Search

The article considers the theoretical foundations for assessing and forecasting the sustainability of the development of the Russian economy using artificial intelligence systems. The relevance of the study lies in the fact that in modern conditions the question of the formation of methodological approaches for a balanced sustainable development of the economy is particularly acute. The novelty of the study lies in the fact that an approach is proposed that involves the use of a certain system of macroeconomic indicators that reflect the dynamics of development of both the real sector of the economy and the parameters of the development of the fi nancial sector to predict the GDP of the Russian Federation using the "Perceptron" AI system. An analysis was made of both the dynamics of macroeconomic indicators of the real sector of the economy and the parameters of the development of the fi nancial sector of the Russian Federation. The authors considered the advantages of using the VaR model indicator as a measure of risk to assess the fi nancial risk of reducing the GDP of the Russian Federation. The "Perceptron" program was developed using the Deductor platform to predict the dynamics of the RF GDP. A wide range of fi nancial mathematics tools proposed by individual authors for assessing and minimizing fi nancial risk, including quantile hedging, defi cit hedging with minimal risk, and optimal quadratic hedging, are considered. The architecture of the neural network, in addition to the input layer, has two hidden layers, and an output layer with one parameter. The neural network "Perceptron" shows a high accuracy of the forecast.

The article considers theoretical approaches to forecasting the development of the domestic economy and the stability of the fi nancial sector. The work uses an artificial intelligence system developed by the authors. The paper puts forward and proves the hypothesis that the use of the risk assessment of the financial sector calculated using the VaR model in the neural network AI system can be used to obtain the forecast value of the country’s GDP in order to provide support for managerial decision-making under conditions of uncertainty. Of great importance is the study of the theoretical foundations for increasing the stability of the Russian economy in modern conditions. The use of neural network models in GDP forecasting is an important tool for assessing and ensuring the sustainability of the Russian economy. The novelty of the study lies in the fact that an approach is proposed to fill the gap that concerns the problem of the lack of a reliable approach in assessing the risk of financial sector stability in order to ensure the sustainable development of the domestic economy. Get a correct forecast of the gross domestic product (GDP) in the face of market uncertainty. According to the VaR model, with a 99 % probability, the absolute value of the fi nancial risk of a reduction in the profits of the banking system of the Russian Federation may amount to 108.5 billion rubles in 2022, or 4.5226 %, and the amount of profi t may reach 2.3 trillion rubles. rub. The resulting neuroforecast of the value of GDP for 2022 is 79 023.4 billion rubles. 33.3 % higher than the 59 262.95 billion rubles, declared by experts, the total forecast value of GDP for 1Q. (28410) and 2 sq. (30853). To compare the accuracy of the forecast, it remains to wait until the actual values of GDP become known. The practical significance of the study lies in the fact that in the course of the study the prerequisites for solving an important national economic problem were formed — forecasting the value of GDP and ensuring the sustainable development of the country’s economy.

The article examines the theoretical foundations for the emergence of overdue debts on loans and forecasting financial risk in Russian banks in modern conditions. The relevance of the study is that the growth of bad debts of commercial banks on loans is currently one of the most acute problems. The collected material made it possible to analyze the dynamics of the volume of overdue debt on loans in commercial banks of the Russian Federation for the period 2013–2021. In the course of the study, it was revealed that many factors influence the volume of bad debts. In order to study the influence of factorial signs on the effective sign — the amount of overdue debt, an attempt was made to use such models as: correlation-regression, AI and VaR. A hypothesis has been put forward and proved that with the help of models: correlation-regression, AI and VaR, it is possible to obtain a forecast of the volume of overdue loans in the portfolio of commercial banks of the Russian Federation. The correlation-regression model, in addition to the resultant sign Y — the growth rate of overdue debt, included such factorial signs as: X1 — the growth rate of GDP per capita; X2 — the growth rate of the average per capita income of the population; X3 — the growth rate of foreign trade surplus; X4 — inflation index; X5 — growth rate of capital outflow; X6 — growth rate of cash; X7 — interest rate on loans; X8 — US dollar exchange rate; X9 — price of a barrel of oil URLS dollars; X10 — wage growth. The study showed that the use of various forecasting tools provides different amounts of arrears, for example, for 2022: the correlation-regression model — 7159.9 billion rubles, the neural network — 4466.251 billion rubles, and the VaR model 5426 .56 billion rubles overdue loans.

The article discusses the theoretical foundations of the analysis and forecasting of financial risk in the banking sector in conditions of market uncertainty. The relevance of the study lies in the fact that the growth of problem debts of commercial banks on loans to legal entities, individual entrepreneurs and individuals is currently the most relevant and debated issue in the banking community. The analysis of the dynamics of assets and the share of overdue loans in 2010–2021 has been carried out, trends in portfolio changes have been identified. The authors considered the advantages of using the VaR indicator as a measure of risk, noting that its weak side is the inability to assess extreme losses (in the tails) if the risk is realized in the range above the confi dence interval. The Perseptron program has been developed for forecasting the dynamics of the share of overdue loans in the portfolio of a commercial bank, which is formed on the Deductor platform. Quantization (grouping) of data was carried out using a neural network on the Deductor platform, which made it possible to identify certain patterns in the change in portfolio quality. It was revealed that the value of the share of overdue loans of commercial banks is influenced by many factors, including factors included in the AI-system, so four parameters are placed on the input layer of the perceptron (%): growth in assets, market share, change in the loan portfolio, dynamics of overdue loans. The output layer has one parameter: forecast of the share of overdue loans (%). In addition to the input layer, the neural network architecture has two hidden layers, and an output layer with one parameter. The neural network shows high forecast accuracy. The error rate for Post Bank was 1.2159 %. The authors consider a wide range of financial mathematics tools proposed by the authors in order to assess and minimize financial risks, including quantile hedging, deficit hedging with minimal risk, and optimal quadratic hedging.

The article discusses the theoretical foundations for forecasting the gross domestic product (GDP) and the sustainability of the development of the Russian economy. The forecast values of GDP were calculated by two methods: based on the use of the AI-system and a multivariate correlationregression model for the purpose of subsequent comparison. The novelty of the study lies in the fact that a comparative assessment of the accuracy of the generated forecasted values of the RF GDP by different methods was carried out: using the AI system and a multivariate correlation-regression model in XLtables using the Microsoft Correlation and Regression packages. Some of the calculations were carried out in the Colab cloud environment using the Python programming language. The initial data for the AI-system and the correlation-regression model are statistical data — chain indices of the development of the main industries taken for the period 2000–2020. Studies have shown that the average forecast error calculated using a linear six-factor regression model between factorial features and the effective one was 0.153553 %, while the average forecast error calculated using the AI system was 0.010082 %. The forecasting accuracy using artificial intelligence algorithms is an order of magnitude higher than the traditional one — using a multivariate regression model. Based on the initial data, a dataset was collected, on which, using the Deductor analytical platform, an AI system was formed in order to obtain the forecast value of the RF GDP for the next calendar year. Using the capabilities of the Colab cloud service, in the Python language, the pairwise correlation coefficients between the factors were calculated, the results of which are presented in a heat matrix. Formed a correlation-regression model in XL-tables using the packages "Correlation" and "Regression" Microsoft. Also, the quality of the regression model, reflecting the dependence of the effective indicator "GDP" on the main factorial indicators, was assessed using the Student’s t-test, F-statistics and others.

The purpose of this article is to present the results of the research on the role of local action groups (LAG) in the development of rural areas in the member states of the European Union (EU). The most important postulate of the investigation is the scientifi cally grounded and practice-tested statement that only by relying on rural communities it is possible to objectively determine the pressing problems that need to be solved within the boundaries of various rural areas. Moreover, with the help of rural population (especially its active part, youth, entrepreneurs, farmers) it is advisable to implement the social, economic, environmental projects developed there, in the positive results of which both rural residents and the state are interested. The scientific work uses analytical and review methods of analysis (including the discourse method), which (together with the widespread use of modern statistical information, reporting materials of the European Agricultural Fund for Rural Development (EAFRD), the content of the relevant legislation) made it possible to formulate a number of practice-oriented conclusions. Most of them relate to the possibilities of European rural communities (institutionalized in a certain way) to solve such problems of the countryside as rising unemployment, limited fi nancing for rural development, reduced public investment in the development of rural areas, sequestration of state spending on the development of the social sphere within the boundaries of rural areas, poverty and social insulation. In general, the European experience of encouraging local participation in achieving the goals and priorities of rural development put forward by society through the support of the LEADER / CLLD initiatives under the EU Common Agricultural Policy (CAP) not only has many positive consequences, but also includes time-tested practices, worthy of implementation in the modern conditions of the domestic environment.

The paper considers the theoretical foundations for the use of artificial intelligence systems for managing bank capital. By growth rates of assets of a commercial bank, one can judge about its success and competitiveness. However, any activity of a bank is associated with the risk of growth, and, depending on a size of the bank, the parameters under consideration will differ. The aim of the study is to develop a neural network model to ensure the capital management of a commercial bank. The results obtained represent an increment in scientific knowledge, since the developed neural network algorithm provides a contribution to the scientific field regarding the use of artificial intelligence systems in solving problems in the banking sector. The hypothesis was advanced and proved that the neural network model "Kohonen Map" can be used to forecast the asset growth. Thus, for example, with the input parameters of the model: asset in 2018 - 26972302745 thousand rubles and asset in 2019 - 27735034904 thousand rubles, the projected value of growth in Sberbank's assets for the next year will be 2.8%, which coincides with the actual growth of the current year. A VaR model has been developed to assess the financial risk of building up a bank asset. The obtained value of X(1) = -4131299748 thousand rubles indicates that in the next year, the growth of assets will not exceed the value of -4131299748 thousand rubles with a 99% probability. And X(5) indicates that over the next five years, the decrease in assets will not fall below -9237867073.53 thousand rubles with a probability of 99%.

Theoretical foundations for artificial intelligence systems being applied in Big Data processing have been considered. Capabilities of the Hadoop framework and Deductor analytic platform have been analyzed and compared. A technology to mine data from news Web sites due to the developed Skraper program has been considered. There were discussed some issues, revealing capabilities of the Word2vec program that was used in scientific research and provided for the retrieved words being vectorized in a 300-dimension format.

The theoretical basis for the SAR-model applied for calculating the financial risk of time series of exchange instruments has been considered. The article describes the AI neural network for assessing financial risk in the course of exchange trading in a SiU8 US dollar futures contract on the Moscow Exchange by the MoExSAR method in order to minimize the risk.

The relevance of the research study is due to the fact that the article attempts to prove or falsify the hypothesis that the "AI-Decision Tree" neural network model makes it possible to obtain a forecast of Russia's GDP for various scenarios.Various aspects of the AI application in the field of big data processing, deep learning and forecasting have been investigated in the article. However, experience has proven that, certain issues of using artificial intelligence require further scientific research in order to achieve a balanced and sustainable growth of the financial and economic system.Theoretical foundations of sustainable economic growth in the country have been studied. The authors have reviewed modern domestic and foreign literature on the topic and paid special attention to the issues of balanced financial and economic system and sustainable economic growth in modern conditions.We demonstrate the factors increasing risk and market uncertainty and other in order to achieve a balanced and sustainable growth of the financial system based on the AI-system "Decision Tree" developed.The trends in functioning of the financial and economic system have been determined; the dynamics of the balanced profit volumes in real sector organizations has been traced quarterly for the period of 2015-2018. Live data of the Federal State Statistics Service showed that the balanced financial result (profit except for loss) of organizations (apart from small business entities, banks, insurance organizations and state and municipal institutions) in current prices decreased by 8.5% in 2017.In order to visualize the dynamics of the effective factor - GDP a neural network model "AI-quantization of data" has been developed.In order to achieve a balanced and sustainable growth of the financial system based on the AI-system "Decision Tree" developed.

The article represents theoretical foundations investigated for application of artificial intelligence systems in Big Data processing. The most comprehensive list of tools for data analysis and machine learning has been considered. A comparative Hadoop framework and Deductor analytical platform opportunity analysis has been performed.An AI-system has been proposed for predicting the cost of innovative products in the context of digitalization of the Russian economy. A hypothesis that a neural network makes it possible to obtain a forecast for the cost of innovative products in the Russian Federation has been put forward and proved. The neural network model included such parameters as GDP (billion rubles), key rate (%), RTS index, output of innovative products (billion rubles), costs of innovative products (billion rubles), dollar exchange rate (rubles), balanced profit (billion rubles), risk (σ), loans originated (billion rubles), VIX-Index and forecast for the volume of innovative products (billion rubles). The list of parameters presented reflects the development of both the economic sphere and Russia's financial sector quarterly for the period of from 2015 to 2018.Based on quantization and subsequent visualization of big data and using a multidimensional diagram, the artificial intelligence system developed allows revealing the GDP trend in Russia depending on the cost of innovative products and the VIX option stock-exchange quotation in the global economic landscape. The AI-system that enables prediction for the cost of innovative products using the "what-if" function in the Deductor platform has been developed.

Modern gas turbines operate at high temperatures, which exceed the endurance limit of material, and therefore the turbines components are cooled by the air taken from the compressor. The cooling provides positive impact on lifetime of GT has negative impact on its performance. In convection-cooled turbine blade the coolant is usually discharged through the trailing edge and it leads to limitation on the minimal size of trailing edge and thereby negatively affects the losses. Moreover, the injection of cooling air in the turbine disturbs the main flow, and may lead to additional increase of losses, and the trailing edge loss is a significant part of the overall loss in modern gas turbines. This study comprises investigations of losses in cooled blades. Four cascades with different unguided part of aerofoil with and without coolant injection were studied both experimentally and numerically. This analysis provides split of losses caused by different factors, and offers the opportunities for efficiency and lifetime improvements of real engine designs/upgrades. In particular it is shown that an increase in the unguided turning angle results in a reduction of loss in case of relatively thick trailing edge. It is also shown that injection through the trailing edge slot parallel to the main flow leads to neutral loss impact and even loss reduction in subsonic range and loss increase in the supersonic range of exit Mach numbers.

This paper presents investigation of nine tip squealer design variants based on full 3D Navier-Stokes CFD calculations. In particular two main design features have been studied: the impact of relative squealer cavity rim extension and the impact of pressure side squealer cavity rim inclination on stage efficiency. All these cases have been compared for two values of relative radial gaps 0.6% and 1.36%. Obtained numerical results were validated against the experimental data measured on the E3 blade cascade test rig given in the open literature. As the overall outcome for these numerical investigations two zones with different vortex structures and different sealing features have been found. Moreover the size of these zones determines the level of the tip clearance leakage and losses for various tip squealer designs. The obtained loss values and corresponding change of the stage efficiency level as well as flow structure details were compared for all studied cases, providing insight into turbine stage aerodynamics with respect to minimal and maximal radial clearance.

The increase of new gas turbine’s efficiency is connected with further rise of turbine inlet temperature and sometimes as well pressure. In these conditions, first cooled turbine stages of a gas turbine engine usually consist of freestanding airfoils, which do not use an integrated shroud, to avoid risk of shroud overheating. In order to better control the radial gap leakage flow between the rotating blade tip and turbine casing, special design features of the airfoil tip need to be considered in the design process to meet the best possible stage performance. In the general engineering practice, a blade tip squealer provides opportunities to control tip clearance loss. In this paper several simplified types of the tip squealer design are investigated to determine the most effective loss control. At this stage of the investigation, blade tip cooling was not taken into account, but aerodynamic effects were analysed in detail. Based on the most common designs of the blade tip in the literature, four geometry types were investigated: (i) a flat tip design as the reference baseline solution, (ii) full tip squealer, (iii) partial squealer along the pressure side (PS) wall with a cut-out at the pressure side near the trailing edge (TE) and (iv) partial squealer along the suction side (SS) wall with a cut-out at the suction side near TE. All these cases have been compared among each other for two relative radial gaps (gap to blade height) of 0.6% and 1.36%. The flow calculations were done with a full 3D Navier-Stokes CFD code. For the flat tip and for full squealer designs, numerical results were validated against well-known experimental data measured on the GE-E3 blade cascade test rig found in the open literature. By using the 3D numerical data, the special attention was considered to confirm reliability and predictive credibility of the blade tip flow obtained from the analytical model. The obtained loss values and flow details were compared for all studied cases, providing insight into turbine stage aerodynamics with respect to minimal and maximal radial clearance.

The aerodynamic loss due to tip leakage vortex of rotor blades represents a significant part of viscous losses in axial flow turbines. The mixing of leakage flow with the main rotor passage flow causes losses and reduces turbine stage efficiency. Many measures have been proposed to reduce the loss in the tip clearance area. In this paper the reduction of the tip clearance loss due to changes made to the blade tip section profile is presented. The blade tip profile was modified to decrease the pressure gradient between pressure surface and suction surface. This approach allows the reduction of tip leakage and tip vortex strength and consequently the reduction of tip clearance losses. A 3D Navier-Stokes solver with q-ω turbulence model is used to analyze the flow in the turbine with various tip section profiles. Test data of three single-stage experimental turbines have been used to validate analytical predictions: • Highly loaded turbine stage with a pressure ratio π0T = 3.2 and reaction degree ρmean = 0.5. • Two turbines with a pressure ratio π0T = 3.9. (One with high degree of reaction ρmean = 0.55; the other with low degree of reaction ρmean = 0.26). The numerical investigation of the influence of various tip section profiles on stage efficiency was carried out in the range of relative tip clearance 0.5%–2.4% with the objective of a decreasing the influence of the tip clearance on the stage efficiency.Copyright © 2010 by Alstom Technology, Ltd.