Skip to main content

Über dieses Buch

This book provides a platform for academics and practitioners for sharing innovative results, approaches, developments, and research projects in computer science and information technology, focusing on the latest challenges in advanced computing and solutions introducing mathematical and engineering approaches. The book presents discussions in the area of advances and challenges of modern computer science, including telecommunications and signal processing, machine learning and artificial intelligence, intelligent control systems, modeling and simulation, data science and big data, data visualization and graphics systems, distributed, cloud and high-performance computing, and software engineering. The papers included are presented at TELECCON 2019 organized by Peter the Great St. Petersburg University during November 18–19, 2019.



Generative Adversarial Network for Detecting Cyber Threats in Industrial Systems

The transition from the information economy to the digital presents new challenges to the community related to the development of breakthrough technologies, a network of cyber-physical systems, artificial intelligence, and big data. When creating digital platforms, a number of difficulties arise: the large dimension of the digital infrastructure and its heterogeneity, poorly established information interaction between the segments, the lack of a common approach to ensuring cybersecurity, and high dependence on personnel qualification and reliability of equipment. The introduction of the digital economy leads to an increase in the risk of cyber threats associated with problems of access control between systems, regulation of information, and control flows. In this paper, for solving cyber threat detection tasks, it is proposed to use generative adversarial neural networks. The paper presents training and testing algorithms of the neural network. The result of the experiments demonstrated high accuracy at cyber threat detection.

Vasiliy Krundyshev, Maxim Kalinin

Detection and Prediction of Safety Faults in Inter-Device Networks Applying a Set of Data-Driven Methods

Digital transformation concerns the safety and reliability for smart homes, industrial IoT, smart building, VANET, WSN, and mesh networks. For common network environment, data confidentiality, integrity, and availability were treated as safety of information, but last decade, due to appearance of dynamic device-to-device cyber spaces, the cyber security is focused on safety, reliability, and sustainability of the connected cyber physical units. Growing variability and amount of controlled data make traditional analytical methods ineffective for safety ensuring. The paper presents our method of cyber faults detection and prediction in smart inter-device infrastructure by data-driven analysis. The proposed approach is founded on the modified method of k-nearest neighbors (kNN) improved with Dempster–Shafer (DS) theory and spatial-temporal correlation of the connected devices. This method shows above 99% level of effectiveness comparing to common safety assurance methods.

Maxim Kalinin, Vasiliy Krundyshev, Viacheslav Belenko, Valery Chernenko

Development of an Algorithm for Determining the Railway Tracks on Video Image

The subject of the work is in the field of creating autopilot railway transport. One of the problems related to railway safety is considered. This paper presents a developed algorithm for determining the railway track from a video image. It is the first step in determining the free path or obstacles in front of the locomotive. The proposed algorithm is a combination of several machine learning algorithms that are applied sequentially (boosting). The first stage of the algorithm is the extraction and classification of features from the image. In this stage, the speeded up robust features or SURF-method is used. At the output of the SURF-stage, we obtain information in the form coordinates of key points and their descriptors. In the second stage, the selected key points are classified. Combinations of two classification methods are used: the K-nearest neighbors or KNN-method and the support vector machines or SVM-method. The final step is the compilation of a railway track mask. For this, the nearest neighbor graph method is used. For practical use of the found mask, the inverse perspective transformation is performed. The efficiency of the developed algorithm is shown experimentally. It can be considered as one of the ways of image segmentation. The main advantages of the algorithm are associated with minimal preparation of the training sample and the ability to analyze its work for further improvement. The results of processing real video images obtained from a video camera mounted on a locomotive are presented.

Ivan Deylid, Sergey Molodyakov, Boris Tyutin

Hardware and Software System for Collection, Storage and Visualization Meteorological Data from a Weather Stand

In the modern world, the use of measuring sensors and microcontrollers is growing every year. Their application ranges from household appliances to industrial automation. The article describes the process of developing and creating a software and hardware system designed for collection, storage and visualization meteorological data. The system includes a stand with weather sensors, a microcontroller and a software. It allows user to receive data from the stand with a microcontroller and to interact with the MongoDB database. As a result of the research and the work, a prototype of a hardware and software system was developed, which provides the tracking of atmospheric environmental indicators, their processing, storage and visualization. In addition, the possibilities for further project development are described.

Pavel Pankov, Igor Nikiforov, Yufeng Zhang

Plant Disease Recognition Based on Multi-dimensional Features of Leaf RGB Images

A system based on the representation of sets of texture parameters of Haralik as multi-dimensional vectors, including the stages of normalizing leaf images, training, testing, recognition. The training stage includes geometric and parametric normalization of the source photo images, calculation of GLCM adjacency matrices, matrix of estimates of mathematical expectations and confidence intervals of the scatter of the Haralik parameters of the components R, G, B, RG, RB, GB of the source RGB photo images; the testing phase, performed by modeling statistical tests in order to determine the required quantity averaging parameters of the diagnosed photo images; recognition stage, based on a correlation comparison of the column vectors of the matrix of estimates of mathematical expectations with the column vector of the averaged values of the parameters of the diagnosed photo images.

Basim Al-Windi, Vladimir Tutygin

Methodology of Service Development with a Single Application Programming Interface

Users of the services can interact with the application using a browser or using mobile devices. The most popular mobile platforms today are iOS and Android. Development of any service includes backend (application logic, database) and frontend (interface) part. Development of frontend part for web, iOS and Android parts is carried out separately, but you can use a single API, which is implemented in the backend part, instead of implementing different backend parts for each platform. In this article, we will consider the architecture of a single API and describe the methodology of its development, which allows you to save resources when creating a service.

Vitaly Monastyrev, Pavel Drobintsev, Petar Kochovski

Using Symbolic Computing to Find Stochastic Process Duration Distribution Laws

Stochastic processes describe the dynamic behavior of systems modeled by formalisms such as queuing networks, network planning models, semi-Markov processes, and some other. Flowgraph models provide an analytical approach to the problem of stochastic process duration distribution law (SPDDL) finding. The problem is solved in two stages. Initially, the moment generating function (MGF) of the process duration is to be obtained using the graphical evaluation and review technique (GERT) or the flowgraph algebra. This stage is straightforward in contrast to the second one—the analytical transition from the MGF to the process duration distribution law in terms of probability distribution function (cumulative or non-cumulative). The transition is nontrivial and is implemented in this study using MATLAB Symbolic Math Toolbox along with various examples of finding SPDDLs when the processes are represented as ordered activity sets. Also, the capabilities of the statistical flowgraph methodology can be extended over the case when flowgraphs have parallel branches. The results of symbolic calculations are validated via simulation using GPSS World software. This study opens up real possibilities of replacing simulation with symbolic mathematics when searching for duration distribution laws of stochastic processes spawned by flowgraph models.

Georgiy Zhemelev, Alexandr Sidnev

Comparison of the Shape of Digital Models of Pump Components

The shape of a flow section determines hydraulic and performance characteristics of dynamic pumps. In order to create digital twins for pumping equipment, a comprehensive project—digital models of pumps—is being implemented in the Hydromechanical Engineering Laboratory of the St. Petersburg Polytechnic University (SPbPU). The engineering issues of the structural optimization are being addressed using a newly developed calculation and design method based on the comparison of component shape. The numerical algorithm is implemented in software codes and is being tested using elements of supercomputing and machine learning. 3D models of pump shafts and impellers with various specific speed rates were used as comparison objects.

Evgeniy Ivanov, Aleksandr Zharkovskii, Igor Borshchev, Arsentiy Klyuyev

Parametric Oscillations of Viscoelastic Orthotropic Rectangular Plates of Variable Thickness

A mathematical model of the problem of parametric vibrations of viscoelastic rectangular orthotropic plates of variable thickness under periodic load is given in the paper on the basis of the Kirchhoff–Love hypothesis in a geometrically nonlinear statement. The mathematical model of this problem is constructed taking into account the propagation of elastic waves. Using the Bubnov–Galerkin method, based on a polynomial approximation of deflection and displacements, the problem is reduced to solving systems of nonlinear integro-differential equations with variable coefficients. The effects of viscoelastic properties of the material and changes in thickness on the oscillation process are studied.

Rustamkhan Abdikarimov, Bakhodir Normuminov, Dadakhan Khodzhaev, Davron Yulchiyev

Methods and Technologies for Protecting Pharmaceutical Products in Polymer Packaging from Counterfeiting

This article considers the problem of protecting pharmaceutical products with polymer packaging from counterfeiting. This issue has grown vital in almost the entire world, as the significant harm can come not only to the producer, but the legitimate producer, but the consumers as well. Due to this, the issue of protecting these products against forgery, and creating and improving existing approaches to anti-forgery protection, becomes a crucial one. The authors suggest methods and technologies for protecting pharmaceutical products’ polymer packaging based on modern ideas from IT and manufacturing such as image recognition, client–server software architecture, mobile apps, digital signatures, luminophores, and PVC film. Testing the authors’ approach showed the effectiveness of the presented methods and technologies. The results should be of interest to companies producing pharmaceuticals.

Tamara Chistyakova, Roman Makaruk, Ilya Sadykov, Christian Kohlert

Solving Multicriteria Optimization Problem for an Oil Refinery Plant

This article describes the process of multicriteria optimization of a complex industrial control object using Pareto efficiency. The object is being decomposed and viewed as a hierarchy of embedded orgraphs. Perfomance indicators and controlling factors lists are created based on the orgraphs and technical specifications of an object, thus allowing to systematize sources of influence. Using statistical data archives to train, the neural network approximates key sensors data to identify the model of the controllable object and optimize it.

Dmitri Kostenko, Vyacheslav Shkodyrev, Vadim Onufriev

Methods and Techniques for Increasing the Accuracy of Continuous Non-invasive Blood Pressure Measurement Under Dynamic Loads

In this paper, various methods for measuring blood pressure (BP) are considered and discussed in detail. Among those, we focus on the indirect BP measurement using the volume clamp method (VC). We considered a hardware solution using the SAKR-2 (Ltd Intox) device for non-invasive BP measurement. Based on our experimental results, this solution is well suited to BP dynamics measure for the patients both at rest and during movement. Specifically, by analyzing the characteristics of multiple measured results in different scenarios, we have proposed several post-processing techniques to remove the systematic errors during measurements (e.g., due to the measurement condition) and enhance accuracy. Compared to the conventional direct BP measurement methods in the radial artery (RA) using commercial device (S5 monitor), both systolic blood pressure (SBP) and diastolic blood pressure (DBP) indicators have strong relationship. The measured BP shifts with respect to hydrostatic changes in blood pressure are fairly match to the theoretical predetermined value during repetitive measurements.

Gleb Zaitsev, Alexei Vassiliev, Quang-Kien Trinh

Computer Modeling of Robust Control of Vibrationless Movement of Multi-mode Flexible Structures

The article is devoted to rejection of unwanted dynamic reactions of flexible structures. The problem of vibrationless movement of an elastic object when natural oscillations are absent and the reaction of the system does not exceed the static reaction is considered. To analyze the oscillations of the system, the maximum response spectrum and the residual spectrum are used. The vibrationless movement property is defined as restrictions on these spectra specified by the control signal. The problem of vibrationless movement is solved by input shaping methods. The shaping filter built on fixed values of the eigenfrequencies of the system is unstable when the frequencies deviate from the set values. To solve this problem, robust modifications of the method are proposed. The high complexity of the considered problem requires computer modeling. In particular, using computer modeling, the problem of the choice of a shaping filter with the property of maximum robustness is solved.

Vladimir A. Prourzin, Kiseon Kim, Georgy Shevlyakov

Feature-Based Plant Seedlings Classification

The application of image features in the plant classification task is studied. The used dataset was created at Aarhus University Flakkebjerg Research. This work aims at different approaches in plant species categorization. In this study, the feature-based approach is used. It allows to perform classifying using less computational resources. The features usage is motivated by the following purpose: to distinguish weeds from other plants by selecting the defining features of all classes of plants. The proposed method combines image thresholding, feature selection, and feature extraction for the further multiclass classification by such well-known machine learning algorithms as the support vector machines, K-nearest neighbors, decision tree, and Naive Bayes. To a greater extent, we use computer vision algorithms for the image processing step. The main classification method we choose for the task is the support vector machines: It shows the best performance among other tested algorithms; the K-nearest neighbors algorithm is slightly worse.

Dmitri Jakovlev, Iuliia Kamaletdinova, Georgy Shevlyakov

Medical Training Simulation in Virtual Reality

Digital technologies have a significant impact on medical training. Growing popularity of virtual and augmented reality changes the trend into virtual simulators. We gave an overview of key market products in the field of medical simulators in order to define main aspects for development of our own system. These aspects are: open surgery, realistic visualization, and haptic feedback. We described in details each of them and how it was implemented in our system. For open surgery was used appendectomy as most common procedure of this type of the surgery. In order to achieve realistic visualization, we implemented three different approaches for creating realistic and accurate 3D models. For haptic feedback, we took Novint Falcon and enhanced it with our custom grip which provides additional degrees of freedom.

Vladimir Ivanov, Sergey Strelkov, Alexander Klygach, Dmitry Arseniev

Application of the Hybrid Model to Numerical Modeling of the Urban Transport Network Topology

Improving road safety, monitoring the vehicles, and controlling their routes require rapid development of intelligent transportation systems (ITS). The main tool for ITS design is computer modeling. Because such modeling is based on either micro-level or macro-level only, the existing models of transportation systems often represent a trade-off between modeling accuracy and computation time. Hybrid models of transportation networks, allowing one to simulate different parts of the network using micro- and macro-models combined with the ability to synchronize them, are currently not available. In the present work, we justify the need for hybrid model development and suggest a way to have a transition between micro- and macro-levels of transportation network model. The paper contains the results of modeling a segment of urban road network with mixed topology and reports computation time required to simulate the traffic behavior in hybrid model. The proposed solution was implemented using the SUMO microscopic simulator of transportation networks and the original continual model of traffic flows. The suggested approach allows one to greatly reduce the computation time required for modeling the traffic flows on large areas without affecting the accuracy.

Vadim Glazunov, Mikhail Chuvatov, Leonid Kurochkin, Mikhail Kurochkin, Alexander Chernyshev, Leonid Hanin

Synchronization Scheme for UWB Wireless Sensor Network System

Ultra-wide band (UWB) is a modern radio technology that is popularly used in wireless sensor networks (WSNs) due to high resistance to multipath fading in conditions of numerous reflective surfaces. However, a cost of an ultimate device can be very high because of problems of ultra-short impulse processing. Hence, we are focused on the synchronization task to employ the UWB technology for a WSN, which has a large number of nodes. We utilize the non-coherent novel approach to detect a signal at the receiving side and for this scheme, it is necessary to choose preamble that has the best properties. In this paper, we compare the auto-correlation and cross-correlation properties of M-sequences and optical orthogonal codes (OOCs) to reveal the best one for transmission by one user and several users asynchronously. Also, we depict the best preamble threshold value using the Bayesian approach.

Iuliia Tropkina, Sergey Zavjalov, Dong Ge

The Deep Survival Forest and Elastic-Net-Cox Cascade Models as Extensions of the Deep Forest

Two new survival models, the deep survival forest and the Elastic-Net-Cox Cascade, are presented in the paper. They can be regarded as a combination of random survival forests and the Elastic-Net-Cox models with the deep forest (DF) proposed by Zhou and Feng. The main ideas to construct the models are to replace the original random forests incorporated into the DF with the corresponding survival analysis models. A stacking algorithm implemented in the deep survival forest and the Elastic-Net-Cox Cascade, which can be regarded as a link between the DF levels, uses quantiles of the random time-to-event and the mean time-to-event computed from the estimated survival functions at every level of the DF. Numerical examples with real data illustrate the proposed models.

Lev Utkin, Andrei Konstantinov, Anna Meldo, Victoria Sokolova, Frank Coolen

An Explanation Method for Siamese Neural Networks

A new method for explaining the Siamese neural network is proposed. It uses the following main ideas. First, the explained feature vector is compared with the prototype of the corresponding class computed at the embedding level (the Siamese neural network output). The important features at this level are determined as features which are close to the same features of the prototype. Second, an autoencoder is trained in a special way in order to take into account the embedding level of the Siamese network, and its decoder part is used for reconstructing input data with the corresponding changes. Numerical experiments with the well-known dataset MNIST illustrate the propose method.

Lev Utkin, Maxim Kovalev, Ernest Kasimov

Hierarchical Multi-agent System for Production Control Using KPI Reconciliation

In this work, the task of production key performance indicators’ values reconciliation is set, that takes into account their hierarchical structure and interrelationships. The structural scheme of a hierarchical multi-agent system for production control was developed where horizontal (sibling) and vertical (multilevel) connections between agents are shown. Then the possible situations of sibling and multilevel interactions are described, such as changing of a task by the controlling agent, sending notifications about the impossibility of a maintaining current operation mode, and others. The agents’ data exchange algorithms, which are used in order to optimize key performance indicators, are shown. The implementation of developed algorithms using client-server architecture is shown, which also includes at the bottom level data exchange between the agents and programmable logic controllers. The single bytes command system for agents’ interactions is described.

Vladislav Kovalevsky, Vadim Onufriev, Anton Dybov

Semi-supervised Learning for Medical Image Segmentation

Semi-supervised learning is a combination of conventional supervised methods with weakly supervised learning. A recent development in neural networks allows to achieve high-quality results but the training requires a large amount of annotated examples. This hinders the applicability of deep learning to some problems, especially medical imaging. In this paper, we present a semi-supervised learning approach based on convolutional neural networks (CNN) for medical image segmentation. A network is trained on a combination of fully labeled samples that have segmentation masks available and weakly labeled samples that only have class labels. We performed experiments that compare the results of the semi-supervised model with the baseline supervised method. Experiment results show the superiority of suggested methods on a low amount of fully annotated samples for lung nodules CT images.

Mikhail Kots, Mikhail Pozigun, Andrei Konstantinov, Viacheslav Chukanov

Developing a New Generation of Reconfigurable Heterogeneous Distributed High Performance Computing System

Restrictions of current architectures of Heterogeneous High-Performance Computing (HPC) systems lead to Reconfigurable Heterogeneous HPC (RH HPC) which are able to adapt to a particular solved task on the hardware level. Highly disruptive technologies like Artificial Intelligence (AI), Internet of Things (IoT), and Machine Learning (ML) are expected to lead not only fundamental shift in classical multi-cores and multi-treads-based approaches to high-performance computing, but also open new direction in designing systems with dynamically reconfigurable runtime environment. The state-of-the-art integrated circuits allow realizing the general architecture of RH HPC at different levels of the distributed HPC: from the level of the supercomputer, to the level of user computers and the remote, built-in units intended for data acquisition, management, and control. The article describes proposed architecture for new generation of Reconfigurable Heterogeneous Distributed HPC (RHD HPC) system, including architectures of each sub-parts and highlights already developed components for such RHD HPC.

Alexander Antonov, Vladimir Zaborovskij, Ivan Kisilev

Usage of a BART Algorithm and Cognitive Services to Research Collaboration Platforms

The ability to predict behavior in complex systems has always interested scientists. With the development of science, there is a gradual complication of systems and data processing methods in them. In this publication, work in the system is achieved by emulating interaction at various levels and nodes through collaboration platforms. The key idea of the whole area is the ability to predict the behavior of the system node based on experience and data obtained at previous stages of work. Improvement of such approaches in the future can have a serious impact on the process of improvement and the evolutionary transition to systems that are currently impossible to imagine. This transition remains impossible until humanity has learned to work effectively in the current collaboration platforms. The paper considers an algorithm for processing the obtained data and its extension using existing cognitive services for the analysis of texts. In the future, the algorithm may be expanded to work with visual information.

Sergey Saradgishvili, Ilia Voronkov

A Computer-Aided Diagnosis System in the Diagnosis of Multiple Sclerosis

This chapter reviews the basics and recent researches of computer-aided diagnosis (CAD) systems for assisting neuroradiologists in the detection, monitoring and prediction of multiple sclerosis (MS) in magnetic resonance (MR) images. The CAD systems consist of image feature extraction based on image processing techniques and machine learning classifiers such as linear discriminant analysis, artificial neural networks, and support vector machines. We introduce useful examples of the CAD systems in the neuroradiology and conclude with possibilities in the future of the CAD systems for MS in MR images.

Polina Andropova, Dmitriy Cheremisin, Anna Meldo

Predicting Students’ Performance on MOOC Using Data Mining Algorithms

This paper describes the results of experiments in predicting students’ performance on a massive open online course (MOOC). Grade reports from MOOC “Data management” on the Russian platform were used for the analysis. It is well known that only a small percent of students who enrolled in MOOCs pass them through. Data mining methods could help to understand the causes of this problem. We tried to predict whether the student will finish an online course or not based on his results during the first weeks. Such prediction if it was performed early enough could help to keep students in the course.

Sergey Nesterov, Elena Smolina, Tigran Egiazarov

On the Implementation of the Planar3D Model Using the Explicit Time Integration Scheme and the Statistical Front Tracking Method

Nowadays, the relevance of problems associated with mathematical modelling of the propagation of hydraulic fractures is growing. In this paper, we consider a Planar3D plane crack model using the explicit numerical integration scheme over time to calculate the dynamics of crack propagation in a multilayer medium. When conducting numerical modelling, the key task is to ensure the speed of calculations while maintaining the stability of the calculation scheme. The present work is devoted to modelling the hydraulic fracturing of a layered formation under the influence of nonuniform injection of non-Newtonian fluid. We present the medium in the form of a set of horizontal layers, each of which is characterized by its own values of minimum compressive stress, toughness and leakage coefficient. To determine the geometry of the crack, we solve the front velocity equation and use universal asymptotes and the explicit time integration scheme.

Egor Starobinskii, Nikita Mushchak, Svetlana Kraeva, Sergei Khlopin, Egor Shel

Fast Fourier Transform in Planar3D Model Using an Explicit Numerical Integration Scheme

In this paper, the authors propose methods to speed up calculations of a fracture propagation model using fast Fourier transform (FFT). We consider the Planar3D model with the explicit numerical integration scheme. Current research decided to implement the radix-2 Cooley–Tukey FFT algorithm in C++ code using STL containers, which provided fast calculations and gave advances to work with memory and cache. We compare the speed of FFT computation with other libraries (FFTW3, GSL, Eigen3). Analysis of results has been shown as a comparable time of calculations. We consider a method for accelerating the calculations of the Planar3D module in the framework of matrix–vector multiplication and processing of input data using a low-pass filter. The considerate model uses the product of matrix–vector multiplication. This procedure engages from time to time throughout the program. In this paper, we implement a modified method for calculating the matrix–vector multiplication product using FFT, which allows us to speed up the calculations. Another technique is used to process input data by the example of averaging lithology layers. In the presence of thin layers with high contrasts of mechanical properties, one can apply a low-pass filter. Such processed layers make it possible to obtain an increase in the computational speed when simulating the evolution of a hydraulic fracture model.

Nikita Mushchak, Egor Starobinskii, Sergei Hlopin, Egor Shel

Employee Performance Analytics Approach Based on Anomaly Detection in User Activity

In this paper, we highlight the distinctive features and critical areas of analytical tool application for the employee performance analytics of user activity. We describe problems of applying data analytics methods and technologies to ensure employee performance analytics. We also discuss the use of user activity time-series data analysis methods and techniques to provide employee performance analytics and describe approaches for processing unstructured data from different sources of user activity for further analytics using anomaly detection methods. Finally, we introduce a new strategy of building features from hybrid data streams from different sources and compare it with current practices.

Aleksey Lukashin, Mikhail Popov, Dmitrii Timofeev, Igor Mikhalev

Deep Predictive Control

Modern control systems are characterized by high complexity and hierarchies of many orders. In the case of a large number of nodes, real-time control becomes a non-trivial task and requires new concepts and paradigms. This paper discusses the concept of using model predictive control (MPC) and deep predictive control (DPC) in the tasks of control complex objects. The ability of deep networks to generalize allows using them to build effective control systems of increased complexity, working in conditions of uncertainty and limited data.

Dmitry Baskakov, Vyacheslav Shkodyrev

On the Computational Complexity of Deep Learning Algorithms

The paper analyzes current research and the state of the industry to assess the complexity of machine learning algorithms. The tasks of deep learning are associated with an extremely high degree of computational complexity, which requires the use, first of all, of new algorithmic methods and an understanding of the assessment of the complexity of the calculations. This area of research is not given due attention for various reasons, but primarily because of the novelty of this paradigm, as well as the use of other advanced methods, which is briefly analyzed in this paper.

Dmitry Baskakov, Dmitry Arseniev

Enactivism in the Conceptual Basis of the Non-classical Theory of Management of Ergatic Systems

The article discusses the problems of managing complex ergatic systems containing symbiotic and environment-oriented forms of control and orientation by interacting agents. The difference in the control methods of complex ergatic systems created in the framework of classical and non-classical ergonomics is shown. The conceptual basis of non-classical and post-non-classical ergonomics is presented. The article discusses the prospects of using the concept of enactivism in the conceptual basis of ergatic systems management, which allows implementing a project of continuously updated ergatic environment, which focuses on the processes of continuous updating, operational control and correction of the parameters of the technical and human parts of the system, taking into account the cyclic processes of self-organization in the actor’s environment of the acting subject achievement of business goals. The forms and properties of intelligent entities embodied in organized environments are presented. A number of general definitions of intelligence and intelligent symbionts are embodied in existing ergatic systems that arise in the process of combining artificial and natural self-organizing systems of the environment of activity. The prospects of using conceptual representations of enactivism and constructivism in the management of complex ergatic systems are shown.

Sergey Sergeev, Vladimir Ivanov, Oleg Ipatov

The Solution of “If-Problem” in Computations with Multi-valued Variables Based on Operator Overloading

This paper presents the exhaustive solution of the “if-problem” in computations with multi-valued variables based on the technique of operation overloading that is widely used in object-oriented programming. The proposed approach addresses the problem of the conditional branches when due to the uncertainty of the operand the result of the logical operation is not determined explicitly and therefore both conditional branches should be executed simultaneously. For the purpose of the study, the code for C/C++ was developed and tested. The results of the performed tests are provided. The proposed solution makes it possible to transfer previously written C/C++ program code for calculations with single-valued types of variables (float, double, etc.) to calculations with any formalism representing the multi-valued variables using minor code modifications. This offers the opportunity to expand the applicability of earlier developed program code to the wider scope of application problems: from the calculations with input variables accurate to round-off errors to the computations with inaccurate or uncertain data.

Vyacheslav Sal’nikov, Konstantin Semenov

The Interval Method of Bisection for Solving the Nonlinear Equations with Interval-Valued Parameters

The article dwells on the interval extension of the bisection approach for solving nonlinear equations with interval-valued parameters, i.e. the ones that might have values from the specified bounds. It is shown that such a procedure allows to obtain an interval of possible values for equation root that is entirely determined by the equation parameters inaccuracy and does not depend on any other factor. The proposed interval bisection method can be easily implemented. All the differences from the traditional bisection approach for solving equations have a clear meaning. The simple stopping rule is proposed. It is shown that considering the interval nature of equation parameters makes it possible to finish the iterative process of equation solving earlier in full accordance with known information on the equation parameters. The proposed approach keeps the important bisection method property—all the intermediate estimates of the bounds of the root’s possible values interval include the exact boundaries. The article provides an illustrative example of how to use the interval bisection.

Konstantin Semenov, Anastasia Tselishcheva

Complex Monitoring Systems for Landfills

The problem of monitoring the state of landfills is described in the article. There are a lot of such objects in the world. At the same time, there are no standard solutions for monitoring the state of these facilities both in Russia and abroad. It is proposed to develop a technical solution based on autonomous sensors for measuring the concentrations of hazardous fumes, radiation background, geotechnical factors and other environmental variables. Such system can be easily installed at any site and can work offline for a long time. The design and implementation of this system are undoubtedly connected with the issue of investment analysis. Positive economic results of the use of such systems can be an important target component in concept of green economy.

Aleksandr Titov, Sergey Krasnov, Andrey Timofeev, Victor Denisov

Modeling the Control Object in the Management System of the Regional Socioeconomic System

The structural and functional model of the control system of the regional socioeconomic system is presented. The author proposes a model of the control object of the regional socioeconomic system in the general context of managing the subject of the Russian Federation of the type “region”. It is shown how the vector matrix calculus apparatus can be used to describe the control object in the control system of the regional socioeconomic system. The task of managing the regional socioeconomic system is formulated as choosing a vector of the controlling action for transition to the desired state of the state matrix of the control object, for which the corresponding subdomain in the range of permissible values is determined. The features of dynamic processes inherent to the controlling action and output coordinates of the control object are considered.

Elena Averchenkova

Transformation, Visualization and Analysis Different Kind of Study Information Contained in the Students’ Electronic Portfolio

Software for accounting and systematization in educational activities is developed for the effective work of teachers and students. In Magnitogorsk State Technical University, every student who has achievements in educational, research, public, cultural, creative or sports activities has the opportunity to be assigned to an increased state academic scholarship (Order of the Ministry of Education and Science of the Russian Federation of December 27, 2016, No. 1663). For this purpose, on the educational portal of the Magnitogorsk State Technical University a student’s portfolio is filled in for each of the activities, which is the link between the student and the teacher. For accounting and systematization of educational activities, a web module was created, integrated into the educational portal. Throughout the course of study, the student has the opportunity to view statistics and achievements in his/her academic activities, add, evaluate and analyze information for further effective study, obtaining increased scholarships and successfully defending graduate qualifying work. Reducing the time and labor costs required to collect and systematize the achievements in student learning activities will make the work of teachers more efficient and productive when interacting with students.

Elena Ilina, Yuliya Kocherzhinskaya, Nikita Dyakonov, Daria Arefeva, Tat’yana Antonova, Il’ya Levandovskii

An Automated Measuring Complex for Research Parameters of Unmanned Aerial Vehicle

In research, design and development of unmanned aerial vehicles (UAV), a key role belongs to automated measuring complexes for its simulation and prototyping electromechanical system with digital control. Design and debugging of such system are suggested and performed by using model-based approach with automated simulation tools. This approach means creating and using in further the simulation model of the measuring complex. In the paper, the simulation model for research parameters of unmanned aerial vehicles is discussed. This model is suggested to develop based on Gough–Stewart platform (six-axes platform manipulator) with UAV mounted on it. The researched model includes the UAV’s trajectory generator, automatic control device for UAV, kinematic manipulator model and decision subsystem to define the current attitude of the manipulator dynamic side. The model allows to evaluate the functioning parameters of proportional–integral–derivative controllers of spatial orientation angles, as well as to automatically obtain program code for implementing both automation of testing and the UAV control device. Also, the presented model can function in conjunction with the user interface of the measuring complex and can be used to evaluate the parameters of complex functioning.

Oleg Drozd, Pavel Avlasko, Semen Bordyugov, Denis Kapulin

Research and Evaluation of the Most Significant Quantitative Characteristics of MPLS Equipment

The article presents an analysis of the characteristics of operator`s equipment for the construction of MPLS networks, which will be important in assessing the proposed procurement units for network modernization. The main characteristics of the devices are considered in the work and the most significant of them are analyzed. Method: An analytical dependence is proposed for the characteristics of L2 VPN services. Core results: A practical experiment was carried out to confirm theoretical conclusions on such characteristic as the number of the LDP neighbors. Practical relevance: The results of the experiment are presented and conclusions are made about the significant characteristics for MPLS equipment for L2 VPN services.

Andrey Krasov, Pavel Karelsky, Igor Zuyev, Max Kovzur, Alexsander Tasyuk

Study of the Microstrip Waveguide Prototype Model for Use as a Retunable Diffraction Grating

The prototype model (hereinafter—the prototype) with strip micromechanical waveguides in a free state with piezoelectric oscillation excitation was studied. The study methodology for the basic functional characteristics of the prototype with strip micromechanical waveguides in a free state was developed. The test bench was made. Visual inspection of the prototype model was performed and its quality was assessed using microscope. Tests on the prototype model were carried out. Diffraction patterns of light reflected from a flexural wave in waveguides at different frequencies of mechanical excitation were obtained. Diffraction patterns were analyzed.

Dmitry Nikulin, Valery Reichert, Sergey Shergin, Igor Karmanov, Vladimir Korneyev, Polina Zvyagintseva

The Use of Digital Cameras for Multispectral Registration with an Unmanned Aircraft

Monitoring of the environment with the help of unmanned aerial vehicles is currently one of the most developing branches of optoelectronic instrument-making, since the digital cameras installed on these devices make it possible to survey the underlying surface in order to further highlight the signs of this surface that carry information about its state. The use of unmanned aerial vehicles for the control of agricultural lands is a particular and very perspective case of such monitoring. The technique of measuring the spectral reflection coefficients of surfaces used to identify the state of vegetation observed in the field of view of multispectral digital camera, which monitors the environment from the board of unmanned aerial vehicles. The method allows to determine the spectral reflectance of the calibration surfaces against the reference surfaces. The results of the work are applied in the analysis and processing of images obtained in the course of the unmanned aviation system that monitors agricultural lands.

Evgenij Gritskevich, Sergei Novikov, Polina Zvyagintseva, Diana Makarova, Marina Egorenko, Aelita Shaburova

Computer Model for Analysis of the Process of Image Construction in Optical-Electronic Visualization Systems

A simulation computer model of an image visualization system is considered. The model allows to analyze the effectiveness of the use of such systems together with the observer’s eye according to the probability criterion for solving the observational problem using Johnson’s equivalent bar. This method of formalizing input actions provides the possibility of their analytical presentation in the form of a set of harmonic signals, which greatly simplifies the modeling task while maintaining the adequacy of the model. The proposed simulation model is intended for computer analysis of any visualization systems that include links of various physical nature. In this case, each link is displayed by its own model, taking into account the features of the link, but the composition of the information signal parameters at the input and output of each link does not change, which allows a uniform methodological approach to the development of software modules that implement link models. At the model output, the probability of solving the observational problem by the observer’s visual apparatus on the device’s screen is calculated at the required degree of decryption of the object. This allows for a given probability to determine the range of the device, and for a given range and probability—to carry out optimal coordination of link parameters. In addition, the model provides the calculation of the most important characteristics of visualization systems, for example, the modulation transfer function, which, combined with the calculation of the effective values of the noise fluctuations of the output image, makes it possible to generate these images for typical objects of observation on the screen of a computer monitor, as when operating a real visualization system.

Evgenij Gritskevich, Marina Egorenko, Diana Makarova, Sergei Novikov, Alexey Polikanin, Aelita Shaburova

Solution of Partial Differential Equations on Radial Basis Functions Networks

The solution of boundary value problems described by partial differential equations on networks of radial basis functions is considered. An analysis of gradient learning algorithms for radial basis functions networks showed that the widely used first-order method, the gradient descent method, does not provide a high learning speed and solution accuracy. The fastest method of the second order, the trust region method, is very complex. A learning algorithm based on the Levenberg–Marquardt method is proposed. The proposed algorithm, with a simpler implementation, showed comparable results in comparison with the trust region method.

Mohie Alqezweeni, Vladimir Gorbachenko

Predicting Personality from Image Preferences: Tendencies, Models and Implementation

This paper describes methods of predicting personality traits from image preferences. Methods of feature selection and machine learning were approved and the results of best models are described in the conclusion. Also the novel approach to the preference data analysis was applied: an attempt to discover general patterns in preference data analysis was made, and several hypotheses about tendencies and precision estimation were formulated and tested, using experiment with images from predefined gallery.

Stanislav Krainikovsky, Mikhail Melnikov, Roman Samarev

Power Consumption Meter for Energy Monitoring and Debugging

This paper presents the results of power consumption measurement research. The main purpose of the research is to develop a low-cost power monitoring device allowing detection of short-time power surges. Power meter should provide high accuracy of measurements along with portability and simplicity of use. This device helps to detect software bugs, caused by power anomalies in electronic equipment functioning. Fixing power bugs keeps application competitive on the market. The structure and electronic design of power meter is presented. The device prototype has been assembled and programmed using purchased components and original software. The testing method has been developed and applied for power meter prototype. The test results have been analyzed and modifications for the next power meter generation.

Nikita Kulikov, Elena Yaitskaya, Arina Shvedova, Vladimir Zhalnin

LoRaWAN Gateway Coverage Evaluation for Smart City Applications

The meaning of the term smart city has undergone changes over the last decades. However, smart city services nowadays are gaining popularity worldwide. Russia is not an exception. Since early 2000s smart city market is growing steadily and, according to forecasts, will continue to grow. Today, the possible applications of smart city services cover a wide range of sectors—from energy production, distribution, and consumption to sustainable mobility and waste management. All the innovative services require millions of monitoring sensors and control devices to be connected between each other and to a management platform. Hence, new types of wireless communication networks that meet the specific requirements of smart city services are needed. LoRaWAN is the most promising (machine-to-machine (M2M) communication technology among other LPWANs such as NB-IoT and Sigfox. Conducted field study of LoRaWAN gateway coverage in different conditions shows the LoRaWAN attenuation coefficient in conditions of city center and outskirt and reveals the factors on which the signal attenuation coefficient depends.

Vadim Shpenst, Andrei Terleev

Fire Resistance Evaluation of Tempered Glass in Software ELCUT

Modern windows and facade glazing elements, depending on the type and purpose of the building, perform various functions. They can be both direct sources of daylighting and elements fireproof structures in external building envelopes. The high temperature in case of fire causes deformations and loss performance properties of structures. Structural calculations and modeling of fire resistance of structures is a hot topic. Designers should know the fire endurance of the structure in the process of solutions development. In this paper, the temperature fields and stress fields under fire exposure are calculated in software package ELCUT. The various thermophysical properties of 6-mm-thick heat-strengthened soda-lime silica float glass as part of single-chamber and two-chamber packages with a cold-framed steel profile and various thickness of glass packs: 6 mm and 24 (with an air gap) taken into account. Berkeley Lab WINDOW software was used to calculate the overall thermal performance of the window. It is shown that ELCUT allows to make a model the fire test of structures and represent temperature fields and stress fields. The interrelation of the occurrence of mechanical stresses from temperature effects with the consideration of the thermophysical properties of materials has been revealed.

Marina Gravit, Nikolay Klimin, Alina Karimova, Evgenia Fedotova, Ivan Dmitriev


Weitere Informationen