We study the following reactiondiffusion system with a crossdiffusion matrix and fractional derivatives in , in , on , , in where is a smooth bounded domain, , the diffusion matrix has semisimple and positive eigenvalues , , is an open nonempty set, and is the characteristic function of . Specifically, we prove that under some conditions over the coefficients , the semigroup generated by the linear operator of the system is exponentially stable, and under other conditions we prove that for all the system is approximately controllable on .
1. Introduction
In this paper we prove controllability for the following reactiondiffusion system with cross diffusion matrix:
where is an open nonempty set of and is the characteristic function of .
We assume the following assumptions.
(H1) is a smooth bounded domain in .
(H2)The diffusion matrix has semisimple and positive eigenvalues
(H3) are real constants, are real constants belonging to the interval
(H4)
(H5)The distributed controls .
Specifically, we prove the following statements.
(i)If and , where is the first eigenvalue of with Dirichlet condition, or if , and then, under the hypotheses (H1)–(H3), the semigroup generated by the linear operator of the system is exponentially stable.
(ii)If and under the hypotheses (H1)–(H5), then, for all and all open nonempty subset of the system is approximately controllable on
This paper has been motivated by the work done in [1] and the work done by H. Larez and H. Leiva in [2]. In the work [1], the auther studies the asymptotic behavior of the solution of the system
supplemeted with the initial conditions
The author proved that in the Banach space where is the space of bounded uniformly continuous real valued functions on , if and are locally Lipshitz and under some conditions over the coefficients , and if then for all Moreover, and satisfy the system of ordinary differential equations
with the initial data
The same result holds for
In the work done in [2], the authers studied the system (1.1) with , and They proved that if the diffusion matrix has semisimple and positive eigenvalues , then if ( is the first eigenvalue of ), the system is approximately controllable on for all open nonempty subset of
2. Notations and Preliminaries
In the following we denote by
the set of matrices with entries from ,
the set of all measurable functions such that ,
the set of all the functions that have generalized derivatives for all ,
the closure of the set in the Hilbert space ,
the set of all the functions that have generalized derivatives for all .
We will use the following results.
Theorem 2.1 (cf. [3]).
Let us consider the following classical boundaryeigenvalue problem for the laplacien:
where is a nonempty bounded open set in and .
This problem has a countable system of eigenvalues and as .
(i)All the eigenvalues have finite multiplicity equal to the dimension of the corresponding eigenspace .
(ii)Let be a basis of the for every then the eigenvectors form a complete orthonormal system in the space Hence for all we have If we put then we get .
(iii)Also, the eigenfunctions , where is the space of infinitely continuously differentiable functions on and compactly supported in .
(iv)For all we have .
(v)The operator generates an analytic semigroup on defined by
Definition 2.2.
Let a real number, the operator is defined by
In particular, we obtain and . Since form a complete orthonormal system in the space then it is dense in , and hence is dense in .
Proposition 2.3 (cf. [4]).
Let be a Hilbert separable space and and two families of bounded linear operators in , with a family of complete orthogonal projections such that
Define the following family of linear operators Then
(a) is a linear and bounded operator if with continiuous for
(b)under the above condition (a), is a strongly continiuous semigroup in the Hilbert space whose infinitesimal generator is given by
Theorem 2.4 (cf. [5]).
Suppose is connected, is a real function in , and on a nonempty open subset of . Then in .
3. Abstract Formulation of the Problem
In this section we consider the following notations.
(i) is a Hilbert space with the inner product
We define
(iii) Let then we can define the linear operator
where
Therefore, for all
If we put
then (3.3) can be written as
and we have for all
Consequently, system (1.1) can be written as an abstract differential equation in the Hilbert space in the following form:
where and is a bounded linear operator from into .
4. Main Results
4.1. Generation of a Semigroup
Theorem 4.1.
If , then, under hypotheses (H1)–(H3), the linear operator defined by (3.3) is the infinitesimal generator of strongly continuous semigroup given by
where
Moreover, if
then the semigoup is exponentially stable, that is, there exist two positives constants such that
Proof.
In order to apply the Proposition 2.3, we observe that can be written as follows:
where
Therefore, and
Now, we have to verify condition (a) of the Proposition 2.3. We shall suppose that Then, there exists a set of complementary projections on such that
If is the matrix passage from the canonical basis of to the basis composed with the eigenvectors of , then
Hence,
We have also
From (4.10)(4.11) into (4.7) we obtain
where
As we get
As as then this implies the existence of a positive number and a real number such that for every Therefore is a strongly continious semigroup given by (4.1). We can even estimate the constants and as follows.
(i) If As , then there exist constants
hence, if we put
we easily obtain
If . If we put
then we find that
Therefore, the linear operator generates a strongly continuous semigroup on given by expression (4.1).
Finally, if we have already proved (4.20). Using (4.20) into (4.1) we get that the semigoup is exponentially stable. The expression (4.5) is verfied with and is defined by (4.19).
Theorem 4.2.
If
then, under the hypotheses (H1)–(H3), the linear operator defined by (3.3) is the infinitesimal generator of strongly continuous semigroup exponentially stable defined by (4.1). Specially, there exist two positives constants such that
To prove this result, we need the following lemma.
Lemma 4.3.
For every two real positives constants and , one has for every
and for every
Proof of Lemma 4.3.
It is easy to verify that for every , for all .
Let and , then we get
Hence, we get (4.23).
Also, it is easy to verify that for every , for all . Let and , then we get
Hence, from (4.26) we get for all and , which gives (4.24).
With the same manner we can prove that for every and every we have
and consequently, for every two real positives constants and and every we have
Now, we are ready to prove Theorem 4.2.
Proof of Theorem 4.2.
By applying Proposition 2.3 we start from formula (4.12) and we put
where
To estimate we have in taking into account
and applying the Lemma 4.3 we get
for all and . But we have , for all Then we get for every that
From (4.31)(4.33) we get
where
and
Applying Lemma 4.3 and taking into account (4.21) we get with the same manner that for every
where
and or every
where
From (4.34)(4.40) into (4.12) we get
where is defined by (4.17) and
Using (4.41) into (4.1) we get that the semigoup generated by is exponentially stable. Expression (4.22) is verfied with and is defined by (4.42).
4.2. Approximate Controllability
Befor giving the definition of the approximate controllabiliy for the sytem (3.9), we have the following known result: for all and the initial value problem (3.9) admits a unique mild solution given by
This solution is denoted by
Definition 4.4.
System (3.9) is said to be approximately controllable at time whenever the set is densely embedded in ; that is,
The following criteria for approximate controllability can be found in [6].
Criteria 1.
System (3.9) is approximately controllable on if and only if
Now, we are ready to formulate the third main result of this work.
Theorem 4.5.
If the following condition
is satisfied; then, under hypotheses (H1)–(H5), for all and all open subset system (3.9) is approximately controllable on .
Proof.
The proof of this theorem relies on the Criteria 1 and the following lemma.
Lemma 4.6.
Let and be sequences of real numbers such that , and , for all , then for any one has
Proof of Lemma 4.6.
By analyticity we get and from this we get . Under the assumptions of the lemma we get as and so If , we divide by and we pass we get . If we divide by and we pass and get . If , we divide by and we pass and get But in this we case we can integrate under the symbol of sommation over the intervall and we get . Hence . Continuing this way we see that for all
We are now ready to prove Theorem 4.5. For this purpose, we observe that
where is the semigroup generated by .
Without lose of generality, we suppose that Hence
where
Now, suppose for that , for all Then
If (4.46) is satisfied, then (4.50) take the form
Then, from lemma 4.6 we obtain that for and all
Since we get that all
On the other hand, from Theorem 2.4 we know that are analytic functions, which implies the analticity of and Then we can conclude that for and all
Hence for all which implies that This completes the proof of Theorem 4.5.
References

Badraoui, S: Asymptotic behavior of solutions to a reactiondiffusion system with a cross diffusion matrix on unbounded domains. Electronic Journal of Differential Equations. 2006(61), 1–13 (2006)

Larez, H, Leiva, H: Interior controllability of a a reactiondiffusion system with cross diffusion matrix. to appear in Boundary Value Problems

Zeidler, E: Applied Functional Analysis, Applied Mathematical Sciences,p. xvi+404. Springer, New York, NY, USA (1995)

Leiva, H: A lemma on semigroups and applications. Quaestiones Mathematicae. 26(3), 247–265 (2003). Publisher Full Text

Axler, S, Bourdon, P, Ramey, W: Harmonic Function Theory, Graduate Texts in Mathematics,p. xii+231. Springer, New York, NY, USA (1992)

Curtain, RF, Zwart, H: An Introduction to InfiniteDimensional Linear Systems Theory, Texts in Applied Mathematics,p. xviii+698. Springer, New York, NY, USA (1995)