Вестник УГАТУ

Применение технологий генеративного искусственного интеллекта к задаче календарного планирования производства

2025-12-01T07:30:46+00:00

The paper addresses the problem of production scheduling under conditions of high dynamics, uncertainty, and resource heterogeneity. Traditional approaches based on centralized algorithms and ERP/MES systems often overlook individual characteristics of equipment and personnel, as well as poorly formalized technological constraints, which reduces the efficiency and adequacy of generated schedules. To overcome these limitations, an original hybrid architecture is proposed that integrates a multi-agent system (MAS) with generative artificial intelligence (GenAI) technologies. Building upon previously developed methodological foundations, a formal MAS model is defined as a tuple comprising resource agents, consumer agents, an ontological knowledge base, semantic constraints, and local/global optimization criteria. The key innovation lies in embedding a large language model (LLM) via the Retrieval-Augmented Generation (RAG) architecture, which enables the system to generate adaptive scheduling strategies, provide human-interpretable explanations, and simulate failure scenarios – while minimizing the risk of hallucinations. A two-stage agent interaction algorithm is proposed: in the first stage, a feasible schedule is constructed based on resource availability; in the second stage, the schedule is optimized through intelligent negotiations facilitated by generative AI. Experimental validation in a simulation environment modeling a machining shop floor demonstrates the effectiveness of the proposed approach. The results confirm that integrating generative AI into multi-agent systems creates a cognitive layer that enhances not only scheduling performance but also user trust – a critical factor for the successful deployment of AI in real-world production and management processes.

Digital Twin of GTD as a new level of modeling capabilities

2025-12-01T15:23:28+00:00

The article describes the current state of information technology in the aircraft engine industry, and shows that the GTD digital twin is a single learnable digital system that includes a set of techniques and mathematical models that describe an engine throughout its entire life cycle: design, testing, production, and operation. The possibility of using the developed methods and tools for computer-aided design, selection of parameters, materials and structures, analysis of control programs for aviation GTE at the early stages of development, for debugging engines during testing,
for diagnostics and assessment of the condition of GTEU (based on GTE) is shown

doi 10.54708/19926502_2026_30111110

Development automation for microgrid modeling agents based on visual programming of hierarchical state machines

2025-12-05T03:43:52+00:00

The article is devoted to the development of a multi-agent system for studying the processes of joint operation of microgrids. An important role in working with such systems is played by the ability to reduce the cost of implementation and redesign of individual agents. The purpose of the presented research is to develop a toolkit for automating the creation of agent behaviour based on extended hierarchical state machines. The toolkit includes a visual diagram editor, a module generator for agent behaviours, and a library of software modules. The key feature of the presented approach is the separation of typical agent functionality into components. Combining the automata paradigm with visual tools and code generation greatly reduces the time required and the necessary expertise of the subject specialist. The toolkit has been successfully used to automate the simulation of interactions between microgrids.

doi 10.54708/19926502_2026_30111122

Methodology for developing a hybrid power plant with an electromechanical energy converter and a permanent magnet generator

2025-12-09T11:25:42+00:00

The article presents a comprehensive methodology for designing hybrid propulsion systems for advanced aircraft, based on the use of high-efficiency electromechanical energy converters with permanent magnets. A comparative analysis of the circuit solutions for propulsion systems, as well as Russian and foreign developments, is conducted. The article also demonstrates the experience of implementing an integrated propulsion system in the field of aviation electric machines. Special attention is given to the determination of primary parameters, the optimization of electromechanical characteristics, and the integration aspects of creating hybrid propulsion systems. The article presents the results of experimental studies that confirm the advantages of the proposed solutions and identify issues for further research.

doi 10.54708/19926502_2026_30111131

Experimental and mathematical modeling of the inflation process for an aerodynamic braking system based on a small spacecraft with a CubeSat form factor

2025-12-24T17:31:33+00:00

This paper investigates the promising application of an aerodynamic braking system based on a small spacecraft with a CubeSat‑type form factor to address the issue of anthropogenic debris in near‑Earth space. A mathematical model of the inflation process was developed for the aerodynamic braking system’s envelope, which is made of polyamide film. Based on the obtained data, experimental studies were conducted in a vacuum chamber to evaluate the system’s operability under conditions simulating the space vacuum.

doi 10.54708/19926502_2026_30111139

Comparative assessment of high-performing electromechanical converters for application as an integrated generator

2025-12-13T10:28:53+00:00

The paper evaluates the types of electromechanical converters used in the aviation industry for application as a high-specific converter for generating electricity. Based on the comparison, a synchronous electric machine with permanent magnets is chosen as the most promising high-specific converter, as it can achieve a higher efficiency with less weight and size. The paper analyzes its components and materials used, highlighting their advantages. Based on the list of materials presented, an electromagnetic calculation was performed for a generator capable of operating at temperatures up to 350°C, with a power overload of 25% exceeding that of a number of existing aircraft generators, while its specific power is more than 8 kW/kg.

doi 10.54708/19926502_2026_30111148

Research on aerodynamic braking technology for deorbiting small spacecraft

2025-12-24T17:12:51+00:00

This paper presents a scientific and methodological approach to addressing the problem of anthropogenic debris in near‑Earth space caused by small‑sized objects, using an aerodynamic braking system integrated into the design of a small spacecraft with a «Cubesat» form factor. A review and analysis of the applicability of existing and prospective methods for space debris removal has been conducted, including the use of optoelectronic and harpoon systems, gravitational catchers, electrodynamic tether systems, and aerodynamic braking systems. An assessment has been performed of the orbital lifetime of a spacecraft in low Earth orbit when employing an aerodynamic braking system, taking into account the influence of aerodynamic factors and solar activity parameters.

doi 10.54708/19926502_2026_30111155

Graph models and operations in evaluating microservice solutions

2026-01-14T18:31:40+00:00

Microservice architecture played an important role as a dominant factor in different software system design, it offers scalability and the ability to maintenance. The increasing complexity of microservice-systems faces significant challenges in understanding and evaluation. This paper proposes a formal and mathematical representation of microservice architecture using graph theory, where each microservice is represented as a vertex and each communication or dependency between two microservices is represented as a directed edge. Mathematical model based on adjacency matrices was introduced, graph theory algorithms is explored. Graph metrics were studied to analyze structural properties, find critical nodes, and detect potential failure points. Other operations are also explored such as analysis of path, finding and detecting cycles, and connectivity evaluation, these operations are important for testing, reliability assessment, and fault tolerance in distributed systems. The proposed approach is tested on a case study of a hypothetical microservice system. The recommendations suggest that graph-theoretic modeling provides a powerful foundation for the formal analysis, monitoring, and optimization of microservice architectures.

doi 10.54708/19926502_2026_30111165

Methodology for estimating the power of a turboprop engine as part of a parallel scheme hybrid power plant with a capacity of up to 800 kW

2026-02-02T09:09:26+00:00

The paper presents a methodology for estimating the power of a turboprop engine as part of a parallel scheme hybrid power plant with a capacity of up to 800 kW. The specific fuel consumption on the throttling coefficient rating is analytically calculated. The dependence of the mass-to-power ratio on the power is given. A demonstration calculation of the power as part of the GSU for the M-101T aircraft has been performed. Based on the methodology, recommendations are given for the use of hybrid power plant on aircraft with a take-off power plant capacity of up to 800 kW.

doi 10.54708/19926502_2026_30111175