MPE Mathematical Problems in Engineering 1563-5147 1024-123X Hindawi Publishing Corporation 818401 10.1155/2013/818401 818401 Research Article Impulsive Vaccination SEIR Model with Nonlinear Incidence Rate and Time Delay Li Dongmei Gui Chunyu Luo Xuefeng Hu Jun Department of Applied Mathematics Harbin University of Science and Technology Harbin 150080 China hrbust.edu.cn 2013 5 12 2013 2013 12 07 2013 22 10 2013 30 10 2013 2013 Copyright © 2013 Dongmei Li et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This paper aims to discuss the delay epidemic model with vertical transmission, constant input, and nonlinear incidence. Some sufficient conditions are given to guarantee the existence and global attractiveness of the infection-free periodic solution and the uniform persistence of the addressed model with time delay. Finally, a numerical example is given to demonstrate the effectiveness of the proposed results.

1. Introduction

Vaccination has been widely used as a method of disease control; inoculate is an effective approach according to the characteristics of the disease which takes the defense in advance. The implementation of inoculate is not continuous but cyclical. As early as in the 1960s, the principle of the stability has been given in  for the impulsive differential equation. Subsequently, a better definition of stability has been proposed in  for the impulsive differential equation. Motivated by the above work, the attention has been made on the application of the pulse immunization in the infectious disease model. For example, the study has been reported in  where the research has been made about inoculating the pulse vaccination for the people who are easily infected. The researchers have realized that the pulse vaccination strategies can eliminate the positive role to measles; see for example [3, 4]. Moreover, the SIV infection model has been studied in . The inoculation ratio and the inoculation interval time to eliminate the influence of the disease have been pointed out. Accordingly, a huge amount of results has appeared concerning the pulse model to study epidemics of infectious diseases; see for example [6, 7].

On the other hand, the pulse SEIR epidemic model with the incubation period has been established in . The qualitative analysis has been given that the local asymptotic stability of infection-free periodic solution is globally stable. Note that the existence of time delay would degrade the desired performance or even result in the instability . As such, a new class of models with vertical transmission delay pulse infectious disease has been given in . By using the impulsive differential inequality, the sufficient conditions have been presented to guarantee the global attractability of infection-free periodic solution and uniform persistence of the disease. Subsequently, the disease delay pulse models with multiple infectious diseases have been addressed in  and the SIR pulse vaccination infectious disease model with time delay has been discussed in . By constructing an appropriate Lyapunov function, the global attractability of the unique positive periodic solution has been discussed in  for the pulse predator-prey system with distributed delay and proliferated. Also, by using the comparison principle of impulsive differential equations, the influence from the time delay, the pulse vaccination, and the other factors on the nature of the model has been tackled in [16, 17].

An impulsive vaccination SEIR epidemic model with saturation infectious and constant input has been studied in , and the sufficient conditions have been established to ensure the stability and the persistence of the disease-free periodic solution. In , the SEIRS pulse infectious disease model with saturated incidence has been investigated. Sufficient conditions of infection-free periodic solution and disease-lasting conclusion have been obtained. Also, it has been shown that the size of the pulse cycle is an important factor affecting the disease extinction. The SIR epidemic model with nonlinear incidence rate and two class infectious diseases containing the pulse effect has been studied in . Sufficient conditions of the global attractability of the disease-free periodic solutions have been given. It has confirmed that the system is uniformly persistent under a certain condition. Meanwhile, the time delay, pulse vaccination, and nonlinear incidence play important roles in the nature of the model. The conditions have been proposed to control two kinds of diseases. In , the pulse vaccination SIQRS epidemic model with constant input and saturated incidence rate has been addressed. However, it is worth mentioning that the unified model with vertical transmission, constant input, and nonlinear incidence has not been investigated.

In this paper, the cases of constant population input and vertical transmission are considered. Patients who contact susceptible people with saturated incidence way are taken into account. A new impulsive vaccination SEIR epidemic model with time delay and nonlinear incidence rate is established. The sufficient condition is given to guarantee the stability of the disease-free periodic solution and the persistence of the model. Compared to existing results, the main contributions lie in the following aspects: (i) the nonlinear incidence rate is considered in the model to describe the spread of the disease which is more close to reality; (ii) all kinds of infectious diseases have the incubation period, and therefore it is necessary to deal with the phenomenon of time delay; (iii) a unified pulse SEIR epidemic model including the nonlinear contract rate, vertical transmission, and time delays is established. As discussed in , the study of the pulse epidemic model conducted in this paper has analyzed the trend of the disease in the theoretical aspect which will contribute to making the strategy of the disease prevention.

2. Model Establishment

We consider an SEIR model by assuming that the input term has a constant population, a nonlinear occurrence rate as βS(t)I(t)/(1+mIh(t)), and the number of sick of which the sick people who birth to the newborn are qμI(t), the time needed which the lurker transfer into the infected people is τ, so after a time τ, the number of survived lurker which into the infected person is (1)βS(t-τ)I(t-τ)1+mIh(t-τ)e-μτ+qμI(t-τ)e-μτ. The effective coverage of the pulse vaccination needle for the newborn that is not infected is denoted by θ; the vaccination cycle is T. Now we can obtain the following impulsive differential equation model: (2)S(t)=A-βS(t)I(t)1+mIh(t)-μS(t)-qμI(t),E(t)=βS(t)I(t)1+mIh(t)-βS(t-τ)I(t-τ)1+mIh(t-τ)e-μτ+qμI(t)-qμI(t-τ)e-μτ-μE(t),I(t)=βS(t-τ)I(t-τ)1+mIh(t-τ)e-μτ+qμI(t-τ)e-μτ-(μ+γ+α)I(t),R(t)=γI(t)-μR(t),R(t)=5--γItnT,S(t+)=S(t)-θμ(S(t)+E(t)+R(t))-θpμI(t),E(t+)=E(t),I(t+)=I(t),R(t+)=R(t)+θμ(S(t)+E(t)+R(t))+θpμI(t),R(t+)=R(t)+55.θμ(S(t)5+E(t)+R(t))t=nT. Here, S(t), E(t), I(t), and R(t) represent susceptible, lurker, disease, and cure at time t, respectively; μ represents the birth rate (the birth rate is equal to death rate); β represents the effective contact number; α is the mortality due to illness; γ denotes the cure rate; q(0<q<1) represents the infected people who give birth to the newborn and who are vertically infected into “the kind of latent” at time t; p=1-q is the proportion of infected people who gave birth to newborn who are not vertically infected at time t; θ(0<θ<1) denotes the succeeded vaccination proportion of newborn for all those who are not infected; A represents input number of the population of constant; τ denotes the time of lurker who become infected people; T is the pulse vaccination cycle; h, m, A, μ, β, and γ are positive constants. In this paper, we consider the property of the model under S(t)0, E(t)0, I(t)0, and R(t)0, and the initial conditions are given as follows: (3)(φ1(s),φ2(s),φ3(s),φ4(s))C([-τ,0],R+4),φi(0)>0(i=1,2,3,4).

Letting N(t)=S(t)+E(t)+I(t)+R(t) and according to (2), we have (4)N(t)=S(t)+E(t)+I(t)+R(t)=A-μN(t)-αI(t). So (2) can be transformed into the following form: (5)S(t)=A-βS(t)I(t)1+mIh(t)-μS(t)-qμI(t),E(t)=βS(t)I(t)1+mIh(t)-βS(t-τ)I(t-τ)1+mIh(t-τ)e-μτ+qμI(t)-qμI(t-τ)e-μτ-μE(t),I(t)=βS(t-τ)I(t-τ)1+mIh(t-τ)e-μτ+qμI(t-τ)e-μτ-(μ+γ+α)I(t),N(t)=A-μN(t)-αI(t),N(t)=A-2μN(t)tnT,S(t+)=S(t)-θμN(t)+θμqI(t),E(t+)=E(t),I(t+)=I(t),N(t+)=N(t),N(t+1)t=nT.

Noticing N(t)A-μN(t) and by the comparison principle, we have N(t)(A/μ)+e-μt(N(0)-(A/μ)) and then we have limtN(t)(A/μ). Then, all solutions (S(t),E(t),I(t),andN(t)) of model (5) eventually enter and remain in the domain Ω={(S,E,I,N)R+4:NA/μ}. Therefore, Ω is the positive invariant set of (5).

To proceed, we introduce the following two lemmas which will play important roles in the remaining parts of this paper.

Lemma 1 (see [<xref ref-type="bibr" rid="B26">25</xref>]).

Consider the following differential equations with delay: (6)w(t)()p(t)w(t)+q(t),ttk,w(tk+)()dkw(tk)+bk,t=tk,kN, where p(t),q(t)C[R+,R], dk0, and bk are constants. Assume that

the sequence {tk} satisfies 0t1<t2 and limttk=;

wPC[R+,R] and w(t) are left continuous at tk(kN), then (7)w(t)()w(t0)t0<tk<tdkexp(t0tp(s)ds)+t0<tk<t(tk<tj<tdjexp(tktp(s)ds))bk+t0ts<tk<tdkexp(stp(θ)dθ)q(s)ds,tt0.

Lemma 2 (see [<xref ref-type="bibr" rid="B27">26</xref>]).

Considering the following differential equations with delay: (8)x(t)=r1x(t-τ)-r2x(t), where r1, r2, and τ are positive constants, and x(t)>0, one has the following

if r1<r2, then limtx(t)=0,

if r1>r2, then limtx(t)=+,

for all t[-τ,0].

3. Main Results

In this section, for model (2), we aim to propose the sufficient conditions to guarantee the existence of the disease-free periodic solutions, the global stability of disease-free periodic solution, and the uniform persistence of the considered model.

3.1. Existence of the Disease-Free Periodic Solutions

Firstly, the analysis result is given to ensure the existence of the disease-free periodic solutions.

Theorem 3.

Ifθμ/(1-e-μT)<1, (10) has a unique and positive periodic solution (S*(t),N*(t)). Moreover (5) has a unique disease-free periodic solution (S*(t),0,0,N*(t)).

Proof.

The existence of the disease-free periodic solution means that the number of sick people is zero, that is I(t)=0 for all t0. Thus (5) is transformed into (9)S(t)=A-μS(t),E(t)=-μE(t),N(t)=A-μN(t),N(t)-.=AtnT,S(t+)=S(t)-θμN(t),E(t+)=E(t),N(t+)=N(t),N(t+5)t=nT. Noticing that E(t) only appears in the second equation of (9), so we only need to consider the first and third equations of (9), (10)S(t)=A-μS(t),N(t)=A-μN(t),N(t)-.=AtnT,S(t+)=S(t)-θμN(t),N(t+)=N(t),N..(t+)t=nT.

Let N(nT) and S(nT) represent the initial value of N(t) and S(t), at the time t=nT, respectively. For brevity, denote Nn=N(nT), Sn=S(nT). Then, we can be integral in the pulse interval [nT,(n+1)T], respectively, and for the total population and infected people, and we have (11)S(t)=Aμ-(Aμ-S(nT))e-μ(t-nT),N(t)=Aμ-(Aμ-N(nT))e-μ(t-nT).

According to (11), we have the following stroboscopic map: (12)(Sn+1Nn+1)=(e-μT-θμ0e-μT)(SnNn)+(Aμ(1-e-μT)Aμ(1-e-μT)). The Jacobi matrix of (12) is (13)J=(e-μT-θμe-μT0e-μT). The matrix J has the characteristic roots λ1=λ2=e-μT with |λi|=e-μt<1(i=1,2). If θμ/(1-e-μT)<1, that is, (A/μ)-(θA/(1-e-μT))>0, then map (12) has a unique and positive fixed point (S*,N*) with S*=(A/μ)-(θA/(1-e-μT)), N*=A/μ. Therefore, the periodic solution of (10) is (14)S*(t)={Aμ-θA1-e-μTe-μ(t-nT),tnT,Aμ-θA1-e-μT,t=nT,N*(t)=Aμ.

By using the second and fifth equations of (9), we have limtE*(t)=0. Thus the proof of this theorem is complete.

3.2. Global Stability of Disease-Free Periodic Solution

In this subsection, the global stability of the disease-free periodic solution is discussed and the sufficient condition is given accordingly.

Theorem 4.

If R1=max{R1,θμ/(1-e-μt)}<1, then the disease-free periodic solution (S*(t),0,0,A/μ) of (9) is globally attractive, where (15)R1=β((A/μ)-(θμA/(μ+α)(1-e-μT)))+qμeμτ(γ+α+μ).

Proof.

By the first and fifth equations of (5) and N(t)A-(μ+α)N(t), we have N(t)A/(μ+α), and then we have (16)S(t)A-μS(t),tnT,nNS(t+)S(t)-θμAμ+α,t=nT,nN. By Lemma 1, we obtain (17)S(t)S(0+)0<nT<te0t-μds+0<nT<tenTt-μds·(-θμAμ+α)+0ts<nT<test-μdθ·AdsS(0+)e-μt+(-θμAμ+α)e-μteμT(1-eμ[t/T]T)1-eμT+Δ¯, where (18)Δ¯=0ts<nT<test-μdθ·Ads=Aμe-μt0ts<nT<teμsdμs=Aμe-μt0t/Tξ<n<t/TeμTξdμTξ=Aμe-μt[01ξ<n<t/TeμTξdμTξ+12ξ<n<t/TeμTξdμTξ=Aμe-μt++[t/T]t/Tξ<n<t/TeμTξdμTξ]=Aμe-μt[(eμT-1)(1-eμT[t/T])1-eμT+eμt-eμT[t/T]]=Aμ-Aμe-μt. So (19)S(t)S(0+)e-μt+(-θμAμ+α)e-μteμT(1-eμ[t/T]T)1-eμT+Aμ-Aμe-μte-μt[S(0+)-Aμ+θμA(μ+α)(1-e-μT)]+Aμ+θμAeμT([t/T]+1-(t/T))(μ+α)(eμT-1)e-μt[S(0+)-Aμ+θμA(μ+α)(1-e-μT)]+Aμ-θμA(μ+α)(1-e-μT), that is, (20)limsuptS(t)Aμ-μθA(μ+α)(1-e-μT). Noticing θμ/(1-e-μT)<1, we have θμ/(1-e-μT)·μ/(μ+α)<1. Moreover, we obtain (A/μ)-(μθA/(μ+α)(1-e-μT))>0. Then, for all ε>0, there exists n1N such that (21)S(t)Aμ-μθA(μ+α)(1-e-μT)+εη for all tn1T.

Subsequently, it follows from (21) and the third equation of (5) that (22)I(t)βηI(t-τ)1+mIh(t-τ)e-μτ+qμI(t-τ)e-μτ-(μ+α+γ)I(t)βηI(t-τ)e-μτ+qμI(t-τ)e-μτ-(μ+α+γ)I(t)=e-μτ(βη+qμ)I(t-τ)-(μ+α+γ)I(t) for t>n1T+τ.

Consider the comparison system of (22) (23)z(t)=e-μτ(βη+qμ)z(t-τ)-(μ+α+γ)z(t). Due to R1<1+ε, the following inequalities hold: (24)e-μτ[β(Aμ-θμA(μ+α)(1-e-μT)+ε)+qμ]-(γ+α+μ)<0,e-μτ(βη+qμ)<(γ+α+μ) for ε>0. By Lemma 2, we have limtz(t)=0. Note that I(s)=z(s)=φ2(s)>0 for all s[-τ,0]. Then, according to the comparison theorem of differential equation, one gets limtI(t)=0.

Without loss of generality, assuming that there exists t0>0, we have 0<I(t)<ε<A/α for all tt0. By the first and the fifth equations of (5), we have (25)S(t)(A-qμε)-(βε+μ)S(t),tnT,nN,S(t+)S(t)-θA,t=nT,nN. Considering the comparison systems of (25), (26)z1(t)=(A-qμε)-(βε+μ)z1(t),z1(t)=(A--qμε)tnT,nN,z1(t+)=z1(t)-θA,t=nT,nN,z1(0+)=S(0+) when nT<t(n+1)T, we obtain (27)z1*(t)={A-qμεβε+μ-θA1-e-(βε+μ)Te-(βε+μ)(t-nT),tnTA-qμεβε+μ-θA1-e-(βε+μ)T,t=nT, where (28)limε0z1*(t)=S*(t). By the comparison theorem of impulsive differential equation, for ε1>0, there exists T1>t0, when t>T1, and we obtain (29)S(t)>z1*(t)-ε1 for all t>T1. As A-μN-αεN(t), tt0, N(t)(A-αε)/μ>0. Then due to the first and the fifth equations of (5), we obtain (30)S(t)A-μS(t),tnT,nN,S(t+)S(t)-θA+(α+μq)θε,t=nT,nN. Considering the comparison systems of (30), (31)z2(t)=A-μz2(t),tnT,nN,z2(t+)=z2(t)-θA+(α+μq)θε,t=nT,nN,z2(0+)=S(0+), we have (32)z2*(t)={Aμ-θA-(α+μq)θε1-e-μTe-μ(t-nT),tnTAμ-θA-(α+μq)θε1-e-μT,t=nT when nT<t(n+1)T.

Similarly, for ε1>0, there exists T2>t0 such that (33)S(t)<z2¯(t)+ε1         for t>T2. Denote T¯=max{T1,T2}. When t>T¯, let ε10. Then it follows from (29) and (33) that S*(t)-ε1<S(t)<S*(t)+ε1, that is, (34)limtS(t)=S*(t). By substituting I(t)=0 into (5), we have limtE(t)=0, limtN(t)=A/μ. Then, the proof of this theorem is complete.

3.3. Uniform Persistence of the Model

In this subsection, the definition of the uniform persistence is first given. Then the sufficient condition is proposed to ensure the uniform persistence of the addressed model.

Definition 5.

If there exists a compact set DΩ, such that (2) at any periodic solution enters into D and finally remains D under the initial condition of (3), then (2) is uniformly persistent.

According to the above definition, we aim to present the analysis result about the uniform persistence for model (2).

Theorem 6.

If R2=min{R2,R3,R4}>1 then the (2) is uniformly persistent, where (35)R2=(A+qμ)(1-e-μT)+θA(μ+β)-(1+mδh)(α+γ+ξ)qμ(1-e-μT)+θAβ,R3=(1-q)(1-e-(δβ+μ)T)(δβ+μ)θ,δ=Aμ,ξ=μ(1-qe-μτ).

Proof.

It follows from Ω that, we obtain N(t)A/μ=δ, then the solutions of (2) have upper bound. So we only need to consider the lower bound solution of (2).

First, we show the existence of a lower bound for I(t). The third equation of (2) transforms the form as follows: (36)I(t)=[βS(t)1+mIh(t)e-μτ-(α+γ+ξ)]I(t)-e-μτddtt-τt(βS(θ)1+mIh(θ)+qμ)I(θ)dθ. Construct V(t)=I(t)+e-μτt-τt((βS(θ)/(1+mIh(θ)))+qμ)I(θ)dθ. So V(t) is a bounded function. Along the solution of (2), we have (37)V(t)=I(t)+e-μτddtt-τt(βS(θ)1+mIh(θ)+qμ)I(θ)dθ=(α+γ+ξ)[βS(t)1+mIh(t)e-μτ·1α+γ+ξ-1]I(t).

By R2>1, there exists m2*=R2-1>0 satisfying (38)m2*=βe-μτA[1-e-μT-θμ]-(1+mδh)(α+γ+ξ)[(1-e-μT)qμ+θAβ]βe-μτ>0. Then, we have (39)βe-μτ(1+mδh)(α+γ+ξ)[A-qμm2*βm2*+μ-θA1-e-μT]=1. Subsequently, there exists ε1>0, such that (40)βe-μτ(1+mδh)(α+γ+ξ)[A-qμm2*βm2*+μ-θA1-e-(βm2*+μ)T-ε1]>1. So (41)βe-μτ(1+mδh)(α+γ+ξ)Δ>1, where (42)Δ=A-qμm2*βm2*+μ-θA1-e-(βm2*+μ)T-ε1.

Now we prove that I(t)m2* is not true. Otherwise, assume that there exists t0>0, and when t>t0, there is I(t)<m2*. By using the first and the fifth equations of (2), we have (43)S(t)(A-qμm2*)-(βm2*+μ)S(t),S(t)(A-1qμm2*)-tnT,nN,S(t)S(t)-θA,t=nT,nN. By Lemma 1, there exists T1t0+τ, and we have (44)S(t)>A-qμm2*βm2*+μ-θA1-e-(βm2*+μ)T-ε1Δ for t>T1. Together with (37) and (44), one gets (45)V(t)>(γ+ξ)[βe-μτ(1+mδh)(α+γ+ξ)·Δ-1]I(t). Denote Il=mint[T1,T1+τ]I(t). Then tT1>t0, that is, I(t)Il. Otherwise there exists T2>0, such that t[T1,T1+τ+T2]. Then there are I(t)Il, I(T1+τ+T2)=Il, and I(T1+τ+T2)0. It follows from the second equation of (2) and (41) that (46)I(T1+τ+T2)(α+γ+ξ)[βe-μτ(1+mδh)(α+γ+ξ)Δ-1]Il>0, which is a contradiction. So, for tT1, according to (36), I(t)Il, V(t)>0, that is limtV(t)=+. This is, a contradiction of which V(t) is bounded. So when t>t0, I(t)<m2* is not tenable.

Two cases to prove are given as follows.

Case  1. If, for t which is sufficiently large, I(t)m2*, hence the statements are proved.

Case  2. If I(t) is shocked near m2*, let m2=min{m2*/2,m2*e-(α+γ+μ)τ}. There exists two constants: t¯ and ϖ satisfying I(t¯)=I(t¯+ϖ)=m2*, and when t¯<t<t¯+ϖ, there is I(t)<m2*. Because I(t) is a continuous and bounded function which is not affected by pulse, then I(t)is uniformly continuous. So, there exists 0<T3<τ (T3 depends on the selection of t¯), met for all t¯<t<t¯+T3, and there is I(t)>m2*/2. If ϖ<T3. The conclusion was established. If T3<ϖ<τ, then, by the second equation of (5) that we have when t¯<t<t¯+ϖ, I(t)-(α+γ+μ)I(t); again by I(t¯)=m2*, there is I(t)m2*e-(α+γ+μ)τ, and obviously I(t)m2 is established. Similarly, if ϖτ, then, by the second equation of (5), when t¯<t<t¯+τ, I(t)m2. Then proof is as follows, when t¯+τtt¯+ϖ, and there is still I(t)m2. If not, there exists T¯0, when t¯tt¯+τ+T¯, and there are I(t)m2, I(t¯+τ+T¯)=m2, and I(t¯+τ+T¯)0. It follows from the second equation of (2) that (47)I(t¯+τ+T¯)[βe-μτ(1+mδh)(α+γ+ξ)Δ-1]m2>0. It is a contradiction. So when t¯<t<t¯+ϖ, there is I(t)m2.

Above all, any positive periodic solution of (2) which for sufficiently large t, it can be concluded that I(t)m2.

It follows from the first and the fifth equations of (2) that (48)S(t)(1-q)A-(δβ+μ)S(t),tnT,nN,S(t+)S(t)-θA,t=nT,nN. It follows from Lemma 1 that (49)S(t)e-(δβ+μ)t[S(0+)-(1-q)Aδβ+μ+θA1-e-(δβ+μ)T]+(1-q)Aδβ+μ-θAe(δβ+μ)(T[t/T]+T-t)e(δβ+μ)T-1. Since 0<T[t/T]+T-t<T, so (50)S(t)e-(δβ+μ)t[S(0+)-(1-q)Aδβ+μ+θA1-e-(δβ+μ)T]+(1-q)Aδβ+μ-θA1-e-(δβ+μ)T. Thus limtS(t)A[((1-q)/(δβ+μ))-(θ/(1-e-(δβ+μ)T))]; it follows from R3>1 that ((1-q)/(δβ+μ))-(θ/(1-e-(δβ+μ)T))>0, that is; for ε>0, there exists T4>0, when t>T4, and there is (51)S(t)A[1-qδβ+μ-θ1-e-(δβ+μ)T]-εm1.

Because of the second equation of (2) (52)E(t)=t-τtβS(u)I(u)1+mIh(u)e-μ(t-u)du+t-τtqμI(u)e-μ(t-u)du(βm1m2μ(1+mδh)+qm2)(1-e-μτ)m3.

Subsequently, it follows from the fourth equation of (2) that (53)R(t)rm2-μR(t). Similarly, there is R(t)rm2/μm4 satisfying (54)D-={(S,E,I,R)m1S(t)δ,m2I(t)δ,-=m3E(t)δ,m4R(t)δ}. Then, D¯ is a bounded compact set, nontrivial for any cycle solution of (2) to enter and stay in D¯ inside. Hence, it completes the proof of this theorem.

Remark 7.

Up till now, we investigate the delay epidemic model with vertical transmission, constant input and nonlinear incidence. The sufficient conditions are given to guarantee the existence of the disease-free periodic solutions, the global stability of disease-free periodic solution, and the uniform persistence of the considered model. It is worth mentioning that, according to the statistics of the suspected patients and patients and using the data identification approach, the parameters in the model can be determined when the disease outbreaks. Subsequently, the illness trend of the epidemic can be predicted. Hence, we can make the reasonable control measures. One of the future research directions would be to apply the developed results to make the control strategy by properly considering the real information on the epidemic data.

4. A Numerical Simulation

For comparisons, we consider the following five cases.

Case 1.

Let the parameters be specified as certain fixed values in (2): A=0.2, μ=0.02, β=0.05, q=0.8, γ=0.5, α=0.3, θ=0.8, τ=10, T=20, and m=h=1; then the result R1=max{0.5137,0.04853}<1 is obtained after calculation. By Theorem 4, we know that the disease-free periodic solution of (2) is globally attractive. Setting the initial value S(0)=1, E(0)=2, I(0)=3, and R(0)=4, the numerical simulation of the diagram is shown in Figure 1.

The global attractability of disease-free periodic solution.

Case 2.

Take the following parameters: τ=1, α=0.01, A=0.5,β=0.12, μ=0.1, q=0.8, m=0.02, h=0.8, γ=0.06, T=10, p=1-q, and θ=0.1; we calculate that R2=min{5.2830,2.8545}>1. By Theorem 6, (2) is uniform persistence. Letting the initial S(0)=0.5, E(0)=1, I(0)=1.5, R(0)=2, the numerical simulation of (2) is shown in Figure 2.

The uniform persistence of disease when β=0.12, q=0.8, and θ=0.1.

Case 3.

The parameter of (2) is the same as that in Figure 1 except for parameter β. Taking β=0.08, we have R2=min{5.4400,3.9730}>1. Then, by Theorem 6, (2) is of uniform persistence. The corresponding numerical simulation of (2) is shown in Figure 3.

The uniform persistence of disease when β=0.08.

Case 4.

If the parameter of (2) is the same as in Figure 1 except q. Taking q=0.1, we obtain R2=min{20.1653,12.8454}>1. By Theorem 6, (2) is of uniform persistence. Accordingly, the simulation of (2) is shown in Figure 4.

The uniform persistence of disease when q=0.1.

Case 5.

If the parameter of (2) is the same as in Figure 1 except θ. Taking θ=0.9, we have R2=min{4.9835,1.2496}>1. By Theorem 6, (2) is of uniform persistence. The numerical simulation of (2) is shown in Figure 5.

It can be seen that I(t) is more influenced by the change of β, p but is less influenced by the change of θ.

The uniform persistence of disease when θ=0.9.

By Figures 2 and 3, we can see that when the contact rate is smaller, the numbers of lurker and infective are reduced, but the numbers of susceptible and removed accordingly changed much. Also, there is a proportional relationship between the size of the extent of epidemic and the contact rate. By comparing Figures 2 and 4, when the vertical transmission rate is bigger, the number of infective is increased by controlling the vertical transmission rate of infected newborn. Moreover, by comparing Figures 2 and 5, the increase of vaccination rate has a certain effect for the control of the disease, but it is not obvious. Therefore, if we reduce the extent of the epidemic, the measure of one is reducing the contact rate and vertical transmission rate. From the simulation the feasibility and usefulness of the proposed main results are confirmed.

5. Conclusions

In this paper, we have discussed the SEIR model with the pulse vaccination, the constant input item of population, and the vertical transmission. By employing the impulsive differential inequality and the stroboscopic map, the existence conditions of the disease-free periodic solution of the model have been given. Also, the sufficient conditions of globally attractive and uniform persistence have been proposed. Finally, a numerical example has been given to illustrate the validity of the proposed results. One of our future research interests is to extend the main results of the analysis and synthesis of gene regulatory networks or complex dynamical systems as discussed in .

Acknowledgments

This work was supported in part by the National Natural Science Foundation of Heilongjiang Province under Grant no. A200502 and the Foundation of Educational Commission of Heilongjiang Province under Grant no. 12521099.

Mil'man V. D. Myškis A. D. On the stability of motion in the presence of impulses Sibirskii Matematicheskii Journal 1960 1 2 233 237 MR0126028 Perestyuk A. N. Samoilenko A. M. Stability of solutions of differential equations with impulse effect Different Equation 1977 13 1 1981 1992 Agur Z. Cojocaru L. Mazor G. Anderson R. M. Danon Y. L. Pulse mass measles vaccination across age cohorts Proceedings of the National Academy of Sciences of the United States of America 1993 90 24 11698 11702 2-s2.0-0027136227 10.1073/pnas.90.24.11698 Gao S. Chen L. Teng Z. Impulsive vaccination of an SEIRS model with time delay and varying total population size Bulletin of Mathematical Biology 2007 69 2 731 745 2-s2.0-33847258383 10.1007/s11538-006-9149-x ZBL1139.92314 Liu H. Yu J. Zhu G. Global asymptotic stable eradication for the SIV epidemic model with impulsive vaccination and infection-age Journal of Systems Science & Complexity 2006 19 3 393 402 10.1007/s11424-006-0393-9 MR2285496 ZBL1114.35031 d'Onofrio A. Mixed pulse vaccination strategy in epidemic model with realistically distributed infectious and latent times Applied Mathematics and Computation 2004 151 1 181 187 10.1016/S0096-3003(03)00331-X MR2037959 ZBL1043.92033 Meng X.-Z. Chen L.-S. Song Z.-T. Global dynamics behaviors for new delay SEIR epidemic disease model with vertical transmission and pulse vaccination Applied Mathematics and Mechanics (English Edition) 2007 28 9 1259 1271 10.1007/s10483-007-0914-x MR2354435 ZBL1231.92057 d'Onofrio A. Stability properties of pulse vaccination strategy in SEIR epidemic model Mathematical Biosciences 2002 179 1 57 72 10.1016/S0025-5564(02)00095-0 MR1908736 ZBL0991.92025 Hu J. Wang Z. Shen B. Gao H. Gain-constrained recursive filtering with stochastic nonlinearities and probabilistic sensor delays IEEE Transactions on Signal Processing 2013 61 5 1230 1238 10.1109/TSP.2012.2232660 MR3028847 Hu J. Wang Z. Gao H. Stergioulas L. K. Robust sliding mode control for discrete stochastic systems with mixed time delays, randomly occurring uncertainties, and randomly occurring nonlinearities IEEE Transactions on Industrial Electronics 2012 59 7 3008 3015 2-s2.0-84857382236 10.1109/TIE.2011.2168791 Hu J. Wang Z. Niu Y. Stergioulas L. K. H sliding mode observer design for a class of nonlinear discrete time-delay systems: a delay-fractioning approach International Journal of Robust and Nonlinear Control 2012 22 16 1806 1826 10.1002/rnc.1785 MR2983311 ZBL06221561 Hu J. Wang Z. Gao H. Stergioulas L. K. Robust H sliding mode control for discrete time-delay systems with stochastic nonlinearities Journal of the Franklin Institute 2012 349 4 1459 1479 10.1016/j.jfranklin.2011.05.018 MR2909497 ZBL1254.93043 Zhang X.-B. Huo H.-F. Sun X.-K. Fu Q. The differential susceptibility SIR epidemic model with time delay and pulse vaccination Journal of Applied Mathematics and Computing 2010 34 1-2 287 298 10.1007/s12190-009-0321-y MR2718786 ZBL1213.34104 Du Y. Xu R. A delayed SIR epidemic model with nonlinear incidence rate and pulse vaccination Journal of Applied Mathematics & Informatics 2010 28 5-6 1089 1099 MR2777935 Zhao Z. Li Z. Chen L. Existence and global stability of periodic solution for impulsive predator-prey model with diffusion and distributed delay Journal of Applied Mathematics and Computing 2010 33 1-2 389 410 10.1007/s12190-009-0292-z MR2647528 ZBL1219.34106 Huang C. An X. An impulsive vaccination SEIRS epidemic model with multi-delay a nonlinear incidence rate Journal of Lanzhou University of Technology 2011 37 1 121 125 Du Y. Xu R. Pulse vaccination of a SEIRS epidemic model with time delays a nonlinear incidence rate Journal of Beihua University (Natural Science) 2011 12 3 258 264 Gao S. J. Teng Z. D. Impulsive vaccination of an SIRS model with saturation infectious force and constant recruitment Journal of Biomathematics 2008 23 2 209 217 MR2445440 Wei C. Chen L. A delayed epidemic model with pulse vaccination Discrete Dynamics in Nature and Society 2008 2008 12 746951 10.1155/2008/746951 MR2403197 ZBL1149.92329 Meng X. Li Z. Wang X. Dynamics of a novel nonlinear SIR model with double epidemic hypothesis and impulsive effects Nonlinear Dynamics 2010 59 3 503 513 10.1007/s11071-009-9557-1 MR2592933 ZBL1183.92072 Xu W. J. The SIQRS epidemic model of impulsive vaccination with constant input and saturation incidence rate Journal of Systems Science and Mathematical Sciences 2010 30 1 43 52 MR2639793 Meng X.-Z. Chen L.-S. Song Z.-T. Global dynamics behaviors for new delay SEIR epidemic disease model with vertical transmission and pulse vaccination Applied Mathematics and Mechanics (English Edition) 2007 28 9 1259 1271 10.1007/s10483-007-0914-x MR2354435 ZBL1231.92057 Pei Y. Z. Liu S. Y. Li C. G. Gao S. J. A pulse vaccination epidemic model with multi-delay and vertical transmission Chinese Annals of Mathematics A 2009 30 5 669 676 MR2582196 Zhu J. Li W. Zhu L. An SIR epidemic model with birth pulse and pulse vaccination Journal of Biomathematics 2011 26 3 490 496 Lakshmikantham V. Baĭnov D. D. Simeonov P. S. Theory of Impulsive Differential Equations 1989 River Edge, NJ, USA World Scientific Publishing MR1082551 Yang K. Delay Differential Equations with Applications in Population Dynamics 1993 San Diego, Calif, USA Academic Press MR1218880 Hu J. Wang Z. Gao H. Stergioulas L. K. Extended Kalman filtering with stochastic nonlinearities and multiple missing measurements Automatica 2012 48 9 2007 2015 10.1016/j.automatica.2012.03.027 MR2956877 ZBL1257.93099 Wang Z. Wu H. Liang J. Liu X. On modeling and state estimation for genetic regulatory networks with polytopic uncertainties IEEE Transactions on NanoBioscience 2013 12 1 13 20 10.1109/TNB.2012.2215626 Hu J. Wang Z. Gao H. Stergioulas L. K. Probability-guaranteed H finite-horizon filtering for a class of nonlinear time-varying systems with sensor saturations Systems & Control Letters 2012 61 4 477 484 10.1016/j.sysconle.2012.01.005 MR2910322 ZBL1250.93121 Wang Z. Wang Y. Liu Y. Global synchronization for discrete-time stochastic complex networks with randomly occurred nonlinearities and mixed time delays IEEE Transactions on Neural Networks 2010 21 1 11 25 2-s2.0-73949136715 10.1109/TNN.2009.2033599 Hu J. Wang Z. Shen B. Gao H. Quantised recursive filtering for a class of nonlinear systems with multiplicative noises and missing measurements International Journal of Control 2013 86 4 650 663 10.1080/00207179.2012.756149 MR3046931 Wang Z. Liu X. Liu Y. Liang J. Vinciotti V. An extended kalman filtering approach to modeling nonlinear dynamic gene regulatory networks via short gene expression time series IEEE/ACM Transactions on Computational Biology and Bioinformatics 2009 6 3 410 419 2-s2.0-68649105450 10.1109/TCBB.2009.5