Professional Documents
Culture Documents
Textbook Mathematical Theory of Compressible Viscous Fluids Analysis and Numerics 1St Edition Eduard Feireisl Ebook All Chapter PDF
Textbook Mathematical Theory of Compressible Viscous Fluids Analysis and Numerics 1St Edition Eduard Feireisl Ebook All Chapter PDF
https://textbookfull.com/product/singular-limits-in-
thermodynamics-of-viscous-fluids-2nd-edition-eduard-feireisl/
https://textbookfull.com/product/handbook-of-mathematical-
analysis-in-mechanics-of-viscous-fluids-1st-edition-yoshikazu-
giga/
https://textbookfull.com/product/elasticity-theory-applications-
and-numerics-fourth-edition-sadd/
https://textbookfull.com/product/elasticity-theory-applications-
and-numerics-4th-edition-martin-h-sadd/
Analysis on Fock spaces and mathematical theory of
quantum fields 1st Edition Asao Arai
https://textbookfull.com/product/analysis-on-fock-spaces-and-
mathematical-theory-of-quantum-fields-1st-edition-asao-arai/
https://textbookfull.com/product/the-analysis-of-drugs-in-
biological-fluids-2nd-edition-chamberlain/
https://textbookfull.com/product/existence-theory-for-
generalized-newtonian-fluids-1st-edition-breit/
https://textbookfull.com/product/theory-and-modeling-of-rotating-
fluids-convection-inertial-waves-and-precession-1st-edition-keke-
zhang/
https://textbookfull.com/product/mathematical-analysis-and-the-
mathematics-of-computation-1st-edition-werner-romisch/
Lecture Notes in Mathematical Fluid Mechanics
Eduard Feireisl
Trygve G. Karper
Milan Pokorný
Mathematical
Theory of
Compressible
Viscous Fluids
Analysis and Numerics
Advances in Mathematical Fluid Mechanics
Editor-in-Chief:
Galdi, Giovanni P
Series Editors
Bresch, D.
John, V.
Hieber, M.
Kukavica, I.
Robinson, J.
Shibata, Y.
Mathematical Theory
of Compressible
Viscous Fluids
Analysis and Numerics
Eduard Feireisl Trygve G. Karper
Institute of Mathematics CAS Department of Mathematical Sciences
Praha, Czech Republic Norwegian University of Science & Tech.
Trondheim, Norway
Milan Pokorný
Charles University
Faculty Mathematics and Physics
Charles University
Praha, Czech Republic
v
vi Preface
to the numerical analysis and the associated real world applications. For the sake
of clarity of presentation, we discuss first the case, where the pressure term grows
sufficiently fast for large value of the density yielding strong energy bounds. The
main novelty of our approach is a detailed existence proof performed in the second
part, where the approximate solutions are constructed by means of a mixed finite
element finite volume numerical scheme. In particular we show that the numerical
solutions, up to a subsequence, converge to a weak solution of the (compressible)
Navier–Stokes system at least for the pressure that increases sufficiently fast for
large values of the density. The final third part of the book is devoted to the
mathematical theory with (nowadays) optimal restriction on the pressure function.
The research of Eduard Feireisl leading to these results has received funding from
the European Research Council under the European Union’s Seventh Framework
Programme (FP7/2007–2013)/ERC Grant Agreement 320078. The Institute of
Mathematics of the Academy of Sciences of the Czech Republic is supported by
RVO:67985840. The work of Milan Pokorný was supported by the GAČR (Czech
Science Foundation) project No. 16-03230S.
The authors thank Martin Michálek for preparing the graphic material and Radim
Hošek and Martin Michálek for critical reading and valuable comments on the final
version of the manuscript.
vii
Contents
ix
x Contents
References .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 179
Index . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 185
Chapter 1
Preliminaries, Notation, and Spaces of Functions
This chapter introduces notation as well as the basic mathematical tools used in the
book such as the function spaces, embedding theorems, and elementary inequalities.
We suppose the reader to be familiar with this material and will refer to it throughout
the text without further specification.
1.1 Notation
Unless otherwise indicated, the symbols and the basic notation will be used as stated
below.
1.1.1 Numbers
The symbols Z, N, and C denote the sets of integers, positive integers, and complex
numbers, respectively. The symbol R denotes the set of real numbers, and RN is the
N-dimensional Euclidean space.
The symbol const, c, or ci will be used for a generic positive constant. These
constants may have different values at different parts of the book. We will also write
< < <
u v if u cv; and u v if u v and v u:
The symbol ˝ RN will be used for a domain—an open connected subset of the
Euclidean space RN , mostly N D 3. The closure of a set Q RN is denoted by Q,
its boundary is @Q. The symbol 1Q stands for the characteristic function of the set
Q. The outer normal vector to @Q is denoted by n.
X
N
abD ai bi ;
iD1
X
N
AWBD Ai;j Bi;j :
i;jD1
a ˝ b D fai bj gNi;jD1 :
a b D .a2 b3 a3 b2 ; a3 b1 a1 b3 ; a1 b2 a2 b1 /:
X
N
ŒA bi D Ai;j bj for i D 1; : : : ; N;
jD1
1.2 Derivatives and Differential Operators 3
AB with components
X
M
ŒABi;j D Ai;k Bk;j :
kD1
The symbol
@g
@yi g.y/ D .y/; i D 1; : : : ; N;
@yi
1.2.1 Gradient
rx g.t; x/ D Œ@x1 g.t; x/; @x2 g.t; x/; @x3 g.t; x/:
rv D ry v D f@yj vi gNi;jD1 I
1.2.2 Divergence
X
N
div v D divy v D @yi vi : (1.1)
iD1
3
X
divx v.t; x/ D @xi vi .t; x/:
iD1
3
X
Œdiv Bi D Œdivx B.t; x/i D @xj Bi;j .t; x/; i D 1; : : : ; 3:
jD1
1.3 Function Spaces 5
x D divx rx :
The vorticity (rotation) curl of a vectorial function v D Œv1 .y/; : : : ; vN .y/ is the
skew-symmetric matrix
n oN
curl v D curly v D rv r T v D @yj vi @yi vj : (1.2)
i;jD1
The vorticity of a vectorial function v D Œv1 .t; x/; : : : ; v3 .t; x/ is a skew-
symmetric matrix
n o3
curlx v D rx v rxT v D @xj vi @xi vj :
i;jD1
If not otherwise stated, all vectorial spaces considered in this book are real. For a
normed linear space X, we denote by k kX the norm on X. The duality pairing
between an abstract vector space X and its dual X is denoted as < I >X IX , or
simply < I > in case the underlying spaces are clearly identified in the context. In
particular, if X is a Hilbert space, the symbol < I > denotes also the scalar product
in X.
The symbol spanfMg, where M is a subset of a vector space X, denotes the space
of all finite linear combinations of vectors contained in M.
For Q RN , the symbol C.Q/ denotes the set of continuous functions on Q. For a
bounded set Q, the symbol C.Q/ denotes the Banach space of functions continuous
on the closure Q endowed with norm
is continuous on Q for any linear form f belonging to the dual space X . We shall
say that gn ! g in Cweak .QI X/ if
and
P
where @˛ u stands for the partial derivative @˛x11 : : : @˛xNN u of order j˛j D NiD1 ˛i . The
spaces Ck; .QI RM / are defined in a similar way. Finally, we set C1 D \1 k
kD0 C .
We recall the standard Arzelà–Ascoli Theorem:
Theorem 1 Let Q RM be compact and X a compact topological metric space
endowed with a metric dX . Let fvn g1
nD1 be a sequence of functions in C.QI X/ that is
equi-continuous, meaning, for any " > 0 there is ı > 0 such that
h i
dX vn .y/; vn .z/ " provided jy zj < ı independently of n D 1; 2; : : : :
1.3 Function Spaces 7
Then fvn g1
nD1 is precompact in C.QI X/, that is, there exists a subsequence (not
relabeled) and a function v 2 C.QI X/ such that
h i
sup dX vn .y/; v.y/ ! 0 as n ! 1:
y2Q
The symbol Cck .QI RM /, k 2 f0; 1; : : : ; 1g denotes the vector space of functions
belonging to Ck .QI RM / and having compact support in Q. If Q RN is an open set,
the symbol D.QI RM / will be used alternatively for the space Cc1 .QI RM / endowed
with the topology induced by the convergence:
if
p
The symbol Lloc .QI X/ denotes the vector space of locally L p -integrable func-
tions, meaning
p
v 2 Lloc .QI X/ if v 2 L p .KI X/ for any compact set K in Q:
For f 2 L1 .Q/ the set of all Lebesgue points is of full measure, meaning its
complement in Q is of zero Lebesgue measure. A similar statement holds for vector
valued functions f 2 L1 .QI X/, where X is a Banach space (see Brezis [8]). If f 2
C.Q/, then identity (1.4) holds for all points a in Q.
Linear functionals on L p .QI X/ are characterized as follows:
Theorem 2 Let Q RN be a measurable set, X a Banach space that is reflexive
and separable, 1 p < 1.
Then any continuous linear form 2 ŒL p .QI X/ admits a unique representation
0
w 2 L p .QI X /,
Z
< I v >.Lp .QIX//IL p .QIX/ D < w .y/I v.y/ >X IX dy for all v 2 L p .QI X/;
Q
1.3 Function Spaces 9
where
1 1
C 0 D 1:
p p
Accordingly, the spaces L p .QI X/ are reflexive for 1 < p < 1 as soon as X is
reflexive and separable.
See Gajewski et al. [45, Chap. IV, Theorem 1.14, Remark 1.9].
Identifying with w , we write
0
ŒL p .QI X/ D L p .QI X /; kkŒL p .QIX/ D kkL p0 .QIX / ; 1 p < 1:
p0
ŒL p .QI X/ D Lweak./ .QI X / for 1 p < 1;
where
p0
Lweak./ .QI X /
n ˇ
ˇ
W Q ! X ˇ y 2 Q 7!< .y/I v >X IX measurable for any fixed v 2 X;
0
o
y 7! k.y/kX 2 L p .Q/
1 1 1
kuvkLr .Q/ kukL p .Q/ kvkLq .Q/ ; D C ; 1 p; q; r 1
r p q
.1/ 1 1
kvkLr .Q/ kvkL p .Q/ kvkLq .Q/ ; D C ; 1 p < r < q 1; 2 .0; 1/
r p q
Then
Z Z Z
r./ b0 exp a.t/ dt C ˇ.t/ exp a.s/ ds dt
0 0 t
A domain ˝ RN is of class C if for each point x 2 @˝, there exist r > 0 and a
mapping W RN1 ! R belonging to a function class C such that—upon rotating
and relabeling the coordinate axes if necessary—we have
9
˝ \ B.xI r/ D fy j .y0 / < yN g \ B.x; r/ =
; where y0 D .y1 ; : : : ; yN1 /:
;
@˝ \ B.xI r/ D fy j .y0 / D yN g \ B.x; r/
Z M Z
X
f dSx D 'i f dSx ; i D @˝ \ B.xi I r/;
A iD1 i
see Nečas [75, Section I.2] or Kufner et al. [64, Sect. 6.3].
A Lipschitz domain ˝ admits the outer normal vector n.x/ for a.a. x 2 @˝. Here
a:a: refers to the surface measure on @˝.
A differential operator @˛ of order j˛j can be identified with a distribution
Z
< @˛ vI ' >D0 .Q/ID.Q/ D .1/j˛j v@˛ ' dy
Q
for any locally integrable function v. The Sobolev spaces W k;p .QI RM /, 1 p 1,
k a positive integer, are the spaces of functions having all distributional derivatives
up to order k in L p .QI RM /. The norm in W k;p .QI RM / is defined as
8P P 1=p 9
ˆ
<
M
k@˛
v k
p
if 1 p < 1 >
=
iD1 j˛jk i L .Q/
p
kvkW k;p .QIRM / D ;
:̂ >
;
˛
max1iM; j˛jk fk@ vi kL1 .Q/ g if p D 1
where the symbol @˛ stands for any partial derivative of order j˛j.
If Q is a bounded domain with boundary of class Ck1;1 , then there exists a
continuous linear operator which maps W k;p .Q/ to W k;p .RN /; it is called extension
operator. For 1 < p < 1 the extension operator exists even for the boundary of
class C0;1 . If 1 p < 1, then W k;p .Q/ is separable and the space Ck .Q/ is its dense
subspace.
The space W 1;1 .Q/, where Q is a bounded domain, is isometrically isomorphic
to the space C0;1 .Q/ of Lipschitz functions on Q.
The symbol W0 .QI RM / denotes the completion of Cc1 .QI RM / with respect to
k;p
0;p
the norm k kW k;p .QIRM / . In what follows, we identify W 0;p .˝I RN / D W0 .˝I RN /
with L p .˝I RN /. R
We denote LP p .Q/ D fu 2 L p .Q/ j Q u dy D 0g and W P 1;p .Q/ D W 1;p .Q/\LPp .Q/.
If Q R is a bounded domain, then LPp .Q/ and W
N P .Q/ can be viewed as closed
1;p
Theorem 3 Let ˝ RN be a domain, and let 1 p < 1. Then the dual space
ŒW0 .˝/ is a proper subspace of the space of distributions D0 .˝/. Moreover, any
k;p
XZ
< f I v >ŒW k;p .˝/ IW k;p .˝/ D .1/j˛j w˛ @˛ v dx; (1.5)
0 0
˝
j˛jk
0 1 1
where w˛ 2 L p .˝/; C 0 D 1:
p p
The dual to the Sobolev space W .˝/ admits formally the same representation
k;p
that vanishes on D.˝/ but generates a nonzero linear form when applied to v 2
W 1;p .˝/.
1.4.3 Embeddings
Np
1 q p D :
N kp
p Np
q> ; where p D if kp < N;
p 1 N kp
q > 1 for kp D N;
or
q 1 if kp > N:
0
Then the space Lq .˝/ is compactly embedded into the space W k;p .˝/, 1=p C
1=p0 D 1.
Q Q jy zjLCˇp
j˛jDk
X
M
k.v'i / ı .I0 ; /kW kCˇ;p .RN1 / :
p p
kvkW kCˇ;p .@˝/ D
iD1
In the above formulas .I0 ; / W RN1 ! RN maps y0 to .y0 ; .y0 //. For more details
see, e.g., Nečas [75, Sect. 3.8].
In the situation when ˝ RN is a bounded Lipschitz domain, the Sobolev-
Slobodeckii spaces admit similar embeddings as classical Sobolev spaces. Namely,
1.4 Sobolev Spaces 15
the embeddings
such that
1
0 .`.v// D v for all v 2 W 1 p ;p .@˝/
0
for any u 2 W 1;p .˝/, v 2 W 1;p .˝/, where n is the outer normal vector to the
boundary @˝.
See Nečas [75, Theorems 5.5, 5.7].
1 1
The dual ŒW 1 p ;p .@˝/ to the Sobolev–Slobodeckii space W 1 p ;p .@˝// D
1
;p p10 ;p0
W p0 .@˝/ is denoted by W .@˝/.
16 1 Preliminaries, Notation, and Spaces of Functions
for
1 1
0 ˛; ˇ; 1; 1 < p; q; r < 1; ˛ D ˇ C .1 /; D C
r p q
The above theorem can be viewed as a particular case of more general results,
for which we refer to Ziemer [99, Chap. 4, Theorem 4.5.1].
The symbol BV.Q/ denotes the space of functions in L1 .Q/, with distributional
derivatives belonging to the space of measures M.Q/. Functions belonging to
BVŒ0; T possess well-defined right and left-hand limits and as such can be defined
at any t 2 Œ0; 1.
More generally, we may define the space BV.Œ0; TI X/ of functions of bounded
variation from a real interval Œ0; T into a metric space X endowed with a metric d,
X
m
VX Œv D sup d.v.ti /I v.ti1 //:
0t1 <<tm T iD1
1.5 Fourier Transform 17
We report the following variant of the celebrated Helly’s Theorem, see Fleischer
and Porter [43, Theorem 2.3]:
Theorem 8 Let fvn g1 nD1 BV.Œ0; TI X/ be sequence ranging in a complete metric
space X that is bounded,
[1
nD1 vn .t/ is precompact in X for any t 2 Œ0; T:
Then, up to a subsequence,
vn .t/ ! v.t/ for any t 2 Œ0; T; where v 2 BV.Œ0; TI X/;
VX Œv sup VX Œvn :
n1
The symbol S.RN / denotes the space of smooth rapidly decreasing (complex
valued) functions, specifically, S.RN / consists of functions u such that
for all s; m D 0; 1; : : : .
Let v D v.x/ be a complex valued function from S.RN /. The Fourier transform
of v is a complex valued function Fx! Œv of the variable 2 RN defined as
N=2 Z
1
Fx! Œv./ D v.x/ exp.i x/ dx: (1.7)
2 RN
The Fourier transform is an operator from S.RN / onto S.RN /. Its inverse reads
1 N=2 Z
1
F!x Œf D f ./ exp.ix /d: (1.8)
2 RN
J.
K.
Kellermann, 255.
L.
Lacretelle, 264.
Lacroix, 237.
Lakanal, 216.
Lally-Tollendal, 100.
Laplanche, 231.
Lecarpentier, 216.
Linguet, 107.
Locke, 110
Loire, war on the, 205.
Louis XI, 5.
M.
Malmesbury, 277.
Manfredini, 42.
Maréchaussée, the, 8.
Maret, 108.
Masséna, 274.
Maury, 99.
Messidor, 228.
Milhaud, 274.
Moreau, 274.
Morelly, 36.
N.
Newton, 110.
Ney, 274.
Nivôse, 227.
P.
Paris, excitement and distress in, 26, 51, 58, 59, 60, 67, 68, 69,
133,
136, 171, 173;
influence of, on finance of Constituent Assembly, 89, 95;
clubs in, 105, 106, 144;
numbers of Jacobins in, 143;
government of, 145-147;
Prison Massacres in, 175-179;
agitation against Louis, 171, 173, 191;
the Girondists and, 193, 196-205;
under the Terror, 222-228, 242, 260;
Reaction in, 263, 264, 271, 272, 273, 279, 281.
Paris, Parlement of, 49, 50.
Payan, 256.
Pluviôse, 228.
Pombal, 42.
Prairial, 228.
See Insurrection.
Q.
Quesnai, 36.
R.
Reynaud, 217.
Richard, 274.
Santerre, 172.
Sémonville, 126.
Sergent, 178.
Soult, 274.
Suvórof, 277.
T.
Talon, 126.
Thermidor, 228;
insurrection of, 259, 260.
Thibeaudeau, 263.
Thugut, 276.
Tronchet, 103.
V.
Vacheron, 231.
Vendémiaire, 227;
insurrection of, 281.
Ventôse, 228.
Victor, 274.
Volney, 107.
W.
Westermann, 237.
Y.
1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside
the United States, check the laws of your country in addition to
the terms of this agreement before downloading, copying,
displaying, performing, distributing or creating derivative works
based on this work or any other Project Gutenberg™ work. The
Foundation makes no representations concerning the copyright
status of any work in any country other than the United States.
1.E.6. You may convert to and distribute this work in any binary,
compressed, marked up, nonproprietary or proprietary form,
including any word processing or hypertext form. However, if
you provide access to or distribute copies of a Project
Gutenberg™ work in a format other than “Plain Vanilla ASCII” or
other format used in the official version posted on the official
Project Gutenberg™ website (www.gutenberg.org), you must, at
no additional cost, fee or expense to the user, provide a copy, a
means of exporting a copy, or a means of obtaining a copy upon
request, of the work in its original “Plain Vanilla ASCII” or other
form. Any alternate format must include the full Project
Gutenberg™ License as specified in paragraph 1.E.1.
• You pay a royalty fee of 20% of the gross profits you derive from
the use of Project Gutenberg™ works calculated using the
method you already use to calculate your applicable taxes. The
fee is owed to the owner of the Project Gutenberg™ trademark,
but he has agreed to donate royalties under this paragraph to
the Project Gutenberg Literary Archive Foundation. Royalty
payments must be paid within 60 days following each date on
which you prepare (or are legally required to prepare) your
periodic tax returns. Royalty payments should be clearly marked
as such and sent to the Project Gutenberg Literary Archive
Foundation at the address specified in Section 4, “Information