Skip to content
Publicly Available Published by De Gruyter January 29, 2021

Existence of local solutions for fractional difference equations with left focal boundary conditions

  • Johnny Henderson and Jeffrey T. Neugebauer EMAIL logo

Abstract

For 1 < ν ≤ 2 a real number and T ≥ 3 a natural number, conditions are given for the existence of solutions of the νth order Atıcı-Eloe fractional difference equation, Δνy(t) + f(t + ν − 1, y(t + ν − 1)) = 0, t ∈ {0, 1, …, T}, and satisfying the left focal boundary conditions Δy(ν − 2) = y(ν + T) = 0.

1 Introduction

In this paper, for 1 < ν ≤ 2 a real number and T ≥ 3 a natural number, we are concerned with local solutions of the nonlinear νth order fractional difference equation,

(1.1)Δνy(t)+f(t+ν1,y(t+ν1))=0,t{0,1,,T},

satisfying the left focal boundary conditions,

(1.2)Δy(ν2)=0,y(ν+T)=0,

where Δ is the forward difference operator with step-size 1, Δν is the Atıcı-Eloe fractional difference, and f(x,r):{ν1,ν,,ν+T1}× is continuous. Our methods rely on application of the Leray-Schauder Nonlinear Alternative [17], which will involve imposing growth restrictions on f as well as restrictions on the natural number T. Since the paper involves restrictions on the size of T, classically such a result is called a local existence result as opposed to a global existence result. Primary motivation for this work is the recent paper by Henderson [18] devoted to local solutions of Dirichlet boundary value problems for fractional difference equations.

A good deal of the current interest in fractional difference equations, devoted to both their theoretical development and their applications, was spawned by the Atıcı and Eloe definitions [1, 2], in the context of discrete domains, of fractional sums and fractional differences. Soon afterwards, Goodrich further developed and extended those papers [9, 10, 11]. Atıcı and Eloe dealt in their first paper with initial value problems for fractional difference equations, and then followed in their second paper with applications of their definitions in obtaining positive solutions for Dirichlet boundary value problems for fractional difference equations. Their second work involved a Guo-Krasnosel’skii fixed point argument requiring the construction of a Green’s function for their fractional problem. Some of Goodrich’s subsequent research that required construction of appropriate Green’s functions has focused on questions extending the Atıcı and Eloe results and can be found in [12, 13, 14, 15, 16]. For some other researchers' closely related works involving boundary value problems for fractional difference equations, we cite [4, 5, 8, 19, 20, 27, 28, 29, 30, 31].

It is natural that discrete fractional calculus and fractional difference equations often appear in modeling natural processes such as found in the paper by Atıcı and Şengül [3] and in the paper by Metzler et al. [22]. Especially prominent is the current use of boundary value problems for discrete fractional difference equations in its applications to discrete control processes; see, for example, this list of monographs devoted to discrete fractional control [6, 7, 21, 23, 24, 25].

2 Some preliminaries and the Green function

We begin this section with the Atıcı-Eloe definitions of fractional sum and fractional difference in the context of a discrete domain.

Definition 2.1

Let n ∈ ℕ and n − 1 < κn be a real number, and let a ∈ ℝ. For t ∈ {a + κ, a + κ + 1, …}, the κth order Atıcı-Eloe fractional sum, Δκu, of the function u is defined by

Δκu(t):=1Γ(κ)s=atκ(ts1)(κ1)u(s),

where t(κ):=Γ(t+1)Γ(t+1κ) is the falling function.

For t ∈ {a + nκ, a + n + 1 − κ, …}, the κth order Atıcı-Eloe fractional difference, Δκu, of the function u is defined by

Δκu(t):=Δn(nκ)u(t):=Δn(Δ(nκ)u(t)),

where Δ is the forward difference defined by Δu(t) = u(t + 1) − u(t), and Δiu(t) = Δ (Δi − 1u(t)), i = 2, 3, ….

Remark 2.1

We note that, for u defined on {a, a + 1, …}, then Δκu is defined on {a + κ, a + κ + 1, …}.

In [11], for 1 < ν ≤ 2, Goodrich constructed the Green function, G(t, s), for −Δνy(t) = 0, t ∈ {0, 1, …, T}, satisfying the left focal boundary conditions (1.2). In particular, by direct computation, Goodrich obtained

G(t,s)=c(ν,T)Γ(ν){Γ(ν)(ν+Ts1)(ν1),t=ν2,ν1,[(2ν)t(ν1)+(ν1)(ν2)]×(ν+Ts1)(ν1),0tνsT,[(2ν)t(ν1)+(ν1)(ν2)]×(ν+Ts1)(ν1)1c(ν,T)(ts1)(ν2),0s<tνT,

with

c(ν,T)=1(ν+T)(ν2)[(ν1)+(2ν)(T+2)].

Goodrich also obtained the following properties of G(t, s) which will be of importance to us:

  1. For each s ∈ {0, …, T},

    ΔtG(ν2,s)=0andG(ν+T,s)=0.
  2. G(t, s) > 0, for (t, s) ∈ {ν − 2, …, ν + T − 1} × {0, …, T}.

  3. maxt∈{ν−2, …, ν+T}G(t, s) = G(s + ν − 1, s), for s ∈ {0, …, T}.

We remark that y is a solution of the linear discrete fractional difference equation, Δνy(t) + h(t + ν − 1) = 0, t ∈ {0, …, T}, and satisfying (1.2), if and only if y:{ν2,,ν+T} has the form

y(t)=s=0TG(t,s)h(s+ν1),t{ν2,,ν+T}.

In the next section, G(t, s) will play the role of the kernel of a completely continuous summation operator.

3 Local existence of solutions

In this section, we make application of the the Leray-Schauder Nonlinear Alternative [17] in obtaining local solutions of (1.1)-(1.2). In doing so, we impose restrictions on f and restrictions on the natural number T.

Theorem 3.1

[Leray-Schauder Nonlinear Alternative]. Let (E, ‖ · ‖) be a Banach space, Kbe a closed and convex subset ofE, Ube a relatively open subset ofKsuch that 0 ∈ U, andN:U¯Kbe completely continuous. Then, either

  1. u = Nuhas a solution inU¯,

or

  1. There existuUand λ ∈ (0, 1) such thatu = λNu.

For 1 < ν ≤ 2 a real number and T ≥ 3 a natural number, assume f(x,r):{ν1,ν,,ν+T1}× is continuous, and let

m:=maxs{0,,T}G(s+ν1,s).

We now present the result of this paper.

Theorem 3.2

Assume

  1. There existσ : {ν − 1, …, T + ν − 1} → [0, ∞) and a nondecreasing functionψ : [0, ∞) → [0, ∞) such that

    |f(x,r)|σ(x)ψ(|r|),(x,r){ν1,,T+ν1}×,

and

  1. There exists>0such that

    mψ()(T+1)maxx{ν1,,T+ν1}σ(x)>1.

Then, (1.1)-(1.2) has a solution defined on {ν − 2, …, T + ν}.

Proof

Let the Banach space E:={h:{ν2,,T+ν}|Δh(ν2)=h(T+ν)=0} be equipped with the norm

h:=maxx{ν2,,T+ν}|h(x)|.

We seek fixed points of the mapping N : EE defined by

(Nh)(t):=s=0TG(t,s)f(s+ν1,h(s+ν1)),hE,t{ν2,,T+ν},

where G(t, s) is the Green function of Section 2. We note that uE is a solution of (1.1)-(1.2) if and only if u is a fixed point of N.

We first show that N maps bounded sets into bounded sets. In that direction, for r > 0, let

Br:={hE|hr}

be a bounded subset of E. Then, by (A), for t ∈ {ν − 2, …, T + ν} and hBr,

|(Nh)(t)|s=0TG(t,s)|f(s+ν1,h(s+ν1))|s=0TG(s+ν1,s)|f(s+ν1,h(s+ν1))|s=0Tmσ(s+ν1)ψ(|h(s+ν1)|)ms=0Tmaxx{ν1,,T+ν1}σ(x)ψ(r)=mψ(r)maxx{ν1,,T+ν1}σ(x)(T+1).

Hence, ‖Nh‖ ≤ mψ(r) maxx∈{ν−1,…,T+ν−1}σ(x)(T + 1), and so N maps Br into a bounded set.

Since {ν − 2, …, T + ν} is a discrete set, it follows immediately that N maps Br into an equicontinuous set. Therefore, by the Arzelà-Ascoli Theorem, N is completely continuous.

Next, we suppose hE and that for some 0 < λ < 1, h = λNh. Then, for t ∈ {ν − 2, …, T + ν}, and again by (A),

|h(t)|=|λ(Nh)(t)|s=0TG(t,s)|f(s+ν1,h(s+ν1))|s=0TG(s+ν1,s)|f(s+ν1,h(s+ν1))|ms=0Tσ(s+ν1)ψ(h)mmaxx{ν1,,T+ν1}σ(x)ψ(h)(T+1),

which yields

hmmaxx{ν1,,T+ν1}σ(x)ψ(h)(T+1)1.

It follows from (B) that h. If we set

U:={hE|h<},

then the operator N:U¯E is completely continuous. From the choice of U, then there is no hU such that h = λNh, for some 0 < λ < 1. It follows from Theorem 3.1 that N has a fixed point yU¯, which is a desired solution of (1.1)-(1.2).

References

[1] F. M. Atıcı and P. W. Eloe, Initial value problems in discrete fractional calculus. Proc. Amer. Math. Soc. 137, No 3 (2009), 981–989.10.1090/S0002-9939-08-09626-3Search in Google Scholar

[2] F. M. Atıcı and P. W. Eloe, Two-point boundary value problems for finite fractional difference equations. J. Differ. Equ. Appl. 17, No 4 (2011), 445–456.10.1080/10236190903029241Search in Google Scholar

[3] F. M. Atıcı and S. Şengül, Modeling with fractional difference equations. J. Math. Anal. Appl. 369 (2010), 1–9.10.1016/j.jmaa.2010.02.009Search in Google Scholar

[4] J. Baoguo, L. Erbe and A. C. Peterson, Convexity for nabla and delta fractional differences. J. Difference Equ. Appl. 21, No 4 (2015), 360–373.10.1080/10236198.2015.1011630Search in Google Scholar

[5] R. Dahal, D. Duncan and C. S. Goodrich, Systems of semipositone discrete fractional boundary value problems. J. Difference Equ. Appl. 20, No 3 (2014), 473–491.10.1080/10236198.2013.856073Search in Google Scholar

[6] S. Das, Functional Fractional Calculus for System Identification and Controls. Springer-Verlag, Berlin-Heidelberg, 2009.Search in Google Scholar

[7] S. Das and I. Pan, Fractional Order Signal Processing: Introductory Concepts and Applications. Springer Briefs in Applied Sciences and Technology, Springer, Heidelberg, 2012.10.1007/978-3-642-23117-9Search in Google Scholar

[8] P. W. Eloe, C. M. Kublik, and J. T. Neugebauer, Comparison of Green's functions for a family of fractional boundary value problems for fractional difference equations. J. Difference Equ. Appl. 25, No 6 (2019), 776–787.10.1080/10236198.2018.1531129Search in Google Scholar

[9] C. S. Goodrich, Continuity of solutions to discrete fractional problems. Comput. Math. Appl. 59 (2010), 3489–3499.10.1016/j.camwa.2010.03.040Search in Google Scholar

[10] C. S. Goodrich, Existence of a positive solution to a class of fractional differential equations. Appl. Math. Lett. 23 (2010), 1050–1055.10.1016/j.aml.2010.04.035Search in Google Scholar

[11] C. S. Goodrich, A comparison result for the fractional differential operator. Int. J. Difference Equ. 1, No 6 (2011), 17–37.Search in Google Scholar

[12] C. S. Goodrich, Existence and uniqueness of solutions to a fractional difference equation with nonlocal conditions. Comput. Math. Appl. 61 (2011), 191–202.10.1016/j.camwa.2010.10.041Search in Google Scholar

[13] C. S. Goodrich, Existence of a positive solution to a system of discrete fractional boundary value problems. Appl. Math. Comput. 217 (2011), 4740–4753.10.1016/j.amc.2010.11.029Search in Google Scholar

[14] C. S. Goodrich, Coercivity of linear functionals on finite dimensional spaces and its application to discrete BVPs. J. Difference Equ. Appl. 22, No 5 (2016), 623–636.10.1080/10236198.2015.1125896Search in Google Scholar

[15] C. S. Goodrich, Summation equations with sign changing kernels and applications to discrete fractional boundary value problems. Comment. Math. Univ. Carolin. 57, No 2 (2016), 201–229.10.14712/1213-7243.2015.164Search in Google Scholar

[16] C. S. Goodrich, Coercive nonlocal elements in fractional differential equations. Positivity21, No 1 (2017), 377–394.10.1007/s11117-016-0427-zSearch in Google Scholar

[17] A. Granas and J. Dugundji, Fixed Point Theory. Springer Briefs in Mathematics, Springer-Verlag, New York, 2003.10.1007/978-0-387-21593-8Search in Google Scholar

[18] J. Henderson, Existence of local solutions for fractional difference equations with Dirichlet boundary conditions. J. Difference Equ. Appl. 25, No 6 (2019), 751–756.10.1080/10236198.2018.1505882Search in Google Scholar

[19] J. Henderson and J. T. Neugebauer, Smallest eigenvalues for a fractional difference equation with right focal boundary conditions. J. Difference Equ. Appl. 23, No 8 (2017), 1317–1323.10.1080/10236198.2017.1321641Search in Google Scholar

[20] Z. Lv, Y. Gong and Y. Chen, Multiplicity and uniqueness for a class of discrete fractional boundary value problems. Appl. Math. 59, No 6 (2014), 673–695.10.1007/s10492-014-0079-xSearch in Google Scholar

[21] R. L. Magin, Fractional Calculus in Bioengineering. Begell House, Inc., Redding, CT, 2006.Search in Google Scholar

[22] F. Metzler, W. Schick, H. G. Kilian and T. F. Nonnenmacher, Relaxation in filled polymers: a fractional calculus approach. J. Chem. Phys. 103 (1995), 7180–7186.10.1063/1.470346Search in Google Scholar

[23] P. Ostalcvzyk, Discrete Fractional Calculus: Applications in Control and Image Processing. World Scientific Publishing Co. Pte. Ltd., Singapore, Hackensack, NJ, 2016.10.1142/9833Search in Google Scholar

[24] A. Oustaloup, Diversity and Non-integer Differentiation for System Dynamics (Control, Systems and Industrial Engineering). Wiley and Sons, Inc., Hoboken, NJ, 2014.10.1002/9781118760864Search in Google Scholar

[25] I. Petrás, Fractional-Order Nonlinear Systems: Modeling, Analysis and Simulation. Series in Nonlinear Physical Science, Springer, Heidelberg, 2011.10.1007/978-3-642-18101-6Search in Google Scholar

[26] M. ur Rehman, F. Iqbal and A. Seemab, On existence of positive solutions for a class of discrete fractional boundary value problems. Positivity21, No 3 (2017), 1173–1187.10.1007/s11117-016-0459-4Search in Google Scholar

[27] J. Xu, X. Gong and C. Hou, Solvability of discrete fractional boundary value problems. Ann. Appl. Math. 31, No 2 (2015), 225–235.Search in Google Scholar

[28] A. Yang and J. Henderson, Comparison of smallest eigenvalues for fractional difference equations. Enlightenment Pure Appl. Math. 2, No 2 (2016), 161–170.Search in Google Scholar

[29] A. Yang, J. Henderson and H. Wang, Parameter dependence for existence, nonexistence and multiplicity of nontrivial solutions for an Atıcı-Eloe fractional difference Lidstone BVP. Electron. J. Qual. Theory Differ. Equ. 2017, No 38 (2017), 1–12.Search in Google Scholar

[30] A. Yang, L. Zhang and J. Henderson, Comparison of smallest eigenvalues for right focal Atıcı-Eloe fractional difference equations. J. Korean Soc. Math. Educ. Ser. B Pure Appl. Math. 24, No 4 (2017), 191–200.Search in Google Scholar

[31] Y. Zhao, S. Sun and Y. Zhang, Existence and uniqueness of solutions to a fractional difference equation with p-Laplacian. J. Appl. Math. Comput. 4, No 1-2 (2017), 183–197.10.1007/s12190-016-1003-1Search in Google Scholar

Received: 2020-04-06
Revised: 2020-12-01
Published Online: 2021-01-29
Published in Print: 2021-02-23

© 2021 Diogenes Co., Sofia

Downloaded on 23.4.2024 from https://www.degruyter.com/document/doi/10.1515/fca-2021-0014/html
Scroll to top button