Computer simulation or physical testing, which is better? Computer modeling

Let's start with the definition of the word modeling.

Modeling is the process of constructing and using a model. A model is understood as a material or abstract object that, in the process of study, replaces the original object, preserving its properties that are important for this study.

Computer modeling as a method of cognition is based on mathematical modeling. A mathematical model is a system of mathematical relationships (formulas, equations, inequalities and signed logical expressions) that reflect the essential properties of the object or phenomenon being studied.

It is very rarely possible to use a mathematical model for specific calculations without using computer technology, which inevitably requires the creation of some computer model.

Let's consider the process computer modeling in more detail.

2.2. Introduction to Computer Modeling

Computer modeling is one of the effective methods studying complex systems. Computer models are easier and more convenient to study due to their ability to conduct computational experiments in cases where real experiments are difficult due to financial or physical obstacles or may give unpredictable results. The logic of computer models makes it possible to identify the main factors that determine the properties of the original object under study (or an entire class of objects), in particular, to study the response of the simulated physical system to changes in its parameters and initial conditions.

Computer modeling how new method scientific research is based on:

1. Construction of mathematical models to describe the processes being studied;

2. Using the latest computers, with high performance (millions of operations per second) and capable of conducting a dialogue with a person.

Distinguish analytical And imitation modeling. In analytical modeling, mathematical (abstract) models of a real object are studied in the form of algebraic, differential and other equations, as well as those involving the implementation of an unambiguous computational procedure leading to their exact solution. In simulation modeling, mathematical models are studied in the form of an algorithm that reproduces the functioning of the system under study by sequentially executing large quantity elementary operations.

2.3. Building a computer model

The construction of a computer model is based on abstraction from the specific nature of phenomena or the original object being studied and consists of two stages - first creating a qualitative and then a quantitative model. Computer modeling consists of conducting a series of computational experiments on a computer, the purpose of which is to analyze, interpret and compare the modeling results with the real behavior of the object under study and, if necessary, subsequent refinement of the model, etc.

So, The main stages of computer modeling include:

1. Statement of the problem, definition of the modeling object:

on at this stage information is collected, a question is formulated, goals are defined, forms for presenting results, and data is described.

2. System analysis and research:

system analysis, meaningful description of the object, development information model, analysis of technical and software, data structure design, development mathematical model.

3. Formalization, that is, the transition to a mathematical model, the creation of an algorithm:

choosing a method for designing an algorithm, choosing a form for writing an algorithm, choosing a testing method, designing an algorithm.

4. Programming:

choosing a programming language or application environment for modeling, clarifying ways to organize data, writing an algorithm in the selected programming language (or in an application environment).

5. Conducting a series of computational experiments:

debugging of syntax, semantics and logical structure, test calculations and analysis of test results, program modification.

6. Analysis and interpretation of results:

modification of the program or model if necessary.

There are many software packages and environments that allow you to build and study models:

Graphics environments

Text editors

Programming environments

Spreadsheets

Math packages

HTML editors

2.4. Computational experiment

An experiment is an experience that is performed with an object or model. It consists of performing certain actions to determine how the experimental sample reacts to these actions. A computational experiment involves carrying out calculations using a formalized model.

Using a computer model that implements a mathematical one is similar to conducting experiments with a real object, only instead of a real experiment with an object, a computational experiment is carried out with its model. By specifying a specific set of values of the initial parameters of the model, as a result of a computational experiment, a specific set of values of the required parameters is obtained, the properties of objects or processes are studied, and they are found optimal parameters and operating modes, specify the model. For example, having an equation that describes the course of a particular process, you can, by changing its coefficients, initial and boundary conditions, study how the object will behave. Moreover, it is possible to predict the behavior of an object in different conditions. To study the behavior of an object with a new set of initial data, it is necessary to conduct a new computational experiment.

To check the adequacy of the mathematical model and the real object, process or system, the results of computer research are compared with the results of an experiment on a prototype full-scale model. The test results are used to adjust the mathematical model or the question of the applicability of the constructed mathematical model to design or research is resolved given objects, processes or systems.

A computational experiment allows you to replace an expensive full-scale experiment with computer calculations. It allows, in a short time and without significant material costs, to study a large number of options for a designed object or process for various modes of its operation, which significantly reduces the time required for the development of complex systems and their implementation in production.

2.5. Simulation in various environments

2.5.1. Simulation in a programming environment

Modeling in a programming environment includes the main stages of computer modeling. At the stage of building an information model and algorithm, it is necessary to determine which quantities are input parameters and which are results, and also determine the type of these quantities. If necessary, an algorithm is drawn up in the form of a block diagram, which is written in the selected programming language. After this, a computational experiment is carried out. To do this, you need to download the program to RAM computer and run it. A computer experiment necessarily includes an analysis of the results obtained, on the basis of which all stages of solving the problem (mathematical model, algorithm, program) can be adjusted. One of the most important stages is testing the algorithm and program.

Debugging a program (the English term debugging means “catching bugs” appeared in 1945, when electrical circuits one of the first Mark-1 computers was hit by a moth and blocked one of the thousands of relays) - this is the process of finding and eliminating errors in the program, carried out based on the results of a computational experiment. During debugging, localization and elimination occurs syntax errors and obvious coding errors.

In modern software systems debugging is carried out using special software tools called debuggers.

Testing is checking the correct operation of the program as a whole or its components. The testing process checks the functionality of the program and does not contain obvious errors.

No matter how carefully the program is debugged, the decisive stage that establishes its suitability for work is monitoring the program based on the results of its execution on the test system. A program can be considered correct if, for the selected system of test input data, correct results are obtained in all cases.

2.5.2. Modeling in Spreadsheets

Modeling in spreadsheets covers a very wide class of problems in different subject areas. Spreadsheets are a universal tool that allows you to quickly perform labor-intensive work of calculation and recalculation quantitative characteristics object. When modeling using spreadsheets, the algorithm for solving the problem is somewhat transformed, hiding behind the need to develop a computing interface. The debugging stage is retained, including the elimination of data errors in connections between cells and in computational formulas. Additional tasks also arise: work on the convenience of presentation on the screen and, if it is necessary to output the received data on paper, on their placement on sheets.

The spreadsheet modeling process is performed using general scheme: goals are defined, characteristics and relationships are identified, and a mathematical model is compiled. The characteristics of the model are necessarily determined by purpose: initial (affecting the behavior of the model), intermediate, and what is required to be obtained as a result. Sometimes the representation of an object is supplemented with diagrams and drawings.

To visually display the dependence of the calculation results on the initial data, charts and graphs are used.

Testing uses a certain set of data for which the exact or approximate result is known. The experiment consists of introducing input data that satisfies the modeling goals. Analysis of the model will make it possible to find out how well the calculations meet the modeling goals.

2.5.3. Modeling in a DBMS environment

Modeling in a DBMS environment usually pursues the following goals:

Storing information and editing it in a timely manner;

Organizing data according to certain criteria;

Creation of various data selection criteria;

Convenient presentation of selected information.

In the process of developing the model, the structure of the future database is formed based on the initial data. The described characteristics and their types are summarized in a table. The number of table columns is determined by the number of object parameters (table fields). The number of rows (table records) corresponds to the number of rows of described objects of the same type. A real database may have not one, but several tables interconnected. These tables describe the objects included in a certain system. After defining and specifying the database structure in computer environment proceed to filling it.

During the experiment, data is sorted, searched and filtered, and calculation fields are created.

A computer information panel provides the ability to create various screen forms and forms for displaying information in printed form - reports. Each report contains information relevant to the purpose of the particular experiment. It allows you to group information according to specified characteristics, in any order, with the introduction of final calculation fields.

If the results obtained do not correspond to the planned ones, you can conduct additional experiments by changing the conditions for sorting and searching for data. If there is a need to change the database, you can adjust its structure: change, add and delete fields. The result is a new model.

2.6. Using a computer model

Computer modeling and computational experiment as a new method scientific research forces to improve the mathematical apparatus used in constructing mathematical models, allows, using mathematical methods, clarify, complicate mathematical models. The most promising for conducting a computational experiment is its use for solving major scientific, technical and socio-economic problems of our time, such as the design of reactors for nuclear power plants, the design of dams and hydroelectric power plants, magnetohydrodynamic energy converters, and in the field of economics - drawing up a balanced plan for the industry, region, country, etc.

In some processes where a natural experiment is dangerous to human life and health, a computational experiment is the only possible one (thermonuclear fusion, space exploration, design and research of chemical and other industries).

2.7. Conclusion

In conclusion, it can be emphasized that computer modeling and computational experiment make it possible to reduce the study of a “non-mathematical” object to a solution mathematical problem. This opens up the possibility of using a well-developed mathematical apparatus combined with powerful computer technology. This is the basis for the use of mathematics and computers to understand the laws of the real world and use them in practice.

3. List of references used

1. S. N. Kolupaeva. Mathematical and computer modeling. Study guide. – Tomsk, School University, 2008. – 208 p.

2. A. V. Mogilev, N. I. Pak, E. K. Henner. Informatics. Study guide. – M.: Center “Academy”, 2000. – 816 p.

3. D. A. Poselov. Informatics. Encyclopedic Dictionary. – M.: Pedagogika-Press, 1994. 648 p.

4. Official website of the publishing house "Open Systems". Internet University of Information Technologies. – Access mode: http://www.intuit.ru/. Date of access: October 5, 2010

There is absolutely no doubt that computer modeling of various physical processes significantly accelerated the process of developing technical products, while saving developers a lot of money on assembling test models. With the help of modern computing power and software, engineers can simulate the operation of individual components and assemblies of complex systems, which will reduce the number of physical tests required before launching a new product. Manufacturers can also calculate the cost of development after CAD modeling, rather than waiting until the end of physical testing of the product.

Modern industry, when launching new products, faces problems such as time to develop a new product and development costs. And in the automotive and aerospace industries it is almost impossible to do without CAD modeling, since modeling helps to significantly speed up development and reduce costs, which is very important for modern market. Historically, the emergence of modern computing systems, which are capable of simulating the dynamic properties of objects under various influences, has pushed into the background the modernization of physical test stands, as well as the development of test methods. Many organizations try to choose modeling because it requires minimal cost and minimal development time. However, in some studies, only the process of physically testing the product can provide an accurate answer. Without greater interaction between electronic models and physical testing, many organizations may become overly dependent on computer models for development, which, if used incorrectly, can subsequently lead to unexpected failures in expensive equipment.

In the automotive industry, computer modeling is becoming an integral part as modern vehicle designs have become much more complex and computer modeling systems have improved significantly. However, unfortunately, many manufacturers reduce physical testing of products to a minimum, relying on computer simulation results.

Physical testing processes have not kept pace with computer modeling in improving techniques. Testing engineers usually try to perform the minimum necessary tests on a product. The result is more frequent test repetitions to obtain more reliable results or their confirmation. Relying purely on computer modeling without physical testing can lead to very serious consequences in the future, since the mathematical model of the product, on the basis of which the process of calculating dynamic properties is carried out, is created with certain assumptions, and real work The product may behave slightly differently than what appears on your monitor.

Computer modeling has a symbiotic relationship with physical testing of equipment, which allows (unlike a computer model) to obtain experimental data. Therefore, lags in technologies for testing finished devices, with such an increase in the capabilities of computer technology, can lead to unnecessary savings on experimental samples with subsequent problems in finished products. The accuracy of the models directly depends on the input data about the behavior of the model (mathematical description) under various conditions.

Of course, model elements cannot include all possible options and the conditions for the behavior of certain components, since the complexity of calculations and the cumbersomeness of the mathematical model would become simply enormous. To simplify the mathematical model, certain assumptions are made that “should not” have a significant impact on the operation of the mechanism. But, unfortunately, the reality is always much harsher. For example, a mathematical model will not be able to calculate how the device will behave if there are microcracks in the material, or if there is a sudden change in weather, which can lead to a completely different load distribution in the structure. Experimental data and calculated data quite often differ from each other. And this must be remembered.

There is another important advantage to the physical testing of equipment. This is the ability to point out flaws to engineers when drawing up mathematical models, and also provides a good opportunity for discovering new phenomena and improving old calculation methods. After all, you must agree that if you put variables into a mathematical formula, the result will depend on the variables, and not on the formula. The formula will always remain constant, and only a real physical test can supplement or change it.

The emergence of new materials in all branches of modern industry creates additional problems for computer modeling. If engineers continued to use time-tested materials and improve them mathematical descriptions then yes, the problems with modeling would be much less. But the emergence of new materials requires mandatory carry out physical tests of finished products with these materials. However, new elements are increasingly appearing on the market and growth trends are only going up.

For example, the aeromobile and automotive industries have rapidly adopted composite materials due to their good strength-to-weight ratio. One of the main problems with computer modeling is the inability of the model to accurately predict the behavior of a material that suffers from certain performance disadvantages compared to the aluminum, steel, plastic and other materials that have long been used in this industry.

Validation of computer models for composite materials is critical during the design phase. After carrying out the calculations, it is necessary to assemble a test stand on a real part. When performing physical tests to measure deformation and load distribution, engineers focus on critical points determined by a computer model. Strain gauges are used to collect information about critical points. This process is only monitored for expected issues that may create blind spots in the testing process. Without comprehensive research, a model may be proven to be authentic when in fact it is not.

There is also a problem with gradually outdated measurement technologies, for example, strain gauges and thermocouples do not allow covering the entire required measurement range. For the most part, traditional sensors are only able to measure the required value by separate areas, not allowing you to deeply penetrate into the essence of what is happening. As a result, scientists are forced to rely on pre-modeled processes that show vulnerabilities and force testers to pay increased attention to one or another node of the system under test. But as always there is one thing. This approach works well for time-tested and well-studied materials, but for designs that include new materials, it can be harmful. Therefore, design engineers in all industries are trying to update old measurement methods as much as possible, as well as introduce new ones that will allow more detailed measurements than older sensors and techniques.

Strain gauge technology has remained virtually unchanged since its invention decades ago. New technologies such as , are capable of measuring full field strength and temperature. Unlike legacy strain gauge technologies, which can only collect information at critical points, fiber optic sensors can collect continuous strain and temperature data. These technologies are much more beneficial when conducting physical testing, as they allow engineers to observe the behavior of the structure under study at and between critical points.

For example, fiber optic sensors can be embedded inside composite materials during downtime to better understand curing processes. A common disadvantage, for example, may be the process of wrinkling in one of the layers of material, which causes internal mechanical stress. These processes are still very poorly understood and there is very little information about the stress and deformation inside composite materials, which makes it almost impossible to apply computer modeling to them.

Outdated strain gauge technologies are quite capable of detecting residual strain in composite materials, but only when the strain field reaches the surface and the sensor is installed exactly in the right place. On the other hand, spatially continuous measurement technologies such as fiber optics can measure all field strength data at and between critical points. It was also previously mentioned that fiber optic sensors can be embedded in composite materials to study internal processes.

The development process is considered complete when the product has passed all tests and has begun to be shipped to consumers. However, the current level allows manufacturers to receive the first reports on their products immediately after users start using them. As a rule, immediately after the release of a serial product, work begins on its modernization.

Computer models and physical tests go hand in hand. They simply cannot exist without each other. Further development technology requires maximum interaction between these design tools. Investments in the advancement of physical research data require initially large investments, but the “return” will also please. But, unfortunately, most developers try to get benefits here and now and do not care at all about long-term prospects, the benefits of which, as a rule, are much greater.

Those looking to secure the long-term future of their products will seek to implement more innovative and reliable methodologies and elements of product testing, such as fiber optic measurements. The combination of computer modeling and physical testing technologies will only grow stronger in the future, because they complement each other.

, astrophysics, mechanics, chemistry, biology, economics, sociology, meteorology, other sciences and applied problems in various fields of radio electronics, mechanical engineering, automotive industry, etc. Computer models are used to obtain new knowledge about the modeled object or to approximate the behavior of systems that are too complex for analytical study.

The main stages of computer modeling include:

There are analytical and simulation modeling. In analytical modeling, mathematical (abstract) models of a real object are studied in the form of algebraic, differential and other equations, as well as those involving the implementation of an unambiguous computational procedure leading to their exact solution. In simulation modeling, mathematical models are studied in the form of an algorithm(s) that reproduces the functioning of the system under study by sequentially performing a large number of elementary operations.

Practical Application

Computer modeling is used for a wide range of tasks, such as:

analysis of the distribution of pollutants in the atmosphere
designing noise barriers to combat noise pollution
vehicle design
flight simulators for pilot training
weather forecasting
emulation of the operation of other electronic devices
forecasting prices in financial markets
study of the behavior of buildings, structures and parts under mechanical load
predicting the strength of structures and their destruction mechanisms
design production processes, for example chemical
strategic management of the organization
study of the behavior of hydraulic systems: oil pipelines, water pipelines
modeling of robots and automatic manipulators
modeling of urban development scenarios
transport systems modeling
simulated crash tests
modeling the results of plastic surgery

Different areas of application of computer models have different requirements for the reliability of the results obtained with their help. Modeling of buildings and aircraft parts requires high precision and confidence, while models of the evolution of cities and socio-economic systems are used to obtain approximate or qualitative results.

Computer simulation algorithms

Component circuit method
State variable method

Links

Wikimedia Foundation. 2010.

See what “Computer modeling” is in other dictionaries:

COMPUTER SIMULATION- Quite literally - using a computer to simulate something. Usually the thinking or behavior of a person is modeled. That is, attempts are being made to program the computer so that it acts in a similar way to how... ... Dictionary in psychology

Modeling is the study of objects of knowledge on their models; building and studying models of real-life objects, processes or phenomena in order to obtain explanations of these phenomena, as well as to predict phenomena of interest... ... Wikipedia

Computer vision is the theory and technology of creating machines that can detect, track and classify objects. As a scientific discipline, computer vision refers to the theory and technology of creating artificial systems,... ... Wikipedia

Social modeling- a scientific method of understanding social phenomena and processes by reproducing their characteristics on other objects, i.e. models specially created for this purpose. The need for M. s. due to increased lately necessity... ... Sociological reference book

Cross section of a simulated volume with a thickness of 15 Mpc/h in the modern Universe (redshift z=0). The density of dark matter is shown, with good ... Wikipedia

M. is an imitation of natural situations, in which a person should ideally behave as if it were a real situation. The advantage of the model is that it allows the subject to react to the situation without facing dangers... ... Psychological Encyclopedia

The request for "Software" is redirected here. See also other meanings. Software (pronunciation software is not recommended, or rather, not recommended), along with hardware, is the most important component of information ... Wikipedia

Modeling- (military), a method of theoretical or technical research of an object (phenomenon, system, process) by creating and studying its analogue (model), in order to obtain information about real system. M. can be physical, logical, mathematical... ... Border Dictionary

Computer modeling is one of the effective methods for studying complex systems. Computer models are easier and more convenient to study due to their ability to carry out the so-called. computational experiments, in cases where real experiments... ... Wikipedia

Modeling is one of the ways to understand the world.

The concept of modeling is quite complex; it includes a huge variety of modeling methods: from creating natural models (reduced and or enlarged copies of real objects) to deriving mathematical formulas.

For various phenomena and processes are appropriate different ways modeling for the purpose of research and knowledge.

The object that is obtained as a result of modeling is called model. It should be clear that this is not necessarily a real object. It could be mathematical formula, graphical representation, etc. However, it may well replace the original when studying it and describing behavior.

Although a model can be an exact copy of the original, most often the models recreate some elements that are important for a given study, and neglect the rest. This simplifies the model. But on the other hand, to create a model - exact copy original - can be an absolutely unrealistic task. For example, if the behavior of an object in space conditions is simulated. We can say that a model is a certain way of describing the real world.

Modeling goes through three stages:

Creating a model.
Studying the model.
Application of research results in practice and/or formulation of theoretical conclusions.

There are a huge number of types of modeling. Here are some examples of model types:

Mathematical models. This iconic models, describing certain numerical relationships.

Graphic models. A visual representation of objects that are so complex that describing them in other ways does not provide a clear understanding to a person. Here the clarity of the model comes to the fore.

Simulation models. They allow you to observe changes in the behavior of elements of the model system and conduct experiments by changing some parameters of the model.

Specialists from different fields can work on creating the model, because In modeling, the role of interdisciplinary connections is quite large.

Features of computer modeling

Improved computing technology and widespread distribution personal computers modeling has opened up enormous prospects for studying the processes and phenomena of the surrounding world, including human society.

Computer modeling is, to a certain extent, the same as the modeling described above, but implemented using computer technology.

For computer modeling, it is important to have certain software.

At the same time software, by means of which computer modeling can be carried out, can be quite universal (for example, ordinary text and GPUs), and very specialized, intended only for a certain type of modeling.

Very often computers are used for mathematical modeling. Here their role is invaluable in performing numerical operations, while the analysis of the problem usually falls on the shoulders of a person.

Typically in computer simulation various types simulations complement each other. So, if the mathematical formula is very complex, which does not provide a clear idea of the processes it describes, then graphical and simulation models come to the rescue. Computer visualization can be much cheaper than actually creating natural models.

With the advent powerful computers Graphic modeling based on engineering systems has spread to create drawings, diagrams, and graphs.

Language is a sign system used for the purposes of communication and cognition.

Languages can be divided into natural And artificial.

Natural (ordinary, spoken) languages develop spontaneously and over time. Artificial languages are created by people for special purposes or for certain groups of people (mathematical language, maritime language, programming languages, etc.). Their characteristic feature is the unambiguous definition of their vocabulary, the rules for the formation of expressions and constructions (strictly formalized). In natural languages they are partially formalized. Each language is characterized by: set of characters used;

The rule for the formation of linguistic constructions from these signs;

A set of syntactic, semantic and pragmatic rules for the use of language constructions.

Alphabet is an ordered set of signs used in a language.

In computer science, we are primarily interested in models that can be created and examined using a computer. Using a computer, you can create and explore many objects: texts, graphs, tables, diagrams, etc. Computer technology are leaving an ever greater imprint on the modeling process, so computer modeling can be considered as a special type of information modeling.

In recent years, thanks to the development GUI and graphic packages, computer, structural and functional modeling has received widespread development. The essence of computer simulation is to obtain quantitative and qualitative results of the functioning of the simulated system according to the existing model. Qualitative conclusions obtained from the analysis of the model make it possible to discover previously unknown properties complex system: its structure, dynamics of development, stability, integrity, etc. Quantitative conclusions are mainly in the nature of a forecast of some future or explanation of past values of parameters characterizing the system.

The subject of computer modeling can be: the economic activity of a company or bank, an industrial enterprise, an information and computer network, process, inflation process, etc.

The goals of computer modeling can be different, but most often it is to obtain data that can be used to prepare and make decisions of an economic, social, organizational or technical nature. The use of the computer even in conceptual modeling has begun, where it is used, for example, in building systems artificial intelligence. Thus, we see that the concept of “computer modeling” is much broader than the traditional concept of “computer modeling” and needs to be clarified, taking into account today's realities.

Let's start with the term "computer model". IN currently under computer model most often understood:

§ a conventional image of an object or some system of objects (or processes), described using interconnected computer tables, flowcharts, diagrams, graphs, drawings, animation fragments, hypertexts, etc. and displaying the structure and relationships between the elements of the object. We will call computer models of this type structural-functional;

§ a separate program, a set of programs, a software package that allows, using a sequence of calculations and graphical display of their results, to reproduce (simulate) the processes of functioning of an object, a system of objects, subject to the influence of various (usually random) factors on the object. We will further call such models simulation models.

Computer simulation - a method for solving the problem of analysis or synthesis of a complex system based on the use of its computer model.

The essence of computer modeling is to obtain quantitative and qualitative results from the existing model. Qualitative conclusions obtained from the results of the analysis make it possible to discover previously unknown properties of a complex system: its structure, dynamics of development, stability, integrity, etc. Quantitative conclusions are mainly in the nature of a forecast of some future or explanation of past values of variables characterizing the system.

Computer modeling for the generation of new information uses any information that can be updated using a computer.

The process of studying the behavior of any object or system of objects on a computer can be divided into next steps:

Construction of a content model;

Construction of a mathematical model;

Construction of an information model and algorithm;

Coding the algorithm in a programming language;

Computer experiment.

Security questions

1. What is a model?

2. What are models used for?

3. What is modeling?

4. How are models classified?