Large-scale systems control
ISSN (print): 1819-2440, ISSN (online): 1819-2467
Media registration certificate: ЭЛ № ФС 77 - 44158 от 09.03.2011
Founder
Institute of Control Sciences of the Russian Academy of Sciences
Editor-in-Chief
Novikov D.A., Academician of RAS, Doctor of Sc.
Frequency / Access
6 issues per year / Open
Included in
White list (4th level), Higher Attestation Commission List, RISC, MathNet
最新一期
编号 116 (2025)
Systems analysis
The task of increasing labor productivity in machinery production
摘要
The results of developments to solve the problem of increasing labor productivity in medium and large enterprises in the basic sectors of the Russian economy are presented. A formal and substantive formulation of the problem is given, and an approach to solution based on determining the efficiency coefficient is outlined. An analysis of known approaches to digital control and optimization of production processes is carried out. Based on the analysis, the conclusion is substantiated that increasing labor productivity, as a rule, is not considered an end in itself. This target process is used to generate indicators of production intensity. The globalization of strategic planning is changing the situation and making planning for sustainable growth in labor productivity for systemically important enterprises in demand. In the process of solving the problem, an information and logical scheme for the formation of complexes of measures to increase labor productivity was developed. In this scheme, activities related to manual labor, automated labor, administration and document flow are distinguished into separate types. In this scheme, activities related to manual and automated work, administration and document flow are distinguished into separate types. Activities of these types correspond to various subtasks of the main task. In this case, the following subtasks are identified: personnel development, increasing motivation, improving tools, developing the provision of resources. Based on the scheme under consideration, two sets of measures are formed and compared in the main directions of achieving the goal.
Large-scale systems control. 2025;(116):6-30
6-30
The problem of redistributing quotas for training between structural divisions of a higher educational institution
摘要
The problem of redistributing quotas for training among structural divisions of a higher education institution is considered. The problem of redistributing quotas has a wide range of applications, in particular, in preventing environmental pollution. The article assumes that a higher education institution has several structural divisions. The management of the higher education institution allocates some quotas for training to structural divisions (SD). The head of each SD can give part of the quota of his SD to another division, receiving some compensation in return. In this case, if the quota is not given, it must be fully used in the joint venture to which it was allocated. The head of each joint venture strives to maximize his gain, expressed by his objective function. For simplicity, it can be considered that the goal of the head of each joint venture is to maximize the amount of funds accumulated in the centralized fund of this joint venture. In this case, the head of the joint venture can manage the share of the quota allocated to him, which he wants to give or receive from another joint venture. It is shown in which cases all structural divisions manage to receive optimal quotas for them as a result of redistribution. Three cases are considered: when the total value of the quota that the divisions want to give is greater than, less than or equal to the total value of the quota that other structural divisions want to acquire. Algorithms for redistributing quotas in each of these cases are proposed. The study is conducted analytically for a particular type of input functions, which are taken as power functions. Numerical examples are given and an analysis of the results obtained is given. A number of meaningful conclusions are made.
Large-scale systems control. 2025;(116):31-46
31-46
Clustering of financial message texts
摘要
The paper is devoted to the problem of clustering financial message texts by machine learning algorithms. Clustering algorithms can be used to identify groups of similar financial messages, identify the same type or suspicious ones, and use the clusters found rather than the message texts themselves in further analysis. Clustering algorithms such as K-means, DBSCAN and the Hierarchical Clustering method are used in the work. Information about bank transactions is used as texts of financial messages in the work. Due to the fact that bank transactions are subject to strict accounting rules established by the Bank of Russia, it is possible to introduce a metric for assessing the quality of clusterization. This metric allows you to rank the quality of clustering using machine learning algorithms, as well as select the parameters used in training these models. Special attention in the article is paid to the specifics of the data used, and how these features can be taken into account in the practical part. In the practical part of the paper, the results of using clustering models are presented, indicating the optimal parameters of these algorithms. In conclusion, it is concluded that the best clustering algorithms are applied to financial texts.
Large-scale systems control. 2025;(116):47-67
47-67
Mathematical control theory
Decomposition of spectral expansions of the quadratic lyapunov function on state-space variables
摘要
Spectral decomposition of the quadratic Lyapunov function is already known. The modal contribution and modal interaction are the main components of this decomposition, which form the basis of Lyapunov modal analysis. This paper presents the results of a further expansion of these spectral components into individual state variables and into their pairwise combinations. The obtained result can also be considered as a decomposition of the quadratic Lyapunov function not only by elements of the spectrum of the dynamical system (by modes), but also by elements of the state space. On the basis of the proposed decomposition method, new Lyapunov modal analysis indices are formulated, which allow one to evaluate the contribution of individual eigenvalues or their pairwise interaction, but in connection only with the part of the external perturbation that is associated with a particular state variable or a pair of such variables. This, in particular, makes it possible to comprehensively evaluate the joint influence of both the mode and the associated state variable on the energy of the system output signal. It is assumed that the main area of application of the new decompositions will be related to the problems of the model order reduction of the large dynamical systems.
Large-scale systems control. 2025;(116):68-89
68-89
Control systems analysis and design
Empirical analysis of interactions of players of the russian oligopoly banking market based on conjectural variations
摘要
The oligopolistic Russian market of bank lending to individuals is considered from the point of view of interaction of players. The Herfindahl-Hirschman index is calculated, proving that the Russian banking market is oligopolistic. The interaction of large banks is described using volumetric hypothetical variations, that is, players' assumptions about the influence of their actions on the behavior of competitors regarding the volumes of loans issued to individuals. Volumetric hypothetical variations in the work are calculated on the basis of empirical data. Statistical data for calculations are taken from official reports of banks. In this case, two cases are considered. In the first case, hypothetical variations of the largest oligopolist in the market in relation to the competitive environment, that is, to other market players or to other banks, are considered. In the second case, hypothetical variations of the largest oligopolist in relation to each individual competitor, that is, a bank, are considered. For both cases, the dependences of the players' volumetric hypothetical variations on the shares of banks' lending volumes are presented. The main result obtained is that the largest bank's volumetric hypothetical variations are positive, i.e., competitors will respond to the bank's expansion of lending volumes by expanding their own lending volumes and vice versa.
Large-scale systems control. 2025;(116):90-115
90-115
Networking in control sciences
The role of the structural properties of local networks in opinion formation
摘要
Many agent-based models of social influence, starting with the classical French - Harary - DeGroot and Voter models, describe the opinion formation processes as a sequence of local subsequent interactions between agents, in which the focal agent’s opinion (influence object) updates subject to its current state and the opinions of the social surrounding, which are taken with certain weights. These weights reflect how influential agents are and may depend on their social status, demographic characteristics, and the strength of ties between agents. At the same time, parallel sociological theories (the theory of social epidemics, the theory of structural proximity, the theory of structural diversity) postulate that the strength of influence can vary depending on the composition of the agent's social environment, as well as the structure of connections herein. In this paper, we test these competing theories using longitudinal data from the social network VKontakte. These data describe the opinion dynamics of a large-scale sample of users (~6 500 000) on a political topic. We study how users with radical political (conservative or liberal) views influence people with moderate opinions. We show that an increase in the number of friends with radical views leads to an increase in influence strength. At the same time, an increase in the density of connections in most cases either has no effect influence strength or consistently leads to a decrease in influence strength, a result that supports the theory of structural diversity.
Large-scale systems control. 2025;(116):116-134
116-134
Control of social-economic systems
Learning in osipov – lanchester models
摘要
Traditionally, in the Osipov – Lanchester models, which describe the dynamics of the number of combatants in terms of differential equations, the combat effectiveness of the parties is considered constant over time. In this paper, we consider the role of learning in modifying the linear and quadratic Osipov – Lanchester models, taking into account two effects: a one-shot introduction of a trained reserve by the initially losing party and the acquisition of combat experience by the parties, which affects the combat effectiveness coefficients. The problems of determining the optimal moment for introducing such a reserve so that its size is minimal, but ensures a "parity"; finding the minimum learning rate of the initially losing party, also ensuring a "parity"; and determining the optimal duration of reserve training before its input, are solved.
Large-scale systems control. 2025;(116):135-153
135-153
Modeling the impact of export restrictions on domestic market prices
摘要
The article considers the problem of analysis and selection of measures of state regulation of the activities of exporting companies. This problem is considered on the example of the grain industry, in particular, wheat. Grain markets, as a rule, have an oligopolistic nature. We consider two types of market participants: many exporting companies and the state, which exercises the function of the market regulator. The goal of the state is to regulate domestic prices for critically important types of products. We use an approach to modeling that takes into account the interaction of market participants and allows us to coordinate the interests of independent producers and the state. A feature of our approach is that each player chooses a strategy for its behavior in the market within the framework of a certain optimization problem, which are related to each other through common variables. A mathematical formulation of the problem under study is given and methods for its solution are proposed. The results of scenario modeling and analysis of the impact of export restriction parameters on domestic prices are presented using the example of the wheat market.
Large-scale systems control. 2025;(116):154-180
154-180
System dynamic model of industry: growth factors and analysis of optimal strategies
摘要
The paper presents a system dynamic model of economic flows in industry developed in IThink. The model allows one to study the influence of investor strategies and state tax policy on industrial growth rates taking into account depreciation of fixed assets and profitability. Modeling has shown that industry is based on the potential for nonlinear growth, but acceleration requires significant time and conditions. An important factor is the level of wear: with high depreciation, a significant part of investments is directed to maintaining capacities, and only net investments ensure expansion. It has been revealed that the effectiveness of state policy is determined by the planning horizon: with a short-term horizon (up to 15 years), it is profitable for the state to set high taxes, but with an increase in the horizon to 30 years, the situation changes - a tax rate of 20% becomes optimal, contributing to the growth of the industry and tax revenues. The key factor of growth is the behavior of investors: with a short-term strategy, they accumulate profits, but lose growth potential and receive significantly less profit in the long term. Long-term strategies of investors allow for a multiple increase in total income. Such investors are beneficial to the state, since they contribute to the sustainable growth of the industry, modernization, and an increase in the tax base. Thus, the coordination of long-term strategies of the state and investors is a condition for industrial development.
Large-scale systems control. 2025;(116):181-202
181-202
Asymptotic representation in stochastic volatility models
摘要
This work is devoted to the study of the asymptotic behavior of the probability density function of stock prices within models with stochastic volatility. The main focus is on generalizing previously known results for the one-factor Heston model to more complex cases, including multivariate models. The research methodology is based on the use of affine principles, which allow analyzing the necessary aspects of asymptotics through solving corresponding Riccati equations near the critical point. The application of high-order Euler estimates ensures accuracy in calculations and robustness of conclusions. Additionally, the method of steepest descent combined with Tauber's principle is used, providing an opportunity to extract precise data about the asymptote of the original function from the analysis of its transformation near the critical point. This allows a deeper understanding of the structure of probability density functions in the context of complex stochastic systems. The obtained results are important for the development of the theory of stochastic differential equations and can find applications in various fields such as financial mathematics and econometrics.
Large-scale systems control. 2025;(116):203-231
203-231
Control of technological systems and processes
Mathematical models of organizational mechanisms for implementing technologies for individual monitoring of the condition of aviation equipment products
摘要
The paper considers the coordination of economic interests of the participants in the maintenance and repair of aircraft and their systems, when changing the approach to their technical operation, due to the application of technologies for individual monitoring of the condition of aircraft objects. The introduction of such technologies is expected to increase the durability of aircraft products and extend their actual lifetime. From the economic point of view, this may cause a decrease in the profit of developers and manufacturers of aircraft products, as well as organizations engaged in their maintenance and repair. The paper considers the economic mechanisms of implementation of these technologies, allowing to ensure their attractiveness both for the above organizations and for the operators. Thus, mutually beneficial price ranges for modified products are determined, and a method of their realization together with maintenance and repair works in the format of selling aircraft operation at a fixed rate per flight hour is proposed. Simulation modeling of scheduled flights in several regions together with aircraft maintenance and repair is carried out. A comparison of the costs of flight fulfillment at different effects from the introduction of such technologies in ensuring the quality-of-service provision has been made. The paper shows that when several operators are combined into a pool in order to determine a single sales rate for all operators, it is possible to offer a single sales rate lower than when operators are serviced separately.
Large-scale systems control. 2025;(116):232-270
232-270
Vehicle control and navigation
Approaches to speed planning for ground-based autonomous vehicles
摘要
The article provides an overview of speed profile planning methods for ground-based unmanned vehicles. Autonomous driving technology is in the active development phase, and many tasks have already been solved, but with insufficient quality for safe operation and scaling of the technology. Thus, this field is promising from a scientific and engineering point of view. The paper examines in detail the formulation of the problem, describes various types of constraints to the final solution, and considers the construction of the task's functionality, depending on the requirements and methods used. Special attention is paid to the ST graph as a tool for modeling interaction with other traffic participants. Practical scenarios of agent interaction in various situations that often arise in traffic are also analyzed in detail. The authors focus on a class of methods for constructing a velocity profile along a known geometric trajectory. Two families of approaches are considered: dynamic programming and reducing a continuous problem to a quadratic programming problem. The existing modern methods based on these approaches are analyzed in detail, and improvements and improvements are proposed in the context of various additional requirements for the solution. The article serves as a methodological basis for developers of autonomous driving systems.
Large-scale systems control. 2025;(116):271-297
271-297
Modeling of plane-parallel motion of a vessel based on a nonlinear dynamic system with varying parameters using the perturbation method and its experimental verification
摘要
The analytical solution of a nonlinear dynamic system, which is a model of plane-parallel motion of a vessel, was obtained by the perturbation method. The solutions obtained allow us to take into account the influence of all the main parameters on linear and angular velocities. Small parameters are analyzed, the choice of which is determined by the specific parameters of the task. The influence of linear velocity on angular velocity due to the nonlinearity of the back is shown. The influence of the static moment on each of the velocity components is shown. The influence of salinity and temperature and, consequently, density on the parameters of the system and, as a result, on the speed of the vessel is considered. The effect of salinity is considered in the form of a piecewise constant function and a piecewise linear function. A modification of the solutions for velocities is shown, which must be done to clarify their values in different sections of the trajectory. An analytical and numerical method for modeling the plane-parallel motion of a vessel is proposed, taking into account the influence of heterogeneity of the water basin on the basis of hydrometeorological data. The obtained solutions can be used to automate the movement of a vessel along a limited fairway using an automatic control system.
Large-scale systems control. 2025;(116):298-320
298-320
Simulation tools
Routing models and methods of information resources on the content delivery network based on domestic software
摘要
Content Delivery Networks are designed to accelerate the loading of websites, applications and other digital content for users located in different regions. The main goal of such networks is to reduce the delay in data transmission between the server and the user, ensuring fast content delivery regardless of the user's geographic location. Import substitution of content delivery networks is a relevant and important scientific and technical task. The paper proposes a method for efficient distribution of digital information resources in a content delivery network implemented on the domestic software product 1C:Bus. The method is based on the problem of optimizing the route for placing content on caching servers, which corresponds to the problem of the minimum path in a directed graph. To solve the problem, we used the ant colony method to find an efficient solution in the computationally complex problem in an acceptable time. Compared with random route generation, the proposed route generation methods can speed up content loading by up to 20%. Multithreaded transmission made it possible to speed up content loading by two more times.
Large-scale systems control. 2025;(116):321-341
321-341
Compensation of interference in airborne electromagnetic systems and modeling of the geological environment by the inversion method in the processing of airborne electrical survey data
摘要
Various methods of induced interference compensation for the EQUATOR system and several modifications of the EM4H onboard electromagnetic system were compared. Two modifications of the EM4H were investigated: the system with a transmitter loop mounted on the fuselage of the An-3 aircraft and the system with a loop towed by the Eurocopter AS350B3 helicopter. Two methods of modeling transmitter signal interference are considered: in the form of a stationary systematic component of measurements and in the form of a stationary field vector rigidly connected to the transmitter. To implement the second approach, two additional dipoles are used in EM4N and EQUATOR to determine the relative positions of the transmitter and receiver. The analysis was performed for data obtained during the flights at high altitudes with no response from the ground. The following statistical parameters of signals remaining after interference compensation were analyzed: standard deviation and the difference between the minimum and maximum values. It is shown that the best compensation result is obtained by an approach that takes into account the movement of the receiver relative to the field source (transmitter). The next step was to provide interpretation of data with the objective of estimating important geophysical characteristics. For each of studied systems Kalman filter algorithm was applied providing resistivity maps, which have been compared with respect to resolution.
Large-scale systems control. 2025;(116):342-365
342-365