In this paper, we consider the existence of periodic solutions for a class of nonlinear difference systems involving classical (ϕ1,ϕ2)-Laplacian. By using the least action principle, we obtain that the system with classical (ϕ1,ϕ2)-Laplacian has at least one periodic solution when potential function is (p,q)-sublinear growth condition, subconvex condition. The results obtained generalize and extend some known works.
Citation: Hai-yun Deng, Jue-liang Zhou, Yu-bo He. Existence of periodic solutions for a class of (ϕ1,ϕ2)-Laplacian discrete Hamiltonian systems[J]. AIMS Mathematics, 2023, 8(5): 10579-10595. doi: 10.3934/math.2023537
[1] | Huiping Zhang, Wangjin Yao . Three solutions for a three-point boundary value problem with instantaneous and non-instantaneous impulses. AIMS Mathematics, 2023, 8(9): 21312-21328. doi: 10.3934/math.20231086 |
[2] | Yang Wang, Yating Li, Yansheng Liu . Multiple solutions for a class of BVPs of fractional discontinuous differential equations with impulses. AIMS Mathematics, 2023, 8(3): 7196-7224. doi: 10.3934/math.2023362 |
[3] | Huiting He, Chungen Liu, Jiabin Zuo . Periodic solutions of a class of non-autonomous second-order discrete Hamiltonian systems. AIMS Mathematics, 2024, 9(2): 3303-3319. doi: 10.3934/math.2024161 |
[4] | Xiying Zheng, Bo Kong, Yao Yu . Quantum codes from $ \sigma $-dual-containing constacyclic codes over $ \mathfrak{R}_{l, k} $. AIMS Mathematics, 2023, 8(10): 24075-24086. doi: 10.3934/math.20231227 |
[5] | Shaomin Wang, Cunji Yang, Guozhi Cha . On the variational principle and applications for a class of damped vibration systems with a small forcing term. AIMS Mathematics, 2023, 8(9): 22162-22177. doi: 10.3934/math.20231129 |
[6] | Jian-Guo Liu, Wen-Hui Zhu, Yan He, Aly R. Seadawy . Complexiton solutions and periodic-soliton solutions for the (2+1)-dimensional BLMP equation. AIMS Mathematics, 2020, 5(1): 421-439. doi: 10.3934/math.2020029 |
[7] | Jean-Guy Caputo, Imene Khames, Arnaud Knippel . Nonlinear normal modes in a network with cubic couplings. AIMS Mathematics, 2022, 7(12): 20565-20578. doi: 10.3934/math.20221127 |
[8] | Ahmed Hussein Msmali . Positive solutions for a system of Hadamard fractional $ (\varrho_{1}, \varrho_{2}, \varrho_{3}) $-Laplacian operator with a parameter in the boundary. AIMS Mathematics, 2022, 7(6): 10564-10581. doi: 10.3934/math.2022589 |
[9] | Tingting Ma, Yuehua He . An efficient linearly-implicit energy-preserving scheme with fast solver for the fractional nonlinear wave equation. AIMS Mathematics, 2023, 8(11): 26574-26589. doi: 10.3934/math.20231358 |
[10] | Jiaying Ma, Yueqiang Song . On multi-bump solutions for a class of Schrödinger-Poisson systems with $ p $-Laplacian in $ \mathbb{R}^{3} $. AIMS Mathematics, 2024, 9(1): 1595-1621. doi: 10.3934/math.2024079 |
In this paper, we consider the existence of periodic solutions for a class of nonlinear difference systems involving classical (ϕ1,ϕ2)-Laplacian. By using the least action principle, we obtain that the system with classical (ϕ1,ϕ2)-Laplacian has at least one periodic solution when potential function is (p,q)-sublinear growth condition, subconvex condition. The results obtained generalize and extend some known works.
In this paper, we investigate the existence of periodic solutions for the following system involving classical (ϕ1,ϕ2)-Laplacian:
{Δϕ1(Δu1(n−1))=∇u1F(n,u1(n),u2(n))Δϕ2(Δu2(n−1))=∇u2F(n,u1(n),u2(n)), | (1.1) |
where Δ is the forward difference operator, n∈Z, F:Z×RN×RN→R and ϕm, m=1,2 satisfy the following condition:
(A0) ϕm is a homeomorphism from RN onto RN such that ϕm(0)=0, ϕm=∇Φm, where Φm∈C1(RN,[0,+∞)) strictly convex and Φm(0)=0, m=1,2.
(F) F(n,x1,x2) is continuously differentiable in (x1,x2), there exist a1,a2∈C(R+,R+), b:Z[1,T]→R+ such that
|F(n,x1,x2)|≤[a1(|x1|)+a2(|x2|]b(n), |
|∇x1F(n,x1,x2)|≤[a1(|x1|)+a2(|x2|]b(n), |
|∇x2F(n,x1,x2)|≤[a1(|x1|)+a2(|x2|]b(n). |
Remark 1.1. Assumption (A0) given in [1] is used to characterize the classical homeomorphisms. Moreover, if Φm:RN→R is coercive (i.e., Φm(x)→+∞ as |x|→∞), then there exists δm=min|x|=1Φm(x)>0,m=1,2 such that
Φm(x)≥δm(|x|−1),x∈RN. | (1.2) |
In recent years, critical point theory (see [2,3,4,5,6]) plays an important role in studying the Hamiltonian systems, nonlinear differential equations, nonlinear difference system, etc (see [7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32]). In [1] and [7], by virtue of critical point theorem, Mawhin investigated the existence and multiplicity of periodic solutions for the following nonlinear difference systems with ϕ-Laplacian:
Δϕ[Δu(n−1)]=∇uF[n,u(n)]+h(n)(n∈Z), | (1.3) |
where ϕ is a homeomorphism from X⊂RN onto Y⊂RN and the following three different types of homeomorphisms were discussed:
(1) classical homeomorphism: if ϕ:RN→RN;
(2) bounded homeomorphism: if ϕ:RN→Ba(a<+∞);
(3) singular homeomorphism: if ϕ:Ba⊂RN→RN,
where Ba is a ball with its center at origin and radius a.
Inspired by [1,7], in [20], by using some abstract critical point theorems, the authors obtained some multiplicity results of periodic solutions for difference systems involving (ϕ1,ϕ2)-Laplacian. In [21], by using Clark theorem, the authors obtained that a class of nonlinear difference systems involving classical (ϕ1,ϕ2)-Laplacian has at least m distinct pairs of homoclinic solutions. In [22], by using an important three critial point theorem, the authors obtained that a class of (ϕ1,ϕ2)-Laplacian system has at least three T-periodic solutions. In [23], by using the least action principle and saddle point theorem, the authors obtained that system with classical and bounded (ϕ1,ϕ2)-Laplacian has at least one periodic solution when F has (p,q)-sublinear growth. In [24], the authors assumed that the nonlinear term F in system satisfies a corresponding sub-linear growth condition in Orlicz-Sobolev space, by using the least action principle, they obtained that nonlinear and non-homogeneous elliptic system involving (ϕ1,ϕ2)-Laplacian has at least a nontrivial solution, and by using the genus theory, obtained that system has infinitely many solutions under an additional symmetric condition.
As Φ1(x)=Φ2(x)=12|x|2 and F(n,x,y)≡F(n,x), system (1.1) reduces to the following second order discrete Hamiltonian system:
Δ2u(n−1)=∇F(n,u(n)),∀n∈Z. | (1.4) |
In [10], Guan and Yang investigated the existence of periodic solutions for system (1.4). By using the variational minimizing method and the saddle point theorem, they obtained that system (1.4) has at least one T-periodic solution.
Theorem A. (see [10]) Suppose that F(n,x)=F1(n,x)+F2(x) satisfies
(H1) F(n,x)∈C1(RN,R) for any n∈Z, F(n+T,x)=F(n,x) for all (n,x)∈Z×RN, T is a positive integer;
(H2) there exist f,g:Z[1,T]→R+ and α∈[0,1) such that
|∇F1(n,x)|≤f(n)|x|α+g(n),forall(n,x)∈Z[1,T]×RN; |
(H3) there exist constants r>0 and γ∈[0,2) such that
(∇F2(x)−∇F2(y),x−y)≥−r|x−y|γ,forallx,y∈RN; |
(H4) liminf|x|→+∞|x|−2αT∑n=1F(n,x)>18sin2πTT∑n=1f2(n).
Then system (1.4) has at least one T-periodic solution.
Theorem B. (see [10]) Suppose that F(n,x)=F1(n,x)+F2(x) satisfies (H1), (H2), (H3) and the following conditions:
(H5) there exist δ∈[0,2) and C>0 such that
(∇F2(x)−∇F2(y),x−y)≤C|x−y|δ,forallx,y∈RN; |
(H6) limsup|x|→+∞|x|−2αT∑n=1F(n,x)<−38sin2πTT∑n=1f2(n).
Then system (1.4) possesses at least one T-periodic solution.
Motivated by [10,20,21,22,23], in this paper, we consider the existence of periodic solutions for system (1.1) and obtain the following main results.
Theorem 1.1. Suppose that F(n,x1,x2)=F1(n,x1,x2)+F2(x1,x2) and the following conditions hold:
(A1) there exist constants d1>1p, d2>1q, p>1 and q>1 such that
Φ1(x1)+Φ2(x2)≥d1|x1|p+d2|x2|q,∀x1,x2∈RN; |
(F0) F(n,x1,x2)∈C1(RN×RN,R) for every n∈Z, and F(n+T,x1,x2)=F(n,x1,x2) for all (n,x1,x2)∈Z×RN×RN;
(F1) there exist fi,hi:Z[1,T]→R+,i=1,2 and α1∈[0,p−1), α2∈[0,q−1) such that
|∇x1F1(n,x1,x2)|≤f1(n)|x1|α1+h1(n), |
|∇x2F1(n,x1,x2)|≤f2(n)|x2|α2+h2(n), |
for all (n,x1,x2)∈Z[1,T]×RN×RN;
(F2) there exist constants ri∈(0,+∞),i=1,2, γ1∈[0,p) and γ2∈[0,q) such that
(∇x1F2(x1,x2)−∇y1F2(y1,y2),x1−y1)≥−r1|x1−y1|γ1 |
and
(∇x2F2(x1,x2)−∇y2F2(y1,y2),x2−y2)≥−r2|x2−y2|γ2 |
for all (x1,x2), (y1,y2)∈RN×RN;
(F3)
lim|x1|+|x2|→∞inf1|x1|qα1+|x2|pα2T∑n=1F(n,x1,x2)>M |
where
M=max{2qα1q[C(p,p′)]qpT∑n=1f1(n)q, 2pα2p[C(q,q′)]pqT∑n=1f2(n)p}, |
C(p,p′)=min{(T−1)2p−1Tp−1,Tp−1Θ(p′,p)(p′+1)p/p′}, 1/p+1/p′=1, |
Θ(p′,p)=T∑n=1[(nT)p′+1+(1−nT+1T)p′+1−2Tp′+1]p/p′. |
Then system (1.1) possesses at least one T-periodic solution.
Theorem 1.2. Suppose that F(n,x1,x2)=F1(n,x1,x2)+F2(x1,x2), (A1), (F0), (F1), (F3) and the following conditions hold:
(F4) there exist constants r1∈[0,pd1−1C(p,p′)), r2∈[0,+∞), r3∈[0,qd2−1C(q,q′)), r4∈[0,+∞), α0∈[0,p) and β0∈[0,q) such that
(∇x1F2(x1,x2)−∇y1F2(y1,y2),x1−y1)≥−r1|x1−y1|p−r2|x1−y1|α0 |
and
(∇x2F2(x1,x2)−∇y2F2(y1,y2),x2−y2)≥−r3|x2−y2|q−r4|x2−y2|β0 |
for all (x1,x2), (y1,y2)∈RN×RN.
Then system (1.1) possesses at least one T-periodic solution.
By Theorem 1.2, it is easy to obtain the following corollary.
Corollary 1.3. Suppose that F(n,x1,x2)=F1(n,x1,x2)+F2(x1,x2), (A1), (F0), (F1), (F3) and the following conditions hold:
(F5) there exist constants r1∈[0,pd1−1C(p,p′)), r2∈[0,+∞), r3∈[0,qd2−1C(q,q′)), r4∈[0,+∞), α0∈[0,p) and β0∈[0,q) such that
|∇x1F2(x1,x2)−∇y1F2(y1,y2)|≥−r1|x1−y1|p−1−r2|x1−y1|α0−1 |
and
|∇x2F2(x1,x2)−∇y2F2(y1,y2)|≥−r3|x2−y2|q−1−r4|x2−y2|β0−1 |
for all (x1,x2), (y1,y2)∈RN×RN.
Then system (1.1) possesses at least one T-periodic solution.
Remark 1.2. Theorem 1.1, Theorem 1.2 and Corollary 1.3 generalize Theorem A. In fact, when Φ1(x)=Φ2(x)=12|x|2 and F(n,x,y)≡F(n,x), System (1.1) reduces to system (1.4) and Theorem 1.1 become Theorem A. Moreover, Theorem 1.2 and Corollary 1.3 is still a new result, which shows that (F2) can be weaken to (F4).
Theorem 1.4. Suppose that F(n,x1,x2)=F1(n,x1,x2)+F2(n,x1,x2), F1 and F2 satisfy (F), (A1), (F0), (F1) and the following conditions hold:
(F6)
F2(n,x1,x2)≥0, |
for all (n,x1,x2)∈Z×RN×RN with |x1|>1, |x2|>1;
(F7) F2(n,.,.) is (λ,μ)-subconvex with λ=(λ1,λ2), μ=(μ1,μ2) and λ1>1p, μ1<2p−1λp1, λ2>1q, μ2<2q−1λq2, that is
F2(n,λ1(x1+y1),λ2(x2+y2))≤μ1F2(n,x1,x2)+μ2F2(n,y1,y2); |
(F8)
lim|x1|+|x2|→∞1|x1|qα1+|x2|pα2(T∑n=1F1(n,x1,x2)+1μ1T∑n=1F2(n,λ1x1,λ2x2))>M |
where M=max{2qα1q[C(p,p′)]qpT∑n=1f1(n)q, 2pα2p[C(q,q′)]pqT∑n=1f2(n)p}.
Then system (1.1) possesses at least one T-periodic solution.
First, we present some basic notations. We wse (⋅,⋅) and |⋅| to denote the inner product and the Euclidean norm in RN. Let
HT={v:={v(n)}|v(n+T)=v(n),v(n)∈RN,n∈Z}. |
For 1<s<+∞ and v∈HT, we define
‖v‖s=(T∑n=1|Δv(n)|s+T∑n=1|v(n)|s)1/s. |
For v∈HT, set
‖v‖[r]=(T∑n=1|v(n)|r)1/r,r>1and‖v‖∞=maxn∈Z[1,T]|v(n)|. |
Let E=HT×HT. For u=(u1,u2)τ∈E, define
‖u‖=‖u1‖p+‖u2‖q. |
For any u∈HT, it can be expressed as u(n)=¯u+˜u(n), where ¯u=1TT∑n=1u(n), and T∑n=1˜u(n)=0.
Lemma 2.1. (see [16])Let ˜u(n)=(˜u1(n),˜u2(n))τ, then
T∑n=1|˜um(n)|s≤C(s,s′)T∑n=1|Δum(n)|s,m=1,2. |
where 1/s+1/s′=1 and s,s′>1.
Lemma 2.2. (see [20])For any u=(u1,u2)τ,v=(v1,v2)τ∈E, the following two equalities hold:
−T∑n=1(Δϕ1(Δu1(n−1)),v1(n))=T∑n=1(ϕ1(Δu1(n)),Δv1(n)), |
−T∑n=1(Δϕ2(Δu2(n−1)),v2(n))=T∑n=1(ϕ2(Δu2(n)),Δv2(n)). |
Define
J(u)=J(u1,u2)=T∑n=1[Φ1(Δu1(n))+Φ2(Δu2(n))+F(n,u1(t),u2(n))]. |
then
⟨J′(u),v⟩=⟨J′(u1,u2),(v1,v2)⟩=T∑n=1[(ϕ1(Δu1(n)),Δv1(n))+(ϕ2(Δu2(n)),Δv2(n))+(∇u1F(n,u1(n),u2(n)),v1(n))+(∇u2F(n,u1(n),u2(n)),v2(n))]=−T∑n=1[(Δϕ1(Δu1(n−1)),v1(n))+(Δϕ2(Δu2(n−1)),v2(n))−(∇u1F(n,u1(n),u2(n)),v1(n))−(∇u2F(n,u1(n),u2(n)),v2(n))] |
and then it is easy to obtain that critical point of J in E is T-periodic solution of system (1.1).
Assume that E is a real Banach space and for φ∈C1(E,R), we say that φ satisfies the Palais-Smale(PS) condition if any sequence (un)⊂E for which φ(un) is bounded and φ′(un)→0 as n→∞ possesses a convergent subsequence.
Lemma 2.3. (see [6])Assume that X is a real Banach space, φ∈C1(X,R) is bounded from below and satisfies the (PS) condition, then c=infu∈Xφ(u) is a critical value of φ.
Proof of Theorem 1.1. It follows from (F1), Hölder inequality, Young inequality and Lemma 2.1 that
|T∑n=1[F1(n,u1(n),u2(n))−F1(n,u1(n),¯u2)]|=|T∑n=1∫10(∇¯u2+s˜u2(n)F1(n,u1(n),¯u2+s˜u2(n)),˜u2(n))ds|≤T∑n=1∫10f2(n)|¯u2+s˜u2(n)|α2|˜u2(n)|ds+T∑n=1∫10h2(n)|˜u2(n)|ds≤T∑n=1f2(n)(|¯u2|+|˜u2(n)|)α2|˜u2(n)|+T∑n=1h2(n)|˜u2(n)|≤T∑n=12α2f2(n)|¯u2|α2|˜u2(n)|+T∑n=12α2f2(n)|˜u2(n)|α2+1+T∑n=1h2(n)|˜u2(n)|≤2α2|¯u2|α2(T∑n=1f2(n)p)1p(T∑n=1|˜u2(n)|q)1q+2α2(T∑n=1f2(n)qq−α2−1)q−α2−1q(T∑n=1|˜u2(n)|q)α2+1q+(T∑n=1h2(n)p)1p(T∑n=1|˜u2(n)|q)1q≤2α2|¯u2|α2(T∑n=1f2(n)p)1p[C(q,q′)]1q(T∑n=1|Δu2(n)|q)1q+2α2(T∑n=1f2(n)qq−α2−1)q−α2−1q[C(q,q′)]α2+1q(T∑n=1|Δu2(n)|q)α2+1q+(T∑n=1h2(n)p)1p[C(q,q′)]1q(T∑n=1|Δu2(n)|q)1q≤2pα2p|¯u2|pα2[C(q,q′)]pqT∑n=1f2(n)p+1qT∑n=1|Δu2(n)|q+2α2[C(q,q′)]α2+1q(T∑n=1f2(n)qq−α2−1)q−α2−1q(T∑n=1|Δu2(n)|q)α2+1q+[C(q,q′)]1q(T∑n=1h2(n)p)1p(T∑n=1|Δu2(n)|q)1q | (3.1) |
and
|T∑n=1[F1(n,u1(n),¯u2)−F1(n,¯u1,¯u2)]|=|T∑n=1∫10(∇¯u1(n)+s˜u1(n)F1(n,¯u1(n)+s˜u1(n),¯u2),˜u1(n))ds|≤T∑n=1∫10f1(n)|¯u1+s˜u1(n)|α1|˜u1(n)|ds+T∑n=1∫10h1(n)|˜u1(n)|ds≤T∑n=1f1(n)(|¯u1|+|˜u1(n)|)α1|˜u1(n)|+T∑n=1h1(n)|˜u1(n)|≤2α1|¯u1|α1(T∑n=1f1(n)q)1q(T∑n=1|˜u1(n)|p)1p+2α1(T∑n=1f1(n)pp−α1−1)p−α1−1p(T∑n=1|˜u1(n)|p)α1+1p+(T∑n=1h1(n)q)1q(T∑n=1|˜u1(n)|p)1p≤2α1|¯u1|α1(T∑n=1f1(n)q)1q[C(p,p′)]1p(T∑n=1|Δu1(n)|p)1p+2α1(T∑n=1f1(n)pp−α1−1)p−α1−1p[C(p,p′)]α1+1p(T∑n=1|Δu1(n)|p)α1+1p+(T∑n=1h1(n)q)1q[C(p,p′)]1p(T∑n=1|Δu1(n)|p)1p≤2qα1q|¯u1|qα1[C(p,p′)]qpT∑n=1f1(n)q+1pT∑n=1|Δu1(n)|p+2α1[C(p,p′)]α1+1p(T∑n=1f1(n)pp−α1−1)p−α1−1p(T∑n=1|Δu1(n)|p)α1+1p+[C(p,p′)]1p(T∑n=1h1(n)q)1q(T∑n=1|Δu1(n)|p)1p. | (3.2) |
Then by (F2), Hölder inequality and Lemma 2.1, we have
T∑n=1[F2(u1(n),u2(n))−F2(u1(n),¯u2)]=T∑n=1∫10(∇¯u2+s˜u2(n))F2(u1(n),¯u2+s˜u2(n)),˜u2(n))ds=T∑n=1∫101s(∇¯u2+s˜u2(n)F2(u1(n),¯u2+s˜u2(n))−∇¯u2F(¯u1,¯u2),s˜u2(n))ds≥−r2T∑n=1∫101s|s˜u2(n)|γ2ds≥−r2γ2T∑n=1|˜u2(n)|γ2≥−r2γ2Tq−γ2q(T∑n=1|˜u2(n)|q)γ2q≥−r2γ2Tq−γ2q[C(q,q′)]γ2q(T∑n=1|Δu2(n)|q)γ2q | (3.3) |
and
T∑n=1[F2(u1(n),¯u2)−F2(¯u1,¯u2)]=T∑n=1∫10(∇¯u1+s˜u1(n)F2(¯u1+s˜u1(n),¯u2),˜u1(n))ds=T∑n=1∫101s(∇¯u1+s˜u1(n)F2(¯u1+s˜u1(n),¯u2−∇¯u1F(¯u1,¯u2),s˜u1(n))ds≥−r1T∑n=1∫101s|s˜u1(n)|γ1ds≥−r1γ1T∑n=1|˜u1(n)|γ1≥−r1γ1Tp−γ1p(T∑n=1|˜u1(n)|p)γ1p≥−r1γ1Tp−γ1p[C(p,p′)]γ1p(T∑n=1|Δu1(n)|p)γ1p. | (3.4) |
Then by (A1), (3.1), (3.2), (3.3), (3.4), we have
J(u1,u2)=T∑n=1[Φ1(Δu1(n))+Φ2(Δu2(n))+F(n,u1(n),u2(n))]≥d1T∑n=1|Δu1(n)|p+d2T∑n=1|Δu2(n)|q+T∑n=1[F1(n,u1(n),u2(n))−F1(n,u1(n),¯u2)]+T∑n=1[F1(n,u1(n),¯u2)−F1(n,¯u1,¯u2)]+T∑n=1[F2(u1(n),u2(n))−F2(u1(n),¯u2)]+T∑n=1[F2(u1(n),¯u2)−F2(¯u1,¯u2)]+T∑n=1F(n,¯u1,¯u2)≥(d1−1p)T∑n=1|Δu1(n)|p+(d2−1q)T∑n=1|Δu2(n)|q−[C(p,p′)]1p(T∑n=1h1(n)q)1q(T∑n=1|Δu1(n)|p)1p−[C(q,q′)]1q(T∑n=1h2(n)p)1p(T∑n=1|Δu2(n)|q)1q−2α1[C(p,p′)]α1+1p(T∑n=1f1(n)pp−α1−1)p−α1−1p(T∑n=1|Δu1(n)|p)α1+1p−2α2[C(q,q′)]α2+1q(T∑n=1f2(n)qq−α2−1)q−α2−1q(T∑n=1|Δu2(n)|q)α2+1q−r1γ1Tp−γ1p[C(p,p′)]γ1p(T∑n=1|Δu1(n)|p)γ1p−r2γ2Tq−γ2q[C(q,q′)]γ2q(T∑n=1|Δu2(n)|q)γ2q−2qα1q|¯u1|qα1[C(p,p′)]qpT∑n=1f1(n)q−2pα2p|¯u2|pα2[C(q,q′)]pqT∑n=1f2(n)p+T∑n=1F(n,¯u1,¯u2)≥(d1−1p)T∑n=1|Δu1(n)|p+(d2−1q)T∑n=1|Δu2(n)|q−[C(p,p′)]1p(T∑n=1h1(n)q)1q(T∑n=1|Δu1(n)|p)1p−[C(q,q′)]1q(T∑n=1h2(n)p)1p(T∑n=1|Δu2(n)|q)1q−2α1[C(p,p′)]α1+1p(T∑n=1f1(n)pp−α1−1)p−α1−1p(T∑n=1|Δu1(n)|p)α1+1p−2α2[C(q,q′)]α2+1q(T∑n=1f2(n)qq−α2−1)q−α2−1q(T∑n=1|Δu2(n)|q)α2+1q−r1γ1Tp−γ1p[C(p,p′)]γ1p(T∑n=1|Δu1(n)|p)γ1p−r2γ2Tq−γ2q[C(q,q′)]γ2q(T∑n=1|Δu2(n)|q)γ2q+(|¯u1|qα1+|¯u2|pα2)(1|¯u1|qα1+|¯u2|pα2T∑n=1F(n,¯u1,¯u2)−M), | (3.5) |
where M=max{2qα1q[C(p,p′)]qpT∑n=1f1(n)q,2pα2p[C(q,q′)]pqT∑n=1f2(n)p}. Note that (F3), α1∈[0,p−1), α2∈[0,q−1), γ1∈[0,p) and γ2∈[0,q). Hence, together with (3.5), implies that
J(u1,u2)→+∞,as‖(u1,u2)τ‖→∞. | (3.6) |
Hence J is bounded from below and (PS) condition holds. Then by Lemma 2.3, it is easy to know that J has at least one critical point u0 such that
J(uo)=infu∈EJ(u). |
Thus the proof is complete.
Proof of Theorem 1.2. It follows from (F4), Hölder inequality, Young inequality and Lemma 2.1 that
T∑n=1[F2(u1(n),u2(n))−F2(u1(n),¯u2)]=T∑n=1∫10(∇¯u2+s˜u2(n)F2(u1(n),¯u2+s˜u2(n)),˜u2(n))ds=T∑n=1∫101s(∇¯u2+s˜u2(n)F2(u1(n),¯u2+s˜u2(n))−∇¯u2F(¯u1,¯u2),s˜u2(n))ds≥T∑n=1∫101s(−r3|s˜u2(n)|q−r4|s˜u2(n)|β0)ds≥−r3qT∑n=1|˜u2(n)|q−r4β0T∑n=1|˜u2(n)|β0≥−r3qC(q,q′)T∑n=1|Δu2(n)|q−r4β0Tq−β0q[C(q,q′)]β0q(T∑n=1|Δu2(n)|q)β0q | (3.7) |
and
T∑n=1[F2(u1(n),¯u2)−F2(¯u1,¯u2)]=T∑n=1∫10(∇¯u1+s˜u1(n)F2(¯u1+s˜u1(n),¯u2),˜u1(n))ds=T∑n=1∫101s(∇¯u1+s˜u1(n)F2(¯u1+s˜u1(n),¯u2−∇¯u1F(¯u1,¯u2),s˜u1(n))ds≥T∑n=1∫101s(−r1|s˜u1(n)|p−r2|s˜u2(n)|α0)ds≥−r1pT∑n=1|˜u1(n)|p−r2α0T∑n=1|˜u1(n)|α0≥−r1pC(p,p′)T∑n=1|Δu1(n)|p−r2α0Tp−α0p[C(p,p′)]α0p(T∑n=1|Δu1(n)|p)α0p | (3.8) |
Then by (A1), (3.1), (3.2), (3.7), (3.8), we have
J(u1,u2)=T∑n=1[Φ1(Δu1(n))+Φ2(Δu2(n))+F(n,u1(n),u2(n))]≥(d1−1+r1C(p,p′)p)T∑n=1|Δu1(n)|p+(d2−1+r3C(q,q′)q)T∑n=1|Δu2(n)|q−[C(p,p′)]1p(T∑n=1h1(n)q)1q(T∑n=1|Δu1(n)|p)1p−[C(q,q′)]1q(T∑n=1h2(n)p)1p(T∑n=1|Δu2(n)|q)1q−2α1[C(p,p′)]α1+1p(T∑n=1f1(n)pp−α1−1)p−α1−1p(T∑n=1|Δu1(n)|p)α1+1p−2α2[C(q,q′)]α2+1q(T∑n=1f2(n)qq−α2−1)q−α2−1q(T∑n=1|Δu2(n)|q)α2+1q−r2α0Tp−α0p[C(p,p′)]α0p(T∑n=1|Δu1(n)|p)α0p−r4β0Tq−β0q[C(q,q′)]β0q(T∑n=1|Δu2(n)|q)β0q−2qα1q|¯u1|qα1[C(p,p′)]qpT∑n=1f1(n)q−2qα2p|¯u2|pα2[C(q,q′)]pqT∑n=1f2(n)p+T∑n=1F(n,¯u1,¯u2)≥(d1−1+r1C(p,p′)p)T∑n=1|Δu1(n)|p+(d2−1+r3C(q,q′)q)T∑n=1|Δu2(n)|q−[C(p,p′)]1p(T∑n=1h1(n)q)1q(T∑n=1|Δu1(n)|p)1p−[C(q,q′)]1q(T∑n=1h2(n)p)1p(T∑n=1|Δu2(n)|q)1q−2α1[C(p,p′)]α1+1p(T∑n=1f1(n)pp−α1−1)p−α1−1p(T∑n=1|Δu1(n)|p)α1+1p−2α2[C(q,q′)]α2+1q(T∑n=1f2(n)qq−α2−1)q−α2−1q(T∑n=1|Δu2(n)|q)α2+1q−r2α0Tp−α0p[C(p,p′)]α0p(T∑n=1|Δu1(n)|p)α0p−r4β0Tq−β0q[C(q,q′)]β0q(T∑n=1|Δu2(n)|q)β0q+(|¯u1|qα1+|¯u2|pα2)(1|¯u1|qα1+|¯u2|pα2T∑n=1F(n,¯u1,¯u2)−M), | (3.9) |
where M=max{2qα1q[C(p,p′)]qpT∑n=1f1(n)q,2pα2p[C(q,q′)]pqT∑n=1f2(n)p}. Note that (F3), r1∈[0,pd1−1C(p,p′)), r3∈[0,qd2−1C(q,q′)), α0∈[0,p) and β0∈[0,q). Hence, together with (3.9), implies that
J(u1,u2)→+∞,as ‖(u1,u2)τ‖→∞. | (3.10) |
Hence J is bounded from below and (PS) condition holds. Then by Lemma 2.3, it is easy to know that J has at least one critical point u0 such that
J(uo)=infu∈EJ(u). |
Thus the proof is complete.
Proof of Theorem 1.3. Let βi=log2λi(2μi), i=1,2 then β1<p, β2<q. For |xi|>1,i=1,2 there exist positive integers l, m such that
l−1≤log2λ1|x1|≤l, m−1≤log2λ2|x2|≤m |
then
|x1|β1>(2λ1)(l−1)β1=(2μ1)l−1, |x1|≤(2λ1)l, |
|x2|β2>(2λ2)(m−1)β2=(2μ2)m−1, |x2|≤(2λ2)m |
by (F), (F6) and (F7), we have
F2(n,x1,x2)=F2(n,λ1(x12λ1+x12λ1),λ2(x22λ2+x22λ2))≤μ1F2(n,x12λ1,x22λ2)+μ2F2(n,x12λ1,x22λ2)≤2μ1F2(n,x12λ1,x22λ2)+2μ2F2(n,x12λ1,x22λ2)≤2μ21F2(n,x1(2λ1)2,x2(2λ2)2)+4μ1μ2F2(n,x1(2λ1)2,x2(2λ2)2)+2μ22F2(n,x1(2λ1)2,x2(2λ2)2)≤(2μ1)2F2(n,x1(2λ1)2,x2(2λ2)2)+(2μ2)2F2(n,x1(2λ1)2,x2(2λ2)2)≤⋅⋅⋅≤(2μ1)KF2(n,x1(2λ1)K,x2(2λ2)K)+(2μ2)KF2(n,x1(2λ1)K,x2(2λ2)K)≤((2μ1)K−l+1|x1|β1+(2μ2)K−m+1|x2|β2)a0b(n) | (3.11) |
for a.e. n∈[0,T] and |x1|,|x2|>1, where a0=max0≤s≤1ai(s),i=1,2. Then, for a.e. n∈[0,T] and all |x1|,|x2|, we have
F2(n,x1,x2)≤((2μ1)K−l+1|x1|β1+(2μ2)K−m+1|x2|β2+1)a0b(n) | (3.12) |
Then by (A1), (3.1), (3.2), (3.12), we have
J(u1,u2)=T∑n=1[Φ1(Δu1(n))+Φ2(Δu2(n))+F(n,u1(n),u2(n))]≥d1T∑n=1|Δu1(n)|p+d2T∑n=1|Δu2(n)|q+T∑n=1[F1(n,u1(n),u2(n))−F1(n,u1(n),¯u2)]+T∑n=1[F1(n,u1(n),¯u2)−F1(n,¯u1,¯u2)]+T∑n=1F1(n,¯u1,¯u2)+T∑n=1F2(n,u1(n),u2(n))≥d1T∑n=1|Δu1(n)|p+d2T∑n=1|Δu2(n)|q−2qα1q|¯u1|qα1[C(p,p′)]qpT∑n=1f1(n)q−1pT∑n=1|Δu1(n)|p−2pα2p|¯u2|pα2[C(q,q′)]pqT∑n=1f2(n)p−1qT∑n=1|Δu2(n)|q−[C(p,p′)]1p(T∑n=1h1(n)q)1q(T∑n=1|Δu1(n)|p)1p−[C(q,q′)]1q(T∑n=1h2(n)p)1p(T∑n=1|Δu2(n)|q)1q−2α1[C(p,p′)]α1+1p(T∑n=1f1(n)pp−α1−1)p−α1−1p(T∑n=1|Δu1(n)|p)α1+1p−2α2[C(q,q′)]α2+1q(T∑n=1f2(n)qq−α2−1)q−α2−1q(T∑n=1|Δu2(n)|q)α2+1q+T∑n=1F1(n,¯u1,¯u2)+1μ1T∑n=1F2(n,λ1¯u1,λ2¯u2)−μ2μ1T∑n=1F2(n,−˜u1(n),−˜u2(n))≥(d1−1p)T∑n=1|Δu1(n)|p+(d2−1q)T∑n=1|Δu2(n)|q−[C(p,p′)]1p(T∑n=1h1(n)q)1q(T∑n=1|Δu1(n)|p)1p−[C(q,q′)]1q(T∑n=1h2(n)p)1p(T∑n=1|Δu2(n)|q)1q−2α1[C(p,p′)]α1+1p(T∑n=1f1(n)pp−α1−1)p−α1−1p(T∑n=1|Δu1(n)|p)α1+1p−2α2[C(q,q′)]α2+1q(T∑n=1f2(n)qq−α2−1)q−α2−1q(T∑n=1|Δu2(n)|q)α2+1q−μ2μ1T∑n=1((2μ1)K−l+1|−˜u1|β1+(2μ2)K−m+1|−˜u2|β2+1)a0b(n)+(|¯u1|qα1+|¯u2|pα2)(1|¯u1|qα1+|¯u2|pα2(T∑n=1F1(n,¯u1,¯u2)+1μ1T∑n=1F2(n,λ1¯u1,λ2¯u2))−M)≥(d1−1p)T∑n=1|Δu1(n)|p+(d2−1q)T∑n=1|Δu2(n)|q−[C(p,p′)]1p(T∑n=1h1(n)q)1q(T∑n=1|Δu1(n)|p)1p−[C(q,q′)]1q(T∑n=1h2(n)p)1p(T∑n=1|Δu2(n)|q)1q−2α1[C(p,p′)]α1+1p(T∑n=1f1(n)pp−α1−1)p−α1−1p(T∑n=1|Δu1(n)|p)α1+1p−2α2[C(q,q′)]α2+1q(T∑n=1f2(n)qq−α2−1)q−α2−1q(T∑n=1|Δu2(n)|q)α2+1q−a0μ2μ1T∑n=1b(n)−C1μ2μ1(T∑n=1|Δu1(n)|p)β1p−C2μ2μ1(T∑n=1|Δu2(n)|q)β2q+(|¯u1|qα1+|¯u2|pα2)(1|¯u1|qα1+|¯u2|pα2(T∑n=1F1(n,¯u1,¯u2)+1μ1T∑n=1F2(n,λ1¯u1,λ2¯u2))−M) | (3.13) |
where C1=(2μ1)K−l+1[T∑n=1b(n)]p−β1p[C(p,p′)]p−β1p, C2=(2μ1)K−m+1[T∑n=1b(n)]q−β2q[C(q,q′)]q−β2q, M=max{2qα1q[C(p,p′)]qpT∑n=1f1(n)q,2pα2p[C(q,q′)]pqT∑n=1f2(n)p}. Hence, together with (3.13), implies that
J(u1,u2)→+∞,as‖(u1,u2)τ‖→∞. | (3.14) |
Hence J is bounded from below and (PS) condition holds. Then by Lemma 2.3, it is easy to know that J has at least one critical point u0 such that
J(uo)=infu∈EJ(u). |
Thus the proof is complete.
The project is supported by the NSF of Hunan Province, China (No: 2022JJ30463), Research Foundation of Education Bureau of Hunan Province, China (Nos. 19C1465, 21C0649, 22A0540), Huaihua University Scientific Research Project, China (Nos. HHUY2019-3, HHUY2018-12, HHUY2022-11) and the Huaihua University Double First-Class Initiative Applied Characteristic Discipline of Control Science and Engineering.
The authors declare that they have no competing interests.
[1] |
J. Mawhin, Periodic solutions of second order nonlinear difference systems with ϕ-Laplacian: a variational approach, Nonlinear Anal., 75 (2012), 4672–4687. https://doi.org/10.1016/j.na.2011.11.018 doi: 10.1016/j.na.2011.11.018
![]() |
[2] | J. Mawhin, M. Willem, Critical point theory and Hamilotonian systems, New York: Springer, 1989. https://doi.org/10.1007/978-1-4757-2061-7 |
[3] | P. H. Rabinowitz, Minimax methods methods in critical point theory with applications to differential equations, CBMS Regional Conference Series in Mathematics, 1986. https://doi.org/10.1090/cbms/065 |
[4] | Y. H. Ding, Variational methods for strongly indefinite problems, World Scientific Publishing, 2007. https://doi.org/10.1142/6565 |
[5] | M. Schechter, Minimax systems and critical point theory, Boston: Birkhäuser, 2009. https://doi.org/10.1007/978-0-8176-4902-9 |
[6] | W. D. Lu, Variational methods in differential equations, Scientific Publishing House in China, 2002. https://doi.org/10.1142/6565 |
[7] |
J. Mawhin, Periodic solutions of second order Lagrangian difference systems with bounded or singular ϕ-Laplacian and periodic potential, Discrete Contin. Dyn. Syst., 6 (2013), 1065–1076. https://doi.org/10.3934/dcdss.2013.6.1065 doi: 10.3934/dcdss.2013.6.1065
![]() |
[8] |
Z. M. Guo, J. S. Yu, The existence of periodic and subharmonic solutions to subquadratic second-order difference equations, J. Lond. Math. Soc., 68 (2003), 419–430. https://doi.org/10.1112/S0024610703004563 doi: 10.1112/S0024610703004563
![]() |
[9] |
Z. M. Guo, J. S. Yu, Existence of periodic and subharmonic solutions for second-order superlinear difference equations, Sci. China Ser. A, 46 (2003), 506–515. https://doi.org/10.1007/BF02884022 doi: 10.1007/BF02884022
![]() |
[10] |
Y. F. Xue, C. L. Tang, Existence of a periodic solution for subquadratic second-order discrete Hamiltonian system, Nonlinear Anal., 67 (2007), 2072–2080. https://doi.org/10.1016/j.na.2006.08.038 doi: 10.1016/j.na.2006.08.038
![]() |
[11] |
T. He, W. Chen, Periodic solutions of second order convex systems involving the p-Laplacian, Appl. Math. Comput., 206 (2008), 124–132. https://doi.org/10.1016/j.amc.2008.08.037 doi: 10.1016/j.amc.2008.08.037
![]() |
[12] |
X. F. He, P. Chen, Homoclinic solutions for second order discrete p-Laplacian systems, Adv. Differ. Equ., 57 (2011), 2011. https://doi.org/10.1186/1687-1847-2011-57 doi: 10.1186/1687-1847-2011-57
![]() |
[13] |
X. Y. Lin, X. H. Tang, Existence of infinitely many homoclinic orbits in discrete Hamiltonian systems, J. Math. Anal. Appl., 373 (2011), 59–72. https://doi.org/10.1016/j.jmaa.2010.06.008 doi: 10.1016/j.jmaa.2010.06.008
![]() |
[14] |
X. H. Tang, X. Y. Zhang, Periodic solutions for second-order discrete Hamiltonian systems, J. Differ. Equ. Appl., 17 (2011), 1413–1430. https://doi.org/10.1080/10236190903555237 doi: 10.1080/10236190903555237
![]() |
[15] |
X. Y. Zhang, X. H. Tang, Existence of solutions for a nonlinear discrete system involing the p-Laplacian, Appl. Math., 57 (2012), 11–30. https://doi.org/10.1007/s10492-012-0002-2 doi: 10.1007/s10492-012-0002-2
![]() |
[16] | X. Y. Zhang, Notes on periodic solutions for a nonlinear discrete system involving the p-Laplacian, Bull. Malays. Math. Sci. Soc., 37 (2014), 499–509. |
[17] |
Q. F. Zhang, X. H. Tang, Q. M. Zhang, Existence of periodic solutions for a class of discrete Hamiltonian systems, Discrete Dyn. Nat. Soc., 2011 (2011), 463480. https://doi.org/10.1155/2011/463480 doi: 10.1155/2011/463480
![]() |
[18] |
M. J. Ma, Z. M. Guo, Homoclinic orbits and subharmonics for nonlinear second order difference equations, Nonlinear Anal., 67 (2007), 1737–1745. https://doi.org/10.1016/j.na.2006.08.014 doi: 10.1016/j.na.2006.08.014
![]() |
[19] |
J. Q. Liu, A generalized saddle point theorem, J. Differ. Equations, 82 (1989), 372–385. https://doi.org/10.1016/0022-0396(89)90139-3 doi: 10.1016/0022-0396(89)90139-3
![]() |
[20] |
Y. Wang, X. Y. Zhang, Multiple periodic solutions for a class of nonlinear difference systems with classical or bounded (ϕ1,ϕ2)-Laplacian, Adv. Differ. Equ., 2014 (2014), 218. https://doi.org/10.1186/1687-1847-2014-218 doi: 10.1186/1687-1847-2014-218
![]() |
[21] |
X. Y. Zhang, C. Zong, H. Y. Deng, L. B. Wang, Existence and multiplicity of homoclinic solutions for difference systems involving classical (ϕ1,ϕ2)-Laplacian and a parameter, Adv. Differ. Equ., 2017 (2017), 380. https://doi.org/10.1186/s13662-017-1419-4 doi: 10.1186/s13662-017-1419-4
![]() |
[22] |
X. Y. Zhang, Y. Wang, Homoclinic solutions for a class of nonlinear difference systems with classical (ϕ1,ϕ2)-Laplacian, Adv. Differ. Equ., 2015 (2015), 149. https://doi.org/10.1186/s13662-015-0467-x doi: 10.1186/s13662-015-0467-x
![]() |
[23] |
H. Y. Deng, X. Y. Zhang, H. Fang, Existence of periodic solutions for a class of discrete systems with classical or bounded (ϕ1,ϕ2)-Laplacian, J. Nonlinear Sci. Appl., 10 (2017), 535–559. http://doi.org/10.22436/jnsa.010.02.19 doi: 10.22436/jnsa.010.02.19
![]() |
[24] |
L. B. Wang, X. Y. Zhang, H. Fang, Existence and multiplicity of solutions for a class of (ϕ1,ϕ2)-Laplacian elliptic system in RN via genus theory, Comput. Math. Appl., 72 (2016), 110–130. https://doi.org/10.1016/j.camwa.2016.04.034 doi: 10.1016/j.camwa.2016.04.034
![]() |
[25] |
D. Pasca, C. L. Tang, Some existence results on periodic solutions of nonautonomous second-order differential systems with (q,p)-Laplacian, Appl. Math. Lett., 23 (2010), 246–251. https://doi.org/10.1016/j.aml.2009.10.005 doi: 10.1016/j.aml.2009.10.005
![]() |
[26] |
D. Pasca, C. L. Tang, Some existence results on periodic solutions of ordinary (q,p)-Laplacian systems, J. Appl. Math. Inform., 29 (2011), 39–48. https://doi.org/10.14317/jami.2011.29.12.039 doi: 10.14317/jami.2011.29.12.039
![]() |
[27] | C. Li, Z. Q. Ou, C. L. Tang, Periodic solutions for non-autonomous second-order differential systems with (q,p)-Laplacian, Electron. J. Differ. Eq., 2014 (2014), 1–13. |
[28] |
X. X. Yang, H. B. Chen, Periodic solutions for autonomous (q,p)-Laplacian system with implusive effects, J. Appl. Math., 2011 (2011), 378389. https://doi.org/10.1155/2011/378389 doi: 10.1155/2011/378389
![]() |
[29] |
Y. K. Li, T. W. Zhang, Infinitely many periodic solutions for second-order (q,p)-Laplacian differential systems, Nonlinear Anal., 74 (2011), 5215–5221. https://doi.org/10.1016/j.na.2011.05.024 doi: 10.1016/j.na.2011.05.024
![]() |
[30] |
X. L. Fan, C. Ji, Existence of infinitely many solutions for a Neumann problem involving the p(x)-Laplacian, J. Math. Anal. Appl., 334 (2007), 248–260. https://doi.org/10.1016/j.jmaa.2006.12.055 doi: 10.1016/j.jmaa.2006.12.055
![]() |
[31] |
Q. Jiang, C. L. Tang, Periodic and subharmonic solutions of a class of subquadratic second-order Hamiltonian systems, J. Math. Anal. Appl., 328 (2007), 380–389. https://doi.org/10.1016/j.jmaa.2006.05.064 doi: 10.1016/j.jmaa.2006.05.064
![]() |
[32] |
C. L. Tang, X. P. Wu, Notes on periodic solutions of subquadratic second order systems, J. Math. Anal. Appl., 285 (2003), 8–16. https://doi.org/10.1016/S0022-247X(02)00417-1 doi: 10.1016/S0022-247X(02)00417-1
![]() |