The application simulated modelling in materials science

Abstract

Formulation of the problem. An important part of the experiment is to obtain a random number sequence with specific characteristics. To simulate any predetermined random process must be able to build enough economical sequence of random numbers corresponding to certain fixed laws of distribution. To obtain the value of the random variable with given distribution, typically use one or more values are uniformly distributed random numbers. Therefore the question of the development of methods for obtaining a uniformly distributed random numbers on a computer has a special meaning, a significant effect on the use of simulation in the practice of solving problems with a large number of variables. Objective. The definition of convergence model with reality. Main part. Model billiard problem allows an experiment aimed at the study of dissipative systems with chaotic layer in the intermediate region, which may occur under certain conditions. At the same time, to give the "most chaotic", in terms of experience has enabled the possibility of job strain sides of billiards by permanent changes in their ellipsoidal, and the constant movement of the balls-reflectors along a predetermined path. The model describing such systems is sensitive to the initial data, and the two close paths over time, which will be removed from each other. As compared to this case, the results of theory and experiment? Small inaccuracies in the determination of the initial state of the system leads to the fact that the difference between the trajectory predicted by theory and experimental data will increase. The reason for this phenomenon is not disadvantages of the model and the nature of the phenomena studied. Conclusions. The way out of this situation is not a comparison of the trajectory of the image point of the object model and at the same times, and some of the more complex characteristics that determine the intrinsic properties of the processes under study.