Show Summary Details
More options …

# Open Mathematics

### formerly Central European Journal of Mathematics

Editor-in-Chief: Gianazza, Ugo / Vespri, Vincenzo

1 Issue per year

IMPACT FACTOR 2017: 0.831
5-year IMPACT FACTOR: 0.836

CiteScore 2017: 0.68

SCImago Journal Rank (SJR) 2017: 0.450
Source Normalized Impact per Paper (SNIP) 2017: 0.829

Mathematical Citation Quotient (MCQ) 2017: 0.32

Open Access
Online
ISSN
2391-5455
See all formats and pricing
More options …
Volume 14, Issue 1

# On the concept of general solution for impulsive differential equations of fractional-order q ∈ (2,3)

Xianmin Zhang
/ Tong Shu
/ Zuohua Liu
/ Wenbin Ding
/ Hui Peng
/ Jun He
Published Online: 2016-07-08 | DOI: https://doi.org/10.1515/math-2016-0042

## Abstract

In this paper, we find the formula of general solution for a generalized impulsive differential equations of fractional-order q ∈ (2, 3).

MSC 2010: 34A08; 34A37

## 1 Introduction

Fractional differential equations play an important part in modeling of many phenomena in various fields of science and engineering, and the subject of fractional differential equations is extensively researched (see [119] and the references therein).

On the other hand, impulsive differential equation is a key tool to describe some systems and processes with impulsive effects. There have appeared many papers focused on the subject of impulsive differential equations with Caputo fractional derivative [2031].

Recently, we have found that there exist general solutions for several kinds of impulsive fractional differential equations in [3238]. Based on these works, we will further study the general solution of the generalized impulsive differential equations of fractional-order q ∈ (2,3).

$aDtqx(t)=f(t,x(t)),q∈(2,3),t∈J=[a,T],t≠ti(i=1,2,…,L)andt≠t¯j(j=1,2,…,M)andt≠t^l(l=1,2,…,N),Δxt=ti=x(ti+)−x(ti−)=Iix(ti−),i=1,2,…,L,Δx′t=t¯j=x′(t¯j+)−x′(t¯j−)=I¯jx(t¯j−),j=1,2,…,M,Δx″t=t^l=x″(t^l+)−x″(t^l−)=I^lx(t^l−),l=1,2,…,N,x(a)=xa,x′(a)=x¯a,x″(a)=x^a.$(1)

where ${}_{a}{D}_{t}^{q}$ denote Caputo fractional derivative of order q in interval [a, t], f : J × ℝ → ℝ and, Ii Īj, Îi : ℝ → ℝ are appropriate functions (here i = 1,2,…, L and j = 1,2,…, M and l = 1,2, …, N, respectively), a = t0 < t1 < … < tL < tL+1 = T, $a={\overline{t}}_{0}<{\overline{t}}_{1}<\cdots {\overline{t}}_{M}<{\overline{t}}_{M+1}=T,a={\stackrel{^}{t}}_{0}<{\stackrel{^}{t}}_{1}<\cdots <{\stackrel{^}{t}}_{N}<{\stackrel{^}{t}}_{N+1}=T$. Here $x\left({t}_{i}^{+}\right)=\underset{\epsilon \to 0}{lim}+x\left({t}_{i}+\epsilon \right)$ and $x\left({t}_{i}^{-}\right)=\underset{\epsilon \to 0}{lim}-x\left({t}_{i}+\epsilon \right)$ represent the right and left limits of x(t) at t = ti, respectively, $\left({x}^{\prime }\left({\overline{t}}_{j}^{+}\right),\phantom{\rule{thinmathspace}{0ex}}{x}^{\prime }\left({\overline{t}}_{j}^{-}\right)$ and $\left({x}^{″}\left({\stackrel{^}{t}}_{j}^{+}\right),\phantom{\rule{thinmathspace}{0ex}}{x}^{″}\left({\stackrel{^}{t}}_{j}^{-}\right)$ have similar meaning for x′(t) at $t={\overline{t}}_{j}$ and x″(t) at $t={\stackrel{^}{t}}_{l}$, respectively).

Next, take $a,{t}_{1},{t}_{2}\cdots ,{t}_{L},{\overline{t}}_{1},{\overline{t}}_{2},\cdots ,{\overline{t}}_{M},{\stackrel{^}{t}}_{1},{\stackrel{^}{t}}_{2},\cdots ,{\stackrel{^}{t}}_{N},T$ to $a={t}_{0}^{\prime }<{t}_{1}^{\prime }<\cdots <{t}_{K}^{\prime }<{t}_{K+1}^{\prime }=T$ such that

$set{t1,t2,⋯,tL,t¯1,t¯2,⋯,t¯M,t^1,t^2,⋯,t^N}=set{t1′,t2′,⋯tK′}$

Let ${J}_{0}^{\prime }=\left[a,{t}_{1}^{\prime }\right]$ and ${J}_{k}^{\prime }=\left({t}_{k}^{\prime },{t}_{k+1}^{\prime },\right]\phantom{\rule{thinmathspace}{0ex}}\left(k=0,1,2,\cdots ,K\right)$. For each $\left[a,{t}_{k}^{\prime }\right]$ (here k = 0,1,2,…, K), assume $\left[a,{t}_{{k}_{0}}\right]\subseteq \left[a,{t}_{k}^{\prime }\right]\subset \left[a,{t}_{{k}_{0}+1}\right]$ (here k0 ∈ {1,2,…, L}) and $\left[a,{t}_{{k}_{1}}^{\prime }\right]\subseteq \left[a,{t}_{k}^{\prime }\right]\subset \left[a,{\overline{t}}_{{k}_{1}+1}\right]$ (here k1 ∈ {1,2,…,M}) and $\left[a,{\stackrel{^}{t}}_{{k}_{2}}\right]\subseteq \left[a,{t}_{k}^{\prime }\right]\subset \left[a,{\stackrel{^}{t}}_{{k}_{2}+1}\right]$ (here k2 ∈ {1,2,…, N}) respectively.

With simplification of system (1), we get

$aDtqx(t)=f(t,x(t)),q∈(2,3),t∈J=[a,T],t≠ti(k=1,2,…,L),Δxt=ti=Iix(ti−),i=1,2,…,L,Δx′t=ti=I¯ix(ti−),i=1,2,…,L,Δx″t=ti=I^ix(ti−),i=1,2,…,L,x(a)=xa,x′(a)=x¯a,x″(a)=x^a.$(2)

Let J0 = [a, t1] and Ji = (ti, ti + 1] (i = 1,2,…, L). Considering some limiting cases in (1), we have

$limIi(x(ti−))→0foralli∈{1,2,…,L}andI¯j(x(t¯j−))→0forallj∈{1,2,…,M}andI^l(x(t^l−))→0foralll∈{1,2,…,N}{impulsesystem(1)}$

$→aDtqx(t)=f(t,x(t)),q∈(2,3),t∈J=[a,T],x(a)=xa,x′(a)=x¯a,x″(a)=x^a.$(3)

$limI¯j(x(t¯j−))→0forallj∈{1,2,…,M}andI^l(x(t^l−))→0foralll∈{1,2,…,N}{impulsesystem(1)}$

$→aDtqx(t)=f(t,x(t)),q∈(2,3),t∈J=[a,T],t≠ti(i=1,2,…,L),Δxt=ti=Iix(ti−),i=1,2,…,L,x(a)=xa,x′(a)=x¯a,x″(a)=x^a.$(4)

$limIi(x(ti−))→0foralli∈{1,2,…,L}andI^l(x(t^l−))→0foralll∈{1,2,…,N}{impulsesystem(1)}$

$→aDtqx(t)=f(t,x(t)),q∈(2,3),t∈J=[a,T],t≠t¯j(j=1,2,…,M),Δx′t=t¯j=I¯jx(t¯j−),j=1,2,…,M,x(a)=xa,x′(a)=x¯a,x″(a)=x^a.$(5)

$limIi(x(ti−))→0foralli∈{1,2,…,L}andI¯j(x(t¯j−))→0forallj∈{1,2,…,M}{impulsesystem(1)}$

$→aDtqx(t)=f(t,x(t)),q∈(2,3),t∈J=[a,T],t≠t^l(l=1,2,…,N),Δx″t=t^l=I^lx(t^l−),l=1,2,…,N,x(a)=xa,x′(a)=x¯a,x″(a)=x^a.$(6)

$limI^l(x(t^l−))→0foralll∈{1,2,…,N}{impulsesystem(1)}$

$→aDtqx(t)=f(t,x(t)),q∈(2,3),t∈J=[a,T],t≠ti(i=1,2,…,L)andt≠t¯j(j=1,2,…,M),Δxt=ti=Iix(ti−),i=1,2,…,L,Δx′t=t¯j=I¯jx(t¯j−),j=1,2,…,M,x(a)=xa,x′(a)=x¯a,x″(a)=x^a.$(7)

$limI¯j(x(t¯j−))→0forallj∈{1,2,…,M}{impulsesystem(1)}$

$→aDtqx(t)=f(t,x(t)),q∈(2,3),t∈J=[a,T],t≠ti(i=1,2,…,L)andt≠t^l(l=1,2,…,N),Δxt=ti=Iix(ti−),i=1,2,…,L,Δx″t=t^l=I^lx(t^l−),l=1,2,…,N,x(a)=xa,x′(a)=x¯a,x″(a)=x^a.$(8)

$limIi(x(ti−))→0foralli∈{1,2,…,L}{impulsesystem(1)}$

$→aDtqx(t)=f(t,x(t)),q∈(2,3),t∈J=[a,T],t≠t¯j(j=1,2,…,M)andt≠t^l(l=1,2,…,N),Δx′t=t¯j=I¯jx(t¯j−),j=1,2,…,M,Δx″t=t^l=I^lx(t^l−),l=1,2,…,N,x(a)=xa,x′(a)=x¯a,x″(a)=x^a.$(9)

This means that the solution of (1) satisfies:

(i) $\underset{\begin{array}{l}I\left(x\left({t}_{j}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}i\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,L\right\}\\ and\phantom{\rule{thinmathspace}{0ex}}{\overline{I}}_{j}\left(x\left({\overline{t}}_{j}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}j\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,M\right\}\\ and\phantom{\rule{thinmathspace}{0ex}}{\stackrel{^}{I}}_{l}\left(x\left({\stackrel{^}{t}}_{j}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}l\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,N\right\}\end{array}}{lim}\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(1\right)\right\}\phantom{\rule{thinmathspace}{0ex}}=\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(3\right)\right),$

(ii) $\underset{\begin{array}{l}{\overline{I}}_{j}\left(x\left({t}_{j}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}j\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,M\right\}\\ and\phantom{\rule{thinmathspace}{0ex}}{\stackrel{^}{I}}_{l}\left(x\left({\stackrel{^}{t}}_{l}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}l\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,N\right\}\end{array}}{lim}\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(1\right)\right\}\phantom{\rule{thinmathspace}{0ex}}=\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(4\right)\right),$

(iii) $\underset{\begin{array}{l}{I}_{i}\left(x\left({t}_{i}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}i\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,L\right\}\\ and\phantom{\rule{thinmathspace}{0ex}}{\stackrel{^}{I}}_{l}\left(x\left({\stackrel{^}{t}}_{l}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}l\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,N\right\}\end{array}}{lim}\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(1\right)\right\}\phantom{\rule{thinmathspace}{0ex}}=\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(5\right)\right),$

(iv) $\underset{\begin{array}{l}{I}_{i}\left(x\left({t}_{i}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}i\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,L\right\}\\ and\phantom{\rule{thinmathspace}{0ex}}{\overline{I}}_{l}\left(x\left({\stackrel{^}{t}}_{j}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}j\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,M\right\}\end{array}}{lim}\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(1\right)\right\}\phantom{\rule{thinmathspace}{0ex}}=\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(6\right)\right),$

(v) $\underset{\begin{array}{l}{\stackrel{^}{I}}_{l}\left(x\left({\stackrel{^}{t}}_{i}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}l\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,N\right\}\end{array}}{lim}\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(1\right)\right\}\phantom{\rule{thinmathspace}{0ex}}=\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(7\right)\right),$

(vi) $\underset{\begin{array}{l}{\overline{I}}_{j}\left(x\left({\overline{t}}_{i}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}j\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,M\right\}\end{array}}{lim}\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(1\right)\right\}\phantom{\rule{thinmathspace}{0ex}}=\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(8\right)\right),$

(vii) $\underset{\begin{array}{l}{I}_{i}\left(x\left({t}_{i}^{-}\right)\right)\to 0\phantom{\rule{1em}{0ex}}for\phantom{\rule{thinmathspace}{0ex}}all\phantom{\rule{thinmathspace}{0ex}}\phantom{\rule{thinmathspace}{0ex}}i\phantom{\rule{thinmathspace}{0ex}}\in \left\{1,2,\cdots ,L\right\}\end{array}}{lim}\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(1\right)\right\}\phantom{\rule{thinmathspace}{0ex}}=\left\{\mathrm{t}\mathrm{h}\mathrm{e}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\phantom{\rule{thinmathspace}{0ex}}\mathrm{o}\mathrm{f}\phantom{\rule{thinmathspace}{0ex}}\mathrm{s}\mathrm{y}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{m}\left(9\right)\right).$

Thus, we present the definition of solution for (1) as follows

A function z(t) : [a, T] → ℝ is said to be the solution of impulsive system (1) if z(a) = xa, ${z}^{\prime }\left(a\right)={\overline{x}}_{a}$ and ${z}^{″}\left(a\right)={\stackrel{^}{x}}_{a}$, the equation condition ${}_{a}{D}_{t}^{q}z\left(t\right)=f\left(t,z\left(t\right)\right)$ for each t ∈ (a, T] is verified, impulsive conditions $\mathrm{\Delta }z{|}_{t={t}_{i}}={I}_{i}\left(z\left({t}_{i}^{-}\right)\right)$(here i = 1,2,…, L), $\mathrm{\Delta }{z}^{\prime }{|}_{t={\overline{t}}_{i}}={\overline{I}}_{i}\left(z\left({\overline{t}}_{i}^{-}\right)\right)$ (here j = 1,2, …,M) and $\mathrm{\Delta }{z}^{″}{|}_{t={\stackrel{^}{t}}_{l}}={\stackrel{^}{I}}_{i}\left(z\left({\stackrel{^}{t}}_{i}^{-}\right)\right)$ (here l = 1,2,…,N) are satisfied, the restriction of to the interval ${J}_{k}^{\prime }$ (here k = 0,1,2,…, K) is continuous, and conditions (i)-(vii) hold.

Define a function

$x~(t)=x(tk+)+x′(tk+)(t−tk)+x″(tk+)2!(t−tk)2+1Γ(q)∫tkt(t−s)q−1f(s,x(s))dsfort∈(tk,tk+1].$

By the definition of Caputo fractional derivative, we have

$[aDtqx~(t)]t∈(tk,tk+1]=aDtqx(tk+)+x′(tk+)(t−tk)+x″(tk+)2!(t−tk)2+1Γ(q)∫tkt(t−s)q−1f(s,x(s))dst∈(tk,tk+1]=aDtq1Γ(q)∫tkt(t−s)q−1f(s,x(s))dst∈(tk,tk+1]=tkDtq1Γ(q)∫tkt(t−s)q−1f(s,x(s))dst∈(tk,tk+1]=f(t,x(t))|t∈(tk,tk+1].$

Therefore, $\stackrel{~}{x}\left(t\right)$ can meet the condition of fractional derivative and impulsive conditions in (1). But, $\stackrel{~}{x}\left(t\right)$ is only considered as an approximate solution of (1) since it doesn’t satisfy conditions (i)-(vii).

Next, we provide some definitions and conclusions in Section 2, and prove the formula of general solution for (1) in Section 3. Finally, an example is provided to expound the main result in Section 4.

## 2 Preliminaries

#### ([2])

The fractional integral of order q for function x is defined as

$aItqx(t)=1Γ(q)∫atx(s)(t−s)1−qds,t>a,q>0,$

where Γ is the gamma function.

#### ([2])

The Caputo fractional derivative of order q for a function x can be written as

$aDtqx(t)=1Γ(n−q)∫atx(n)(s)(t−s)q+1−nds=aItn−qx(n)(t),t>a,0≤n−1

#### ([39])

If the function h(t, x) is continuous, then the initial value problem

$aDtqx(t)=h(t,x(t)),q∈(n,n+1],n∈R+∪{0},x(k)(a)=xak,k=0,1,2,…,n.$

is equivalent to the following nonlinear Volterra integral equation of the second kind,

$x(t)=∑k=0nxakk!(t−a)k+1Γ(q)∫at(t−s)q−1h(s,x(s))ds,$

and its solutions are continuous.

#### ([32])

Let q ∈ (0, 1) and ξ is a constant. Impulsive system

$0Dtqu(t)=h(t,u(t)),t∈J=[0,T],t≠tk,k=1,2,…,m,Δut=tk=Iku(tk−),t=tk,k=1,2,…,m,u(0)=u0.$

is equivalent to the fractional integral equation

$u(t)=u0+1Γ(q)∫0t(t−s)q−1h(s,(u(s))ds,fort∈[0,t1],u0+∑k=1nIk(u(tk−))+1Γ(q)∫0t(t−s)q−1h(s,u(s))ds+ξΓ(q)∑k=1nIk(u(tk−))∫0tk(tk−s)q−1h(s,u(s))ds+∫tkt(t−s)q−1h(s,u(s))ds−∫0t(t−s)q−1h(s,u(s))dsfort∈(tn,tn+1],1≤n≤m,$(10)

provided that the integral in (10) exists.

## 3 Main results

For convenience, let f = f(s, x(s)) and ${\sum }_{i=1}^{0}{y}_{i}=0$ in this section.

Let q ∈ (2, 3) and ξ0 is a constant. System (4) is equivalent to the fractional integral equation

$x(t)=xa+x¯a(t−a)+x^a2!(t−a)2+∑i=1kIix(ti−)+1Γ(q)∫at(t−s)q−1fds+∑i=1kξ0Ii(x(ti−))1Γ(q)∫ati(ti−s)q−1fds+∫tit(t−s)q−1fds−∫at(t−s)q−1fds+(t−ti)Γ(q−1)∫ati(ti−s)q−2fds+(t−ti)22!Γ(q−2)∫ati(ti−s)q−3fdsfort∈(tk,tk+1],k=0,1,2,…,L.$(11)

provided that the integral in (11) exists.

With similarity to Lemma 3.1, the following two conclusions can be proved.

Let q ∈ (2,3) and ξ1 is a constant. System (5) is equivalent to the fractional integral equation

$x(t)=xa+x¯a(t−a)+x^a2!(t−a)2+∑j=1kI¯jx(t¯j−)(t−t¯j)+1Γ(q)∫at(t−s)q−1fds+∑j=1kξ1I¯j(x(t¯j−))1Γ(q)∫at¯j(t¯j−s)q−1fds+∫t¯jt(t−s)q−1fds−∫at(t−s)q−1fds+(t−t¯j)Γ(q−1)∫at¯j(t¯j−s)q−2fds+(t−t¯j)22!Γ(q−2)∫at¯j(t¯j−s)q−3fdsfort∈(t¯k,t¯k+1],k=0,1,2,...,M.$(32)

provided that the integral in (32) exists.

Let q ∈ (2, 3) ξ2 and is a constant. System (6) is equivalent to the fractional integral equation

$x(t)=xa+x¯a(t−a)+x^a2!(t−a)2+12!∑l=1kI^lx(t^l−)(t−t^l)2+1Γ(q)∫at(t−s)q−1fds+∑l=1kξ2I^l(x(t^l−))1Γ(q)∫at^l(t^l−s)q−1fds+∫t^lt(t−s)q−1fds−∫at(t−s)q−1fds+(t−t^l)Γ(q−1)∫at^l(t^l−s)q−2fds+(t−t^l)22!Γ(q−2)∫at^l(t^l−s)q−3fdsfort∈(t^k,t^k+1],k=0,1,2,...,N.$(33)

provided that the integral in (33) exists.

Let q ∈ (2, 3) and ξ0 is a constant. If a function x is the general solution of system (4) then

$x″(t)=x^a+1Γ(q−2)∫at(t−s)q−3f(s,x(s))ds+ξ0∑i=1kIi(x(ti−))Γ(q−2)∫ati(ti−s)q−3fds+∫tit(t−s)q−3fds−∫at(t−s)q−3fdsfort∈(tk,tk+1],k=0,1,2,...,L.$

Let q ∈ (2, 3) and ξ1 is a constant. If a function x is the general solution of system (5) then

$x″(t)=x^a+1Γ(q−2)∫at(t−s)q−3fds+ξ1∑j=1kI¯j(x(t¯j−))Γ(q−2)∫at¯j(t¯j−s)q−3fds+∫t¯jt(t−s)q−3fds−∫at(t−s)q−3fdsfort∈(t¯k,t¯k+1],k=0,1,2,...,M.$

Let q ∈ (2, 3) and ξ2 is a constant. If a function x is the general solution of system (6) then

$x″(t)=x^a+∑l=1kI^lx(t^l−)+1Γ(q−2)∫at(t−s)q−3fds+ξ2∑l=1kI^l(x(t^l−))Γ(q−2)∫at^l(t^l−s)q−3fds+∫t^lt(t−s)q−3fds−∫at(t−s)q−3fdsfort∈(t^k,t^k+1],k=0,1,2,...,N.$

Impulses ${\mathrm{\Delta }x|}_{t={t}_{i}},\phantom{\rule{thickmathspace}{0ex}}{\mathrm{\Delta }{x}^{\prime }|}_{t={\overline{t}}_{j}}$ and ${\mathrm{\Delta }{x}^{″}|}_{t={\stackrel{^}{t}}_{l}}$ have similar effect on x″(t) of (3) by Corollaries 3.4-3.6. Thus, we will consider ${\mathrm{\Delta }x|}_{t={t}_{i}}$ and ${\mathrm{\Delta }{x}^{\prime }|}_{t={\overline{t}}_{j}}$ as some special impulses ${\mathrm{\Delta }{x}^{″}|}_{t={\stackrel{^}{t}}_{l}}$ for system (3).

Let q ∈ (2, 3) and ξb (where b ∈ {0,1,2}) are three constants. If a function x is the general solution of system (1) then

$x″(t)=x^a+∑l=1k2I^lx(t^l−)+1Γ(q−2)∫at(t−s)q−3fds+ξ0∑i=1k0Ii(x(ti−))Γ(q−2)∫ati(ti−s)q−3fds+∫tit(t−s)q−3fds−∫at(t−s)q−3fds+ξ1∑j=1k1I¯j(x(t¯j−))Γ(q−2)∫at¯j(t¯j−s)q−3fds+∫t¯jt(t−s)q−3fds−∫at(t−s)q−3fds+ξ2∑l=1k2I^l(x(t^l−))Γ(q−2)∫at^l(t^l−s)q−3fds+∫t^lt(t−s)q−3fds−∫at(t−s)q−3fdsfort∈Jk′,k=0,1,2,...,K.$(34)

Let q ∈ (2,3) and ξb (where b ∈ {0, 1, 2}) are three constants. System (1) is equivalent to the fractional integral equation

$x(t)=xa+x¯a(t−a)+x^a2!(t−a)2+∑i=1k0Iix(ti−)+∑j=1k1I¯jx(t¯j−)(t−t¯j)+12!∑l=1k2I^lx(t^l−)(t−t^l)2+1Γ(q)∫at(t−s)q−1fds+∑i=1k0ξ0Ii(x(ti−))1Γ(q)∫ati(ti−s)q−1fds+∫tit(t−s)q−1fds−∫at(t−s)q−1fds+(t−ti)Γ(q−1)∫ati(ti−s)q−2fds+(t−ti)22!Γ(q−2)∫ati(ti−s)q−3fds+∑j=1k1ξ1I¯j(x(t¯j−))1Γ(q)∫at¯j(t¯j−s)q−1fds+∫t¯jt(t−s)q−1fds−∫at(t−s)q−1fds+(t−t¯j)Γ(q−1)∫at¯j(t¯j−s)q−2fds+(t−t¯j)22!Γ(q−2)∫at¯j(t¯j−s)q−3fds+∑l=1k2ξ2I^l(x(t^l−))1Γ(q)∫at^l(t^l−s)q−1fds+∫t^lt(t−s)q−1fds−∫at(t−s)q−1fds+(t−t^l)Γ(q−1)∫at^l(t^l−s)q−2fds+(t−t^l)22!Γ(q−2)∫at^l(t^l−s)q−3fdsfort∈Jk′,$(35)

here k ∈{0, 1, 2,…, K}, provided that the integral in (35) exists.

For impulsive system (1), we have

$→x(3)(t)=f(t,x(t)),t∈J=[a,T],t≠ti(i=1,2,…,L)andt≠t¯j(j=1,2,…,M)andt≠t^l(l=1,2,…,N),Δxt=ti=Iix(tj−)i=1,2,…,L,Δx′t=t¯j=I¯ix(t¯j−)j=1,2,…,M,Δx″t=t^l=I^lx(t^l−)l=1,2,…,N,x(a)=xa,x′(a)=x¯a,x″(a)=x^a.$(36)

On the other hand, using (35), we get

$limq→3−⁡x(t)=xa+x¯a(t−a)+x^a2!(t−a)2+12∫at(t−s)2fds,fort∈J0′,xa+x¯a(t−a)+x^a2!(t−a)2+∑i=1k0Iix(ti−)+∑j=1k1I¯jx(t¯j−)(t−t¯j)+12!∑l=1k2I^lx(t^l−)(t−t^l)2+12∫at(t−s)2fds,fort∈Jk′,k=1,2,…K.$(37)

Moreover, we can verify that Eq. (37) is the solution of (36), and indirectly supports our results.

Let q ∈ (2, 3) and ξb (where b ∈ {0,1,2}) are three constants. System (2) is equivalent to the fractional integral equation

$x(t)=xa+x¯a(t−a)+x^a2!(t−a)2+∑i=1kIix(ti−)+I¯ix(t¯i−)(t−t¯i)+12!I^ix(ti−)(t−ti)2+1Γ(q)∫at(t−s)q−1fds+∑i=1kξ0Iix(ti−)+ξ1I¯ix(ti−)+ξ2I^ix(ti−)×1Γ(q)∫ati(ti−s)q−1fds+∫tit(t−s)q−1fds−∫at(t−s)q−1fds+(t−ti)Γ(q−1)∫ati(ti−s)q−2fds+(t−ti)22!Γ(q−2)∫ati(ti−s)q−3fdsfort∈Jk,k=0,1,2,…,K.$(38)

provided that the integral in (38) exists.

## 4 Example

The analytical solution of system (1) is difficult to obtain when f is a nonlinear function in (1). So, let us consider a linear impulsive system.

$0Dt9/4x(t)=t,t∈[0,2]∖{1},x(1+)=x(1−)+I(x(1−)),x′(1+)=x′(1−)+I¯(x(1−)),x″(1+)=x″(1−)+I^(x(1−)),x(0)=x0,x′(0)=x¯0,x″(0)=x^0.$(39)

Next, we give the general solution by

$x(t)=x0+x¯0t+x^02!t2+1Γ(94)169×13t134fort∈[0,1],x0+x¯0t+x^02!t2+I(x(1−))+I¯(x(1−))(t−1)+12!I^(x(1−))(t−1)2+1Γ(94)169×13t134t≥0+ξ0I(x(1−))+ξ1I¯(x(1−))+ξ2I^(x(1−))1Γ(94)169×13+1Γ(94)99×13(t−1)94(4t+9)t≥1−1Γ(94)169×13t134t≥0+(t−1)Γ(94+1)t≥1+(t−1)22!Γ(94)t≥1fort∈(1,2].$(40)

where ξ0, ξ1 and ξ2 are arbitrary constants.

Next, for Eq. (40), we have

$0Dt94x(t)=0Dt94x0+x¯0t+x^02!t2+1Γ(94)169×13t134=tfort∈[0,1],$

and

$0Dt94x(t)=0Dt94x0+x¯0t+x^02!t2+I(x(1−))+I¯(x(1−))(t−1)+12!I^(x(1−))(t−1)2+1Γ(94)169×13t134t≥0+ξ0I(x(1−))+ξ1I¯(x(1−))+ξ2I^(x(1−))1Γ(94)169×13+1Γ(94)49×13(t−1)94(4t+9)t≥1−1Γ(94)169×13t134t≥0+(t−1)Γ(94+1)t≥1+(t−1)22!Γ(94)t≥1t∈(1,2]=0Dt941Γ(94)169×13t134t≥0+ξ0I(x(1−))+ξ1I¯(x(1−))+ξ2I^(x(1−))Γ(94)49×13(t−1)94(4t+9)t≥1−4t134t≥0t∈(1,2]=tt∈[0,2]t∈(1,2]+ξ0I(x(1−))+ξ1I¯(x(1−))+ξ2I^(x(1−))[tt∈(1,2]−tt∈[0,2]]t∈(1,2]=t,t∈(1,2].$

So, Eq. (40) satisfies Caputo fractional derivative condition in (39).

Secondly, we can verify that Eq. (40) satisfies

$x(1+)=x(1−)+I(x(1−)),x′(1+)=x′(1−)+I¯(x(1−))andx″(1+)=x″(1−)+I^(x(1−)).$

Moreover, we can verify that Eq. (40) satisfies the corresponding conditions (i)-(vii) of (39). Therefore, Eq. (40) is the general solution of (39).

## Acknowledgement

The authors are deeply grateful to the anonymous referees for their kind comments, correcting errors and improving written language, which have been very useful for improving the quality of this paper. The work described in this paper is financially supported by the National Natural Science Foundation of China (Grant No. 21576033) and the Natural Science Foundation of Jiangxi Province (Grant No. 20151BAB207013) and Jiujiang University Research Foundation (Grant No. 8400183).

## References

• [1]

Podlubny I., Fractional Differential Equations, Academic Press, San Diego, 1999 Google Scholar

• [2]

Kilbas A.A., Srivastava H.H., Trujillo J.J., Theory and Applications of Fractional Differential Equations, Elsevier, Amsterdam, 2006 Google Scholar

• [3]

Baleanu D., Diethelm K., Scalas E., Trujillo J.J., Fractional Calculus Models and Numerical Methods, Series on Complexity, Nonlinearity and Chaos, World Scientific, Singapore, 2012Google Scholar

• [4]

Ye H., Gao J., Ding Y., A generalized Gronwall inequality and its application to a fractional differential equation, J. Math. Anal. Appl., 2007, 328 (2), 1075-1081Google Scholar

• [5]

Benchohra M., Henderson J., Ntouyas S.K., Ouahab A., Existence results for fractional order functional differential equations with infinite delay, J. Math. Anal. Appl., 2008, 338 (2), 1340–1350Google Scholar

• [6]

Agarwal R.P., Benchohra M., Hamani S., A survey on existence results for boundary value problems of nonlinear fractional differential equations and inclusions, Acta. Appl. Math., 2010, 109 (3), 973–1033Google Scholar

• [7]

Odibat Z.M., Analytic study on linear systems of fractional differential equations, Comput. Math. Appl., 2010, 59 (3), 1171-1183Google Scholar

• [8]

Ahmad B., Nieto J.J., Existence of solutions for anti-periodic boundary value problems involving fractional differential equations via Leray-Schauder degree theory, Topol. Methods Nonlinear Anal., 2010, 35(2), 295-304Google Scholar

• [9]

Bai Z., On positive solutions of a nonlocal fractional boundary value problem, Nonlinear Anal.: TMA, 2010, 72 (2), 916-924Google Scholar

• [10]

Mophou G.M., N’Guérékata G.M., Existence of mild solutions of some semilinear neutral fractional functional evolution equations with infinite delay, Appl. Math. Comput., 2010, 216 (1), 61–69Google Scholar

• [11]

Deng W., Smoothness and stability of the solutions for nonlinear fractional differential equations, Nonlinear Anal.: TMA., 2010, 72 (3-4), 1768-1777Google Scholar

• [12]

Kilbas A.A., Hadamard-type fractional calculus, J. Korean Math. Soc., 2001, 38(6), 1191-1204Google Scholar

• [13]

Butzer P.L., Kilbas A.A., Trujillo J.J., Compositions of Hadamard-type fractional integration operators and the semigroup property, J. Math. Anal. Appl., 2002, 269, 387-400Google Scholar

• [14]

Butzer P.L., Kilbas A.A., Trujillo J.J., Mellin transform analysis and integration by parts for Hadamard-type fractional integrals, J. Math. Anal. Appl., 2002, 270, 1-15Google Scholar

• [15]

Thiramanus P., Ntouyas S.K., Tariboon J., Existence and Uniqueness Results for Hadamard-Type Fractional Differential Equations with Nonlocal Fractional Integral Boundary Conditions, Abstr. Appl. Anal., 2014, 2014, Article ID 902054, 9 pagesGoogle Scholar

• [16]

Klimek M., Sequential fractional differential equations with Hadamard derivative, Commun. Nonlinear Sci. Numer. Simul., 2011, 16, 4689-4697Google Scholar

• [17]

Ahmad B., Ntouyas S.K., A fully Hadamard type integral boundary value problem of a coupled system of fractional differential equations, Fract. Calc. Appl. Anal., 2014, 17, 348-360Google Scholar

• [18]

• [19]

• [20]

Ahmad B., Sivasundaram S., Existence results for nonlinear impulsive hybrid boundary value problems involving fractional differential equations, Nonlinear Anal. Hybrid Syst., 2009, 3, 251-258 Google Scholar

• [21]

Ahmad B., Sivasundaram S., Existence of solutions for impulsive integral boundary value problems of fractional order, Nonlinear Anal. Hybrid Syst., 2010, 4, 134-141Google Scholar

• [22]

Tian Y., Bai Z., Existence results for the three-point impulsive boundary value problem involving fractional differential equations, Comput. Math. Appl., 2010, 59, 2601-2609Google Scholar

• [23]

Cao J., Chen H., Some results on impulsive boundary value problem for fractional differential inclusions, Electron. J. Qual. Theory Differ. Equ., 2011, 11, 1-24Google Scholar

• [24]

Wang X., Impulsive boundary value problem for nonlinear differential equations of fractional order, Comput. Math. Appl., 2011, 62, 2383-2391Google Scholar

• [25]

Stamova I., Stamov G., Stability analysis of impulsive functional systems of fractional order, Commun Nonlinear Sci Numer Simulat., 2014, 19, 702-709Google Scholar

• [26]

Abbas S., Benchohra M., Impulsive partial hyperbolic functional differential equations of fractional order with state-dependent delay, Fract. Calc. Appl. Anal., 2010, 13(3), 225-244Google Scholar

• [27]

Abbas S., Benchohra M., Upper and lower solutions method for impulsive partial hyperbolic differential equations with fractional order, Nonlinear Anal. Hybrid Syst., 2010, 4(3), 406-413Google Scholar

• [28]

Abbas S., Agarwal R.P., Benchohra M., Darboux problem for impulsive partial hyperbolic differential equations of fractional order with variable times and infinite delay, Nonlinear Anal. Hybrid Syst., 2010, 4(4), 818-829Google Scholar

• [29]

Abbas S., Benchohra M., Gorniewicz L., Existence theory for impulsive partial hyperbolic functional differential equations involving the Caputo fractional derivative, Sci. Math. Jpn., 2010, 72(1), 49-60Google Scholar

• [30]

Benchohra M., Seba D., Impulsive partial hyperbolic fractional order differential equations in banach spaces, J. Fract. Calc. Appl., 2011, 1 (4), 1-12Google Scholar

• [31]

Guo T., Zhang K., Impulsive fractional partial differential equations, Appl. Math. Comput., 2015, 257, 581-590 Google Scholar

• [32]

Zhang X., Zhang X., Zhang M., On the concept of general solution for impulsive differential equations of fractional order q ∈ (0,1), Appl. Math. Comput., 2014, 247, 72-89 Google Scholar

• [33]

Zhang X., The general solution of differential equations with Caputo-Hadamard fractional derivatives and impulsive effect, Adv. Differ. Equ., 2015, 2015, Article ID 215, 16 pagesGoogle Scholar

• [34]

Zhang X., On the concept of general solutions for impulsive differential equations of fractional order q ∈ (1, 2), Appl. Math. Comput., 2015, 268, 103-120Google Scholar

• [35]

Zhang X., Agarwal P., Liu Z., Peng H., The general solution for impulsive differential equations with Riemann-Liouville fractionalorder q ∈ (1,2), Open Math., 2015, 13, 908-923Google Scholar

• [36]

Zhang X., Shu T., Cao H., Liu Z., Ding W., The general solution for impulsive differential equations with Hadamard fractional derivative of order q ∈ (1,2), Adv. Differ. Equ., 2016, 2016, Article ID 14, 36 pagesGoogle Scholar

• [37]

Zhang X., Zhang X., Liu Z., Peng H., Shu T., Yang S., The General Solution of Impulsive Systems with Caputo-Hadamard Fractional Derivative of Order q ∈ ℂ(ℜ(q) ∈ (1,2)), Math. Prob. Eng., 2016, 2016, Article ID 8101802, 20 pagesGoogle Scholar

• [38]

Zhang X., Zhang X., Liu Z., Ding W., Cao H., Shu T., On the general solution of impulsive systems with Hadamard fractional derivatives, Math. Prob. Eng., (in press), http://www.hindawi.com/journals/mpe/aip/2814310/

• [39]

Diethelm K., Ford N.J., Analysis of fractional differential equations, J. Math. Anal. Appl., 2002, 265 (2), 229-248 Google Scholar

E-mail: ,

Accepted: 2016-06-08

Published Online: 2016-07-08

Published in Print: 2016-01-01

Citation Information: Open Mathematics, Volume 14, Issue 1, Pages 452–473, ISSN (Online) 2391-5455,

Export Citation