Volume 2013, Issue 1 619068
Research Article
Open Access

The Local Strong Solutions and Global Weak Solutions for a Nonlinear Equation

Meng Wu

Corresponding Author

Meng Wu

Department of Mathematics, Southwestern University of Finance and Economics, Chengdu 610074, China swufe.edu.cn

Search for more papers by this author
First published: 20 May 2013
Citations: 1
Academic Editor: Shaoyong Lai

Abstract

The existence and uniqueness of local strong solutions for a nonlinear equation are investigated in the Sobolev space C([0, T); Hs(R)) ∩C1([0, T); Hs−1(R)) provided that the initial value lies in Hs(R) with s > 3/2. Meanwhile, we prove the existence of global weak solutions in L([0, ∞); L2(R)) for the equation.

1. Introduction

Coclite and Karlsen [1] investigated the well posedness in classes of discontinuous functions for the generalized Degasperis-Procesi equation:
()
which is subject to the condition
()
or
()
where c is a positive constant. The existence and L1 stability of entropy weak solutions belonging to the class L1(R)∩BV(R) are established for (1) in paper [1].
In this work, we study the following model:
()
where m is a positive constant and h(u) ∈ C3. If m = 4 and h(u) = u2/2, (4) reduces to the classical Degasperis-Procesi model (see [213]). Here, we notice that assumptions (2) and (3) do not include the case h(u) = u3. In this paper, we will study the case h(u) = u3, and m is an arbitrary positive constant.
In fact, the Cauchy problem of (4) in the case h(u) = u3 is equivalent to the following system:
()
Using the operator to multiply the first equation of the problem (5), we obtain
()

It is shown in this work that there exists a unique local strong solution in the Sobolev space C([0, T); Hs(R))∩C1([0, T); Hs−1(R)) by assuming that the initial value u0(x) belongs to Hs(R) with s > 3/2. In addition, we prove the existence of global weak solutions in L([0, ); L2(R)) for the system (6).

This paper is organized as follows. Section 2 investigates the existence and uniqueness of local strong solutions. The result about global weak solution is given in Section 3.

2. Local Existence

In this section, we will use the Kato theorem in [14] for abstract differential equation to establish the existence of local strong solution for the problem (6). Let us consider the following problem:
()
Let X and Y be Hilbert spaces such that Y is continuously and densely embedded in X, and let Q:YX be a topological isomorphism. Let L(Y, X) be the space of all bounded linear operators from Y to X. In the case of X = Y, we denote this space by L(X). We illustrate the following conditions in which σ1, σ2, σ3, and σ4 are constants depending only on max {∥yY, ∥zY}.
  • (i)

    H(y) ∈ L(Y, X) for yX with

    ()
    and H(y) ∈ G(X, 1, β) (i.e., H(y) is quasi-m-accretive), uniformly on bounded sets in Y.

  • (ii)

    QH(y)Q−1 = H(y) + A(y), where A(y) ∈ L(X) is bounded, uniformly on bounded sets in Y. Moreover,

    ()

  • (iii)

    g : YY extends to a map from X into X, is bounded on bounded sets in Y, and satisfies

    ()
    ()

Kato Theorem (see [14].)Assume that conditions (i), (ii), and (iii) hold. If v0Y, there is a maximal T > 0 depending only on and a unique solution v to the problem (7) such that

()
Moreover, the map v0v(·, v0) is a continuous map from Y to the following space:
()

In order to apply the Kato theorem to establish the local well posedness for the problem (6), we let H(u) = 3u2x, Y = Hs(R), X = Hs−1(R), , g(u) = (m − 1)Λ−2x(u3), and Q = Λs. We know that Q is an isomorphism of Hs onto Hs−1. Now, we cite the following Lemmas.

Lemma 1. The operator A(u) = u2x with uHs(R), s > 3/2 belongs to G(Hs−1(R), 1, β).

Lemma 2. Assume that H(u) = 3u2x with uHs(R) and s > 3/2. Then, H(u) ∈ L(Hs(R), Hs−1(R)) for all uHs(R). Moreover,

()

Lemma 3. For s > 3/2, u, zHs(R) and wHs−1, it holds that A(u) = [Λs, 3u2xsL(Hs−1) for uHs and

()

The above three Lemmas can be found in Ni and Zhou [15].

Lemma 4. Let u, zHs with s > 3/2 and g(u) = (m − 1)Λ−2x(u3). Then, g is bounded on bounded sets in Hs and satisfies

()

Proof. For s0 > 1/2, we know that . Consequently, we have

()

Using the Kato Theorem, Lemmas 14, we immediately obtain the local well-posedness theorem.

Theorem 5. Assume that u0Hs(R) with s > 3/2. Then, there exists a T > 0 such that the system (5) or the problem (6) has a unique solution u(t, x) satisfying

()

3. Weak Solutions

In this section, our aim is to establish the existence of global weak solutions for the system (6). Firstly, we prove that the solution of the problem (5) is bounded in the space L2(R) and L(R).

Lemma 6. The solution of the problem (5) with m > 0 satisfies

()
where , and . Moreover, there exist two constants c1 > 0 and c2 > 0 depending only on m such that
()

Proof. Setting and and using the first equation of the problem (5), we obtain u = myyxx and

()
Using the Parseval identity and (21), we obtain (19) and (20).

From Theorem 5, we know that for any u0Hs(R) with s > 3/2, there exists a maximal T = T(u0) > 0 and a unique strong solution u to the problem (6) such that
()
Firstly, we study the following differential equation:
()

Lemma 7. Let u0Hs, s > 3, and let T > 0 be the maximal existence time of the solution to the problem (6). Then, the problem (23) has a unique solution pC1([0, T) × R, R). Moreover, the map p(t, ·) is an increasing diffeomorphism of R with px(t, x) > 0 for (t, x)∈[0, T) × R.

Proof. Using Theorem 5, we obtain uC1([0, T); Hs−1(R)) and Hs−1C1(R). Therefore, we know that functions u(t, x) and ux(t, x) are bounded, Lipschitz in space, and C1 in time. Using the existence and uniqueness theorem for ordinary differential equations derives that the problem (23) has a unique solution pC1([0, T) × R, R).

Differentiating (23) with respect to x gives rise to the following:

()
from which we obtain
()
For every T < T, using the Sobolev imbedding theorem yields that
()

It is inferred that there exists a constant K0 > 0 such that for (t, x)∈[0, T) × R. It completes the proof.

Lemma 8. Assume that u0Hs(R), s > 3/2. Let T be the maximal existence time of the solution u to the problem (6). Then, it has

()
where c > 0 is a constant independent of t.

Proof. Let ξ(x) = (1/2)e−|x|, we have for all gL2(R) and u = ξK1(t, x). Using a simple density argument presented in [7], it suffices to consider s = 3 to prove this lemma. Let T be the maximal existence time of the solution u to the problem (6) with the initial value u0H3(R) such that uC([0, T), H3(R))∩C1([0, T), H2(R)). From (6), we have

()
Since
()
from (29), we have
()
Using Lemma 6 and (30) derives that
()
where c is a positive constant independent of t. Using (31) results in the following:
()
Therefore,
()
Using the Sobolev embedding theorem to ensure the uniform boundedness of ux(s, η) for (s, η)∈[0, t] × R with t ∈ [0, T), from Lemma 7, for every t ∈ [0, T), we get a constant C(t) such that
()
We deduce from (34) that the function p(t, ·) is strictly increasing on R with lim x→±p(t, x) = ± as long as t ∈ [0, T). It follows from (33) that
()
Using the Gronwall inequality and (35) derives that (27) holds.

For a real number s with s > 0, suppose that the function u0(x) is in Hs(R), and let uε0 be the convolution uε0 = ϕεu0 of the function ϕε(x) = ε−1/4ϕ(ε−1/4x) and u0 such that the Fourier transform of ϕ satisfies , , and for any ξ ∈ (−1,1). Then, we have uε0(x) ∈ C. It follows from Theorem 5 that for each ε satisfying 0 < ε < 1/2, the Cauchy problem,
()
has a unique solution uε(t, x) ∈ C([0, T); H). Using Lemmas 6 and 8, for every t ∈ [0, T), we obtain
()
Sending tT, we know that inequalities (37) are still valid. This means that for t ∈ [0, ), (37) hold.

Now, we state the concepts of weak solutions.

Definition 9 (weak solution). We call a function u : R+ × RR a weak solution of the Cauchy problem (5) provided that

  • (i)

    uL(R+; L2(R));

  • (ii)

    in D([0, ) × R), that is, for all there holds the following identity:

    ()

Theorem 10. Let u0(x) ∈ L2(R). Then, there exists a weak solution u(t, x) ∈ L([0, ); L2(R)) to the problem (5).

Proof. Consider the problem (36). For an arbitrary T > 0, choosing a subsequence εn → 0, from (37), we know that is bounded in L and is uniformly bounded in L2(R). Therefore, we obtain that is bounded in L2(R). Therefore, there exist subsequences and , still denoted by and , are weakly convergent to v in L2(R). Noticing (38) completes the proof.

Acknowledgment

This work is supported by the Fundamental Research Funds for the Central Universities (JBK120504).

      The full text of this article hosted at iucr.org is unavailable due to technical difficulties.