Документ взят из кэша поисковой машины. Адрес оригинального документа : http://xray.sai.msu.ru/~mystery/articles/review/node36.html
Дата изменения: Wed Feb 26 14:32:33 1997
Дата индексирования: Tue Oct 2 14:43:38 2012
Кодировка:
Поисковые слова: р п р п п п п п п п п п п п п п п п

Principles of Scenario Machine The Scenario Machine: Operational Dependence on the Scenario

How to Catch ``Rare'' Tracks

The only way to define the origin of a ``rare'' binary system is to calculate evolutionary tracks of different binaries until one come across the evolutionary stage of interest. Note, that the term ``rare'' in a simulation experiment does not entirely refer to the rate of occurrence of the corresponding object in the Galaxy. The systems are rarely encountered in a real galaxy for two possible reasons:

1. binary systems are born frequently but spend little time at the selected stage (for example, SS 433-like  objects, common envelopes, etc.),
2. binary systems are born on rare occasions, but lives for a long period of time at the selected stage (X-ray pulsars,  X-ray novae, etc.).

Clearly, both rare-born and short-living binary systems are practically unobservable in real galaxies.

In the statistical experiment, the first factor can easily be taken into account. Problems start to arise when search is made for evolutionary tracks of the second type. If we possess some a priori information (or assumptions) about the initial volume in the parameter space where the studied binary came from, we can accelerate the search by generating the parameters only in the most probable region. Another way to shorten the time spent in looking for ``rare'' tracks is to use super-uniform quasi-random sequences (see Sobol', 1973[183]). Finding one of these tracks allows us to answer both questions as to what binary type the progenitor of the investigated system was and what kind of binary will result from it. Of course, the discovery of a track does not mean that the investigated binary cannot be formed in another way. However now we are able to study in more detail the occurrence rate of the detected track and to find limits of the initial parameter volume , and then, by generating the initial parameters only within these limits, we can estimate the occurrence rate and distributions upon parameters with a higher accuracy by using less extended calculations.

 
Figure 14: A cartoon scheme of the Scenario Machine 

Principles of Scenario Machine The Scenario Machine: Operational Dependence on the Scenario