Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46 https://doi.org/10.1186/s13661-018-0969-z

RESEARCH

Open Access

On nonlocal Robin boundary value problems for Riemann–Liouville fractional Hahn integrodifference equation Nichaphat Patanarapeelert1 and Thanin Sitthiwirattham2* *

Correspondence: [email protected] 2 Mathematics Department, Faculty of Science and Technology, Suan Dusit University, Bangkok, Thailand Full list of author information is available at the end of the article

Abstract In this paper, we study a nonlocal Robin boundary value problem for fractional Hahn integrodifference equation. Our problem contains three fractional Hahn difference operators and a fractional Hahn integral with different numbers of q, ω and order. The existence and uniqueness result is proved by using the Banach fixed point theorem. In addition, the existence of at least one solution is obtained by using Schauder’s fixed point theorem. MSC: 39A10; 39A13; 39A70 Keywords: Fractional Hahn integral; Riemann–Liouville fractional Hahn difference; Boundary value problems; Existence

1 Introduction Recently, many researchers have extensively studied calculus without limit that deals with a set of non-differentiable functions, the so-called quantum calculus. Many types of quantum difference operators are employed in several applications of mathematical areas such as the calculus of variations, particle physics, quantum mechanics, and theory of relativity (see [1–12] and the references therein for some applications and new results of the quantum calculus). In this paper, we study the Hahn quantum calculus that is one type of quantum calculus. Hahn [13] introduced the Hahn difference operator Dq,ω in 1949 as follows:

Dq,ω f (t) =

f (qt + ω) – f (t) , t(q – 1) + ω

t = ω0 :=

ω . 1–q

The Hahn difference operator is a combination of two well-known difference operators: the forward difference operator and the Jackson q-difference operator. Notice that Dq,ω f (t) = ω f (t) whenever q = 1,

Dq,ω f (t) = Dq f (t) whenever ω = 0 and

Dq,ω f (t) = f  (t) whenever q = 1, ω → 0. © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 2 of 16

The Hahn difference operator has been employed to construct families of orthogonal polynomials and investigate some approximation problems (see [14–16] and the references therein). In 2009, Aldwoah [17, 18] defined the right inverse of Dq,ω in terms of both the Jackson q-integral containing the right inverse of Dq [19] and the Nörlund sum containing the right inverse of ω [19]. In 2010, Malinowska and Torres [20, 21] introduced the Hahn quantum variational calculus. In 2013, Malinowska and Martins [22] studied the generalized transversality conditions for the Hahn quantum variational calculus. Later, Hamza and Ahmed [23, 24] studied the theory of linear Hahn difference equations, and investigated the existence and uniqueness results for the initial value problems for Hahn difference equations by using the method of successive approximations. Moreover, they proved Gronwall’s and Bernoulli’s inequalities with respect to the Hahn difference operator and established the mean value theorems for this calculus. In 2016, Hamza and Makharesh [25] investigated the Leibnitz’s rule and Fubini’s theorem associated with Hahn difference operator. In the same year, Sitthiwirattham [26] considered a nonlinear Hahn difference equation with nonlocal boundary value conditions of the form   D2q,ω x(t) + f t, x(t), Dp,θ x(pt + θ ) = 0,

t ∈ [ω0 , T]q,ω ,

x(ω0 ) = ϕ(x),

(1.1) η ∈ (ω0 , T)q,ω ,

x(T) = λx(η),

ω 0 where 0 < q < 1, 0 < ω < T, ω0 := 1–q , 1 ≤ λ < T–ω , p = qm , m ∈ N, θ = ω( 1–p ), f : η–ω0 1–q [ω0 , T]q,ω × R × R → R is a given function, and ϕ : C([ω0 , T]q,ω , R) → R is a given functional. In 2017, Sriphanomwan et al. [27] considered a nonlocal boundary value problem for second-order nonlinear Hahn integrodifference equation with integral boundary condition of the form

  D2q,ω x(t) = f t, x(t), Dp,θ x(pt + θ ), Ψp,θ x(pt + θ ) , t ∈ [ω0 , T]q,ω ,  T x(ω0 ) = x(T), x(η) = μ g(s)x(s) dq,ω s, η ∈ (ω0 , T)q,ω ,

(1.2)

ω0

T ω where 0 < q < 1, 0 < ω < T, ω0 := 1–q , μ ω0 g(r) dq,ω r = 1, μ ∈ R, p = qm , m ∈ N, θ = ω( 1–p ), 1–q f ∈ C([ω0 , T]q,ω × R × R × R, R), and g ∈ C([ω0 , T]q,ω , R+ ) are given functions, and for ϕ ∈ C([ω0 , T]q,ω × [ω0 , T]q,ω , [0, ∞)), 

t

Ψp,θ x(t) :=

ϕ(t, ps + θ )x(ps + θ ) dp,θ s. ω0

In 2010, Čermák and Nechvátal [28] proposed the fractional (q, h)-difference operator and the fractional (q, h)-integral for q > 1. In 2011, Čermák et al. [29] studied discrete Mittag–Leffler functions in linear fractional difference equations for q > 1, and Rahmat [30, 31] studied the (q, h)-Laplace transform and some (q, h)-analogues of integral inequalities on discrete time scales for q > 1. In 2016, Du et al. [32] presented the monotonicity

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 3 of 16

and convexity for nabla fractional (q, h)-difference for q > 0, q = 1. However, we realize that Hahn difference operator requires the condition 0 < q < 1. Therefore, to fill the gap, Brikshavana and Sitthiwirattham [33] have introduced the fractional Hahn difference operators for 0 < q < 1. In 2017, Patanarapeelert and Sitthiwirattham [34] considered a Riemann–Liouville fractional Hahn difference boundary value problem for a fractional Hahn integrodifference equation of the form   γ Dαq,ω u(t) = F t, u(t), r,φ u(t) ,

t ∈ [ω0 , T]q,ω ,

u(ω0 ) = u(T), β

(1.3) β

Dp,θ u(ω0 ) = Dp,θ u(pT + θ ), and a fractional Hahn integral boundary value problem for a Caputo fractional Hahn difference equation of the form C

  γ Dαq,ω u(t) = G t, u(t), C Dr,φ u(rt + φ) ,

t ∈ [ω0 , T]q,ω ,

u(ω0 ) = A(u), β Ip,θ u(T) =

(1.4)

1 p (β)



T

T – σp,ω (s)

β–1 p,θ

ω0

u(s) dp,θ s = B (u),

where [ω0 , T]q,ω := {qk T + ω[k]q : k ∈ N0 } ∪ {ω0 }; α ∈ (1, 2], β, γ ∈ (0, 1], ω > 0, p, q, r ∈ 1–r (0, 1), p = qm , r = qn , m, n ∈ N, θ = ω( 1–p ), φ = ω( 1–q ); F, G ∈ C([ω0 , T]q,ω × R × R, R) 1–q is a given function; A, B : C([ω0 , T]q,ω , R) → R are given functionals; and for ϕ ∈ C([ω0 , T]q,ω × [ω0 , T]q,ω , [0, ∞)), define  γ  γ

r,φ u(t) := Ir,φ ϕu (t) =

1 r (γ )



t

t – σr,φ (s)

ω0

γ –1 r,φ

ϕ(t, s)u(s) dr,φ s.

Presently, Patanarapeelert et al. [35] studied the boundary value problem for fractional Hahn difference equation containing a sequential Caputo fractional Hahn integrodifference equation with nonlocal Dirichlet boundary conditions  C

Dαq,ω C Dβq,ω

Eσq,ω ρq,ω (t)

   γ + qDq,ω u(t) = F t, u(t), q,ω u(t) ,

t ∈ [ω0 , T]q,ω ,

u(ω0 ) = φ(u),

(1.5)

ρq,ω (T)u(T) = ρq,ω (η)u(η) = ψ(u),

η ∈ (ω0 , T)q,ω ,

T := {qk T + ω[k]q : k ∈ N0 } ∪ {ω0 }; ω > 0, q ∈ (0, 1); α, β, γ ∈ (0, 1]; where [ω0 , T]q,ω = Iq,ω T the shift operator Eσq,ω u(t) := u(σq,ω (t)); F ∈ C(Iq,ω × R × R, R) is a given function; φ, ψ : T T T × Iq,ω , [0, ∞)), we define C(Iq,ω , R) → R are given functionals; and for ϕ ∈ C(Iq,ω

γ u(t) :=

q,ω

 γ  Iq,ω ϕu (t) =

1 q,ω (γ )



t ω0

t – σq,ω (s)

γ –1 q,ω

ϕ(t, s)u(s) dq,ω s.

(1.6)

In quantum calculus, there are apparently few research works related to boundary value problems of fractional Hahn difference equations (see [34, 35]). Therefore, in this paper,

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 4 of 16

we devote ourselves to studying a boundary value problem for fractional Hahn difference equation. Our problem is a nonlocal Robin boundary value problem for a fractional Hahn integrodifference equation of the form   γ Dαq,ω u(t) = F t, u(t), r,φ u(t), Dνm,χ u(t) , β

T t ∈ Iq,ω ,

T η ∈ Iq,ω – {ω0 , T},

λ1 u(η) + λ2 Dp,θ u(η) = φ1 (u),

(1.7)

β

μ1 u(T) + μ2 Dp,θ u(T) = φ2 (u), T := {qk T + ω[k]q : k ∈ N0 } ∪ {ω0 }; α ∈ (1, 2], β, γ , ν ∈ (0, 1], ω > 0, p, q, r ∈ (0, 1), p = where Iq,ω 1–r qa , r = qb , m = qc , a, b, c ∈ N, θ = ω( 1–p ), φ = ω( 1–q ), χ = ω( 1–m ); λ1 , λ2 , μ1 , μ2 ∈ R+ ; F ∈ 1–q 1–q T T C(Iq,ω × R × R × R, R) is a given function; φ1 , φ2 : C(Iq,ω , R) → R are given functionals, γ

and r,φ u(t) is defined as (1.6). In the next section, we briefly recall some definitions and lemmas used in this research work. In Sect. 3, we prove the existence and uniqueness of a solution to problem (1.7) by using the Banach fixed point theorem. In Sect. 4, we show the existence of at least one solution to problem (1.7) by using Schauder’s fixed point theorem. Finally, an example is provided to illustrate our results in the last section.

2 Preliminaries In this section, we present the notations, definitions, and lemmas used in the main results. Let q ∈ (0, 1), ω > 0 and define

[n]q :=

1 – qn = qn–1 + · · · + q + 1 and 1–q

[n]q ! :=

n  1 – qk k=1

1–q

,

n ∈ R.

n

The q-analogue of the power function (a – b)q with n ∈ N0 := [0, 1, 2, . . .] is

(a – b)0q := 1,

(a – b)nq :=

n–1  

 a – bqk ,

a, b ∈ R.

k=0 n

The q, ω-analogue of the power function (a – b)q,ω with n ∈ N0 := [0, 1, 2, . . .] is

(a – b)0q,ω := 1,

(a – b)nq,ω :=

n–1 

 

a – bqk + ω[k]q ,

a, b ∈ R.

k=0

More generally, if α ∈ R, we have

(a – b)αq = aα

∞  1 – ( ab )qn n=0

(a – b)αq,ω

1 – ( ab )qα+n

= (a – ω0 )

α

,

a = 0,

b–ω0 n ∞  1 – ( a–ω )q 0 b–ω0 α+n n=0 1 – ( a–ω0 )q

 α = (a – ω0 ) – (b – ω0 ) q ,

a = ω0 .

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

α

Page 5 of 16

α

α

α

Note that aq = aα and (a – ω0 )q,ω = (a – ω0 )α . We also use the notation (0)q = (ω0 )q,ω = 0 for α > 0. The q-gamma and q-beta functions are defined by x–1

(1 – q)q , x ∈ R \ {0, –1, –2, . . .}, (1 – q)x–1  1 q (x)q (s) . t x–1 (1 – qt)s–1 Bq (x, s) := q dq t = q (x + s) 0 q (x) :=

Definition 2.1 For q ∈ (0, 1), ω > 0 and f defined on an interval I ⊆ R which contains ω ω0 := 1–q , the Hahn difference of f is defined by Dq,ω f (t) =

f (qt + ω) – f (t) t(q – 1) + ω

for t = ω0 ,

and Dq,ω f (ω0 ) = f  (ω0 ). Provided that f is differentiable at ω0 , we call Dq,ω f the q, ωderivative of f , and say that f is q, ω-differentiable on I. Remarks (1) Dq,ω [f (t) + g(t)] = Dq,ω f (t) + Dq,ω g(t). (2) Dq,ω [αf (t)] = αDq,ω f (t). (3) Dq,ω [f (t)g(t)] = f (t)Dq,ω g(t) + g(qt + ω)Dq,ω f (t). g(t)Dq,ω f (t) – f (t)Dq,ω g(t) f (t) (4) Dq,ω [ ]= . g(t) g(t)g(qt + ω) Letting a, b ∈ I ⊆ R with a < ω0 < b and [k]q = q, ω-interval by

1–qk ,k 1–q

∈ N0 := N ∪ {0}, we define the

  [a, b]q,ω := qk a + ω[k]q : k ∈ N0 ∪ qk b + ω[k]q : k ∈ N0 ∪ {ω0 } = [a, ω0 ]q,ω ∪ [ω0 , b]q,ω = (a, b)q,ω ∪ {a, b} = [a, b)q,ω ∪ {b} = (a, b]q,ω ∪ {a}. ∞ k k We observe that for each s ∈ [a, b]q,ω , the sequence {σq,ω (s)}∞ k=0 = {q s + ω[k]q }k=0 is uniformly convergent to ω0 . k In addition, we define the forward jump operator σq,ω (t) := qk t + ω[k]q and the backward k (t) := jump operator ρq,ω

t–ω[k]q qk

for k ∈ N.

Definition 2.2 Let I be any closed interval of R that contains a, b, and ω0 . Letting f : I → R be a given function, we define q, ω-integral of f from a to b by 



b

a



b

f (t) dq,ω t :=

a

f (t) dq,ω t – ω0

f (t) dq,ω t, ω0

where 

x

ω0

∞ 

 f (t) dq,ω t := x(1 – q) – ω qk f xqk + ω[k]q , k=0

x ∈ I,

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 6 of 16

and the series converges at x = a and x = b. We call f q, ω-integrable on [a, b], and the sum to the right-hand side of the above equation is called the Jackson–Nörlund sum. Note that the actual domain of function f is defined on [a, b]q,ω ⊂ I. We next introduce the fundamental theorem of Hahn calculus. Lemma 2.1 ([17]) Let f : I → R be continuous at ω0 and define 

x

F(x) :=

x ∈ I.

f (t) dq,ω t, ω0

Then F is continuous at ω0 . Furthermore, Dq,ω0 F(x) exists for every x ∈ I and Dq,ω F(x) = f (x). Conversely, 

b

Dq,ω F(t) dq,ω t = F(b) – F(a) for all a, b ∈ I.

a

Lemma 2.2 ([26]) Let q ∈ (0, 1), ω > 0 and f : I → R be continuous at ω0 . Then 

t





r

t



t

x(s) dq,ω sdq,ω r = ω0

ω0

x(s) dq,ω rdq,ω s. ω0

qs+ω

Lemma 2.3 ([26]) Let q ∈ (0, 1) and ω > 0. Then 



t

dq,ω s = t – ω0

and

ω0

t ω0

(t – ω0 )2 t – σq,ω (s) dq,ω s = . 1+q

Particulary, we introduce fractional Hahn integral and fractional Hahn difference of Riemann–Liouville type as follows. Definition 2.3 For α, ω > 0, q ∈ (0, 1) and f defined on [ω0 , T]q,ω , the fractional Hahn integral is defined by α Iq,ω f (t) :=

1 q (α)



t

t – σq,ω (s)

ω0

α–1 q,ω

f (s) dq,ω s

∞

=

α–1  n  [t(1 – q) – ω] n  n+1 q t – σq,ω (t) q,ω f σq,ω (t) , q (α) n=0

0 f )(t) = f (t). and (Iq,ω

Definition 2.4 For α, ω > 0, q ∈ (0, 1), and f defined on [ω0 , T]q,ω , the fractional Hahn difference of the Riemann–Liouville type of order α is defined by   N–α Dαq,ω f (t) := DN q,ω Iq,ω f (t)  t  –α–1 1 t – σq,ω (s) q,ω f (s) dq,ω s, = q (–α) ω0

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 7 of 16

and D0q,ω f (t) = f (t), where N is the smallest integer that is greater than or equal to α. T Lemma 2.4 ([33]) Letting α > 0, q ∈ (0, 1), ω > 0, and f : Iq,ω → R, we get α Iq,ω Dαq,ω f (t) = f (t) + C1 (t – ω0 )α–1 + · · · + CN (t – ω0 )α–N

for some Ci ∈ R, i = N1,N and N – 1 < α ≤ N, N ∈ N. Next, we give some auxiliary lemmas used for simplifying calculations. Lemma 2.5 ([33]) Letting α, β > 0, p, q ∈ (0, 1), and ω > 0, we have  

t ω0 t ω0

t – σq,ω (s)



x

α–1 q,ω

β

(s – ω0 )q,ω dq,ω s = (t – ω0 )α+β Bq (β + 1, α),

t – σp,ω (x)

α–1 

ω0

p,ω

x – σq,ω (s)

β–1 q,ω

dq,ω s dp,ω x =

(t – ω0 )α+β Bp (β + 1, α). [β]q

The following lemma deals with the linear variant of problem (1.7) and gives a representation of the solution. Lemma 2.6 Let α ∈ (1, 2], β ∈ (0, 1], ω > 0, p, q ∈ (0, 1), p = qm , m ∈ N, θ = ω( 1–p ); 1–q + T T λ1 , λ2 , μ1 , μ2 ∈ R ; h ∈ C(Iq,ω , R) is a given function; φ1 , φ2 : C(Iq,ω , R) → R are given functionals. Then the problem T t ∈ Iq,ω ,

Dαq,ω u(t) = h(t), β

T η ∈ Iq,ω – {ω0 , T},

λ1 u(η) + λ2 Dp,θ u(η) = φ1 (u),

(2.1)

β

μ1 u(T) + μ2 Dp,θ u(T) = φ2 (u) has the unique solution 1 u(t) = q (α)



t

t – σq,ω (s)

ω0

α–1 q,ω

h(s) dq,ω s

(t – ω0 )α–1  BT η [φ1 , h] – Bη T [φ2 , h]  (t – ω0 )α–2  AT η [φ1 , h] – Aη T [φ2 , h] , + 

–

(2.2)

where the functionals η [φ1 , h], T [φ2 , h] are defined by η [φ1 , h] := φ1 (u) –  ×

λ1 q (α)

η  x



η

η – σq,ω (s)

ω0

η – σp,θ (s)

ω0

ω0

α–1 q,ω

h(s) dq,ω s –

λ2 q (α)p (–β)

–β–1  α–1 x – σq,ω (s) q,ω h(s) dq,ω s dp,θ x, p,θ

 T  α–1 μ1 μ2 T – σq,ω (s) q,ω h(s) dq,ω s – q (α) ω0 q (α)p (–β)  T x  –β–1  α–1 × T – σp,θ (s) p,θ x – σq,ω (s) q,ω h(s) dq,ω s dp,θ x,

(2.3)

T [φ2 , h] := φ2 (u) –

ω0

ω0

(2.4)

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 8 of 16

and the constants Aη , AT , Bη , BT , and  are defined by Aη := λ1 (η – ω0 )α–1 +

λ2 p (–β)



η

η – σp,θ (s)

ω0

–β–1 p,θ

(s – ω0 )α–1 dp,θ s

  λ2 (η – ω0 )–β p (α) = (η – ω0 )α–1 λ1 + , p (α – β)  T  –β–1 μ2 T – σp,θ (s) p,θ (s – ω0 )α–1 dp,θ s AT := μ1 (T – ω0 )α–1 + p (–β) ω0   μ2 (T – ω0 )–β p (α) = (T – ω0 )α–1 μ1 + , p (α – β)  η  –β–1 λ2 α–2 Bη := λ1 (η – ω0 ) + η – σp,θ (s) p,θ (s – ω0 )α–2 dp,θ s p (–β) ω0   λ2 (η – ω0 )–β p (α – 1) α–2 λ1 + , = (η – ω0 ) p (α – β – 1)  T  –β–1 μ2 α–2 BT := μ1 (T – ω0 ) + T – σp,θ (s) p,θ (s – ω0 )α–2 dp,θ s p (–β) ω0   μ2 (T – ω0 )–β p (α – 1) α–2 μ1 + , = (T – ω0 ) p (α – β – 1)  := AT Bη – Aη BT .

(2.5)

(2.6)

(2.7)

(2.8) (2.9)

Proof Taking fractional Hahn q, ω-integral of order α for (2.1), we obtain α h(t) u(t) = C1 (t – ω0 )α–1 + C2 (t – ω0 )α–2 + Iq,ω  t  α–1 1 t – σq,ω (s) q,ω h(x) dq,ω s. = C1 (t – ω0 )α–1 + C2 (t – ω0 )α–2 + q (α) ω0

(2.10)

Then we take fractional Hahn p, θ -difference of order β for (2.10) to get β

Dp,θ u(t) =

1 q (α)



t

t – σq,ω (s)

ω0

1 + q (α)p (–β)



t

α–1 q,ω



ω0

x ω0

C1 (s – ω0 )α–1 + C2 (s – ω0 )α–2 dq,ω s

t – σq,ω (x)

–β–1  p,θ

x – σq,ω (s)

α–1 q,ω

h(s) dq,ω s dp,θ x.

(2.11)

Substituting t = η into (2.10) and (2.11) and employing the first condition of (2.1), we have Aη C1 + Bη C2 = η [φ1 , h].

(2.12)

Taking t = T into (2.10) and (2.11) and employing the second condition of (2.1), we have AT C1 + BT C2 = T [φ2 , h].

(2.13)

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 9 of 16

The constants C1 and C2 are revealed from solving the system of equations (2.12)–(2.13) as C1 =

Bη T – BT η 

and C2 =

AT η – Aη T . 

Substituting the constants C1 , C2 into (2.10), we obtain (2.2). On the other hand, it is easy to show that (2.2) is the solution of problem (2.1). By taking fractional Hahn q, ω-difference of order α for (2.2), we obtain (2.1). This completes the proof.  We next introduce Schauder’s fixed point theorem used to prove the existence of a solution of problem (1.7). Lemma 2.7 ([36] Arzelá–Ascoli theorem) A set of functions in C[a, b] with the sup norm is relatively compact if and only if it is uniformly bounded and equicontinuous on [a, b]. Lemma 2.8 ([36]) If a set is closed and relatively compact, then it is compact. Lemma 2.9 ([37] Schauder’s fixed point theorem) Let (D, d) be a complete metric space, U be a closed convex subset of D, and T : D → D be the map such that the set Tu : u ∈ U is relatively compact in D. Then the operator T has at least one fixed point u∗ ∈ U: Tu∗ = u∗ .

3 Existence and uniqueness result In this section, we consider the existence and uniqueness result for problem (1.7). Let T C = C(Iq,ω , R) be a Banach space of all function u with the norm defined by    u C = max u(t), Dm,χ u(t) , T t∈Iq,ω

where α ∈ (1, 2], β, γ ∈ (0, 1], ω > 0, p, q, r ∈ (0, 1), p = qa , r = qb , m = qc , a, b, c ∈ N, θ = 1–r ω( 1–p ), φ = ω( 1–q ), χ = ω( 1–m ); λ1 , λ2 , μ1 , μ2 ∈ R+ . Define an operator F : C → C by 1–q 1–q 1 (F u)(t) := q (α)



t

t – σq,ω (s)

ω0

α–1   γ F s, u(s), r,φ u(s), Dνm,χ u(s) dq,ω s q,ω

(t – ω0 )α–1  BT ∗η [φ1 , Fu ] – Bη ∗T [φ2 , Fu ]  (t – ω0 )α–2  AT ∗η [φ1 , Fu ] – Aη ∗T [φ2 , Fu ] , + 

–

(3.1)

where the functionals ∗η [φ1 , Fu ], ∗T [φ2 , Fu ] are defined by ∗η [φ1 , Fu ] := φ1 (u) –

λ1 q (α)



η

η – σq,ω (s)

ω0

α–1   γ F s, u(s), r,φ u(s), Dνm,χ u(s) dq,ω s q,ω

 η x  –β–1  α–1 λ2 η – σp,θ (s) p,θ x – σq,ω (s) q,ω – q (α)p (–β) ω0 ω0   γ × F s, u(s), r,φ u(s), Dνm,χ u(s) dq,ω s dp,θ x,

(3.2)

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

∗T [φ2 , Fu ] := φ2 (u) –

μ1 q (α)



T

T – σq,ω (s)

ω0

Page 10 of 16

α–1   γ F s, u(s), r,φ u(s), Dνm,χ u(s) dq,ω s q,ω

 T x  –β–1  α–1 μ2 T – σp,θ (s) p,θ x – σq,ω (s) q,ω q (α)p (–β) ω0 ω0   γ × F s, u(s), r,φ u(s), Dνm,χ u(s) dq,ω s dp,θ x, –

(3.3)

and the constants Aη , AT , Bη , BT ,  are defined by (2.5)–(2.9), respectively. We find that problem (1.7) has a solution if and only if the operator F has a fixed point. T T T × R × R × R → R is continuous, ϕ : Iq,ω × Iq,ω → [0, ∞) Theorem 3.1 Assume that F : Iq,ω T T is continuous with ϕ0 = max{ϕ(t, s) : (t, s) ∈ Iq,ω × Iq,ω }. In addition, suppose that the following conditions hold: T and u, v ∈ R, (H1 ) There exist constants 1 , 2 , 3 > 0 such that, for each t ∈ Iq,ω

     F t, u, γ u, Dν u – F t, v, γ v, Dν v  m,χ m,χ r,φ r,φ   γ  γ  ≤ 1 |u – v| + 2  r,φ u – r,φ  + 3 Dνm,χ u – Dνm,χ v. (H2 ) There exist constants ϑ1 , ϑ2 > 0 such that, for each u, v ∈ C ,   φ1 (u) – φ1 (v) ≤ ϑ1 u – v C

  and φ2 (u) – φ2 (v) ≤ ϑ2 u – v C .

(H3 ) O < 1, where

L := 1 + 2 ϕ0

(T – ω0 )γ , r (γ + 1)

(3.4)

 λ1 (η – ω0 )α λ2 (η – ω0 )α–β q (–β) + , q (α + 1) q (α – β + 1)p (–β)   μ1 (T – ω0 )α μ2 (T – ω0 )α–β q (–β) + , 2 := q (α + 1) q (α – β + 1)p (–β) 

(3.5)

1 :=

O :=

(3.6)

(L + 3 )(T – ω0 )α [ϑ1 + (L + 3 )1 ]  + |BT |(T – ω0 )α–1 + |AT |(T – ω0 )α–2 q (α + 1) || +

[ϑ2 + (L + 3 )2 ]  |Bη |(T – ω0 )α–1 + |Aη |(T – ω0 )α–2 . ||

T Then problem (1.7) has a unique solution in Iq,ω .

Proof To show that F is a contraction, we denote that      γ γ H|u – v|(t) := F t, u(t), r,φ u(t), Dνm,χ u(t) – F t, v(t), r,φ v(t), Dνm,χ v(t)  T for each t ∈ Iq,ω and u, v ∈ C . We find that

  ∗  [φ1 , Fu ] – ∗ [φ1 , Fv ] η η   λ1 ≤ φ1 (u) – φ1 (v) + q (α)



η

ω0

η – σq,ω (s)

α–1 q,ω

H|u – v|(s) dq,ω s

(3.7)

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 11 of 16

 η x  –β–1  α–1 λ2 η – σp,θ (s) p,θ x – σq,ω (s) q,ω H|u – v|(s) dq,ω s dp,θ x + q (α)p (–β) ω0 ω0  γ    γ  ≤ ϑ1 u – v C + 1 |u – v| + 2  r,φ u – r,φ  + 3 Dνm,χ u – Dνm,χ v    λ1 (η – ω0 )α λ2 (η – ω0 )α–β q (–β)    + × q (α + 1) q (α – β + 1)p (–β)      ν  (T – ω0 )γ ν   |u – v| + 3 Dm,χ u – Dm,χ v 1 ≤ ϑ1 u – v C + 1 + 2 ϕ0 r (γ + 1)

≤ ϑ1 + (L + 3 )1 u – v C . Similarly,   ∗

 [φ2 , Fu ] – ∗ [φ2 , Fv ] ≤ ϑ2 + (L + 3 )2 u – v C . T T Next, we have   (F u)(t) – (F v)(t)  T  α–1 1 ≤ T – σq,ω (s) q,ω H|u – v|(s) dq,ω s q (α) ω0 +

    (T – ω0 )α–1  |BT |∗η [φ1 , Fu ] – ∗η [φ1 , Fv ] + |Bη |∗T [φ2 , Fu ] – ∗T [φ2 , Fv ] ||

+

    (T – ω0 )α–2  |AT |∗η [φ1 , Fu ] – ∗η [φ1 , Fv ] + |Aη |∗T [φ2 , Fu ] – ∗T [φ2 , Fv ] ||

 ≤

(L + 3 )(T – ω0 )α [ϑ1 + (L + 3 )1 ]  + |BT |(T – ω0 )α–1 + |AT |(T – ω0 )α–2 q (α + 1)   [ϑ2 + (L + 3 )2 ]  |Bη |(T – ω0 )α–1 + |Aη |(T – ω0 )α–2 u – v C + ||

= O u – v C .

(3.8)

We take fractional Hahn m, χ -difference of order γ for (3.1) to obtain 

 Dγm,χ F u (t) =

 t x  –γ –1  α–1 1 t – σm,χ (x) m,χ x – σq,ω (s) q,ω m (–γ )q (α) ω0 ω0   γ × F s, u(s), r,φ u(s), Dνm,χ u(s) dq,ω s dm,χ x   t –γ –1 1 – BT ∗η [φ1 , F] – Bη ∗T [φ2 , F] t – σm,χ (s) m,χ (s – ω0 )α–1 dm,χ s m (–γ ) ω0  1 AT ∗η [φ1 , F] – Aη ∗T [φ2 , F] m (–γ )  t  –γ –1 t – σm,χ (s) m,χ (s – ω0 )α–2 dm,χ s. × +

ω0

(3.9)

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 12 of 16

Using the same argument as above, we have  γ      D F u (t) – Dγ F v (t) < O u – v C . m,χ m,χ

(3.10)

(3.8) and (3.10) imply that F u – F v C ≤ O u – v C . By (H3 ) we can conclude that F is a contraction. Therefore, by using the Banach fixed T point theorem, F has a fixed point which is a unique solution of problem (1.7) on Iq,ω . 

4 Existence of at least one solution In this section, we present the existence of a solution to (1.7) by using Schauder’s fixed point theorem. Theorem 4.1 Suppose that (H1 ) and (H3 ) hold. Then problem (1.7) has at least one soluT tion on Iq,ω . Proof We divide the proof into three steps as follows. Step I. Verify that F maps bounded sets into bounded sets in BR = {u ∈ C : u C ≤ R}. T We consider BR = {u ∈ C(Iq,ω ) : u C ≤ R}. Set maxt∈Iq,ω T |F(t, 0, 0, 0)| = K , supu∈C |φ1 (u)| = M1 , supu∈C |φ2 (u)| = M2 and choose a constant R≥

M1 + M2 +

K || [(T

– ω0 )α–1 (|BT |1 + |Bη |2 ) + (T – ω0 )α–2 (|AT |1 + |Aη |2 )] 1–O

. (4.1)

Denote that         S (t, u, 0) = F t, u(t), γ u(t), Dγ u(t) – F(t, 0, 0, 0) + F(t, 0, 0, 0). m,χ r,φ T For each t ∈ Iq,ω and u ∈ BR , we obtain

 ∗   [φ1 , Fu ] η ≤ M1 +

λ1 q (α)



η

η – σq,ω (s)

ω0

 α–1  S (s, u, 0)dq,ω s q,ω

 η x   –β–1  α–1  λ2 η – σp,θ (s) p,θ x – σq,ω (s) q,ω S (s, u, 0)dq,ω s dp,θ x q (α)p (–β) ω0 ω0     ν  (T – ω0 )γ   |u| + 3 Dm,χ u + K 1 ≤ M1 + 1 + 2 ϕ0 r (γ + 1) +

≤ ϑ1 + K1 + (L + 3 )1 u C ≤ ϑ1 + K1 + (L + 3 )1 R.

(4.2)

Similarly,  ∗   [φ2 , Fu ] ≤ M2 + K2 + (L + 3 )2 R. T

(4.3)

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 13 of 16

From (4.2)–(4.3), we find that   (F u)(t) ≤ R + M1 + M2 + K ||    

α–1 |BT |1 + |Bη |2 + (T – ω0 )α–2 |AT |1 + |Aη |2 × (T – ω0 ) ≤ R.

(4.4)

In addition, we obtain  γ    D F u (t) < R. m,χ

(4.5)

Therefore, F u C ≤ R, which implies that F is uniformly bounded. Step II. We can conclude that the operator F is continuous on BR by the continuity of F. T Step III. In this step, we examine that F is equicontinuous on BR . For any t1 , t2 ∈ Iq,ω with t1 < t2 , we have   (F u)(t2 ) – (F u)(t1 ) ≤

 F  (t2 – ω0 )α – (t1 – ω0 )α  q (α + 1) +

|(t2 – ω0 )α–1 – (t1 – ω0 )α–1 |  |BT |∗η [φ1 , F] + |Bη |∗T [φ2 , F] ||

+

|(t2 – ω0 )α–2 – (t1 – ω0 )α–2 |  |AT |∗η [φ1 , F] + |Aη |∗T [φ2 , F] ||

(4.6)

and  γ      D F u (t1 ) – Dγ F u (t2 ) m,χ m,χ ≤

  F q (–γ ) (t2 – ω0 )α–γ – (t1 – ω0 )α–γ  m (–γ )q (α – γ + 1)  q (α)q (–γ ) |BT |∗η [φ1 , F] + |Bη |∗T [φ2 , F] ||m (–γ )q (α – γ )   × (t2 – ω0 )α–γ –1 – (t1 – ω0 )α–γ –1  +

 q (α – 1)q (–γ ) |AT |∗η [φ1 , F] + |Aη |∗T [φ2 , F] ||m (–γ )q (α – γ – 1)   × (t2 – ω0 )α–γ –2 – (t1 – ω0 )α–γ –2 . +

(4.7)

We observe that the right-hand side of (4.7) tends to be zero when |t2 – t1 | → 0. So F is relatively compact on BR . This implies that the set F (BR ) is an equicontinuous set. As a consequence of Steps I to III together with the Arzelá–Ascoli theorem, we get that F : C → C is completely continuous. By Schauder’s fixed point theorem, we can conclude that problem (1.7) has at least one solution. 

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 14 of 16

5 Example Consider the following boundary value problem for fractional Hahn difference equation: 4

D 31 , 2 u(t) = 2 3

 –3t  2  –(π +sin2 π t)  12 1  1 7 u(t) u e + 2|u| + e (1000e3 + t 2 )(1 + |u(t)|) 8,6    2 

4 2 , 10 , + e–(1+cos π t) D 51 , 2 u(t) , t ∈ 1,2 3 2 3

    ∞ 3 15 Ci |u(ti )| 1 15 4 u + 100eD 1 ,1 u = , 10e 8 8 1 + |u(ti )| 4 i=0

 i 1 2 + [i] 1 , ti ∈ 10 2 3 2

 i 1 2 + [i] 1 , ti ∈ 10 2 3 2

∞

200πu(10) +

(5.1)

2 3

Di |u(ti )| 3 1 D 41 ,1 u(10) = , 10π 4 1 + |u(ti )| i=0

∞ e–2|s–t| 1 π where ϕ(t, s) = (t+10) 3 and Ci , Di are given constants with 500t 3 ≤ i=0 Ci ≤ 500t 3 and  1 π ≤ ∞ i=0 Di ≤ 1000t 2 . 1000t 2 4 We provide α = 3 , β = 34 , γ = 12 , ν = 25 ,q = 12 , p = 14 , r = 18 , m = 12 , ω = 23 , θ = 1, φ = 76 , χ = 2 ω 1 1 , ω0 = 1–q = 43 , T = 10, η = 10( 12 )4 + 23 [4] 1 = 15 , λ1 = 10e , λ2 = 100e, μ1 = 200π, μ2 = 10π , 3 8 2 ∞ Ci |u(ti )| ∞ Di |u(ti )| γ 1 ν φ1 (u) = i=0 1+|u(ti )| , φ2 = i=0 1+|u(ti )| , and F(t, u(t), r,φ u(t), Dm,χ u(t)) = (1000e3 +t2 )(1+|u(t)|) × 2 π t)

[e–3t (u2 + 2|u|) + e–(π +sin

1

2 π t)

| 12 , 7 u(t)| + e–(1+cos 8 6

We can find that

2

|D 51 , 2 u(t)|]. 2 3

|Aη | = 173.1815,

|AT | = 1290.6198,

|BT | = 148.9158

and || = 1.7202 × 106 .

|Bη | = 1312.8836,

For all t ∈ [ 43 , 10] 1 , 2 and u, v ∈ R, we have 2 3

     F t, u, γ u, Dν u – F t, v, γ v, Dν v  m,χ m,χ r,φ r,φ ≤

1 e4 (1000e3 +

+

16 ) 9

1 e(1000e3 +

|u – v| +

16 ) 9

1 eπ (1000e3

+

16 ) 9

  γ  u – γ v r,φ r,φ

 ν  D u – Dν v. m,χ m,χ

Thus, (H1 ) holds with 1 = 9.118 × 10–7 , 2 = 2.1513 × 10–6 , and 3 = 0.0000183. For all u, v ∈ C ,   φ1 (u) – φ1 (v) =

π u – v C , 500t 3   e φ2 (u) – φ2 (v) = u – v C . 1000t 2 So, (H2 ) holds with ϑ1 = 0.00265 and ϑ2 = 0.00153. Also, we find that

L = 9.163 × 10–7 ,

1 = 440.682,

2 = 248.882.

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 15 of 16

Therefore, (H3 ) holds with

O ≈ 0.000307 < 1. Hence, by Theorem 3.1, problem (5.1) has a unique solution.

6 Conclusion We have proved the existence and uniqueness result of the nonlocal Robin boundary problem for a fractional Hahn integrodifference equation (1.7) by using the Banach fixed point theorem, and the existence of at least one solution by Schauder’s fixed point theorem. Our problem contains three fractional Hahn difference operators and a fractional Hahn integral with different numbers of q, ω and order, which is a new idea. Funding This research was funded by King Mongkut’s University of Technology North Bangkok. Contract No. KMUTNB-ART-60-33. List of abbreviations Not applicable. Availability of data and materials Not applicable. Ethics approval and consent to participate Not applicable. Competing interests The authors declare that they have no competing interests. Consent for publication Not applicable. Authors’ contributions The authors declare that they carried out all the work in this manuscript, and read and approved the final manuscript. Author details 1 Department of Mathematics, Faculty of Applied Science, King Mongkut’s University of Technology North Bangkok, Bangkok, Thailand. 2 Mathematics Department, Faculty of Science and Technology, Suan Dusit University, Bangkok, Thailand.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Received: 16 January 2018 Accepted: 26 March 2018 References 1. Annaby, M.H., Mansour, Z.S.: q-Fractional Calculus and Equations. Springer, Berlin (2012) 2. Kac, V., Cheung, P.: Quantum Calculus. Springer, New York (2002) 3. Jagerman, D.L.: Difference Equations with Applications to Queues. Dekker, New York (2000) 4. Aldowah, K.A., Malinowska, A.B., Torres, D.F.M.: The power quantum calculus and variational problems. Dyn. Contin. Discrete Impuls. Syst., Ser. B, Appl. Algorithms 19, 93–116 (2012) 5. Birto da Cruz, A.M.C., Martins, N., Torres, D.F.M.: Symmetric differentiation on time scales. Appl. Math. Lett. 26(2), 264–269 (2013) 6. Cruz, B., Artur, M.C.: Symmetric Quantum Calculus. Ph.D. thesis, Aveiro University (2012) 7. Wu, G.C., Baleanu, D.: New applications of the variational iteration method from differential equations to q-fractional difference equations. Adv. Differ. Equ. 2013, 21 (2013) 8. Tariboon, J., Ntouyas, S.K.: Quantum calculus on finite intervals and applications to impulsive difference equations. Adv. Differ. Equ. 2013, 282 (2013) 9. Álvarez-Nodarse, R.: On characterization of classical polynomials. J. Comput. Appl. Math. 196, 320–337 (2006) 10. Agarwal, R.P., Baleanu, D., Hedayati, V., Rezapour, S.: Two fractional derivative inclusion problems via integral boundary condition. Appl. Math. Comput. 257, 205–212 (2015) 11. Agarwal, R.P., Baleanu, D., Rezapour, S., Salehi, S.: The existence of solution for some fractional finite difference equations via sum boundary conditions. Adv. Differ. Equ. 2014, 282 (2014)

Patanarapeelert and Sitthiwirattham Boundary Value Problems (2018) 2018:46

Page 16 of 16

12. Nyamoradi, N., Baleanu, D., Agarwal, R.P.: Existence and uniqueness of positive solutions to fractional boundary value problems with nonlinear boundary conditions. Adv. Differ. Equ. 2013, 266 (2013) 13. Hahn, W.: Über Orthogonalpolynome, die q-Differenzenlgleichungen genügen. Math. Nachr. 2, 4–34 (1949) 14. Costas-Santos, R.S., Marcellán, F.: Second structure relation for q-semiclassical polynomials of the Hahn tableau. J. Math. Anal. Appl. 329, 206–228 (2007) 15. Kwon, K.H., Lee, D.W., Park, S.B., Yoo, B.H.: Hahn class orthogonal polynomials. Kyungpook Math. J. 38, 259–281 (1998) 16. Foupouagnigni, M.: Laguerre–Hahn orthogonal polynomials with respect to the Hahn operator: fourth-order difference equation for the rth associated and the Laguerre–Freud equations recurrence coefficients. Ph.D. thesis, Université Nationale du Bénin, Bénin (1998) 17. Aldwoah, K.A.: Generalized time scales and associated difference equations. Ph.D. thesis, Cairo University (2009) 18. Annaby, M.H., Hamza, A.E., Aldwoah, K.A.: Hahn difference operator and associated Jackson–Nörlund integrals. J. Optim. Theory Appl. 154, 133–153 (2012) 19. Jackson, F.H.: Basic integration. Q. J. Math. 2, 1–16 (1951) 20. Malinowska, A.B., Torres, D.F.M.: The Hahn quantum variational calculus. J. Optim. Theory Appl. 147, 419–442 (2010) 21. Malinowska, A.B., Torres, D.F.M.: Quantum Variational Calculus. Springer Briefs in Electrical and Computer Engineering-Control, Automation and Robotics. Springer, Berlin (2014) 22. Malinowska, A.B., Martins, N.: Generalized transversality conditions for the Hahn quantum variational calculus. Optimization 62(3), 323–344 (2013) 23. Hamza, A.E., Ahmed, S.M.: Theory of linear Hahn difference equations. J. Adv. Math. 4(2), 441–461 (2013) 24. Hamza, A.E., Ahmed, S.M.: Existence and uniqueness of solutions of Hahn difference equations. Adv. Differ. Equ. 2013, 316 (2013) 25. Hamza, A.E., Makharesh, S.D.: Leibniz’ rule and Fubinis theorem associated with Hahn difference operator. J. Adv. Math. 12(6), 6335–6345 (2016) 26. Sitthiwirattham, T.: On a nonlocal boundary value problem for nonlinear second-order Hahn difference equation with two different q, ω -derivatives. Adv. Differ. Equ. 2016, 116 (2016) 27. Sriphanomwan, U., Tariboon, J., Patanarapeelert, N., Ntouyas, S.K., Sitthiwirattham, T.: Nonlocal boundary value problems for second-order nonlinear Hahn integro-difference equations with integral boundary conditions. Adv. Differ. Equ. 2017, 170 (2017) ˇ 28. Cermák, J., Nechvátal, L.: On (q, h)-analogue of fractional calculus. J. Nonlinear Math. Phys. 17(1), 51–68 (2010) ˇ 29. Cermák, J., Kisela, T., Nechvátal, L.: Discrete Mittag–Leffler functions in linear fractional difference equations. Abstr. Appl. Anal. 2011, Article ID 565067 (2011) 30. Rahmat, M.R.S.: The (q, h)-Laplace transform on discrete time scales. Comput. Math. Appl. 62, 272–281 (2011) 31. Rahmat, M.R.S.: On some (q, h)-analogues of integral inequalities on discrete time scales. Comput. Math. Appl. 62, 1790–1797 (2011) 32. Du, F., Jai, B., Erbe, L., Peterson, A.: Monotonicity and convexity for nabla fractional (q, h)-difference. J. Differ. Equ. Appl. 22(9), 1224–1243 (2016) 33. Brikshavana, T., Sitthiwirattham, T.: On fractional Hahn calculus with the delta operators. Adv. Differ. Equ. 2017, 354 (2017) 34. Patanarapeelert, N., Sitthiwirattham, T.: Existence results for fractional Hahn difference and fractional Hahn integral boundary value problems. Discrete Dyn. Nat. Soc. 2017, Article ID 7895186 (2017) 35. Patanarapeelert, N., Brikshavana, T., Sitthiwirattham, T.: On nonlocal Dirichlet boundary value problem for sequential Caputo fractional Hahn integrodifference equations. Bound. Value Probl. 2018, 6 (2018) 36. Griffel, D.H.: Applied Functional Analysis. Ellis Horwood, Chichester (1981) 37. Guo, D., Lakshmikantham, V.: Nonlinear Problems in Abstract Cone. Academic Press, Orlando (1988)