Rambler's Top100Astronet    
  po tekstam   po klyuchevym slovam   v glossarii   po saitam   perevod   po katalogu
 

Na pervuyu stranicu << 1.2 Superpotencial Kaca-Krushciela ... | Oglavlenie | Literatura k lekcii 1 >>

Razdely


1.3 Postroenie sohranyayushihsya tokov i superpotencialov na proizvol'nom vspomogatel'nom fone. Problema edinstvennosti

Rassmotrim proizvol'nuyu kovariantnuyu teoriyu s lagranzhianom [20] (snachala on v neyavno kovariantnom vide): $\hat L = \hat L (A_B; A_{B,\alpha}; A_{B,\alpha\beta})$, gde AB -- eto nabor dinamicheskih peremennyh, proizvol'nyh tenzornyh plotnostei. Vneshnee zadannoe fonovoe prostranstvo-vremya vvedem prostoi tozhdestvennoi zamenoi: $A_{B,\tau} \equiv
\overline D_\tau A_B - \overline \Gamma^\sigma_{\tau\rho} \left. A_B \right\vert _\sigma^\rho$. Togda lagranzhian perepishetsya v yavno kovariantnoi forme:

\begin{displaymath}
\hat L =
{\hat{\cal L}}_c = {{\hat{\cal L}}}_c (A_B; \overline D_\alpha A_{B};
\overline D_\beta \overline D_\alpha A_{B}).
\end{displaymath} (1.36)

My sleduem Mickevichu [8], chtoby opredelit' proizvodnuyu Li:
\begin{displaymath}
{\pounds_\xi} A_B \equiv - \xi^\alpha \overline D_\alpha A_B...
...left.A_B\right\vert^\alpha_\beta
\overline D_\alpha \xi^\beta.
\end{displaymath} (1.37)

Poskol'ku lagranzhian (1.36) -- eto skalyarnaya plotnost', sostavlyaem dlya nego tozhdestvo Neter: ${\pounds_\xi} {\hat{\cal L}}_c +\partial_\alpha \left(\xi^\alpha {\hat{\cal L}}_c\right) \equiv 0$, kotoroe tozhdestvenno preobrazuetsya v
  - $\displaystyle \left[{{\delta {\hat{\cal L}}_c} \over {\delta A_B}} \overline D_...
...} \over {\delta A_B}} \left.A_B\right\vert^\beta_\alpha\right)\right]\xi^\alpha$  
  + $\displaystyle \overline D_\alpha \left[\hat u^\alpha_\sigma\xi^\sigma +
\hat m^...
...\beta}_\sigma \overline D_{\beta} \overline D_{\tau}\xi^\sigma\right] \equiv 0,$ (1.38)

gde koefficienty odnoznachno opredeleny lagranzhianom sleduyushim obrazom:
$\displaystyle \hat u^\alpha_\sigma$ $\textstyle \equiv$ $\displaystyle {\hat{\cal L}}_c \delta^\alpha_\sigma +
{{\delta {\hat{\cal L}}_c} \over {\delta A_B}} \left.A_B\right\vert^\alpha_\sigma$  
  - $\displaystyle \left[{{\partial {\hat{\cal L}}_c} \over {\partial (\overline D_\...
...ine D_\beta \overline D_\alpha A_{B})}}\right) \right]
\overline D_\sigma A_{B}$  
  - $\displaystyle {{\partial {\hat{\cal L}}_c} \over {\partial (\overline D_\beta \...
...A_{B}- \hat n^{\alpha\tau\beta}_\lambda
\overline R^\lambda_{ \tau\beta\sigma},$ (1.39)


$\displaystyle \hat m^{\alpha\tau}_\sigma$ $\textstyle \equiv$ $\displaystyle \left[{{\partial {\hat{\cal L}}_c} \over {\partial (\overline D_\...
...ta \overline D_\alpha A_{B})}}\right)\right]
\left.A_{B}\right\vert^\tau_\sigma$  
  - $\displaystyle {{\partial {\hat{\cal L}}_c} \over {\partial (\overline D_\tau \o...
...rline D_\alpha A_{B})}}
\overline D_\beta (\left.A_{B}\right\vert^\tau_\sigma),$ (1.40)


\begin{displaymath}
\hat n^{\alpha\tau\beta}_\sigma \equiv
{\textstyle{\frac{1}{...
...D_\alpha A_{B})}}
\left.A_{B}\right\vert^\beta_\sigma\right].
\end{displaymath} (1.41)

1.3.1 Obobshennyi tok

Podrobnyi analiz (1.38) daet drugoe sil'noe tozhdestvo:

\begin{displaymath}
{{\delta {\hat{\cal L}}_c} \over {\delta A_B}} \overline D_\...
...delta A_B}} \left.A_B\right\vert^\beta_\alpha\right) \equiv 0,
\end{displaymath} (1.42)

ispol'zovanie kotorogo v tozhdestve (1.38) privodit ego k vidu
\begin{displaymath}
\overline D_\alpha \left[\hat u^\alpha_\sigma\xi^\sigma +
\h...
...erline D_{\beta}
\overline D_{\tau}\xi^\sigma\right] \equiv 0.
\end{displaymath} (1.43)

Vyrazhenie pod divergenciei v etom tozhdestve vpolne mozhet byt' interpretirovano kak obobshennyi tok:
\begin{displaymath}
\hat i^\alpha \equiv - \left[\hat u^\alpha_\sigma\xi^\sigma ...
...sigma \overline D_{\beta} \overline D_{\tau}\xi^\sigma\right],
\end{displaymath} (1.44)

a samo tozhdestvo Neter (1.43) perepisyvaetsya v vide:
\begin{displaymath}
\overline D_\alpha \hat i^\alpha
\equiv \partial_\alpha \hat i^\alpha \equiv 0.
\end{displaymath} (1.45)

1.3.2 Obobshennyi superpotencial

Perepishem tozhdestvo (1.43) v vide:

\begin{displaymath}
\hat{\cal O}_\sigma\xi^\sigma + \hat{\cal O}^\alpha_\sigma \...
...ta}_\sigma \overline D_{(\alpha\tau\beta)}\xi^\sigma \equiv 0.
\end{displaymath} (1.46)

Poskol'ku komponenty $\xi^\mu$ i proizvodnye ot nih proizvol'ny v kazhdoi tochke, to koefficienty v (1.46) po otdel'nosti takzhe tozhdestvenno ravny nulyu. Eto daet nabor tozhdestv:
$\displaystyle \overline D_\alpha \hat u^\alpha_\sigma +
{\textstyle{\frac{1}{2}...
...\tau\beta}_\lambda \overline D_\beta
\overline R^{ \lambda}_{\sigma \tau\alpha}$ $\textstyle \equiv$ 0,  
$\displaystyle \hat u^\alpha_\sigma +
\overline D_\lambda \hat m^{\lambda \alpha...
...r 3}} \hat n^{\lambda\tau\beta}_\sigma
\overline R^{\alpha}_{ \tau\beta\lambda}$ $\textstyle \equiv$ 0,  
$\displaystyle \hat m^{(\alpha\tau)}_\sigma +
\overline D_\lambda \hat n^{\lambda(\alpha\tau)}_\sigma$ $\textstyle \equiv$ 0,  
$\displaystyle \hat n^{(\alpha\tau\beta)}_\sigma$ $\textstyle \equiv$ 0. (1.47)

Sama forma tozhdestva (1.45) govorit o tom, chto obobshennyi tok dolzhen vyrazhat'sya v vide:
\begin{displaymath}
\hat i^\alpha \equiv
\overline D_\tau \hat \Phi^{\alpha\tau}(\xi),
\end{displaymath} (1.48)

gde superpotencial v pravoi chasti tozhdestvenno udovletvoryaet: $\overline D_{\alpha\tau} \hat \Phi^{\alpha\tau}(\xi) \equiv 0$. Deistvitel'no, ispol'zuya tozhdestva (1.47) v opredelenii toka (1.44) my predstavlyaem ego v vide (1.48), gde obobshennyi superpotencial imeet vid:
\begin{displaymath}
\hat \Phi^{\alpha\tau} (\xi) \equiv
\left(\hat m^{\tau\alpha...
...n^{[\alpha\tau]\lambda}_\sigma \overline D_\lambda \xi^\sigma.
\end{displaymath} (1.49)

Uchet tret'ego iz tozhdestv (1.47) govorit, chto superpotencial (1.49) yavno antisimmetrichen.

1.3.3 Vklad ot divergencii v lagranzhiane

Prostym, no ochen' vazhnym yavlyaetsya vopros o vklade v tok i superpotencil ot divergencii v lagranzhiane. Otvet byl dan eshe Kacem [14]. Zdes' my kratko daem rezul'tat. Poskol'ku $ \Delta_{(div)} \hat L = \partial_\nu \hat k^\nu$ takzhe yavlyaetsya skalyarnoi plotnost'yu, to dlya etoi dobavki otdel'no vypolnyaetsya tozhdestvo Neter: $
\partial_\alpha ({\pounds_\xi} \hat k^\alpha + \xi^\alpha \partial_\nu \hat k^\nu ) \equiv 0 $, prostoi analiz kotorogo privodit k dobavkam v koefficientah (1.39) - (1.41) i v superpotenciale (1.49):

$\displaystyle \Delta_{(div)}\hat u^\alpha_\sigma$ = $\displaystyle 2 \overline D_\tau\left(\delta^{[\alpha}_\sigma \hat k^{\tau]} \right),$  
$\displaystyle \Delta_{(div)}\hat m^{\alpha\tau}_\sigma$ = $\displaystyle 2 \left(\delta^{[\alpha}_\sigma \hat k^{\tau]} \right),$  
$\displaystyle \Delta_{(div)}\hat n^{\alpha\tau\beta}_\sigma$ = 0,  
$\displaystyle \Delta_{(div)}\hat \Phi^{\alpha\tau}$ = $\displaystyle -2 \left(\xi^{[\alpha} \hat k^{\tau]} \right).$ (1.50)

Neobhodimo otmetit', chto forma etih dobavok nikak ne zavisit ot struktury vektornoi plotnosti $\hat k^{\tau}$!

1.3.4 Neopredelennost' i edinstvennost' v opredelenii tokov i superpotencialov

Horosho izvestno, chto bez izmeneniya tozhdestva (1.45) k toku $\hat i^\alpha$ mozhet byt' dobavlena proizvol'naya velichina $\Delta \hat i^\alpha(\xi)$, lish' by divergenciya ot nee tozhdestvenno obrashalas' v nul': $\partial_\alpha \left(\Delta \hat i^\alpha(\xi)\right) \equiv 0 $. Odnako, ,,isporchennyi'' tok tem zhe samym sposobom mozhet byt' ,,ispravlen''. Velichina $\Delta \hat i^\alpha(\xi)$ ne svyazana s lagranzhianom i proceduroi Neter. S drugoi storony, v opredelenii (1.44) vse koefficenty (1.39) - (1.41) strogo opredeleny lagranzhianom, zadannym v obshei (neyavnoi) forme. V formule (1.44) net ni odnogo chlena divergenciya ot kotorogo yavno by obrashalas' v nul'. Tozhdestvo (1.45) vypolnyaetsya kak vse vyrazhenie v celom, kak i dolzhno byt' dlya edinogo sohranyayushegosya toka Neter. Takim obrazom, mozhno sdelat' utverzhdenie:



Podobnye vyvody spravedlivy i dlya opredeleniya superpotenciala (1.49). V principe, bez izmeneniya toka $\hat i^\alpha$ k superpotencialu v (1.48) mozhet byt' dobavlena proizvol'naya velichina $\Delta \hat \Phi^{\alpha\tau}(\xi)$, lish' by divergenciya ot nee tozhdestvenno obrashalas' v nul': $\partial_\tau \left(\Delta \hat \Phi^{\alpha\tau}(\xi)\right) \equiv 0 $. Odnako, eta dobavka nikak ne svyazana s lagranzhianom i proceduroi Neter. Kak ee dobavili, tochno takzhe legko unichtozhit'. S drugoi storony, v opredelenii (1.49) vse chleny strogo opredeleny lagranzhianom i opisannoi proceduroi -- ih nevozmozhno otkinut'. Poetomu my delaem utverzhdenie:



Sdelav eto utverzhdenie, my, krome togo, proillyustriruem, chto dvusmyslennost' v superpotenciale ne opasna, vo vsyakom sluchae dlya postroeniya global'nyh sohranyayushihsya velichin. Tak, tozhdestvo $\partial_\tau \left(\Delta \hat \Phi^{\alpha\tau}(\xi)\right) \equiv 0 $ v svoyu ochered' govorit o tom, chto dobavka dolzhna byt' vyrazhena kak $\Delta \hat \Phi^{\alpha\beta}(\xi) \equiv
\partial_\gamma \hat j^{\alpha\beta\gamma}(\xi)$, gde $\partial_{\beta\gamma} \hat j^{\alpha\beta\gamma}(\xi) \equiv 0$. Sochetanie trebovaniya kovariantnosti i neobhodimosti ispol'zovat' chastnye proizvodnye dlya postroeniya zakonov sohraneniya i global'nyh velichin privodit k tomu, chto velichina $\hat j^{\alpha\beta\gamma}(\xi) =
\hat j^{[\alpha\beta\gamma]}(\xi)$, to est' dolzhna byt' antisimmetrichna po vsem indeksam. Teper' perepishem zakon sohraneniya (1.48) v izmenennom vide:

   
  $\displaystyle \hat i^\alpha \equiv \partial_\beta \left[\hat \Phi^{\alpha\beta}(\xi)+
\partial_\gamma \hat j^{\alpha\beta\gamma}(\xi)\right],$  

chto daet vozmozhnost' postroit' sohranyayushiesya velichiny analogichno (1.30):
   
  $\displaystyle {\cal P}(\xi) =
\int_{\Sigma} \hat i^0(\xi)d^3 x =
\oint_{\partia...
...[\hat \Phi^{[0n]}(\xi)+
\partial_m \left(\hat j^{[0nm]}(\xi)\right)\right]ds_n.$  

V silu teremy Stoksa poslednii chlen ne daet vklada v integral i, takim obrazom, ne izmenyaet velichinu ${\cal P}(\xi)$.

1.3.5 Edinstvennost' tokov i superpotencialov v opredelenii KBL

Vernemsya k voprosu b) v konce predydushei chasti o edinstvennosti velichin v opredelenii KBL. Prezhde vsego obsudim lagranzhian (1.27). Perepishem skalyarnuyu kriviznu v yavno ,,kovariantizovannom'' vide:

   
  $\displaystyle \hat R = \hat g^{\theta\sigma}\left(
\overline D_\rho \Delta^\rho...
...lta^\eta_{\theta\rho}\right)
+ \hat g^{\theta\sigma}\overline R_{\theta\sigma},$  

gde, napomnim,


$\hat k^\mu = \hat g^{\mu\rho} \Delta^\sigma_{\rho\sigma} -
\hat g^{\rho\sigma} \Delta^\mu_{\rho\sigma}$ i


$\Delta^\alpha_{\mu\nu} \equiv
\Gamma^\alpha_{\mu\nu} - \overline{\Gamma^\alpha_...
...ta\nu} +
\overline D_{\nu} g_{\beta\mu} -
\overline D_{\beta} g_{\mu\nu}\right)$.


S uchetom togo, chto my znaem kakoi vklad dayut divergencii (sm. (1.50)), lagranzhian (1.27) yavlyaetsya kak raz togo vida, chto i lagranzhian (1.36) v etoi chasti. Sledovatel'no, lagranzhian (1.27) mozhet byt' podstavlen v obshie formuly etoi chasti. Pryamaya podstanovka pokazyvaet, chto vse formuly KBL poluchayutsya etim putem. Tak, podstanovka (1.27) v obobshennyi tok (1.44) daet KBL tok (1.32). Bolee detal'no: Podstanovka v koefficient (1.39) i uchet uravnenii Einshteina daet tenzor energii-impul'sa (1.33). Podstanovka v (1.40) daet spinovyi koefficient (1.35). Podstanovka v (1.41) daet tochno Z-chlen v (1.32). (Neobhodimo tol'ko uchityvat' raznicu znakov v formulah (1.39) - (1.41) i v opredelenii toka (1.44).) Takoe zhe sootvetstvie i mezhdu superpotencialami: (1.49) perehodit tochno v (1.31). Iz skazannogo ostaetsya sdelat' vyvod:




<< 1.2 Superpotencial Kaca-Krushciela ... | Oglavlenie | Literatura k lekcii 1 >>

Publikacii s klyuchevymi slovami: zakony sohraneniya - Obshaya teoriya otnositel'nosti - gravitaciya
Publikacii so slovami: zakony sohraneniya - Obshaya teoriya otnositel'nosti - gravitaciya
Sm. takzhe:
Vse publikacii na tu zhe temu >>

Ocenka: 2.7 [golosov: 106]
 
O reitinge
Versiya dlya pechati Raspechatat'

Astrometriya - Astronomicheskie instrumenty - Astronomicheskoe obrazovanie - Astrofizika - Istoriya astronomii - Kosmonavtika, issledovanie kosmosa - Lyubitel'skaya astronomiya - Planety i Solnechnaya sistema - Solnce


Astronet | Nauchnaya set' | GAISh MGU | Poisk po MGU | O proekte | Avtoram

Kommentarii, voprosy? Pishite: info@astronet.ru ili syuda

Rambler's Top100 Yandeks citirovaniya