Next Article in Journal
Novel Bioinspired Approach Based on Chaotic Dynamics for Robot Patrolling Missions with Adversaries
Next Article in Special Issue
Conditional Gaussian Systems for Multiscale Nonlinear Stochastic Systems: Prediction, State Estimation and Uncertainty Quantification
Previous Article in Journal
Characterization of the Stroke-Induced Changes in the Variability and Complexity of Handgrip Force
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Dynamics Analysis of a Nonlinear Stochastic SEIR Epidemic System with Varying Population Size

1
College of Mathematics and Systems Science, Shandong University of Science and Technology, Qingdao 266590, China
2
State Key Laboratory of Mining Disaster Prevention and Control Co-founded by Shandong Province and the Ministry of Science and Technology, Shandong University of Science and Technology, Qingdao 266590, China
*
Author to whom correspondence should be addressed.
Submission received: 16 April 2018 / Revised: 14 May 2018 / Accepted: 16 May 2018 / Published: 17 May 2018
(This article belongs to the Special Issue Information Theory and Stochastics for Multiscale Nonlinear Systems)

Abstract

:
This paper considers a stochastic susceptible exposed infectious recovered (SEIR) epidemic model with varying population size and vaccination. We aim to study the global dynamics of the reduced nonlinear stochastic proportional differential system. We first investigate the existence and uniqueness of global positive solution of the stochastic system. Then the sufficient conditions for the extinction and permanence in mean of the infectious disease are obtained. Furthermore, we prove that the solution of the stochastic system has a unique ergodic stationary distribution under appropriate conditions. Finally, the discussion and numerical simulation are given to demonstrate the obtained results.

1. Introduction

Since the pioneering work of Kermack and Mckendrick [1], mathematical modeling for the dynamics of epidemic transmission has a realistic significance in predicting and controlling the spread of infectious diseases in the field of epidemiological research [2,3,4,5,6,7,8]. Recently, stochastic differential equations have been widely applied to physics, engineering, chemistry, and biology [9,10,11,12,13,14,15,16,17,18,19,20,21], which have obtained some novel results.
In fact, with the development of modern medicine, vaccination has become an important strategy for disease control. Then numerous scholars have investigated the effect of vaccination on disease [22,23,24,25,26,27]. The epidemic model with a constant population size is relatively effective for diseases with a low mortality and short duration. However, it is clearly untenable for diseases with a high mortality and varying populations. Thus epidemic models with varying population size seem to be more reasonable, which have attracted much interest from the research scientists [28,29,30]. Moreover, many infectious diseases incubate inside the hosts for a period of time before becoming infectious, so it is very meaningful to consider the effect of the incubation period. Based on the above considerations, Sun et al. [28] studied an SEIR model with varying population size and vaccination. The system can be described by
S ˙ = b N ( 1 δ ) β S I N δ ( 1 p ) β S I N δ p S μ S , E ˙ = ( 1 δ ) β S I N + δ ( 1 p ) β S I N α E μ E , I ˙ = α E ( ε + γ + μ ) I , R ˙ = δ p S + γ I μ R ,
where S ( t ) , E ( t ) , I ( t ) and R ( t ) , respectively, stand for the densities of the susceptible, the exposed, the infective and recovered individuals at time t, the total population size is denoted by N ( t ) = S ( t ) + E ( t ) + I ( t ) + R ( t ) . b represents the inflow rate (including birth and immigration), μ denotes the outflow rate (including natural death and emigration). The function β S I N stands for the standard incidence rate, here β represents the transmission rate of disease. δ ( 0 δ < 1 ) is the vaccine coverage rate of susceptible individuals, p ( 0 p 1 ) is the vaccine efficacy, α represents the rate at which the exposed individuals become infectious, ε is the rate of disease-related death and γ stands for the recovery rate of infective individuals. The parameters δ and p are all non-negative constants and b, μ , β , α , ε and γ are positive constants. Moreover, the differential equation of total population size N ( t ) is given by N ˙ = ( b μ ) N ε I . The authors [28] explored the proportions of individuals in the four epidemiological classes, namely
s ˜ = S N , e ˜ = E N , i ˜ = I N , r ˜ = R N .
It is easy to get that the variables s ˜ , e ˜ , i ˜ and r ˜ satisfy the following system of differential equations
s ˜ ˙ = b ( 1 δ p ) β s ˜ i ˜ ( δ p + b ) s ˜ + ε s ˜ i ˜ , e ˜ ˙ = ( 1 δ p ) β s ˜ i ˜ ( α + b ) e ˜ + ε e ˜ i ˜ , i ˜ ˙ = α e ˜ ( ε + γ + b ) i ˜ + ε i ˜ 2 , r ˜ ˙ = δ p s ˜ + γ i ˜ b r ˜ + ε i ˜ r ˜ .
Since variable r ˜ does not appear in the first, second, third equations of the above system. Then the above system becomes the following reduced system
s ˜ ˙ = b ( 1 δ p ) β s ˜ i ˜ ( δ p + b ) s ˜ + ε s ˜ i ˜ , e ˜ ˙ = ( 1 δ p ) β s ˜ i ˜ ( α + b ) e ˜ + ε e ˜ i ˜ , i ˜ ˙ = α e ˜ ( ε + γ + b ) i ˜ + ε i ˜ 2
which is subject to the constraint r ˜ = 1 s ˜ e ˜ i ˜ . In the region = s ˜ , e ˜ , i ˜ R + 3 | 0 s ˜ + e ˜ + i ˜ 1 , they established the epidemiological threshold condition R 0 , which determines disease extinction or permanence, where
R 0 = b α β ( 1 δ p ) ( α + b ) ( δ p + b ) ( ε + γ + b ) .
Meanwhile, they analyzed the global dynamics of system (3) and derived the equilibria (including the disease-free equilibrium and the endemic equilibrium) and their global stability. In addition, the parameter restrictions for uniform permanence were obtained.
Nevertheless, the biological populations in the ecosystem are inevitably subjected to uncertain environmental perturbations. It is worth noting that this phenomenon is ubiquitous in the natural environment. So various stochastic epidemic models have been proposed and studied [31,32,33,34,35,36]. To the best of our knowledge, there are not too many researches on global dynamics of the stochastic SEIR epidemic model with varying population size and vaccination yet. In this paper, to make this epidemic model (1) more reasonable and realistic, we suppose the stochastic perturbations are directly proportional to s ˜ , e ˜ , i ˜ and r ˜ under the influence of white noise type, influenced on the s ˜ ˙ ( t ) , e ˜ ˙ ( t ) , i ˜ ˙ ( t ) and r ˜ ˙ ( t ) in system (1), respectively. This implies the stochastic effects of white noise on the birth and death rates of S , E , I , R . Then corresponding to system (1), a stochastic version can be reached by
d S = b N ( 1 δ ) β S I N δ ( 1 p ) β S I N δ p S μ S d t + σ 1 S d B 1 ( t ) , d E = ( 1 δ ) β S I N + δ ( 1 p ) β S I N α E μ E d t + σ 2 E d B 2 ( t ) , d I = α E ( ε + γ + μ ) I d t + σ 3 I d B 3 ( t ) , d R = δ p S + γ I μ R d t + σ 4 R d B 4 ( t ) ,
where B i ( t ) ( i = 1 , 2 , 3 , 4 ) is the standard Wiener processes with B i ( 0 ) = 0 a.s. σ i ( t ) ( i = 1 , 2 , 3 , 4 ) stands for a continuous and bounded function for any t 0 and σ i 2 ( t ) ( i = 1 , 2 , 3 , 4 ) represents the intensities of Wiener processes. Furthermore, the differential equation of total population size N ( t ) is given by the following form
d N = [ ( b μ ) N ε I ] d t + σ 1 S d B 1 ( t ) + σ 2 E d B 2 ( t ) + σ 3 I d B 3 ( t ) + σ 4 R d B 4 ( t ) .
From (2), the system (4) becomes the following proportional system
d s ˜ = b ( 1 δ p ) β s ˜ i ˜ ( δ p + b ) s ˜ + ε s ˜ i ˜ σ 1 2 s ˜ 2 + s ˜ σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 r ˜ 2 d t + σ 1 s ˜ ( 1 s ˜ ) d B 1 ( t ) σ 2 s ˜ e ˜ d B 2 ( t ) σ 3 s ˜ i ˜ d B 3 ( t ) σ 4 s ˜ r ˜ d B 4 ( t ) , d e ˜ = ( 1 δ p ) β s ˜ i ˜ ( α + b ) e ˜ + ε e ˜ i ˜ σ 2 2 e ˜ 2 + e ˜ σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 r ˜ 2 d t σ 1 s ˜ e ˜ d B 1 ( t ) + σ 2 e ˜ ( 1 e ˜ ) d B 2 ( t ) σ 3 e ˜ i ˜ d B 3 ( t ) σ 4 e ˜ r ˜ d B 4 ( t ) , d i ˜ = α e ˜ ( ε + γ + b ) i ˜ + ε i ˜ 2 σ 3 2 i ˜ 2 + i ˜ σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 r ˜ 2 d t σ 1 s ˜ i ˜ d B 1 ( t ) σ 2 e ˜ i ˜ d B 2 ( t ) + σ 3 i ˜ 1 i ˜ d B 3 ( t ) σ 4 i ˜ r ˜ d B 4 ( t ) , d r ˜ = δ p s ˜ + γ i ˜ b r ˜ + ε i ˜ r ˜ σ 4 2 r ˜ 2 + r ˜ σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 r ˜ 2 d t σ 1 s ˜ r ˜ d B 1 ( t ) σ 2 e ˜ r ˜ d B 2 ( t ) σ 3 i ˜ r ˜ d B 3 ( t ) + σ 4 r ˜ ( 1 r ˜ ) d B 4 ( t ) .
It is worthy to note that, the variables s ˜ , e ˜ , i ˜ and r ˜ satisfy the relation r ˜ = 1 s ˜ e ˜ i ˜ , we can omit analysis of the fourth equation of system (5) and explore the following reduced system
d s ˜ = [ b ( 1 δ p ) β s ˜ i ˜ ( δ p + b ) s ˜ + ε s ˜ i ˜ σ 1 2 s ˜ 2 + s ˜ ( σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 ) ] d t + σ 1 s ˜ ( 1 s ˜ ) d B 1 ( t ) σ 2 s ˜ e ˜ d B 2 ( t ) σ 3 s ˜ i ˜ d B 3 ( t ) σ 4 s ˜ 1 s ˜ e ˜ i ˜ d B 4 ( t ) , d e ˜ = [ ( 1 δ p ) β s ˜ i ˜ ( α + b ) e ˜ + ε e ˜ i ˜ σ 2 2 e ˜ 2 + e ˜ ( σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 ) ] d t σ 1 s ˜ e ˜ d B 1 ( t ) + σ 2 e ˜ ( 1 e ˜ ) d B 2 ( t ) σ 3 e ˜ i ˜ d B 3 ( t ) σ 4 e ˜ 1 s ˜ e ˜ i ˜ d B 4 ( t ) , d i ˜ = α e ˜ ( ε + γ + b ) i ˜ + ε i ˜ 2 σ 3 2 i ˜ 2 + i ˜ σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 d t σ 1 s ˜ i ˜ d B 1 ( t ) σ 2 e ˜ i ˜ d B 2 ( t ) + σ 3 i ˜ 1 i ˜ d B 3 ( t ) σ 4 i ˜ 1 s ˜ e ˜ i ˜ d B 4 ( t )
with the initial value s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) R + 3 and s ˜ ( 0 ) + e ˜ ( 0 ) + i ˜ ( 0 ) < 1 .
Since system (6) is a three-dimensional stochastic system with many high-order nonlinear terms, this makes the stochastic analysis novel and more complex than [34,36].
Throughout this article, unless otherwise specified, let ( Ω , F , { F } t 0 , P ) be a complete probability space with a filtration { F t } t 0 satisfying the usual conditions (i.e., it is increasing and right continuous while F 0 contains all P -null sets). Further suppose B i ( t ) ( i = 1 , 2 , 3 , 4 ) stands for the mutually independent standard Wiener processes defined on the complete probability space Ω . For an integrable function x ( t ) on [ 0 , + ) , let us define x ( t ) = 1 t 0 t x ( r ) d r .

2. Global Positive Solution

The following Itô’s formula will be used frequently in the sequel.
Lemma 1.
[37] Assume that X ( t ) R + is an Itô’s process of the form
d X ( t ) = F ( X ( t ) , t ) d t + G ( X ( t ) , t ) d B ( t ) ,
where F : R n × R + × S R n and G : R n × R + × S R n are measurable functions.
Given V C 2 , 1 ( R n × R + × S ; R + ) , we define the operator L V by
L V ( X , t ) = V t ( X , t ) + V X ( X , t ) F ( X , t ) + 1 2 trace G T ( X , t ) V X X ( X , t ) G ( X , t ) ,
where
V t ( X , t ) = V X ( X , t ) t , V X ( X , t ) = V X ( X , t ) X 1 , , V X ( X , t ) X n , V X X ( X , t ) = 2 V X ( X , t ) X i X j n × n .
Then the generalized Itô’s formula is given by
d V ( X , t ) = L V ( X , t ) d t + V X ( X , t ) G ( X , t ) d B ( t ) .
To explore the dynamical behaviors of a population system, we first concern the global existence and positivity of the solutions of system (6).
Lemma 2.
For any given initial value s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) R + 3 and s ˜ ( 0 ) + e ˜ ( 0 ) + i ˜ ( 0 ) < 1 , the system (6) has a unique positive local solution ( s ˜ ( t ) , e ˜ ( t ) , i ˜ ( t ) ) for t [ ω , τ e ) , where τ e is the explosion time [37].
Theorem 1.
For any given initial value s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) R + 3 and s ˜ ( 0 ) + e ˜ ( 0 ) + i ˜ ( 0 ) < 1 , the system (6) has a unique positive solution s ˜ ( t ) , e ˜ ( t ) , i ˜ ( t ) R + 3 on t > 0 a.s.
Proof. 
The following proof is divided into two parts.
Part I. Since the coefficients of the system (6) satisfy local Lipschitz condition, from Lemma 2, it is easy to see that the system (6) has a unique positive local solution s ˜ ( t ) , e ˜ ( t ) , i ˜ ( t ) for any given initial value s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) R + 3 and s ˜ ( 0 ) + e ˜ ( 0 ) + i ˜ ( 0 ) < 1 .
Part II. Now we prove that the positive solution is global, that is τ e = a.s. Let k 0 0 be sufficiently large such that s ˜ ( 0 ) , e ˜ ( 0 ) and i ˜ ( 0 ) all lie in 1 k 0 , k 0 . For each integer k k 0 , let us define the stopping time
τ k = inf t [ ω , τ e ) : s ˜ ( t ) 1 k , k , e ˜ ( t ) 1 k , k or i ˜ ( t ) 1 k , k ,
where we define inf = (∅ stands for the empty set). Evidently, τ k is strictly increasing when k . Let τ = lim k τ k , thus τ τ e a.s. So we just need to show that τ = a.s. If τ = is untrue, then there exist two constants T > 0 and ϵ ( 0 , 1 ) such that P { τ T } > ϵ . Hence, there exists k 1 k 0 ( k 1 N + ) such that
P { τ k T } ϵ , k k 1 .
Define a C 2 -function V: R + 3 R + by
V s ˜ , e ˜ , i ˜ = ln 1 s ˜ e ˜ i ˜ ln s ˜ ln e ˜ ln i ˜ 3 .
The non-negativity of V ( s ˜ , e ˜ , i ˜ ) can be obtained from m 1 ln m 0 , m > 0 .
In terms of the multi-dimensional Itô’s formula and system (6), we have
d V = L V d t + σ 1 4 s ˜ 1 d B 1 ( t ) + σ 2 4 e ˜ 1 d B 2 ( t ) + σ 3 4 i ˜ 1 d B 3 ( t ) + σ 4 3 4 s ˜ 4 e ˜ 4 i ˜ d B 4 ( t ) ,
where L V is given in Appendix A in detail. Then we have
L V β + δ p + 4 b + ε + α + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 + 1 2 σ 4 2 : = M 0 ,
where M 0 is a positive constant.
So we get
d V M 0 d t + σ 1 4 s ˜ 1 d B 1 ( t ) + σ 2 4 e ˜ 1 d B 2 ( t ) + σ 3 4 i ˜ 1 d B 3 ( t ) + σ 4 3 4 s ˜ 4 e ˜ 4 i ˜ d B 4 ( t ) .
Integrating both sides of (8) from 0 to τ k T and then taking the expectation yield
E V s ˜ ( τ k T ) , e ˜ ( τ k T ) , i ˜ ( τ k T ) V s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) + E 0 τ k T M 0 d t V s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) + M 0 T .
Let Ω k = { τ k T } , k k 1 and from (7), we have P ( Ω k ) ϵ . Notice that for every ω Ω k , there exists s ˜ ( τ k , ω ) , e ˜ ( τ k , ω ) or i ˜ ( τ k , ω ) equals either 1 k or k. Thus
V s ˜ ( τ k , ω ) , e ˜ ( τ k , ω ) , i ˜ ( τ k , ω ) 1 k 1 ln 1 k ( k 1 ln k ) .
By virtue of (9) and (10), we have
V s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) + M 0 T E 1 Ω k ( ω ) V s ˜ ( τ k , ω ) , e ˜ ( τ k , ω ) , i ˜ ( τ k , ω ) ϵ 1 k 1 ln 1 k ( k 1 ln k ) ,
here 1 Ω k ( ω ) represents the indicator function of Ω k ( ω ) .
Let k , which implies
> V s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) + M 0 T =
is a contradiction. Obviously, we get that τ = . The proof of Theorem 1 is complete.  ☐

3. Extinction

For a population system, the parameter conditions of disease extinction and permanence have become an important issue that attracts more and more attention in real life. In this section, we mainly investigate the extinction of disease and leave the argument of permanence to the next section.
Theorem 2.
Let s ˜ ( t ) , e ˜ ( t ) , i ˜ ( t ) be the solution of system (6) with the initial value s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) R + 3 and s ˜ ( 0 ) + e ˜ ( 0 ) + i ˜ ( 0 ) < 1 . If the parameter conditions
M 1 < 2 ( 1 ϱ ) α , ε ( 1 δ p ) β < ε + γ + b
hold, then
lim   sup t ln e ˜ ( t ) + ϱ i ˜ ( t ) t ( ϱ 1 ) α + M 1 2 < 0 a . s . ,
where
ϱ = ( ε + γ + b α ) + ( ε + γ + b α ) 2 + 4 ( 1 δ p ) α β 2 α
and
M 1 = max σ 1 2 , 1 ϱ 2 σ 3 2 , σ 4 2 ,
namely, e ˜ ( t ) and i ˜ ( t ) tend to zero exponentially a.s. That is to say, the exposed and infective individuals go to extinction almost surely.
Proof. 
Let us define a differentiable function V by
V = ln e ˜ ( t ) + ϱ i ˜ ( t ) ,
here ϱ is a positive constant to be determined later. According to the Itô’s formula and system (6), we have
d V = L V d t σ 1 s ˜ d B 1 ( t ) + σ 2 e ˜ 1 e ˜ ϱ i ˜ e ˜ + ϱ i ˜ d B 2 ( t ) + σ 3 i ˜ ϱ e ˜ ϱ i ˜ e ˜ + ϱ i ˜ d B 3 ( t ) σ 4 1 s ˜ e ˜ i ˜ d B 4 ( t ) ,
where L V is given in Appendix B in detail. One can derive that
d V ( ϱ 1 ) α + M 1 2 d t σ 1 s ˜ d B 1 ( t ) + σ 2 e ˜ 1 e ˜ ϱ i ˜ e ˜ + ϱ i ˜ d B 2 ( t ) + σ 3 i ˜ ϱ e ˜ ϱ i ˜ e ˜ + ϱ i ˜ d B 3 ( t ) σ 4 1 s ˜ e ˜ i ˜ d B 4 ( t ) ,
here M 1 = max σ 1 2 , 1 ϱ 2 σ 3 2 , σ 4 2 . Then, integrating from 0 to t and dividing by t on both sides of (12) yield
ln e ˜ ( t ) + ϱ i ˜ ( t ) t ( ϱ 1 ) α + M 1 2 + ln e ˜ ( 0 ) + ϱ i ˜ ( 0 ) t + M ˜ ( t ) t ,
here
M ˜ ( t ) = σ 1 0 t s ˜ ( r ) d B 1 ( r ) + σ 2 0 t e ˜ ( r ) 1 e ˜ ( r ) ϱ i ˜ ( r ) e ˜ ( r ) + ϱ i ˜ ( r ) d B 2 ( r ) + σ 3 0 t i ˜ ( r ) ϱ e ˜ ( r ) ϱ i ˜ ( r ) e ˜ ( r ) + ϱ i ˜ ( r ) d B 3 ( r ) σ 4 0 t 1 s ˜ ( r ) e ˜ ( r ) i ˜ ( r ) d B 4 ( r ) .
In a similar way as [38], making use of the strong law of large numbers [37] yields
lim t M ˜ ( t ) t = 0 a . s .
Therefore,
lim   sup t ln e ˜ ( t ) + ϱ i ˜ ( t ) t ( ϱ 1 ) α + M 1 2 < 0 a . s . ,
which shows that
lim t e ˜ ( t ) = 0 , lim t i ˜ ( t ) = 0 a . s .
The proof of Theorem 2 is complete. ☐

4. Permanence in Mean

Theorem 3.
Let s ˜ ( t ) , e ˜ ( t ) , i ˜ ( t ) be the solution of system (6) with the initial value s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) R + 3 and s ˜ ( 0 ) + e ˜ ( 0 ) + i ˜ ( 0 ) < 1 . If the parameter condition
b ( 1 δ p ) β α 3 > δ p + 3 b + α + ε + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 3
holds, then
i ˜ ̲ lim   inf t i ˜ ( t ) lim   sup t i ˜ ( t ) i ˜ ¯ a . s . ,
where
i ˜ ̲ = 3 b ( 1 δ p ) β α 3 δ p + 3 b + α + ε + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 ( 1 δ p ) β
and
i ˜ ¯ = 2 b + α + ε + γ + 1 2 σ 2 2 + 1 2 σ 3 2 2 ε ,
that is to say, the infective individuals i ˜ ( t ) are permanent in mean almost surely.
Proof. 
The following proof is divided into two steps.
Step I. According to the Itô’s formula and system (6), we have
d ln s ˜ + ln e ˜ + ln i ˜ = [ b s ˜ ( 1 δ p ) β i ˜ + 3 ε i ˜ + ( 1 δ p ) β s ˜ i ˜ e ˜ + α e ˜ i ˜ + 3 2 σ 1 2 s ˜ 2 + 3 2 σ 2 2 e ˜ 2 + 3 2 σ 3 2 i ˜ 2 + 3 2 σ 4 2 1 s ˜ e ˜ i ˜ 2 ( δ p + 3 b + α + ε + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 ) ] d t + σ 1 1 3 s ˜ d B 1 ( t ) + σ 2 1 3 e ˜ d B 2 ( t ) + σ 3 1 3 i ˜ d B 3 ( t ) 3 σ 4 1 s ˜ e ˜ i ˜ d B 4 ( t ) .
Integrating from 0 to t and dividing by t on both sides of (13) lead to
ln s ˜ ( t ) t + ln e ˜ ( t ) t + ln i ˜ ( t ) t 3 b ( 1 δ p ) β α 3 ( 1 δ p ) β i ˜ ( δ p + 3 b + α + ε + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 ) + ln s ˜ ( 0 ) t + ln e ˜ ( 0 ) t + ln i ˜ ( 0 ) t + M ̲ ( t ) t ,
here
M ̲ ( t ) = σ 1 0 t 1 3 s ˜ ( r ) d B 1 ( r ) + σ 2 0 t 1 3 e ˜ ( r ) d B 2 ( r ) + σ 3 0 t 1 3 i ˜ ( r ) d B 3 ( r ) 3 σ 4 0 t 1 s ˜ ( r ) e ˜ ( r ) i ˜ ( r ) d B 4 ( r ) .
The detail derivation process for the above inequality is given in Appendix C.
In a similar way as [38], making use of the strong law of large numbers [37] leads to
lim t M ̲ ( t ) t = 0 a . s .
Then, by virtue of < ln s ˜ ( t ) < 0 , < ln e ˜ ( t ) < 0 , < ln i ˜ ( t ) < 0 ( s ˜ + e ˜ + i ˜ + r ˜ = 1 ) and δ p < 1 , it is easy to get that
lim   inf t i ˜ ( t ) i ˜ ̲ > 0 a . s .
Step II. Similarly, using the Itô’s formula and system (6), we have
d ln e ˜ + ln i ˜ = [ ( 1 δ p ) β s ˜ i ˜ e ˜ + α e ˜ i ˜ + 2 ε i ˜ + σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 2 b + α + ε + γ + 1 2 σ 2 2 + 1 2 σ 3 2 ] d t 2 σ 1 s ˜ d B 1 ( t ) + σ 2 1 2 e ˜ d B 2 ( t ) + σ 3 1 2 i ˜ d B 3 ( t ) 2 σ 4 1 s ˜ e ˜ i ˜ d B 4 ( t ) .
Integrating from 0 to t and dividing by t on both sides of (14) result in
ln e ˜ ( t ) t + ln i ˜ ( t ) t = ( 1 δ p ) β s ˜ i ˜ e ˜ + α e ˜ i ˜ + 2 ε i ˜ + σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 2 b + α + ε + γ + 1 2 σ 2 2 + 1 2 σ 3 2 + ln e ˜ ( 0 ) t + ln i ˜ ( 0 ) t + M ^ ( t ) t 2 ε i ˜ 2 b + α + ε + γ + 1 2 σ 2 2 + 1 2 σ 3 2 + ln e ˜ ( 0 ) t + ln i ˜ ( 0 ) t + M ^ ( t ) t ,
here
M ^ ( t ) = 2 σ 1 0 t s ˜ ( r ) d B 1 ( r ) + σ 2 0 t 1 2 e ˜ ( r ) d B 2 ( r ) + σ 3 0 t 1 2 i ˜ ( r ) d B 3 ( r ) 2 σ 4 0 t 1 s ˜ ( r ) e ˜ ( r ) i ˜ ( r ) d B 4 ( r ) .
In a similar way as [38], using the strong law of large numbers [37], we have
lim t M ^ ( t ) t = 0 a . s .
Therefore,
lim   sup t i ˜ ( t ) i ˜ ¯ > 0 a . s .
The proof of Theorem 3 is complete. ☐

5. Stationary Distribution and Ergodicity

Recently, the stationary distribution attract deep research interests of many authors [32,33,34,35]. The ergodicity is one of the most important properties for the stochastic system, and geometric ergodicity for finite-dimensional systems has been shown in detail and well-developed in many earlier works [39,40]. In this section, based on the theory of Khasminskii [41] and the Lyapunov function method, we explore the conditions of the existence of an ergodic stationary distribution, which shows that the epidemic disease will prevail.
Assume X ( t ) be a time-homogeneous Markov process in D n R n , which is described by the stochastic differential equation
d X ( t ) = b ( X ) d t + η = 1 n σ η ( X ) d B η ( t ) ,
here D n stands for a n-dimensional Euclidean space.
The diffusion matrix is as follows:
A ( x ) = ( a i j ( x ) ) , a i j ( x ) = η = 1 n σ η i ( x ) σ η j ( x ) .
Assumption 1.
Assume that there exists a bounded domain U D n with regular boundary Γ such that U ¯ D n ( U ¯ is the closure of U ) , satisfying the following properties:
(i) In the domain U and some neighborhood thereof, the smallest eigenvalue of the diffusion matrix A ( x ) is bounded away from zero.
(ii) If x D n \ U , the mean time τ at which a path issuing from x reaches the set U is finite, and sup x Θ E x τ < for every compact subset Θ D n .
Lemma 3.
[41] When Assumption 1 holds, then the Markov process X ( t ) has a stationary distribution π ( · ) . In addition, when f ( · ) is a function integrable with respect to the measure π, then
P x lim T 1 T 0 T f ( X ( t ) ) d t = D n f ( x ) π ( d x ) = 1
for all x D n .
Remark 1.
To prove Assumption 1 ( i ) [42], it suffices to demonstrate that F is uniformly elliptical in any bounded domain H, here
F u = b ( x ) u x + 1 2 trace ( A ( x ) u x x ) ,
namely, there exists a positive number Z such that
i , j = 1 n a i j ( x ) ξ i ξ j Z | ξ | 2 , x H ¯ , ξ R n .
To prove Assumption 1 ( i i ) [43], it suffices to demonstrate that there exist some neighborhood U and a nonnegative C 2 -function V such that x D n \ U , L V ( x ) < 0 .
Making use of the Lemma 3, we can obtain the main results as follows.
Theorem 4.
Let s ˜ ( t ) , e ˜ ( t ) , i ˜ ( t ) be the solution of system (6) with the initial value s ˜ ( 0 ) , e ˜ ( 0 ) , i ˜ ( 0 ) R + 3 and s ˜ ( 0 ) + e ˜ ( 0 ) + i ˜ ( 0 ) < 1 . If the parameter condition
b ( 1 δ p ) β α 3 > 2 δ p + 5 b + α + 2 ε + 2 γ + 2 σ 1 2 + 2 σ 2 2 + 2 σ 3 2 + 1 2 σ 4 2 3
holds, then the system (6) has a unique stationary distribution π ( · ) and it has ergodic property.
Proof. 
Now let us define a positive-definite function V by
V = ln s ˜ + e ˜ + i ˜ ln s ˜ ln e ˜ ln i ˜ ln r ˜ .
Using the Itô’s formula yields
L V b s ˜ ( 1 δ p ) β s ˜ i ˜ e ˜ α e ˜ i ˜ b s ˜ + e ˜ + i ˜ δ p s ˜ r ˜ γ i ˜ r ˜ + M 1 ,
here
M 1 = ( 1 δ p ) β + 2 δ p + 5 b + α + 2 ε + 2 γ + 2 σ 1 2 + 2 σ 2 2 + 2 σ 3 2 + 1 2 σ 4 2 .
The detail derivation process for the above inequality of L V is given in Appendix D.
Next let us construct the following compact subset U:
U = s ˜ , e ˜ , i ˜ U ˜ : ψ 1 s ˜ < 1 , ψ 2 e ˜ < 1 , ψ 3 i ˜ < 1 , ψ 4 s ˜ + e ˜ + i ˜ 1 ψ 4 ,
where
U ˜ = 0 < s ˜ < 1 , 0 < e ˜ < 1 , 0 < i ˜ < 1 , 0 < s ˜ + e ˜ + i ˜ < 1
and ψ i ( 0 , 1 ) ( i = 1 , 2 , 3 , 4 ) is a sufficiently small constant satisfying the following conditions:
ψ 2 = ψ 1 2 ψ 3 , ψ 4 = ψ 1 2 = ψ 3 2 ,
b ψ 1 + M 1 1 ,
( 1 δ p ) β ψ 1 + M 1 1 ,
( 1 δ p ) β ψ 3 b < 0 ,
b ψ 4 + M 1 1 ,
δ p ψ 1 γ ψ 3 + M 1 1 .
Then
U ˜ \ U = U 1 U 2 U 3 U 4 U 5 ,
with
U 1 = s ˜ , e ˜ , i ˜ U ˜ : 0 < s ˜ < ψ 1 , U 2 = s ˜ , e ˜ , i ˜ U ˜ : ψ 1 s ˜ < 1 , 0 < e ˜ < ψ 2 , ψ 3 i ˜ < 1 ,
U 3 = s ˜ , e ˜ , i ˜ U ˜ : 0 < i ˜ < ψ 3 , U 4 = s ˜ , e ˜ , i ˜ U ˜ : 0 < s ˜ + e ˜ + i ˜ < ψ 4 ,
U 5 = s ˜ , e ˜ , i ˜ U ˜ : ψ 1 s ˜ < 1 , ψ 3 i ˜ < 1 , 1 ψ 4 < s ˜ + e ˜ + i ˜ < 1 .
Now we prove the negativity of L V for any U ˜ \ U .
Case I. If s ˜ , e ˜ , i ˜ U 1 , it follows from (A1) and (17) that
L V b s ˜ + M 1 b ψ 1 + M 1 1 .
Case II. If s ˜ , e ˜ , i ˜ U 2 , (16) and (18) derive that
L V ( 1 δ p ) β s ˜ i ˜ e ˜ + M 1 ( 1 δ p ) β ψ 1 ψ 3 ψ 2 + M 1 = ( 1 δ p ) β ψ 1 + M 1 1 .
Case III. If s ˜ , e ˜ , i ˜ U 3 , (A1) and (19) yield that
L V b b s ˜ ( 1 δ p ) β s ˜ i ˜ e ˜ α e ˜ i ˜ + ( 1 δ p ) β ψ 3 + 2 δ p + 5 b + α + 2 ε + 2 γ + 2 σ 1 2 + 2 σ 2 2 + 2 σ 3 2 + 1 2 σ 4 2 3 b ( 1 δ p ) β α 3 + 2 δ p + 5 b + α + 2 ε + 2 γ + 2 σ 1 2 + 2 σ 2 2 + 2 σ 3 2 + 1 2 σ 4 2 < 0 .
Case IV. If s ˜ , e ˜ , i ˜ U 4 , (A1) and (20) imply that
L V b s ˜ + e ˜ + i ˜ + M 1 b ψ 4 + M 1 1 .
Case V. If s ˜ , e ˜ , i ˜ U 5 , it follows from (A1), (16) and (21) that
L V δ p s ˜ r ˜ γ i ˜ r ˜ + M 1 δ p ψ 1 ψ 4 γ ψ 3 ψ 4 + M 1 = δ p ψ 1 γ ψ 3 + M 1 1 .
Define
ϕ = max 1 , 3 b ( 1 δ p ) β α 3 + 2 δ p + 5 b + α + 2 ε + 2 γ + 2 σ 1 2 + 2 σ 2 2 + 2 σ 3 2 + 1 2 σ 4 2 < 0 .
Obviously, one can see that L V ϕ < 0 for all s ˜ , e ˜ , i ˜ U ˜ \ U , which shows that Assumption 1 ( i i ) is satisfied. On the other hand, there exists a positive number
Z = min { σ 1 2 1 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 s ˜ 2 , ( σ 1 2 s ˜ 2 + σ 2 2 1 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 × 1 s ˜ e ˜ i ˜ 2 ) e ˜ 2 , σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 1 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 i ˜ 2 , s ˜ , e ˜ , i ˜ U ˜ }
such that
i , j = 1 3 a i j ξ i ξ j = σ 1 2 1 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 s ˜ 2 ξ 1 2 + ( σ 1 2 s ˜ 2 + σ 2 2 1 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 ) e ˜ 2 ξ 2 2 + σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 1 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 i ˜ 2 ξ 3 2 Z | ξ | 2 , s ˜ , e ˜ , i ˜ U ˜ , ξ R 3 ,
which shows that Assumption 1 ( i ) is satisfied. Consequently, the system (6) has a unique stationary distribution π ( · ) and it has ergodic property. The proof of Theorem 4 is complete. ☐

6. Simulations and Conclusions

6.1. Simulations

Next, in order to support the results of the above theorems, we carry out some computer simulations.
In Figure 1, take s ˜ ( 0 ) = 0.3 , e ˜ ( 0 ) = 0.25 , i ˜ ( 0 ) = 0.15 , b = 0.15 , β = 0.5 , γ = 0.3 , α = 0.2 , δ = 0.25 , p = 0.2 , ε = 0.15 and σ 1 = σ 2 = σ 3 = σ 4 = 0.25 . Then
M 1 = max σ 1 2 , 1 ϱ 2 σ 3 2 , σ 4 2 = 0.0625 < 2 ( 1 ϱ ) α = 0.0652
and
ε = 0.15 < ( 1 δ p ) β = 0.475 < ε + γ + b = 0.6
satisfy the parameter conditions in Theorem 2, we can get that the exposed and infective individuals go to extinction almost surely. Obviously, Figure 1 validates our results of the Theorem 2.
In Figure 2, take s ˜ ( 0 ) = 0.15 , e ˜ ( 0 ) = 0.2 , i ˜ ( 0 ) = 0.15 , b = 0.02 , β = 0.9 , γ = 0.01 , α = 0.1 , δ = 0.02 , p = 0.02 , ε = 0.16 , σ 1 = 0.05 , σ 2 = 0.05 , σ 3 = 0.05 and σ 4 = 0.1 . Obviously,
0.1216 = b ( 1 δ p ) β α 3 > δ p + 3 b + α + ε + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 3 = 0.1114
satisfies the parameter condition in Theorem 3, then
0.0342 = i ˜ ̲ lim   inf t i ˜ ( t ) lim   sup t i ˜ ( t ) i ˜ ¯ = 0.9766 ,
we can get that the infective individuals i ˜ ( t ) are permanent in mean almost surely. As expected, Figure 2 confirms our results of the Theorem 3.
From Figure 2 and Figure 3, a set of large stochastic parameter values σ 1 = σ 2 = σ 3 = σ 4 = 0.25 can lead to infective individuals go to extinction (see Figure 2), while infective individuals can be permanent in mean under the condition of a set of small stochastic parameter values σ 1 = σ 2 = σ 3 = 0.05 and σ 4 = 0.1 (see Figure 3).
In Figure 3, take s ˜ ( 0 ) = 0.15 , e ˜ ( 0 ) = 0.2 , i ˜ ( 0 ) = 0.15 , b = 0.02 , β = 2.1 , γ = 0.01 , α = 0.2 , δ = 0.02 , p = 0.02 , ε = 0.1 and σ 1 = σ 2 = σ 3 = σ 4 = 0.01 . Then
0.2033 = b ( 1 δ p ) β α 3 > 2 δ p + 5 b + α + 2 ε + 2 γ + 2 σ 1 2 + 2 σ 2 2 + 2 σ 3 2 + 1 2 σ 4 2 3 = 0.1738
satisfies the parameter condition in Theorem 4, we can get that the stochastic system (6) has a unique stationary distribution π ( · ) and it has ergodic property. Figure 3 indicates that the solution of system (6) swings up and down in a small neighborhood. According to the density functions in Figure 3d–f, we can see that there exists a stationary distribution. As expected, Figure 3 supports our results of the Theorem 4.
The Figure 1, Figure 2 and Figure 3 above show that the large white noise value can lead to infectious diseases to go to extinction, which implies that stochastic fluctuations can suppress the disease outbreak, while the small white noise value can cause infectious diseases to be persistent. In addition, The Figure 3 also shows the stochastic system (6) has a unique ergodic stationary distribution under appropriate conditions. Therefore, the numerical simulation examples are completely consistent with the theoretical results of the Theorems 2–4.

6.2. Conclusions

In this paper, we apply stochastic analysis methods to study the global dynamics of a high-dimensional stochastic reduced proportional SEIR epidemic system which makes the analysis novel and complex. We obtain the existence of a unique global positive solution and parameter conditions of extinction or permanence in mean. Furthermore, the solution of the stochastic system has a unique ergodic stationary distribution under certain sufficient parameter conditions. Cubic terms of s ˜ , e ˜ , i ˜ and multiple stochastic terms for d B i ( t ) ( i = 1 , 2 , 3 , 4 ) in system (6) make the analysis more difficult and complex than the models in [34,36]. Some ingenious inequality techniques are used to deal with cubic terms of s ˜ , e ˜ , i ˜ of system (6). Therefore, compare with previous methods and research results, we develop previous methods and improve the main results of previous studies.
We summarize the main conclusions as follows:
(I) When
M 1 < 2 ( 1 ϱ ) α , ε ( 1 δ p ) β < ε + γ + b
hold, then
lim   sup t ln e ˜ ( t ) + ϱ i ˜ ( t ) t < 0 a . s .
That is to say, the exposed and infective individuals go to extinction almost surely.
(II) When
b ( 1 δ p ) β α 3 > δ p + 3 b + α + ε + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 3
holds, then
i ˜ ̲ lim   inf t i ˜ ( t ) lim   sup t i ˜ ( t ) i ˜ ¯ a . s .
That is to say, the infective individuals i ˜ ( t ) are permanent in mean almost surely.
(III) When
b ( 1 δ p ) β α 3 > 2 δ p + 5 b + α + 2 ε + 2 γ + 2 σ 1 2 + 2 σ 2 2 + 2 σ 3 2 + 1 2 σ 4 2 3
holds, then the system (6) has a unique stationary distribution π ( · ) and it has ergodic property.
By comparing the above conclusions (II) and (III), we can see that when system (6) has a ergodic stationary distribution, then the infective individuals i ˜ ( t ) are permanent in mean almost surely. However, it is not applicable in reverse. The above results of Theorems 2–4 show a large stochastic disturbance can cause infectious diseases to go to extinction, in other words, the persistent infectious disease of a deterministic system can become extinct due to the white noise stochastic disturbance. This implies that stochastic fluctuations can suppress the disease outbreak.

Author Contributions

The work presented in this paper has been accomplished through contributions of all authors. All authors contributed equally to the writing of this paper. All authors read and approved the final manuscript.

Acknowledgments

This work are supported by National Natural Science Foundation of China (11561004), the SDUST Research Fund (2014TDJH102), Joint Innovative Center for Safe And Effective Mining Technology and Equipment of Coal Resources, Shandong Province, the Research Fund for the Taishan Scholar Project of Shandong Province of China, SDUST Innovation Fund for Graduate Students (SDKDYC180226).

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A.

Proof of Theorem 1.
Define a C 2 -function V: R + 3 R + by
V s ˜ , e ˜ , i ˜ = ln 1 s ˜ e ˜ i ˜ ln s ˜ ln e ˜ ln i ˜ 3 .
d V = L V d t + σ 1 4 s ˜ 1 d B 1 ( t ) + σ 2 4 e ˜ 1 d B 2 ( t ) + σ 3 4 i ˜ 1 d B 3 ( t ) + σ 4 3 4 s ˜ 4 e ˜ 4 i ˜ d B 4 ( t ) ,
where
L V = b 2 s ˜ + e ˜ + i ˜ 1 s ˜ 1 s ˜ e ˜ i ˜ ( 1 δ p ) β i ˜ 2 s ˜ + e ˜ + i ˜ 1 1 s ˜ e ˜ i ˜ ( δ p + b ) 2 s ˜ + e ˜ + i ˜ 1 1 s ˜ e ˜ i ˜ + ε i ˜ 4 s ˜ + 4 e ˜ + 4 i ˜ 3 1 s ˜ e ˜ i ˜ + σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 1 s ˜ e ˜ i ˜ 2 4 s ˜ + 4 e ˜ + 4 i ˜ 3 1 s ˜ e ˜ i ˜ σ 1 2 s ˜ 2 s ˜ + e ˜ + i ˜ 1 1 s ˜ e ˜ i ˜ + ( 1 δ p ) β s ˜ i ˜ s ˜ + 2 e ˜ + i ˜ 1 e ˜ 1 s ˜ e ˜ i ˜ ( α + b ) s ˜ + 2 e ˜ + i ˜ 1 1 s ˜ e ˜ i ˜ σ 2 2 e ˜ s ˜ + 2 e ˜ + i ˜ 1 1 s ˜ e ˜ i ˜ + α e ˜ s ˜ + e ˜ + 2 i ˜ 1 i ˜ 1 s ˜ e ˜ i ˜ ( ε + γ + b ) s ˜ + e ˜ + 2 i ˜ 1 1 s ˜ e ˜ i ˜ σ 3 2 i ˜ s ˜ + e ˜ + 2 i ˜ 1 1 s ˜ e ˜ i ˜ + 1 2 σ 1 2 4 s ˜ 2 2 s ˜ + 1 + 1 2 σ 2 2 4 e ˜ 2 2 e ˜ + 1 + 1 2 σ 3 2 4 i ˜ 2 2 i ˜ + 1 + 1 2 σ 4 2 3 1 s ˜ e ˜ i ˜ 2 + s ˜ + e ˜ + i ˜ 2 ,
since r ˜ = 1 s ˜ e ˜ i ˜ and δ p < 1 , thus
L V b s ˜ r ˜ s ˜ r ˜ ( 1 δ p ) β i ˜ s ˜ r ˜ r ˜ ( δ p + b ) s ˜ r ˜ r ˜ + ε i ˜ 1 4 r ˜ r ˜ + ( 1 δ p ) β s ˜ i ˜ e ˜ r ˜ e ˜ r ˜ ( α + b ) e ˜ r ˜ r ˜ + α e ˜ i ˜ r ˜ i ˜ r ˜ ( ε + γ + b ) i ˜ r ˜ r ˜ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 + 1 2 σ 4 2 = b r ˜ b s ˜ β s ˜ i ˜ r ˜ + β i ˜ + β δ p s ˜ i ˜ r ˜ β δ p i ˜ δ p s ˜ r ˜ + δ p b s ˜ r ˜ + b + ε i ˜ r ˜ 4 ε i ˜ ε i ˜ r ˜ + ε + β s ˜ i ˜ r ˜ β s ˜ i ˜ e ˜ β δ p s ˜ i ˜ r ˜ + β δ p s ˜ i ˜ e ˜ α e ˜ r ˜ + α b e ˜ r ˜ + b + α e ˜ r ˜ α e ˜ i ˜ γ i ˜ r ˜ + γ b i ˜ r ˜ + b + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 + 1 2 σ 4 2 b r ˜ b s ˜ r ˜ b e ˜ r ˜ b i ˜ r ˜ + β i ˜ + δ p + 3 b + ε + β s ˜ i ˜ e ˜ ( δ p 1 ) + α + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 + 1 2 σ 4 2 β + δ p + 4 b + ε + α + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 + 1 2 σ 4 2 : = M 0 ,
where M 0 is a positive constant.  ☐

Appendix B.

Proof of Theorem 2.
L V = 1 e ˜ + ϱ i ˜ ( 1 δ p ) β s ˜ i ˜ ( α + b ) e ˜ + ε e ˜ i ˜ + ϱ α e ˜ ϱ ( ε + γ + b ) i ˜ + ϱ ε i ˜ 2 + 1 2 σ 1 2 s ˜ 2 + 1 2 1 1 e ˜ + ϱ i ˜ 2 σ 2 2 e ˜ 2 + 1 2 1 ϱ 2 e ˜ + ϱ i ˜ 2 σ 3 2 i ˜ 2 + 1 2 σ 4 2 1 s ˜ e ˜ i ˜ 2 ,
since s ˜ = 1 e ˜ i ˜ r ˜ and δ p < 1 , thus
L V 1 e ˜ + ϱ i ˜ [ ( β δ p β ϱ ( ε + γ + b ) ) i ˜ + ( ε β + δ p β ) e ˜ i ˜ + ( ϱ ε β + δ p β ) i ˜ 2 + ( ϱ 1 ) α e ˜ ] + 1 2 σ 1 2 s ˜ 2 + 1 2 1 1 e ˜ + ϱ i ˜ 2 σ 2 2 e ˜ 2 + 1 2 1 ϱ 2 e ˜ + ϱ i ˜ 2 σ 3 2 i ˜ 2 + 1 2 σ 4 2 1 s ˜ e ˜ i ˜ 2 .
Take
ϱ = ( ε + γ + b α ) + ( ε + γ + b α ) 2 + 4 ( 1 δ p ) α β 2 α
such that ϱ ( ϱ 1 ) α = β δ p β ϱ ( ε + γ + b ) . Here, it is easy to see that ϱ ( 0 , 1 ) . Then one can derive that
d V ( ϱ 1 ) α + 1 2 σ 1 2 s ˜ 2 + 1 2 1 ϱ 2 σ 3 2 i ˜ 2 + 1 2 σ 4 2 1 s ˜ e ˜ i ˜ 2 d t σ 1 s ˜ d B 1 ( t ) + σ 2 e ˜ 1 e ˜ ϱ i ˜ e ˜ + ϱ i ˜ d B 2 ( t ) + σ 3 i ˜ ϱ e ˜ ϱ i ˜ e ˜ + ϱ i ˜ d B 3 ( t ) σ 4 1 s ˜ e ˜ i ˜ d B 4 ( t ) ( ϱ 1 ) α + M 1 2 d t σ 1 s ˜ d B 1 ( t ) + σ 2 e ˜ 1 e ˜ ϱ i ˜ e ˜ + ϱ i ˜ d B 2 ( t ) + σ 3 i ˜ ϱ e ˜ ϱ i ˜ e ˜ + ϱ i ˜ d B 3 ( t ) σ 4 1 s ˜ e ˜ i ˜ d B 4 ( t ) ,
here M 1 = max σ 1 2 , 1 ϱ 2 σ 3 2 , σ 4 2 .  ☐

Appendix C.

Proof of Theorem 3.
ln s ˜ ( t ) t + ln e ˜ ( t ) t + ln i ˜ ( t ) t = b 1 s ˜ ( 1 δ p ) β i ˜ + 3 ε i ˜ + ( 1 δ p ) β s ˜ i ˜ e ˜ + α e ˜ i ˜ + 3 2 σ 1 2 s ˜ 2 + 3 2 σ 2 2 e ˜ 2 + 3 2 σ 3 2 i ˜ 2 + 3 2 σ 4 2 1 s ˜ e ˜ i ˜ 2 δ p + 3 b + α + ε + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 + ln s ˜ ( 0 ) t + ln e ˜ ( 0 ) t + ln i ˜ ( 0 ) t + M ̲ ( t ) t b 1 s ˜ + ( 1 δ p ) β s ˜ i ˜ e ˜ + α e ˜ i ˜ ( 1 δ p ) β i ˜ ( δ p + 3 b + α + ε + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 ) + ln s ˜ ( 0 ) t + ln e ˜ ( 0 ) t + ln i ˜ ( 0 ) t + M ̲ ( t ) t 3 b ( 1 δ p ) β α 3 ( 1 δ p ) β i ˜ ( δ p + 3 b + α + ε + γ + 1 2 σ 1 2 + 1 2 σ 2 2 + 1 2 σ 3 2 ) + ln s ˜ ( 0 ) t + ln e ˜ ( 0 ) t + ln i ˜ ( 0 ) t + M ̲ ( t ) t .
 ☐

Appendix D.

Proof of Theorem 4.
Now let us define a positive-definite function V by
V = ln s ˜ + e ˜ + i ˜ ln s ˜ ln e ˜ ln i ˜ ln r ˜ : = V 1 + V 2 + V 3 + V 4 + V 5 .
From the Itô’s formula yields
L V 1 = 1 s ˜ + e ˜ + i ˜ [ b ( δ p + b ) s ˜ + ε s ˜ i ˜ b e ˜ + ε e ˜ i ˜ ( ε + γ + b ) i ˜ + ε i ˜ 2 σ 1 2 s ˜ 2 σ 2 2 e ˜ 2 σ 3 2 i ˜ 2 + s ˜ + e ˜ + i ˜ σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 r ˜ 2 ] + r ˜ 2 σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 s ˜ + e ˜ + i ˜ 2 2 s ˜ + e ˜ + i ˜ 2 = b s ˜ + e ˜ + i ˜ + δ p s ˜ + ( ε + γ ) i ˜ s ˜ + e ˜ + i ˜ ε i ˜ + σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 s ˜ + e ˜ + i ˜ σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + 1 2 σ 4 2 r ˜ 2 + r ˜ 2 σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 2 s ˜ + e ˜ + i ˜ 2 + b .
Similarly,
L V 2 = b s ˜ + ( 1 δ p ) β i ˜ ε i ˜ + δ p + b + 1 2 σ 1 2 1 2 σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 r ˜ 2 , L V 3 = ( 1 δ p ) β s ˜ i ˜ e ˜ ε i ˜ + α + b + 1 2 σ 2 2 1 2 σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 r ˜ 2 , L V 4 = α e ˜ i ˜ ε i ˜ + ε + γ + b + 1 2 σ 3 2 1 2 σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 r ˜ 2 , L V 5 = δ p s ˜ r ˜ γ i ˜ r ˜ ε i ˜ + b + 1 2 σ 4 2 1 2 σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 + σ 4 2 r ˜ 2 .
Therefore,
L V = b s ˜ + e ˜ + i ˜ b s ˜ ( 1 δ p ) β s ˜ i ˜ e ˜ α e ˜ i ˜ 5 ε i ˜ δ p s ˜ r ˜ γ i ˜ r ˜ + δ p s ˜ + ( ε + γ ) i ˜ s ˜ + e ˜ + i ˜ + σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 s ˜ + e ˜ + i ˜ + r ˜ 2 σ 1 2 s ˜ 2 + σ 2 2 e ˜ 2 + σ 3 2 i ˜ 2 2 s ˜ + e ˜ + i ˜ 2 + ( 1 δ p ) β i ˜ ( 3 σ 1 2 s ˜ 2 + 3 σ 2 2 e ˜ 2 + 3 σ 3 2 i ˜ 2 + 5 2 σ 4 2 r ˜ 2 ) + 1 2 σ 1 2 + σ 2 2 + σ 3 2 + σ 4 2 + δ p + 5 b + α + ε + γ b s ˜ ( 1 δ p ) β s ˜ i ˜ e ˜ α e ˜ i ˜ b s ˜ + e ˜ + i ˜ δ p s ˜ r ˜ γ i ˜ r ˜ + M 1 ,
here
M 1 = ( 1 δ p ) β + 2 δ p + 5 b + α + 2 ε + 2 γ + 2 σ 1 2 + 2 σ 2 2 + 2 σ 3 2 + 1 2 σ 4 2 .
 ☐

References

  1. Kermack, W.O.; McKendrick, A.G. Contributions to the mathematical theory of epidemics-I. Bull. Math. Biol. 1991, 53, 33–55. [Google Scholar] [PubMed]
  2. Cai, Y.; Kang, Y.; Banerjee, M.; Wang, W. A stochastic SIRS epidemic model with infectious force under intervention strategies. J. Differ. Equat. 2015, 259, 7463–7502. [Google Scholar] [CrossRef]
  3. Sun, G.; Jusup, M.; Jin, Z.; Wang, Y.; Wang, Z. Pattern transitions in spatial epidemics: Mechanisms and emergent properties. Phys. Life. Rev. 2016, 19, 43–73. [Google Scholar] [CrossRef] [PubMed]
  4. Zhang, T.; Meng, X.; Zhang, T. Global analysis for a delayed SIV model with direct and environmental transmissions. J. Appl. Anal. Comput. 2016, 6, 479–491. [Google Scholar]
  5. Zhou, L.; Fan, M. Dynamics of an SIR epidemic model with limited medical resources revisited. Nonlinear Anal. Real World Appl. 2012, 13, 312–324. [Google Scholar] [CrossRef]
  6. Huang, G.; Takeuchi, Y.; Ma, W.; Wei, D. Global stability for delay SIR and SEIR epidemic models with nonlinear incidence rate. Bull. Math. Biol. 2010, 72, 1192–1207. [Google Scholar] [CrossRef] [PubMed]
  7. Jiang, Z.; Ma, W.; Wei, J. Global Hopf bifurcation and permanence of a delayed SEIRS epidemic model. Math. Comput. Simul. 2016, 122, 35–54. [Google Scholar] [CrossRef]
  8. Wang, W.; Zhang, T.Q. Caspase-1-Mediated Pyroptosis of the Predominance for Driving CD4+ T Cells Death: A Nonlocal Spatial Mathematical Model. Bull. Math. Biol. 2018, 80, 540–582. [Google Scholar] [CrossRef] [PubMed]
  9. Zhang, S.; Meng, X.; Feng, T.; Zhang, T. Dynamics analysis and numerical simulations of a stochastic non-autonomous predator-prey system with impulsive effects. Nonlinear Anal. Hybrid Syst. 2017, 26, 19–37. [Google Scholar] [CrossRef]
  10. Zhang, S.; Meng, X.; Wang, X. Application of stochastic inequalities to global analysis of a nonlinear stochastic SIRS epidemic model with saturated treatment function. Adv. Differ. Equ. 2018, 2018, 50. [Google Scholar] [CrossRef]
  11. Liu, M.; Fan, M. Permanence of stochastic Lotka-Volterra systems. J. Nonlinear Sci. 2017, 27, 425–452. [Google Scholar] [CrossRef]
  12. Li, X.; Lin, X.; Lin, Y. Lyapunov-type conditions and stochastic differential equations driven by G-Brownian motion. J. Math. Anal. Appl. 2016, 439, 235–255. [Google Scholar] [CrossRef]
  13. Meng, X.; Wang, L.; Zhang, T. Global dynamics analysis of a nonlinear impulsive stochastic chemostat system in a polluted environment. J. Appl. Anal. Comput. 2016, 6, 865–875. [Google Scholar]
  14. Ma, H.; Jia, Y. Stability analysis for stochastic differential equations with infinite Markovian switchings. J. Math. Anal. Appl. 2016, 435, 593–605. [Google Scholar] [CrossRef]
  15. Liu, G.; Wang, X.; Meng, X.; Gao, S. Extinction and persistence in mean of a novel delay impulsive stochastic infected predator-prey system with jumps. Complexity 2017, 2017, 1–15. [Google Scholar] [CrossRef]
  16. Zhao, Y.; Zhang, W. Observer-based controller design for singular stochastic Markov jump systems with state dependent noise. J. Syst. Sci. Complex. 2016, 29, 946–958. [Google Scholar] [CrossRef]
  17. Ma, H.; Hou, T. A separation theorem for stochastic singular linear quadratic control problem with partial information. Acta Math. Appl. Sin. Engl. Ser. 2013, 29, 303–314. [Google Scholar] [CrossRef]
  18. Liu, X.; Li, Y.; Zhang, W. Stochastic linear quadratic optimal control with constraint for discrete-time systems. Appl. Math. Comput. 2014, 228, 264–270. [Google Scholar] [CrossRef]
  19. Wei, F.; Geritz, S.A.H.; Cai, J. A stochastic single-species population model with partial pollution tolerance in a polluted environment. Appl. Math. Lett. 2017, 63, 130–136. [Google Scholar] [CrossRef]
  20. Liu, Q.; Jiang, D. Stationary distribution and extinction of a stochastic SIR model with nonlinear perturbation. Appl. Math. Lett. 2017, 73, 8–15. [Google Scholar] [CrossRef]
  21. Liu, M.; He, X.; Yu, J. Dynamics of a stochastic regime-switching predator-prey model with harvesting and distributed delays. Nonlinear Anal. Hybrid Syst. 2018, 28, 87–104. [Google Scholar] [CrossRef]
  22. Gao, S.; Chen, L.; Nieto, J.; Torres, A. Analysis of a delayed epidemic model with pulse vaccination and saturation incidence. Vaccine 2006, 24, 6037–6045. [Google Scholar] [CrossRef] [PubMed]
  23. Miao, A.; Zhang, J.; Zhang, T. Sampath Aruna Pradeep, B.G. Threshold dynamics of a stochastic SIR model with vertical transmission and vaccination. Comput. Math. Methods Med. 2017, 2017, 4820183. [Google Scholar] [CrossRef] [PubMed]
  24. Zhao, Y.; Jiang, D. The threshold of a stochastic SIS epidemic model with vaccination. Appl. Math. Comput. 2014, 243, 718–727. [Google Scholar] [CrossRef]
  25. Meng, X.; Chen, L.; Wu, B. A delay SIR epidemic model with pulse vaccination and incubation times. Nonlinear Anal. Real World Appl. 2010, 11, 88–98. [Google Scholar] [CrossRef]
  26. Gao, S.; Zhong, D.; Zhang, Y. Analysis of novel stochastic switched SILI epidemic models with continuous and impulsive control. Phys. A 2018, 495, 162–171. [Google Scholar] [CrossRef]
  27. Leng, X.; Feng, T.; Meng, X. Stochastic inequalities and applications to dynamics analysis of a novel SIVS epidemic model with jumps. J. Inequal. Appl. 2017, 1, 138. [Google Scholar] [CrossRef] [PubMed]
  28. Sun, C.; Hsieh, Y.H. Global analysis of an SEIR model with varying population size and vaccination. Appl. Math. Model. 2010, 34, 2685–2697. [Google Scholar] [CrossRef]
  29. Zhao, Y.; Jiang, D.; Mao, X.; Gray, A. The threshold of a stochastic SIRS epidemic model in a population with varying size. Discrete Contin. Dyn. Syst. Ser. B 2015, 20, 1277–1295. [Google Scholar] [CrossRef]
  30. Chen, L.; Wei, F. Persistence and distribution of a stochastic susceptible-infected-removed epidemic model with varying population size. Phys. A 2017, 483, 386–397. [Google Scholar] [CrossRef]
  31. Meng, X.; Zhao, S.; Feng, T.; Zhang, T. Dynamics of a novel nonlinear stochastic SIS epidemic model with double epidemic hypothesis. J. Math. Anal. Appl. 2016, 433, 227–242. [Google Scholar] [CrossRef]
  32. Zhao, Y.; Yuan, S.; Zhang, T. The stationary distribution and ergodicity of a stochastic phytoplankton allelopathy model under regime switching. Commun. Nonlinear Sci. Numer. Simul. 2016, 37, 131–142. [Google Scholar] [CrossRef]
  33. Feng, T.; Meng, X.; Liu, L.; Gao, S. Application of inequalities technique to dynamics analysis of a stochastic eco-epidemiology model. J. Inequal. Appl. 2016, 2016, 327. [Google Scholar] [CrossRef]
  34. Liu, Q.; Jiang, D.; Shi, N.; Hayat, T.; Alsaedi, A. Stationary distribution and extinction of a stochastic SEIR epidemic model with standard incidence. Phys. A 2017, 476, 58–69. [Google Scholar] [CrossRef]
  35. Yu, J.; Liu, M. Stationary distribution and ergodicity of a stochastic food-chain model with Lévy jumps. Phys. A 2017, 482, 14–28. [Google Scholar] [CrossRef]
  36. Li, F.; Meng, X.; Wang, X. Analysis and Numerical Simulations of a Stochastic SEIQR Epidemic System with Quarantine-Adjusted Incidence and Imperfect Vaccination. Comput. Math. Methods Med. 2018, 2018, 7873902. [Google Scholar] [CrossRef] [PubMed]
  37. Mao, X. Stochastic Differential Equations and Applications, 2nd ed.; Horwood: Chichester, UK, 2008. [Google Scholar]
  38. Chang, Z.; Meng, X.; Lu, X. Analysis of a novel stochastic SIRS epidemic model with two different saturated incidence rates. Phys. A 2017, 472, 103–116. [Google Scholar] [CrossRef]
  39. Mattingly, J.C.; Stuart, A.M. Geometric ergodicity of some hypoelliptic diffusion for particle motions. Markov Process Related Fields 2002, 8, 199–214. [Google Scholar]
  40. Mattingly, J.C.; Stuart, A.M.; Higham, D.J. Ergodicity for SDEs and approximations: locally Lipschitz vector fields and degenerate noise. Stoch. Process. Appl. 2002, 101, 185–232. [Google Scholar] [CrossRef] [Green Version]
  41. Khasminskii, R.Z.; Milstein, G.N.; Nevelon, M.B. Stochastic Stability of Differential Equations; Springer: Berlin, Germany, 2012. [Google Scholar]
  42. Gard, T.C. Introduction to Stochastic Differential Equations; Dekker: New York, NY, USA, 1988. [Google Scholar]
  43. Zhu, C.; Yin, G. Asymptotic properties of hybrid diffusion systems. SIAM J. Control Optim. 2007, 46, 1155–1179. [Google Scholar] [CrossRef]
Figure 1. Time sequence diagram of system (6) for extinctions of the exposed and infective individuals.
Figure 1. Time sequence diagram of system (6) for extinctions of the exposed and infective individuals.
Entropy 20 00376 g001
Figure 2. Time sequence diagram of system (6) for permanence in mean of the infective individuals.
Figure 2. Time sequence diagram of system (6) for permanence in mean of the infective individuals.
Entropy 20 00376 g002
Figure 3. (ac) represent the solutions of system (6); (de) stand for the density functions of s ˜ ( t ) , e ˜ ( t ) and i ˜ ( t ) , respectively.
Figure 3. (ac) represent the solutions of system (6); (de) stand for the density functions of s ˜ ( t ) , e ˜ ( t ) and i ˜ ( t ) , respectively.
Entropy 20 00376 g003

Share and Cite

MDPI and ACS Style

Han, X.; Li, F.; Meng, X. Dynamics Analysis of a Nonlinear Stochastic SEIR Epidemic System with Varying Population Size. Entropy 2018, 20, 376. https://0-doi-org.brum.beds.ac.uk/10.3390/e20050376

AMA Style

Han X, Li F, Meng X. Dynamics Analysis of a Nonlinear Stochastic SEIR Epidemic System with Varying Population Size. Entropy. 2018; 20(5):376. https://0-doi-org.brum.beds.ac.uk/10.3390/e20050376

Chicago/Turabian Style

Han, Xiaofeng, Fei Li, and Xinzhu Meng. 2018. "Dynamics Analysis of a Nonlinear Stochastic SEIR Epidemic System with Varying Population Size" Entropy 20, no. 5: 376. https://0-doi-org.brum.beds.ac.uk/10.3390/e20050376

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop