Software & Systems

Articles of journal № 1 at 2014 year.

Order result by: Public date | Title | Authors

31. Software-hardware complex for research, designing and control of processing the plastics masses [№1 за 2014 год]
Authors: Ravichev L.V., Loginov V.Ya., Bespalov A.V.
Visitors: 11746
The developed software-hardware complex «Research and processing of plastics» is intended for researchers, technologists, designers, experts in control and automation who are engaged in scientific research, designing and control processing of the plastics masses. The operational environment of the developed software-hardware complex has prominent fea-tures and advantages and creates possibilities for: full statistical analysis of experimental data, construction of the adequ ate mathematical description, operative transition from experimental researches to designing and back, the analysis and ordering of the experimental data received during various time, various researchers with various techniques. Listed advantages of a software-hardware complex are especially useful at research, designing and control processing of the plastics masses using computer modeling.

32. Software and algorithms for information processing at resistance thermometers calibration [№1 за 2014 год]
Author: Yaparova N.M.
Visitors: 10651
The paper is devoted to parametric identification problem and estimation of the temperature error in the calibration of resistance thermometers. The operating principle of the resistance thermometers is based on the dependence of the electrical resistance on the temperature. In the operation of the device, the resistance is directly measured quantity. Since, it is required to obtain the temperature values from the resistance measurements. The mathematical model describing the de-pendence temperature on the resistance is given by the system of equations with unknown parameters. To solve formulated problem, the calibration of instruments is performed. The goals of calibration are to obtain the parameters of the model and to estimate the temperature errors. To solve the parametric identification problem, we propose a method based on the Tikhonov regularization. Approach for estimation of the temperature error is also expounded. The proposed methods provide the basis for development the methods of the information processing in the calibration. The processing results are the basis for determining of the criteria for choosing optimal parameters in the calibration process. In the second part of the paper we describe the software to implement the proposed algorithms. The proposed methods and software were used for the computational experiments and for the creation of software for experimental data processing.

33. Program software of recovering values in multivariate data based on fuzzy modeling methods [№1 за 2014 год]
Authors: Peremitina T.O., Yashchenko I.G., Luchkova S.V.
Visitors: 9186
The article describes the features of fuzzy systems to solve the problem of restoring missing values in multi-variate data. It considers the Singleton fuzzy system and triangular membership function of the fuzzy linguistic terms. Gen-eral algorithm of an evolutionary strategy method to identify fuzzy system is provided. The article considers two types of this algorithm to assess the advantages and disadvantages of using. Principle of fuzzy system parameter identification based on the evolutionary strategy is described. The article also describes the requirements for the input data and the algorithm of re-storing missing values using fuzzy system based on the evolutionary strategy. There is the experimental efficiency evaluation of the presented algorithm of fuzzy system on the basis of the evolutionary strategy for the problem of recovering missing values. It is tested on a test set of oil properties data. Optimal parameters of fuzzy system and a method of the evolutionary strategy were selected for test data. The article investigates influence of the main parameters of the fuzzy system and metho d parameters of the evolutionary strategy on the recovery of multivariate data.

34. Development of algorithm of directed recognition considering information about relief by example of satellite imagery and Earth remote sensing data [№1 за 2014 год]
Authors: Ovechkin M.V., Altynbaev R.B., Gafurov A.S.
Visitors: 11358
A remote sensing problem concerning unavoidable distortion of the images is researched. It is discovered that if satellite's ground path and distortion mechanism is known, such distortions can be compensated. Additional image adjus t-ment using methods of comparison of control points with geographical map is described. It is mentioned that geographical reference and geometric correction can be combined into one operation of alignment of the details of satellite image and map. The alignment method applied to the image of the reach of the Ural river is given as an example. Directed recognition algo-rithm including information about terrain based on outlining of the image objects is suggested. Differentiation formulas for displacement estimation are provided. The choice of 2-D Laplace operator adjusted to the analysis of satellite images is justi-fied. It is suggested that algorithm speed can be increased with preliminarily worked out block of addresses of image and succeeding operations with addresses of elements resulting 5–8 times higher algorithm speed comparing to direct access to image pixels, that is currently important when processing packets of such high resolution images as satellite images.

35. Synthesis gas telemedicine networks [№1 за 2014 год]
Authors: Margolis B.I., Muzanna M.M.
Visitors: 10272
The article considers the graph model of structural-functional synthesis of backbone telecommunication net-work. It also describes the problem of the backbone structure synthesis. This problem is in minimizing the distances when passing information to the regional switching nodes. The optimal routes for information passing to the regional nodes of backbone network are obtained using a matrix of distances between backbone nodes and the graphshortestpath function in MatLab. The problem of parametric synthesis of backbone network is considered. It consists in determining the capacities of the main communication channels using synthetic topological structure and values of the input streams from subscribers. It is shown that the problem of parametric synthesis of backbone network belongs to a class of nonlinear optimization problems with constraints like inequality and equations. Inequality constraints are formulated for capacity of channels and the type of conservation equations for saving flows in the network nodes. The article presents a developed program that allows finding the optimal information flows in the backbone communication channels using criteria of average messages delay and maxi-mum network load. The program is developed on the basis of the standard fmincon function in MatLab. A methodology for calculating capacity for maximum information flows in the main channels to regional switching nodes. There are the results of calculation of optimal channels capacities for backbone telecommunication network backbone of the Republic of Yemen. An example of passing information to the particular backbone source node and the regional destination node is shown.

36. Synthesis of the subscriber telecommunication networks structure [№1 за 2014 год]
Authors: Dmitriev G.A., Margolis B.I., Muzanna M.M.
Visitors: 9447
The need to separate the telecommunication network in terms of topological structure on the terminal (su b-scriber) and a core (backbone) network is described. The problem of synthesis of a subscriber network is stated. It represent s minimizing the sum of the distances from the subscribers to the main switching nodes with limitations of the ability to con-nect each subscriber to a single site. A complete matrix of distances from the subscribers to the switching nodes is presented. It is shown that the problem of subscribers’ optimal connection to sites switching refers to the class of linear integer pr o-gramming problems with equality constraints. The equality constraints for each subscriber’s connection conditions to one node switching is formulated. Based on the MatLab bintprog standard function, the authors have developed the program that allows synthesizing the topological structure of a user's network using a plurality of users, multiple nodes and switching ma-trix of distances between them. The paper shows the results of calculating the optimal connection of subscribers with the cri-terion of minimum sum of distances to the switching nodes for the telemedicine network of the Republic of Yemen. The art i-cle presents a method for calculating capacity of subscriber communication channels based on the supposed information flow for the received structure of subscriber network and its maximum load. Results of bandwidth calculation for the resulting structure of subscribers network of the Republic of Yemen that is received from synthesis.

37. Decision-making support system UNIQUE [№1 за 2014 год]
Authors: Butenko L.N., Oleynikov D.P., Oleynikov S.P.
Visitors: 9234
Most of the solved problems in the beginning of design process are mixed formulation problems. Type of input conditions can be quantitative (objective evaluation), qualitative (subjective assessment) and mixed. In the case of mixed type, when using standard decision-making methods there is a choice to transform qualitative data into quantitative or vice versa. The consequence is reducing the accuracy of decision-making because of the information loss related to the process of converting information. This is caused by the lack of decision-making methods that could have used the available different-nature information as much as possible. Authors have developed a new method of decision-making to resolve this problem. It is called “UNIQUE” and allows solving problems of mixed nature. The method “UNIQUE” includes considering criteria de-pendences by a hierarchical structure, and a partial accounting of mismatches of expert judgments by using the method of matching clustered rankings. The software package “UNIQUE” was designed for the purpose of automating the method “UNIQUE”. The software package “UNIQUE” is decision-making environment, which consists of structuring subsystem, survey solutions subsystem, justify solutions subsystem, accounting inconsistencies subsystem and verbalization subsystem. The paper contains solutions of common tasks from different domains to demonstrate the operability of software package “UNIQUE”.

38. Technological models for management of company automated control system development [№1 за 2014 год]
Author: Loginov I.V.
Visitors: 12854
The article considers technological models of company automated control system on informatizied business -process level for development management. Technological models application is based on the necessity of development management automation based on continuous effectiveness evaluation of management activities (maintenance, modernization and adaptation of company automated control system considering environment conditions changing). The article extracts four levels of technological models that indicate qualitative assessment of state and effectiveness for informatizied business-processes. These technological models allow estimating instantaneous and integrated effectiveness characteristic values of automatization and informatization for achievement company goals and tasks within a lifecycle. An effect from applying technological models allows increasing management effectiveness. It includes timely detecting uneffective and loweffective automatization projects and utilization of obsolete business-processes.

39. Investment risks management for construction organizations based on bayesian information technologies [№1 за 2014 год]
Authors: Vetrov A.N., Prokopchina S.V., Nesterov А.О
Visitors: 9426
The article considers problems of risk management for investment projects in construction under uncertainty and when productive-economic situation is unstabilized. The authors suggest a complex model of united services “Builder-Client-Contractor” and methods for organizing process approach to risk management of construction projects. The methods are implemented on the basis of "Infoanalitik_stroyinvest" platform. The article describes the methods for assessment and derisking for construction projects under uncertainty. The hierarchical model of risks assessment is based on Bayesian infor-mation technologies (BIT). A procedure of receiving the regularized Bayesian estimates (RBE) based on dynamic restrictions models is described, as well as a possibility of changing object properties and connections between them while managing the object. The main expressions for investment risks assessment for construction projects based on BITS are offered. The con-volution technology of posterior estimates of various indicators for receiving generally integrated assessment in the form of RBE is shown. The paper also shows a possibility of using a scale with dynamic constraints. This type of scale for measure-ment the properties of one-dimensional indicator is two-dimensional. One axis represents indicator values in numerical or linguistic form. The other shows a measure of belief for a result. A multi -dimensional scale is made in multidimensional space of parameters with adding controlled indicators. When switching to a new higher level of hyerarhy, a scale changes in-to integrated index with forming a new two-dimensional scale. In automatic mode the system gives recommendations for changing the situation. Risk assessment is accompanied by counting metrological indexes which are united into complexes of metrological properties. Developed model of risk management for investment projects was tested and had shown good results.

40. Numerical particle-in-cell code for CPU-GPU hybrid systems [№1 за 2014 год]
Authors: Popova N.N., Nikishin N.G.
Visitors: 8827
The paper presents efficiency analysis of parallel programs for ions evolution analysis in particle-in-cell model. Parallel programs for hybrid computing systems with CPU (Central Processing Units, central processors) and GPU (Graphic Processing Units, graphic accelerators) devices were developed. The programs are applied for ions behavior direct modeling in mass spectrometers traps based on Fourier transform. The paper demonstrates the possibilities of using GPU for speeding-up repeated Poisson’s equation with Dirichlet boundary conditions solution, which is based on Fast Fourier Transform, that is implemented in cuFFT library – FFT procedures library for CUDA (Compute Unified Device Architecture). Effective performance and memory bandwidth are compared with GPU peak characteristics for different devices. The chosen algorithm scales according to its complexity estimate. The programs for trapping fields calculation between random shaped electrodes on hybrid systems with simultaneous CPU and GPU data processing were developed. Every parallel process of the field calculation program solves a system of linear equations using GPU with LAPACK procedures that are implemented in CULA library. Calculations on “Lomonosov” supercomputer show that parallel GPU usage efficiency significly depends on choosing of parallel program process distribution nodes. Accelerators can be efficiently used for finding coulomb interaction between ions with solution of Poisson’s equation boundary problem. Parallel computing of fields that inducted by every single surface on CPU can be done when solving linear equations systems on GPU achieving good efficiency.

◄ ← Preview | 1 | 2 | 3 | 4