This paper is concerned with the existence and uniqueness of solutions for a boundary value problem of nonlinear fractional q-difference equations with nonlocal integral boundary conditions. The existence results are obtained by applying some well-known fixed point theorems and illustrated with examples.
1. Introduction
Several kinds of boundary value problems of fractional-order have recently been investigated by many researchers. Fractional derivatives appear naturally in the mathematical modelling of dynamical systems involving fractals and chaos. In fact, the concept of fractional calculus has played a key role in improving the work based on integer-order (classical) calculus in several diverse disciplines of science and engineering. This might have been due to the fact that fractional-differential operators help to understand the hereditary phenomena in many materials and processes in a better way than the corresponding integer-order differential operators. Examples include physics, chemistry, biology, biophysics, blood flow phenomena, control theory, signal and image processing, and economics [1–4]. For some recent results on the topic, see a series of papers [5–12] and the references therein.
Fractional q-difference equations, regarded as fractional analogue of q-difference equations, have been studied by several researchers. For some earlier work on the subject, we refer to [13, 14], whereas the recent development on the existence theory of fractional q-difference equations can be found in [15–25]. In a recent paper [26], the authors investigated a nonlocal boundary value problem of nonlinear fractional q-difference equations:
(1)Dqαcx(t)=f(t,x(t)),0≤t≤1,1<α≤2,α1x(0)-β1Dqx(0)=γ1x(η1),α2x(1)+β2Dqx(1)=γ2x(η2),
where f∈C([0,1]×ℝ,ℝ),cDqα is the fractional q-derivative of the Caputo type, Dq is q-derivative, and αi,βi,γi∈ℝ,ηi∈(0,1),i=1,2.
The purpose of the present paper is to study the following nonlocal boundary value problem of nonlinear fractional q-difference equations:
(2)cDqβx(t)+λx(t)=f(t,x(t)),t∈[0,1],1<β≤2,0<q<1,x(0)=0,x(1)=Iqx(η)=∫0ηx(s)dqs,0<η<1,
where f∈C([0,1]×ℝ,ℝ),cDqβ is the fractional q-derivative of the Caputo type, and λ is a real number.
The paper is organized as follows. Section 2 contains some necessary background material on the topic, while the main results are presented in Section 3. We make use of Banach’s contraction principle, Krasnoselskii’s fixed point theorem, and Leray-Schauder nonlinear alternative to establish the existence results for the problem at hand.
2. Preliminaries on Fractional q-Calculus
Here we recall some definitions and fundamental results on fractional q-calculus.
Definition 1 (see [13]).
Let β≥0, let 0<q<1, and let f be a function defined on [0,1]. The fractional q-integral of the Riemann-Liouville type is (Iq0f)(t)=f(t) and
(3)(Iqβf)(t)=∫0t(t-qs)(β-1)Γq(β)f(s)dq(s),β>0,t∈[0,1],
where
(4)Γq(β)=(1-q)(β-1)(1-q)β-1,0<q<1.
Recall that Γq(β+1)=⌈β⌉qΓq(β), with
(5)⌈β⌉q=qβ-1q-1,(1-q)(0)=1,(1-q)(n)=∏k=0n-1(1-qk+1),n∈ℕ.
More generally, if β∈ℝ, then
(6)(1-q)(β)=∏i=0∞(1-qi+1)(1-qβ+i+1).
Lemma 2 (see [27]).
For β∈ℝ+, λ∈(-1,∞), the following is valid:
(7)Iqβ((t-a)(λ))=Γq(λ+1)Γq(β+λ+1)(t-a)(β+λ),0<a<t<b.
In particular, for λ=0,a=0, using q-integration by parts, we have
(8)(Iqβ1)(t)=1Γq(β)∫0t(t-qs)(β-1)dqs=1Γq(β)∫0tDq((t-s)(β))-[β]qdqs=-1Γq(β+1)∫0tDq((t-s)(β))dqs=1Γq(β+1)t(β).
For 0<q<1, we define the q-derivative of a real valued function f as
(9)Dqf(t)=f(t)-f(qt)(1-q)t,t≠0,Dqf(0)=limn→∞f(sqn)-f(0)sqn,s≠0.
For more details, see [28].
Definition 3 (see [29]).
The fractional q-derivative of the Riemann-Liouville type of order β≥0 is defined by (Dq0f)(t)=f(t) and
(10)(Dqβf)(t)=(Dq[β]Iq[β]-βf)(t),β>0,
where [β] is the smallest integer greater than or equal to β.
Definition 4 (see [29]).
The fractional q-derivative of the Caputo type of order β≥0 is defined by
(11)(Dcqβf)(t)=(Iq[β]-βDq[β]f)(t),β>0,
where [β] is the smallest integer greater than or equal to β.
Now we state some known results involving q-derivatives and q-integrals.
Lemma 5 (see [29]).
Let β,γ≥0, and let f be a function defined on [0,1]. Then
Let β≥0 and n∈ℕ. Then, the following equality holds:
(12)(IqβDqnf)(t)=DqnIqβf(t)-∑k=0[β]-1tβ-n+kΓq(β-n+k)(Dqkf)(0).
In the forthcoming analysis, the following lemma plays a pivotal role.
Lemma 7.
Let h∈C([0,1],ℝ) be a given function. Then the unique solution of the boundary value problem,
(13)Dcqβx(t)+λx(t)=h(t),t∈[0,1],1<β≤2,0<q<1,x(0)=0,x(1)=∫0ηx(s)dqs,0<η<1,
is given by
(14)x(t)=∫0t(t-qs)(β-1)Γq(β)(h(s)-λx(s))dqs+δ1t∫0η(η-qs)(β)Γq(β+1)(h(s)-λx(s))dqs-δ1t∫01(1-qs)(β-1)Γq(β)(h(s)-λx(s))dqs,
where
(15)δ1=1+q1+q-η2.
Proof.
Using Lemma 5, we can write the solution of fractional q-difference equation in (13) as
(16)x(t)=∫0t(t-qs)(β-1)Γq(β)(h(s)-λx(s))dqs+c0t+c1,t∈[0,1].
Using the boundary conditions of (13) in (16), we get c1=0 and
(17)c0=δ1[∫0η(η-qs)(β)Γq(β+1)(h(s)-λx(s))dqs--∫01(1-qs)(β-1)Γq(β)(h(s)-λx(s))dqs],
where δ1 is given by (15). Substituting the values of c0,c1 in (16), we obtain (14).
Let 𝒞:=C([0,1],ℝ) denote the Banach space of all continuous functions from [0,1] to ℝ endowed with the norm defined by ∥x∥=sup{|x(t)|:t∈[0,1]}.
By virtue of Lemma 7, we define an operator 𝒢:𝒞→𝒞 as
(18)(𝒢x)(t)=∫0t(t-qs)(β-1)Γq(β)(f(s,x(s))-λx(s))dqs+δ1t∫0η(η-qs)(β)Γq(β+1)(f(s,x(s))-λx(s))dqs-δ1t∫01(1-qs)(β-1)Γq(β)(f(s,x(s))-λx(s))dqs
and note that the given problem (2) has solutions only if the operator equation 𝒢x=x has fixed points.
3. Main Results
In the sequel, we assume that
f:[0,1]×ℝ→ℝ is a continuous function and that there exists a q-integrable function ζ:[0,1]→ℝ such that |f(t,x)-f(t,y)|≤ζ(t)|x-y|,t∈[0,1],x,y∈ℝ.
For computational convenience, we introduce the notations:
(19)Λ=μ1+|λ|μ2,
where
(20)μ1:=(1+|δ1|)(Iqβζ)(1)+|δ1|(Iqβ+1ζ)(η),μ2:=(1Γq(β+1)(1+|δ1|)+|δ1|ηβ+1Γq(β+2)).
Our first existence result is based on Leray-Schauder nonlinear alternative.
Lemma 8 (nonlinear alternative for single valued maps [30]).
Let E be a Banach space, C a closed, convex subset of E, and V an open subset of C with 0∈V. Suppose that 𝒢:V¯→C is a continuous, compact (i.e., 𝒢(V¯) is a relatively compact subset of C) map. Then either
(i)𝒢 has a fixed point in V¯,
or
(ii) there is x∈∂V (the boundary of V in C) and ε∈(0,1) with x=ε𝒢(x).
Theorem 9.
Suppose that f:[0,1]×ℝ→ℝ is a continuous function. In addition, it is assumed that
there exist functions h1,h2∈L1([0,1],ℝ+) and a nondecreasing function ϑ:ℝ+→ℝ+ such that |f(t,x)|≤h1(t)ϑ(|x|)+h2(t), for (t,x)∈[0,1]×ℝ.
Then the problem (2) has at least one solution on [0,1] if there exists a positive number ω such that
(21)ω>ϑ(ω)ν1+ν21-|λ|μ2,
where
(22)|λ|≠1μ2,νi:=(1+|δ1|)(Iqβhi)(1)+|δ1|(Iqβ+1hi)(η),i=1,2.
Proof.
As a first step, it will be shown that the operator 𝒢:𝒞→𝒞 (defined by (18)) maps bounded sets into bounded sets in C([0,1],ℝ). Notice that 𝒢 is continuous. For a positive number σ, let B-σ={x∈C([0,1],ℝ):∥x∥≤σ} be a bounded set in C([0,1],ℝ). Then, for any x∈B-σ, we find that
(23)|(𝒢x)(t)|≤∫0t(t-qs)(β-1)Γq(β)(|f(s,x(s))|+|λ||x(s)|)dqs+|δ1||t|∫0η(η-qs)(β)Γq(β+1)(|f(s,x(s))|+|λ||x(s)|)dqs+|δ1||t|∫01(1-qs)(β-1)Γq(β)(|f(s,x(s))|+|λ||x(s)|)dqs≤∫0t(t-qs)(β-1)Γq(β)[h1(s)ϑ(∥x∥)+h2(s)]dqs+|δ1||t|∫0η(η-qs)(β)Γq(β+1)[h1(s)ϑ(∥x∥)+h2(s)]dqs+|δ1||t|∫01(1-qs)(β-1)Γq(β)[h1(s)ϑ(∥x∥)+h2(s)]dqs≤ϑ(σ){∫01(1-qs)(β-1)Γq(β)h1(s)dqs+|δ1|∫0η(η-qs)(β)Γq(β+1)h1(s)dqs+|δ1|∫01(1-qs)(β-1)Γq(β)h1(s)dqs}+{∫01(1-qs)(β-1)Γq(β)h2(s)dqs++|δ1|∫0η(η-qs)(β)Γq(β+1)h2(s)dqs+|δ1|∫01(1-qs)(β-1)Γq(β)h2(s)dqs}+σ|λ|{∫01(1-qs)(β-1)Γq(β)dqs+|δ1|∫0η(η-qs)(β)Γq(β+1)dqs+|δ1|+|δ1|∫01(1-qs)(β-1)Γq(β)dqs}≤ϑ(σ)ν1+ν2+σ|λ|μ2.
This establishes our assertion.
Next, we show that the operator 𝒢 maps bounded sets into equicontinuous sets of C([0,1],ℝ). Taking t1,t2∈[0,1] with t1<t2 and x∈B-σ together with the inequality: (t2-qs)(β-1)-(t1-qs)(β-1)≤(t2-t1) for 0<t1<t2 (see [19] p. 4), we obtain
(24)|(𝒢x)(t2)-(𝒢x)(t1)|≤|∫0t2(t2-qs)(β-1)Γq(β)(f(s,x(s))-λx(s))dqs-∫0t1(t1-qs)(β-1)Γq(β)(f(s,x(s))-λx(s))dqs|+|δ1|(t2-t1)∫0η(η-qs)(β)Γq(β+1)+|δ1|(t2-t1)××(|f(s,x(s))|+|λ||x(s)|)dqs.+|δ1|(t2-t1)∫01(1-qs)(β-1)Γq(β)+|δ1|(t2-t1)××(|f(s,x(s))|+|λ||x(s)|)dqs.≤|∫0t1(t2-t1)Γq(β)[h1(s)ϑ(σ)+h2(s)+|λ|σ]dqs+∫t1t2(t2-qs)(β-1)Γq(β)[h1(s)ϑ(σ)+h2(s)+|λ|σ]dqs|+|δ1|(t2-t1)∫0η(η-qs)(β)Γq(β+1)+|δ1|(t2-t1)××[h1(s)ϑ(σ)+h2(s)+|λ|σ]dqs.+|δ1|(t2-t1)∫01(1-qs)(β-1)Γq(β)+|δ1|(t2-t1)××[h1(s)ϑ(σ)+h2(s)+|λ|σ]dqs.
It is obvious that the right hand side of the above inequality tends to zero independently of x∈B-σ as t2→t1. Therefore the operator 𝒢 is completely continuous by the Arzelá-Ascoli theorem.
Thus the operator 𝒢 satisfies the hypothesis of Lemma 8 and hence by its conclusion, either condition (i) or condition (ii) holds. We claim that the conclusion (ii) is not possible.
Let V={x∈C([0,1],ℝ):∥x∥<ω} with ω given by (21). Then we will show that ∥𝒢x∥<ω. Indeed, by means of (A4), we get
(25)∥𝒢x∥≤ϑ(∥x∥){∫01(1-qs)(β-1)Γq(β)h1(s)dqs≤ϑ(∥x∥)+|δ1|+|δ1|∫0η(η-qs)(β)Γq(β+1)h1(s)dqs≤ϑ(∥x∥)+|δ1|+|δ1|∫01(1-qs)(β-1)Γq(β)h1(s)dqs}+{∫01(1-qs)(β-1)Γq(β)h2(s)dqs++|δ1|∫0η(η-qs)(β)Γq(β+1)h2(s)dqs++|δ1|∫01(1-qs)(β-1)Γq(β)h2(s)dqs}+ω|λ|{∫01(1-qs)(β-1)Γq(β)dqs+|δ1|∫0η(η-qs)(β)Γq(β+1)dqs+M|λ|+++|δ1|∫01(1-qs)(β-1)Γq(β)dqs}≤ϑ(ω)ν1+ν2+ω|λ|μ2<ω.
Assume that there exist x∈∂V and ε∈(0,1) such that x=ε𝒢x. Then for such a choice of x and ε, we get a contradiction:
(26)ω=∥x∥=ε∥𝒢x∥<ϑ(∥x∥)ν1+ν2+ω|λ|μ2=ϑ(ω)ν1+ν2+ω|λ|μ2<ω.
Thus it follows by Lemma 8 that 𝒢 has a fixed point x∈V¯ which is a solution of the problem (2). This completes the proof.
Remark 10.
In case we take h1,h2 in (A2) to be continuous, then νi=μ2∥hi∥,i=1,2, where μ2 is given by (20).
Our next result deals with existence and uniqueness of solutions for the problem (2) and is based on Banach’s fixed point theorem.
Theorem 11.
Suppose that the assumption (A1) holds and that Λ<1, where Λ is given by (19). Then the boundary value problem (2) has a unique solution.
Proof.
Fix N=supt∈[0,1]|f(t,0)|<∞, r≥(Nμ2/1-Λ) (μ2 is given by (20)), and define B-r={x∈𝒞:∥x∥≤r}. We will show that 𝒢B-r⊂B-r, where 𝒢 is defined by (18). For x∈B-r,t∈[0,1], it follows by the assumption (A1) that
(27)|f(t,x(t))|≤|f(t,x(t))-f(t,0)|+|f(t,0)|≤ζ(t)|x(t)|+|f(t,0)|≤ζ(t)r+N.
Then, for x∈B-r,t∈[0,1], we have
(28)|(𝒢x)(t)|≤∫0t(t-qs)(β-1)Γq(β)[ζ(s)r+N]dqs+r|λ|∫0t(t-qs)(β-1)Γq(β)dqs+|t||δ1|×∫0η(η-qs)(β)Γq(β+1)[ζ(s)r+N]dqs+r|λ|∫0η(η-qs)(β)Γq(β+1)dqs+|t||δ1|×∫01(1-qs)(β-1)Γq(β)[ζ(s)r+N]dqs+r|λ|∫01(1-qs)(β-1)Γq(β)dqs≤r{∫01(1-qs)(β-1)Γq(β)ζ(s)dqs++|δ1|∫0η(η-qs)(β)Γq(β+1)ζ(s)dqs++|δ1|∫01(1-qs)(β-1)Γq(β)ζ(s)dqs}+N{∫01(1-qs)(β-1)Γq(β)dqs+|δ1|∫0η(η-qs)(β)Γq(β+1)dqs|δ1|++|δ1|∫01(1-qs)(β-1)Γq(β)dqs}+r|λ|{∫01(1-qs)(β-1)Γq(β)dqs+|δ1|∫0η(η-qs)(β)Γq(β+1)dqs|δ1|++|δ1|∫01(1-qs)(β-1)Γq(β)dqs}≤r{(Iqβζ)(1)+|δ1|(Iqβ+1ζ)(η)+|δ1|(Iqβζ)(1)}+N{1Γq(β+1)+|δ1|ηβ+1Γq(β+2)+|δ1|Γq(β+1)}+r|λ|{1Γq(β+1)+|δ1|ηβ+1Γq(β+2)+|δ1|Γq(β+1)},
which, in view of (19), implies that
(29)∥𝒢x∥≤Nμ2+rΛ≤r.
This shows that 𝒢B-r⊂B-r.
Now, for x,y∈𝒞, we have
(30)∥(𝒢x)-(𝒢y)∥≤supt∈[0,1]{∫0t(t-qs)(β-1)Γq(β)××(|f(s,x(s))-f(s,y(s))|×+|λ||x(s)-y(s)|)dqs≤supt∈[0,1]++|δ1||t|∫0η(η-qs)(β)Γq(β+1)(|f(s,x(s))-f(s,y(s))|+|δ1||t|(s)-y(s)+++|λ||x(s)-y(s)|)dqs+|δ1||t|∫01(1-qs)(β-1)Γq(β)≤supt∈[0,1]×××(|f(s,x(s))-f(s,y(s))|≤supt∈[0,1]+|λ||x(s)-y(s)|)dqs∫0t(t-qs)(β-1)Γq(β)}≤∥x-y∥supt∈[0,1]{∫0t(t-qs)(β-1)Γq(β)ζ(s)dqs≤∥x-y∥supt∈[0,1]++|δ1||t|∫0η(η-qs)(β)Γq(β+1)ζ(s)dqs≤∥x-y∥supt∈[0,1]++|δ1||t|∫01(1-qs)(β-1)Γq(β)ζ(s)dqs≤∥x-y∥supt∈[0,1]++|λ|(∫0t(t-qs)(β-1)Γq(β)dqs≤∥x-y∥supt∈[0,1]++|λ|+|δ1||t|∫0η(η-qs)(β)Γq(β+1)dqs≤∥x-y∥supt∈[0,1]++|λ|+|δ1||t|∫01(1-qs)(β-1)Γq(β)dqs)},
which, by (19), takes the form
(31)∥(𝒢x)(t)-(𝒢y)(t)∥≤Λ∥x-y∥.
As Λ<1 (the given assumption), therefore 𝒢 is a contraction. Hence, by Banach’s contraction mapping principle, the problem (2) has a unique solution.
If we take ζ(t)=L (L is a positive constant), the condition Λ<1 becomes L<((1/μ2)-|λ|) and Theorem 11 can be phrased as follows.
Corollary 12.
Let there exists a constant L∈(0,(1/μ2)-|λ|) with μ2 given by (20), and let f:[0,1]×ℝ→ℝ be a continuous function satisfying the assumption: |f(t,x)-f(t,y)|≤L|x-y|, t∈[0,1], x,y∈ℝ. Then the boundary value problem (2) has a unique solution.
Our last result relies on Krasnoselskii’s fixed point theorem [31].
Lemma 13 (Krasnoselskii).
Let Y be a closed, bounded, convex, and nonempty subset of a Banach space X. Let Q1,Q2 be two operators such that (i) Q1x+Q2y∈Y whenever x,y∈Y; (ii) Q1 is compact and continuous; (iii) Q2 is a contraction mapping. Then there exists z∈Y such that z=Q1z+Q2z.
Theorem 14.
Let f:[0,1]×ℝ→ℝ be a continuous function satisfying (A1). Furthermore, it is assumed that
there exist a function ξ∈C([0,1],ℝ+) and a nondecreasing function χ∈C([0,1],ℝ+) such that
(32)|f(t,x)|≤ξ(t)χ(|x|),(t,x)∈[0,1]×ℝ.
If
(33)|δ1|[∫0η(η-qs)(β)Γq(β+1)ζ(s)dqs+∫01(1-qs)(β-1)Γq(β)ζ(s)dqs]+|λ||δ1|[ηβ+1Γq(β+2)+1Γq(β+1)]<1,
with |λ|μ2<1, then the boundary value problem (2) has at least one solution on [0,1].
Proof.
Let us consider the set B-r¯={x∈𝒞:∥x∥≤r¯}, where r¯ is given by
(34)r¯≥χ(r¯)∥ξ∥1-|λ|μ2,∥ξ∥=supt∈[0,t]|ξ(t)|,
and introduce the operators 𝒢1 and 𝒢2 on B-r¯ as
(35)(𝒢1x)(t)=∫0t(t-qs)(β-1)Γq(β)(f(s,x(s))-λx(s))dqs,λx(s)dqs=∫0t(t-qs)(β-1)Γq(β)t∈[0,1],(𝒢2x)(t)=δ1t∫0η(η-qs)(β)Γq(β+1)(f(s,x(s))-λx(s))dqs-δ1t∫01(1-qs)(β-1)Γq(β)(f(s,x(s))-λx(s))dqs,λx(s)dqs-δ1t-λx∫01(1-qs)(β-1)Γq(β)11t∈[0,1].
In order to show the hypothesis of Krasnoselskii’s fixed point theorem, we proceed as follows.
For x,y∈B-r¯, we find that
(36)|(𝒢1x+𝒢2y)(t)|≤∫0t(t-qs)(β-1)Γq(β)(ξ(s)χ(|x(s)|)+|λ||x(s)|)dqs+|δ1|∫0η(η-qs)(β)Γq(β+1)(ξ(s)χ(|y(s)|)+|λ||y(s)|)dqs+|δ1|∫01(1-qs)(β-1)Γq(β)(ξ(s)χ(|y(s)|)+|λ||y(s)|)dqs≤χ(r¯)∥ξ∥[1Γq(β+1)(1+|δ1|)+|δ1|ηβ+1Γq(β+2)]+r¯|λ|[1Γq(β+1)(1+|δ1|)+|δ1|ηβ+1Γq(β+2)]≤χ(r¯)∥ξ∥μ2+r¯|λ|μ2≤r¯.
This implies that 𝒢1x+𝒢2y∈B-r¯.
From the continuity of f, it follows that the operator 𝒢1 is continuous. Also, 𝒢1 is uniformly bounded on B-r¯ as
(37)∥𝒢1x∥≤χ(r¯)∥ξ∥Γq(β+1)+|λ|r¯Γq(β+1).
Next, for any x∈B-r¯, and τ1,τ2∈[0,1] with τ1<τ2, we have
(38)|(𝒢1x)(τ2)-(𝒢1x)(τ1)|=|∫0τ2(τ2-qs)(β-1)Γq(β)(f(s,x(s))-λx(s))dqs-∫0τ1(τ1-qs)(β-1)Γq(β)(f(s,x(s))-λx(s))dqs|≤|∫0τ1(τ2-qs)(β-1)-(τ1-qs)(β-1)Γq(β)××(f(s,x(s))-λx(s))dqs∫0τ1(τ2-qs)(β-1)-(τ1-qs)(β-1)Γq(β)|+|∫τ1τ2(τ2-qs)(β-1)Γq(β)(f(s,x(s))-λx(s))dqs|≤(χ(r¯)∥ξ∥+|λ|r¯)×[∫0τ1(τ2-qs)(β-1)-(τ1-qs)(β-1)Γq(β)dqs+∫τ1τ2(τ2-qs)(β-1)Γq(β)dqs]
which is independent of x and tends to zero as τ2→τ1. Thus, 𝒢1 is equicontinuous. So 𝒢1 is relatively compact on B-r¯. Hence, by the Arzelá-Ascoli theorem, 𝒢1 is compact on B-r¯.
From (A1) and (33) it follows that 𝒢2 is a contraction mapping.
Thus all the conditions of Lemma 13 are satisfied. Hence, by the conclusion of Lemma 13, the problem (2) has at least one solution on [0,1].
As a special case, for χ(x)≡1, there always exists a positive r¯ such that (34) holds true. In consequence, we have the following corollary.
Corollary 15.
Let f:[0,1]×ℝ→ℝ be a continuous function satisfying (A1). Furthermore, |f(t,x)|≤ξ(t), ∀(t,x)∈[0,1]×ℝ, and ξ∈C([0,1],ℝ+). Then the boundary value problem (2) has at least one solution on [0,1] provided (33) holds.
4. ExamplesExample 1.
Consider the problem
(39)cDq3/2x(t)+14x(t)=15x+cosx,0≤t≤1,x(0)=0,x(1)=Iqx(13),
where q=1/2,λ=1/4,η=1/3. It is easy to find that δ1=27/25, and
(40)|f(t,x)|=|15x+cosx|≤15|x|+1.
Obviously h1=1/5,h2=1,ϑ(ω)=ω. In consequence, ν1=0.3002935758,ν2=1.501467879, and condition (21) yields ω>4.629306104. Thus, the hypothesis of Theorem 9 is satisfied. Hence it follows by the conclusion of Theorem 9 that there exists at least one solution for the problem (39).
Example 2.
Consider the nonlocal boundary value problem given by
(41)cDq3/2x(t)+14x(t)=L(cost+tan-1x),0≤t≤1,x(0)=0,x(1)=Iqx(13).
Here q=1/2,λ=1/4,η=1/3 and L is a constant to be fixed later on. With the given data, it is found that δ1=27/25, |f(t,x)-f(t,y)|≤L|x-y| and
(42)Λ=1Γ1/2(5/2)(1+273(42-1)+62253(42-1))(L+14).
Letting
(43)L<[1Γ1/2(5/2)(1+273(42-1)+62253(42-1))]-1-14,
all the conditions of Corollary 12 are satisfied. Therefore, the conclusion of Corollary 12 applies to the problem (41).
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
The authors thank the referees for their useful suggestions that led to the improvement of the original paper. This work has been partially supported by Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah, Saudi Arabia.
KilbasA. A.SrivastavaH. M.TrujilloJ. J.2006204Amsterdam, The NetherlandsElsevier Science B.V.North-Holland Mathematics StudiesMR221807310.1016/S0304-0208(06)80001-0PodlubnyI.1999San Diego, Calif, USAAcademic PressMR1658022SabatierJ.AgrawalO. P.MachadoJ. A. T.2007Dordrecht, The NetherlandsSpringerSamkoS. G.KilbasA. A.MarichevO. I.1993Yverdon, SwitzerlandGordon and Breach ScienceMR1347689BăleanuD.MustafaO. G.On the global existence of solutions to a class of fractional differential equations20105951835184110.1016/j.camwa.2009.08.028MR2595957ZBL1189.34006BăleanuD.MustafaO. G.AgarwalR. P.On Lp-solutions for a class of sequential fractional differential equations201121852074208110.1016/j.amc.2011.07.024MR2831483AgarwalR. P.AhmadB.Existence theory for anti-periodic boundary value problems of fractional differential equations and inclusions20116231200121410.1016/j.camwa.2011.03.001MR2824708ZBL1228.34009AhmadB.NietoJ. J.Riemann-Liouville fractional integro-differential equations with fractional nonlocal integral boundary conditions20112011, article 36MR2851530ZBL06200107BhalekarS.Daftardar-GejjiV.BaleanuD.MaginR.Fractional Bloch equation with delay20116151355136510.1016/j.camwa.2010.12.079MR2773409ZBL1217.34123AkyildizF. T.BelloutH.VajraveluK.Van GorderR. A.Existence results for third order nonlinear boundary value problems arising in nano boundary layer fluid flows over stretching surfaces20111262919293010.1016/j.nonrwa.2011.02.017MR2832937ZBL1231.35155AhmadB.NietoJ. J.Sequential fractional differential equations with three-point boundary conditions201264103046305210.1016/j.camwa.2012.02.036MR2989333ZBL1268.34006GraefJ. R.KongL.KongQ.Application of the mixed monotone operator method to fractional boundary value problems20112554567AgarwalR. P.Certain fractional q-integrals and q-derivatives196966365370MR0247389Al-SalamW. A.Some fractional q-integrals and q-derivatives196615135140MR021884810.1017/S0013091500011469AticiF. M.EloeP. W.Fractional q-calculus on a time scale200714333334410.2991/jnmp.2007.14.3.4MR2350094AticiF. M.EloeP. W.Linear systems of fractional nabla difference equations201141235337010.1216/RMJ-2011-41-2-353MR2794443ZBL1218.39003AtıcıF. M.EloeP. W.Two-point boundary value problems for finite fractional difference equations201117444545610.1080/10236190903029241MR2783359ZBL1215.39002El-ShahedM.Al-AskarF. M.Positive solutions for boundary value problem of nonlinear fractional q-difference equation2011201112385459MR2793792FerreiraR. A. C.Positive solutions for a class of boundary value problems with fractional q-differences201161236737310.1016/j.camwa.2010.11.012MR2754144FerreiraR. A. C.Nontrivial solutions for fractional q-difference boundary value problems201070110MR2740675GoodrichC. S.Existence and uniqueness of solutions to a fractional difference equation with nonlocal conditions201161219120210.1016/j.camwa.2010.10.041MR2754129ZBL1211.39002GraefJ. R.KongL.Positive solutions for a class of higher order boundary value problems with fractional q-derivatives2012218199682968910.1016/j.amc.2012.03.006MR2916148MaJ.YangJ.Existence of solutions for multi-point boundary value problem of fractional q-difference equation201192110MR2861421WuG. C.BaleanuD.New applications of the variational iteration methodfrom differential equations to q-fractional difference equations20132013, article 21JaradF.AbdeljawadT.BaleanuD.Stability of q-fractional non-autonomous systems201314178078410.1016/j.nonrwa.2012.08.001MR2969872AhmadB.NtouyasS. K.PurnarasI. K.Existence results for nonlocal boundary value problems of nonlinear fractional q-difference equations20122012, article 140RajkovićP. M.MarinkovićS. D.StankovićM. S.Fractional integrals and derivatives in q-calculus20071131132310.2298/AADM0701311RMR2316607AnnabyM. H.MansourZ. S.20122056Heidelberg, GermanySpringerLecture Notes in Mathematics10.1007/978-3-642-30898-7MR2963764RajkovićP. M.MarinkovićS. D.StankovićM. S.On q-analogues of Caputo derivative and Mittag-Leffler function2007104359373MR2378985GranasA.DugundjiJ.2005New York, NY, USASpringerSpringer Monographs in MathematicsMR1987179SmartD. R.1980Cambridge University PressMR0467717