Web-Books
im Austria-Forum
Austria-Forum
Web-Books
Naturwissenschaften
Physik
Differential Geometrical Theory of Statistics
Seite - 29 -
  • Benutzer
  • Version
    • Vollversion
    • Textversion
  • Sprache
    • Deutsch
    • English - Englisch

Seite - 29 - in Differential Geometrical Theory of Statistics

Bild der Seite - 29 -

Bild der Seite - 29 - in Differential Geometrical Theory of Statistics

Text der Seite - 29 -

Entropy2016,18, 370 configuration space is a smoothmanifoldN, described in theLagrangian formalismbya smooth time-independenthyper-regularLagarangianL :TN→Ror, in theHamiltonianformalism,bythe associatedHamiltonianHL : T∗N→R. Let t → −−→ x(t)beamotionof that system,−→x0 = −−→ x(t0) and−→x1 = −−→ x(t0)betheconfigurationsof thesystemfor thatmotionat times t0 and t1. Thereexistsanother motion t →−−→x′(t)of thesystemforwhich−−−→x′(t0)=−→x1 and −−−→ x′(t1)=−→x0: since theequationsofmotion are invariantby timereversal, themotion t →−−→x′(t) isobtainedsimplybytakingas initial conditionat time t0 −−−→ x′(t0)= −−→ x(t1)and d −−→ x′(t) dt ∣∣∣ t=t0 =−d −−→ x(t) dt ∣∣∣ t=t1 .Anothermoreseriousargumentagainstakind of thermodynamicbehaviourofLagarangianorHamiltoniansystemsrestsonthe famousrecurrence theoremduetoPoincaré [51]. This theoremasserts indeedthatwhentheusefulpartof thephasespace of thesystemisofafinite totalmeasure,almostallpoints inanarbitrarilysmallopensubsetof the phasespaceare recurrent, i.e., themotionstartingof suchapointat time t0 repeatedlycrosses that opensubsetagainandagain, infinitelymanytimeswhen t→+∞. Letusnowconsider, insteadofperfectlydefinedstates, i.e.,points inphasespace,statisticalstates, andaskthequestion:Whenat time t= t0 aHamiltoniansystemonasymplecticmanifold (M,ω) is in astatisticalstategivenbysomeprobabilitymeasureofdensityρ0withrespect to theLiouvillemeasure λω,does its statistical stateconverge,when t→+∞, towards theprobabilitymeasureofaGibbsstate? This question shouldbemademoreprecise by specifyingwhatphysicalmeaninghas a statistical stateandinwhatmathematical senseastatistical statecanconverge towards theprobabilitymeasure ofaGibbsstate. Apositivepartial answerwasgivenbyLudwigBoltzmannwhen,developinghis kinetic theoryofgases,heprovedhis famous(butcontroversed)Êta theoremstatingthat theentropy of thestatistical stateofagasofsmallparticles isamonotonously increasingfunctionof time. This question, linkedwithtimeirreversibility inphysics, is still thesubjectof important researches,both byphysicists andbymathematicians. The reader is referred to thepaper [50]byBalian foramore thoroughdiscussionof thatquestion. 6.3. ExamplesofThermodynamicEquilibria 6.3.1.ClassicalMonoatomic IdealGas Inclassicalmechanics,adilutegascontainedinavesselat rest inaGalileanreference frameis mathematicallydescribedbyaHamiltoniansystemmadebya largenumberofverysmallmassive particles,which interactbyverybrief collisionsbetweenthemselvesorwith thewallsof thevessel, whosemotionsbetweentwocollisionsare free. Letusfirstassumethat theseparticlesarematerial points and that no external field is acting on them, other than that describing the interactions by collisionswith thewallsof thevessel. TheHamiltonianofoneparticle inapartof thephasespace inwhich itsmotion is free is simply 1 2m ‖−→p‖2= 1 2m (p21+p 2 2+p 2 3) , with −→p =m−→v , wherem is themassof theparticle,−→v itsvelocityvectorand−→p its linearmomentumvector (in the consideredGalileanreference frame), p1, p2 and p3 thecomponentsof −→p inafixedorhtonormalbasis of thephysical space. LetNbethe totalnumberofparticles,whichmaynothaveall thesamemass.Weusea integer i∈{1, 2, . . . , N} to label theparticlesanddenotebymi,−→xi ,−→vi ,−→pi themassandthevectorsposition, velocityandlinearmomentumof the i-thparticle. TheHamiltonianof thegas is therefore H= N ∑ i=1 1 2mi ‖−→pi‖2+ terms involvingthecollisionsbetweenparticlesandwith thewalls. 29
zurück zum  Buch Differential Geometrical Theory of Statistics"
Differential Geometrical Theory of Statistics
Titel
Differential Geometrical Theory of Statistics
Autoren
Frédéric Barbaresco
Frank Nielsen
Herausgeber
MDPI
Ort
Basel
Datum
2017
Sprache
englisch
Lizenz
CC BY-NC-ND 4.0
ISBN
978-3-03842-425-3
Abmessungen
17.0 x 24.4 cm
Seiten
476
Schlagwörter
Entropy, Coding Theory, Maximum entropy, Information geometry, Computational Information Geometry, Hessian Geometry, Divergence Geometry, Information topology, Cohomology, Shape Space, Statistical physics, Thermodynamics
Kategorien
Naturwissenschaften Physik
Web-Books
Bibliothek
Datenschutz
Impressum
Austria-Forum
Austria-Forum
Web-Books
Differential Geometrical Theory of Statistics