The research paper presents the synthesis of a fuzzy quasi-optimal controller model and an analysis of its effectiveness compared to a known train speed controller for short-term deviations from the specified operating mode. The task of controlling an underactuated system is of particular importance for railway transport, especially for high-speed transportation. Mechanical systems as control objects are essentially nonlinear dynamical systems of high order. In addition, the complexity of optimizing the operating modes of such systems is due to the fact that even detailed modeling does not accurately predict the cumulative effect of all dynamic factors acting on a dynamic system under operating conditions. Traditionally used in practice linear control laws with constant coefficients are designed to stabilize only one specific mode of motion, which makes them ineffective in conditions of control deficit and a priori uncertainty. Using the reduction of the Lagrange optimization problem to the isoperimetric one makes it possible to obtain a quasi-optimal solution to the structural synthesis problem, which increases control efficiency compared to known methods. The use of the fuzzy logic apparatus allows for parametric synthesis of control, providing adaptability to a priori uncertain operating conditions.

The paper considers the problem of placement, specialization, and development of a network of repair sites for special self-propelled rolling stock as a spatially distributed complex of railway transport infrastructure. The relevance of the task is substantiated. A formal formulation of the placement problem in a multi-period form is proposed, which differs from the previous one by simultaneously considering a binary tolerance filter, the profitability threshold for the production site, and an explicit axis of planning periods, which allows for the description of seasonal unevenness in the repair campaign. The equivalence of the problem (when the binary activation variables are removed) to the minimum cost flow problem on a time-extended network is proven, and the inclusion of activation variables and the breakeven point transforms it into a problem of minimizing the cost of a flow with a fixed cost of an object. The operational and tactical level (integer linear programming) is complemented by the strategic level of multi-criteria selection of a network development scenario using hierarchical analysis and ordering based on proximity to the ideal solution with robust assessment of the range of the track maintenance program. The approach has been tested under the conditions of the Eastern Polygon of JSC Russian Railways. The sustainable result is the targeted priority for the development of the eastern part of the network.

The patterns of changes in equivalent resistance during parallel connection of additional conductors to active-inductive circuits are investigated. Based on analytical dependencies, cases of connection of an active resistance, an active-inductive element without an inter-inductive coupling and conductors connected by mutual induction are considered. The calculations performed have shown the existence of a parameter range in which the connection of an additional conductor leads to an increase, rather than a decrease, in the active component of the equivalent resistance. The boundary conditions for the occurrence of this effect are established, and its effect on the current distribution in the system is shown. The results obtained are extended to a traction network with a shielding wire. It is proved that the use of a shielding conductor does not always provide a reduction in active power losses. With certain correlations of the parameters of the track circuit and the shielding wire, it is possible to increase the active resistance of an equivalent reverse current circuit, despite a decrease in the total network resistance. The conclusion is made about the need for an individual assessment of the effectiveness of the use of shielding wires for each specific section of the railway, taking into account the parameters of the traction network and its operating modes.

The article is devoted to the development of a mathematical framework of an automated system for monitoring the weight of overhead contact wires intended for early detection of ice and frost formations. Previous studies have derived relationships between the branch deflection angles, compensator parameters, and measured forces for two- and three-block compensating devices, and substantiated the minimum number of strain gauges required for accurate ice load monitoring. This paper transitions from relationships for the branch deflection angles to formulas directly linking the additional ice load to the force recorded by the strain gauge. It is shown that the difference in load transfer coefficients in these systems should be taken into account when selecting sensor sensitivity and configuring automated wire weight monitoring (AWM) data processing algorithms.

In the context of climate change, ensuring the sustainable operation of transport infrastructure located in permafrost zones has become particularly urgent. For structures operated according to Principle I (maintaining the frozen state of the ground), rising average annual temperatures create the conditions for permafrost degradation. This leads to the development of cryogenic deformations and reduction in the reliability and safety of the structures. This article presents the results of geocryological modeling of the "structure-foundation" system on the Chum-Labytnangi railway section in the Yamalo-Nenets Autonomous Okrug, conducted using specialized software packages. The possibility of extending the basic geocryological model to other sections of transport infrastructure was confirmed by comparing numerical modeling results with engineering-geocryological monitoring data; including temperature measurements in boreholes not exceeding 0.1°C. Based on the analysis of temperature fields and deformations, a criterion for the feasibility of adaptation measures is substantiated. The obtained results have practical significance for substantiating a long-term strategy for adapting transport infrastructure to climate change.

The railway track is analyzed as a complex technical system, the management of which is based on regulatory approaches that limit the predictive control of critical sections. A set of system analysis methods has been applied: structural decomposition, functional and structural analysis, classification and multi-criteria analysis. The hierarchical interrelation-ships of the elements of the upper and lower structures have been revealed, and the cascading nature of degradation has been established, where the malfunction of the underlying element acts as the primary factor in the destruction of the overlying ones. The division of components into critical (providing instant safety) and fundamental (determining long-term stability) is justified. A multilevel life cycle model is proposed, which establishes a discrete scale of states correlating with the probability of failure. The concept of a hybrid digital twin integrating stationary sensor networks and data from mobile diagnostic complexes has been developed.

This study quantitatively evaluates the combined effects of moisture content, compaction coefficient, and the number of freeze-thaw cycles on the deformation properties of clayey soils. The study was conducted under laboratory conditions. Soil samples with varying moisture contents ranging from 14% to 25% and compaction coefficients from 0.93 to 1.06 were prepared and subjected to varying numbers of freeze-thaw cycles (1–6 cycles). Control samples not subjected to freeze-thaw cycles were also tested. All samples were tested in a compression tester with loads ranging from 0 to 0.6 MPa. Based on the processing and analysis of test reports, dependences of sample deformations on load were obtained for different combinations of initial soil sample properties. It was noted that all the factors considered influence the increase in deformations, and the relationship between deformations and moisture content for all compaction coefficients considered increases with increasing freezing cycles, while the relationship between deformations and compaction coefficients decreases across the entire moisture content range.

One of the goals of railway transport digitalization is to create digital twins of wagons, which allow for the management of their life cycle. The initial stages of creating a digital twin of a wagon include, firstly, the development of digital models of all modifications of the main components of a specific model of wagon and, secondly, the expansion of the functions of these models to ensure the ability to predict the performance of wagon components under specific operating conditions with specified parameters and possible defects, which makes these models digital shadows of specific components of a specific wagon. In addition to predicting the performance of the units and, consequently, the entire car, their digital shadow allows for the reasonable identification of the causes of malfunctions that prevent the operation, derailments, accidents, and crashes, which can be used in forensic examinations in disputes between business entities regarding liability for the occurrence of malfunctions or the investigation of incidents. The initial stage of developing a digital shadow of a gondola is the creation of a parametric digital model of the body of a gondola with cargo. The issue of using the results of a commercial inspection of a gondola with loose cargo to model the position of the total center of mass of the body and cargo is being considered. An algorithm for determining the maximum load on the spring set coils of a gondola bogie has been proposed. The found value of the maximum load on the spring is used to calculate the stress-strain state of the spring and determine the safety factor according to various criteria.

This article proposes the architecture and operating algorithm of an automated system for assessing the residual life of NB-514E traction electric motor (TEM) insulation on 3ES5K electric locomotives using operational load cyclograms. The need for such a system arises from the fact that the periodic insulation resistance monitoring and local monitoring of individual parameters used in practice do not allow for the consideration of the motor's thermal history and the timely assessment of accumulated thermal deterioration of the insulation. The system's computing core utilizes a digital twin of the thermal state, enabling the calculation of overheating in controlled and hard-to-reach areas, the generation of virtual temperature estimates, the comparison of calculated and standard values, and the prediction of the remaining life. The system structure includes an operational data collection subsystem, a cyclogram generation module, a data comparison unit, and a decision-making module. A distinctive feature of the proposed approach is the use of the locomotive's actual current load to periodically update the TEM's thermal state and the subsequent transition from overheating diagnostics to a predictive assessment of insulation deterioration. The result of the work is a formalized structure and algorithm of the system, oriented towards application in tasks of monitoring, diagnostics and support of operational decisions.

The railcar maintenance and repair system (MRS) includes scheduled repairs (major repairs) and routine repairs and maintenance between scheduled repairs (hereinafter referred to as inter-repair periods, or IRPs). This paper examines the MRS of refrigerated rail-cars for potential improvement. Based on an analysis of the existing repair system, the scheduled repair intervals are varied. An optimization problem is solved, in which the best combination of these intervals is determined based on the adopted criterion. This makes it possible to both improve the economic efficiency of railcar operation and extend its service life. A new term, "unit profit per mileage" (hereinafter referred to as UPM), is introduced, which denotes the ratio of the total profit from railcar operation to its overall service life. This indicator, characterizing the economic efficiency of railcar operation, will be adopted as a criterion for optimizing the MRS of refrigerated railcars.

The article presents the results of a statistical analysis of the microclimate parameters (temperature, the environmental heat load index – EHL-index) in the cabins of electric locomotive drivers of the EP1M, 2ES5K and VL-80 series. The study is based on data of a special assessment of electric locomotive crews’ working conditions carried out at different times of the year. The methods of descriptive statistics and data visualization revealed that the empirical distributions of the studied parameters differ significantly from the normal one and represent a mixture of several components, each of which corresponds to a certain class of working conditions and is close to the normal distribution. It has been established that the effect of a heating microclimate poses a greater risk to locomotive crew members compared to the cooling one. The results of the work serve as the basis for the development and validation of mathematical models of heat and mass transfer, as well as for the design of microclimate normalization systems and devices in drivers' cabins to ensure workplace safety and safe railway operation of rolling stock.

The article is devoted to the urgent task of improving the efficiency of a modern domestic AC electric locomotive of the “Ermak” series. This is the basis for heavy-haul train traffic on heavily loaded railway lines in the Eastern Polygon (Trans-Siberian Railway and Baikal-Amur Mainline). In conditions of difficult mountainous terrain and protracted descents, the efficiency of these electric locomotives during regenerative braking plays a special role.

In the course of the study, a problem was identified that limits the braking characteristics of an electric locomotive through the control channel of a rectifier-inverter converter at the level of 3.5 zones due to the design specifics of the construction of electrical power circuits. A proposed solution allows for expanding the voltage regulation range of a rectifier-inverter converter beyond the accepted limit, up to a full four zones.

To confirm the effectiveness of the proposed method's implementation, the calculation of its operation was carried out for a modern domestic AC electric locomotive of the 2ES5K “Ermak” series. The proposed method will improve the locomotive's braking performance, increase its downhill speed, and increase the return of electrical energy to the overhead contact system.

The article considers the problem of improving the operational efficiency of traction rolling stock based on the analysis of information on completed train rides performed by locomotive crews and obtained from on-board movement parameter recorders, microprocessor-based control and diagnostic systems, as well as safety and automatic train operation devices. The paper proposes a formulation of the problem of rational selection of a train operation model and assignment of a locomotive crew or an automated control system algorithm to a trip, taking into account the train category, train operation parameters, and the priority of particular control quality criteria. The possibility of representing this problem as a multicriteria decision-making problem and a combinatorial optimization problem is shown. The practical significance of the approach is associated with forming a basis for an automated decision support system under the conditions of digitalization of locomotive operation.

The study is devoted to the development of a methodological approach to the digital transformation of transport and logistics business processes in regional clusters, taking into account subcontracting relationships and risk management. Using the Sverdlovsk region as an example, the impact of digitalization on the efficiency of transport and logistics business processes under subcontracting conditions is analyzed. The methodology includes systematic analysis of cluster structures, economic and mathematical modeling (linear programming, stochastic risk management models) and correlation-regression analysis. The results demonstrate that the volume of loading on the Sverdlovsk Railway for January-November 2025 amounted to 114.7 million tons, investments in the transport and logistics sector in the first half of 2025 exceeded 50 billion rubles. The developed hybrid model for managing transport and logistics business processes integrates traditional optimization methods with artificial intelligence technologies, the Internet of Things and blockchain. The projected economic effect is estimated at 40-60 billion rubles in annual savings in logistics costs with GRP growth of 5-6%.

The article examines methods for determining key reliability indicators applicable to assessing cargo placement and securing methods in railcars. It provides examples of calculations for determining the reliability indicators of approved methods of approved cargo placement and securing methods based on statistical data on identified commercial malfunctions and failures. The article also proposes a tool for monitoring the reliability of cargo placement and securing methods based on the results of transportation. Based on the processed statistical data and key reliability indicators, it is proposed to issue a conclusion regarding the further extension of the validity period and the online conversion of local and unspecified technical specifications.

The article presents the author's approach to the selection and evaluation of classification features of hub transport organizational and technical systems (HTOTS) based on the proposed conceptual framework. The classification is based on five key features that reflect the dynamic and synergistic properties of HTOTS: the stage of the life cycle, the type of development dynamics, the ability to self-organize, the type of attractor, and the phase state. To determine the weighting significance of the classification features, the method of hierarchical analysis (MHA) by T. Saaty is used, which allows for a mathematical justification of their contribution to the integral assessment. The consistency ratio (CR = 0.017) confirms the consistency of the expert judgments. A hierarchical system of private indicators (23 indicators) has been developed to ensure objective evaluation. Based on the proposed methodology, fifty HTOTS of the Russian Federation were quantified, their integral development indices were obtained, and a ranking and comparative analysis were performed for the purposes of strategic planning of transport infrastructure development and management decision-making.

The article discusses the improvement of the process model for managing the fleet of empty freight cars in Russian Railways. An analysis of the current model, developed based on the Regulations for the process of planning the performance indicators of the fleet of empty freight cars on the railways of JSC Russian Railways No. 1397/r dated 02.07.2025, was conducted, which revealed two systemic shortcomings: the verification of the throughput and processing capacities of the infrastructure using a dynamic model of infrastructure loading only at the final stage of reviewing applications for railway cargo transportation, as well as the lack of a formalized mechanism for selecting stations for empty car parking after unloading. To eliminate the identified shortcomings, the authors proposed measures to transfer the data input to the DM ZI to the first stage of processing applications and introduce a new structural element «Control over unloading and further work with empty car fleet» into the process model. Implementation of these measures will reduce application review times, decrease empty mileage, increase section speed, line throughput, and station handling capacity.

The article deals with an urgent issue related to improving the accuracy of calculating the capacity of railway transport facilities. To solve this problem, the authors have proposed a new approach to determining the loading of switch throat – in addition to the occupation of movement and scheduled maintenance, the developed algorithm will take into account structural (technological) losses. The essence of the algorithm is an element-by-element analysis of the occupation of switch groups – time intervals are considered when the switch was not actually occupied by movement, but could have been occupied if no incident had occurred. Using the method of simulation modeling on real objects, the verification of the new approach was carried out. A new approach was verified using the simulation method on real objects. A comparison of the results obtained using various calculation methods is provided: analytical (according to the current Instruction of JSC Russian Railways No. 545/r), model-based, and model-based calculations using both the model and the proposed algorithm.

This article examines the prohibition of placing railway stations on reverse curves in existing design standards. Previously applicable station and junction design standards on Russian railways, as well as international and domestic experience in operating railway stations on reverse curves, are analyzed. A comparative analysis of regulatory documentation establishes a cause-and-effect relationship between the presence of inverse curves and deterioration in visibility, and identifies the conditions under which the use of inverse curves was permitted.
A brief overview of modern technical means for compensating for reduced visibility is provided.
A conclusion is reached regarding the advisability of revising current design standards.

A model is proposed for assessing the residual life of structural elements under random vibration loading, based on a one-dimensional convolutional neural network (1D-CNN) that takes the stress-process maxima distribution as input and outputs the relative durability. A training set of 12,000 realizations of stationary and non-stationary random processes – including processes with shock impacts and real operational records – was generated; the reference durability values were obtained by the rainflow method with linear damage summation, and the hyperparameters were tuned by Bayesian optimization (minimum error RMSE = 0.0403). Experimental verification on MA-15 light-alloy specimens tested on an electrodynamic shaker at three loading levels yielded a prediction error of 6 to 25 %, increasing at high levels owing to low-cycle fatigue. Taking the scatter of the experimental data into account, the model is considered verified.

This study investigates the structural features, mechanical, and tribological properties of a multilayer metal-ceramic coating (CrAlSiN/Cr) deposited by physical vapor deposition (PVD) onto 40ХН steel substrates. The coating's architecture comprises a Cr adhesive layer, a nanostructured functional CrAlSiN layer, an intermediate Cr layer with up to 12 alternating layers, and a final CrAlSiN layer. These layers are designed to increase load-bearing capacity and reduce crack propagation under sliding contact. The microstructure, elemental composition, hardness, and elastic modulus were investigated using scanning electron microscopy, energy-dispersive spectroscopy, and nanoindentation. Tribological tests were conducted using a ballon-disk setup, with an Al2O3 ball serving as the counterweight. These tests allowed us to determine the friction coefficient and wear rate under dry friction.
Preliminary tribological tests have confirmed the stability of the friction coefficient and high wear resistance, which is due to the multilayer architecture and optimal mechanical properties of the material. The alternation of hard CrAlSiN ceramic layers with ductile Cr interlayers creates an effective system of internal barriers that prevent crack propagation and localize plastic deformation. The combination of structural features and mechanical properties of the CrAlSiN/Cr coating renders it an exceptionally effective solution for components operating under high loads and limited lubrication.

The paper considers the vibration process of a fine-module gear transmission operating in the tooth profile separation mode. The main groups of factors determining the vibration behavior are analyzed: excitation sources (manufacturing and assembly errors, kinematic eccentricity, shaft rotation irregularity, wear) and the inertia–stiffness parameters together with damping. A dynamic model of the gear pair is developed and a system of differential equations is derived that accounts for variable damping, backlash, and the multivalued kinematic error function. The system is simplified for engineering calculations, yielding an expression for the meshing force with an explicit dynamic load component. It is shown that the likelihood of transition into the profile separation mode increases as the rotation speed rises and the transmitted torque decreases.

When constructing structures with a foundation consisting of piles, pile drivers are used. This process consists of several steps: 1– preparing the site; 2 – transporting the piles to their placement location; 3 – installing the pile in the hammer; 4 – driving the piles at designated points. These tasks require the use of a range of mechanized equipment, including loaders and transport vehicles. A disadvantage of this process is the limited automation of the equipment. It is proposed to automate the technological process by using an automated loading and transport unit with manipulators. The design of the unit is proposed. The technological process of pile driving with loading and transport operations is described. The kinematic calculation of the manipulator's executive mechanism is performed relative to the specified position. The dependence of rotation angles of the manipulator's links and the changes in the speeds of the executive drive rods is found.

Video sequences redundant frame removal currently plays an important role in various video preprocessing applications related to video classification, change detection, and object recognition (e.g., defects, faults, human actions, etc.). This paper proposes a video sequence subsampling method based on estimating the similarity of two consecutive black-and white frames using a "sign-sign" correlation function and associativity coefficients, verifying the compliance of the calculated metric values with threshold intervals, and making decisions regarding frame removal. A generalized scheme of the proposed method is considered with a detailed description of the stages, and an algorithm for finding similarities and differences between two consecutive black-and-white frames of video sequences is described. This method reduces the size of video sequences by removing redundant or repeated frames, thereby reducing the amount of stored information and increasing the efficiency of detecting key moments in video images obtained in uncontrolled environments. Research results are presented, demonstrating that the proposed method can halve frame redundancy without losing important information or improving the quality of video sequences.