Ph. Cl'ement, H.J.A.M. Heijmans et al. "One-Parameter Semigroups," North Holland, CWI Monograph 5, Amsterdam, 1987.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....attractivity of center manifolds for quasilinear equations rely on maximal regularity results. These in turn are intimately connected with the continuous interpolation spaces introduced by Da Prato and Grisvard in [20] In this section we present some results about maximal regularity. We refer to [20,27,17,10,35]. To describe what maximal regularity is about, we consider the linear Cauchy problem (CP ) A;f;x) t u Au = f(t) u(0) x : 2:1) on a Banach space X: Here, GammaA is the generator of a strongly continuous analytic semigroup on X; denoted by fe ; t 0g: It is well known that A is ....

....which turns it into a Banach space. It can be shown that the mapping (X 0 ; X 1 ) X ff ; ff 2 (0; 1) 2:11) assigning to each pair (X 0 ; X 1 ) the intermediate space X ff ; defines an exact interpolation method of exponent ff : This interpolation method was introduced in [20] cf. also [27,17,10], and is called the continuos interpolation method. Besides this definition, the continuous interpolation spaces can also be introduced in another way. Indeed, due to [22] we have ff;1 = X ff : 2:12) Here, X 0 ; X 1 ) ff;1 is obtained by assigning to each pair (X 0 ; X 1 ) of densely ....

Ph. Cl'ement, H.J.A.M. Heijmans et al. "One-Parameter Semigroups," North Holland, CWI Monograph 5, Amsterdam, 1987.

....implies that s(A 0 (#) #(A 0 (#) This is proved in [5, Theorem 7.4] as a consequence of the Pringsheim Landau Lemma. The discussion of the spectrum of A 0 (#) in [8, Section 3] shows that for # # r 0. 5792 we have that s(A 0 (#) #(A 0 (#) Hence the general abstract result in [5] shows that (23) is not true for # # r . Using the structure of the problem we have found the sharp estimate # 0 for the loss of the invariance (23) 7. Monotone iterations. Having established the validity of a strong maximum principle in Section 5 we will briefly discuss the application of ....

Ph. Clement, H.J.A.M. Heijmans et al., "One-Parameter-Semigroups," CWI Monographs, Vol., 5, North-Holland, Elsevier, Amsterdam, 1987.

....of the problem. Theorem 1. Let e tA be a strongly continuous positive semigroup on L p( F ; where( F ; is a sigma nite measure space, and 1 p 1. Then (A) s(A) In order to show this result, we will make use of the following lemmas. The rst result may be derived from [C], Theorem 7.4 (the reader may like to know that a proof of the Pringsheim Landau Theorem used in [C] may be found on page 59 of [Wi] Lemma 2. Let e tA be a strongly continuous positive semigroup on a Banach lattice X, and let g 2 X. Then for any s(A) we have that ( A) 1 g = Z 1 0 ....

....where( F ; is a sigma nite measure space, and 1 p 1. Then (A) s(A) In order to show this result, we will make use of the following lemmas. The rst result may be derived from [C] Theorem 7. 4 (the reader may like to know that a proof of the Pringsheim Landau Theorem used in [C] may be found on page 59 of [Wi] Lemma 2. Let e tA be a strongly continuous positive semigroup on a Banach lattice X, and let g 2 X. Then for any s(A) we have that ( A) 1 g = Z 1 0 e s(A ) g ds: Here the right hand side is taken in the sense of an improper integral. The next ....

Ph. Clement, H.J.A.M. Heijmans, et al., One Parameter Semigroups, North-Holland, 1987.

....are in oe ess f Gamma (S) as part of the definition. But if a closed operator has finite codimension then it automatically follows that it has closed range, so the additional clause is not required. 2) The definition of the Browder essential spectrum given is equivalent to the following (See [2], Theorem A.3.3. oe ess b (S) oe(A) n f 2 C : is an isolated point of oe(A) such that the spectral projection of fg has finite rank.g: Definition 5.4 Let L 2 B(X) We define the essential spectral radius of L by r ess (L) supfjj : 2 oe ess b (L)g: Let A be a closed linear ....

.... Gamma = exp( oe ess (A) Gamma ; where (1) oe ess = oe ess b , 2) oe ess = oe ess f , or (3) oe ess = oe ess f . Proof. Without loss of generality we may assume that ffi(T ) 0 (T ) 0. Let 0. 1) By the spectral inclusion theorem for the Browder essential spectrum ([2], Proposition 8.4) it is sufficient to prove that whenever 2 oe ess b (T ( 20 with jj e ffi(T ) then = e for some 2 oe ess b (A) Let 2 oe ess b (T ( with jj e ffi(T ) Choose 0 2 C such that e 0 = and define S : f 2 oe(A) e = g f 0 2in : n 2 ....

Ph.Cl'ement et al. One-Parameter Semigroups. CWI Monographs 5, North Holland, 1987.

....u) def = Z D g d i 1 t ( ffi v Gamma ffi u ) j = d ( t ( d (ffi v Gamma ffi u ) 1. 5) The technique applied to achieve such results is placed around the TrotterKato theorem and the Duhamel perturbation formula for contraction operator semigroups on Banach spaces (e.g. 6] [4]) Roughly speaking, the Trotter Kato theorem tells us that a sequence of contraction operator semigroups converges to its limit semigroup if the corresponding sequence of generators converges to the limit generator, while the Duhamel formula expresses the rate of semigroup convergence through the ....

Clement, Ph. et al. (1987) One-Parameter Semigroups, North-Holland, Amsterdam--New York--Oxford--Tokyo.

....which real part is bounded above. Then T (t)f(x) e tm(x) f(x) is a C 0 semigroup with the generator Af(x) m(x)f(x) D(A) ff 2 X;mf 2 Xg. The semigroup T is called a multiplication semigroup. Remark. From many books dealing with continuous semigroups we have choosen as representatives [C], E N] G] H P] Kr] Pa] 3,2 Cauchy Problem We recall the definition of a mild and classical solution (Definition 2.2) of the Cauchy problem (3.2) u(t) Au(t) u(0) x ; and generalize it for a non homogeneous equation (3.3) u(t) Au(t) f(t) u(0) x : Definition 3.2 A function u ....

....and R( Gamma A) X for some 0. This proposition yields stronger results than those in Example 5,6, namely that A generates a positive contractive semigroup. Positive semigroups share similar properties with positive matrices, e.g. a theorem of Perron Frobenius type is valid ( see e.g. [C], LN] 3,4 Special Semigroups This section is devoted to special classes of semigroups which have important specific properties concerning mainly the regularity in time and or in a space variable. The first special class can be considered as a generalization of semigroups in Hilbert spaces ....

Ph.Cl'ement et al., One-Parameter Semigroups, North-Holland 1987.

....: 9C 1 with kT t k Ce t for all t 0 g: In [9] the following theorem was proved. Theorem 1: If T t is a positive c 0 semigroup on L p( Omega ; 1 p 1, then s(A) T t ) The case p = 2 is due to Gearhart and Greiner Nagel, the case p = 1 is due to Derndinger (see [7] 8] or [3], theorem 9.5 and 9.7) but the general case remained an open problem for about 10 years. The proof in [9] used a new spectral mapping theorem for the evolutionary semigroup I Omega T t on L q (L p ) by Latushkin and Montgomery Smith [5] and an extrapolation procedure for the Yosida ....

.... 1 that is based on a boundedness result for positive convolutions on mixed norm spaces L p (L q ) which may be of independent interest (see theorem 2 below) With this convolution result we can reduce theorem 1 to a well known characterization of the spectral bound in terms of weak integrability ([3], theorem 7.4) Finally, we point out, that recent counterexamples concerning stability of semigroups (see e.g. 1] can be transplanted onto L p spaces. At the end of this note we give an example of a semigroup T t on L p (0; 1) 1 p 1; p 6= 2, for which s(A) s 1 (A) T t ) Here ....

[Article contains additional citation context not shown here]

Ph. Clement, H. J. A. M. Heijmans, et al, One Parameter Semigroups, North Holland 1987

....f0g and C(D) are the only invariant closed ideals for all T t (p) t 0: A simple sufficient condition of the irreducibility consists in dimL(b 1 (#; b q (#; d Gamma 2 for any (#; 2 D (5.18) where L denotes the linear hull spanned by the given vector fields. It follows due to [10] [8] that the spectrum oe(A(p) of the generator A(p) of the positive semigroup T t (p) is not empty and s(A(p) supfRe : 2 oe(A(p) g = maxf 2 R : 2 oe(A(p) g; Gamma1 s(A(p) 1 Moreover, the resolvent R( A(p) is strongly positive for s(A(p) because T t (p) is irreducible, and ....

....by g(p) we get (5.25) and the equality A (p) p = g(p) p (5. 26) Further, as ( Gamma s(A(p) Gamma1 is a pole of the resolvent of the operator R( A(p) the number s(A(p) g(p) is a pole of R( A(p) see [10] In such a case the generalized Perron Frobenius theorem [10] see also [8]) sets not only (5.25) and (5.26) but also sets that all the points from oe(A(p) with real part g(p) are g(p) iffk; k = 0; Sigma1; Sigma2; for some ff 0, and they are all simple isolated eigenvalues of A(p) Thus, the abovementioned assertion is justified. We underline that the noted ....

Ph. Cl'ement, H.J.A.M. Heijmans, et al. One-Parameter Semigroups. North-Holland, Amsterdam, 1987.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC