Message Board

Dear readers, authors and reviewers,you can add a message on this page. We will reply to you as soon as possible!

2020 Volume 45 Issue 5
Article Contents

Jia-li XING, Ai-min ZHAO, Gui-rong LIU. Qualitative Analysis of Stochastic SIRI Model with Lévy Noise and Media Reports[J]. Journal of Southwest China Normal University(Natural Science Edition), 2020, 45(5): 39-44. doi: 10.13718/j.cnki.xsxb.2020.05.007
Citation: Jia-li XING, Ai-min ZHAO, Gui-rong LIU. Qualitative Analysis of Stochastic SIRI Model with Lévy Noise and Media Reports[J]. Journal of Southwest China Normal University(Natural Science Edition), 2020, 45(5): 39-44. doi: 10.13718/j.cnki.xsxb.2020.05.007

Qualitative Analysis of Stochastic SIRI Model with Lévy Noise and Media Reports

More Information
  • Corresponding author: Gui-rong LIU
  • Received Date: 18/04/2019
    Available Online: 20/05/2020
  • MSC: O211.63

  • Consider a stochastic SIRI model with Lévy noise and media coverage. In the Lyapunov function method and the Itô formula, existence and uniqueness of the global positive solution of model have been given. Then the asymptotic of the solution of this model have been studied around disease-free equilibrium and endemic equilibrium of the corresponding deterministic model. Finally, the theoretical results of this paper have been verified by numerical simulation.
  • 加载中
  • [1] ZHANG J, JIN Z, SUN G Q, et al.Analysis of Rabies in China:Transmission Dynamics and Control[J].Plos One, 2011, 6(7):e20891. doi: 10.1371/journal.pone.0020891

    CrossRef Google Scholar

    [2] MA Z, ZHOU Y, WU J.Modeling and Dynamics of Infectious Diseases[M].Singapore:World Scientific Publishing, 2009.

    Google Scholar

    [3] 舒梦诗, 王稳地, 姚苗然, 等.具有饱和免疫作用的病毒动力学模型稳定性分析[J].西南大学学报(自然科学版), 2017, 39(3):102-107.

    Google Scholar

    [4] SALLY B.Modelling the Genital Herpes Epidemic[J].Herpes the Journal of the Ihmf, 2004, 11(suppl3):138A-146A.

    Google Scholar

    [5] TUDOR D.A Deterministic Model for Herpes Infections in Human and Animal Populations[J].SIAM Review, 1990, 32(1):136-139. doi: 10.1137/1032003

    CrossRef Google Scholar

    [6] TOMAS C, MOHAMED E F.A Stochastic SIRI Epidemic Model with Relapse and Media Coverage[J].Discrete and Continuous Dynamical Systems Series B, 2018, 23:3483-3501. doi: 10.3934/dcdsb.2018250

    CrossRef Google Scholar

    [7] MAO X R.Stochastic Differential Equations and Applications[M].Chichester:Horwood Pub, 2007.

    Google Scholar

    [8] LIU Y, CUI J A.The Impact of Media Coverge on the Dynamics of Infectious Disease[J].International Journal of Biomathematics, 2008, 1(1):65-74. doi: 10.1142/S1793524508000023

    CrossRef Google Scholar

    [9] KLOEADEN P E, Pearson R A.The Numerical Solution of Stochastic Differential Equations[J].The ANZIAM Journal, 1977, 20(1):8-12.

    Google Scholar

    [10] HIGHAM D J.An Algorithmic Introduction to Numerical Simulation of Stochastic Differential Equations[J].SIAM Rev, 2001, 43(3):525-546. doi: 10.1137/S0036144500378302

    CrossRef Google Scholar

  • 加载中
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

Figures(2)

Article Metrics

Article views(1709) PDF downloads(388) Cited by(0)

Access History

Other Articles By Authors

Qualitative Analysis of Stochastic SIRI Model with Lévy Noise and Media Reports

    Corresponding author: Gui-rong LIU

Abstract: Consider a stochastic SIRI model with Lévy noise and media coverage. In the Lyapunov function method and the Itô formula, existence and uniqueness of the global positive solution of model have been given. Then the asymptotic of the solution of this model have been studied around disease-free equilibrium and endemic equilibrium of the corresponding deterministic model. Finally, the theoretical results of this paper have been verified by numerical simulation.

1.   模型建立
  • 近年来,各种传染病严重威胁着人类的健康,数学模型被认为是研究各种传染病动态的有力工具[1-3].一些传染病暂时或永久免疫,用SIR,SIRS模型来描述;其他一些传染病,康复后的个体可能恢复成感染者,用SIRI模型来描述,如肺结核、疱疹[4-5].最近,各种各样的数学模型都开始考虑媒体报道的影响.文献[5]提出下列带有媒体报道的随机SIRI模型:

    其中:S(t)表示t时刻易感者的数量,I(t)表示t时刻感染者的数量,R(t)表示t时刻恢复者的数量;Λ表示单位时刻补充的数量,μ表示死亡率,α表示恢复率,γ为免疫失去率,Λμγαλm都为正常数;β1是媒体报道前的接触率,β2是媒体报道后的接触率,由于报道不能完全阻止疾病的传播,所以β1β2>0;Bi(t)表示有滤波${\left\{ {{\mathscr{F}_t}} \right\}_{t \ge 0}}$的完备概率空间(Ω$\mathscr{F}$${\left\{ {{\mathscr{F}_t}} \right\}_{t \ge 0}}$$\mathbb{P}$)上的标准布朗运动;σi表示白噪声强度,i=1,2,3.

    现实中,不仅传染病的传播会受到环境噪声的干扰,而且在传染病传播的过程中可能受到严重的环境扰动,如非典、洪水等.这些现象不可能用随机连续模型描述,进而引入Lévy跳来刻画这些现象.此外本文用更一般的函数(β1-β2f(I))SI来描述媒体报道的影响.

    基于上面两点,在文献[6]的基础上,考虑带有Lévy噪声和媒体报道的随机SIRI模型:

    其中:f(I)满足f′(I)≥0,f(0)=0,$\underset{I+\infty }{\mathop{\text{lim}}}\, \text{ }f\left(I \right)=1$X(t-)表示X(t)在t处的左极限;N(dt,dy)是有平稳分布v(dy)dt的泊松计数过程且$\widetilde{N}$(dt,dy)=N(dt,dy)-v(dy)dt${{\int }_{Y}}[\text{ln}(1+{{q}_{i}}\left(y \right))]v\left(\text{d}y \right)>-\infty $${{\int }_{Y}}[{{q}_{i}}\left(y \right)-\text{ln}(1+{{q}_{i}}\left(y \right))]v\left(\text{d}y \right) < +\infty $i=1,2,3;v定义在[0,∞)的可测子集Y上,且v(Y) < ∞;qiY×Ω (-1,+∞)上的连续有界函数,i=1,2,3.模型(2)中其他参数的定义与模型(1)中相同.

2.   全局正解的存在唯一性
  • 定理1 对于任意初值(S(0),I(0),R(0))T$\mathbb{R}_{+}^{3}$,模型(2)存在唯一的全局正解(S(t),I(t),R(t))T,即对任意t∈[0,∞),(S(t),I(t),R(t))T$\mathbb{R}_{+}^{3}$a. s.

     模型(2)的系数满足局部Lipschitz条件,对于任意的初值(S(0),I(0),R(0))T$\mathbb{R}_{+}^{3}$,模型(2)在区间[0, τe)存在唯一局部解x(t)=(S(t),I(t),R(t))T,其中τe是爆破时刻.存在充分大的正整数k0,使得S(0),I(0)和R(0)都在区间$\left(\frac{1}{{{k}_{0}}}~, {{k}_{0}} \right)$内,对于所有的正整数kk0,定义停时

    此外,规定infØ=∞.显然{τk}是单调递增序列.设τ=limk +∞τk.易知ττe.若能证明τ=∞ a. s.则τe=∞ a. s.且(S(t),I(t),R(t))T$\mathbb{R}_{+}^{3}$ a. s.因此下面需证明τ=∞ a. s.假设τ < ∞,则存在T∈(0,∞)和ε∈(0,1),使得P(τT)>ε.

    $V(S, I, R)=\left(S-a-a\text{ln}\frac{S}{a} \right)+\left(I-1-\text{ln}I \right)+\left(R-1-\text{ln}R \right)$,其中a为待定正常数.由Itô公式[6]

    其中

    其中${{K}_{1}}={{\int }_{Y}}\left[a{{q}_{1}}\left(y \right)-a\text{ln}\left(1+{{q}_{1}}\left(y \right) \right) \right]v\left(\text{d}y \right)+$ ${{\int }_{Y}}\left[{{q}_{2}}\left(y \right)-\text{ln}\left(1+{{q}_{2}}\left(y \right) \right) \right]v\left(\text{d}y \right)+{{\int }_{Y}}\left[{{q}_{3}}\left(y \right)-\text{ln}\left(1+{{q}_{3}}\left(y \right) \right) \right]v(\text{d}y)$.取$a=\frac{\mu }{{{\beta }_{1}}}~$,则由(3)式得0≤EV(x(Tτk))≤V(x(0))+KT.

    Ωk={τkT}.当kk0时,P(Ωk)>ε,并且有V(x(Tτk))≥0.因此EV(x(Tτk))≥E(1ΩkV(x(Tτk))),其中1ΩkΩk的示性函数.对于任意ωΩk,由停时的定义,S(τkω),I(τkω),R(τkω)中至少有一个等于k$\frac{\text{1}}{k}$,因此

    所以V(x(0))+KTE(1ΩkV(x(Tτk)))P(Ωk)=EV(x(τkω))P(Ωk)≥Akε,当k→∞时,有∞>V(x(0))+KT=∞,矛盾,故假设不成立,即τ=∞ a. s.从而定理1得证.

3.   解的渐近性质
  • 模型(2)对应的确定性模型如下:

    利用文献[8]中计算基本再生数的方法,易知模型(4)的基本再生数${{R}_{0}}=\frac{{{\beta }_{1}}\mathit{\Lambda }\left(\mu +\gamma \right)}{{{\mu }^{2}}\left(\mu +\gamma +\alpha \right)}~$.当R0≤1时,模型(4)仅存在无病平衡点${{\mathrm{E}}_{0}}=\left(\frac{\mathit{\Lambda }}{\mu }, 0, 0 \right)~$;当R0> 1时,模型(4)不仅存在无病平衡点${{\mathrm{E}}_{0}}=\left(\frac{\mathit{\Lambda }}{\mu }, 0, 0 \right)~$,还存在地方病平衡点E*=(S*I*R*).

    定理2  如果R0≤1,且$-\mu +{{\sigma }_{1}}^{2}+3{{\int }_{Y}}{{q}_{1}}^{2}\left(y \right)v\left(\text{d}y \right) < 0$$-\mu +\frac{1}{2}{{\sigma }_{i}}^{2}+\frac{3}{2}{{\int }_{Y}}{{q}_{i}}^{2}\left(y \right)v\left(\text{d}y \right) < 0$i=2,3,那么对于任意初值(S(0),I(0),R(0))T$\mathbb{R}_{+}^{3}$,模型(2)的解x(t)=(S(t),I(t),R(t))T满足

    其中$M=\{\text{min}~\mu -{{\sigma }_{1}}^{2}-3{{\int }_{Y}}{{q}_{1}}^{2}\left(y \right)v\left(\text{d}y \right)$$\mu -\frac{1}{2}{{\sigma }_{2}}^{2}-\frac{3}{2}{{\int }_{Y}}{{q}_{2}}^{2}\left(y \right)v\left(\text{d}y \right)$$\mu -\frac{1}{2}{{\sigma }_{3}}^{2}-\frac{3}{2}{{\int }_{Y}}{{q}_{3}}^{2}\left(y \right)v\left(\text{d}y \right)\text{ }\!\!\}\!\!\text{ }$${{K}_{2}}={{\sigma }_{1}}^{2}~{{\left(\frac{\mathit{\Lambda }}{\mu } \right)}^{2}}+3{{\left(\frac{\mathit{\Lambda }}{\mu } \right)}^{2}}{{\int }_{Y}}{{q}_{1}}^{2}\left(y \right)v\left(\text{d}y \right)$.

     令${{V}_{1}}\left(S, I, R \right)=\frac{1}{2}\left(S-\frac{\mathit{\Lambda }}{\mu }+I+R \right){{~}^{2}}+{{a}_{1}}I+{{a}_{2}}R$,其中a1a2为待定正常数.由Itô公式[6]

    其中

    ${{a}_{1}}=\frac{2\mu }{{{\beta }_{1}}}$${{a}_{2}}=\frac{2\mu \gamma }{{{\beta }_{1}}\left(\mu +\gamma \right)}~$,则

    易知$0\le E{{V}_{1}}\left(x\left(t \right) \right)\le {{V}_{1}}\left(x\left(0 \right) \right)+E\int_{0}^{t}{~}\left(-M\left({{\left(S-\frac{\mathit{\Lambda }}{\mu } \right)}^{2}}+{{I}^{2}}+{{R}^{2}}+{{K}_{2}} \right) \right)\text{d}t$.从而

    定理2得证.

    定理3 如果R0>1且A>0,B>0,C>0,那么对于任意初值(S(0),I(0),R(0))T$\mathbb{R}_{+}^{3}$,模型(2)的解x(t)=(S(t),I(t),R(t))T满足

     令${{V}_{2}}\left(S, I, R \right)=\frac{1}{2}{{\left(S-{{S}^{*}}+I-{{I}^{*}}+R-{{R}^{*}} \right)}^{2}}+$ $\frac{p}{2}{{\left(S-{{S}^{*}}+I-{{I}^{*}} \right)}^{2}}+b\left(I-{{I}^{*}}-{{I}^{*}}\text{ln}~\frac{I}{{{I}^{*}}} \right)~+a\left(R-{{R}^{*}}-{{R}^{*}}\text{ln }\!\!~\!\!\text{ }\frac{R}{{{R}^{*}}} \right)$,其中abp为待定正常数.由Itô公式[6]

    其中

    $a=\frac{\gamma {{R}^{*}}}{\alpha {{I}^{*}}}$$p=\frac{2\mu }{\gamma }$$b=\frac{2\mu }{\gamma }\cdot \frac{\gamma +2\mu +\alpha }{{{\beta }_{1}}-{{\beta }_{2}}f\left({{I}^{*}} \right)~}$,则

    易知$0\le E{{V}_{2}}\left(x\left(t \right) \right)\le {{V}_{2}}\left(x\left(0 \right) \right)+E\int_{0}^{t}{{}}\left(-N\left({{\left(S-{{S}^{*}} \right)}^{2}}+{{\left(I-{{I}^{*}} \right)}^{2}}+{{\left(R-{{R}^{*}} \right)}^{2}} \right)+{{K}_{3}} \right))\text{d}t$,从而

    定理3得证.

     在模型(2)中,令f(I)= $\frac{\mathit{I}}{\mathit{m}\text{+}\mathit{I}}$qi(y)=0,i=1,2,3,则模型(2)转化为模型(1).容易验证,定理2与3推广了文献[6]中的定理3.1与4.1.

4.   数值模拟
  • 本节利用Milsteins方法[9-10]来验证本文主要理论结果.

    例1  对于模型(2),在无病平衡点附近,令初值S(0)=1.5,I(0)=0.5,R(0)=0.6,Λ=0.3,β1=0.005,β2=0.003,μ=0.2,α=0.043,γ=0.033,σ1=0.1,σ2=0.1,σ3=0.1,q1(y)=q2(y)=q3(y)=0.25,v=0.8,易得R0=0.04 < 1,且定理2的其他条件满足,因此定理2成立.如图 1所示.

    例2 对于模型(2),在地方病平衡点附近,令初值S(0)=2,I(0)=0.8,R(0)=0.1,Λ=0.3,β1=0.3,β2=0.25,μ=0.15,α=0.285 7,γ=0.285 7,σ1=0.02,σ2=0.01,σ3=0.02,q1(y)=q2(y)=q3(y)=0.15,v=0.05,易得R0=2.42>1,且定理3的其他条件满足,因此定理3成立.如图 2所示.

Figure (2)  Reference (10)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return