• Aucun résultat trouvé

Derivation of relaxional transport equations for a gas of pseudo-Maxwellian molecules

N/A
N/A
Protected

Academic year: 2021

Partager "Derivation of relaxional transport equations for a gas of pseudo-Maxwellian molecules"

Copied!
5
0
0

Texte intégral

(1)

HAL Id: jpa-00246476

https://hal.archives-ouvertes.fr/jpa-00246476

Submitted on 1 Jan 1992

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Derivation of relaxional transport equations for a gas of pseudo-Maxwellian molecules

Alexander Orlov

To cite this version:

Alexander Orlov. Derivation of relaxional transport equations for a gas of pseudo-Maxwellian molecules. Journal de Physique I, EDP Sciences, 1992, 2 (3), pp.229-232. �10.1051/jp1:1992137�.

�jpa-00246476�

(2)

Classification Physics Abstracts

05.20D 05.60 47.45

Short Communication

Derivation of relaxional transport equations

for

a

gas of pseudo-Maxwellian molecules

Alexander V.

Orlov(*)

Dept. 9, Institute for High Temperatures

(USSR

Acad.

Sci.), 13/19

Izhorskaya St., Moscow 127412, Russia

(Received

16 October 1991, revised 9 December 1991, accepted 20 December

1991)

Abstract. In this brief note the moment and energy transport equations of relaxational type are derived by the proposed iteration scheme of the solution of the Boltzmann equation for

pseudo-Maxwellian molecules.

The Boltzmann

equation

fif/fit

+ v V

f

=

II / f'f( f fi) ga(g, x)

sin x

dXded~vi

for a gas of

pseudc-Maxwellian

molecules

[ii

is

fif/fit

+ v

Vf

=

(go/4~) II1 f'f(

sin x

dXed~vi

gun

f, (I)

where

go/4~

=

ga(g, x)

= const, n =

d~v f.

We shall solve

equation (I) iteratively

with the use of the

following

scheme:

fi

fk+i/fit

+ gun

fk+i

"

(go/4~) II1 f( fk)(

sin x

dXded~vi

v V

fk

Similar method was

applied

in reference [2] to the linearized Boltzmann

equation.

The first iteration is

fi

fi/fit

+ gun

ii

=

(go/4~) / / / ii fo)(

sin x

dxded~vi

v V

fo

" gun

fo

v V

fo, (2)

(*)

Address for correspondence: Tashkentskaya St. 10-2-39, Moscow 109444, Russia.

(3)

230 JOURNAL DE PHYSIQUE I N°3

where

fo

"

n(2~kT/m)~~/~

exp

(-m(v u)~/2kT) (3)

n, u and T here and afterwards are real parameters of the gas. After introduction of

~a =

ii fo

we can rewrite

equation (2)

as follows:

fi~a/fit

+ gon~a = -fi

fo/fit

v V

fo

"

"

~ i~ ~

~ ~'~ ~

i~ ~~

~

~~~

~

~~~ ~~it~

~

~~ ~

~

+d (cc

~I)

Vu + c

(fl

~

Vln

j

,

(4)

kT 3 2kT 2

where

d/dt

=

alai

+ u V, c = v u.

In order to obtain the relaxation

equations

for the fluxes we

multiply equation (4) by

the

corresponding polynonfial

of molecular

velocity

and

integrate

over velocities. The

right-hand

side of the

resulting equation

is

expressed

in terms of

temporal

and

spatial

derivatives of the parameters of

fo,

I-e- n, u, T. These parameters do not contain any ~a contributions and hence

obey

Euler

equations

as follows from

equation (4).

This follows from

multiplication

of

equation (4) by I,

c, and c~ and

integration

over c.

Since the tensor of viscous stresses is

~a~ = m cac~~a

d~c

= m

/

vav~~a

d~v

/

we can obtain in this way

fi~a~ /fit

+ gon~a~ =

-(film) (p6a~

+

pttatt~) V~ (pttatt~tt~

+

ptt~6a~)

?fl (P"al

?a

(PUfl) (5)

Taking

into account the Euler

equations

tip/fit

+ T7

(pu)

=

0, pdu/df

+

VP

=

0, dp/df

+

(5/3)pT7

u = 0

we can derive from

equation (5)

that

fi~a~ /fit

= -pea~ gon~a~, ea~ =

Vatt~

+

V~tta (2/3)6a~V~tt~. (6)

If the left-hand side is

neglected

we receive the well-known linear Navier-Stokes law

~«~ =

-~oe«~/ngo

= -»e«~.

(6'j

The left-hand side describes flows for which the viscous stresses

change quickly. Equation (6)

has a formal solution

t t

~a~ =

~w

exp -go

~' n(r)dr (pea~),,

dt'

(7)

Here the

subscript

t' shows that the

hydrodynamical

values in

parentheses

should be calculated at the moment

t',

I.e. earlier than at t.

Equation (7)

is the sc-called transport

equation

with

(4)

delay (or

with

memory)

[3], but contrary to [2, 3] in the

present

note the memory kernel is

expressed explicitly.

We can do similar calculations to derive the relaxational

equation

for the heat flux

q =

(m/2) / cc~ f d~v.

In order to do this we

multiply equation (4) by (m/2)vav~

and

integrate

over

velocity. Taking

into account that

(m/2) /

vav~~a

d~v

= qa + u~~a~

we have the

equation

similar to the

preceding

one:

(~

+

go) (q«

+

u~~a~)

=

()pua

+

)pu~ua)

+

+$7~ ((~

~~ +

)u~) 6a~

+

(jp

+

)pu~) au~j

l'~

We

simplify

the

right-hand

side of this

equation using

the Euler

equations

as earlier:

()

=

-~pV« $

+

?VfIP

gonq«.

(8)

The solution of this

equation

can be

expressed

as the transport law with memory similar to

equation (7).

If we

neglect

the left-hand side we obtain in the case u

= 0 the classical Fourier law

go =

(5p/2gon) Va(kT/m)

= AVaT.

(9)

Notice that

A/pcv

=

5/3,

I-e- the Eucken relation [4]

A/pcv

=

5/2

is hence not valid for

pseudc-Maxwellian

molecules.

Expressions

similar to our

equations (6), (8)

were derived

by

another

approximate

method in reference [5], but the

equation

for qa obtained in that paper does not contain the term with

~a~.

Conclusion.

Relaxational

equations

for momentum and energy fluxes

generalizing

well-known Navier-Stokes and Fourier laws are obtained

using

the first iteration of the

proposed

scheme of solution of the Boltzmann

equation.

This short note appears to be another

proof

of the fact that these

equations

can be obtained from the Boltzmann

equation

with the same accuracy as the classical

Ones.

Acknowledgements.

Valuable discussions with Dr. A-G. Bashkirov and Dr. A-D. Khonkin are

gratefully appreciated.

The author thanks very much the referee for his

penetrating

criticism and very useful notes that

helped

the author to understand the

problem

more

deeply.

(5)

232 JOURNAL DE PHYSIQUE I N°3

References

[ii

I~OGAN M.N., Rarefied Gas Dynalrdcs

(Plenum,

N.Y.,

1969).

[2j ZUBAREV D-N- and KHONKIN A-D-, Theor. Math. Phys. II

(1972)

601.

[3j ZUBAREV D-1i, and TISHCHENKO S-V-, Physica 59

(1972)

285.

[4] FERzIGER J.H. and I~APER H-G-, Mathematical Theory of Transport Processes in Gases

(North-

Holland, Amsterdarr L.,

1972).

[5j KHONKIN A.D-, Fluid Mech. SOC. Res. 9

(1980)

93.

Références

Documents relatifs

They are entirely devoted to the initial value problem and the long-time behavior of solutions for the two-dimensional incompressible Navier–Stokes equations, in the particular

Under reasonable assumptions on the growth of the nonhomogeneous term, we obtain the asymptotic behavior of the solutions.. as t --~

§5, existence of very weak solution for the Navier-Stokes system is obtained, using a fixed point technique over the Oseen system, first for the case of small data and then

An essential feature of the studied scheme is that the (discrete) kinetic energy remains controlled. We show the compactness of approximate sequences of solutions thanks to a

The best result we know about (partial) regularity of weak solutions has been given in 1982 by Caffarelli, Kohn and Nirenberg [Ca, La].. Their result is based on the notion of

Keywords: Navier–Stokes equations; Uniqueness; Weak solution; Fourier localization; Losing derivative estimates.. In three dimensions, however, the question of regularity and

NIRENBERG, Partial regularity of suitable weak solutions of the Navier-Stokes equations. FRASCA, Morrey spaces and Hardy-Littlewood maximal

In Section 2, we prove a few orthogonality results concerning the Navier- Stokes equations with profiles as initial data, which will be used in the proof ofTheorem 2.. Section 3