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

Fractional order COVID-19 model with transmission rout infected through environment

  • Received: 10 September 2021 Revised: 02 November 2021 Accepted: 21 November 2021 Published: 04 January 2022
  • MSC : 37C75, 65L07, 93B05

  • In this paper, we study a fractional order COVID-19 model using different techniques and analysis. The sumudu transform is applied with the environment as a route of infection in society to the proposed fractional-order model. It plays a significant part in issues of medical and engineering as well as its analysis in community. Initially, we present the model formation and its sensitivity analysis. Further, the uniqueness and stability analysis has been made for COVID-19 also used the iterative scheme with fixed point theorem. After using the Adams-Moulton rule to support our results, we examine some results using the fractal fractional operator. Demonstrate the numerical simulations to prove the efficiency of the given techniques. We illustrate the visual depiction of sensitive parameters that reveal the decrease and triumph over the virus within the network. We can reduce the virus by the appropriate recognition of the individuals in community of Saudi Arabia.

    Citation: Shao-Wen Yao, Muhammad Farman, Maryam Amin, Mustafa Inc, Ali Akgül, Aqeel Ahmad. Fractional order COVID-19 model with transmission rout infected through environment[J]. AIMS Mathematics, 2022, 7(4): 5156-5174. doi: 10.3934/math.2022288

    Related Papers:

    [1] Fahad Alsharari, Ahmed O. M. Abubaker, Islam M. Taha . On r-fuzzy soft γ-open sets and fuzzy soft γ-continuous functions with some applications. AIMS Mathematics, 2025, 10(3): 5285-5306. doi: 10.3934/math.2025244
    [2] Jinguo Zhang, Dengyun Yang, Yadong Wu . Existence results for a Kirchhoff-type equation involving fractional p(x)-Laplacian. AIMS Mathematics, 2021, 6(8): 8390-8403. doi: 10.3934/math.2021486
    [3] Cenap Ozel, M. A. Al Shumrani, Aynur Keskin Kaymakci, Choonkil Park, Dong Yun Shin . On δb-open continuous functions. AIMS Mathematics, 2021, 6(3): 2947-2955. doi: 10.3934/math.2021178
    [4] Ibtesam Alshammari, Islam M. Taha . On fuzzy soft β-continuity and β-irresoluteness: some new results. AIMS Mathematics, 2024, 9(5): 11304-11319. doi: 10.3934/math.2024554
    [5] Sang-Eon Han . Semi-topological properties of the Marcus-Wyse topological spaces. AIMS Mathematics, 2022, 7(7): 12742-12759. doi: 10.3934/math.2022705
    [6] Samirah Alzahrani, A. A. Nasef, N. Youns, A. I. EL-Maghrabi, M. S. Badr . Soft topological approaches via soft γ-open sets. AIMS Mathematics, 2022, 7(7): 12144-12153. doi: 10.3934/math.2022675
    [7] Nicky K. Tumalun, Philotheus E. A. Tuerah, Marvel G. Maukar, Anetha L. F. Tilaar, Patricia V. J. Runtu . An application of generalized Morrey spaces to unique continuation property of the quasilinear elliptic equations. AIMS Mathematics, 2023, 8(11): 26007-26020. doi: 10.3934/math.20231325
    [8] Dina Abuzaid, Samer Al-Ghour . Supra soft Omega-open sets and supra soft Omega-regularity. AIMS Mathematics, 2025, 10(3): 6636-6651. doi: 10.3934/math.2025303
    [9] Syafrizal Sy, Rinovia Simanjuntak, Tamaro Nadeak, Kiki Ariyanti Sugeng, Tulus Tulus . Distance antimagic labeling of circulant graphs. AIMS Mathematics, 2024, 9(8): 21177-21188. doi: 10.3934/math.20241028
    [10] Mesfer H. Alqahtani, Alaa M. Abd El-latif . Separation axioms via novel operators in the frame of topological spaces and applications. AIMS Mathematics, 2024, 9(6): 14213-14227. doi: 10.3934/math.2024690
  • In this paper, we study a fractional order COVID-19 model using different techniques and analysis. The sumudu transform is applied with the environment as a route of infection in society to the proposed fractional-order model. It plays a significant part in issues of medical and engineering as well as its analysis in community. Initially, we present the model formation and its sensitivity analysis. Further, the uniqueness and stability analysis has been made for COVID-19 also used the iterative scheme with fixed point theorem. After using the Adams-Moulton rule to support our results, we examine some results using the fractal fractional operator. Demonstrate the numerical simulations to prove the efficiency of the given techniques. We illustrate the visual depiction of sensitive parameters that reveal the decrease and triumph over the virus within the network. We can reduce the virus by the appropriate recognition of the individuals in community of Saudi Arabia.



    Since shunting inhibitory cellular neural networks were proposed by Bouzerdoum and Pinter [1] as a new type of neural networks, they have received more and more attention and have been widely applied in optimisation, psychophysics, speech and other fields. At the same time, since time delays are ubiquitous, many research results have been obtained on the dynamics of shunting inhibitory cellular neural networks with time delays [2,3,4,5,6].

    On the one hand, the quaternion is a generalization of real and complex numbers [7]. The skew field of quaternions is defined by

    H:={q=qR+iqI+jqJ+kqK},

    where qR,qI,qJ,qKR and the elements i,j and k obey the Hamilton's multiplication rules:

    ij=jk=k,jk=kj=i,ki=ik=j,i2=j2=k2=1.

    For q=qR+iqI+jqJ+kqK, we denote ˇq=iqI+jqJ+kqK and qR=qˇq. The norm of q is defined by qH=(qR)2+(qI)2+(qJ)2+(qK)2. For y=(y1,y2,,yn)THn, we define yHn=max1pn{yH}, then (Hn,Hn) is a Banach space. As we all know, quaternion-valued neural networks include real-valued neural networks and complex-valued neural networks as their special cases. Compared with complex-valued neural networks, quaternion-valued neural networks only needs half of the connection weight parameters of complex-valued neural networks when dealing with multi-level information [8]. In recent years, quaternion-valued neural networks have attracted the attention of many researchers, and their various dynamic behaviors, including fractional-order and stochastic quaternion-valued neural networks, have been extensively studied [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24].

    On the other hand, because periodic and almost periodic oscillations are important dynamics of neural networks, the periodic and almost periodic oscillations of neural networks have been studied a lot in the past few decades [25,26,27,28,29,30,31,32,33]. Weyl almost periodicity is a generalization of Bohr almost periodicity and Stepanov almost periodicity [34,35,36,37]. It is a more complex recurrent oscillation. Because the spaces composed of Bohr almost periodic functions and Stepanov almost periodic functions are Banach spaces, it brings some convenience to study the existence of almost periodic solutions in these two senses of differential equations. Therefore, many results have been obtained on the Bohr almost periodic oscillation and Stepanov almost periodic oscillation of neural networks. However, the space composed of Weyl almost periodic functions is incomplete [38]. Therefore, the results of Weyl almost periodic solutions of neural networks are still very rare. Therefore, it is a meaningful and challenging work to study the existence of Weyl almost periodic solutions of neural networks.

    Motivated by the above, in this paper, we consider the following shunting inhibitory cellular neural networks with time-varying delays:

    ˙xij(t)=aij(t)xij(t)CklNr(i,j)Bklij(t)fij(xkl(t))xij(t)CklNs(i,j)Cklij(t)gij(xkl(tτkl(t)))xij(t)+Iij(t), (1.1)

    where ij{11,12,,1n,,m1,m2,,mn}:=Λ, Cij denotes the cell at the (i,j) position of the lattice. The r-neighborhood Nr(i,j) of Cij is given as

    Nr(i,j)={Ckl:max(|ki|,|lj|)r,1km,1ln},

    and Ns(i,j) is similarly specified; xij(t)H denotes the activity of the cell of Cij, Iij(t)H is the external input to Cij, aij(t)H is the coefficient of the leakage term, which represents the passive decay rate of the activity of the cell Cij, Bklij(t)0 and Cklij(t)0 represent the connection or coupling strength of postsynaptic of activity of the cell transmitted to the cell Cij, the activity functions fij,gij:HH are continuous functions representing the output or firing rate of the cell Cij, τkl(t) corresponds to the transmission delay and satisfies 0τkl(t)τ.

    The purpose of this paper is to use the fixed point theorem and a variant of Gronwall inequality to establish the existence and global exponential stability of Weyl almost periodic solutions for a class of quaternion-valued shunting inhibitory cellular neural networks whose coefficients of the leakage terms are quaternions. This is the first paper to study the existence and global exponential stability of Weyl almost periodic solutions of system (1.1) by using the fixed point theorem and a variant of Gronwall inequality. Our result of this paper is new, and our method can be used to study other types of quaternion-valued neural networks.

    For convenience, we introduce the following notations:

    am=minijΛ{inftR{aij(t)}},aM=maxijΛ{suptR{aij(t)}},ˇaMij=suptRˇaij(t)H,
    BklMij=suptR{Bklij(t)},CklMij=suptR{Cklij(t)},τMij=suptR{τij(t)},τ=maxklΛ{suptR{τkl(t)}}.

    The initial condition of system (1.1) is given by

    xij(s)=φij(s),s[τ,0],

    where φijC(R,H),ijΛ.

    Throughout this paper, we assume that:

    (H1) For ij,klΛ, functions aRij,Bklij,CklijAP(R,R+), ˇaijAP(R,H), IijAPWp(R,H), τklAP(R,R+)C1(R,R) and τ<1.

    (H2) For ijΛ, there exist positive constants Lfij and Lgij such that for all x,yH,

    fij(x)fij(y)HLfijxyH,gij(x)gij(y)HLgijxyH,

    and fij(0)=gij(0)=0.

    The rest of this paper is arranged as follows. In Section 2, we introduce some definitions and lemmas. In Section 3, we study the existence and global exponential stability of Weyl almost periodic solutions of (1.1). In Section 4, an example is given to verify the theoretical results. This paper ends with a brief conclusion in Section 5.

    Let (X,X) be a Banach space and BC(R,X) be the set of all bounded continuous functions from R to X.

    Definition 2.1. [38] A function fBC(R,X) is said to be almost periodic, if for every ϵ>0, there exists a constant l=l(ϵ)>0 such that in every interval of length l(ϵ) contains at least one σ such that

    f(t+σ)f(t)X<ϵ,tR.

    Denote by AP(R,X) the set of all such functions.

    For p[1,), we denote by Lploc(R,X) the space of all functions from R into X which are locally p-integrable. For fLploc(R,X), we define the following seminorm:

    fWp=limr+supβR(1rβ+rβf(t)pXdt)1p.

    Definition 2.2. [38] A function fLploc(R,X) is said to be p-th Weyl almost periodic (Wp-almost periodic for short), if for every ϵ>0, there exists a constant l=l(ϵ)>0 such that in every interval of length l(ϵ) contains at least one σ such that

    f(t+σ)f(t)Wp<ϵ.

    This σ is called on ϵ-translation number of f. The set of all such functions will be denoted by APWp(R,X).

    Remark 2.1. By Definitions 2.1 and 2.2, it is easy to see that if fAP(R,X), then fAPWp(R,X).

    Similar to the proofs of the lemma on page 83 and the lemma on page 84 of [39], it is not difficult to prove the following two lemmas.

    Lemma 2.1. If fAPWp(R,X), then f is bounded and uniformly continuous on R withrespect to the seminorn Wp.

    Using the argumentation contained in the proof of Proposition 3.21 in [38], one can easily prove the following.

    Lemma 2.2. If fkAPWp(R,X), k=1,2,,n. Then, for every ϵ>0, there exist common ϵ-translation numbers for these functions.

    Lemma 2.3. [40]Let g:RR be a continuous function such that, for every tR,

    0g(t)ρ(t)+γ1teη1(ts)g(s)ds++γnteηn(ts)g(s)ds (2.1)

    for some locally integrable function ρ:RR, and for some constantsγ1,,γn0, and some constants η1,,ηnγ, whereγ=np=1γp. We assume that the integrals in the right hand side of (2.1)are convergent. Let η=min{η1,,ηn}. Then, for every ξ(0,ηγ]such that 0eξsρ(s)ds converges, we have, for every tR,

    g(t)ρ(t)+γteξ(ts)ρ(s)ds.

    In particular, if ρ(t) is constant, we have

    g(t)ρηηγ.

    Let BUC(R,Hm×n) be a collection of bounded and uniformly continuous functions from R to Hm×n, then, the space BUC(R,Hm×n) with the norm x=suptRx(t)Hm×n is a Banach space, where xBUC(R,Hm×n).

    Denote ϕ0=(ϕ011,,ϕ01n,ϕ021,,ϕ02n,,ϕ0m1,,ϕ0mn)T, where

    ϕ0ij(t)=tetsaRij(v)dvIij(s)ds,ijΛ.

    We will show that ϕ0 is well defined under assumption (H1). In fact, by IijAPWp(R,H) and Lemma 2.1, there exists a constant M>0 such that IijWpM for all ijΛ. According to the Hölder inequality, one has

    ϕ0ij(t)Hteam(ts)Iij(s)dsHr=0(trt(r+1)eamq(ts)ds)1q(trt(r+1)Iij(s)pHds)1pr=0eamrM<+, (3.1)

    where 1p+1q=1, which means that ϕ0 is well defined.

    Take a positive constant αϕ0. Let

    Ω={ϕBUC(R,Hm×n)|ϕϕ0α}.

    Then, for every ϕΩ, one has

    ϕϕϕ0+ϕ02α.

    Theorem 3.1. Assume (H1)(H2) hold. Furthermore, suppose that

    (H3)

    κ=maxijΛ{2am[ˇaMij+CklNr(i,j)2BklMijLfijα+CklNs(i,j)2CklMijLgijα]}<1,

    (H4)for p>2,

    maxijΛ{24(2p4amp)p2(4amp)2[2(ˇaMij)p+2(CklNr(i,j)2BklMijLfijα)p+(1+2ep4amτ1τ)(CklNs(i,j)2CklMijLgijα)p]}<1,

    and, for p=2,

    maxijΛ{24(2am)2[2(ˇaMij)2+2(CklNr(i,j)2BklMijLfijα)2+(1+2e12amτ1τ)(CklNs(i,j)2CklMijLgijα)2]}<1,

    then system (1.1) has a unique Wp-almost periodic solution in Ω.

    Proof. It is easy to check that if x=(x11,,x1n,x21,,x2n,,xm1,,xmn)TΩ is a solution of the integral equation

    xij(t)=tetsaRij(v)dv[ˇaij(s)xij(s)CklNr(i,j)Bklij(s)fij(xkl(s))xij(s)CklNs(i,j)Cklij(s)gij(xkl(sτkl(s)))xij(s)+Iij(s)]ds,ijΛ, (3.2)

    then x is a solution of system (1.1).

    Define an operator T:ΩHm×n by

    (Tϕ)(t)=((T11ϕ)(t),,(T1nϕ)(t),(T21ϕ)(t),,(T2nϕ)(t),,(Tm1ϕ)(t),,(Tmnϕ)(t))T,

    where

    (Tijϕ)(t)=tetsaRij(v)dv[ˇaij(s)ϕij(s)CklNr(i,j)Bklij(s)fij(ϕkl(s))ϕij(s)CklNs(i,j)Cklij(s)gij(ϕkl(sτkl(s)))ϕij(s)+Iij(s)]ds,ijΛ.

    Now, we will prove that Tϕ is well defined. Actually, by (H1)(H3) and (3.1), for ijΛ, one deduces that

    (Tijϕ)(t)HtetsaRij(v)dvˇaij(s)ϕij(s)CklNr(i,j)Bklij(s)(fij(ϕkl(s))fij(0))ϕij(s)CklNs(i,j)Cklij(s)(gij(ϕkl(sτkl(s)))gij(0))ϕij(s)Hds+tetsaRij(v)dvIij(s)Hds1am(ˇaMij+CklNr(i,j)BklMijLfijϕ+CklNs(i,j)CklMijLgijϕ)ϕ+tetsaRij(v)dvIij(s)Hds<+. (3.3)

    That is, Tϕ is well defined.

    We will divide the rest of the proof into four steps.

    Step 1, we will prove that TϕBUC(R,Hm×n), for every ϕΩ.

    In fact, by (3.3), we see that Tϕ is bounded on R. So, we only need to show that Tϕ is uniformly continuous on R. Based on the Hölder inequality for 0h1 and q1 with 1p+1q=1, one has

    (Tijϕ)(t+h)(Tijϕ)(t)H=t+het+hsaRij(v)dv(ˇaij(s)ϕij(s)CklNr(i,j)Bklij(s)fij(ϕkl(s))ϕij(s)CklNs(i,j)Cklij(s)×gij(ϕkl(sτkl(s)))ϕij(s)+Iij(s))dstetsaRij(v)dv(ˇaij(s)ϕij(s)CklNr(i,j)Bklij(s)×fij(ϕkl(s))ϕij(s)CklNs(i,j)Cklij(s)gij(ϕkl(sτkl(s)))ϕij(s)+Iij(s))dsHt|et+hsaRij(v)dvetsaRij(v)dv|ˇaij(s)ϕij(s)+CklNr(i,j)Bklij(s)fij(ϕkl(s))ϕij(s)+CklNs(i,j)Cklij(s)×gij(ϕkl(sτkl(s)))ϕij(s)Hds+t|et+hsaRij(v)dvetsaRij(v)dv|Iij(s)Hds+t+htet+hsaRij(v)dvˇaij(s)ϕij(s)+CklNr(i,j)Bklij(s)fij(ϕkl(s))ϕij(s)+CklNs(i,j)Cklij(s)×gij(ϕkl(sτkl(s)))ϕij(s)Hds+t+htet+hsaRij(v)dvIij(s)Hds(ˇaMij+CklNr(i,j)BklMijLfijϕ+CklNs(i,j)CklMijLgijϕ)ϕt(etsaRij(v)dvt+htaRij(v)dv)ds+t(etsaRij(v)dvt+htaRij(v)dv)Iij(s)Hds+(ˇaMij+CklNr(i,j)BklMijLfijϕ+CklNs(i,j)CklMijLgijϕ)ϕt+htetsaRij(v)dvds+t+htetsaRij(v)dvIij(s)HdsaMh(ˇaMij+CklNr(i,j)BklMijLfijϕ+CklNs(i,j)CklMijLgijϕ)ϕteam(ts)ds+aMhtetsaRij(v)dvIij(s)Hds+(ˇaMij+CklNr(i,j)BklMijLfijϕ+CklNs(i,j)CklMijLgijϕ)ϕ×t+hteam(ts)ds+(t+hteqam(ts)ds)1q(t+htIij(s)pHds)1paMam(ˇaMij+CklNr(i,j)BklMijLfijϕ+CklNs(i,j)CklMijLgijϕ)ϕh+aMhtetsaRij(v)dvIij(s)Hds+eamh(ˇaMij+CklNr(i,j)BklMijLfijϕ+CklNs(i,j)CklMijLgijϕ)ϕh+eamhIijWph,

    where ijΛ. Hence, letting h0+, by (3.1), we have

    (Tijϕ)(t+h)(Tijϕ)(t)H0,

    which means that (Tijϕ) is uniformly continuous on R, ijΛ. Therefore, TϕBUC(R,Hm×n).

    Step 2, we will prove that T is a self-mapping from Ω to Ω.

    Actually, for arbitrary ϕΩ, from (H2)(H3), we have

    Tϕϕ0suptR{maxijΛ[tetsaRij(v)dvˇaij(s)ϕij(s)+CklNr(i,j)Bklij(s)(fij(ϕkl(s))f(0))ϕij(s)+CklNs(i,j)Cklij(s)(gij(ϕkl(sτkl(s)))gij(0))ϕij(s)Hds]}maxijΛ{1am[ˇaMij+CklNr(i,j)2BklMijLfijα+CklNs(i,j)2CklMijLgijα]}ϕκαα,

    which implies that TϕΩ. Consequently, T is a self-mapping from Ω to Ω.

    Step 3, we will prove T is a contraction mapping.

    As a matter of fact, in view of (H1)(H2), for any ϕ,νΩ, we can get

    TϕTνsuptR{maxijΛ[team(ts)(ˇaMijϕij(s)νij(s)H+CklNr(i,j)BklMijfij(ϕkl(s))(ϕij(s)νij(s))+(fij(ϕkl(s))fij(νkl(s)))νij(s)H+CklNs(i,j)CklMijgij(ϕkl(sτkl(s)))(ϕij(s)νij(s))+(gij(ϕkl(sτkl(s)))gij(νkl(sτkl(s))))νij(s)H)ds]}suptR{maxijΛ[team(ts)[ˇaMijϕij(s)νij(s)H+CklNr(i,j)BklMij(2Lfijαϕij(s)νij(s)H+2Lfijαϕkl(s)νkl(s)H)+CklNs(i,j)CklMij(2Lgijαϕij(s)νij(s)H2Lgijαϕkl(sτkl(s))νkl(sτkl(s))H)]ds]}maxijΛ{1am[ˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)4CklMijLgijα]}ϕν.

    From this and (H3), one has

    ||TϕTν||κϕν.

    Noticing that κ<1, T is a contraction mapping. Consequently, system (1.1) has a unique solution x in Ω.

    Step 4, we will prove that the unique solution xΩ is Wp-almost periodic.

    Indeed, since x=(x11,,x1n,x21,,x2n,,xm1,,xmn)TΩ, x is bounded and uniformly continuous. Hence, for every ϵ>0, there exists a δ(0,ϵ) such that for any t1,t2R with |t1t2|<δ and ijΛ, we have

    xij(t1)xij(t2)H<ϵ. (3.4)

    Also, for this δ, in view of (H1) and Lemma 2.2, we see that there exists a common δ-translation number σ such that

    limr+supβR(1rβ+rβIij(t+σ)Iij(t)pHdt)1p<δ<ϵ, (3.5)
    |Bklij(t+σ)Bklij(t)|<ϵ, (3.6)
    |Cklij(t+σ)Cklij(t)|<ϵ, (3.7)
    |aRij(t+σ)aRij(t)|<ϵ,ˇaij(t+σ)ˇaij(t)H<ϵ (3.8)

    and

    |τij(t+σ)τij(t)|<δ, (3.9)

    where ij,klΛ. Consequently, from (3.4) and (3.9), we get

    xij(tτij(t+σ))xij(tτij(t))H<ϵ. (3.10)

    Since x is a solution of system (1.1), by (3.2), for ijΛ, we have

    xij(t+σ)xij(t)HtetsaRij(v+σ)dv(ˇaij(s+σ)xij(s+σ)ˇaij(s)xij(s))dsH+tetsaRij(v+σ)dvCklNr(i,j)[Bklij(s+σ)fij(xkl(s+σ))xij(s+σ)Bklij(s)fij(xkl(s))xij(s)]dsH+tetsaRij(v+σ)dvCklNs(i,j)[Cklij(s+σ)×gij(xkl(s+στkl(s+σ)))xij(s+σ)Cklij(s)gij(xkl(sτkl(s)))xij(s)]dsH+tetsaRij(v+σ)dv[Iij(s+σ)Iij(s)]dsH+t(etsaRij(v+σ)dvetsaRij(v)dv)ˇaij(s)xij(s)dsH+t(etsaRij(v+σ)dvetsaRij(v)dv)CklNr(i,j)Bklij(s)fij(xkl(s))xij(s)dsH+t(etsaRij(v+σ)dvetsaRij(v)dv)CklNs(i,j)Cklij(s)gij(xkl(sτkl(s)))xij(s)dsH+t(etsaRij(v+σ)dvetsaRij(v)dv)Iij(s)dsH:=8l=1Blij(t). (3.11)

    When p>2, it follows from Hölder's inequality (2p,p2p), Hölder's inequality (12,12) and (H2) that

    B2ij(t)team(ts)CklNr(i,j)Bklij(s+σ)fij(xkl(s+σ))(xij(s+σ)xij(s))Hds+team(ts)CklNr(i,j)Bklij(s+σ)(fij(xkl(s+σ))fij(xkl(s)))xij(s)Hds+team(ts)CklNr(i,j)(Bklij(s+σ)Bklij(s))fij(xkl(s))xij(s)Hds(tep2p4am(ts)ds)p2p[tep4am(ts)(CklNr(i,j)2BklMijLfijαxij(s+σ)xij(s))H)p2ds]2p+(tep2p4am(ts)ds)p2p[tep4am(ts)(CklNr(i,j)2BklMijLfijαxkl(s+σ)xkl(s)H)p2ds]2p+(tep2p4am(ts)ds)p2p[tep4am(ts)(CklNr(i,j)4Lfijα2|Bklij(s+σ)Bklij(s)|ds)p2ds]2p(2p4amp)p2p{[tep4am(ts)(CklNr(i,j)2BklMijLfijαxij(s+σ)xij(s))H)p2ds]2p+[tep4am(ts)(CklNr(i,j)2BklMijLfijαxkl(s+σ))xkl(s)H)p2ds]2p+[tep4am(ts)(CklNr(i,j)4Lfijα2|Bklij(s+σ)Bklij(s)|ds)p2ds]2p}(2p4amp)p2p{[(tep4am(ts)ds)12(tep4am(ts)(CklNr(i,j)2BklMijLfijαxij(s+σ)xij(s))H)pds)12]2p+[(tep4am(ts)ds)12(tep4am(ts)(CklNr(i,j)2BklMijLfijα×xkl(s+σ))xkl(s)H)pds)12]2p+[(tep4am(ts)ds)12(tep4am(ts)×(CklNr(i,j)4Lfijα2|Bklij(s+σ)Bklij(s)|)pds)12]2p}(2p4amp)p2p(4amp)1p{[tep4am(ts)(CklNr(i,j)2BklMijLfijαxij(s+σ)xij(s))H)pds]1p+[tep4am(ts)(CklNr(i,j)2BklMijLfijαxkl(s+σ))xkl(s)H)pds]1p+[tep4am(ts)(CklNr(i,j)4Lfijα2|Bklij(s+σ)Bklij(s)|)pds]1p}, (3.12)

    for ijΛ. Similarly, we have

    B1ij(t)(2p4amp)p2p(4amp)1p{ˇaMij[tep4am(ts)xij(s+σ)xij(s)pHds]1p+2α[tep4am(ts)ˇaij(s+σ)˘aij(s)pHds]1p}, (3.13)
    B3ij(t)(2p4amp)p2p(4amp)1p{[tep4am(ts)(CklNs(i,j)2CklMijLgijαxij(s+σ)xij(s))H)pds]1p+[tep4am(ts)(CklNs(i,j)2CklMijLgijαxkl(s+στkl(s+σ))xkl(sτkl(s))H)pds]1p+[tep4am(ts)(CklNs(i,j)4Lgijα2|Cklij(s+σ)Cklij(s))|)pds]1p} (3.14)

    and

    B4ij(t)(2p4amp)p2p(4amp)1p(tep4am(ts)Iij(s+σ)Iij(s)pHds)1p, (3.15)

    for ijΛ.

    Besides, combining with Hölder's inequality (2p,p2p), Hölder's inequality (12,12) and (H2) that

    B5ij(t)2ˇaMijαteam(ts)(ts|aRij(v+σ)aRij(v)|dv)ds2ˇaMijα(tep2p4am(ts)ds)p2p[tep4am(ts)(ts|aRij(v+σ)aRij(v)|dv)p2ds]2p(2p4amp)p2p2ˇaMijα[(tep4am(ts)ds)12(tep4am(ts)(ts|aRij(v+σ)aRij(v)|dv)pds)12]2p(2p4amp)p2p(4amp)1p2ˇaMijα[tep4am(ts)(ts|aRij(v+σ)aRij(v)|dv)pds]1p, (3.16)

    for ijΛ. In a similar way, one can get that

    B6ij(t)(2p4amp)p2p(4amp)1pCklNr(i,j)4BklMijLfijα2[tep4am(ts)(ts|aRij(v+σ)aRij(v)|dv)pds]1p, (3.17)
    B7ij(t)(2p4amp)p2p(4amp)1pCklNs(i,j)4CklMijLgijα2[tep4am(ts)(ts|aRij(v+σ)aRij(v)|dv)pds]1p (3.18)

    and

    B8ij(t)(2p4amp)p2p(4amp)1p[tep4am(ts)(ts|aRij(v+σ)aRij(v)|dv)pIij(s)pHds]1p, (3.19)

    for ijΛ. Hence, together with a change of variables, Fubini's theorem, Hölder's inequality, (3.8) and (3.13), we derive that

    1rβ+rβBp1ij(t)dt(2p4amp)p24amp{1rβ+rβ[[ˇaMijtep4am(ts)xij(s+σ)xij(s)pHds]1p+2α[tep4am(ts)ˇaij(s+σ)˘aij(s)pHds]1p]pdt}(2p4amp)p28amp{(ˇaMij)p1rβ+rβtep4am(ts)xij(s+σ)xij(s)pHdsdt+(2α)p1rβ+rβtep4am(ts)ˇaij(s+σ)˘aij(s)pHdsdt}(2p4amp)p28amp{(ˇaMij)pβep4am(βs)(1rs+rsxij(t+σ)xij(t)pHdt)ds+(2α)pβep4am(βs)(1rs+rsˇaij(t+σ)˘aij(t)pHdt)ds}=(ˇaMij)p(2p4amp)p28ampβep4am(βs)Θσ,r(s)ds+ρ1ij,

    where

    Θσ,r(s):=1rs+rsx(t+σ)x(t)pHndt

    and

    ρ1ij:=2(2p4amp)p2(4amp)2(2αϵ)p, (3.20)

    and, together with a change of variables, Fubini's theorem, Hölder's inequality, (3.6) and (3.12), we derive that

    1rβ+rβBp2ij(t)dt(2p4amp)p24amp{1rβ+rβ[(tep4am(ts)(CklNr(i,j)2BklMijLfijαxij(s+σ)xij(s))H)pds)1p+(tep4am(ts)(CklNr(i,j)2BklMijLfijαxkl(s+σ)xkl(s)H)pds)1p+(tep4am(ts)(CklNr(i,j)4Lfijα2|Bklij(s+σ)Bklij(s)|)pds)1p]pdt}(2p4amp)p212amp{1rβ+rβtep4am(ts)(CklNr(i,j)2BklMijLfijαxij(s+σ)xij(s)H)pdsdt+1rβ+rβtep4am(ts)(CklNr(i,j)2BklMijLfijαxkl(s+σ)xkl(s)H)pdsdt+1rβ+rβtep4am(ts)(CklNr(i,j)4Lfijα2|Bklij(s+σ)Bklij(s)|)pdsdt}(2p4amp)p212amp[(CklNr(i,j)2BklMijLfijα)p1rβ+rβtep4am(ts)x(s+σ)x(s)pHm×ndsdt+(CklNr(i,j)2BklMijLfijα)p1rβ+rβtep4am(ts)x(s+σ)x(s)Hm×ndsdt+1rβ+rβtep4am(ts)(4mnLfijα2ϵ)pdsdt](2p4amp)p212amp[(CklNr(i,j)2BklMijLfijα)pβep4am(βs)(1rs+rsx(t+σ)x(t)pHm×ndt)ds+(CklNr(i,j)2BklMijLfijα)pβep4am(βs)(1rs+rsx(t+σ)x(t)Hm×ndt)ds+4amp(4mnLfijα2ϵ)p]=(2p4amp)p224amp(CklNr(i,j)2BklMijLfijα)pβep4am(βs)Θσ,r(s)ds+ρ2ij,

    where

    ρ2ij:=3(2p4amp)p2(4amp)2(4mnLfijα2ϵ)p. (3.21)

    Moreover, based on a change of variables, Fubini's theorem, Hölder's inequality, (3.7), (3.10) and (3.14), we deduce that

    1rβ+rβBp3ij(t)dt(2p4amp)p24amp{1rβ+rβ[(tep4am(ts)(CklNs(i,j)2CklMijLgijαxij(s+σ)xij(s))H)pds)1p+(tep4am(ts)(CklNs(i,j)2CklMijLgijαxkl(s+στkl(s+σ))xkl(sτkl(s))H)pds)1p+(tep4am(ts)(CklNs(i,j)4Lgijα2(Cklij(s+σ)Cklij(s))H)pds)1p]pdt}(2p4amp)p212amp{1rβ+rβtep4am(ts)(CklNs(i,j)2CklMijLgijαxij(s+σ)xij(s))H)pdsdt+1rβ+rβtep4am(ts)(CklNs(i,j)2CklMijLgijαxkl(s+στkl(s+σ))xkl(sτkl(s))H)pdsdt+1rβ+rβtep4am(ts)(CklNs(i,j)4Lgijα2|Cklij(s+σ)Cklij(s))|)pdsdt}(2p4amp)p212amp{1rβ+rβtep4am(ts)(CklNs(i,j)2CklMijLgijαxij(s+σ)xij(s))H)pdsdt+1rβ+rβ[2tep4am(ts)(CklNs(i,j)2CklMijLgijαxkl(s+στkl(s+σ))xkl(sτkl(s+σ))H)pds+2tep4am(ts)(CklNs(i,j)4CklMijLgijα2xkl(sτkl(s+σ))xkl(sτkl(s))H)pds]dt+1rβ+rβtep4am(ts)(4mnLgijα2ϵ)pdsdt}(2p4amp)p212amp{1rβ+rβtep4am(ts)(CklNs(i,j)2CklMijLgijαxij(s+σ)xij(s))H)pdsdt+1rβ+rβ[21τtτkl(t+σ)ep4am(tuτ)(CklNs(i,j)2CklMijLgijαxkl(u+σ)xkl(u)H)pdu+2(CklNs(i,j)2CklMijLgijαϵ)ptep4am(ts)ds]dt+4amp(4mnLgijα2ϵ)p}(2p4amp)p212amp{1rβ+rβtep4am(ts)(CklNs(i,j)2CklMijLgijαxij(s+σ)xij(s))H)pdsdt+1rβ+rβ[2ep4amτ1τtep4am(ts)(CklNs(i,j)2CklMijLgijαxkl(s+σ)xkl(s)H)pds]dt+8amp(CklNs(i,j)2CklMijLgijαϵ)p+4amp(4mnLgijα2ϵ)p}(2p4amp)p212amp[(CklNs(i,j)2CklMijLgijα)pβep4am(βs)(1rs+rsx(t+σ)x(t))Hm×ndt)ds+2ep4amτ1τ(CklNs(i,j)2CklMijLgijα)pβep4am(βs)(1rs+rsx(t+σ)x(t)pHm×ndt)ds+8amp(CklNs(i,j)2CklMijLgijαϵ)p+4amp(4mnLgijα2ϵ)p]=(2p4amp)p212amp(1+2ep4amτ1τ)(CklNs(i,j)2CklMijLgijα)pβep4am(βs)Θσ,r(s)ds+ρ3ij,

    where

    ρ3ij:=3(2p4amp)p2(4amp)2[2(CklNs(i,j)2CklMijLgijα)p+(4mnLgijα2)p]ϵp. (3.22)

    In view of (3.5), (3.8) and (3.15)–(3.19), we can easily obtain that

    1rβ+rβBp4ij(t)dt(2p4amp)p24ampβep4am(βs)(1rs+rsIij(t+σ)Iij(t)pHdt)ds(2p4amp)p2(4amp)2ϵp:=ρ4ij, (3.23)
    1rβ+rβBp5ij(t)dt(2p4amp)p24amp(2ˇaMijα)p[1rβ+rβtep4am(ts)(ts|aRij(v+σ)aRij(v)|dv)pdsdt](2p4amp)p24amp(2ˇaMijαϵ)p0ep4amsspds:=ρ5ij, (3.24)
    1rβ+rβBp6ij(t)dt(2p4amp)p24amp(CklNr(i,j)4BklMijLfijα2)p[1rβ+rβtep4am(ts)×(ts|aRij(v+σ)aRij(v)|dv)pdsdt](2p4amp)p24amp(CklNr(i,j)4BklMijLfijα2ϵ)p0ep4amsspds:=ρ6ij, (3.25)
    1rβ+rβBp7ij(t)dt(2p4amp)p24amp(CklNs(i,j)4CklMijLgijα2)p[1rβ+rβtep4am(ts)×(ts|aRij(v+σ)aRij(v)|dv)pdsdt](2p4amp)p24amp(CklNs(i,j)4CklMijLgijα2ϵ)p0ep4amsspds:=ρ7ij (3.26)

    and

    1rβ+rβBp8ij(t)dt(2p4amp)p24amp[1rβ+rβtep4am(ts)(ts|aRij(v+σ)aRij(v)|dv)pIij(s)pHdsdt]ϵp(2p4amp)p24amp[1rβ+rβtep4am(ts)(ts)pIij(s)pHdsdt](2p4amp)p24ampIijpWpϵp0ep4amsspds:=ρ8ij, (3.27)

    for ijΛ. Consequently, combining with (3.11) and (3.20)–(3.27), we obtain

    Θσ,r(β)maxijΛ{88l=11rβ+rβBplij(t)dt}ρ+γβeη(βs)Θσ,r(s)ds,

    where η=p4am,

    ρ=86l=1maxijΛ{ρlij}=8maxijΛ{(2p4amp)p24amp[8amp(2α)p+12amp(4mnLfijα2)p+24amp(CklNs(i,j)2CklMijLgijα)P+12amp(4mnLgijα2)p+4amp+[(2ˇaMijα)p+(CklNr(i,j)4BklMijLfijα2)p+(CklNs(i,j)4CklMijLgijα2)p+IijpWp]0ep4amsspds]}ϵp

    and

    γ=8maxijΛ{(2p4amp)p212amp[23(ˇaMij)p+2(CklNr(i,j)2BklMijLfijα)p+(1+2ep4amτ1τ)(CklNs(i,j)2CklMijLgijα)p]}.

    By (H4), we have γ<η. Thus, it follows from Lemma 2.3 that

    1rβ+rβx(t+σ)x(t)pHm×ndtρηηγ.

    Hence, xAPWp(R,Hm×n).

    When p=2, similar to the proof of the case of p>2, one can obtain

    Θσ,r(β)˜ρ+˜γβe˜η(βs)Θσ,r(s)ds,

    where ˜η=am2,

    ˜ρ=88l=1maxijΛ{˜ρlij}=maxijΛ{16am[4am(2α)2+6am(4mnLfijα2ϵ)2+12am(2CklNs(i,j)CklMijLgijα)2+6am(4mnLgijα2)2+2am+[(2ˇaMijα)2+(CklNr(i,j)4BklMijLfijα2)2+(CklNs(i,j)4CklMijLgijα2)2+Iij2W2]0e12amss2ds]}ϵ2

    and

    ˜γ=maxijΛ{48am[23(ˇaMij)2+2(CklNr(i,j)2BklMijLfijα)2+(1+2e12amτ1τ)(CklNs(i,j)2CklMijLgijα)2]}.

    By (H4), we have ˜γ<˜η. Thus, it follows from Lemma 2.3 that

    1rβ+rβx(t+σ)x(t)pHm×ndt˜ρ˜η˜η˜γ,

    which means that xAPW2(R,Hm×n). The proof is complete.

    Definition 3.1. [14] Let x be a solution of system (1.1) with the initial value φ and y be an arbitrary solution of system (1.1) with the initial value ψ, respectively. If there exist positive constants λ and M such that

    x(t)y(t)Hm×nMeλtφψτ,tR+,

    where φψτ=supt[τ,0]φ(t)ψ(t)Hm×n. Then the solution x of system (1.1) is said to be globally exponentially stable.

    Theorem 3.2. Assume that (H1)(H3) hold, then system (1.1) has a unique Wp-almost periodic solution that is globallyexponentially stable.

    Proof. Let x(t) be the Wp-almost periodic solution with the initial value φ(t) and y(t) be an arbitrary solution with the initial value ψ(t). Taking

    zij(t)=xi(t)yij(t),ϕij(t)=φij(t)ψij(t),ijΛ,

    we have

    ˙zij(t)=aij(t)zij(t)CklNr(i,j)Bklij(t)(fij(xkl(t))xij(t)fij(ykl(t))yij(t)CklNs(i,j)Cklij(t)(gij(xkl(tτkl(t)))xij(t)gij(ykl(tτkl(t)))yij(t)),ijΛ. (3.28)

    For ijΛ, we define the following functions:

    Πij(u)=amu(ˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)2CklijLgijαeuτMij+CklNs(i,j)2CklijLgijα).

    From (H3), we get

    Πij(0)=am(ˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)4CklMijLgijα)>0,ijΛ.

    Since Πij(u) is continuous on [0,+) and Πij(u) as u, there exists ζij>0 such that Πij(ζij)=0 and Πij(u)>0, for u(0,ζij), ijΛ. Let ς=minijΛ{ζij}, then we have Πij(ς)0,ijΛ. Hence, we can choose a positive constant λ such that 0<λ<min{ς,am} and Πij(λ)>0. Thus, one has

    1amλ(ˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)2CklijLgijαeλτMij+CklNs(i,j)2CklijLgijα)<1, (3.29)

    where ijΛ. Take a constant M=maxiΛ{amˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)4CklMijLgijα}, then by (H3), we have M>1. Thus,

    1M1amλ(ˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)2CklijLgijαeλτMij+CklNs(i,j)2CklijLgijα)<0,ijΛ. (3.30)

    From (3.28), we have

    ˙zij(t)+aRij(t)zij(t)=ˇaij(t)zij(t)CklNr(i,j)Bklij(t)(fij(xkl(t))xij(t)fij(ykl(t))yij(t))CklNs(i,j)Cklij(t)(gij(xkl(tτkl(t)))xij(t)gij(ykl(tτkl(t)))yij(t)),ijΛ. (3.31)

    Multiplying both sides of (3.31) by et0aRij(v)dv and integrating over [0,t], we have

    zij(t)=ϕij(0)et0aRij(v)dv+t0etsaRij(v)dv[ˇaij(t)zij(s)CklNr(i,j)Bklij(t)(fij(xkl(t))xij(t)fij(ykl(t))yij(t))CklNs(i,j)Cklij(t)(gij(xkl(tτkl(t)))xij(t)gij(ykl(tτkl(t)))yij(t))]ds,ijΛ.

    Hence, for any ϵ>0, it is easy to see that

    z(t)Hm×n<(ϕτ+ϵ)eλt<M(ϕτ+ϵ)eλt,t(τ,0].

    We claim that

    z(t)Hm×n<M(ϕτ+ϵ)eλt,t[0,+). (3.32)

    Otherwise, there exists t>0 such that

    z(t)Hm×n=M(ϕτ+ϵ)eλt (3.33)

    and

    z(t)Hm×n<M(ϕτ+ϵ)eλt,t<t. (3.34)

    From (3.29), (3.30) and (3.34), we get

    zij(t)Hϕij(0)Hetam+t0etam[ˇaMijzij(s)H+CklNr(i,j)Bklij(t)×(fij(xkl(t))(xij(t)yij(t))H+(fij(xkl(t))fij(ykl(t)))yij(t)H)+CklNs(i,j)Cklij(t)(gij(xkl(tτkl(t)))(xij(t)yij(t))H+(gij(xkl(tτkl(t)))gij(ykl(tτkl(t))))yij(t)H)]ds(ϕτ+ϵ)eamt+M(ϕτ+ϵ)t0e(ts)am(ˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)2CklijLgijαeλτMij+CklNs(i,j)2CklijLgijα)eλsds(ϕτ+ϵ)eλte(λam)t+M(ϕτ+ϵ)eλt(1e(λam)t)amλ(ˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)2CklijLgijαeλτMij+CklNs(i,j)2CklijLgijα)=M(ϕτ+ϵ)eλt[e(λam)tM+1e(λam)tamλ(ˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)2CklijLgijαeλτMij+CklNs(i,j)2CklijLgijα)]=M(ϕτ+ϵ)eλt[(1M1amλ(ˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)2CklijLgijαeλτMij+CklNs(i,j)2CklijLgijα)e(λam)t+1amλ(ˇaMij+CklNr(i,j)4BklMijLfijα+CklNs(i,j)2CklijLgijαeλτMij+CklNs(i,j)2CklijLgijα)]<M(ϕτ+ϵ)eλt,ijΛ,

    which contradicts the Eq (3.33). Hence, (3.32) holds. Letting ϵ0+, from (3.32), we have

    z(t)Hm×nMϕτeλt,t>0.

    Therefore, the Wp-almost periodic solution of system (1.1) is globally exponentially stable. This completes the proof.

    Example 4.1. In system (1.1), let i,j=1,2, r=s=1 and take

    xij(t)=xRij(t)+ixIij(t)+jxJij(t)+kxKij(t)H,
    a11(t)=32sin2(3t)+i25cos(2t)j25sin(5t)+k1100cos2(3t),
    a21(t)=39|cos(2t)|i210sin(2t)+j310sin2(7t)+k3100cos(t),
    a12(t)=33cos2(t)i36sin(5t)j14cos3(5t)+k16cos2(3t)
    a22(t)=37sin4(3t)+i14cos(7t)j2916|sin(11t)|+k18cos2(5t),
    τ11(t)=232sin8(2t),τ12(t)=116cos2(3t),τ21(t)=141sin2(25t),τ22(t)=129sin4(37t),
    B11(t)=13|cos(2t)|,B12(t)=16sin(4t)+12B21(t)=34sin(2t)+1,B22(t)=14sin2(5t),
    C11(t)=cos(3t)+3,C12(t)=|sin(5t)|,C21(t)=2sin2(7t),C22(t)=14sin(2t)+74,
    f11(x)=f12(x)=15sin(14xR+36xK)i15|324xI+35xJ|+k14sin(15xR),
    f21(x)=f22(x)=14|23xI+37xK|i28sin(15xR+25xK)+k23sin(15xIj),
    g11(x)=g12(x)=120sin(922xR)i340sin(75xK)+j220sin(13xJ+34xI),
    g21(x)=g22(x)=225|13xJ+23xI|i340sin(75xK+xR)+j121sin(42xR),
    I11(t)=I21(t)=2sint+i43e|t|+k23cos(12t),
    I12(t)=I22(t)=i23sin2tj15cos2t+ke|t|.

    By computing, we obtain

    Lf11=Lf12=310,Lf21=Lf22=16,Lg11=Lg12=920,Lg21=Lg22=821,
    am=32,ˇaM11=12,ˇaM12=25,ˇaM21=512,ˇaM22=716,τ=116,τ=12,
    CklN1(1,1)BklM11=CklN1(1,2)BklM12=CklN1(2,1)BklM21=CklN1(2,2)BklM22=4,
    CklN1(1,1)CklM11=CklN1(1,2)CklM12=CklN1(2,1)CklM21=CklN1(2,2)CklM22=8.

    Choose a constant α=116ϕ0, for i,j=1,2, we obtain that

    κ=maxijΛ{2am[ˇaMij+CklNr(i,j)2BklMijLfijα+CklNs(i,j)2CklMijLgijα]}=0.06875<1.

    For i,j=1,2, take p=3, it is easy to obtain that

    maxijΛ{96(23am)3[23(ˇaMij)3+2(CklNr(i,j)2BklMijLfijα)3+(1+2e34amτ1τ)(CklNs(i,j)2CklMijLgijα)3]}=0.0016<1.

    Take p=2, we have

    maxijΛ{24(2am)2[23(ˇaMij)2+2(CklNr(i,j)2BklMijLfijα)2+(1+2e12amτ1τ)(CklNs(i,j)2CklMijLgijα)2]}=0.2832<1.

    Thus, conditions (H1)(H4) of Theorems 3.1 and Theorems 3.2 are satisfied. Hence, system (1.1) has a unique Wp-almost periodic solution that is globally exponentially stable (see Figures 1, 2).

    Figure 1.  Curves of (xl11(t),xl12(t))T of system (1.1) with the initial values (xl11(0),xl12(0))T=(1,2)T,(3,4)T,(5,5)T,(7,7)T,(9,9)T, l = R, I, J, K.
    Figure 2.  Curves of (xl21(t),xl22(t))T of system (1.1) with the initial values (xl21(0),xl22(0))T=(8,1)T,(9,3)T,(5,5)T,(7,2)T,(9,4)T, l = R, I, J, K.

    Remark 4.1. No known results are available to give the results of Example 4.1.

    In this paper, the existence and global exponential stability of Weyl almost periodic solutions for a class of quaternion-valued neural networks with time-varying delays are established. Even when the system we consider is a real-valued system, our results are brand-new. In addition, the method in this paper can be used to study the existence of Weyl almost periodic solutions for other types of neural networks.

    This work is supported by the National Natural Science Foundation of China under Grant 11861072.

    The authors declare no conflict of interest in this paper.



    [1] I. Podlubny, Fractional differential equations: An introduction to fractional derivatives, fractional differential equations, to methods of their solution and some of their applications, Elsevier, 1998.
    [2] A. Atangana, Non validity of index law in fractional calculus, a fractional differential operator with markovian and non-markovian properties, Physica A, 505 (2018), 688–706. http://dx.doi.org/10.1016/j.physa.2018.03.056 doi: 10.1016/j.physa.2018.03.056
    [3] D. Baleanu, A. Jajarmi, E. Bonyah, M. Hajipour, New aspects of poor nutrition in the life cycle within the fractional calculus, Adv. Differ. Equ., 2018 (2018), 230. http://dx.doi.org/10.1186/s13662-018-1684-x doi: 10.1186/s13662-018-1684-x
    [4] S. Ullah, M. A. Khan, M. Farooq, A fractional model for the dynamics of TB virus, Chaos Soliton. Fract., 116 (2018), 63–71. http://dx.doi.org/10.1016/j.chaos.2018.09.001 doi: 10.1016/j.chaos.2018.09.001
    [5] S. Ullah, M. A. Khan, M. Farooq, Modeling and analysis of the fractional HBV model with atangana-baleanu derivative, Eur. Phys. J. Plus, 133 (2018), 313. http://dx.doi.org/10.1140/epjp/i2018-12120-1 doi: 10.1140/epjp/i2018-12120-1
    [6] Fatmawati, M. A. Khan, M. Azizah, Windarto, S. Ullah, A fractional model for the dynamics of competition between commercial and rural banks in indonesia, Chaos Soliton. Fract., 122 (2019), 32–46. http://dx.doi.org/10.1016/j.chaos.2019.02.009 doi: 10.1016/j.chaos.2019.02.009
    [7] M. Farman, M. U. Saleem, M. F. Tabassum, A. Ahmad, M. O. Ahmad, A linear control of composite model for glucose insulin glucagon, Ain Shams Eng. J., 10 (2019), 867–872. http://dx.doi.org/10.1016/j.asej.2019.04.001 doi: 10.1016/j.asej.2019.04.001
    [8] K. A. Golmankhaneh, C. Tunç, Sumudu transform in fractal calculus, Appl. Math. Comput., 350 (2019), 386–401. http://dx.doi.org/10.1016/j.amc.2019.01.025 doi: 10.1016/j.amc.2019.01.025
    [9] M. Goyal, H. Mehmet Baskonus, A. Prakash, An efficient technique for a time fractional model of lassa hemorrhagic fever spreading in pregnant women, Eur. Phys. J. Plus, 134 (2019), 482. http://dx.doi.org/10.1140/epjp/i2019-12854-0 doi: 10.1140/epjp/i2019-12854-0
    [10] S. Zhao, H. Chen, Modeling the epidemic dynamics and control of covid-19 outbreak in China, Quant. Biol., 8 (2020), 11–19. http://dx.doi.org/10.1007/s40484-020-0199-0 doi: 10.1007/s40484-020-0199-0
    [11] S. Ullah, M. A. Khan, M. Farooq, Z. Hammouch, D. Baleanu, A fractional model for the dynamics of tuberculosis infection using caputo-fabrizio derivative, Discrete Cont. Dyn. Sys. S, 13 (2020), 975–993. http://dx.doi.org/10.3934/dcdss.2020057 doi: 10.3934/dcdss.2020057
    [12] M. A. Khan, A. Atangana, Modeling the dynamics of novel coronavirus (2019-ncov) with fractional derivative, Alex. Eng. J., 59 (2020), 2379–2389. http://dx.doi.org/10.1016/j.aej.2020.02.033 doi: 10.1016/j.aej.2020.02.033
    [13] M. Rahman, M. Arfan, k. Shah, J. F. Gómez-Aguilar, Investigating a nonlinear dynamical model of covid-19 disease under fuzzy Caputo, random and ABC fractional-order derivative, Chaos Soliton. Fract., 140 (2020), 110232. http://dx.doi.org/10.1016/j.chaos.2020.110232 doi: 10.1016/j.chaos.2020.110232
    [14] N. H. Tuan, H. Mohammadi, S. Rezapour, A mathematical model for covid-19 transmission by using the Caputo fractional derivative, Chaos Soliton. Fract., 140 (2020), 110107. http://dx.doi.org/10.1016/j.chaos.2020.110107 doi: 10.1016/j.chaos.2020.110107
    [15] F. S. Alshammari, A mathematical model to investigate the transmission of COVID-19 in the kingdom of Saudi Arabia, Comput. Math. Methods Med., 2020 (2020), 9136157. http://dx.doi.org/10.1155/2020/9136157 doi: 10.1155/2020/9136157
    [16] A. Prakash, M. Goyal, H. M. Baskonus, S. Gupta, A reliable hybrid numerical method for a time dependent vibration model of arbitrary order, AIMS Mathematics, 5 (2020), 979–1000. http://dx.doi.org/10.3934/math.2020068 doi: 10.3934/math.2020068
    [17] C. Tunc, A. K. Golmankhaneh, On stability of a class of second alpha-order fractal differential equations, AIMS Mathematics, 5 (2020), 2126–2142. http://dx.doi.org/10.3934/math.2020141 doi: 10.3934/math.2020141
    [18] M. Hamid, M. Usman, R. U. Haq, W. Wang, A Chelyshkov polynomial based algorithm to analyze the transport dynamics and anomalous diffusion in fractional model, Physica A, 551 (2020), 124227. http://dx.doi.org/10.1016/j.physa.2020.124227 doi: 10.1016/j.physa.2020.124227
    [19] M. Goyal, H. M. Baskonus, A. Prakash, Regarding new positive, bounded and convergent numerical solution of nonlinear time fractional HIV/AIDS transmission model, Chaos Soliton. Fract., 139 (2020), 110096. http://dx.doi.org/10.1016/j.chaos.2020.110096 doi: 10.1016/j.chaos.2020.110096
    [20] M. Hamid, M. Usman, W. Wang, Z. Tian, Hybrid fully spectral linearized scheme for time‐fractional evolutionary equations, Math. Meth. Appl. Sci., 44 (2021), 3890–3912. http://dx.doi.org/10.1002/mma.6996 doi: 10.1002/mma.6996
    [21] M. Farman, M. U. Saleem, A. Ahmad, S. Imtiaz, M. F. Tabassum, S. Akram, et al., A control of glucose level in insulin therapies for the development of artificial pancreas by Atangana Baleanu fractional derivative, Alex. Eng. J., 59 (2020), 2639–2648. http://dx.doi.org/10.1016/j.aej.2020.04.027 doi: 10.1016/j.aej.2020.04.027
    [22] M. U. Saleem, M. Farman, A. Ahmad, E. U. Haque, M. O. Ahmad, A Caputo Fabrizio fractional order model for control of glucose in insulintherapies for diabetes, Ain Shams Eng. J., 11 (2020), 1309–1316. http://dx.doi.org/10.1016/j.asej.2020.03.006 doi: 10.1016/j.asej.2020.03.006
    [23] M. Farman, A. Akgül, A. Ahmad, D. Baleanu, M. U. Saleem, Dynamical transmission of coronavirus model with analysis and simulation, CMES-Comput. Model. Eng. Sci., 127 (2021), 753–769. http://dx.doi.org/10.32604/cmes.2021.014882 doi: 10.32604/cmes.2021.014882
    [24] M. Farman, A. Ahmad, A. Akgül, M. U. Saleem, M. Naeem, D. Baleanu, Epidemiological analysis of the coronavirus disease outbreak with random effects, CMC-Comput. Mater. Con., 67 (2021), 3215–3227. http://dx.doi.org/10.32604/cmc.2021.014006 doi: 10.32604/cmc.2021.014006
    [25] M. Aslam, M. Farman, A. Akgül, M. Sun, Modeling and simulation of fractional order COVID-19 model with quarantined-isolated people, Math. Meth. Appl. Sci., 44 (2021), 6389–6405. http://dx.doi.org/10.1002/mma.7191 doi: 10.1002/mma.7191
    [26] M. Aslam, M. Farman, A. Akgül, A. Ahmad, M. Sun, Generalized form of fractional order COVID-19 model with Mittag-Leffler kernel, Math. Meth. Appl. Sci., 44 (2021), 8598–8614. http://dx.doi.org/10.1002/mma.7286 doi: 10.1002/mma.7286
    [27] M. Bohner, O. Tunç, C. Tunç, Qualitative analysis of caputo fractional integro-differential equations with constant delays, Comp. Appl. Math., 40 (2021), 214. http://dx.doi.org/10.1007/s40314-021-01595-3 doi: 10.1007/s40314-021-01595-3
    [28] V. Padmavathi. A. Prakash, K. Alagesan, N. Magesh, Analysis and numerical simula-tion of novel coronavirus (COVID-19) model with Mittag-Leffler Kernel, Math. Meth. Appl. Sci., 44 (2021), 1863–1877. http://dx.doi.org/10.1002/mma.6886 doi: 10.1002/mma.6886
    [29] M. Usman, M. Hamid, M. Liu, Novel operational matrices-based finite difference/spectral algorithm for a class of time-fractional Burger equation in multidimensions, Chaos Soliton. Fract., 144 (2021), 110701. http://dx.doi.org/10.1016/j.chaos.2021.110701 doi: 10.1016/j.chaos.2021.110701
    [30] A. Prakash, H. Kaur, Analysis and numerical simulation of fractional Biswas–Milovic model, Math. Comput. Simulat., 181 (2021), 298–315. http://dx.doi.org/10.1016/j.matcom.2020.09.016 doi: 10.1016/j.matcom.2020.09.016
    [31] M. Hamid, M. Usman, R. Haq, Z. Tian, A spectral approach to analyze the nonlinear oscillatory fractional-order differential equations, Chaos Soliton. Fract., 146 (2021), 110921. http://dx.doi.org/10.1016/j.chaos.2021.110921 doi: 10.1016/j.chaos.2021.110921
    [32] M. Mehmood, M. Hamid, S. Ashraf, Z. Tian, Galerkin time discretization for transmission dynamics of HBV with non-linear saturated incidence rate, Appl. Math. Comput., 410 (2021), 126481. http://dx.doi.org/10.1016/j.amc.2021.126481 doi: 10.1016/j.amc.2021.126481
    [33] M. Amin, M. Farman, A. Akgül, R. T. Alqahtani, Effect of vaccination to control COVID-19 with fractal fractional operator, Alex. Eng. J., 61 (2022), 3551–3557. http://dx.doi.org/10.1016/j.aej.2021.09.006 doi: 10.1016/j.aej.2021.09.006
    [34] M. Higazy, F. M. Allehiany, E. E. Mahmoud, Numerical study of fractional order COVID-19 pandemic transmission model in context of ABO blood group, Results Phys., 22 (2021), 103852. http://dx.doi.org/10.1016/j.rinp.2021.103852 doi: 10.1016/j.rinp.2021.103852
    [35] M. Caputo, M. Fabrizio, A new definition of fractional derivatives without singular kernel, Prog. Fract. Differ. Appl., 1 (2015), 1–13.
    [36] A. Atangana, D. Baleanu, New fractional derivatives with nonlocal and non-singular kernel theory and application to heat transfer model, Therm. Sci., 20 (2016), 763–769.
    [37] A. Atangana, E. Bonyah, A. A. Elsadany, A fractional order optimal 4D chaotic financial model with Mittag-Leffler law, Chinese J. Phys., 65 (2020), 38–53. http://dx.doi.org/10.1016/j.cjph.2020.02.003 doi: 10.1016/j.cjph.2020.02.003
    [38] E. Alzahrani, M. M. El-Dessoky, D. Baleanu, Mathematical modeling and analysis of the novel Coronavirus using Atangana-Baleanu derivative, Results Phys., 25 (2021), 104240. http://dx.doi.org/10.1016/j.rinp.2021.104240 doi: 10.1016/j.rinp.2021.104240
  • This article has been cited by:

    1. Adnène Arbi, Najeh Tahri, New results on time scales of pseudo Weyl almost periodic solution of delayed QVSICNNs, 2022, 41, 2238-3603, 10.1007/s40314-022-02003-0
    2. Marko Kostić, Weyl ρ-Almost Periodic Functions in General Metric, 2023, 73, 1337-2211, 465, 10.1515/ms-2023-0035
    3. Qi Shao, Yongkun Li, Almost periodic solutions for Clifford-valued stochastic shunting inhibitory cellular neural networks with mixed delays, 2024, 9, 2473-6988, 13439, 10.3934/math.2024655
    4. Adnène Arbi, Najeh Tahri, Synchronization analysis of novel delayed dynamical Clifford-valued neural networks on timescales, 2024, 18, 1748-3018, 10.1177/17483026241241492
  • Reader Comments
  • © 2022 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(2395) PDF downloads(101) Cited by(18)

Figures and Tables

Figures(6)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog