Siberian Aerospace Journal

We consider the problem of analyzing low-intensity streaming data (one measurement per day) to identify hidden anomalies in the behavior of complex objects using taxpayers' fiscal data obtained from cash registers as an example. A methodology for constructing a multidimensional feature space is proposed, incorporating statistical, structural, and dynamic characteristics of time series. The methodology is based on a system of five clearly formulated working hypotheses: a change in the stationary operating mode of the object, the concentration of the total indicator volume on a small number of time samples, a binary (two-mode) structure of the value distribution, the presence of long periods of inactivity, and increased variability of indicators. Each hypothesis is formalized as a set of quantitative features with accompanying mathematical expressions. Feature selection methods are described in detail: correlation analysis with a target variable (threshold |r| > 0.2), nonparametric Kolmogorov – Smirnov test (p < 0.05), one-way ANOVA (p < 0.01), removal of multicollinear features (|r| > 0.8), and a combined approach. A comparative analysis of eight classification models was conducted on fiscal data (2200 objects, 365 days). Combined feature selection made it possible to reduce the dimensionality from 96 to 28 while increasing the ROC-AUC from 0.85 to 0.94. Validation on an independent sample confirmed the effectiveness of the approach: the proportion of confirmed anomalies was 84 %. The proposed methodology can be scaled up to other anomaly detection tasks in technical and economic systems with low-intensity data streams.

This article develops mathematical routing models for hybrid communication systems that integrate ground, stratospheric, and space segments. The study addresses networks where topology, demand matrices, link capacities, loss levels, and delay characteristics vary at the same time. The paper aims to formulate a multilevel mathematical routing concept that combines three core ideas: fractional multicommodity flow, path-limited routing, and delay minimization. The study uses multicommodity flow models on directed graphs, path-based formulations with a bounded number of routes per demand, convex delay-aware objectives, and an analysis of modern approximation algorithms. The results show that fractional multicommodity flow defines the upper level for estimating throughput, fairness, and priority-aware service; path-limited formulations translate this solution into engineering policies that a routing plane can install and maintain; and delay-oriented models account for quality-of-service requirements and temporal dynamics. The paper also shows how this concept links routing with radio-resource allocation, structural adaptation of the network, and routing-information dissemination. The results support a multistage routing logic in which a fractional formulation estimates the theoretical upper bound, a path-limited model compresses this solution into an installable routing policy, and a delay-oriented stage refines the decision for hybrid-network operation. The proposed concept applies to the design and control of communication systems that link spacecraft, airborne platforms, and terrestrial infrastructure. The article concludes that this concept can provide a theoretical basis for routing in hybrid communication systems and can naturally extend to lossy transmission models, dynamic network scenarios, and integrated network-control problems.

The development and practical application of low-thrust rocket engines with enhanced reliability and functional performance characteristics is a promising scientific, technical, production and technological task in the creation of new-generation spacecraft and upper stages. In the process of implementing the Cooperation Agreement concluded by the State Scientific Center of the Russian Federation "M. V. Keldysh Research Center", the Siberian State University of Science and Technology named after M. F. Reshetnev, with the participation of Polychrome LLC, is conducting a complex of research and experimental work on the design, manufacture by additive 3D printing technologies and bench tests of prototypes of low-thrust rocket engines powered by gaseous fuels with a regenerative chamber cooling system.

At this stage of the joint work, two small thrust engine chambers were selected as the object of research. These chambers were manufactured using the ASTRA 420 3D printer from the heat-resistant alloy Inconel 718 in the form of monoblock products with meridional (longitudinal) and helical (spiral) cooling channels. The thermodynamic and gas-dynamic calculations, as well as the calculation of the regenerative cooling of the chambers using methane gas, were performed using the results of experimental and analytical studies conducted by the Keldysh Research Center. The article describes the sequence of development and optimization of additive printing technology for cameras, which ensures the required product quality.

The results of bench and experimental versions of two chambers with the same configuration of mixing heads and different channel orientations of the cooling system are presented in an expanded volume. When analyzing the energy and thermal parameters of the chambers obtained during fire tests, the operability of experimental designs was demonstrated, confirming the prospects of the selected technical solutions. The principal possibility of manufacturing low-thrust rocket engines using additive technology of selective laser melting from a heat-resistant alloy is shown.

The paper presents a methodology for determining the mounting places of balancing weights inside spacecraft compartments, based on the use of a combination of methods of analytical and computational geometry, mathematical programming and computer graphics. The use of balancing weights is necessary to ensure the required position of the center of gravity of the compartment and the product as a whole. When using the methodology, the problems of ensuring a minimum mass of balancing weights and reducing the labor intensity of developing options for their installation are solved to speed up the preparation and approval of design documentation. The balancing weight placement zone is considered as a set of spatial regions free from compartment structural elements and other component parts. To minimize the overall mass of the balancing weights by determining their placement locations on a coordinate grid, the balancing weight placement problem is proposed to be represented as a linear programming problem. For testing, a conical compartment of a product with a spherical bottom was used as an example. It was determined that the balancing weight placement zone should be located near the junction of the bottom and the hull shell. The configuration of the placement zone was identified, taking into account the surrounding structural elements. The coordinates for placing the balancing weights were determined, and their masses were selected. Testing has showed the performance of the proposed methodology and the algorithm based on it. Effective use of the methodology is possible with the availability of a specialized calculation software package.

Shells of rotation made of composite materials are often used as force elements of structures in the production of rocket and space technology. Composite shells have a high degree of weight perfection provided by high specific mechanical characteristics of composites. They are manufactured by the method of continuous winding of composite fibers on a mandrel of the required shape. The method is widespread due to its manufacturability and guarantees reliable provision of design parameters of the shells.

Anisogrid cylindrical and conical shells in recent years have also started to be used in RCT. For example, at Reshetnev JSC, this type of shells are used in the designs of adapters intended for launching spacecraft and satellites into orbit. Adapter structural elements are similar, but differing in purpose, dimensions and bearing capacity, they have a unique combination of a large number of design parameters, the exact determination of which every time results in a complex scientific task. The solution to this problem involves numerous design calculations of the stress-strain state, critical loads and stiffness parameters. For this purpose, a digital twin is formed on the basis of a finite element model of anisogrid shells of rotation, manufactured by continuous winding of composite fiber, with the help of which the solution of the optimal design problem is performed in interactive mode.

An algorithm and a program for constructing a composite anisohydric cylindrical shell with concentrated mass on one base and with a rigidly fixed second base are developed. Numerical analysis of stability, stiffness and stress-strain state of the structure under different variants of inertial action and at variation of parameters of its mesh structure formation is carried out with the help of FEM.

This paper addresses the problem of modeling an optimal motion trajectory of a spacecraft within a spacecraft cluster consisting of a target and a deputy spacecraft during an undocking maneuver followed by a fly-around at a safe distance. The relative motion of the deputy spacecraft is described in a local orbital reference frame using a linearized dynamical model based on the Hill – Clohessy – Wiltshire equations. An integral performance index is introduced to characterize the tendency of motion near the boundary of the admissible relative position region. The original optimal control problem with an integral cost functional and bounded control inputs proves to be challenging for both analytical and numerical analysis, which limits the direct application of standard optimization techniques. To facilitate the study, a decomposition approach is employed, allowing the problem to be divided into two consecutive stages: transfer to the boundary of the safe zone and subsequent motion along this boundary. The Pontryagin maximum principle is applied to investigate the properties of optimal solutions and to derive analytical expressions for the adjoint variables, making it possible to analyze the structure of optimal control laws. Numerical analysis is carried out using both direct and indirect optimal control methods. A set of quality criteria is introduced to assess the obtained solutions, including the fulfillment of necessary optimality conditions, the Hamiltonian behavior, and the sensitivity of the solution to variations in problem parameters. The comparative analysis shows that direct optimization methods yield more stable and reproducible solutions, while the tangential boundary approach improves the agreement between the two stages of motion at the expense of a moderate increase in the transfer time to the boundary.

The paper presents the results of plasma spheroidization of metal and ceramic powders. The aim of the study was to conduct a comparative analysis of two methods for feeding powders into a plasma jet – under the nozzle section (plasma torch F-4) and along the axis of the plasma jet (plasma torch PM-1). Metallic (Cu, NiCr) and ceramic (Al₂O₃, ZrO₂) powders were used as objects of research. To implement the spheroidization process, a specialized water-cooled reactor has been developed that provides controlled cooling of molten particles. Numerical modeling of thermal processes in the reactor confirmed the efficiency of the cooling system: the stationary thermal regime is established 286 seconds after the start of operation of the plasma torch, and the increase in the temperature of water in the reactor, measured experimentally, is consistent with the data obtained by calculation and is 5 °C. It has been established that the PM-1 plasma torch provides homogeneous and finely dispersed powders, however, when processing materials with a low melting point (copper), the powder begins to melt in the inlet channel of the plasma torch, settles on the walls of the nozzle and, over time, clogs it. Spheroidization with the supply of powder under the nozzle section makes it possible to process any materials, however, larger particles with a wide granulometric distribution are formed, especially when processing refractory powders. Based on a comprehensive analysis of the microstructure and granulometric composition of spheroidized powders, technological recommendations have been formulated for choosing the type of plasma torch depending on the required powder properties: the PM-1 plasma torch is preferred for obtaining homogeneous fine powders, and the F-4 plasma torch is preferred for processing low-melting materials and ensuring the maximum degree of spheroidization.

Spacecraft electronics and onboard computer microchips are made of semiconductors. Radiation increases in near-Earth orbit, especially during solar flares, where the flux of high-energy particles and gamma radiation increases. This leads to defects in semiconductor transistors and failure of electronic devices. Therefore, replacing field-effect transistors with spintronics, which utilizes the spin degree of freedom of electrons, is becoming a pressing issue. Transport characteristics can be controlled by a magnetic field using samarium-substituted manganese sulfides. The conductivity of a sample was studied at low current in a magnetic field applied at an angle to the current, varying from 0° to 360°. Without a magnetic field, the conductivity remains constant. When a magnetic field is applied and the sample rotates, a change in conductivity is observed. In a magnetic field, conductivity decreases and reaches a minimum within a certain angular range. Upon heating, conductivity decreases in a magnetic field and reaches one order of magnitude near the magnetic phase transition. The current-voltage characteristics of Sm_xMn_1–xS with a concentration of x = 0.1 were measured without a magnetic field of H = 0 kOe and in a magnetic field of H = 12 kOe, directed along the current and perpendicular to the current. The dependence of current on voltage is nonlinear and is associated with electrically inhomogeneous states in the sample. From the current-voltage characteristics, the dependence of the change in conductivity in a magnetic field on the current (voltage) and temperature was found. The maximum decrease in conductivity in a magnetic field was found at 200 K. Above room temperature, conductivity decreases by several percent due to the Hall contribution. Heating and increasing current lead to a decrease in magnetoconductivity. A comparison of the two methods for measuring conductivity in a magnetic field indicates that the regulation of conductivity by a magnetic field depends on the current value at which the conductivity is measured.