Annotation: For the design of an aircraft wing, algorithms are required to determine the geometric parameters of the designed lifting surface in order to provide the necessary values of integral design parameters that significantly affect flight performance. In this paper, methods for creating metamodels for problems of aerodynamics based on the approximation of data from numerical experiments are investigated. The use of these methods significantly reduces the time spent on searching for optimal geometric characteristics of the wing, bypassing long-term experimental studies on the influence of geometric parameters on the power characteristics of the considered lifting surface. The aircraft development process consists of several stages, the most important of which is the preliminary design, at which stage a decision is made on the viability of the aircraft development project. The first stage is to develop a variant of the geometry of the aircraft, based on expert experience, then the calculation of the main aerodynamic configuration parameters is carried out taking into account various flight conditions using software tools and engineering methods. Experimental data obtained as a result of testing an aircraft model in a wind tunnel is one of the important sources of information about aerodynamic properties. The third stage consists in a detailed analysis of the obtained flight characteristics of the aircraft. Currently, the aviation industry needs to move to a new technological approach based on the use of software for automated optimal design of aircraft aerodynamics, taking into account their design features and limitations. Parametric modeling in specialized computer-aided design systems allows you to speed up the process of generating technical proposals. Making design decisions is an important stage, the purpose of which is to choose one of the possible options for the values of the geometric model parameters. This choice is usually a compromise, as each option has its advantages and disadvantages. In this paper, a method for constructing a surrogate model based on the kriging interpolation method is developed. The proposed model makes it possible to analyze complex multiparametric problems of aerodynamics. Using the example of the problem of two-dimensional flow around an airfoil, nonlinear relationships of the main aerodynamic characteristics with the geometric parameters of the profile at different speeds of the incoming flow were found.

Annotation: The article discusses the possibilities of rational use of artificial intelligence in the business processes of project management for the banking industry, and a conceptual approach to the rational use of artificial intelligence is proposed, which includes subprocesses of analysis, planning, implementation and control of IT projects. The main goal of the article is to study the use of current artificial intelligence tools (hereinafter - AI) in the banking industry for project management, determine the main risks from the point of view of the use of artificial intelligence, as well as argue the feasibility and the need to improve control over this process. The prospects of using artificial intelligence algorithms and improve the digitalization of large banks are presented.

Annotation: The article examines the problem of analyzing changes in software functions due to changes in data types and proposes a method of static source code analysis aimed at solving it. The developed method is based on traversing the abstract syntax tree of source code in the C++ programming language. Approaches for applying this method in development processes that based on flexible methodologies are also proposed.

Annotation: The article considers an original approach to clustering of single-dimensional streaming data, based on the principles of density based clustering. This allows to work in conditions of information noise in order to cut off outliers and uninformative anomalous data. To implement this approach, an algorithm was developed consisting of several functional blocks and involving the search for single-dimensional cluster boundaries using machine learning technologies, which effectively uses information about the appearance of new clusters, preserving only significant data elements, which has a positive effect on the requirements for computing resources. To further improve the efficiency of the proposed algorithm, an approach of adaptive splitting of data from the input stream into frames of various sizes with subsequent processing based on a heuristic approach that takes into account the features of single-dimensional feature space and the cumulative nature of information about the presence of clusters is applied. The resulting algorithm demonstrates high efficiency in terms of data processing speed and memory usage. Its computational complexity tends to be linear over time. The authors also managed to achieve high clustering quality indicators, evaluated according to the criteria of compactness and separability of clusters, which are universal for any clustering algorithms based on the density of data distribution in the feature space. These advantages were confirmed by an experiment on 20 sets of test data, the results of which are also presented in the framework of this work. The presented algorithm occupies a rare niche of algorithms for clustering streaming data in conditions of information noise, optimized for working with one-dimensional data. Individually, each of the tasks of clustering streaming data and clustering one-dimensional data has been considered by the scientific community for quite a long time, however, their totality remains without due attention, despite the obvious benefits, for example, for solving problems of searching for stable states or clearing anomalous and noise values when analyzing one-dimensional signals, sensor readings, etc.

Annotation: This paper proposes an approach to improve the quality of Extreme Learning Machine (ELM) predictions using a genetic algorithm that implements an evolutionary process for neuron activation functions in the hidden layer of the model. To select the best candidate in the selection breeding process, both its computational complexity and the quality of the results obtained using it are considered. To validate the proposed approach, 4 different datasets are considered and used to evaluate the quality of different models for classification and regression tasks. Experimental results confirm that ELM model shows better results with activation functions obtained using Gene Expression Programming (GEP) algorithm than with classical activation functions used to solve similar problems.

Annotation: The paper presents the results of research in the field of trusted design of computing systems and their elements. The problem of using physically unclonable functions to protect against illegal copying of integrated circuits that are used in information computing systems for various purposes is considered. The problem of creating universal software and hardware for studying physically unclonable functions based on programmable logic integrated circuits is being solved, due to which the approach to the practical implementation of protective equipment is fundamentally changed by introducing a software add-on for correctly determining the identifiers of elements of digital computer devices. A physically unclonable function of the arbiter type was chosen as the object of study. An original system architecture is proposed and the implementation of complex functional blocks that form the basis of this system is considered. A unified hardware and software solution has been developed for collecting and analyzing data in order to find unique identifiers of integrated circuits. The format of commands for the interaction of hardware and software parts of the developed tools is considered. The hardware of the system is described in detail, including interfaces and transition graphs of finite state machines of complex functional blocks. Possible nuances that must be taken into account during design are reflected. A software solution has been developed with the help of which the results are collected, analyzed and saved in text form according to a given format. Algorithms of the main operating modes of this software are described. Examples of using the system to search for reference identifier values are given.