SIMULATION IN LIFE-CONTROL SYSTEMS
With the acceleration of the pace of economic development and the intensification of production processes, automation is increasingly introduced in enterprises: from the organizational management of shops and sectors to the management of technological processes for the production of various products. The most promising direction is the creation of flexible automated production facilities and production systems that enable the use of modern robotic complexes, numerical control machines, computer facilities to move quickly to produce new products, monitoring the dynamics of needs and the conjuncture of the world market. Management in such flexible systems can most effectively be implemented on the basis of local EMW networks, providing interaction and coordination of all information and computing resources for the management of individual units in the system and enabling real-time processing of information.
Features of real-time control.
The progress of the development of the national economy is now increasingly associated with the extent to which the accumulation, exchange and delivery of information to various users (administrative management personnel, designers and designers, researchers, service workers, etc.) is taking place. As a matter of fact, on the basis of modern means of computer technology and communication technology, a real "industry" is being created. production and consumption of information, which requires large information and computing resources and rapid access to them. The most promising is the consolidation of all information and computing resources with the help of digital integrated service networks, which allow the transfer of various types of information (operational, dialogue data and computer files, speech, television signals, etc.) in a single digital form. To effectively meet the requirements of various users for the quality and timeliness of information delivery, the management of integrated service networks should be implemented in real time.
There are other examples of systems that need to be managed in real time. All these systems are united by the fact that they belong to the class of large systems (see Chapter 1), which complicates the solution of the management tasks. When developing control systems for such objects, there is usually no a priori information about the conditions of their operation. This makes the construction of adaptive control systems promising (see § 9.2).
In the examples discussed in the previous paragraphs, the modeling method was used for the purposes of studying the characteristics of the systems & pound; in interaction with the external environment E, design (synthesis) of the structure, algorithms and parameters of the system, etc. In all these cases, as a rule, there were no strict limitations on the time between the start of the simulation and the obtaining of the result, the use of high-performance computers and GVK.
Forecasting and decision making.
The main purpose of modeling is forecasting in the broadest sense of the word. Modeling allows to draw a conclusion about the principle operability of an object (system 5), to evaluate its potentially possible characteristics, to establish the dependence of characteristics on various parameters and variables, to determine the optimal values of parameters, etc. Machine models L m Used as simulators and simulators, make it possible to predict the behavior of the system & pound; in conditions of interaction with the external environment Е.
Using the modeling method to obtain a forecast in making decisions in the real-time management system puts forward the task of fulfilling the constraint on the time resource of modeling the process of the functioning of the system 5. Therefore, we will consider in more detail the features of the prediction based on the machine model M < sub> n in real time.
To control an object, either information about the states (situations) of system 5 and the environment E y or information about the output characteristics (behavior) of system 5 in interaction can be used in the system with the external environment Е. This circumstance defines also the purposes of modeling. One case requires
estimate the changes in the states Zk ^ Z y to - 1, ng, during the prediction time t n ( we call such a simulation situational). In another
If you want to estimate the output characteristics y/e Y, 1, l y , on the time interval (0, T) (we call this modeling behavioral).
Thus, the goal of situational modeling is to obtain the prediction of the state vector r (/) (see 2.3)), and the goal of behavioral modeling is to estimate the vector of output characteristics y (0 [see (2.2) and (2.5)] For example, if the 0 <-circuit y is used as the conceptual model of the x then in case of situational modeling it is required to predict such states as the number of applications in storage devices, the number of occupied channels, etc., and in case of behavioral modeling, in this case it is necessary to estimate such as the probability of losing an application, the average delay time of an application in the system, etc. In accordance with the goals of situational and behavioral modeling, the approach to the development and implementation of modeling algorithms should differ, although the principles of their construction (the "A /" principle and principle dg ) are saved.
Another feature of modeling for making decisions on the management of an object in real time is the substantial limitation of computational resources, since such control systems, and, consequently, machine models A/ m , are realized, as a rule , on the basis of mini- and micro-computers or specialized microprocessor sets, when there is a limitation on the speed and the amount of memory. This requires a careful approach to minimizing the cost of resources for real-time modeling [12, 29, 52].
In addition, it should be borne in mind that the reliability and accuracy of the solution of the problem of modeling (prediction of the situation or behavior) of the system essentially depend on the number of realizations N, which are spent on obtaining a statistical forecast (see Chapter 7) . Thus, there is the problem of finding a compromise between the need to increase the time spent on modeling, i.e., the number of implementations N [on the interval (0, T )] to improve accuracy and reliability the results of modeling (forecasting), and the need to reduce the expenditure of computer time from the control conditions in real time.
Using the computer model M m in the system management loop 5 in real time, there is also a problem of prompt updating of information both in the object database and in the database about the experiment, ie, in this case about a specific forecast.
Let's consider in more detail the possibilities of constructing modeling algorithms for situational and behavioral models. In case of situational modeling, it is important not to lose information about the change of states of the system 5, since the effectiveness of control depends on this. Therefore, the construction of deterministic modeling algorithms, when the A principle is used, results in either an increase in modeling time with decreasing A /, or a decrease in the reliability of the state forecast with increasing A /. This speaks in favor of the use of stochastic algorithms, namely those that are most simply implemented, that is, asynchronous sporadic algorithms.
In behavioral modeling, it is important to obtain an average statistical estimate of the characteristics of the system S on the interval (0, T). Therefore, in the construction of simulation algorithms, it is important to choose the simplest realizable algorithm that requires minimal time and RAM for its run. In this case, both stochastic and deterministic modeling algorithms can be effective. The choice of the principle of constructing a modeling algorithm for decision-making in the control system can be carried out only taking into account the specific features of the concrete 5.
From the point of view of programming models M m for real-time modeling, there are also a number of features. This is primarily due to the lack or inability to use YON and YAM for software implementation of models based on the capabilities of the software of mini- and micro-computers and strict limitations on the counting time using the modeling algorithm. In this case, the main application is to find low-level languages, which complicates the process of developing simulation software in real time, but usually allows you to obtain sufficiently effective simulation work programs. To accelerate the process of developing simulation software in real time and improving its quality, it is rational to develop appropriate application packages that generate simulation work programs using high-performance computing resources.
Thus, modeling the process of functioning of systems for real-time control purposes has a number of specific features, but the modeling technique and the principles of implementing modeling algorithms are preserved.
Also We Can Offer!
- Argumentative essay
- Best college essays
- Buy custom essays online
- Buy essay online
- Cheap essay
- Cheap essay writing service
- Cheap writing service
- College essay
- College essay introduction
- College essay writing service
- Compare and contrast essay
- Custom essay
- Custom essay writing service
- Custom essays writing services
- Death penalty essay
- Do my essay
- Essay about love
- Essay about yourself
- Essay help
- Essay writing help
- Essay writing service reviews
- Essays online
- Fast food essay
- George orwell essays
- Human rights essay
- Narrative essay
- Pay to write essay
- Personal essay for college
- Personal narrative essay
- Persuasive writing
- Write my essay
- Write my essay for me cheap
- Writing a scholarship essay