Selected or created computational algorithms, as well as the introduction of
simplifying assumptions into the model, should not substantially distort the
basic properties of the object under study, and they should be adaptable to the
peculiarities of the tasks being solved and the computational tools used;
– creation of software for the implementation of the model and algorithm on the
computer. Software must take into account the specifics of mathematical
modeling associated with the use of a hierarchy of mathematical models and
multivariate calculations. This fact implies the possibility of using software
packages developed, in particular, using object-oriented programming;
– carrying out a computational experiment, which allows to obtain systematic
calculated results determined by various input parameters of the problem;
– processing of calculation results, their analysis, comparison (if possible) with the
results of a full-scale experiment, formulation of conclusions, and recommendations. It should be noted that at this stage it may be necessary to clarify the
mathematical model in order to obtain an adequate practical result.
The methodological universality of computer modeling makes it possible, on the
basis of accumulated experience in the development of mathematical models,
numerical methods, computational algorithms, and means of analyzing the results
obtained, to quickly and effectively solve various applied problems.
In practice, for the study of complex multicomponent systems, a computational
experiment is often carried out using specialized software packages. When solving
problems of a specific domain, certain requirements arise for the packages used.
The textbook discusses the modeling of dynamic systems (class of tasks related
to the macro-level). We study problems whose mathematical models are described
by differential and differential–algebraic equations. To solve this class of problems,
computer math packages such as MATLAB, Simulink [1], MapleSim [2], Rand
Model Designer [3], ISMA [4], and Wolfram SystemModeler [5] can be used.
Each of these packages, having powerful functionality, including visual modeling tools, can best solve specialized applied tasks of a certain type. In this regard,
when solving a specific engineering problem, the question arises which of the
packages of computer mathematics would be more effective for studying a particular problem. To answer this question, you must either have experience with each
of the packages or have information about the features of these packages. There are
works devoted to the comparative analysis of visual modeling tools, thanks to
which the user can decide on the choice of a tool according to his needs.
In the textbook, the Wolfram SystemModeler software environment [5] is
considered as a tool for computer modeling. A review of the functionality of the
package is made, its specificity is discussed in comparison with other packages of
computer mathematics, and the possibilities of the package for the solution of
applied engineering problems of mechanics are considered.
Wolfram SystemModeler is an interactive graphical environment designed for
mathematical and computer simulation of multi-dimensional systems using the
Modelica language.
x
Introduction
Précédent

- 9/274

Suivant