## Boundary Value Problems

Impact Factor 0.819

Open Access

# Positive solutions of third-order nonlocal boundary value problems at resonance

Boundary Value Problems20122012:102

DOI: 10.1186/1687-2770-2012-102

Received: 21 April 2012

Accepted: 3 September 2012

Published: 18 September 2012

## Abstract

In this paper, we investigate the existence of positive solutions for a class of third-order nonlocal boundary value problems at resonance. Our results are based on the Leggett-Williams norm-type theorem, which is due to O’Regan and Zima. An example is also included to illustrate the main results.

MSC:34B10, 34B15.

### Keywords

third-order nonlocal at resonance positive solution

## 1 Introduction

This paper is devoted to the existence of positive solutions for the following third-order nonlocal boundary value problem (BVP for short):
$\left\{\begin{array}{c}{x}^{\mathrm{\prime }\mathrm{\prime }\mathrm{\prime }}\left(t\right)+f\left(t,x\left(t\right)\right)=0,\phantom{\rule{1em}{0ex}}0
(1.1)
where $0<{\xi }_{1}<{\xi }_{2}<\cdots <{\xi }_{m-2}<1$, ${\alpha }_{i}\in {R}^{+}$ ($i=1,2,\dots ,m-2$) and ${\sum }_{i=1}^{m-2}{\alpha }_{i}=1$. The problem (1.1) happens to be at resonance in the sense that the associated linear homogeneous BVP
$\left\{\begin{array}{c}-{x}^{\mathrm{\prime }\mathrm{\prime }\mathrm{\prime }}\left(t\right)=0,\phantom{\rule{1em}{0ex}}0
(1.2)

has nontrivial solutions. Clearly, the resonant condition is ${\sum }_{i=1}^{m-2}{\alpha }_{i}=1$. Third-order differential equations arise in a variety of different areas of applied mathematics and physics, e.g., in the deflection of a curved beam having a constant or varying cross section, a three-layer beam, electromagnetic waves or gravity-driven flows and so on [1].

Recently, the existence of positive solutions for third-order two-point or multi-point BVPs has received considerable attention; we mention a few works: [211] and the references therein. However, all of the papers on third-order BVPs focused their attention on the positive solutions with non-resonance cases. It is well known that the problem of the existence of positive solutions to BVPs is very difficult when the resonant case is considered. Only few papers deal with the existence of positive solutions to BVPs at resonance, and just to second-order BVPs [1215]. It is worth mentioning that Infante and Zima [13] studied the existence of positive solutions for the second-order m-point BVP
$\left\{\begin{array}{c}-{x}^{\mathrm{\prime }\mathrm{\prime }}\left(t\right)+f\left(t,x\left(t\right)\right)=0,\phantom{\rule{1em}{0ex}}0
(1.3)

by means of the Leggett-Williams norm-type theorem due to O’Regan and Zima [16], where $0<{\eta }_{1}<{\eta }_{2}<\cdots <{\eta }_{m-2}<1$, ${\alpha }_{i}\in {R}^{+}$ ($i=1,2,\dots ,m-2$) and ${\sum }_{i=1}^{m-2}{\alpha }_{i}=1$.

However, third-order or higher-order derivatives do not have the convexity; to the best of our knowledge, no results are available for the existence of positive solutions for third-order or higher- order BVPs at resonance. The main purpose of this paper is to fill the gap in this area. Motivated greatly by the above-mentioned excellent works, in this paper we will investigate the third-order nonlocal BVP (1.1) at resonance, where $0<{\xi }_{1}<{\xi }_{2}<\cdots <{\xi }_{m-2}<1$, ${\alpha }_{i}\in {R}^{+}$ ($i=1,2,\dots ,m-2$) and ${\sum }_{i=1}^{m-2}{\alpha }_{i}=1$. Some new existence results of at least one positive solution are established by applying the Leggett-Williams norm-type theorem due to O’Regan and Zima [16]. An example is also included to illustrate the main results.

## 2 Some definitions and a fixed point theorem

For the convenience of the reader, we present here the necessary definitions and a new fixed point theorem due to O’Regan and Zima.

Definition 2.1 Let X and Z be real Banach spaces. A linear operator $L:domL\subset X\to Z$ is called a Fredholm operator if the following two conditions hold:
1. (i)

KerL has a finite dimension, and

2. (ii)

ImL is closed and has a finite codimension.

Throughout the paper, we will assume that

1L is a Fredholm operator of index zero, that is, ImL is closed and $dimKerL=codimImL<\mathrm{\infty }$.

From Definition 2.1, it follows that there exist continuous projectors $P:X\to X$ and $Q:Z\to Z$ such that
$ImP=KerL;\phantom{\rule{2em}{0ex}}KerQ=ImL;\phantom{\rule{2em}{0ex}}X=KerL\oplus KerP;\phantom{\rule{2em}{0ex}}Z=ImL\oplus ImQ$
and that the isomorphism
$L{|}_{domL\cap KerP}:domL\cap KerP\to ImL$
is invertible. We denote the inverse of $L{|}_{domL\cap KerP}$ by ${K}_{P}:ImL\to domL\cap KerP$. The generalized inverse of L denoted by ${K}_{P,Q}:Z\to domL\cap KerP$ is defined by ${K}_{P,Q}={K}_{P}\left(I-Q\right)$. Moreover, since $dimImQ=codimImL$, there exists an isomorphism $J:ImQ\to KerL$. Consider a nonlinear operator $N:X\to Z$. It is known (see [17, 18]) that the coincidence equation $Lx=Nx$ is equivalent to
$x=\left(P+JQN\right)x+{K}_{P,Q}Nx.$
Definition 2.2 Let X be a real Banach space. A nonempty closed convex set P is said to be a cone provided that
1. (1)

$\lambda x\in P$ for all $x\in P$ and $\lambda \ge 0$, and

2. (2)

$x,-x\in P$ implies $x=\theta$.

Note that every cone $P\subset X$ induces a partial order ≤ in X by defining $x\le y$ if and only if $y-x\in P$. The following property is valid for every cone in a Banach space.

Lemma 2.3 ([19])

Let P be a cone in X. Then for every$u\in P\mathrm{\setminus }\left\{\theta \right\}$, there exists a positive number$\sigma \left(u\right)$such that$\parallel x+u\parallel \ge \sigma \left(u\right)\parallel x\parallel$for all$x\in P$.

Let $\gamma :X\to P$ be a retraction, that is, a continuous mapping such that $\gamma \left(x\right)=x$ for all $x\in P$. Set
$\mathrm{\Psi }=P+JQN+{K}_{P,Q}N$
and
${\mathrm{\Psi }}_{\gamma }=\mathrm{\Psi }\circ \gamma .$

Our main results are based on the following theorem due to O’Regan and Zima.

Theorem 2.4 ([16])

Let C be a cone in X and let${\mathrm{\Omega }}_{1}$, ${\mathrm{\Omega }}_{2}$be open bounded subsets of X with${\overline{\mathrm{\Omega }}}_{1}\subset {\mathrm{\Omega }}_{2}$and$C\cap \left({\overline{\mathrm{\Omega }}}_{2}\mathrm{\setminus }{\mathrm{\Omega }}_{1}\right)\ne \theta$. Assume that 1is satisfied and if the following assumptions hold:

(H1) $QN:X\to Z$is continuous and bounded, and${K}_{P,Q}:X\to X$is compact on every bounded subset of X;

(H2) $Lx\ne \lambda Nx$for all$x\in C\cap \partial {\mathrm{\Omega }}_{2}\cap domL$and$\lambda \in \left(0,1\right)$;

(H3) γ maps subsets of${\overline{\mathrm{\Omega }}}_{2}$into bounded subsets of C;

(H4) ${d}_{B}\left(\left[I-\left(P+JQN\right)\gamma \right]{|}_{KerL},KerL\cap {\mathrm{\Omega }}_{2},\theta \right)\ne 0$, where${d}_{B}$stands for the Brouwer degree;

(H5) there exists${u}_{0}\in C\mathrm{\setminus }\left\{\theta \right\}$such that$\parallel x\parallel \le \sigma \left({u}_{0}\right)\parallel \mathrm{\Psi }\left(x\right)\parallel$for$x\in C\left({u}_{0}\right)\cap \partial {\mathrm{\Omega }}_{1}$, whereand$\sigma \left({u}_{0}\right)$is such that$\parallel x+{u}_{0}\parallel \ge \sigma \left({u}_{0}\right)\parallel x\parallel$for all$x\in C$;

(H6) $\left(P+JQN\right)\gamma \left(\partial {\mathrm{\Omega }}_{2}\right)\subset C$;

(H7) ${\mathrm{\Psi }}_{\gamma }\left({\overline{\mathrm{\Omega }}}_{2}\mathrm{\setminus }{\mathrm{\Omega }}_{1}\right)\subset C$,

then the equation$Lx=Nx$has a solution in the set$C\cap \left({\overline{\mathrm{\Omega }}}_{2}\mathrm{\setminus }{\mathrm{\Omega }}_{1}\right)$.

## 3 Main results

For simplicity of notation, we set
${h}_{i}\left(s\right)=\left\{\begin{array}{cc}\frac{2{\xi }_{i}s-{\xi }_{i}^{2}s-{s}^{2}}{2},\hfill & 0\le s\le {\xi }_{i},\hfill \\ \frac{{\xi }_{i}^{2}\left(1-s\right)}{2},\hfill & {\xi }_{i}\le s\le 1,\hfill \end{array}$
where $i=1,2,\dots ,m-2$, and
$G\left(t,s\right)=\left\{\begin{array}{cc}\frac{\left(1-s\right)\left({s}^{2}-2s+3{t}^{2}\right)}{6}-\frac{{\left(t-s\right)}^{2}}{2}+\frac{4{t}^{3}-6{t}^{2}+23}{2{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}{\sum }_{i=1}^{m-2}{\alpha }_{i}{h}_{i}\left(s\right),\hfill & 0\le s\le t\le 1,\hfill \\ \frac{\left(1-s\right)\left({s}^{2}-2s+3{t}^{2}\right)}{6}+\frac{4{t}^{3}-6{t}^{2}+23}{2{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}{\sum }_{i=1}^{m-2}{\alpha }_{i}{h}_{i}\left(s\right),\hfill & 0\le t\le s\le 1.\hfill \end{array}$
It is easy to check that $G\left(t,s\right)\ge 0$, $t,s\in \left[0,1\right]$, and since $0\le {h}_{i}\left(s\right)\le \frac{1}{2}s\left(1-s\right)$, we get
$1-\frac{12\delta }{{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}\sum _{i=1}^{m-2}{\alpha }_{i}{h}_{i}\left(s\right)\ge 0,\phantom{\rule{1em}{0ex}}s\in \left[0,1\right]$

for every $\delta \in \left(0,\frac{2{\xi }_{1}^{2}\left(3-2{\xi }_{1}\right)}{3}\right]$. We also let $\delta :=min\left\{\frac{2{\xi }_{1}^{2}\left(3-2{\xi }_{1}\right)}{3},\frac{1}{{max}_{t,s\in \left[0,1\right]}G\left(t,s\right)},1\right\}$.

We can now state our result on the existence of a positive solution for the BVP (1.1).

Theorem 3.1 Assume that $f:\left[0,1\right]×\left[0,+\mathrm{\infty }\right)\to R$ is continuous and
1. (1)

there exists a constant $M\in \left(0,\mathrm{\infty }\right)$ such that $f\left(t,x\right)>-\delta x$ for all $\left(t,x\right)\in \left[0,1\right]×\left[0,M\right]$;

2. (2)
there exist $r\in \left(0,M\right)$, ${t}_{0}\in \left[0,1\right]$, $a\in \left(0,1\right]$, $b\in \left(0,1\right)$ and continuous functions $g:\left[0,1\right]\to \left[0,\mathrm{\infty }\right)$, $h:\left(0,r\right]\to \left[0,\mathrm{\infty }\right)$ such that $f\left(t,x\right)\ge g\left(t\right)h\left(x\right)$ for $\left(t,x\right)\in \left[0,1\right]×\left(0,r\right]$, $\frac{h\left(x\right)}{{x}^{a}}$ is non-increasing on $\left(0,r\right]$ with
$\frac{h\left(r\right)}{{\left(r\right)}^{a}}{\int }_{0}^{1}G\left({t}_{0},s\right)g\left(s\right)\phantom{\rule{0.2em}{0ex}}ds\ge \frac{1-b}{{b}^{a}}.$

3. (3)

${f}_{\mathrm{\infty }}={lim inf}_{x\to +\mathrm{\infty }}{min}_{t\in \left[0,1\right]}\frac{f\left(t,x\right)}{x}<0$.

Then the resonant BVP (1.1) has at least one positive solution on$\left[0,1\right]$.

Proof Consider the Banach spaces $X=Z=C\left[0,1\right]$ with $\parallel x\parallel ={max}_{t\in \left[0,1\right]}|x\left(t\right)|$.

Let $L:domL\subset X\to Z$ and $N:X\to Z$ with
$domL=\left\{x\in X:{x}^{‴}\in C\left[0,1\right],x\left(0\right)=\sum _{i=1}^{m-2}{\alpha }_{i}x\left({\xi }_{i}\right),{x}^{\mathrm{\prime }}\left(0\right)={x}^{\mathrm{\prime }}\left(1\right)=0\right\}$
be given by $\left(Lx\right)\left(t\right)=-{x}^{‴}\left(t\right)$ and $\left(Nx\right)\left(t\right)=f\left(t,x\left(t\right)\right)$ for $t\in \left[0,1\right]$. Then
and
$ImL=\left\{y\in Z:\sum _{i=1}^{m-2}{\alpha }_{i}{\int }_{0}^{1}{h}_{i}\left(s\right)y\left(s\right)\phantom{\rule{0.2em}{0ex}}ds=0\right\}.$
Clearly, $dimKerL=1$ and ImL is closed. It follows from ${Z}_{1}=Z\mathrm{\setminus }ImL$ that
${Z}_{1}=\left\{{y}_{1}\in Z:{y}_{1}=\frac{12}{2{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}\sum _{i=1}^{m-2}{\alpha }_{i}{\int }_{0}^{1}{h}_{i}\left(s\right)y\left(s\right)\phantom{\rule{0.2em}{0ex}}ds,y\in Z\right\}.$
In fact, for each $y\in Z$, we have
$\begin{array}{c}\sum _{i=1}^{m-2}{\alpha }_{i}{\int }_{0}^{1}{h}_{i}\left(s\right)\left(y\left(s\right)-{y}_{1}\left(s\right)\right)\phantom{\rule{0.2em}{0ex}}ds\hfill \\ \phantom{\rule{1em}{0ex}}=\left(1-\frac{12}{{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}\sum _{i=1}^{m-2}{\alpha }_{i}{\int }_{0}^{1}{h}_{i}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds\right)\sum _{i=1}^{m-2}{\alpha }_{i}{\int }_{0}^{1}{h}_{i}\left(s\right)y\left(s\right)\phantom{\rule{0.2em}{0ex}}ds=0,\hfill \end{array}$

which shows that $y-{y}_{1}\in ImL$, which together with ${Z}_{1}\cap ImL=\left\{\theta \right\}$ implies that $Z={Z}_{1}\oplus ImL$. Note that $dim{Z}_{1}=1$ and thus $codimImL=1$. Therefore, L is a Fredholm operator of index zero.

Next, define the projections $P:X\to X$ by
$Px={\int }_{0}^{1}x\left(s\right)\phantom{\rule{0.2em}{0ex}}ds$
and $Q:Z\to Z$ by
$Qy=\frac{12}{{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}\sum _{i=1}^{m-2}{\alpha }_{i}{\int }_{0}^{1}{h}_{i}\left(s\right)y\left(s\right)\phantom{\rule{0.2em}{0ex}}ds.$
Clearly, $ImP=KerL$, $KerQ=ImL$ and $KerL=\left\{x\in X:{\int }_{0}^{1}x\left(s\right)\phantom{\rule{0.2em}{0ex}}ds=0\right\}$. Note that for $y\in ImL$, the inverse ${K}_{P}$ of ${L}_{P}$ is given by
$\left({K}_{P}y\right)\left(t\right)={\int }_{0}^{1}k\left(t,s\right)y\left(s\right)\phantom{\rule{0.2em}{0ex}}ds,$
where
$k\left(t,s\right)=\left\{\begin{array}{cc}\frac{\left(1-s\right)\left({s}^{2}-2s+3{t}^{2}\right)}{6}-\frac{{\left(t-s\right)}^{2}}{2},\hfill & 0\le s\le t\le 1,\hfill \\ \frac{\left(1-s\right)\left({s}^{2}-2s+3{t}^{2}\right)}{6},\hfill & 0\le t\le s\le 1.\hfill \end{array}$

Considering that f can be extended continuously to $\left[0,1\right]×R$, it is easy to check that $QN:X\to Z$ is continuous and bounded, and ${K}_{P,Q}N:X\to X$ is compact on every bounded subset of X, which ensures that (H1) of Theorem 2.4 is fulfilled.

Define the cone of nonnegative functions
Let
${\mathrm{\Omega }}_{1}=\left\{x\in X:b\parallel x\parallel <|x\left(t\right)|
and
${\mathrm{\Omega }}_{2}=\left\{x\in X:\parallel x\parallel
Clearly, ${\mathrm{\Omega }}_{1}$ and ${\mathrm{\Omega }}_{2}$ are bounded and open sets and
${\overline{\mathrm{\Omega }}}_{1}=\left\{x\in X:b\parallel x\parallel \le |x\left(t\right)|\le r,t\in \left[0,1\right]\right\}\subset {\mathrm{\Omega }}_{2}.$

Moreover, $C\cap \left({\overline{\mathrm{\Omega }}}_{2}\mathrm{\setminus }{\mathrm{\Omega }}_{1}\right)\ne \theta$. Let $J=I$ and $\left(\gamma x\right)\left(t\right)=|x\left(t\right)|$ for $x\in X$. Then γ is a retraction and maps subsets of ${\overline{\mathrm{\Omega }}}_{2}$ into bounded subsets of C, which means that (H3) of Theorem 2.4 holds.

Let $0<{r}^{\prime }=min\left\{{r}_{1},{r}_{2}\right\}$, where ${r}_{1}$ and ${r}_{2}$ will be defined in the following proof.

Suppose that there exist ${x}_{0}\in C\cap \partial {\mathrm{\Omega }}_{2}\cap domL$ and ${\lambda }_{0}\in \left(0,1\right)$ such that $L{x}_{0}={\lambda }_{0}N{x}_{0}$. Then
${x}_{0}^{‴}\left(t\right)+{\lambda }_{0}f\left(t,{x}_{0}\left(t\right)\right)=0.$

Let ${x}_{0}\left({t}_{1}\right)=\parallel {x}_{0}\parallel ={max}_{t\in \left[0,1\right]}|{x}_{0}\left(t\right)|=M$. Now, we verify that ${t}_{1}\ne 0$ and ${t}_{1}\ne 1$.

First, we show ${t}_{1}\ne 0$. Suppose, on the contrary, that ${x}_{0}\left(t\right)$ achieves maximum value M only at $t=0$. Then ${x}_{0}\left(0\right)={\sum }_{i=1}^{m-2}{\alpha }_{i}{x}_{0}\left({\xi }_{i}\right)$ in combination with ${\sum }_{i=1}^{m-2}{\alpha }_{i}=1$ yields that ${max}_{1\le i\le m-2}\left\{{x}_{0}\left({\xi }_{i}\right)\right\}\ge M$, which is a contradiction.

Next, we show ${t}_{1}\ne 1$. It follows from ${x}_{0}^{\prime }\left(0\right)={x}_{0}^{\prime }\left(1\right)=0$ that there is a constant $\eta \in \left(0,1\right)$ such that ${x}_{0}^{″}\left(\eta \right)=0$, and thus ${x}_{0}^{″}\left(t\right)=-{\lambda }_{0}{\int }_{\eta }^{t}f\left(s,{x}_{0}\left(s\right)\right)\phantom{\rule{0.2em}{0ex}}ds$. By the condition $\left(3\right)$ we have, for $x\ge {r}^{\prime }$ ($M\ge {r}^{\prime }>r>0$), there exists $\sigma \in \left(0,\delta \right)$ such that
$f\left(t,x\right)\le -\sigma x,\phantom{\rule{1em}{0ex}}\mathrm{\forall }t\in \left[0,1\right].$
(3.1)

Suppose, on the contrary, that ${x}_{0}\left(1\right)=M$. The step is divided into two cases:

Case 1. Assume that ${x}_{0}\left(t\right)>0$ on $\left[\eta ,1\right]$. Let ${r}_{1}={min}_{t\in \left[\eta ,1\right]}{x}_{0}\left(t\right)$. Then (3.1) yields
${x}_{0}^{″}\left(1\right)=-{\lambda }_{0}{\int }_{\eta }^{1}f\left(s,{x}_{0}\left(s\right)\right)\phantom{\rule{0.2em}{0ex}}ds\ge {\lambda }_{0}\sigma {\int }_{\eta }^{1}{x}_{0}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds>0,$

which implies ${x}_{0}^{\prime }\left(t\right)$ is increasing close to 1. This together with ${x}_{0}^{\prime }\left(1\right)=0$ induces ${x}_{0}^{\prime }\left(t\right)<0$ (t close to 1), that is, ${x}_{0}\left(t\right)$ is decreasing close to 1, which contradicts ${x}_{0}\left(1\right)=M$.

Case 2. Assume that ${x}_{0}\left(t\right)$ has zero points on $\left[\eta ,1\right]$; we may choose ${\eta }^{\prime }$ nearest to 1 with ${x}_{0}\left({\eta }^{\prime }\right)=0$. Then there is a constant $\eta \in \left({\eta }^{\prime },1\right)$ such that ${x}_{0}^{″}\left(\eta \right)=0$. Similar to the above arguments, we easily get a contradiction too.

Hence, we can choose ${t}_{1}\in \left(0,1\right)$ so that ${x}_{0}\left({t}_{1}\right)=M$. This gives ${x}_{0}^{\prime }\left({t}_{1}\right)=0$ and ${x}_{0}^{″}\left({t}_{1}\right)\le 0$. By ${x}_{0}\in domL$, we know ${x}_{0}\left(t\right)\ge 0$ and ${x}_{0}^{\prime }\left(0\right)=0$. Similarly, we also divide the part of the proof into two cases.

Case 1. If ${x}_{0}\left(t\right)>0$ on $\left[0,{t}_{1}\right]$, then there is a constant $\eta \in \left(0,{t}_{1}\right)$ such that ${x}_{0}^{″}\left(\eta \right)=0$. Thus we have
${x}_{0}^{″}\left(t\right)=-{\lambda }_{0}{\int }_{\eta }^{t}f\left(s,{x}_{0}\left(s\right)\right)\phantom{\rule{0.2em}{0ex}}ds.$
Let ${r}_{2}={min}_{t\in \left[\eta ,{t}_{1}\right]}{x}_{0}\left(t\right)$. Then it follows from (3.1) that
$0\ge {x}_{0}^{″}\left({t}_{1}\right)=-{\lambda }_{0}{\int }_{\eta }^{{t}_{1}}f\left(s,{x}_{0}\left(s\right)\right)\phantom{\rule{0.2em}{0ex}}ds\ge {\lambda }_{0}\sigma {\int }_{\eta }^{{t}_{1}}{x}_{0}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds>0,$

which is a contradiction.

Case 2. If ${x}_{0}\left(t\right)$ has zero points on $\left[0,{t}_{1}\right]$, we may choose ${\eta }^{\prime }$ nearest to ${t}_{1}$ with ${x}_{0}\left({\eta }^{\prime }\right)=0$. Then there is a constant $\eta \in \left({\eta }^{\prime },{t}_{1}\right)$ such that ${x}_{0}^{″}\left(\eta \right)=0$. Thus we have
${x}_{0}^{″}\left(t\right)=-{\lambda }_{0}{\int }_{\eta }^{t}f\left(s,{x}_{0}\left(s\right)\right)\phantom{\rule{0.2em}{0ex}}ds.$
Let ${r}_{2}={min}_{t\in \left[\eta ,{t}_{1}\right]}{x}_{0}\left(t\right)$. Then it follows from (3.1) that
$0\ge {x}_{0}^{″}\left({t}_{1}\right)=-{\lambda }_{0}{\int }_{\eta }^{{t}_{1}}f\left(s,{x}_{0}\left(s\right)\right)\phantom{\rule{0.2em}{0ex}}ds\ge {\lambda }_{0}\sigma {\int }_{\eta }^{{t}_{1}}{x}_{0}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds>0,$

which is a contradiction. Therefore, (H2) of Theorem 2.4 holds.

Consider $x\in KerL\cap {\mathrm{\Omega }}_{2}$. Then $x\left(t\right)\equiv c$ on $\left[0,1\right]$. Similar to [13], we define
Suppose $H\left(x,\lambda \right)=0$. Then in view of (1), we obtain
$\begin{array}{rcl}c& =& \lambda c+\frac{12\lambda }{{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}\sum _{i=1}^{m-2}{\alpha }_{i}{\int }_{0}^{1}{h}_{i}\left(s\right)f\left(s,|c|\right)\phantom{\rule{0.2em}{0ex}}ds\\ \ge & \lambda c-\frac{12\lambda }{{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}\sum _{i=1}^{m-2}{\alpha }_{i}{\int }_{0}^{1}{h}_{i}\left(s\right)\delta |c|\phantom{\rule{0.2em}{0ex}}ds\ge \lambda |c|\left(1-\delta \right)\ge 0.\end{array}$
Hence, $H\left(x,\lambda \right)=0$ implies $c\ge 0$. Furthermore, if $H\left(M,\lambda \right)=0$, then we have
$0\le M\left(1-\lambda \right)=\frac{12\lambda }{{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}\sum _{i=1}^{m-2}{\alpha }_{i}{\int }_{0}^{1}{h}_{i}\left(s\right)f\left(s,M\right)\phantom{\rule{0.2em}{0ex}}ds,$
contradicting (3.1). Thus $H\left(x,\lambda \right)=0$ for $x\in \partial {\mathrm{\Omega }}_{2}$ and $\lambda \in \left[0,1\right]$. Therefore,
${d}_{B}\left(H\left(x,0\right),KerL\cap {\mathrm{\Omega }}_{2},\theta \right)={d}_{B}\left(H\left(x,1\right),KerL\cap {\mathrm{\Omega }}_{2},\theta \right).$
However,
${d}_{B}\left(H\left(x,0\right),KerL\cap {\mathrm{\Omega }}_{2},\theta \right)={d}_{B}\left(I,KerL\cap {\mathrm{\Omega }}_{2},\theta \right).$
This gives
${d}_{B}\left(\left[I-\left(P+JQN\right)\gamma \right]{|}_{KerL},KerL\cap {\mathrm{\Omega }}_{2},\theta \right)={d}_{B}\left(H\left(x,1\right),KerL\cap {\mathrm{\Omega }}_{2},\theta \right)\ne 0,$

which shows that (H4) of Theorem 2.4 holds.

Let $x\in {\overline{\mathrm{\Omega }}}_{2}\mathrm{\setminus }{\mathrm{\Omega }}_{1}$ and $t\in \left[0,1\right]$. Then
From (1), we know that
$\begin{array}{rcl}\left({\mathrm{\Psi }}_{\gamma }\right)\left(t\right)& \ge & {\int }_{0}^{1}|x\left(s\right)|\phantom{\rule{0.2em}{0ex}}ds-\delta {\int }_{0}^{1}G\left(t,s\right)|x\left(s\right)|\phantom{\rule{0.2em}{0ex}}ds\\ =& {\int }_{0}^{1}\left(1-\delta G\left(t,s\right)\right)|x\left(s\right)|\phantom{\rule{0.2em}{0ex}}ds\ge 0.\end{array}$
Hence ${\mathrm{\Psi }}_{\gamma }\left(\overline{{\mathrm{\Omega }}_{2}}\mathrm{\setminus }{\mathrm{\Omega }}_{1}\right)\subset C$. Moreover, for $x\in \partial {\mathrm{\Omega }}_{2}$, we have
$\begin{array}{rcl}\left(P+JQN\right)\gamma x& =& {\int }_{0}^{1}|x\left(s\right)|\phantom{\rule{0.2em}{0ex}}ds+\frac{12}{{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}\sum _{i=1}^{m-2}{\alpha }_{i}{\int }_{0}^{1}{h}_{i}\left(s\right)f\left(s,|x\left(s\right)|\right)\phantom{\rule{0.2em}{0ex}}ds\\ \ge & {\int }_{0}^{1}\left[1-\frac{12\delta }{2{\sum }_{i=1}^{m-2}{\alpha }_{i}\left(3{\xi }_{i}^{2}-2{\xi }_{i}^{3}\right)}\sum _{i=1}^{m-2}{\alpha }_{i}{h}_{i}\left(s\right)\right]|x\left(s\right)|\phantom{\rule{0.2em}{0ex}}ds\ge 0,\end{array}$

which shows that $\left(P+JQN\right)\gamma \left(\partial {\mathrm{\Omega }}_{2}\right)\subset P$. These ensure that (H6), (H7) of Theorem 2.4 hold. It remains to show that (H5) is satisfied.

Taking ${u}_{0}\left(t\right)\equiv 1$ on $\left[0,1\right]$ and $\sigma \left({u}_{0}\right)=1$, we confirm that
Let $x\in C\left({u}_{0}\right)\cap \partial {\mathrm{\Omega }}_{1}$. Then we have $x\left(t\right)>0$, $t\in \left[0,1\right]$, $0<\parallel x\parallel and $x\left(t\right)\ge b\parallel x\parallel$, $t\in \left[0,1\right]$. Therefore, in view of (2), for all $x\in C\left({u}_{0}\right)\cap \partial {\mathrm{\Omega }}_{1}$, we obtain
$\begin{array}{rcl}\left(\mathrm{\Psi }x\right)\left({t}_{0}\right)& =& {\int }_{0}^{1}x\left(s\right)\phantom{\rule{0.2em}{0ex}}ds+{\int }_{0}^{1}G\left({t}_{0},s\right)f\left(s,x\left(s\right)\right)\phantom{\rule{0.2em}{0ex}}ds\\ \ge & b\parallel x\parallel +{\int }_{0}^{1}G\left({t}_{0},s\right)g\left(s\right)h\left(x\left(s\right)\right)\phantom{\rule{0.2em}{0ex}}ds\\ =& b\parallel x\parallel +{\int }_{0}^{1}G\left({t}_{0},s\right)g\left(s\right)\frac{h\left(x\left(s\right)\right)}{{x}^{a}\left(s\right)}{x}^{a}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds\\ \ge & b\parallel x\parallel +\frac{h\left(r\right)}{{\left(r\right)}^{a}}{\int }_{0}^{1}G\left({t}_{0},s\right)g\left(s\right){x}^{a}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds\\ \ge & b\parallel x\parallel +\frac{h\left(r\right)}{{\left(r\right)}^{a}}{\int }_{0}^{1}G\left({t}_{0},s\right)g\left(s\right){b}^{a}{\parallel x\parallel }^{a}\phantom{\rule{0.2em}{0ex}}ds\\ \ge & b\parallel x\parallel +\left(1-b\right)\parallel x\parallel =\parallel x\parallel .\end{array}$

That is, $\parallel x\parallel \le \sigma \left({u}_{0}\right)\parallel \mathrm{\Psi }x\parallel$ for all $x\in C\left({u}_{0}\right)\cap \partial {\mathrm{\Omega }}_{1}$, which shows that (H5) of Theorem 2.4 holds.

Summing up, all the hypotheses of Theorem 2.4 are satisfied. Therefore, the equation $Lx=Nx$ has a solution $x\in C\cap \left({\overline{\mathrm{\Omega }}}_{2}\mathrm{\setminus }{\mathrm{\Omega }}_{1}\right)$. And so, the resonant BVP (1.1) has at least one positive solution on $\left[0,1\right]$. □

## 4 An example

Consider the BVP
$\left\{\begin{array}{c}{x}^{\mathrm{\prime }\mathrm{\prime }\mathrm{\prime }}\left(t\right)+f\left(t,x\left(t\right)\right)=0,\phantom{\rule{1em}{0ex}}0
(4.1)
Here ${\alpha }_{1}=1$, ${\xi }_{1}=\frac{1}{2}$, $g\left(t\right)=-\frac{1}{4}\left({t}^{2}-t-1\right)$, $h\left(x\right)=\sqrt{x+1}$ and
$f\left(t,x\right)=\left\{\begin{array}{cc}-\frac{1}{4}\left({t}^{2}-t-1\right)\left({x}^{2}-8x+15\right)\sqrt{x+1},\hfill & t\in \left[0,1\right],x\in \left[0,2\right],\hfill \\ -\frac{1}{4}\left({t}^{2}-t-1\right)\left(2-\frac{1}{2}\sqrt{x-1}\right)\sqrt{x+1},\hfill & t\in \left[0,1\right],x\in \left(2,+\mathrm{\infty }\right).\hfill \end{array}$
By a simple computation, we get
$\delta =\frac{1}{3},\phantom{\rule{2em}{0ex}}\frac{1}{4}\le g\left(t\right)\le \frac{5}{16},\phantom{\rule{1em}{0ex}}t\in \left[0,1\right],\phantom{\rule{2em}{0ex}}{x}^{2}-8x+15\ge -x,\phantom{\rule{1em}{0ex}}x\in \left[0,3\right].$
We may choose $M=3$, $r=\frac{5}{2}$, ${t}_{0}=0$, $a=1$, $b=\frac{13}{14}$. It is easy to check
1. (1)

$f\left(t,x\right)>-\frac{1}{3}x$ for all $\left(t,x\right)\in \left[0,1\right]×\left[0,3\right]$;

2. (2)
$f\left(t,x\right)\ge g\left(t\right)h\left(x\right)$ for $\left(t,x\right)\in \left[0,1\right]×\left(0,\frac{5}{2}\right]$, $\frac{\sqrt{x+1}}{x}$ is non-increasing on $\left(0,\frac{5}{2}\right]$ with
$\frac{h\left(r\right)}{{\left(r\right)}^{a}}{\int }_{0}^{1}G\left(0,s\right)g\left(s\right)\phantom{\rule{0.2em}{0ex}}ds\ge \frac{7\sqrt{14}}{320}\ge \frac{1}{13}=\frac{1-b}{{b}^{a}};$

3. (3)

${f}_{\mathrm{\infty }}={lim inf}_{x\to +\mathrm{\infty }}{min}_{t\in \left[0,1\right]}\frac{f\left(t,x\right)}{x}=-\frac{1}{8}$.

Thus, all the conditions of Theorem 3.1 are satisfied. Then the resonant problem (4.1) has at least one positive solution on $\left[0,1\right]$.

## Declarations

### Acknowledgements

This work is supported by the National Natural Science Foundation of China (10801068) and the Education Scientific Research Foundation of Tangshan College (120186). The authors would like to thank the anonymous referees very much for helpful comments and suggestions which led to the improvement of presentation and quality of the work.

## Authors’ Affiliations

(1)
Department of Basic Teaching, Tangshan College
(2)
Department of Applied Mathematics, Lanzhou University of Technology

## References

1. Gregus M Math. Appl. In Third Order Linear Differential Equations. Reidel, Dordrecht; 1987.
2. Anderson DR: Green’s function for a third-order generalized right focal problem. J. Math. Anal. Appl. 2003, 288: 1-14. 10.1016/S0022-247X(03)00132-X
3. Anderson DR, Davis JM: Multiple solutions and eigenvalues for three-order right focal boundary value problems. J. Math. Anal. Appl. 2002, 267: 135-157. 10.1006/jmaa.2001.7756
4. Du ZJ, Ge WG, Zhou MR: Singular perturbations for third-order nonlinear multi-point boundary value problem. J. Differ. Equ. 2005, 218: 69-90. 10.1016/j.jde.2005.01.005
5. El-Shahed M: Positive solutions for nonlinear singular third order boundary value problems. Commun. Nonlinear Sci. Numer. Simul. 2009, 14: 424-429. 10.1016/j.cnsns.2007.10.008
6. Li S: Positive solutions of nonlinear singular third-order two-point boundary value problem. J. Math. Anal. Appl. 2006, 323: 413-425. 10.1016/j.jmaa.2005.10.037
7. Liu ZQ, Debnath L, Kang SM: Existence of monotone positive solutions to a third-order two-point generalized right focal boundary value problem. Comput. Math. Appl. 2008, 55: 356-367. 10.1016/j.camwa.2007.03.021
8. Palamides AP, Smyrlis G: Positive solutions to a singular third-order three-point boundary value problem with an indefinitely signed Green’s function. Nonlinear Anal. 2008, 68: 2104-2118. 10.1016/j.na.2007.01.045
9. Sun JP, Zhang HE: Existence of solutions to third-order m -point boundary value problems. Electron. J. Differ. Equ. 2008., 2008: Article ID 125Google Scholar
10. Sun Y: Positive solutions for third-order three-point nonhomogeneous boundary value problems. Appl. Math. Lett. 2009, 22: 45-51. 10.1016/j.aml.2008.02.002
11. Yao Q: The existence and multiplicity of positive solutions for a third-order three-point boundary value problem. Acta Math. Appl. Sin. 2003, 19: 117-122. 10.1007/s10255-003-0087-1
12. Bai C, Fang J: Existence of positive solutions for boundary value problems at resonance. J. Math. Anal. Appl. 2004, 291: 538-549. 10.1016/j.jmaa.2003.11.014
13. Infante G, Zima M: Positive solutions of multi-point boundary value problems at resonance. Nonlinear Anal. 2008, 69: 2458-2465. 10.1016/j.na.2007.08.024
14. Liang SQ, Mu L: Multiplicity of positive solutions for singular three-point boundary value problem at resonance. Nonlinear Anal. 2009, 71: 2497-2505. 10.1016/j.na.2009.01.085
15. Yang L, Shen CF: On the existence of positive solution for a kind of multi-order boundary value problem at resonance. Nonlinear Anal. 2010, 72: 4211-4220. 10.1016/j.na.2010.01.051
16. O’Regan D, Zima M: Leggett-Williams norm-type theorems for coincidence. Arch. Math. 2006, 87: 233-244. 10.1007/s00013-006-1661-6
17. Mawhin J: Equivalence theorems for nonlinear operator equations and coincidence degree theory for some mappings in locally convex topological vector spaces. J. Differ. Equ. 2010, 72: 4211-4220.Google Scholar
18. Santanilla J: Some coincidence theorems in wedges, cones and convex sets. J. Math. Anal. Appl. 1985, 105: 357-371. 10.1016/0022-247X(85)90053-8
19. Petryshyn WV: On the solvability of$x\in Tx+\lambda Fx$ in quasinormal cones with T and F k -set contractive. Nonlinear Anal. 1981, 5: 585-591. 10.1016/0362-546X(81)90105-X