 Research
 Open access
 Published:
Generalized monotone iterative method for nonlinear boundary value problems with causal operators
Boundary Value Problems volume 2014, Article number: 192 (2014)
Abstract
This paper discusses nonlinear boundary value problems for causal differential equations where the righthand side is the sum of two monotone functions. We develop the monotone iterative technique and establish the existence results of the extremal solutions. The results obtained include several special cases and extend previous results; two examples satisfying the assumptions are also presented.
MSC: 34B15, 39B12.
1 Introduction
As we all know, the monotone iterative technique is an effective and a flexible method, and it provides a useful mechanism to prove existence results for nonlinear differential equations, for detail see for monographs [1], papers [2]–[11], and the references therein. The basic idea of this method is that using the upper and lower solutions as an initial iteration, one can construct monotone sequences from a corresponding linear equations, and these sequences converge monotonically to the minimal and maximal solutions of the nonlinear equations.
In 2004, West [12] developed this method, considered the generalized monotone iterative method for initial value problems, obtained the existence of extremal solutions for differential equations where the forcing function is the sum of two monotone functions, one of which is monotone nondecreasing and the other is nonincreasing.
Recently, this method has been extended to causal differential equations. Its theory has the powerful quality of unifying ordinary differential equations, integro differential equations, differential equations with finite or infinite delay, Volterra integral equations and neutral equations. We refer to the monograph by Lakshmikantham [13] and papers [14], [15].
In 2009, Lakshmikantham discussed in [13] the following problem with causal operators: {x}^{\prime}(t)=(Px)(t)+(Qx)(t), x(0)={x}_{0}, where t\in J=[0,T], P,Q:E\to E=C(J,\mathbb{R}) are causal operators. However, we notice that the results are only valid for initial value problems. Motivated by the above excellent work, we extend the notion of casual operators to nonlinear boundary value problems and develop the monotone iterative technique.
In this paper, we deal with the following causal differential equation:
where Q,S:E\to E=C(J,\mathbb{R}) are causal operators.
Note that the nonlinear boundary value problem (1.1) reduce to periodic boundary value problems for g(u(0),u(T))=u(0)u(T), initial value problems for g(u(0),u(T))=u(0){u}_{0} which has been studied in [13] and other general conditions such as g(u(0),u(T))=h(u(0))u(T). Thus problem (1.1) can be regarded as a generalization of the boundary value problems mentioned above.
The rest of this paper is organized as follows. In Section 2, we develop the monotone technique for (1.1); four theorems and several special cases are given. In Section 3, we give two examples to illustrate the results obtained. Finally, a brief summary is given in Section 4.
2 Main results
Let J=[0,T], Q,S:E\to E=C(J,\mathbb{R}). In order to prove general results, we need the following definitions.
Definition 2.1
For (1.1), the functions {y}_{0},{z}_{0}\in {C}^{1}[J,\mathbb{R}] are said to be

(1)
natural lower and upper solutions if
\{\begin{array}{c}{y}_{0}^{\prime}(t)\le (Q{y}_{0})(t)+(S{y}_{0})(t),\phantom{\rule{2em}{0ex}}g({y}_{0}(0),{y}_{0}(T))\le 0,\hfill \\ {z}_{0}^{\prime}(t)\ge (Q{z}_{0})(t)+(S{z}_{0})(t),\phantom{\rule{2em}{0ex}}g({z}_{0}(0),{z}_{0}(T))\ge 0;\hfill \end{array}

(2)
coupled lower and upper solutions of type I, if
\{\begin{array}{c}{y}_{0}^{\prime}(t)\le (Q{y}_{0})(t)+(S{z}_{0})(t),\phantom{\rule{2em}{0ex}}g({y}_{0}(0),{y}_{0}(T))\le 0,\hfill \\ {z}_{0}^{\prime}(t)\ge (Q{z}_{0})(t)+(S{y}_{0})(t),\phantom{\rule{2em}{0ex}}g({z}_{0}(0),{z}_{0}(T))\ge 0;\hfill \end{array}

(3)
coupled lower and upper solutions of type II, if
\{\begin{array}{c}{y}_{0}^{\prime}(t)\le (Q{z}_{0})(t)+(S{y}_{0})(t),\phantom{\rule{2em}{0ex}}g({y}_{0}(0),{y}_{0}(T))\le 0,\hfill \\ {z}_{0}^{\prime}(t)\ge (Q{y}_{0})(t)+(S{z}_{0})(t),\phantom{\rule{2em}{0ex}}g({z}_{0}(0),{z}_{0}(T))\ge 0;\hfill \end{array}

(4)
coupled lower and upper solutions of type III, if
\{\begin{array}{c}{y}_{0}^{\prime}(t)\le (Q{z}_{0})(t)+(S{z}_{0})(t),\phantom{\rule{2em}{0ex}}g({y}_{0}(0),{y}_{0}(T))\le 0,\hfill \\ {z}_{0}^{\prime}(t)\ge (Q{y}_{0})(t)+(S{y}_{0})(t),\phantom{\rule{2em}{0ex}}g({z}_{0}(0),{z}_{0}(T))\ge 0.\hfill \end{array}
If we set {y}_{0}(t)\le {z}_{0}(t), t\in J, (Qu) is nondecreasing and (Su) is nonincreasing, then the natural lower and upper solutions and the coupled lower and upper solutions of type III satisfy type II. Thus, we only need to consider the case of the coupled lower and upper solutions of type I and II for (1.1).
Definition 2.2
Relative to the causal differential equation (1.1):

(1)
A function U\in {C}^{1}[J,\mathbb{R}] is said to be a natural solution if it satisfies (1.1).

(2)
U,V\in {C}^{1}[J,\mathbb{R}] are said to be coupled solutions of type I, if
\{\begin{array}{c}{U}^{\prime}(t)=(QU)(t)+(SV)(t),\phantom{\rule{2em}{0ex}}g(U(0),U(T))=0,\hfill \\ {V}^{\prime}(t)=(QV)(t)+(SU)(t),\phantom{\rule{2em}{0ex}}g(V(0),V(T))=0.\hfill \end{array}

(3)
U, V are said to be coupled solutions of type II, if
\{\begin{array}{c}{U}^{\prime}(t)=(QV)(t)+(SU)(t),\phantom{\rule{2em}{0ex}}g(U(0),U(T))=0,\hfill \\ {V}^{\prime}(t)=(QU)(t)+(SV)(t),\phantom{\rule{2em}{0ex}}g(V(0),V(T))=0.\hfill \end{array}

(4)
U, V are said to be coupled solutions of type III, if
\{\begin{array}{c}{U}^{\prime}(t)=(QV)(t)+(SV)(t),\phantom{\rule{2em}{0ex}}g(U(0),U(T))=0,\hfill \\ {V}^{\prime}(t)=(QU)(t)+(SU)(t),\phantom{\rule{2em}{0ex}}g(V(0),V(T))=0.\hfill \end{array}
Definition 2.3
Coupled solutions \rho ,r\in {C}^{1}[J,\mathbb{R}], are said to be couple minimal and maximal solutions of (1.1), if for any coupled solutions U, V, we have \rho \le U,V\le r.
Theorem 2.1
We suppose that the following hypotheses hold:
H_{1}: {y}_{0},{z}_{0}\in {C}^{1}(J,\mathbb{R})are the coupled lower and upper solutions of type I for (1.1) with{y}_{0}(t)\le {z}_{0}(t)on J;
H_{2}: the operators Q, S in (1.1) are such thatQ,S:E\to E, (Qu)is nondecreasing in u and(Su)is nonincreasing in u;
H_{3}: the functiong(u,v)\in C({\mathbb{R}}^{2},\mathbb{R})is nonincreasing in the second variable and there exists a constantM>0such that
for{y}_{0}(0)\le {u}_{2}\le {u}_{1}\le {z}_{0}(0), {y}_{0}(T)\le v\le {z}_{0}(T).
Then there exist two monotone sequences \{{y}_{n}(t)\}, \{{z}_{n}(t)\} such that {lim}_{n\to \mathrm{\infty}}{y}_{n}(t)=\rho (t), {lim}_{n\to \mathrm{\infty}}{z}_{n}(t)=r(t) uniformly and monotonically on J and that ρ, r are coupled minimal and maximal solutions of type I for (1.1). Furthermore, if u is any natural solution of (1.1) such that {y}_{0}\le u\le {z}_{0} on J, then \rho \le u\le r on J.
Proof
We consider the following linear problem:
This is an adequate definition since by general results on the initial value problem of causal differential equations [13] the existence and uniqueness of solution for (2.1) and (2.2) are guaranteed.
First, we show that {y}_{0}\le {y}_{1}\le {z}_{1}\le {z}_{0}, putting n=0 in (2.1) and setting p={y}_{0}{y}_{1}, we acquire
It follows that p(t)\le 0 on J, which implies {y}_{0}\le {y}_{1} on J. Similarly, we may obtain {z}_{1}\le {z}_{0} on J.
Next, take p={y}_{1}{z}_{1}, then from the hypotheses H_{2}, H_{3}, and the fact {y}_{0}\le {z}_{0}, one attains
This implies that p(t)\le 0 on J, and {y}_{1}\le {z}_{1}.
In the following, we shall show that {y}_{1}, {z}_{1} are the coupled lower and upper solutions of type I for (1.1). Following H_{2} and {y}_{0}\le {y}_{1}, {z}_{1}\le {z}_{0}, we obtain
and by means of the facts that {y}_{0}\le {y}_{1}, {z}_{1}\le {z}_{0}, H_{1}, and H_{3}, we have
Similarly, we can get
from (2.3)(2.5), we show that {y}_{1}, {z}_{1} are coupled lower and upper solutions of type I for (1.1).
Now employing the mathematical induction, assume that, for some integer k>1,
We need to show that
For this purpose, let p={y}_{k}{y}_{k+1} and use H_{2}, H_{3}; we note that
This implies {y}_{k}\le {y}_{k+1} on J. Similarly, we can prove that {z}_{k+1}\le {z}_{k} on J by using (2.2) and H_{2}, H_{3}. To prove {y}_{k+1}\le {z}_{k+1}, set p={y}_{k+1}{z}_{k+1}, then by using H_{2}, H_{3}, and the fact that {y}_{k}\le {z}_{k}, we get
That yields p(t)\le 0, i.e.{y}_{k+1}\le {z}_{k+1}. From the above discussion, we have
Obviously, the constructed sequences \{{y}_{n}\}, \{{z}_{n}\} are equicontinuous and uniform bounded. Thus, by the AscoliArzela theorem, we have \{{y}_{n}\}\to \rho, \{{z}_{n}\}\to r on J. Since the sequences \{{y}_{n}\}, \{{z}_{n}\} are monotone, the entire sequences converge uniformly and monotonically to ρ, r on J, respectively.
Using the definition of (2.1), (2.2), and passing to the limit when n\to \mathrm{\infty}, we obtain the result that ρ, r are coupled solutions of type I for (1.1).
It remains to show that ρ, r are coupled minimal and maximal solutions of type I for (1.1). Let {u}_{1},{u}_{2}\in [{y}_{0},{z}_{0}] be any coupled solutions of type I for (1.1). Assume that there exists a positive integer k such that {y}_{k}\le {u}_{1},{u}_{2}\le {z}_{k} on J. Then, putting p={y}_{k+1}{u}_{1}, and employing H_{2} and H_{3}, we arrive at
That implies p(t)\le 0, which proves {y}_{k+1}\le {u}_{1} on J. Using similar arguments we can conclude {y}_{k+1}\le {u}_{1},{u}_{2}\le {z}_{k+1} on J. Since {y}_{0}\le {u}_{1},{u}_{2}\le {z}_{0}, by the principle of induction, {y}_{n}\le {u}_{1},{u}_{2}\le {z}_{n} holds for all n. Taking the limit as n\to \mathrm{\infty}, we have \rho \le {u}_{1},{u}_{2}\le r on J proving ρ, r are coupled minimal and maximal solutions of type I for (1.1). Since any natural solution u of (1.1) can be considered as (u,u) coupled solutions of type I, we also have \rho \le u\le r on J. This completes the proof. □
Theorem 2.2
Let the hypotheses of Theorem 2.1hold. Then, for any natural solution u of (1.1) with{y}_{0}\le u\le {z}_{0}, there exist alternating sequences\{{y}_{2n},{z}_{2n+1}\}\to \rho, \{{z}_{2n},{y}_{2n+1}\}\to runiformly on J with{y}_{0}\le {z}_{1}\le \cdots \le {y}_{2n}\le {z}_{2n+1}\le u\le {y}_{2n+1}\le {z}_{2n}\le \cdots \le {y}_{1}\le {z}_{0}. Here ρ, r are the coupled minimal and maximal solutions of type I for (1.1). Also, \rho \le u\le ron J.
Proof
Consider the following initial problem:
First we show that {y}_{0}\le {y}_{1} and {z}_{1}\le {z}_{0}.
Set p={y}_{0}{y}_{1}, following (2.6) and the hypotheses of Theorem 2.1, we obtain
Thus we prove p(t)\le 0, which gives {y}_{0}\le {y}_{1} on J. Similarly, we can show {z}_{1}\le {z}_{0}.
Now we wish to prove that
Setting p=u{y}_{1}, we get
using the monotone nature of the operators Q, S, and the fact {y}_{0}\le u\le {z}_{0} on J, u being any natural solution of (1.1). Also, by using H_{3}, we have
Hence p(t)\le 0 on J, i.e., u\le {y}_{1} on J. A similar argument yields {z}_{1}\le u. In order to avoid repetition, we can prove each of the following: {y}_{2}\le u, u\le {z}_{2}, u\le {y}_{3}, and {z}_{3}\le u.
Now we shall show that {y}_{0}\le {z}_{1}\le {y}_{2}\le {z}_{3} and {y}_{3}\le {z}_{2}\le {y}_{1}\le {z}_{0}. Take p={y}_{0}{z}_{1}, and one attains
Then p(t)\le 0 on J, thus {y}_{0}\le {z}_{1}. Similarly, we can obtain {y}_{1}\le {z}_{0} on J.
Next, take p={z}_{1}{y}_{2}, by H_{2} and H_{3}, we derive
This implies that p(t)\le 0 on J, i.e., {z}_{1}\le {y}_{2} on J. Using similar arguments we can show {y}_{2}\le {z}_{3}, {y}_{3}\le {z}_{2}, {z}_{2}\le {y}_{1}. Combining all these arguments, we now have the desired relations (2.7).
Now, suppose there exists an integer k\ge 2 such that {z}_{2k1}\le {y}_{2k}\le {z}_{2k+1}\le u\le {y}_{2k+1}\le {z}_{2k}\le {y}_{2k1} on J. Thus, we need to show
Setting p={z}_{2k+1}{y}_{2k+2}, then for t\in J and utilizing the hypotheses {y}_{2k}\le {z}_{2k+1} and H_{2}, H_{3}, we may get
This implies that p\le 0 and {z}_{2k+1}\le {y}_{2k+2}. Similarly, we obtain {z}_{2k+2}\le {y}_{2k+1}, {y}_{2k+2}\le {z}_{2k+3}, and {y}_{2k+3}\le {z}_{2k+2}.
To prove {y}_{2k+2}\le u, by using H_{2}, H_{3}, and the inequalities {z}_{2k+1}\le u, consider the relations
and
one attains p\le 0 on J, i.e.{y}_{2k+2}\le u. Hence, as before, we can conclude that u\le {z}_{2k+2}, {z}_{2k+3}\le u, and u\le {y}_{2k+3} on J. Now, with the principle of induction, we have
By employing a reasoning similar to that of Theorem 2.1, we get the sequences \{{y}_{2n},{z}_{2n+1}\}, \{{z}_{2n},{y}_{2n+1}\} which converge uniformly and monotonically to ρ, r on J, respectively. Thus, ρ, r are coupled solutions of type I for (1.1).
Finally, to prove that ρ, r are coupled minimal and maximal solutions of (1.1), let {u}_{1},{u}_{2}\in [{y}_{0},{z}_{0}] be any coupled solutions of type I for (1.1). Similar to the proof of the above, if {z}_{2k1}\le {y}_{2k}\le {z}_{2k+1}\le {u}_{1},{u}_{2}\le {y}_{2k+1}\le {z}_{2k}\le {y}_{2k1} for some positive integer k, we can easily see that {z}_{2k+1}\le {y}_{2k+2}\le {z}_{2k+3}\le {u}_{1},{u}_{2}\le {y}_{2k+3}\le {z}_{2k+2}\le {y}_{2k+1} on J, by the induction, one has {y}_{2n}\le {z}_{2n+1}\le {u}_{1},{u}_{2}\le {y}_{2n+1}\le {z}_{2n} holds on J for all n. Taking the limit as n\to \mathrm{\infty}, we have \rho \le {u}_{1},{u}_{2}\le r on J proving ρ, r are coupled minimal and maximal solutions of type I for (1.1). Since we have already shown that {y}_{2n}\le {z}_{2n+1}\le u\le {y}_{2n+1}\le {z}_{2n} holds on J for all n. Now taking the limit as n\to \mathrm{\infty}, we get \rho \le u\le r on J. This completes the proof. □
Corollary 2.1
Under the hypotheses of Theorem 2.1, let(Su)\equiv 0, then{y}_{0}, {z}_{0}are natural lower and upper solutions of (1.1), and we can get the following results:

(i)
There exist two monotone sequences \{{y}_{n}\}, \{{z}_{n}\} such that {y}_{0}\le {y}_{1}\le \cdots \le {y}_{n}\le {z}_{n}\le \cdots \le {z}_{1}\le {z}_{0} which converge uniformly to the minimal and maximal solutions of (1.1), respectively.

(ii)
There exist alternating sequences \{{y}_{2n},{z}_{2n+1}\}, \{{z}_{2n},{y}_{2n+1}\} such that {y}_{0}\le {z}_{1}\le \cdots \le {y}_{2n}\le {z}_{2n+1}\le {y}_{2n+1}\le {z}_{2n}\le \cdots \le {y}_{1}\le {z}_{0} which converge uniformly to the minimal and maximal solutions of (1.1), respectively.
Corollary 2.2
If(Su)\equiv 0and(Qu)is not nondecreasing, then(\tilde{Q}u)=(Qu)+Luis nondecreasing for someL>0, and we can consider the following problem:
we see that it can be seen as (1.1) with(Qu)replaced by\tilde{Q}uand(Su)replaced byLu. Thus we get the same conclusions as for Theorem 2.1and Theorem 2.2.
Corollary 2.3
Under the hypotheses of Theorem 2.1, let(Qu)\equiv 0, if u is any natural solution of (1.1) such that{y}_{0}\le u\le {z}_{0}on J, then we have

(i)
there exist the monotone sequences \{{y}_{n}\}, \{{z}_{n}\} converging to ρ and r, where (\rho ,r) are coupled minimal and maximal solutions of (1.1), respectively, satisfying
\{\begin{array}{c}{\rho}^{\prime}(t)=(Sr)(t),\phantom{\rule{2em}{0ex}}g(\rho (0),\rho (T))=0,\hfill \\ {r}^{\prime}(t)=(S\rho )(t),\phantom{\rule{2em}{0ex}}g(r(0),r(T))=0,\hfill \end{array}
fort\in Jand\rho \le u\le r;

(ii)
there exist alternating sequences \{{y}_{2n},{z}_{2n+1}\} converging to ρ, \{{z}_{2n},{y}_{2n+1}\} converging to r, where (\rho ,r) are coupled minimal and maximal solutions of (1.1), respectively, satisfying
\{\begin{array}{c}{\rho}^{\prime}(t)=(Sr)(t),\phantom{\rule{2em}{0ex}}g(\rho (0),\rho (T))=0,\hfill \\ {r}^{\prime}(t)=(S\rho )(t),\phantom{\rule{2em}{0ex}}g(r(0),r(T))=0,\hfill \end{array}
fort\in Jand\rho \le u\le r.
Corollary 2.4
If(Qu)\equiv 0and(Su)is not nonincreasing, then(\tilde{S}u)=(Su)Nuis nonincreasing for someN>0, we can consider the following problem:
which is the same as (1.1) with(Qu)replaced by Nu and(Su)replaced by(\tilde{S}u). Hence the conclusions of Theorem 2.1and Theorem 2.2remain valid.
Corollary 2.5
Suppose(Qu)is nondecreasing but(Su)is not nonincreasing, then(\tilde{Q}u)=(Qu)+Nuis nondecreasing and(\tilde{S}u)=(Su)Nuis nonincreasing for someN>0, we can consider the following problem:
which is the same as (1.1) with(Qu)replaced by(\tilde{Q}u)and(Su)replaced by(\tilde{S}u)and the conclusions of Theorem 2.1and Theorem 2.2hold.
Corollary 2.6
If(Qu)is not nondecreasing but(Su)is nonincreasing, then(\tilde{Q}u)=(Qu)+Luis nondecreasing and(\tilde{S}u)=(Su)Luis nonincreasing for someL>0, we can consider the following problem:
which is the same as (1.1) with(Qu)replaced by(\tilde{Q}u)and(Su)replaced by(\tilde{S}u)and the conclusions of Theorem 2.1and Theorem 2.2hold.
Corollary 2.7
If(Qu)is not nondecreasing and(Su)is not nonincreasing, then for someL>0, such that(\tilde{Q}u)=(Qu)+Luis nondecreasing and(\tilde{S}u)=(Su)Luis nonincreasing, we can get the conclusions of Theorem 2.1and Theorem 2.1with(Qu)replaced by(\tilde{Q}u)and(Su)replaced by(\tilde{S}u).
We can always construct coupled upper and lower solutions of type II as in this paper. To avoid repetition, we will merely state the next two theorems without proof since it follows along the same lines as Theorems 2.1 and 2.2.
Theorem 2.3
Assume conditions H_{2}, H_{3}of Theorem 2.1hold, let{y}_{0},{z}_{0}\in {C}^{1}(J,\mathbb{R})be the coupled lower and upper solutions of type II with{y}_{0}(t)\le {z}_{0}(t)on J, we have the iterates\{{y}_{n}\}, \{{z}_{n}\}satisfying
provided{y}_{0}\le {y}_{1}and{z}_{1}\le {z}_{0}on J, where the iterates are developed by
Moreover, the monotone sequences\{{y}_{n}\}, \{{z}_{n}\}converge uniformly to ρ and r, respectively, where(\rho ,r)are coupled minimal and maximal solutions of type II for (1.1), they satisfy the coupled system
fort\in Jand if u is any natural solution of (1.1) such that{y}_{0}\le u\le {z}_{0}on J, then\rho \le u\le ron J.
Theorem 2.4
Assume the hypotheses H_{2}, H_{3}of Theorem 2.1to hold, and let{y}_{0},{z}_{0}\in {C}^{1}(J,\mathbb{R})be the coupled lower and upper solutions of type II with{y}_{0}(t)\le {z}_{0}(t)on J, then for any natural solution u of (1.1) with{y}_{0}\le u\le {z}_{0}on J, we get the alternating sequences\{{y}_{2n},{z}_{2n+1}\}, \{{z}_{2n},{y}_{2n+1}\}satisfying
provided{y}_{0}\le {y}_{1}and{z}_{1}\le {z}_{0}on J, for everyn\ge 1, where the iterative schemes are constructed by
Moreover, the monotone sequences\{{y}_{2n},{z}_{2n+1}\}converge to ρ and\{{z}_{2n},{y}_{2n+1}\}converge to r on J, where(\rho ,r)are coupled minimal and maximal solutions of type I for (1.1), respectively, satisfying the coupled system
fort\in Jand\rho \le u\le ron J.
3 Examples
In this section, we give two simple but illustrative examples, thereby validating the proposed theorems.
Example 3.1
Consider the following problem:
We construct a pair of coupled upper and lower solutions of type I for (3.1),
Obviously, {y}_{0}(t)\le {z}_{0}(t), and
It proves that {y}_{0}(t), {z}_{0}(t) are coupled lower and upper solutions of problem (3.1). Then assumptions H_{1}, H_{2}, and H_{3} hold with M=1. By Theorem 2.1, we obtain the existence of monotone sequences that approximate the extremal solutions of (3.1) in the sector [0,t+\frac{1}{10}]. By Theorem 2.2, we obtain the existence of alternating sequences that also converge to the extremal solutions.
Example 3.2
Consider the following problem:
Put {y}_{0}(t)=1, {z}_{0}(t)=1, t\in J. Obviously, {y}_{0}(t)\le {z}_{0}(t), then
Functions {y}_{0}(t), {z}_{0}(t) are coupled lower and upper solutions of type II for problem (3.2). It is easy to see that H_{1}, H_{2}, and H_{3} hold with M=1. Consider the function {y}_{1}(t)=\frac{t}{2}, {z}_{1}(t)=\frac{t}{2} satisfy {y}_{0}\le {y}_{1}, {z}_{1}\le {z}_{0}, respectively. Therefore, applying Theorem 2.3, by using (2.1) and (2.2), we obtain the existence of monotone sequences that approximate the extremal solutions of (3.2) in the sector [1,1]. From Theorem 2.4, we obtain the existence of alternating sequences that also converge to the extremal solutions.
4 Conclusions
In this paper, we have developed monotone iterative method for casual differential equations with nonlinear boundary conditions. The method is based on the new concepts of lower and upper solutions (resp. coupled lower and upper solutions). We have constructed monotone sequences and alternating sequences from a corresponding linear equations. It was proven that these sequences converge uniformly to the coupled minimal and maximal solutions of the problems.
The condition on the function g(u,v) was necessary to carry out the proofs. A future research direction would be to consider this method for nonlinear conditions of B(u(0),u)=0 or B(u,u(T))=0. Also, since the theory of causal differential equations has recently gained more attention, much work can be done on its theoretical research.
References
Ladde GS, Lakshmikantham V, Vatsala AS: Monotone Iterative Techniques for Nonlinear Differential Equations. Pitman, London; 1985.
Agarwal RP, Benchohra N, Hamani S: A survey on existing results for boundary value problems of nonlinear fractional differential equations and inclusions. Acta Appl. Math. 2010, 109: 9731033. 10.1007/s1044000893566
Agarwal RP, Franco D, O’Regan D: Singular boundary value problems for first and second order impulsive differential equations. Aequ. Math. 2005, 69: 8396. 10.1007/s0001000427359
AlRefai M, Ali Hajji M: Monotone iterative sequences for nonlinear boundary value problems of fractional order. Nonlinear Anal. 2011, 74: 35313539. 10.1016/j.na.2011.03.006
Benchohra M, Hamani S: The method of upper and lower solutions and impulsive fractional differential inclusions. Nonlinear Anal. Hybrid Syst. 2009, 3: 433440. 10.1016/j.nahs.2009.02.009
Bhaskar TG, McRae FA: Monotone iterative techniques for nonlinear problems involving the difference of two monotone functions. Appl. Math. Comput. 2002, 133: 187192. 10.1016/S00963003(01)002429
Cabada A, Grossinhob M, Minhosc F: Extremal solutions for thirdorder nonlinear problems with upper and lower solutions in reversed order. Nonlinear Anal. 2005, 62: 11091121. 10.1016/j.na.2005.04.023
Corduneanu C: Functional Equations with Causal Operators Stability, and Control. Taylor & Francis, London; 2002.
Cui Y, Zou Y: Monotone iterative method for differential systems with coupled integral boundary value problems. Bound. Value Probl. 2013., 2013: 10.1186/168727702013245
Franco D, Nieto JJ, O’Regan D: Existence of solutions for first order differential equations with nonlinear boundary conditions. Appl. Math. Comput. 2004, 153: 793802. 10.1016/S00963003(03)006787
Wang W, Yang X, Shen J: Boundary value problems involving upper and lower solutions in reverse order. J. Comput. Appl. Math. 2009, 230: 17. 10.1016/j.cam.2008.10.040
West IH, Vatsala AS: Generalized monotone iterative method for initial value problems. Appl. Math. Lett. 2004, 17: 12311237. 10.1016/j.aml.2004.03.003
Lakshmikantham V, Leela S, Drici Z, McRae FA: Theory of Causal Differential Equations. World Scientific, Singapore; 2009.
Drici D, McRae FA, Devi JV: Monotone iterative technique for periodic boundary value problems with causal operators. Nonlinear Anal. 2006, 64: 12711277. 10.1016/j.na.2005.06.033
Li Y, Sun S, Yang D, Han Z: Threepoint boundary value problems of fractional functional differential equations with delay. Bound. Value Probl. 2013., 2013: 10.1186/16872770201338
Acknowledgements
The authors would like to thank the reviewers and the editors for their valuable suggestions and comments. This work was supported by the NNSF of China (No. 61073121), the Fundamental Research Funds for the Central Universities (No. 13ZD19), and the Higher School Science Research of Hebei Province of China (No. Z2013038).
Author information
Authors and Affiliations
Corresponding author
Additional information
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
All authors contributed equally to the writing of this paper. All authors read and approved the final manuscript.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0), which permits use, duplication, adaptation, distribution, and reproduction in any medium or format, as long as 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.
About this article
Cite this article
Wang, WL., Tian, JF. Generalized monotone iterative method for nonlinear boundary value problems with causal operators. Bound Value Probl 2014, 192 (2014). https://doi.org/10.1186/s1366101401925
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s1366101401925