The article considers the problems of construction and maintenance of road facilities in the permafrost region related to the degradation of permafrost soils and the formation of soft soils at the base of structures. For stabilization of soft soils in the considered conditions it is proposed to use the method of pressure injection of soil-cement mortars in the mode of hydraulic fractures, overcoming its main disadvantages by optimizing the compositions of the mixtures used. The main direction of optimization is substantiated – reduction of the amount of cement used in mortars, which will reduce their cost, as well as the amount of associated heat generation, additionally thawing permafrost soils in the base.

Laboratory tests were supplied and carried out, which allowed to determine the dependence of the main characteristics of soil-cement injection mortars on the amount of cement (in the minimum possible range of values), as well as to determine the minimum binder concentration that provides the stability of the mixture and hardened mortar.

An algorithm for optimizing the compositions of injection mortars for stabilizing soft bases of the operating subgrade sections under the considered conditions is proposed.

The article analyzes the current state of research in the field of laser post-processing of ceramic and tungsten carbide coatings applied using thermal spraying (TS) technologies. It is shown that laser processing with surface melting is the most appropriate. Laser remelting makes it possible to obtain coatings with improved strength properties. It was found that the structure of thermal spray coatings (TSC) changed significantly, the dispersion and porosity decreased, and their performance characteristics, such as wear resistance, hardness, and corrosion resistance, increased. The reasons for the observed changes in the structural and strength characteristics of coatings after laser heat treatment (LHT) of HVOF coatings with remelting are indicated and recommendations for choosing processing modes are given.

In order to ensure traffic safety and monitor the parameters of the wagons en route, various diagnostic tools for the components are installed when approaching the maintenance depot (PTO). Currently, as part of the implementation of projects to organize the operation of "unmanned stations" on public tracks, equipment for monitoring parts and assemblies of freight wagons (body, chassis, braking equipment) is being placed. The detection of brake equipment malfunctions consists of detecting missing, dragging parts and jamming of wheelsets, for example, if the brake lever gear is incorrectly adjusted. The article considers the issue of using KTSM-03 diagnostic tools to detect increased heating of the wheelsets when the train speed decreases by a specified value within the control. The temperature ranges on the rim of the wheel of a freight car involved in the braking process is determined, depending on all possible ratios of the parameters of the braking equipment that are within acceptable operating limits. The calculated value of the maximum heating of the wheel from pressing the pads to the rolling surface will be the threshold rejection value when monitoring the serviceability of the braking system using KSTM-03 diagnostic tools.

Surface grinding machines are mainly used to reduce the surface roughness of workpieces and to obtain accurate dimensions, that is, to perform finishing operations of technological processing processes. It should be noted that the general layout of the oscillatory system of surface grinding machines, the geometric dimensions of the grinding wheels, workpieces and the methods of their fastening have significant differences from other types of grinding machines. This circumstance determines the need to substantiate the models of vibroacoustic dynamics of those elements of the general oscillatory system of the objects of study, the intensity of sound radiation of which leads to an excess of sound pressure levels at the workplaces of machine operators over the maximum permissible values. The acoustic models considered in the article allow theoretically calculating the spectral sound pressure levels of the above sources of noise discomfort at the stage of designing and installing equipment, as well as identifying frequency intervals and the values of excess over sanitary standards.

The study considers the issue of developing an automated control system for a wheel lathe to restore the rolling profile of wheel sets of freight wagons, and establishes the causes of premature failure of high-speed milling cutters of specialized wheel-rolling machines. The readiness function of the control system for the rotation speed of the machine spindle and the working tool feed was determined. The work analyzes the values of the modulus of the metal cutting speed during the restoration of the rolling profile of the wheelset and the amplitude of the self-oscillation of the working tool of the machine, as well as the results of calculations of the readiness indicators of the restored objects of the dynamic system of the wheel machine and the main characteristics of the elements of the dynamic system of the machine. The nature of the oscillations of the working tool of the machine is investigated and the factors that cause these oscillations are analyzed. A block diagram of an automated wheel-rolling machine control system is proposed and a set of basic elements of the main contour of the control system for a specialized wheel lathe is presented.

A convolutional neural network is considered for classifying surface types encountered by mobile robots during navigation tasks. The classification of five surface types encountered in Iraq is analyzed: clay, hills, potholes, asphalt roads, and concrete pavements. The architecture of the convolutional neural network (CNN) consists of three convolution blocks with normalization and ReLU activation layers, a pooling layer, a fully connected classification layer after the CNN. The training with 96.62 % accuracy convinced that it is effective. The ray plots show a sharp decrease in loss and improvement in classification accuracy, and the cross matrix confirms successful recognition of most surface types with no misclassification of hills. CNN allows these robots to adapt faster to such complex terrain by dynamically adjusting navigation paths, which significantly improves reliability and autonomy in realworld operations.

A new approach to classifying the operation modes of a railway freight marshalling yard is proposed based on objective data obtained by merging information from various sensors and station systems. A concept for classifying station operation "from the wheel", i. e. based on data from automatic reading devices, is proposed, as opposed to the existing system of calculating station efficiency based on manually entered information. The new concept is described by formalizing such key station indicators as detention of car wagon downtime, operating fleet, and train arrival rate, which are essentially quantitative variables whose values reflect station operation. The application of the proposed approach is shown using statistical data on the operation of the Chelyabinsk-Glavny station as an example, and the station operating modes identified based on an expert assessment of the calculated indicators are described. The development of the given approach as a tool for predictive analytics of the station operation and targeted development of measures to implement the effective functioning of railway marshalling yards is proposed.

This study is a continuation of the previous works of the authors and is aimed at assessing the region's needs for the development of transport infrastructure and forecasting the volume of traffic in conditions of its shortage.

A graphical-analytical model of the interaction between the transport infrastructure and the economic sectors of the region is presented and the basic principles of forecasting freight transportation in the region and assessing infrastructure shortages are described. The results of forecasting the value of transport demand in a particular region are presented and the moment of occurrence of a shortage of transport infrastructure is determined. A methodology for predicting the loading of a region with priority industry access to transport infrastructure in conditions of its scarcity is developed. The magnitude of economic losses due to the shortage of transport infrastructure in the region is determined with different options for prioritizing industries.

In modern conditions of active development of digital information and communication technologies, the management structures of logistics transport systems are being transformed. In turn, logistics transport systems in the context of the globalization of the transport services market are transformed into integrated logistics transport systems that combine the processes of cargo delivery on several modes of transport. In this case, they need to be considered as multilevel complex systems, since the management level inevitably dominates the levels of implementation of technological processes. A promising direction in the development of management structures for integrated logistics transport systems is the creation of digital information and analytical platforms, which in this case aggregate management functions. The creation of new structures for managing information flows in logistics transport systems requires the transformation of existing tools in order to match their quality and the capabilities realized through digital technologies.

This article describes the models for determining the conditions of transportation for various types of cargo by rail, road, air and sea transport. To build the models, the regulatory acts and reference books governing the process of cargo transportation, the principles of cargo classification for air, sea, rail and road transport were studied. Based on the classification, the conditions of transportation for each specific type of cargo by the corresponding type of transport were considered. Based on the analysis, a general model for determining the conditions of cargo transportation for all modes of transport was built, and a conceptual data model was defined that determines the relationships between the main objects of transportation necessary to determine the conditions. The presented models are used as the basis of algorithms of the new IT tool designed by the authors – the logistics navigator, intended for obtaining an optimal logistics route for cargo transportation using various types of transport.

The article is aimed to the development of measures to reduce losses of bulk cargo due to blow-out during its transportation by open-top cars. The characteristics of the model range of open-top cars in use are analyzed, and the most important ones for organizing the selection of rolling stock, which affect the degree of filling of the body with cargo, are determined. The characteristics of bulk cargo that influence on the loading height into an opentop car are analyzed, and the most significant ones are identified, which determine both the degree of filling of the car body and the configuration of the cargo surface.

An aerodynamic model of the process of blowing out bulk cargo during transportation in open-top cars has been developed using specialized software in the field of Computational fluid dynamics (CFD). Based on the results of the model study, an algorithm has been developed for selecting optimal options for combining the characteristics of open-top cars used for transportation and specific brands and fractions of bulk cargo, which ensures minimization of cargo losses from blowing out.

A brief historical overview of the problem of unproductive detention of cars and their excessive accumulation on the railway network is presented. Proposals have been made to solve the current problems of wagon accumulation through the creation of specialized stations and fleets for their temporary storage.

For stations for temporary wagon storage, methodological recommendations are formulated for determining the number of tracks in receiving and dispatching parks, illustrated by a diagram developed during the study for determining the coefficient of unevenness.

Circuit and technological solutions are presented for a new specialized station for temporary wagon storage, as well as for an existing large station when arranging a fleet of long-term storage of cars on it.

This article considers an approach to assessing the reliability of a simulation model of cargo delivery based on taking into account the location of agents performing technological operations during transportation. Particular attention is paid to reference points such as container loading and unloading, as well as the stages of folding and unfolding a transformable container. Accurately determining the parameters of these processes allows for a more reliable reflection of the real dynamics of the system, taking into account the interaction of various participants in the logistics chain. This makes it possible to conduct experiments with a significantly lower level of capital investment, as well as more flexibly configure the systems being studied. The article determines the level of reliability of a computer simulation model of cargo delivery in an urbanized environment, created in the Anylogic software package.

The results of computer modeling of possible faults in the electrical part of an electric locomotive based on the digital twin technology using Matlab Simulink software are presented. For approbation of the developed digital twin of the electric part of the 2ES6 electric locomotive the validation of the computer model on the basis of retrospective data measured with the help of the onboard recorder of motion parameters and autotracking (RMPA) was carried out. The results of the research can be used for machine learning of the model for predicting of the state of electric rolling stock, namely for the purpose of generating a set of data with different parameters of electric locomotive motion due to the small number of records of real trips with RMPA. Prospects of practical application of the obtained results are related to the developed system of prescriptive monitoring and diagnostics for automatic prevention of electric locomotive faults on the way.

The problem of making non-trivial decisions taking into account the temperament and psycho-emotional status of the decision maker has been formulated. This required clarification and expansion of the categorical apparatus of the study. In particular, the concepts of subject-independent and subject-dependent feature spaces have been studied. Bioinspired methods of decision making, the basic model of which is finite state machines, have been analyzed. The concept of "finite state machines with a preferential direction of transitions" has been introduced. A mechanism for identifying the feature space (FS) of the researcher has been developed, including: selection of the type of proximity measure, an algorithm for calculating the proximity measure. A scheme for training a teacher and a machine has been proposed. The possibility of immersing finite state machines in a subject-dependent feature space has been shown. Research prospects and illustrative examples are given.

The connection of the design feature of high-current semiconductor thermistors (SCT) with some characteristics of the material of their body is considered. The purpose of this research is to study the non-uniform temperature field in the mass of a thermistor, since under certain conditions this can affect the stability of its operation. The results obtained confirm the position that the temperature gradients in the semiconductor mass of the thermistor obey a linear law both with respect to current and with respect to the distance from the axis of the thermistor. As a result of the research, the theoretical statement that temperature gradients are determined only by design parameters has been confirmed. As a consequence, coaxial type thermistors meet the basic requirements for improving the reliability of the SCT.

In modern conditions of operation of the traction network, there is a need for reliable protection against short circuits. An effective solution to this problem was the introduction of microprocessor relay protections, which are integrated into programmable logic controllers. These devices have a high degree of automation and are able to quickly respond to changes in the electrical network, which significantly increases the level of safety. However, despite the high degree of reliability of such systems, there are a number of questions regarding their functionality. The article considers the possibility and expediency of functional diagnostics of relay protection by increasing the number of measuring units included in the protection system. The criteria for assessing the effectiveness of the proposed method are the probability of detecting a failure to operate a measuring unit before the moment when it should perform its functions and the probability of detecting a false (excessive) operation. The first criterion helps to assess changes in the hardware and software reliability of the protection system when implementing functional control. The second is necessary to assess the effectiveness of control from the point of view of changing the functional reliability of protection, as well as the logical and information compatibility of protection with other automation functions.

The paper considers an approach to determining the possibility of integrating a fiber-optic cable with a conductive sheath into a group grounding system as a means of increasing the reliability and safety of DC railroad operation. The criterion for decision making is the heating temperature of the cable elements, determined depending on the magnitude of the flowing current and the time of relay protection operation. As a tool for calculating the temperature of all cable elements, including the optical module with optical fibers, a program using the finite element method as the main algorithm is proposed. The initial data for performing the modeling are the current in the cable, data on the cable design, as well as the technical characteristics of the cable elements.

The article is devoted to the study of the possibility and feasibility of constructing fiber-optic communication lines of railways using fiber-optic cable built into a lightning protection cable. The research work presents the results of a comparison of the cost of one kilometer of cable used today and the cable proposed as an alternative solution. The results of business correspondence with the manufacturer of cable products, as well as information available in the public domain on the Internet, are used as a source of information in the article. The expediency of changing the type of fiber-optic cable to a conductive one is substantiated, the economic effect based on reducing the cost and increasing the reliability of communication lines is shown. This solution allows increasing the technical and economic indicators of sections of electrified railways with alternating current due to the integration of the conductive sheath of the fiber-optic cable OPGW into the traction power supply system.

An electric drive with asynchronous electric motors with a short-circuited rotor is widespread enough due to operational reliability and sufficiently high energy characteristics. The authors propose a method for further improving the energy characteristics of an asynchronous electric motor in dynamic modes with control of the amplitude of the supply voltage. Based on a mathematical model of a generalized electric machine in fixed axes α and β, an algorithm has been developed to reduce electrical energy losses in the windings of an asynchronous electric drive in transient start and braking modes. Using the maximum principle and the Newton-Raphson method, the optimal functional dependence of the amplitude of the supply voltage of an asynchronous motor on the time of the transient process has been determined. The minimum square of the voltage supplying the electric motor is selected as an optimization criterion. Based on the calculation results, graphs of changes in currents, the electromagnetic moment of the angular velocity of rotation and electrical losses in the windings of the asynchronous motor with optimal control actions are obtained.

The article considers the issues of equipping the divisions of Transenergo and the Traction Directorate with devices for mechanical removal of ice, vibrating pantographs and pneumatic drums. The method for assessing the risk of damage to JSC Russian Railways from the formation of ice on the contact network wires has been improved due to the lack of devices for removing ice from the contact wire in the required quantity in the structural divisions of Transenergo and the Traction Directorate. The levels of failure frequency and severity of their consequences due to ice formation have been proposed. Damage risk matrices and recommendations for risk reduction for each level have been compiled. Determination of the need to retrofit the divisions of Transenergo and the Traction Directorate with devices for removing ice is based on a comparison of the costs of restoring the infrastructure under current operating conditions and after retrofitting.

The article considers the structure of the automatic speed control system for electric trains. The proportional-integral control law with a mismatch prediction is used, which ensures the required quality of control in steady-state and transient motion modes. The quality criteria for speed control are determined. By means of simulation modeling in the Embarcadero C++ Builder software environment, a comparative analysis of the efficiency of the automatic control system is carried out with different structures of its control law: constant (proportional-integral with a mismatch prediction is used in all motion modes) and variable (the proportional law is used during acceleration or braking, and the proportional-integral law with a mismatch prediction is used in the speed stabilization mode). For this purpose, the problem of parametric synthesis of the system is solved using the Nelder-Mead method, as a result of which rational values of the coefficients of the control law are found for each of its structures under study. It is established that the adaptation of the system control law to a change in the driving mode due to the use of its variable structure ensures an increase in the quality of electric train speed control.

The paper considers the improvement of the system and technology of operation of traction rolling stock based on the analysis of operational data obtained by drivers of various skill levels in the process of locomotive control. To assess the specified data, objective control tools were used – on-board motion parameter recorders, microprocessor control and diagnostic systems, as well as safety and automatic driving devices. A methodology for assessing compliance with the technology of rolling stock operation has been developed and implemented, as well as a rating system for assessing drivers, which made it possible to reduce operational risks and improve the efficiency of rolling stock operation. The method has been implemented in the operational locomotive depots of the North Caucasus Traction Directorate and used in the concept of the model of differentiated selection and assignment of locomotive drivers to trips depending on train categories.

The existing problem of cracks in cast parts of freight car bogies periodically detected during operation, threatening traffic safety, requires new approaches to theoretical and practical assessment of the crack formation process in car structures. The problem of predicting crack resistance of structures at the design stage is of relevance. Method Theoretical and experimental studies of crack formation in cast elements of a bogie with an axial load of 25 t / axle were performed. Results: The sizes and configuration of initial defects, at which crack growth is possible, were determined based on the stress intensity factors at the crack tips and crack-like defects in the axle box assembly of the side frame. The conditions for crack opening and propagation were modeled and analyzed. Experimental studies of the resource of side frames of a bogie with an axial load of 25 t / axle were carried out and compared with the modeling results. Practical significance: Approaches to assessing the crack resistance of side frames of freight car bogies are proposed, allowing more accurate prediction of the possibility of crack formation in cast elements of the bogie during operation. Individual research results are used in developing new side frames for bogies with an axle load of 25 t/axle.

The modern concept of brake disc development provides for two main tasks: ease of manufacture and efficiency of operation with minimal economic costs. As a rule, both of these tasks are incompatible and have different goals. The operating principle of all ventilation devices of brake discs is the same, heat removal by passing cooled air through the channels. However, the efficiency of this physical process varies depending on the design of the disk. The influence of the opening angle of diffuser-type ventilation channels on the efficiency of the brake disc ventilation device is considered. The study presents a comparative characteristic of various, most common designs of ventilation apparatus, and a new method for solving the tasks is proposed.