.3j - 0.2k} \end{array}} \right)\\ + \left( {\begin{array}{*{20}{c}} {0.4\Big(\sin t + i\sin t + j\sin t + k\sin t\Big)}&{0.4\Big(\cos t + i\cos t + j\cos t + k\cos t\Big)}\\ {0.4\Big(\sin t + i\cos t + j\sin t + k\cos t\Big)}&{0.4\Big(-\cos t - i\sin t - j\sin t + k\cos t\Big)} \end{array}} \right). \end{array} \end{align*} |
The response system is
{}_{{t_0}}^CD_t^\mu {o_p}(t) = - {a_p}{o_p}(t) + \sum\limits_{w = 1}^2 {\Big({b_{pw}} + \Delta {b_{pw}}(t)\Big)} {g_w}\Big({o_w}(t)\Big) + \sum\limits_{w = 1}^2 {\Big({c_{pw}} + \Delta {c_{pw}}(t)\Big){g_w}\\ \Big({o_w}(t - {\tau _1}(t))\Big)}\\ + \sum\limits_{w = 1}^2 {\Big({d_{pw}} + \Delta {d_{pw}}(t)\Big)\int_{t - {\tau _2}(t)}^t {{g_w}\Big({o_w}(s)\Big)} ds} + {I_p}(t)+U_p(t), \ p = 1, 2, | (4.2) |
where {o_p}(t) = o_p^R(t) + Io_p^I(t) + Jo_p^J(t) + Ko_p^K(t) \in \mathbb{Q}. The initial conditions of Systems (4.1) and (4.2) are chosen as
\begin{align*} \begin{aligned} &{k_1}(0) = 0.1 + 0.4i + 0.3j +0.6 k, \ {k_2}(0) = - 0.1 - 0.1i - 0.4j + 0.2k, \\ &{w_1}(0) = - 0.1 - 0.4i -0.3j -0.6k, \ {w_2}(0) = 0.1 + 0.1i + 0.4j -0.2 k. \end{aligned} \end{align*} |
Example 4.1. To realize Fin-TS of Systems (4.1) and (4.2) , for the self feedback controller (3.11) , the following control gains are taken as \delta = 0.9, \ \sigma = 0.1, \ \varepsilon = 2.1, \ h = r = 1.5, \ {l_p} = {l'_p} = {l''_p} = 1. By calculation, we can get
\begin{equation*} \begin{aligned} 9.8 = \alpha_1\geqslant& 2l_1^2 +0.5\Big({\left|b_{11} \right|}^2 +{\left| \hat {b_{11} } \right|}^2+{\left|b_{12} \right|}^2 +{\left| \hat {b_{12} } \right|}^2 +{\left|c_{11} \right|}^2 +{\left| \hat {c_{11} } \right|}^2+{\left|c_{12} \right|}^2 +{\left| \hat {c_{12} } \right|}^2\\ &+{\left|d_{11} \right|}^2 +{\left| \hat {d_{11} } \right|}^2+{\left|d_{12} \right|}^2 +{\left| \hat {d_{12} } \right|}^2\Big)-a_1+2{l'_1}^2h + 2{l''_1}^2r\\ = &9.7450, \end{aligned} \end{equation*} |
\begin{equation*} \begin{aligned} 9.6 = \alpha_2\geqslant& 2l_2^2 +0.5\Big({\left|b_{21} \right|}^2 +{\left| \hat {b_{21} } \right|}^2+{\left|b_{22} \right|}^2 +{\left| \hat {b_{22} } \right|}^2 +{\left|c_{21} \right|}^2 +{\left| \hat {c_{21} } \right|}^2+{\left|c_{22} \right|}^2 +{\left| \hat {c_{22} } \right|}^2\\ &+{\left|d_{21} \right|}^2 +{\left| \hat {d_{21} } \right|}^2+{\left|d_{22} \right|}^2 +{\left| \hat {d_{22} } \right|}^2\Big)-a_2+2{l'_2}^2h + 2{l''_2}^2r\\ = &9.5500. \end{aligned} \end{equation*} |
From Theorem 3.2, the settling time can be estimated as \overline T \approx 4.5123 . Figure 1 shows the state trajectories of derive-response Systems (4.1) and (4.2) without controller. Figures 2 and 3 describe Fin-TS trajectories and error norm of Systems (4.1) and (4.2) under controller (3.11) when the order is 0.73. Figures 4 and 5 depict Fin-TS trajectories and error norm of Systems (4.1) and (4.2) under controller (3.11) when the derivative order is 0.53. Figures 2–5 reveal that the numerical simulations are consistent with Theorem 3.2.
Example 4.2. To realize Fix-TS between Systems (4.1) and (4.2) , an adaptive controller (3.20) is designed based on Theorem 3.1. Taking
\begin{align*} {k_{11}} = {k_{12}} = 0.1, \ {k_{21}} = {k_{22}} = 0.3, \ \delta = -0.31, \ r = h = 1.5, \\ \varepsilon = 0.9, \ {\lambda _p} = {R_p} = {S_p} = 1, \ \sigma _1^* = \sigma _2^* = 10.0, \ H_1 = 0.2, H_2 = 0.6, \end{align*} |
the other model parameters are consistent with Example 4.1. By computation, the conditions of Theorem 3.3 are satisfied. The settling time can be estimated as T\approx5.2198 . Figures 6 and 7 show Fix-TS trajectories and error norm of Systems (4.1) and (4.2) under controller (3.23) when the order is 0.73. Figures 8 and 9 characterize Fix-TS trajectories and error norm of Systems (4.1) and (4.2) under controller (3.23) when the derivative order is 0.53. Figures 6–9 verify that the numerical simulations are coincident with Theorems 3.1 and 3.3.
Remark 4.1. Figures 10 and 11 depict the relationship between the settling time \overline{T} of Fin-TS with the order \mu and the parameters \sigma , \delta and \varepsilon . Figures 12 and 13 illustrate the relationship between the settling time T of Fix-TS with the order \mu and the parameters \delta , \varepsilon . For Fin-TS, it can be observed from Figures 10 and 11 that the settling time is positively correlated with \sigma , \varepsilon and negatively correlated with \delta , \mu . For Fix-TS the settling time is positively correlated with \delta and negatively correlated with \varepsilon , \mu . Therefore, the settling time of synchronisation is closely related to the order of the equation, regardless of whether it is Fin-TS or Fix-TS.
This paper has investigated Fin-TS and Fix-TS issues for derive-response Systems (2.1) and (2.2) . Firstly, a new Caputo fractional differential inequality (3.1) is constructed, then Fix-TS settling time of the positive definite function V(t) is estimated, which is very convenient to derive Fix-TS condition to Systems (2.1) and (2.2) . By designing the appropriate self feedback controller (3.11) and adaptive controller (3.23) , the algebraic discriminant conditions (3.12) and (3.25) to achieve Fin-TS and Fix-TS on Systems (2.1) and (2.2) are established in terms of quaternion direct method, Lyapunov stability theory, extended Cauchy Schwartz inequality, Jensen inequality. Finally, the correctness and validity of Theorems 3.1–3.3 under the orders \mu = 0.53 and \mu = 0.73 are verified by two numerical examples. The dynamics of QDNNs models including impulsive effect and multiple time delays will be our next research issue.
This work was supported by the Natural Science Foundation of Anhui Province of China (No. 1908085MA01) and Research on green prevention and control of tea diseases and insect pests and automatic processing technology based on 5G network (No. 200118).
The authors declare that they have no conflict of interest.
[1] |
K. S. Chiu, T. X. Li, New stability results for bidirectional associative memory neural networks model involving generalized piecewise constant delay, Math. Comput. Simul., 194 (2022), 719–743. https://doi.org/10.1016/j.matcom.2021.12.016 doi: 10.1016/j.matcom.2021.12.016
![]() |
[2] |
A. M. Alimi, C. Aouiti, E. A. Assali, 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
![]() |
[3] |
Z. Yang, D. Wang, X. Sun, J. Wu, Speed sensorless control of a bearingless induction motor with combined neural network and fractional sliding mode, Mechatronics, 82 (2022), 102721. https://doi.org/10.1016/j.mechatronics.2021.102721 doi: 10.1016/j.mechatronics.2021.102721
![]() |
[4] |
H. Zhang, R. Ye, S. Liu, J. Cao, A. Alsaedi, X. Li, LMI-based approach to stability analysis for fractional-order neural networks with discrete and distributed delays, Int. J. Syst. Sci., 49 (2018), 537–545. https://doi.org/10.1080/00207721.2017.1412534 doi: 10.1080/00207721.2017.1412534
![]() |
[5] |
F. Zhang, T. Huang, Q. Wu, Z. Zeng, Multistability of delayed fractional-order competitive neural networks, Neural Network, 140 (2021), 325–335. https://doi.org/10.1016/j.neunet.2021.03.036 doi: 10.1016/j.neunet.2021.03.036
![]() |
[6] |
X. D. Li, D. W.C. Ho, J. Cao, Finite-time stability and settling-time estimation of nonlinear impulsive systems, Automatica, 99 (2019), 361–368. https://doi.org/10.1016/j.automatica.2018.10.024 doi: 10.1016/j.automatica.2018.10.024
![]() |
[7] |
X. D. Li, S. J. Song, J. H. Wu, Exponential stability of nonlinear systems with delayed impulses and applications, IEEE Trans. Automat. Control, 64 (2019), 4024–4034. https://doi.org/10.1109/TAC.2019.2905271 doi: 10.1109/TAC.2019.2905271
![]() |
[8] |
X. D. Li, D. X. Peng, J. Cao, Lyapunov stability for impulsive systems via event-triggered impulsive control, IEEE Trans. Automat. Control, 65 (2020), 4908–4913. https://doi.org/10.1109/TAC.2020.2964558 doi: 10.1109/TAC.2020.2964558
![]() |
[9] |
L. M. Wang, Z. G. Zeng, M. F. Ge, A disturbance rejection framework for finite-time and fixed-time stabilization of delayed memristive neural networks, IEEE Trans. Syst. Man Cybern. Syst., 51 (2019), 905–915. https://doi.org/10.1109/TSMC.2018.2888867 doi: 10.1109/TSMC.2018.2888867
![]() |
[10] |
C. Hu, H. J. Jiang, Special functions-based fixed-time estimation and stabilization for dynamic systems, IEEE Trans. Syst. Man Cybern. Syst., 52 (2022), 3251–3262. https://doi.org/10.1109/TSMC.2021.3062206 doi: 10.1109/TSMC.2021.3062206
![]() |
[11] | W. Yang, Y. W. Wang, I. C. Morarescu, X. K. Liu, Y. H. Huang, Fixed-time synchronization of competitive neural networks with multiple time scales, IEEE Trans. Neural Network Learn. Syst., 2021 (2021). https://doi.org/10.1109/TNNLS.2021.3052868 |
[12] |
Y. Zhao, S. Ren, J. Kurths, Finite-time and fixed-time synchronization for a class of memristor-based competitive neural networks with different time scales, Chaos Solitons Fractals, 148 (2021), 111033. https://doi.org/10.1016/j.chaos.2021.111033 doi: 10.1016/j.chaos.2021.111033
![]() |
[13] | C. J. Xu, Z. X. Liu, C. Aouiti, P. L. Li, L. Y. Yao, J. L. Yan, New exploration on bifurcation for fractional-order quaternion-valued neural networks involving leakage delays, Cogn. Neurodyn., 2022 (2022). https://doi.org/10.1007/s11571-021-09763-1 |
[14] |
T. Li, G. Wang, D. Yu, Q. Ding, Y. Jia, Synchronization mode transitions induced by chaos in modified MorrisLecar neural systems with weak coupling, Nonlinear Dyn., 108 (2022), 2611–2625. https://doi.org/10.1007/s11071-022-07318-5 doi: 10.1007/s11071-022-07318-5
![]() |
[15] |
S. Lin, X. Liu, Synchronization and control for directly coupled reaction-diffusion neural networks with multiple weights and hybrid coupling, Neurocomputing, 487 (2022), 144–156. https://doi.org/10.1016/j.neucom.2022.02.061 doi: 10.1016/j.neucom.2022.02.061
![]() |
[16] | E. Viera-Martin, J. F. Gomez-Aguilar, J. E. Solis-Perez, J. A. Hernandez-Perez, R. F. Escobar-Jimenez, Artificial neural networks: a practical review of applications involving fractional calculus, Eur. Phys. J. Spec. Top., 2022 (2022). https://doi.org/10.1140/epjs/s11734-022-00455-3 |
[17] |
X. He, Z. Hu, Optimization design of fractional-order Chebyshev lowpass filters based on genetic algorithm, Int. J. Circ. Theor. App., 50 (2022), 1420–1441. https://doi.org/10.1002/cta.3224 doi: 10.1002/cta.3224
![]() |
[18] |
X. Zhang, R. Li, J. Y. Hong, X. Zhou, N. Xin, Q. Li, Image-enhanced single-pixel imaging using fractional calculus, Opt. Express, 30 (2022), 81–91. https://doi.org/10.1364/OE.444739 doi: 10.1364/OE.444739
![]() |
[19] | C. J. Xu, Z. X. Liu, C. Aouiti, et al, New exploration on bifurcation for fractional-order quaternion-valued neural networks involving leakage delays, Cogn. Neurodyn., 2022 (2022). https://doi.org/10.1007/s11571-021-09763-1 |
[20] |
S. B. Rao, T. W. Zhang, L. J. Xu, Exponential stability and synchronization of fuzzy Mittag-Leffler discrete-time Cohen-Grossberg neural networks with time delays, Int. J. Syst. Sci., 2022 (2022), 1–23. https://doi.org/10.1080/00207721.2022.2051093 doi: 10.1080/00207721.2022.2051093
![]() |
[21] |
N. Huo, B. Li, Y. K. Li, Global exponential stability and existence of almost periodic solutions in distribution for Clifford-valued stochastic high-order Hopfield neural networks with time-varying delays, AIMS Math., 7 (2022), 3653–3679. https://doi.org/ 10.3934/math.2022202 doi: 10.3934/math.2022202
![]() |
[22] |
H. L. Li, C. Hu, L. Zhang, H. J. Jiang, J. Cao, Non-separation method-based robust finite-time synchronization of uncertain fractional-order quaternion-valued neural networks, Appl. Math. Comput., 409 (2021), 126377. https://doi.org/10.1016/j.amc.2021.126377 doi: 10.1016/j.amc.2021.126377
![]() |
[23] |
W. Zhang, H. Zhang, J. Cao, F. E. Alsaadi, D. Chen, Synchronization in uncertain fractional-order memristive complex-valued neural networks with multiple time delays, Neural Network, 110 (2019), 186–198. https://doi.org/10.1016/j.neunet.2018.12.004 doi: 10.1016/j.neunet.2018.12.004
![]() |
[24] |
J. Pan, Z. Y. Pan, Novel robust stability criteria for uncertain parameter quaternionic neural networks with mixed delays: Whole quaternionic method, Appl. Math. Comput., 407 (2021), 126326. https://doi.org/10.1016/j.amc.2021.126326 doi: 10.1016/j.amc.2021.126326
![]() |
[25] |
W. Wei, J. Yu, L. Wang, C. Hu, H. Jiang, Fixed/Preassigned-time synchronization of quaternion-valued neural networks via pure power-law control, Neural Network, 146 (2022), 341–349. https://doi.org/10.1016/j.neunet.2021.11.023 doi: 10.1016/j.neunet.2021.11.023
![]() |
[26] |
A. Kashkynbayev, A. Issakhanov, M. Otkel, J. Kurths, Finite-time and fixed-time synchronization analysis of shunting inhibitory memristive neural networks with time-varying delays, Chaos Solitons Fractals, 156 (2022), 111866. https://doi.org/10.1016/j.chaos.2022.111866 doi: 10.1016/j.chaos.2022.111866
![]() |
[27] | R. Q. Tang, H. S. Su, Y. Zou, X. S. Yang, Finite-time synchronization of Markovian coupled neural networks with delays via intermittent quantized control: Linear programming approach, IEEE Trans. Neural Network Learn. Syst., 2021 (2021). https://doi.org/10.1109/TNNLS.2021.3069926 |
[28] |
S. Yang, J. Yu, C. Hu, H. J. Jiang, Finite-time synchronization of memristive neural networks with fractional-order, IEEE Trans. Syst. Man Cybern. Syst., 51 (2021), 3739–3750. https://doi.org/10.1109/TSMC.2019.2931046 doi: 10.1109/TSMC.2019.2931046
![]() |
[29] |
T. Q. Hou, J. Yu, C. Hu, H. J. Jiang, Finite-time synchronization of fractional-order complex-variable dynamic networks, IEEE Trans. Syst. Man Cybern. Syst., 51 (2021), 4297–4307. https://doi.org/10.1109/TSMC.2019.2931339 doi: 10.1109/TSMC.2019.2931339
![]() |
[30] |
X. S. Yang, J. Lam, D. W.C. Ho, Z. G. Feng, Fixed-time synchronization of complex networks with impulsive effects via non-chattering control, IEEE Trans. Automat. Contr., 62 (2017), 5511–5521. https://doi.org/10.1109/TAC.2017.2691303 doi: 10.1109/TAC.2017.2691303
![]() |
[31] |
Y. Chen, X. Zhang, Y. Xue, Global exponential synchronization of high-order quaternion Hopfield neural networks with unbounded distributed delays and time-varying discrete delays, Math. Comput. Simul., 193 (2022), 173–189. https://doi.org/10.1016/j.matcom.2021.10.012 doi: 10.1016/j.matcom.2021.10.012
![]() |
[32] | S. Rao, T. Zhang, L. Xu, Exponential stability and synchronization of fuzzy Mittag-Leffler discrete-time Cohen-Grossberg neural networks with time delays, Int. J. Syst. Sci., 2022 (2022). https://doi.org/10.1080/00207721.2022.2051093 |
[33] |
D. Lin, X. Chen, G. Yu, Z. Li, Y. Xia, Global exponential synchronization via nonlinear feedback control for delayed inertial memristor-based quaternion-valued neural networks with impulses, Appl. Math. Comput., 401 (2021), 126093. https://doi.org/10.1016/j.amc.2021.126093 doi: 10.1016/j.amc.2021.126093
![]() |
[34] |
Y. Wang, Y. Tian, X. Li, Global exponential synchronization of interval neural networks with mixed delays via delayed impulsive control, Neurocomputing, 420 (2021), 290–298. https://doi.org/10.1016/j.neucom.2020.09.010 doi: 10.1016/j.neucom.2020.09.010
![]() |
[35] |
A. Kumar, S. Das, V. K. Yadav, Global exponential synchronization of complex-valued recurrent neural networks in presence of uncertainty along with time-varying bounded and unbounded delay terms, Int. J. Dyn. Control, 2021 (2021), 1–15. https://doi.org/10.1007/s40435-021-00838-9 doi: 10.1007/s40435-021-00838-9
![]() |
[36] |
H. Zhang, Y. Cheng, H. M. Zhang, W. Zhang, J. Cao, Hybrid control design for Mittag-Leffler projective synchronization on FOQVNNs with multiple mixed delays and impulsive effects, Math. Comput. Simul., 197 (2022), 341–354. https://doi.org/10.1016/j.matcom.2022.02.022 doi: 10.1016/j.matcom.2022.02.022
![]() |
[37] |
Z. Li, Y. Zhang, The boundedness and the global Mittag-Leffler synchronization of fractional-Order inertial Cohen-Grossberg neural networks with time delays, Neural Process. Lett., 54 (2022), 597–611. https://doi.org/10.1007/s11063-021-10648-x doi: 10.1007/s11063-021-10648-x
![]() |
[38] |
S. Aadhithiyan, R. Raja, J. Alzabut, Q. Zhu, M. Niezabitowski, Robust nonfragile Mittag-Leffler synchronization of fractional order nonlinear complex dynamical networks with constant and infinite distributed delays, Math. Methods Appl. Sci., 45 (2022), 2166–2189. https://doi.org/10.1002/mma.7915 doi: 10.1002/mma.7915
![]() |
[39] | Y. Zhang, Q. Li, Finite-time projective synchronization of neural networks with uncertainties via adaptive fault-tolerant control, in 2021 IEEE 7th International Conference on Control Science and Systems Engineering (ICCSSE), (2021), 29–33. https://doi.org/10.1109/ICCSSE52761.2021.9545093 |
[40] |
W. Chen, Y. Yu, X. Hai, G. Ren, Adaptive quasi-synchronization control of heterogeneous fractional-order coupled neural networks with reaction-diffusion, Appl. Math. Comput., 427 (2022), 127145. https://doi.org/10.1016/j.amc.2022.127145 doi: 10.1016/j.amc.2022.127145
![]() |
[41] | F. Ma, X. Gao, Synchronization and quasi-synchronization of delayed fractional coupled memristive neural networks, Neural Process. Lett., 2022 (2022). https://doi.org/10.1007/s11063-021-10698-1 |
[42] |
L. M. Wang, H. B. He, Z. G. Zeng, Global synchronization of fuzzy memristive neural networks with discrete and distributed delays, IEEE Trans. Fuzzy Syst., 28 (2020), 2022–2034. https://doi.org/10.1109/TFUZZ.2019.2930032 doi: 10.1109/TFUZZ.2019.2930032
![]() |
[43] |
C. Hu, H. B. He, H. J. Jiang, Edge-based adaptive distributed method for synchronization of intermittently coupled spatiotemporal networks, IEEE Trans. Automat. Control, 67 (2022), 2597–2604. https://doi.org/10.1109/TAC.2021.3088805 doi: 10.1109/TAC.2021.3088805
![]() |
[44] |
J. Xiao, J. Cheng, K. Shi, R. Zhang, A general approach to fixed-time synchronization problem for fractional-order multi-dimension-valued fuzzy neural networks based on memristor, IEEE Trans. Fuzzy Syst., 30 (2022), 968–977. https://doi.org/ 10.1109/TFUZZ.2021.3051308 doi: 10.1109/TFUZZ.2021.3051308
![]() |
[45] |
M. Dutta, B. K. Roy, A new memductance-based fractional-order chaotic system and its fixed-time synchronization, Chaos Solitons Fractals, 145 (2021), 110782. https://doi.org/10.1016/j.chaos.2021.110782 doi: 10.1016/j.chaos.2021.110782
![]() |
[46] | W. Zhang, H. Zhang, J. Cao, H. M. Zhang, F. E. Alsaadi, A. Alsaedi, Global projective synchronization in fractional-order quaternion valued neural networks. Asian J. Control, 24 (2022), 227–236. https://doi.org/10.1002/asjc.2485 |
[47] |
H. Zhang, J. S. Cheng, H. M. Zhang, W. W. Zhang, J. Cao, Quasi-uniform synchronization of Caputo type fractional neural networks with leakage and discrete delays, Chaos Solitons Fractals, 152 (2021), 111432. https://doi.org/10.1016/j.chaos.2021.111432 doi: 10.1016/j.chaos.2021.111432
![]() |
[48] | M. S. Ali, M. Hymavathi, S. Senan, V. Shekher, S. Arik, Global asymptotic synchronization of impulsive fractional-order complex-valued memristor-based neural networks with time varying delays. Commun. Nonlinear Sci. Numer. Simul., 78 (2019), 104869. https://doi.org/10.1016/j.cnsns.2019.104869 |
[49] |
H. Li, J. Cao, H. Jiang, A. Alsaedi, Graph theory-based finite-time synchronization of fractional-order complex dynamical networks, J. Franklin Inst., 355 (2018), 5771–5789. https://doi.org/10.1016/j.jfranklin.2018.05.039 doi: 10.1016/j.jfranklin.2018.05.039
![]() |
[50] |
H. Zhang, M. Ye, R. Ye, J. Cao, Synchronization stability of Riemann-Liouville fractional delay-coupled complex neural networks, Phys. A, 508 (2018), 155–165. https://doi.org/10.1016/j.physa.2018.05.060 doi: 10.1016/j.physa.2018.05.060
![]() |
[51] | A. A. Kilbas, H. M. Srivastava, J. J. Trujillo, Theory and Applications of Fractional Differential Equations, Elsevier, Amsterdam, 2006. |
[52] |
W. Chen, H. G. Yu, X. D. Hai, G. J. Ren, Adaptive quasi-synchronization control of heterogeneous fractional-order coupled neural networks with reaction-diffusion, Appl. Math. Comput., 427 (2022), 127145. https://doi.org/10.1016/j.amc.2022.127145 doi: 10.1016/j.amc.2022.127145
![]() |
1. | 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, 2022, 20, 1551-0018, 52, 10.3934/mbe.2023004 |