Method for monitoring condition of track
The system addresses the limitations of existing track condition detection methods by using vibration sensors and NB-IoT communication to analyze and compare track data in real-time, enhancing defect identification and power efficiency.
Patent Information
- Authority / Receiving Office
- RU · RU
- Patent Type
- Patents
- Current Assignee / Owner
- OBSHCHESTVO S OGRANICHENNOJ OTVETSTVENNOSTYU PROIZVODSTVENNO-KOMMERCHESKAYA FIRMA EVROKHIM REZINOTEKHNIKA
- Filing Date
- 2025-11-11
- Publication Date
- 2026-07-07
Smart Images

Figure 00000001_ABST
Abstract
Description
[0001] The invention relates to systems and methods for monitoring the condition of extended objects and can be used for remote detection of dangerous changes in the condition of a track, including rail defects.
[0002] Technology Level
[0003] Local analysis systems for loads and geometry of solid structural elements on railways are devices that allow monitoring the condition of infrastructure elements, identifying defects and optimizing track operation.
[0004] Such systems use various methods, such as:
[0005] - measuring wheel-rail contact loads. Strain gauges, accelerometers, and other devices are used to record forces and accelerations during rolling stock movement;
[0006] - Track geometry monitoring. Track measuring trolleys with sensors are used to determine the position of the rails, track width, track profile, and other parameters;
[0007] - Remote diagnostics. Laser scanners are used to scan the route, record the beam's direction and travel time, and create a 3D model of the route based on the results.
[0008] Local analysis systems for loads and geometry on railways solve the following problems:
[0009] - Identification of local faults, such as rail defects and track gauge deformations.
[0010] - Track condition assessment and forecasting. The systems help determine the need for track maintenance and speed restrictions.
[0011] The monitoring results are used to plan track maintenance and repair work, as well as to resolve emergency situations. Measurement results are also used to optimize alignment, tamping, straightening, ballast cleaning, and other work.
[0012] An acoustic method for detecting track faults while a train is moving along a railway is known. According to this method, acoustic pulses are generated "on board" the rolling stock, transmitted through the wheelset into the rails, the presence of reflected pulses is recorded at the point of their formation, based on the fact of their occurrence, a decision is made about the faulty track of the train ahead (RU Patent No. 2126339, IPC: B61K 9 / 10, published on 20.02.1999). A disadvantage of this method is the low reliability of fault detection, caused by the presence in the rail of powerful vibroacoustic signals from the interaction of wheels with joints and defects in the rails, which allows the detection of only those defects in the rails that are characterized by the high reflectivity of vibroacoustic signals.
[0013] A known method for monitoring the critical state of rolling stock on a track and recording its derailment is based on continuously recording vertical and transverse oscillatory accelerations on bogies and acoustic emission energy from the wheel-rail contact zone, comparing the recorded functions with background ones, and, based on the comparison results, identifying the rolling stock state as "normal," "critical," or "derailment" (RU Patent No. 2399524, IPC: B60T 7 / 12, published September 20, 2010). However, the time for stopping a moving train when transitioning from the "critical" to the "derailment" mode, based on real-world conditions, is insufficient.
[0014] Of the known technical solutions, the closest in terms of the set of essential features to the claimed method for monitoring the condition of a track is the invention “Method for remotely detecting changes in the condition of a track in front of a moving train” (RU Patent No. 2490153, IPC: B61K1 9 / 08, published on 20.08.2013), based on the recording of vibroacoustic signals at different points of the track, the analysis of these signals and the adoption of a decision based on the results of the analysis, characterized in that only vibroacoustic signals from the interaction of a wheel with a rail joint at equally distant, if possible, joints of the rail line from it are recorded, the signals recorded when a train passes such a joint are subjected to filtering and accumulation operations, the signals obtained after these operations from the joint when the first train passes a known good track are corrected until maximum similarity between them is obtained, and the parameters of the correcting circuit are preserved, when the next train passes the joint - the signal generator, the corrected signals are compared, and the result of the comparison is compared with the established threshold, upon exceeding which a decision is made about the presence of a change in the track in front of the moving train.If there is no train on the monitored section of track, the recording circuits are switched to a low-sensitivity mode, which is characterized by actuation only when a moving wheel interacts with the joint at the location where the recording circuit is installed. Based on these actuation signals, the recording circuits closest to the joint along the track are switched to a high-sensitivity mode.
[0015] However, recording and analyzing vibrations specifically at track joints during a train's passage does not provide sufficiently accurate results regarding the condition of the track section under investigation. This is due to gaps between rails and the presence of connecting elements (pads), preventing timely detection of track defects, particularly in real time. A disadvantage of this method is the difficulty of detecting foreign objects on the track due to avalanches, rockfalls, quicksand, etc., as well as the results of sabotage operations, including the installation of munitions. Furthermore, the method is difficult to implement.
[0016] Concepts and Definitions
[0017] LoRa (Long Range) is a long-range wireless communication technology designed to transmit data over long distances with minimal power consumption and low data transfer rates.
[0018] NB-IoT (Narrow Band Internet of Things) is a cellular communications standard for low-bandwidth telemetry devices.
[0019] Accelerometer is a device that measures the acceleration of an object relative to its initial position or the change in its speed.
[0020] Stray currents are currents generated in the ground when it is used as a conductive medium. The main sources of stray currents in underground metal structures are electrified railways (main and suburban), trams, industrial, quarry, and mine transport.
[0021] Vibration acceleration is the acceleration of movement of the controlled point of the equipment or the rate of change of speed.
[0022] A gyroscope is a device that can respond to changes in the orientation angles of the body on which it is installed relative to an inertial reference frame.
[0023] A vibration sensor is a device that generates an electrical signal proportional to the measured vibration parameter. When measuring vibration parameters, the following sensors are used: proximometers – for measuring vibration displacement; velocimeters – for measuring vibration velocity; accelerometers – generating a signal proportional to vibration acceleration.
[0024] A microcontroller (MCU) is a microcircuit for software control of electronic devices. It is typically manufactured as a single crystal containing the functions of a microprocessor core, command and data buses, peripherals, RAM, and ROM. Essentially, it is a computer on a chip, acting as a peripheral processor.
[0025] Random Access Memory (RAM) is the volatile portion of a computer's memory system that stores executable machine code (programs) as well as input, output, and intermediate data processed by the processor. It is a temporary data storage device that provides rapid access to information for the processor.
[0026] A server is a device that provides services or resources to other devices, called clients, over a network. The primary function of a server is to process client requests and provide them with the necessary information or capabilities. Typically, a server is a powerful computer capable of processing and storing large amounts of information. Special software is installed on the computer, which gives it server functionality. Depending on the server's purpose, specific software is installed on it to meet the necessary requirements and requests. Typically, a server is a dedicated or specialized computer for running service software; a network computer that processes requests from other computers on a local or global network.
[0027] Flash memory is a solid-state, semiconductor, non-volatile, and rewritable memory capable of retaining data even after power is removed. It is used in a variety of devices, including USB drives, SSDs, memory cards, and embedded systems in routers, televisions, and IoT devices.
[0028] A real-time clock (RTC) is an electronic circuit designed to record chronometric data (current time, date, day of the week, etc.), consisting of a system consisting of an independent power source and a recording device. Real-time clocks are most often found in computers, although, in fact, RTCs are present in virtually all electronic devices that need to keep time.
[0029] Ethernet is a family of data transmission technologies used to build local area networks (LANs). It provides physical and channel data transfer between devices within a single network segment or over switched / routed networks.
[0030] The purpose of the proposed technical solution is to improve the quality and reliability of remote detection of dangerous changes in the condition of the track in real time while simplifying the system and method for monitoring the track condition.
[0031] In order to achieve the stated objective, the method for monitoring the condition of a track based on recording vibration signals at different points of the track, analyzing these signals and making a decision based on the results of the analysis consists in the fact that devices for recording physical parameters in an amount of at least one, located under the rails at an equal distance from each other, record vibration signals from a passing vehicle using vibration signal recording sensors, carry out their preliminary processing using a microcontroller and send them to a temporary storage facility, where data packets corresponding to passing vehicles are formed from them, after which the data packets are returned to the microcontroller and, using a wireless communication device, the devices for recording physical parameters send data packets to a server configured with the possibility of processing, analyzing and storing data, on which they are compared with nominal or other,The values established as reference values for a given track are compared to determine the condition of the monitored track section, identify its defects, and then send information about the condition of the monitored track section to the end user's user device. The condition of the monitored track section is determined based on data from vibration sensors. Processed data from the processing device (microcontroller) is sent to a temporary storage device in CSV files, each file containing data for one vehicle. The server synchronizes the device clocks and the operation of the track defect monitoring device components. Monitoring is performed in power-saving mode, periodically, at set intervals. Control commands from the user device are sent directly to the physical parameter recording device.
[0032] The essence of the invention is explained by the following description and the accompanying diagrams.
[0033] Figure 1 shows a block diagram of the claimed device.
[0034] Fig. 2 shows the conventional placement of the sensor under the sleeper.
[0035] Figure 3 shows a tram connected to a sleeper via an elastic support as a triangle. The sleeper is connected to the ground via elastic ties. These ties are not rigidly defined and can be adjusted over time. F1, F2, and F3 are the forces generated by rail defects, determined by the system. Knowing the sleeper's support reactions, the system allows one to determine the forces acting on the design model.
[0036] Fig. 4 shows an image of the propagation of vibrations from a passing tram.
[0037] Where:
[0038] I – device for recording physical parameters;
[0039] 1 – vibration sensor (accelerometer, gyroscope, etc.);
[0040] 2 – microcontroller;
[0041] 3 – temporary data storage device (flash memory);
[0042] 4 – NB-IoT communication module;
[0043] 5 – power supply device (battery);
[0044] 6 – DC / DC stray current charger;
[0045] 7 – Local Area Network (LoRa) module;
[0046] 8 – data analysis tool (server);
[0047] 9 – user device.
[0048] Static System Description
[0049] The method for monitoring the condition of a track consists in the fact that devices for recording physical parameters (I) in a quantity of not less than one, located under the rails at an equal distance from each other, record vibration signals from a passing vehicle using vibration signal recording sensors (1), carry out their preliminary processing using a microcontroller (2) and send them to a temporary storage facility (3), where they form data packets corresponding to passing vehicles, after which the data packets are returned to the microcontroller (2) and, using a wireless communication device (4), the devices for recording physical parameters (I) send data packets to a server (8) configured with the possibility of processing, analyzing and storing data, on which they are compared with nominal or other values established as reference for a given track, based on the results of the comparison, the state of the monitored section of the track is determined,its defects are identified, after which information about the state of the monitored section of the route is sent to the user device (9) to the end user.
[0050] The condition of the monitored track section is determined, in particular, based on data from the vibration sensor (1). Processed data from the microcontroller is transferred to a temporary storage device, implemented in the form of flash memory, in "csv" format files, each file containing data for one vehicle. The server (8) synchronizes the clocks of the devices and the operation of the track condition monitoring system elements. Track condition monitoring is performed in power-saving mode, periodically, at set intervals. Control commands to the microcontroller (2) can be sent via an additional communication device (7) directly to the physical parameter recording devices (I).
[0051] The method for monitoring the condition of the track is implemented as follows.
[0052] At least one physical parameter recording device (I) shall be placed on the track, for example under the sleepers, to prevent their damage or loss, at equal distances, every 20–100 m, which is determined by the distance covered by a specific vibration sensor, taking into account some overlap of their action. Additional devices may be installed on the following track structural elements: sub-ballast vibration-isolating mats, sleeper pads, reinforced concrete slabs, sleepers, profiles, and clamps. Each device (I) contains at least one sensor (1) for recording vibrations from passing vehicles, which is connected to a microcontroller (2), which in turn is connected to wireless communication devices (4, 7) with a power source (5, 6), and a temporary data storage device (3).The physical parameter recording device (I) can be implemented as a unit containing structural elements mounted on a substrate. The physical parameter recording device (I) can include various sensors that capture vibration signals.
[0053] When a vehicle moves, the wheelset impacts the rail, causing vibrations that reflect mechanical track defects. These vibrations are transmitted to the sleeper and, through the sleeper, to the vibration sensors (1). The vibration characteristics depend on the type of defect—cracks or track wear. Furthermore, the presence of track defects can be indicated by the temperature of the rails when the vehicle passes, which can also be taken into account when determining track defects.
[0054] Vibration sensors (1) can be mounted on the module board and include a three-axis gyroscope and a three-axis accelerometer with an analog output and wide bandwidth, as well as an ambient temperature sensor. Together, these sensors provide sufficient data to identify track defects. The vibration sensors (1) – accelerometers – are capable of measuring acceleration: 2g, 4g, 8g, and 16g – with a resolution of 18 bits. The actual accuracy of the measured acceleration value is 1 mg. This allows for vibration analysis in the frequency range of 0…2000 Hz.
[0055] The sensors detect vibrations from a passing vehicle. The received data is sent to a microcontroller (2) based on 32-bit Arm® architecture. The microcontroller is housed in a housing and contains RAM.
[0056] The microcontroller (2) communicates with the displacement sensor (1) via the SPI protocol ( EnglishSerial Peripheral Interface, SPI bus – serial peripheral interface, SPI bus), configuring it as needed and receiving data from it.
[0057] The microcontroller (2) receives measured data from the sensor (1), processes the received data, records it in its RAM and transfers it to a temporary data storage device (3), implemented in the form of flash memory, with which the microcontroller also communicates via the SPI protocol.
[0058] In the temporary data storage device (3), the received data from a passing vehicle is accumulated and formed into a data packet, which is returned to the microcontroller (2). Forming a data packet in the temporary data storage device (3) optimizes the data collection process and relieves the controller (2), thereby increasing the reliability of the method.
[0059] Sensors (1) measure track vibrations caused by passing vehicles at a frequency specified by the microcontroller (2). The microcontroller (2) uses software to determine how much information should be stored in flash memory (3) before a specific data packet is generated, based on specific criteria for track condition monitoring purposes. The microcontroller (2) controls all components of the physical parameter recording device (I).
[0060] The microcontroller (2) of the physical parameters recording device (I) sends the received formed data packet to the remote server (8) via the NB-IoT communication module (4). The microcontroller (2) communicates with the server (8) via the UART protocol ( English The Universal Asynchronous Receiver-Transmitter (UART) uses NB-IoT (Narrow Band Internet of Things) data transmission, a cellular communications standard for low-volume telemetry devices. Since the NB-IoT module's output power does not exceed 23 dBm, power losses in the control system are minimized.
[0061] After the data packet is successfully sent to the server (8) and confirmed by the server, the sent data is deleted from the flash memory (3), making room for other data.
[0062] The microcontroller (2) controls the sensors (1), configures the device elements, in particular the sensors (1) (determines the frequency at which vibrations from passing vehicles are recorded), the temporary data storage device – flash memory (3), communication modules (4, 7), receives the set (synchronized) time from the server (8), sends data packets to the server (8), and sets the "sleep" power saving mode using software. The microcontroller (2) monitors the time, configures the system elements (1, 3, 4, 7) for switching on and off, distributes which units and when to turn on or switch to low-power mode to optimize energy consumption.
[0063] The server (8) receives data packets from the microcontroller (2) of the physical parameters recording device (I). The server (8) can operate using the computer program "LANGETEK "E" System for Local Analysis of Loads and Geometry of Solid Structural Elements" (registration No. 2025683690, registration date: 09 / 05 / 2025), which ensures the operation of the server (8), receives data, compares them according to the rules with the nominal values established for the tracks, which are in the server (8) database, identifies deviations, determines their compliance with the track condition categories and their defects and notifies the user device (9) of the identified discrepancies in the condition and defects of the track. The data received from the sensors (1) can be compared with the data obtained during the track test, corresponding to the established requirements, taken as the reference (nominal). The reference data is entered into the database.After receiving data packets from the microprocessor (2), the server (8) processes and analyzes them using the installed "LANGETEK "E" program. The program is designed to monitor the condition of the track and rolling stock based on vibration analysis. The program receives vibration data packets, enters them into a database, interprets the data, compares them according to the rules with the nominal values established for the track, identifies deviations, determines their compliance with track condition categories and defects, and notifies the user of any identified track condition discrepancies and defects.The "LANGETEK "E" program evaluates the status of data received from the physical parameter recording device (I) and, if packets containing errors or inaccuracies are received, returns them to the processing unit (2) of the physical parameter recording device (I), implemented in the form of a microprocessor (2) via a wireless communication device – module (4) – for correction. If the data is correct and no defects are detected, the server (8) acknowledges receipt of the data packet, processes it, compares it with data on the nominal condition of the rails (nominal values) stored in the database, and identifies rail defects and their locations. The database contains data on vibrations corresponding to the normal operating condition of the track. This data is obtained either as a result of inspection and data collection from the track during its commissioning, or on the basis of calculated and / or standard data.
[0064] The nature of the received vibrations allows us to identify track defects. For example, in the presence of "wave-like rail wear," the vehicle's movement causes vertical vibrations. In the presence of a "step defect," the sensor records a sharp increase in vertical acceleration. Other track defects are also identified by changes in vibration characteristics.
[0065] To ensure the smooth operation of the device, the microcontroller (2) receives and analyzes the response (data receipt confirmation) from the server (8), synchronizing its real-time clock with the time specified in the server's response, which is considered the system's precise time. All clocks in the system's components operate in accordance with the time set by the server. All components of the physical parameter recording device (I) are synchronized with each other with an accuracy of 0.001 seconds. The server (8) determines the operating mode and conditions of all components of the device.
[0066] The data received by the server (8) is entered into the database.
[0067] After processing the received data, the server (8) receives conclusions about the detected track defects or their absence and sends the information to the end user (9). The collected track condition data is converted into the required information presentation format (image, graph, diagram, map, etc.) using the appropriate program and transmitted to the user's receiving devices (9).
[0068] The device's operation is ensured by communication modules (4, 7). The communication module (4) ensures the connection and data transfer from the physical parameter recording device (I) to the server (8) and vice versa. The communication module (7) receives control signals from the user device (9) and enables the operational control of the working elements (1, 2, 3) of the physical parameter recording device (I), bypassing the control server (8).
[0069] The control system, specifically the physical parameter recording device (I), is powered by a lithium-ion battery (5) and a stray current charger (6). Stray currents are extracted using a copper pin driven into the ground with a contact, and the second contact is attached to the rail. Stray currents are present in the rails, which the device uses to power itself. These currents are influenced by many factors, including soil moisture, distance from the power supply station, and tram power.
[0070] The microcontroller (2) monitors the battery voltage (5) and, if critically discharged, does not perform any actions—measurements, data transfer, etc.—until the charge is restored, but instead switches the device to low-power mode. All device components are powered by the battery through a 3.0-volt linear regulator. The charger (6) uses rectified and smoothed stray currents to charge the battery.
[0071] The device is in power-saving mode almost all the time. Only the real-time clock, which is located in the microcontroller (2), is operational. When a specified mode is reached, for example, searching for vehicles, in particular trams, the sensors (1) take readings at the specified time intervals. The monitoring period is set by the administrator, the system user (9). The frequency depends on the section's occupancy. As a rule, the minimum occupancy during tram track monitoring is two trams per hour. The microcontroller (2) switches the vibration sensor (1) to vibration search mode, setting its vibration response threshold, and then returns to sleep mode. When a vehicle passes, the vibration sensor (1) wakes the microcontroller (2) from sleep mode, which reads from it at the specified frequency and writes the data to flash memory (3) for storage.After measuring data from a specified number of vehicles, such as trams, the measured data is sent to the server (8) using the NB-IoT communication module (4).
[0072] If necessary, the user device (9), controlled by the administrator or another user, changes the frequency or number of measurements, or changes the wake-up and sleep times in the program by sending the corresponding commands to the server (8). In this case, the physical parameter recording device (I), upon the first connection to the server, receives updated settings from the server (8) or using the local LoRa communication module (7) - a radio module - when the administrator is directly located next to the device. This radio module (7) is also designed to update the sensor software. The radio module (7) allows the administrator to connect to the device located nearby and change its settings, bypassing the server. The user device (9), represented by a user with certain access or an administrator, can set the number of passing trams per day (from 1 to 255 units).) on the server (8) either using special software programmer or via LoRa (7).
[0073] The data analysis tool (8), implemented in the form of a server, can be connected to at least two devices for recording physical parameters (I), the number of which is determined by practical necessity and the technical capabilities of the system elements.
[0074] The claimed method allows to determine
[0075] - wear of the working surfaces of the rails;
[0076] - weld defects;
[0077] - track width widening;
[0078] - unacceptable angles in the plan;
[0079] - curvature of paths in the horizontal plane;
[0080] - vertical differences;
[0081] - vertical vibration;
[0082] - horizontal vibration;
[0083] - noise level;
[0084] - break in rail strings.
[0085] The control method can be implemented using standard devices. The method has been tested on tram infrastructure facilities in Moscow and cities in Western Siberia.
Claims
1. A method for monitoring the condition of a track based on recording vibration signals at different points of the track, analyzing these signals and making a decision based on the results of the analysis, which consists in the fact that devices for recording physical parameters (I) in an amount of at least one, located under the rails at an equal distance from each other, record vibration signals from a passing vehicle using vibration signal recording sensors (1), carry out their preliminary processing using a microcontroller (2) and send them to a temporary storage facility (3), where data packets corresponding to passing vehicles are formed from them, after which the data packets are returned to the microcontroller (2) and, using a wireless communication device (4), the devices for recording physical parameters (I) send data packets to a server (8) configured with the possibility of processing, analyzing and storing data, on which they are compared with nominal or other,values established as reference values for a given route, based on the results of the comparison, the state of the monitored section of the route is determined, its defects are identified, after which information about the state of the monitored section of the route is sent to the user device (9) to the end user.
2. A method for monitoring the condition of a track according to paragraph 1, characterized in that the condition of the monitored section of track is determined based on data from vibration sensors.
3. A method for monitoring the condition of a track according to paragraph 1, characterized in that the processed data from the microcontroller (2) is sent to a temporary storage device (3) in “csv” format files, each file containing data from one vehicle.
4. The method for monitoring the condition of a track according to paragraph 1, characterized in that the server means (8) for processing, analyzing and storing data synchronizes the clocks of the devices and the operation of the elements of the track defect monitoring device.
5. A method for monitoring the condition of a track according to paragraph 1, characterized in that the monitoring is carried out in energy saving mode, periodically, at set periods of time.
6. A method for monitoring the condition of a track according to paragraph 1, characterized in that control commands from the user device are sent directly to the device for recording physical parameters (I).