Architecture of modeling languages., Requirements...

Architecture of modeling languages.

The architecture of the YAM Y ie the concept of the interrelationships of language elements as a complex system, and the technology of transition from system 5 to its machine model M m can be represented as follows: 1) modeling objects ( S systems) are described (displayed in the language) using some language attributes; 2) the attributes interact with processes that are adequate to actual phenomena in the simulated system 5; 3) processes require specific conditions that determine the logical basis and sequence of interaction of these processes in time; 4) the conditions affect the events taking place within the simulation object (the S system) and when interacting with the external environment & pound ;; 5) events change the states of the model of the system M in space and time.

A typical scheme of the architecture of the nuclear power plant and the technology of its use in modeling systems is shown in Fig. 5.1.

Fig. 5.1. A typical scheme of the architects of the Institute of Nuclear Physics and the technology of its use

In most cases, with the help of computer models, the characteristics and behavior of system 5 are investigated on a certain time interval, therefore one of the most important tasks in creating the model of the system and choosing the programming language of the model is the realization of two functions: 1) the correction of the time coordinate of the state of the system ( ; promotion of "time, organization of hours"); 2) ensuring the consistency of various blocks and events in the system (synchronization in time, coordination with other blocks).

Thus, the functioning of the M m model should take place in an artificial time (not in real and not in machine), ensuring the occurrence of events in the order and with appropriate time intervals between them. It should be taken into account that the elements of the real system 5 function simultaneously (in parallel), and the components of the machine model M m act sequentially, as they are realized with the help of a sequential computer. Since events can occur simultaneously in different parts of the modeling object, in order to maintain the adequacy of cause-effect time relationships, it is necessary to create a "mechanism" in I AM. set the time for synchronization of the actions of the elements of the system model [17, 46].

Setting the time in the machine model. As already noted in Ch. 3, there are two basic approaches to time assignment: using constant and variable time intervals, which correspond to two principles of implementing modeling algorithms, ie, the "A/ and the " 3g" principle.

Consider the appropriate methods of time management in the model of the system M (5) using the example shown in Fig. 5.2, where the sequence of events in the system {^} is postponed along the real-time axis, and events and 5 occur simultaneously (Figure 5.2, a). Under the action of events the state of the model changes at the instant of time and such a change occurs abruptly bz.

In a model built on the principle of (Figure 5.2, b), the moments of the system

time will consistently take the values ​​= A /,/' * 2D /, /' S /, 1 ^ -4Dr, - 5Dr. These system time moments/ y '(D /) are in no way related to the moments of occurrence of the .r/events, which are simulated in the system model. The system time thus receives a constant increment, selected and set before the start of the simulation experiment.

In the model constructed according to the principle 5m (Figure 5.2, c), the time change occurs at the moment of changing the state of the system, and the sequence of moments of the system time has the form *; - Tx2, | ; -1 A C • ** •

-/ X 5, i.e., the moments of the system time f '(k pound; r) are directly related to the moments of occurrence of events in the system 1 /

Fig. 5.2. Methods of time management in the system model

Each of these methods has its advantages in terms of adequate reflection of real events in system 5 and the cost of machine resources for modeling. Using the principle 5d , events are processed sequentially and the time is shifted each time forward until the next event begins. In a model built on the "A /" principle, event processing occurs in groups, packets, or sets of events. In this case, the choice of A/exerts a significant influence on the course of the process and the results of the simulation, and if A/is not specified correctly, the results can turn out to be unreliable, since all events appear at the point corresponding to the upper boundary of each simulation interval. When applying the principle of 5d , simultaneous processing of events in the model takes place only when these events appear simultaneously in the real system. This avoids the need to artificially introduce the ranking of events when they are processed at the end of the interval A /.

If you are modeling on the A we can achieve a good approximation: for this, Ar must be small so that two non-simultaneous events do not fall into the same time interval. But a decrease in Ar leads to an increase in the expenditure of computer time for modeling, since a significant portion is spent on adjusting the "hours" and tracking of events, which in most intervals may not be. In this case, even with a strong compression A/two non-simultaneous events can fall into the same time interval Ar, which creates a false impression of their simultaneity.

To choose the principle of constructing the machine model M m and, respectively, the NIM, it is necessary to know: the purpose and purpose of the model; the required accuracy of simulation results; the expenditure of computer time when using a particular principle; the necessary amount of machine memory for the realization of the model constructed according to the principle of Ar and bgr y the complexity of programming the model and its debugging.

Requirements for simulation languages.

Thus, a number of specific difficulties arise in the development of system models, therefore, a set of software tools and concepts that do not occur in ordinary NON should be provided in the NAM.

Combination. In parallel, the processes occurring in real systems 5 are represented using a computer that runs sequentially. The modeling languages ​​make it possible to bypass this difficulty by introducing the concept of system time used to represent time-ordered events.

Size. Most simulated systems have a complex structure and behavioral algorithms, and their models are large in volume. Therefore, the dynamic memory allocation is used when the components of the model of the system M m appear in the main memory of the computer or leave it depending on the current state. An important aspect of the realizability of the M m model on a computer in this case is the blockiness of its construction, i.e., the possibility of splitting the model into blocks, subblocks, etc.

Changes. Dynamic systems are motion-related and characterized by the development of the process, resulting in a spatial configuration of these systems undergoing changes over time. Therefore, in all nuclear weapons, the processing of lists reflecting changes in the states of the functioning of the system being simulated is envisaged. 5

Interconnection. The conditions necessary to accomplish various events in the M and model of the functioning of system 5 can be very complex due to the presence a large number of mutual relationships between the components of the model. To resolve the related issues, most YAMs include the appropriate logical possibilities and concepts of set theory.

Stochasticity. To model random events and processes, special programs are used to generate sequences of pseudo-random numbers that are quasi- uniformly distributed over a given interval, on the basis of which stochastic effects on the model D/ m simulated by random variables with the corresponding distribution law can be obtained.

Analysis. To obtain a clear and convenient answer to the questions solved by the method of computer simulation, it is necessary to obtain statistical characteristics of the process of functioning of the model of the system A/(5). Therefore, methods of statistical processing and analysis of simulation results are provided in modeling languages.

The following requirements for the study and design of various systems S correspond to the most well-known modeling languages ​​for discrete events, such as SIMULA, SIMSCRIPT, GPSS, SOL, CSL , etc.

thematic pictures

Also We Can Offer!

Other services that we offer

If you don’t see the necessary subject, paper type, or topic in our list of available services and examples, don’t worry! We have a number of other academic disciplines to suit the needs of anyone who visits this website looking for help.

How to ...

We made your life easier with putting together a big number of articles and guidelines on how to plan and write different types of assignments (Essay, Research Paper, Dissertation etc)