The article discusses the features of identifying novelty in data, as well as general methods for identifying it. Since the absence of noise in the training information is a determining factor for building high-quality classifiers on it in supervised machine learning, such a practically important special case of the search for novelty is considered, when it is determined in separate classes of training data after all outliers have been eliminated in these data. For greater definiteness, when searching for novelty, its geometric interpretation in the space of object feature values.

Keywords: data, classifier, outliers, novelty, novelty detection, geometric approach, statistical criterion

Concrete has unique properties that allow it to occupy a leading position in the market of building materials. General-purpose structural concrete is used for the construction of buildings and structures that can provide sufficient strength and elasticity for load-bearing and enclosing structures. Concrete production technologies are in constant development, inextricably linked with the solution of the most important task – reducing the mass of building materials, and, as a result, reducing the material consumption of construction

Keywords: capital construction, structural concrete, heavy concrete, light structural concrete, resource saving, economic feasibility of using light concrete

The aim of this study was to analyze the possibility of using a mathematical model of logistic regression for the recognition of malignant neoplasms on digital images of the skin. The study used a database containing 6594 digital skin images. At the first stage of the study, digital skin images were segmented to isolate the object under study, in which morphometric and color characteristics corresponding to the parameters of the classical ABCD system were determined. At the second stage, the characteristics were used in the classification into malignant and benign neoplasms using logistic regression. When classifying images, the highest value of the accuracy indicator (67.9 [66.9; 68.8]%) was obtained during classification using logistic regression, built on the basis of the reverse Wald stepwise method. Thus, the logistic regression built on the basis of the reverse stepwise Wald method can be applied in the classification of malignant neoplasms on digital images of the skin, but further research and determination of the optimal parameters are required.

Keywords: mathematical model, digital skin images, logistic regression, image classification, skin cancer

The article proposes a special device to transport earth when building an underground part by the «top-down» and «top and down» technique. The developed device is fully electrically powered. The proposed technology enables to completely exclude nonproductive movements of excavating and transportation devices to relocate the earth from the working site to the dumping site onto trucks above ground. The device is adapted to work as part of the production line when realizing the “top-down” and «top and down» technology. To exclude load lifting mechanisms for lifting and dumping of the earth onto the trucks above ground, it is proposed to use two lifting rails with hoisting equipment fixed on a metal frame.

Keywords: “top and down” technology, earth works, excavation works, subsurface part of a building, earth movement, tier of a subsurface part of a building

The article discusses the experimental implementation of a cloud data center based on the principles of sharing economy. The research presents an algorithm and implementation of resource management for shared consumers. The article demonstrates a qualitative assessment of the system performance and basic performance tests of web applications deployed according to the proposed model.

Keywords: cloud technologies, web application, consumer economy, virtua

When creating a model of a nonlinear electrical system based on differential equations, it is important to keep in mind that ordinary differentiation does not give an adequate mathematical idea of ​​the nonlinear physical processes taking place. Even in the simplest case of a conductor with a current, the construction of an adequate model is only possible if the tensor approach is applied. From this point of view, any electrical system is an object of tensor methodology. A typical example of a non-linear electrical system is an AC electromagnetic drive. The nonlinear nature of the processes in such devices is due to a number of reasons. The main reasons are: current displacement in the winding of the AC electromagnetic drive and saturation of the steel of the magnetic circuit in the process of magnetization. Both phenomena admit tensor description, which is confirmed by the research presented in the article.

Keywords: dynamic characteristic of magnetization, electromagnet, mathematical model, tensor methodology

Noise in training samples, the main part of which is made up of outliers and novelty, is considered. The analysis of the main causes of outliers in training samples is given. The essence of the main existing approaches to determining outliers in training samples is considered. Based on the use of the nearest neighbors method, a modified method for comparing generalized distances from objects to classes is proposed. For the main types of metrics used in the spaces of feature values, the justified values ​​of the safety factors used in this technique are found. For a programmatic assessment of the quality of the training sample and a reasonable choice of the method for correcting outliers, it is proposed to use the permissible fractions of corrected and removed outliers in it. An algorithm for analyzing the presence of outliers in a set of training examples is given. An estimate of the complexity of the algorithm by the length of the input of the problem is given. An algorithm for evaluating and correcting training samples has been developed.

Keywords: classification problem, classifier, decision function, training sample, precedent, erroneous data, analysis, correction, artificial intelligence, compactness hypothesis, novelty, learningg

Presents the results of a study of the strength of inclined sections of reinforced concrete beams, reinforced with double-sided carbon fiber clamps. An assessment of the effectiveness of composite reinforcement is given when the value of the cut span is changed, which was equal to 1.5h0; 2h0 and 2.5 h0. At the same time, the degree of influence of the initial inclined cracks on the bearing capacity of the inclined section was analyzed.

Keywords: reinforced concrete, beam, reinforcement, composite materials, carbon fiber reinforced plastic, inclined sections, clamps, reinforcement

The article considers a model for evaluating the performance indicators of information resource exchange in corporate systems, designed to analyze the ability of systems with a distributed registry to ensure the timeliness and completeness of information exchange. As the main indicator, it is proposed to take into account the probability of failure of a segment of the corporate system for each epoch. To find this indicator, we use probabilistic estimates of sums bounded from above by hypergeometric and binomial distributions with probability boundaries of Chebyshev, Heffding, and Schwatal. A numerical and comparative analysis of the proposed estimates is carried out.

Keywords: distributed registry technology, information system, segmentation, block chain

The technology of well cleaning using the KOS-02 complex and a multifunctional tool that has the ability to reciprocate and rotational motion. The scheme of force effect of the cleaning tool is given and the power parameters of the process are estimated.

Keywords: destruction of the face zone, sand plug , equipment for cleaning wells, cleaning tool, flushing technology

The article discusses the methodology for ensuring the timeliness and completeness of the exchange of information resources in corporate networks built on the basis of distributed registry technology, which takes into account the variability of the strategy of the behavior of the distributed registry system during information exchange. The technique takes into account non-standard functions, such as the formation of a branch of the processed data and the impact of an attacker, and also allows you to determine the average value of the block generation delay time by adjusting the number of operations required to solve the block. The use of this method allows to increase the value of the indicator of timeliness and completeness of the exchange of information resources in the corporate network by 30% compared to the existing system of information exchange.

Keywords: distributed registry technology, corporate network, information resources

A method for measuring the magnetic characteristics of high-coercive permanent magnets made of rare-earth materials, such as neodymium-iron-boron and samarium-cobalt, is proposed. The essence of the method is as follows: the data on the measured supply voltage and the characteristics of the magnetizing installation base a mathematical model; the measured shape of the magnetizing current pulse is fed into the mathematical model and compared with the simulated current pulse in the comparison device; further, if the difference in current characteristics does not satisfy the measurement accuracy, then adjustments are made to the mathematical model. Thus, the mathematical model determines the magnetic characteristics of a permanent high-coercive magnet. The method is developed based on full-scale model tests using a multiphysical computer model. The implementation of the measurement method is based on modern IT solutions, the multiphysical model is built in the COMSOL Multiphysics software package.

Keywords: magnet, magnetic parameter, multiphysical model, magnetization, highly coercive material, neodymium, wavelet, Fourier analysis, COMSOL

Actuator drives are widely used both in consumer electronics and in industrial systems: as all kinds of locks, couplings; as part of valves that control the flow of liquids or gases; in robotics; alarm systems, etc. In the event of a failure of the actuator drive, it is not typical for it to restore the operating mode. As a consequence, predicting reliability plays an important role. The article discusses the development of a method for processing the weber-ampere characteristics of actuator drives. Weber-ampere characteristics are converted into ampere-weber ones, reflected by the maximum flux linkage value, and interpolated by the current values ​​that are the same for all characteristics. The characteristics obtained in this way are transformed by the method of principal components and are represented by a point in three-dimensional space. By the location of the points of the center of mass and the dispersion of the groups of characteristics of one electromagnetic drive, a forecast is made about its state by the time of the next check according to the regulations. This method was applied to the characteristics of 10 actuator drives, each of which was subjected to mechanical resistance during operation. The characteristics transferred into the space of the principal components forms a figure resembling a boomerang. The movement of points from the upper edge of the "boomerang" through the center to the lower edge reflects the deterioration of its working properties.

Keywords: prediction, reliability, actuator drive, weber-ampere characteristic, principal components, flux linkage, data processing, duty cycle, malfunction, Statistica, points cloud, performance data

In modern times due to the growth of mass production of electronic devices, it becomes necessary to quickly and accurately verify their characteristics. At the same time, device testing, as a rule, includes a set of different types of checks performed at different stages of the production process. In particular, the analog signature analysis method is used for these purposes. It allows you to quickly detect faults at the component level, which makes it useful in various stages of production, such as incoming inspection, testing printed circuit boards or repairing faulty devices. This article discusses the basic principles of the method, as well as the features of its application.

Keywords: analog signature analysis, component level testing, current-voltage characteristic, fault localization

The main results of studies of the anodic behavior and chemical composition of the surface layer of the aluminum film before and after implantation of silicon ions (20 keV), conducted using the method of measuring potentiostatic polarization curves, are presented. The decrease in the anode polarization currents and the expansion of the passivation region depending on the dose of the introduced ions, due to the formation of oxide compounds with the implanted silicon impurity, are shown.

Keywords: ion implantation, films, aluminum, polarization curves, potentiostat, anode behavior, auger electron spectra, electrolyte, silicon ions