ISSN 0236-235X (P)
ISSN 2311-2735 (E)


Next issue

Publication date:
16 September 2020

Articles of journal № 2 at 2015 year.

Order result by:
Public date | Title | Authors

1. Information support system of fuzzy evaluation and consistent optimization [№2 за 2015 год]
Authors: Veselkov A.N., Kuznetsov V.N., Doropey V.N.
Visitors: 5597
Enterprise management is considered as a serial or network decision-making process in fuzzy condi-tions. People operate with fuzzy concepts and fuzzy instr uctions. Fuzziness is the main source of uncertainty in the process of coordinated optimization. People use fuzzy concepts and perform fuzzy instructions when making dec i-sions. In addition, in many cases they have different opinions on specific problems an d ways to solve them. In activi-ties, people tend to operate with vague concepts and perform vague instructions. Fuzzy sets and approach to decision -making in fuzzy conditions allows creating a decision support system for information technology such as "1C: Enter-prise 8.3". A decision-making group makes decisions. This group consists of the Center (operation researcher), who is respon-sible for making decisions for the entire enterprise, and agents, who make decisions for their units. The Center and agents are users of a local network for group decision-making support. If the Center solves the problem of optimal so-lutions on the basis of interests, this solution, in general, cannot be optimal for the elements. Elements in accordance with their interests will distort the information provided to the Center, or will not use the information system. To avoid that, the Center should provide the elements with solutions, which are favorable to them. But to do this for the Cent er and agents at the same time is not always possible. Therefore it is better to apply consistent optimization solutions. A method of system analysis for decision-making to create an electronic document management system is based on the consistent control principle and the rules of consistent planning; it includes heuristic procedures and fuzzy logic.

2. An estimate of the of interdistrict passenger traffic distribution in electric transport [№2 за 2015 год]
Authors: Druzhinina N.G., Trofimova O.G., Trofimov S.P.
Visitors: 4643
The paper shows the developed analysis system of urban electric transport passenger traffic based on infor-mation about of electronic fare payment and navigation data of mobile unit motion. The objective of the study is to identify the traffic flow under uncertainty based on the mapping information, electronic fare payment (transaction) and the navigation system of the mobile unit motion (the moments of passing through the checkpoints – stops). Thhe authors introduce the con-cept of transactivator as a set of transactions, which allow determining the movement of passengers paying fares by E-card. The paper shows the possibility of “bonding” the transplants transaction set in a single itinerary for a passenger. It solves the problem of “last mile”: it means that unknown place and time of arrival of the passenger is defined. The authors received sample distributions of urban electric transport passenger traffic containing basic information about the movement of passen-ger: time and place of the start and end points of the trip. Estimate of the urban electric transport passenger traffic distribution is carried out between the city administrative regions with reference to the routing scheme of tram movement. The number of stops in each administrative district is different; however this allowed us to determine the direction of passenger traffic movement by districts and crowding of routes that pass through several parts of the city. The analysis is performed for the re-al passenger traffic flow in weekdays and days off. The system works with incomplete input data (only one type of public transport is considered). Based on the grouping of stops by administrative districts, OLAP-cube with information about the route of the passenger is built. Statistical analysis of this cube allows defining the extremes of passenger traffic and the city administration can mitigate these extremes through modernization of urban transport network.

3. Investigation of the load prediction methods in computer and computer systems [№2 за 2015 год]
Authors: Brazhnikova Yu.S., Goritsky Yu.A., Kutepov V.P., Pankov N.A.
Visitors: 4529
The problem of process and resource management in large scale computer systems with thousands of components is important and has not been solved yet. Thus, a user has to determine the amount of resources needed for a computer system in advance and distribute parallel program fragments to achieve a desired acceleration effect and minimize resource usage. However, even for simple paralleling logic problems such a static planning strategy results in average resource utilization is no more than 15–20 %. The paper is devoted to the problem of computing system load level forecasting to create adaptive methods and algorithms for resource dynamic management and optimization algorithms. The paper presents the data of an experimental investigation of predicting workload of processors. It is based on various filtration methods of high-frequency signal which is processes assessable workload. The results show that among well known filters the median filters are the most precise. The results are used for the development of adaptive algorithms intended for resources optimization, in particular, for allocation of processor resources in computers and large computer systems.

4. Modeling surface disturbance from the object moving in fluids [№2 за 2015 год]
Authors: Barulin А.V., Kulakov I.S.
Visitors: 4364
The problem of calculating environmental disturbance caused by object motion can be stated in many practical cases and dif-ferent objectives that can be: assessment of bogy characteristics and body flow when designing an object, calculation of obj ect motion pa-rameters (natural experiment is frequently changed by numerical simulation), research of flows and dam breaks, creating technologies of flu-id material processing. This problem can also be solved as one of 3D image synthesis stages of object motion in outside envir onment. When considering objects moving across water surface, we can face the problem of creating a dynamic image of interface disturbance (water and air). Talking about real-time image synthesis, an acceptable solution should be made in limited time. The results of this research were r e-ceived after searching synthesis methods of realistic 3D video reflection of object dynamic motion. However, the methodology as a base can possess communality. The paper considers an object moving with not high speed (about several meters per second) in fluid in a surface layer with some (any and in general changeable) deepening. It is required to calculate liquid surface disturbance, which caused by object motion.

5. Early warning system of drinking water quality parameters violation [№2 за 2015 год]
Author: Bubyr D.S.
Visitors: 6855
Nowadays water treatment process is significant and of great interest due to the difficult environmental situa-tion caused by industrial and household activities of people. It is important to maintain the water treatment quality at the proper level, as it directly affects human health. Eventually the state of the water source may be changed, so it is necessary to adapt the water treatment process to these changes. It is necessary to constantly monitor not only the state of an incoming wa-ter source, but also the results of each water treatment. An early warning system was developed to response to a possible ab-normal situation, in which drinking water quality parameters are out of rage, in time. Physical and chemical parameters of water source and drinking water obtained after purification as well as operated factors, which affect the water treatment quality, are controlled regularly and form time series system. The procedure of possible abnormal situation detection consists of two stages. The first stage includes modeling and forecasting the water source state factors based on vector autoregression approach. At the second stage piecewise linear regression dependences of quality factors (with a response break) from physi-cal and chemical parameters of water source and operated factors, which characterize the work of water treatment system, are used to forecast drinking water quality parameters. In addition, for each indicator of drinking water quality there is its own model of optimal order based on the modeling sample of the optimal volume. The system quality is estimated using a test sample.

6. On distance metric for the system of automatic classification of the eee devices by production batches [№2 за 2015 год]
Authors: Kazakovtsev L.A., Stupina A.A., Orlov V.I.
Visitors: 5587
Packaging electronic units of complex technical systems with high quality electronic components is an essential condition of increasing quality of the whole system. Elements of the same type should have equal characteristics, which is achieved if they are produced as one produc-tion batch from a one batch of raw materials. Electronic units vary in the requirements concerning the number of pr o-duction batches of devices. This article considers the problem of discovering the quantity of the production ba tches in a lot shipped by a supplier of electronic devices based on testing results as a problem of cluster analysis. The authors propose using the rectangular metric in the k-means clustering problem. They also show the necessary modification of a local search procedure. The results of running system of EEE devices automatic classification by production batches are given. Such results are provided for a k-median problem with squared Euclidean and rectangular metrics. Tests da-ta of the electronic chips were used as example data. Data dimension is up to 2500 data vectors, each of them contains the results of measurement that are up to 230 parameters. An MDS method (Multidimensional Scaling) was used for visual representation of the multidimensional vectors classification results.

7. The problems of automation technological process of drilling oil and gas wells [№2 за 2015 год]
Author: S.G. Chernyi
Visitors: 6067
Nowadays the problem of IT development and controlling equipment in the oil and gas recovery industry is very important. In addition to the system-managed automated equipment with its characteristics and compatibility problems, an important component of the automation process is the adequacy of its information support. Accounting for all the comp o-nents of the information system requires systematization approach (automation of drilling is a complex process; compatibility problems; the problem of information uncertainty; intelligent decision support systems; the expert knowledge base; automat-ed brake control; automated speed control; automation and control of fluids while drilling; the problem of energy supply). The work of a drilling platform (automation and information aspects), especially during the development of the great depths, faces many uncertainties. This constantly causes accidents and emergency situations that o ccur with a frequency which is not typical for other industries. In addition to the lack of data, there is also the problem of high speed changes. While moving to the project depth, the character of the rocks, pressure, temperature and other important para meters vary sharply and this re-quires continuous collection of information about them. Under uncertainty and high variability of existing decisions on fu r-ther drilling operations management the use of intelligent systems for decision -making based on fuzzy logic and soft compu-ting is considered to be perspective. A new group of automation facilities can help to overcome many problems typical of the oil and gas industries, to improve the efficiency of oilfield development. This group can include new developments in the field of downhole sensors that will provide measurements of high quality; new developments in the design of drill pipes with cable or fiber optic wiring that will help to provide large amounts of received information to the surface.

8. Fractaltheory of information technologies of processing, analysis and classification of large astronomical data flows [№2 за 2015 год]
Author: Myshev A.V.
Visitors: 4944
The paper considers new approaches for constructing models and logic algorithms and procedures of information technologies of processing, analysis and classification of large astronomical dataflow of small bodies’ orbits and trajectori es. The au-thor based the methodology for constructing such models and schemes on fractal theory for constructing estimates of proximity and connectivity criteria of orbits and trajectories in the space of possible states. Logical, algorithmic and informative essenc e of meth-ods and techniques of the theory is as follows. First, processing and analyzing data flow of orbits to determine whether it forms a fractal structure. If yes, it is necessary to identify centers of flow fractal connectivity and obtain estimates of the index of orbits or trajectories information connectedness. Second, selection of monofractal structures in the stream and their classification ac cording to belonging to percolating fractal or fractal aggregate classes. In contrast to traditional methods, fractal theory of information technologies of processing, analysis and classification of large dataflows allows taking into account the properties of regular and irregular structure of the dataflow information state spac e scale as well as dynamic information and their relatedness. Formalism and logics of this theory allow fuller decoding the information that is hidden in dataflows of astronomical observations or measurements. In this case, we receive more complete and informative unde-formed semantic information picture of the research object, which allows giving more complete description of the mathematical and logical space of possible states of the objects observed in the trajectories of their evolution in terms of closure, restrict ions, exchange and uncertainty. On the other hand, it allows giving a mathematical description and the reflection distribution of the observed object in the real three-dimensional space in the local time moments and in the evolutionary development. Methods and approaches of the theory in logical and algorithmic schemes of cognitive information processing technologies and data flow analysis of astronomical observations and modeling results allow reflecting, describing and interpreting them as genetic information that is a carrier of domi-nant and recessive traits. These features are defined as dominant for an information decoding technology contained in the observa-tional data and modeling results. This approach to a mathematical description and presentation, logical organization and management of observations dataflows and modeling results reveals hidden patterns, which analysis can not be obtained by traditional ana-lyzing methods.

9. Decision support system development when determining the level of an enviromental fine of an industrial enterprise [№2 за 2015 год]
Author: Kislyakov I.М.
Visitors: 5679
The article considers a problem of decision support system (DSS) development for taking into account the en-vironmental factor by an administrative center when considering a new investment project, including both opening new man-ufactures and expansion of existing facilities. The author proves the necessity of DSS development. The requirements of the user to DSS are given, as a result there also is its composition: an information component (searching and gathering infor-mation for system operation), a modeling component (mathematical model development and its implementation as a comput-er program), an expert component (the analysis of results obtained in the course of the modeling component, giving recom-mendations). The paper describes the role of every component in DSS. The author gives an example of developed DSS im-plementation which helps an end-user to make searching of exact, objective, science-based result easier. This result should help a territory administration to conduct balanced policy on economic development with minimal damage to the environ-ment. The developed program complex “Ecological balancer” shows an example of the graph which reflects the dependence of an investment project net present value from the rate size of an environmental fine when changing specific emission per unit of new production.

10. A genetic algorithm for computer-aided design of preparatory forging transitions [№2 за 2015 год]
Authors: Kanyukov S.I., Konovalov A.V.
Visitors: 6110
The experience of development and implementation of forging CAPP for different types of hammer and press forgings shows that a considerable part in the list of problems to be solved during computer-aided design is so-called difficult-to-formalize problems. An ex-perienced technologist solves these problems under conditions of noncomputerized designing fairly easily, but CAPP developers have to complement the systems by approximation algorithms and offer users-technologists to make their corrections in solutions obtained in an ac-tive dialogue. This approach supposes that technologist’s production experience is used by systems only in a specific version of a design and is not extended to other versions. At the same time the generated database of technological processes previously designed wit h or without human participation implicitly reflects technologists' production experience and the specific character of forging production at an enterprise. The problem is to use this information when solving the tasks which are difficult to formalize. The application of ideas and methods of the genetic algorithm theory which allow a system to use the experience of previous decisions and to develop itself on this basis while in use seems promising in this case. The paper describes a genetic algorithm developed to solve the problem of designing preparatory transitions in a CAPP system of shaft press-forging. This algorithm has the properties of self-improvement and auto-adjustment to actual manufacturing conditions. This allows saving CAPP users from the necessity of making numerous corrections in the design outputs. The abovementioned approach to the development of genetic algorithms can be applied to solving other difficult-to-formalize tasks of forging design.

| 1 | 2 | 3 | Next →