Skip to main content

Advertisement

Existence of positive periodic solutions for a generalized predator-prey model with diffusion feedback controls

Article metrics

  • 1593 Accesses

  • 2 Citations

Abstract

By employing the continuation theorem of coincidence degree theory, we derive a sufficient condition for the existence and attractivity of at least a positive periodic solution of the generalized predator-prey model with exploited term.

MSC:92D25, 34C25, 34K15.

1 Introduction

In recent years, the existence of positive periodic solutions of the prey-predator model has been widely studied [13]. The qualitative analysis of predator-prey systems is an interesting mathematical problem and has attracted a great attention of many mathematicians and biologists [4, 5]. Recently, Xu and Chen [6] investigated the two-species ratio-dependent predator-prey different model with time delay. Since a realistic model requires the inclusion of the effect of changing environment, recently, Shihua and Feng [7] have considered the following model:

{ x 1 ( t ) = x 1 ( t ) ( a 1 ( t ) a 11 ( t ) x 1 ( t ) a 13 ( t ) x 3 ( t ) m ( t ) x 3 ( t ) + x 1 ( t ) ) x 1 ( t ) = + D 1 ( t ) ( x 2 ( t ) x 1 ( t ) ) h 1 ( t ) , x 2 ( t ) = x 2 ( t ) ( a 2 ( t ) a 22 ( t ) x 2 ( t ) ) + D 2 ( t ) ( x 1 ( t ) x 2 ( t ) ) h 2 ( t ) , x 3 ( t ) = x 3 ( t ) ( a 3 ( t ) a 4 ( t ) x 3 ( t ) + a 31 ( t ) x 1 ( t τ ) m ( t ) x 3 ( t τ ) + x 1 ( t τ ) ) h 3 ( t ) ,
(1.1)

where D i (t) (i=1,2), a i (t) (i=1,2,3), a 11 (t), a 13 (t), a 22 (t), a 31 (t) and m(t) are strictly positive continuous w-periodic functions.

In the paper, we will study the following model:

{ x 1 ( t ) = x 1 ( t ) ( g 1 ( t , x 1 ( t ) ) a 13 ( t ) x 3 ( t ) m ( t ) x 3 ( t ) + x 1 ( t ) ) x 1 ( t ) = + D 1 ( t ) ( x 2 ( t ) x 1 ( t ) ) h 1 ( t ) , x 2 ( t ) = x 2 ( t ) ( g 2 ( t , x 2 ( t ) ) + D 2 ( t ) ( x 1 ( t ) x 2 ( t ) ) h 2 ( t ) , x 3 ( t ) = x 3 ( t ) ( a 3 ( t ) a 4 ( t ) x 3 ( t ) + a 31 ( t ) x 1 ( t τ ) m ( t ) x 3 ( t τ ) + x 1 ( t τ ) ) h 3 ( t ) ,
(1.2)

where D i (t) (i=1,2), a 13 (t), a 31 (t), m(t) are the same as in model (1.1). Some assumptions on the above functions on g i (t,x) (i=1,2) will be given in next section.

Our aim in this paper is to establish a sufficient condition for the existence and attractivity of at least a positive w-periodic solution of model (1.2).

2 Main result

To obtain the existence of positive periodic solutions of system (1.2), we summarize some concepts and results from [5] that will be basic for this section.

Let X, Z be Banach spaces, let L:DomLXY be a linear mapping, and let N:XZ be a continuous mapping. The mapping L will be called a Fredholm mapping of index zero if dimKerL=codimImL<+ and ImL is closed in Z. If L is a Fredholm mapping of index zero, there exist continuous projectors P:XZ and Q:ZZ such that ImP=KerL and ImL=KerQ=Im(IQ). It follows that L/DomLKerP:(IP)XImL is invertible. We denote the inverse of that map by Kp. If Ω is an open-bounded subset of X, the mapping N will be called L-compact on Ω ¯ if QN( Ω ¯ ) is bounded and Kp(IQ)N: Ω ¯ X is compact. Since ImQ is isomorphic to KerL, there exists an isomorphism J:ImQKerL.

In the proof of our existence theorem, we will use the continuation theorem of Gaines and Mawhin [8].

Lemma 2.1 [8]

Let L be a Fredholm mapping of index zero and let N be L-compact on Ω ¯ . Suppose the following:

  1. (i)

    for each λ(0,1), every solution x of Lx=λNx is such that x ¯ Ω;

  2. (ii)

    QNx0 for each xΩKerL;

  3. (iii)

    deg{JQN,ΩKerL,0}0.

The Lx=Nx has at least one solution in DomL Ω ¯ .

For convenience, we introduce the notations

f ¯ = 1 w 0 w f(t)dt, f L = min t [ 0 , w ] | f ( t ) | , f M = max t [ 0 , w ] | f ( t ) | ,

where f is a continuous w-periodic function.

Our main result on the global existence of a positive periodic solution of system (1.2) is stated in the following theorem.

Theorem 2.1 Assume that

( H 1 ) there exists a constant A such that for xR, tR, when xA,

g 1 ( t , e x ) 0;

( H 2 ) there exists a constant B such that for xR, when xB,

g 2 ( t , e x ) 0;

( H 3 ) there exists a constant C (C<A) such that for xR, tR, when xC,

e x g 1 ( t , e x ) ( a 13 m ) M e x + h 1 M ;

( H 4 ) there exists a constant D (D<B) such that for xR, tR, when xD,

e x g 2 ( t , e x ) h 2 M ;

( H 5 ) a 31 M e d 1 > h 3 l m l .

Then system (1.1) has at least one positive w-periodic solution.

Proof

Consider the system

{ u 1 ( t ) = g 1 ( t , e u 1 ( t ) ) a 13 ( t ) e u 3 ( t ) m ( t ) e u 3 ( t ) + e u 1 ( t ) u 1 ( t ) = + D 1 ( t ) ( e u 2 ( t ) u 1 ( t ) 1 ) h 1 ( t ) e u 1 ( t ) , u 2 ( t ) = g 2 ( t , e u 2 ( t ) ) + D 2 ( t ) ( e u 1 ( t ) u 2 ( t ) 1 ) h 2 ( t ) e u 2 ( t ) , u 3 ( t ) = a 3 ( t ) a 4 ( t ) e u 3 ( t ) + a 3 ( t ) e u 1 ( t τ ) m ( t ) e u 3 ( t ) + e u 1 ( t τ ) h 3 ( t ) e u 3 ( t ) .
(2.1)

Let x i (t)= e u i ( t ) , i=1,2,3, then system (1.2) changes into system (2.1). Hence it is easy to see that system (2.1) has a w-periodic solution ( u 1 ( t ) , u 2 ( t ) , u 3 ( t ) ) T , then ( e u 1 ( t ) , e u 2 ( t ) , e u 3 ( t ) ) T is a positive w-periodic solution of system (1.2). Therefore, for (1.2) to have at least one positive w-periodic solution, it is sufficient that (2.1) has at least one w-periodic solution. In order to apply Lemma 2.1 to system (2.1), we take

X=Z= { u ( t ) = ( u 1 ( t ) , u 2 ( t ) , u 3 ( t ) ) T C ( R , R 3 ) , u ( t + w ) = u ( t ) }

and

u= ( u 1 ( t ) , u 2 ( t ) , u 3 ( t ) ) T = i = 1 3 max t [ 0 , w ] | u i ( t ) |

for any uZ (or Z). Then X and Z are Banach spaces with the norm . Let

N u = [ g 1 ( t , e u 1 ( t ) ) a 13 ( t ) e u 3 ( t ) m ( t ) e u 3 ( t ) + e u 1 ( t ) + D 1 ( t ) ( e u 2 ( t ) u 1 ( t ) 1 ) h 1 ( t ) e u 1 ( t ) g 2 ( t , e u 2 ( t ) ) + D 2 ( t ) ( e u 1 ( t ) u 2 ( t ) 1 ) h 2 ( t ) e u 2 ( t ) a 3 ( t ) a 4 ( t ) e u 3 ( t ) + a 3 ( t ) e u 1 ( t τ ) m ( t ) e u 3 ( t ) + e u 1 ( t τ ) h 3 ( t ) e u 3 ( t ) ] , u X , L u = u = d u ( t ) d t , p u = 1 w 0 w u ( t ) d t , u X ; Q z = 1 w 0 w z ( t ) d t , z Z .

Then it follows that

Ker L = R 3 , Im L = { z Z : 0 w z ( t ) d t = 0 }  is closed in  Z , dim Ker L = 3 = codim Im L ,

and P, Q are continuous projectors such that

ImP=KerL,KerQ=ImL=Im(IQ).

Therefore, L is a Fredholm mapping of index zero. Furthermore, the generalized inverse (to L) Kp:ImLKerPDomL is given by

Kp(z)= 0 t z(s)ds 1 w 0 w 0 t z(s)dsdt.

Thus

QNu= [ 1 / w 0 w F 1 ( s ) d s 1 / w 0 w F 2 ( s ) d s 1 / w 0 w F 3 ( s ) d s ]

and

Kp(IQ)Nu= [ 0 t F 1 ( s ) d s 1 / w 0 w 0 t F 1 ( s ) d s d t + ( 1 / 2 t / w ) 0 w F 1 ( s ) d s 0 t F 2 ( s ) d s 1 / w 0 w 0 t F 2 ( s ) d s d t + ( 1 / 2 t / w ) 0 w F 2 ( s ) d s 0 t F 3 ( s ) d s 1 / w 0 w 0 t F 3 ( s ) d s d t + ( 1 / 2 t / w ) 0 w F 3 ( s ) d s ] ,

where

F 1 ( s ) = g 1 ( s , e u 1 ( s ) ) a 13 ( s ) e u 3 ( s ) m ( s ) e u 3 ( s ) + e u 1 ( s ) + D 1 ( s ) ( e u 2 ( s ) u 1 ( s ) 1 ) h 1 ( s ) e u 1 ( s ) , F 2 ( s ) = g 2 ( s , e u 2 ( s ) ) + D 2 ( s ) ( e u 1 ( s ) u 2 ( s ) 1 ) h 2 ( s ) e u 2 ( s )

and

F 3 (s)= a 3 (s) a 4 (s) e u 3 ( s ) + a 3 ( s ) e u 1 ( s τ ) m ( s ) e u 3 ( s τ ) + e u 1 ( s τ ) h 3 (s) e u 3 ( s ) .

Obviously, QN and Kp(IQ)N are continuous. It is not difficult to show that Kp(IQ)N( Ω ¯ ) is compact for any open bounded ΩX by using the Arzela-Ascoli theorem. Moreover, QN( Ω ¯ ) is clearly bounded. Thus, N is L-compact on Ω ¯ with any open bounded set ΩX.

Now we reach the point where we search for an appropriate open bounded subset Ω for the application of the continuation theorem (Lemma 2.1). Corresponding to the operator equation Lx=λNx, λ(0,1), we have

{ u 1 ( t ) = λ [ g 1 ( t , e u 1 ( t ) ) a 13 ( t ) e u 3 ( t ) m ( t ) e u 3 ( t ) + e u 1 ( t ) u 1 ( t ) = + D 1 ( t ) ( e u 2 ( t ) u 1 ( t ) 1 ) h 1 ( t ) e u 1 ( t ) ] , u 2 ( t ) = λ [ g 2 ( t , e u 2 ( t ) ) + D 2 ( t ) ( e u 1 ( t ) u 2 ( t ) 1 ) h 2 ( t ) e u 2 ( t ) ] , u 3 ( t ) = λ [ a 3 ( t ) a 4 ( t ) e u 3 ( t ) + a 31 ( t ) e u 1 ( t τ ) m ( t ) e u 3 ( t ) + e u 1 ( t τ ) h 3 ( t ) e u 3 ( t ) ] .
(2.2)

Assume that u=u(t)X is a solution of system (2.2) for a certain λ(0,1).

Because of ( u 1 ( t ) , u 2 ( t ) , u 3 ( t ) ) T X, there exist ξ i , η i [0,w] such that

u i ( ξ i )= max t [ 0 , w ] u i (t), u i ( η i )= min t [ 0 , w ] u i (t),i=1,2,3.

It is clear that

u i ( ξ i )=0, u i ( η i )=0,i=1,2,3.

From this and system (2.2), we obtain

g 1 ( ξ 1 , e u 1 ( ξ 1 ) ) a 13 ( ξ 1 ) e u 3 ( ξ 1 ) m ( ξ 1 ) e u 3 ( ξ 1 ) + e u 1 ( ξ 1 ) + D 1 ( ξ 1 ) ( e u 2 ( ξ 1 ) u 1 ( ξ 1 ) 1 ) h 1 ( ξ 1 ) e u 1 ( ξ 1 ) =0,
(2.3)
g 2 ( ξ 2 , e u 2 ( ξ 2 ) ) + D 2 ( ξ 2 ) ( e u 1 ( ξ 2 ) u 2 ( ξ 2 ) 1 ) h 2 ( ξ 2 ) e u 2 ( ξ 2 ) =0,
(2.4)
a 3 ( ξ 3 ) a 4 ( ξ 3 ) e u 3 ( ξ 3 ) + a 31 ( ξ 3 ) e u 1 ( ξ 3 τ ) m ( ξ 3 ) e u 3 ( ξ 3 τ ) + e u 1 ( ξ 3 τ ) h 3 ( ξ 3 ) e u 3 ( ξ 3 ) =0,
(2.5)
g 1 ( η 1 , e u 1 ( η 1 ) ) a 13 ( η 1 ) e u 3 ( η 1 ) m ( η 1 ) e u 3 ( η 1 ) + e u 1 ( η 1 ) + D 1 ( η 1 ) ( e u 2 ( η 1 ) u 1 ( η 1 ) 1 ) h 1 ( η 1 ) e u 1 ( η 1 ) = 0 ,
(2.6)
g 2 ( η 2 , e u 2 ( η 2 ) ) + D 2 ( η 2 ) ( e u 1 ( η 2 ) u 2 ( η 2 ) 1 ) h 2 ( η 2 ) e u 2 ( η 2 ) =0,
(2.7)
a 3 ( η 3 ) a 4 ( η 3 ) e u 3 ( η 3 ) + a 31 ( η 3 ) e u 1 ( η 3 τ ) m ( η 3 ) e u 3 ( η 3 τ ) + e u 1 ( η 3 τ ) h 3 ( η 3 ) e u 3 ( η 3 ) =0.
(2.8)

There are two cases to consider for (2.3) and (2.4).

Case 1. Assume that u 1 ( ξ 1 ) u 2 ( ξ 2 ), then u 1 ( ξ 1 ) u 2 ( ξ 1 ).

From this and (2.3), we have

g 1 ( ξ 1 , e u 1 ( ξ 1 ) ) = a 13 ( ξ 1 ) e u 3 ( ξ 3 ) m ( ξ 1 ) e u 3 ( ξ 1 ) + e u 1 ( ξ 1 ) D 1 ( ξ 1 ) ( e u 2 ( ξ 1 ) u 1 ( ξ 1 ) 1 ) + h 1 ( ξ 1 ) e u 1 ( ξ 1 ) >0,

which, together with condition ( H 1 ) in Theorem 2.1, gives

u 1 ( ξ 1 )<A.
(2.9)

Thus

u 2 ( ξ 2 ) u 1 ( ξ 1 )<A.
(2.10)

Case 2. Assume that u 1 ( ξ 1 ) u 2 ( ξ 2 ), then u 1 ( ξ 2 )< u 2 ( ξ 2 ).

From this and (2.4), we have

g 2 ( ξ 2 , e u 2 ( ξ 2 ) ) = D 2 ( ξ 2 ) ( e u 1 ( ξ 2 ) u 2 ( ξ 2 ) 1 ) + h 2 ( ξ 2 ) e u 2 ( ξ 2 ) >0,

which, together with condition ( H 2 ) in Theorem 2.1, gives

u 2 ( ξ 2 )<B.
(2.11)

Thus

u 1 ( ξ 1 ) u 2 ( ξ 2 )<B.
(2.12)

From Case 1 and Case 2, we obtain

u 1 ( ξ 1 )<max{A,B} = def d 1 ,
(2.13)
u 2 ( ξ 2 )<max{A,B}= d 2 .
(2.14)

From (2.5), we get

a 4 e u 3 ( ξ 3 ) a 4 ( ξ 3 ) e u 3 ( ξ 3 ) a 31 ( ξ 3 ) e u 1 ( ξ 3 τ ) m ( ξ 3 ) e u 3 ( ξ 3 τ ) + e u 1 ( ξ 3 τ ) < a 31 M .

Thus

u 3 ( ξ 3 )ln ( a 31 M a 4 l ) = def d 3 .
(2.15)

There are two cases to consider for (2.6) and (2.7).

Case 1. Assume that u 1 ( η 1 ) u 2 ( η 2 ), then u 1 ( η 1 )< u 2 ( η 1 ). From this and (2.5), we have

g 1 ( η 1 , e u 1 ( η 1 ) ) = a 13 ( η 1 ) e u 3 ( η 1 ) m ( η 1 ) e u 3 ( η 1 ) + e u 1 ( η 1 ) D 1 ( η 1 ) ( e u 2 ( η 1 ) u 1 ( η 1 ) 1 ) + h 1 ( η 1 ) e u 1 ( η 1 ) < a 13 ( η 1 ) e u 3 ( η 1 ) m ( η 1 ) e u 3 ( η 1 ) + e u 1 ( η 1 ) + h 1 ( η 1 ) e u 1 ( η 1 ) < ( a 13 m ) M + h 1 M e u 1 ( η 1 ) ,

which, together with condition ( H 3 ) in Theorem 2.1, gives

u 1 ( η 1 )>C.
(2.16)

Hence

u 2 ( η 2 )> u 1 ( η 1 )>C.
(2.17)

Case 2. Assume that u 1 ( η 1 ) u 2 ( η 2 ), then u 1 ( η 2 ) u 2 ( η 2 ). From this and (2.7), we have

g 2 ( η 2 , e u 2 ( η 2 ) ) = D 2 ( η 2 ) ( e u 1 ( η 2 ) u 2 ( η 2 ) 1 ) + h 2 ( η 2 ) e u 2 ( η 2 ) < h 2 M e u 2 ( η 2 ) ,

which, together with condition ( H 4 ) in Theorem 2.1, gives

u 2 ( η 2 )>D.
(2.18)

Hence

u 1 ( η 1 )> u 2 ( η 2 )>D.
(2.19)

From Case 1 and Case 2, we have

u 1 ( η 1 )>min{C,D} = def ρ 1 ,
(2.20)
u 2 ( η 2 )>min{C,D}= ρ 1 .
(2.21)

From Theorem 2.1( H 5 ), we get

u 3 ( η 3 )>ln ( h 3 l a 31 M a 3 l ) .
(2.22)

From (2.11)-(2.22), we obtain, for tR,

| u 1 ( t ) | max { | d 1 | , | ρ 1 | } = def R 1 , | u 2 ( t ) | max { | d 1 | , | ρ 1 | } = def R 2

and

| u 3 ( t ) | max { | a 3 | , | ρ 3 | } = def R 3 .

Clearly, R i (i=1,2,3) are independent of λ. Denote M= i = 1 3 R i + R 0 , here R 0 is taken sufficiently large such that each solution ( α , β , γ ) T of the system

g 1 ( t , e α ) a ¯ 13 e γ m ( t 3 ) e γ + e α + D ¯ 1 ( e β α 1 ) h ¯ 1 e α = 0 , g 2 ( t 2 , e β ) + D ¯ 2 ( e α β 1 ) h ¯ 2 e β = 0 , a ¯ 3 a 4 e r + a ¯ 31 e α m ( t 4 ) e γ + e α h ¯ 3 e γ = 0
(2.23)

satisfies ( α , β , γ ) T =| α |+| β |+| γ |<M, provided that system (2.23) has a solution or a number of solutions, and that

max { | d 1 | , | ρ 1 | } +max { | d 1 | , | ρ 1 | } +max { | d 3 | , | ρ 3 | } <M,

where t i (0,w) will appear in QNu below.

Now we take Ω={u= ( u 1 ( t ) , u 2 ( t ) , u 3 ( t ) ) T X:u<M}. This satisfies condition (i) of Lemma 2.1. When uΩKerL=Ω R 3 , u is a constant vector in R 3 with i = 1 3 | u i |=M. If system (2.23) has one or more solutions, then

QNu= [ g 1 ( t 1 , e u 1 ) a ¯ 13 e u 3 m ( t 3 ) e u 3 + e u 1 + D ¯ 1 ( e u 2 u 1 1 ) h ¯ 1 e u 1 g 2 ( t 2 , e u 2 ) + D ¯ 2 ( e u 1 u 2 1 ) h ¯ 2 e u 2 a ¯ 3 a ¯ 4 e u 3 + a ¯ 13 e u 1 m ( t 4 ) e u 3 + e u 1 h ¯ 3 e u 3 ] ( 0 , 0 , 0 ) T ,

where t i (0,w) are one constant.

If system (2.23) does not have a solution, then naturally

QNu ( 0 , 0 , 0 ) T .

This shows that condition (ii) of Lemma 2.1 is satisfied finally. We will prove that condition (iii) of Lemma 2.1 is satisfied. We only prove that when uΩKerL=Ω R 3 , deg{JQNu,ΩKerL, ( 0 , 0 , 0 ) T }0. When uΩKerL=Ω R 3 , u is a constant vector in R 3 with i = 1 3 | u i |=M. Our proof will be broken into three steps as follows.

Step 1. We prove

deg { J Q N u , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ( g ( t 1 , e u 1 ) , g ( t 2 , e u 2 ) , a ¯ 3 a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 h 3 e u 3 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } .

To this end, we define the mapping ϕ 1 :DomL×[0,1]X by

ϕ 1 ( u 1 , u 2 , u 3 , μ 1 ) = [ g 1 ( t 1 , e u 1 ) g 2 ( t 2 , e u 2 ) a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ] + μ 1 [ a ¯ 13 e u 3 m ( t 3 ) e u 3 + e u 1 + D ¯ 1 ( e u 2 u 1 1 ) h ¯ 1 e u 1 D ¯ 2 ( e u 1 u 2 1 ) h ¯ 2 e u 2 a ¯ 3 h ¯ 3 e u 3 ] ,

where μ 1 [0,1] is a parameter, when u= ( u 1 , u 2 , u 3 ) T ΩKerL=Ω R 3 , u is a constant vector in R 3 with i = 1 3 | u i |=M. We will show that when uΩKerL, ϕ 1 ( u 1 , u 2 , u 3 , μ 1 )0, if the conclusion is not true, i.e., the constant vector u with i = 1 3 | u i |=M satisfies ϕ 1 ( u 1 , u 2 , u 3 , μ 1 )=0, then from

g 1 ( t 1 , e u 1 ) + μ 1 ( a ¯ 13 e u 3 m ( t 3 ) e u 3 + e u 1 + D ¯ 1 ( e u 2 u 1 1 ) ) h ¯ 1 e u 1 = 0 , g 2 ( t 2 , e u 2 ) + μ 1 ( D ¯ 2 ( e u 1 u 2 1 ) h ¯ 2 e u 2 ) = 0 , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 μ 1 ( a ¯ 3 + h ¯ 3 e u 3 ) = 0

it follows the arguments of (2.11)-(2.22) that

| u i |< R i ,i=1,2,3.

Thus

i = 1 3 | u i |< i = 1 3 R i <M,

which contradicts the fact that i = 1 3 | u i |=M.

According to topological degree theory, we have

deg { ( J Q N , Ω Ker L , ( 0 , 0 , 0 ) T ) } = deg { ϕ 1 ( u 1 , u 2 , u 3 , 1 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ϕ 1 ( u 1 , u 2 , u 3 , 0 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ( g 1 ( t 1 , e u 1 ) , g 2 ( t 2 , e u 2 ) , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } .

Step 2. We prove

deg { ( g 1 ( t 1 , e u 1 ) , g 2 ( t 2 , e u 2 ) , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ( a ¯ 1 a ¯ 11 e u 1 , g ( t 2 , e u 2 ) , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } ,

where a ¯ 1 , a ¯ 11 are two chosen positive constants such that

C<ln a ¯ 1 a ¯ 11 <A.

To this end, we define the mapping ϕ 2 :DomL×[0,1]X by

ϕ 2 ( u 1 , u 2 , u 3 , μ 2 ) = μ 2 [ a ¯ 1 a ¯ 11 e u 1 g 2 ( t 2 , e u 2 ) a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ] + ( 1 μ 2 ) [ g 1 ( t 1 , e u 1 ) g 2 ( t 2 , e u 2 ) a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ] = [ μ 2 ( a ¯ 1 a ¯ 11 e u 1 ) + ( 1 μ 2 ) g 1 ( t 1 , e u 1 ) g 2 ( t 2 , e u 2 ) a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ] ,

where μ 2 [0,1] is a parameter. We will prove that when uΩKerL, ϕ 2 ( u 1 , u 2 , u 3 , μ 2 ) ( 0 , 0 , 0 ) T . When uΩKerL=Ω R 3 , u is a constant vector in R 3 with i = 1 3 | u i |=M. Now we consider two possible cases:

(i) u 1 A;(ii) u 1 <A.
  1. (i)

    When u 1 A, from condition (iii) in Theorem 2.1, we have g( t 1 , e u 1 )0. Moreover, a ¯ 1 a ¯ 11 e u 1 a ¯ 1 a ¯ 11 e A <0, thus μ 2 ( a ¯ 1 a ¯ 11 e u 1 )+(1 μ 2 )g( t 1 , e u 1 )<0. Therefore, ϕ 1 ( u 1 , u 2 , u 3 , μ 2 ) ( 0 , 0 , 0 ) T .

  2. (ii)

    When u 1 <A, if u 1 C, from condition ( H 3 ) in Theorem 2.1, we have g( t 1 , e u 1 )>0. However, a ¯ 1 a ¯ 11 e u 1 a ¯ 1 a ¯ 11 e C >0. Therefore, ϕ 1 ( u 1 , u 2 , u 3 , μ 2 ) ( 0 , 0 , 0 ) T . If u 1 >C, we also consider two possible cases: (a) u 2 B; (b) u 2 <B. (a) When u 2 B, from condition ( H 2 ) in Theorem 2.1, we have

    g 2 ( t 2 , e u 2 ) <0.

Therefore ϕ 1 ( u 1 , u 2 , u 3 , μ 2 ) ( 0 , 0 , 0 ) T . (b) When u 2 <B, if u 2 D, then from condition ( H 4 ) in Theorem 2.1, we obtain g 2 ( t 2 , e u 2 )>0. Consequently, ϕ 2 ( u 1 , u 2 , u 3 , μ 2 ) ( 0 , 0 , 0 ) T . If u 2 >D, we can claim when uΩKerL=Ω R 3 , ϕ 2 ( u 1 , u 2 , u 3 , μ 2 ) ( 0 , 0 , 0 ) T . Otherwise, from

a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 =0

we have

e u 3 < a ¯ 31 a ¯ 4

and

e u 3 > a ¯ 4 e ρ 1 + ( a ¯ 4 e ρ 1 ) 2 + 4 a ¯ 4 m ( t 4 ) a ¯ 31 e ρ 1 2 a ¯ 4 m ( t 4 ) >0,

i.e.,

u 3 < ln a ¯ 31 ln a ¯ 4 , u 3 > ln a ¯ 4 e ρ 1 + ( a ¯ 4 e ρ 1 ) 2 + 4 a ¯ 4 m ( t 4 ) a ¯ 31 e ρ 1 2 a ¯ 4 m ( t 4 ) .

Thus

| u 1 | < max { | d 1 | , | ρ 1 | } , | u 2 | < max { | d 1 | , | ρ 1 | }

and

| u 3 |<max { | d 3 | , | ρ 3 | } .

Therefore

i = 1 3 | u i | < max { | d 1 | , | ρ 1 | } + max { | d 1 | , | ρ 1 | } + max { | d 3 | , | ρ 3 | } < M ,

which contradicts the fact that i = 1 3 | u i |=M. Based on the above discussion, for any uΩKerL, we have ϕ 2 ( u 1 , u 2 , u 3 , μ 2 ) ( 0 , 0 , 0 ) T . According to topological degree theory, we obtain

deg { ( g 1 ( t 1 , e u 1 ) , g 2 ( t 2 , e u 2 ) , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ϕ 2 ( u 1 , u 2 , u 3 , 1 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ϕ 2 ( u 1 , u 2 , u 3 , 0 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ( a 1 a 11 e u 1 , g 2 ( t 2 , e u 2 ) , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } .

Step 3. We prove

deg { ( a 1 a 11 e u 1 , g 2 ( t 2 , e u 2 ) , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ( a 1 a 11 e u 1 , a 2 a 22 e u 2 , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } .

To this end, we define the mapping ϕ 3 :DomL×[0,1]X by

ϕ 3 ( u 1 , u 2 , u 3 , μ 3 ) = μ 3 [ a 1 a 11 e u 1 a 2 a 22 e u 2 a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ] + ( 1 μ 3 ) [ a 1 a 11 e u 1 g 2 ( t 2 , e u 2 ) a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ] = [ a 1 a 11 e u 1 μ 3 ( a ¯ 2 a ¯ 22 e u 2 ) + ( 1 μ 3 ) g ( t 2 , e u 2 ) a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ] ,

where μ 3 [0,1] is a parameter and a 2 , a 22 are two chosen positive constants such that D<ln a 2 a 22 <B. We will prove that when uΩKerL, ϕ 3 ( u 1 , u 2 , u 3 , μ 2 ) ( 0 , 0 , 0 ) T . If it is not true, then the constant vector u satisfies ϕ 3 ( u 1 , u 2 , u 3 , μ 2 ) ( 0 , 0 , 0 ) T with i = 1 3 | u i |=M. Thus we have

(2.24) implies

C< u 1 =ln a 1 a 11 <A.
(2.27)

We claim that u 2 <B; otherwise, if u 2 B, then from condition ( H 2 ) in Theorem 2.1, we have

(1 u 3 ) g 2 ( t 2 , e u 2 ) <0.

Consequently,

μ 3 ( a 2 a 22 e u 2 ) +(1 μ 3 ) g 2 ( t 2 , e u 2 ) <0,

which contradicts (2.23). We also claim that u 2 >D. If u 2 D, then g 2 ( t 2 , e u 2 )>0. However, a 2 a 22 e u 2 > a 2 a 22 e D >0.

Thus

u 3 ( a 2 a 22 e u 2 ) +(1 μ 3 ) g 2 ( t 2 , e u 2 ) >0,

which contradicts (2.24). (2.26) gives

a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 =0,

that is,

u 3 < ln a ¯ 31 ln a ¯ 4 , u 3 > ln a ¯ 4 e ρ 1 + ( a ¯ 4 e ρ 1 ) 2 + 4 a ¯ 4 m ( t 4 ) a ¯ 31 e ρ 1 2 a ¯ 4 m ( t 4 ) .

Thus

| u 1 | < max { | d 1 | , | ρ 1 | } , | u 2 | < max { | d 1 | , | ρ 1 | }

and

| u 3 |<max { | d 3 | , | ρ 3 | } .

Therefore

i = 1 3 | u i | < max { | d 1 | , | ρ 1 | } + max { | d 1 | , | ρ 1 | } + max { | d 3 | , | ρ 3 | } < M ,

which leads to a contradiction. Therefore, by means of topological degree theory, we have

deg { ( a 1 a 11 e u 1 , g 2 ( t 2 , e u 2 ) , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ϕ 3 ( u 1 , u 2 , u 3 , 1 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ϕ 3 ( u 1 , u 2 , u 3 , 0 ) T , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ( a 1 a 11 e u 1 , a 2 a 22 e u 2 , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) , Ω Ker L , ( 0 , 0 , 0 ) T } .

From the proof of the three steps above, we obtain

deg { J Q N u , Ω Ker L , ( 0 , 0 , 0 ) T } = deg { ( a 1 a 11 e u 1 , a 2 a 22 e u 2 , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) , Ω Ker L , ( 0 , 0 , 0 ) T } .

Because of condition ( H 5 ) in Theorem 2.1, the system of algebraic equations

{ a 1 a 11 x = 0 , a 2 a 22 y = 0 , a ¯ 4 I + a ¯ 31 x m ( t 4 ) z + x = 0

has a unique solution ( x , y , z ) T which satisfies

x = a 1 a 11 >0, y = a 2 a 22 >0, z = a 4 x + ( a 4 x ) 2 + 4 a 4 m ( t 4 ) a ¯ 31 x 2 a ¯ 4 m ( t 4 ) >0.

Thus

deg { ( a 1 a 11 e u 1 , a 2 a 22 e u 2 , a ¯ 4 e u 3 + a ¯ 31 e u 1 m ( t 4 ) e u 3 + e u 1 ) , Ω Ker L , ( 0 , 0 , 0 ) T } = sign | a 11 x 0 0 0 a 22 y 0 0 a ¯ 4 z a ¯ 31 m ( t 4 ) x z [ m ( t 4 ) z + x ] 2 | = 1 .

Therefore, from (2.20), we have

deg { J Q N u , Ω Ker L , ( 0 , 0 , 0 ) T } =1.

This completes the proof of Theorem 2.1. □

3 An example

Consider the system

{ x 1 = x 1 ( t ) ( a 1 ( t ) a 11 ( t ) x 1 ( t ) a 13 ( t ) x 3 ( t ) m ( t ) x 3 ( t ) + x 1 ( t ) ) + D 1 ( t ) ( x 2 ( t ) x 1 ( t ) ) , x 2 = x 2 ( t ) ( a 2 ( t ) a 22 ( t ) x 2 ( t ) ) + D 2 ( t ) ( x 1 ( t ) x 2 ( t ) ) , x 3 = x 3 ( t ) ( a 3 ( t ) + a 31 ( t ) x 1 ( t T ) m ( t ) x 3 ( t T ) + x 1 ( t T ) ) ,
(3.1)

where τ>0 is a positive constant, all the parameters are positive continuous w-periodic functions with periodic w>0.

In Theorem 2.1, g 1 (t, e x )= a 1 (t) a 11 (t) e x , g 2 (t, e x )= a 2 (t) a 22 (t) e x . It is easily shown that if xln( a 1 m a 11 l )=A, then g 1 (t, e x )0 and if xln( a 2 M a 22 l )=B, then g 2 (t, e x )0. We also can show if

x ln a 1 M ( a 13 m ) M a 11 l = C , g 1 ( t , e x ) ( a 11 m ) M

and if xln a 2 M a 22 l =D, then g 2 (t, e x )>0.

( H 1 ) a 1 m > ( a 13 m ) M ,

( H 2 ) a ¯ 31 > a ¯ 3 .

By Theorem 2.1, we have the following theorem.

Theorem 3.1 If ( H 1 ) and ( H 2 ) hold, the system (3.1) has at least one positive w-periodic solution. Consider the system

{ x 1 = x 1 ( t ) ( a 1 ( t ) a 11 ( t ) x 1 ( t ) a 13 ( t ) x 1 ( t ) x 3 ( t ) m ( t ) x 3 2 ( t ) + x 1 2 ( t ) ) + D 1 ( t ) ( x 2 ( t ) x 1 ( t ) ) , x 2 = x 2 ( t ) ( a 2 ( t ) a 22 ( t ) x 2 ( t ) ) + D 2 ( t ) ( x 1 ( t ) x 2 ( t ) ) , x 3 = x 3 ( t ) ( a 3 ( t ) + a 31 ( t ) x 1 2 ( t τ ) m ( t ) x 3 2 ( t τ ) + x 1 2 ( t τ ) ) ,
(3.2)

where z>0 is a positive constant, all the parameters are positive continuous w-periodic functions with periodic w>0.

In Theorem 2.1, g 1 (t, e x )= a 1 (t) a 11 (t) e x , g 2 (t, e x )= a 2 (t) a 22 (t) e x . It is easily shown that if xln a 1 M a 11 l =A, then g 1 (t, e x )0 and if xln a 2 M a 22 l =B, then g 2 (t, e x )0. We also can show if xln a 1 M ( a 13 2 m ) M a 11 l =C, g 1 (t, e x ) ( a 13 2 m ) m and if xln a 2 M a 22 l =D, then g 2 (t, e x )0.

( H 1 ) a 1 M > ( a 13 2 m ) M ,

( H 2 ) a ¯ 31 > a ¯ 3 .

By Theorem 2.1, we have the following theorem.

Theorem 3.2 If ( H 1 ) and ( H 2 ) hold, system (3.2) has at least one positive w-periodic solution.

References

  1. 1.

    Li YK: Positive periodic solution of neutral predator-prey system. Appl. Math. Mech. 1999, 20(5):545-550. (in Chinese) 10.1007/BF02463752

  2. 2.

    Zhang ZQ, Wang ZC: The existence of a periodic solution for a generalized prey-predator system with delay. Math. Proc. Camb. Philos. Soc. 2004, 137: 475-487. 10.1017/S0305004103007527

  3. 3.

    Zhang ZQ, Zeng XW: Periodic solution of a nonautonomous stage-structured single species model with diffusion. Q. Appl. Math. 2005, 63(2):277-289.

  4. 4.

    Jose F, Santiago V: An approximation for prey-predator models with time delay. Physica D 1997, 110(3-4):313-322. 10.1016/S0167-2789(97)00124-3

  5. 5.

    Xiao YN, Chen LS: Modeling and analysis of predator-prey model with disease in the prey. Math. Biosci. 2001, 171(1):59-82. 10.1016/S0025-5564(01)00049-9

  6. 6.

    Xu R, Chen LS: Persistence and stability for two-species ratio-dependent predator-prey system with time delay in a two-patch environment. Comput. Math. Appl. 2000, 40: 577-588. 10.1016/S0898-1221(00)00181-4

  7. 7.

    Shihua C, Feng W: Positive periodic solution of two-species ratio-dependent predator-prey system with time delay in two-patch environment. Appl. Math. Comput. 2004, 150: 737-748. 10.1016/S0096-3003(03)00303-5

  8. 8.

    Gaines RE, Mawhin JL: Coincidence Degree and Non-Linear Differential Equations. Springer, Berlin; 1977.

Download references

Acknowledgements

This research is partially supported by the National Natural Science Foundation of China (11071222, 11101126), the Hunan provincial Natural Science Foundation of China (12JJ3008) and the Research Fund of Hunan provincial Education Department (11B113).

Author information

Correspondence to Chaoxiong Du.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

Each of the authors, CD and YW, contributed to each part of this study equally and read and approved the final version of the manuscript.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License (https://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and Permissions

About this article

Keywords

  • prey-predator model
  • diffusion
  • positive periodic solution
  • coincidence degree