Multi-source data fusion rail train speed monitoring and early warning method, system and equipment

Abstract

The invention discloses a multi-source data fusion rail train speed monitoring and early warning method, system and device, and relates to the related field of rail traffic safety monitoring, and the method comprises the steps: obtaining the line key parameters of a train test line, and constructing a train test line electronic map; matching a sensor array according to scene features of different operation sections; in the running process of the train, the real-time position and speed of the train are obtained through a real-time matching sensor, and multi-source sensor data are obtained; the ground dispatching center carries out fusion correction to obtain a corrected position and a corrected speed; performing prediction to obtain a predicted speed value and a position at the next moment; according to the train operation scene corresponding to the predicted position, scene speed limit parameters are extracted, and if the predicted speed value is not met, a train-ground cooperative early warning mechanism is triggered to conduct early warning. The technical problems that existing rail train speed monitoring and early warning are inaccurate in monitoring and not timely in early warning are solved, and the technical effect of improving the accuracy and timeliness of rail train speed monitoring and early warning is achieved.

Description

Technical Field

[0001] The present application relates to the field related to rail transit safety monitoring, and in particular to a rail train speed monitoring and early warning method, system and equipment using multi-source data fusion. Background Art

[0002] Rail train speed monitoring and early warning are crucial for ensuring train operation safety. Accurate speed monitoring and early warning can effectively prevent dangerous situations such as train overspeeding and reduce accident risks. Currently, the main approach to solving rail train speed monitoring and early warning issues is to monitor train speed and position using a single sensor or a simple combination of sensors, and then issue simple early warnings based on the monitoring data. However, due to the complex and changing operating environment of rail trains, single sensors or simple combination sensors are significantly affected by environmental interference, making monitoring data prone to errors. Furthermore, the lack of effective fusion of data from different sensors leads to inaccurate monitoring results, making it difficult to fully reflect the actual operating status of the train. This in turn affects the accuracy and timeliness of early warnings, making it impossible to meet the increasingly stringent requirements for train operation safety.

[0003] Among the current related technologies, rail train speed monitoring and early warning have technical problems such as inaccurate monitoring and untimely early warning. Summary of the Invention

[0004] The present application provides a rail train speed monitoring and early warning method, system and equipment with multi-source data fusion, adopts the method of obtaining key line parameters to construct an electronic map of the train test line, matches the sensor array according to the scene characteristics of different sections of the electronic map, determines the real-time matching sensor, and when the train is running, the real-time matching sensor obtains the real-time position and speed of the train to form multi-source sensor data. The ground dispatching center fuses and corrects the multi-source sensor data to obtain the corrected position and speed, makes a prediction based on the corrected position and speed, obtains the predicted speed value and position at the next moment, matches the speed limit parameters of the scene corresponding to the predicted position, and triggers the vehicle-ground collaborative early warning mechanism if the predicted speed value is not satisfied. The technical means and the like achieve the technical effect of improving the accuracy and timeliness of rail train speed monitoring and early warning.

[0005] The present application provides a rail train speed monitoring and early warning method using multi-source data fusion, comprising: obtaining key line parameters of a train test line, and constructing an electronic map of the train test line based on the key line parameters; matching a sensor array based on scene features of different operating sections in the electronic map of the train test line to obtain real-time matching sensors; obtaining the real-time position and real-time speed of the train through the real-time matching sensors during train operation to obtain multi-source sensor data; a ground dispatching center fusing and correcting the multi-source sensor data to obtain a corrected position and a corrected speed; making a prediction based on the corrected position and the corrected speed to obtain a predicted speed value and a predicted position at the next moment; matching and extracting scene speed limit parameters based on the train operation scene corresponding to the predicted position; and triggering a vehicle-ground collaborative early warning mechanism to issue an early warning reminder if the predicted speed value does not meet the scene speed limit parameters.

[0006] In a possible implementation, key line parameters of a train test line are obtained, and an electronic map of the train test line is constructed based on the key line parameters, and the following processing is performed: according to preset safety requirements, speed limit section parameters of the track line are obtained; according to the starting point, end point and tunnel environment information of the tunnel, tunnel location parameters are obtained; according to the slope change of the track, slope parameters are obtained; according to the curvature of the track, curve radius parameters are obtained; according to the regional signal strength, signal blocking area parameters are obtained; the speed limit section parameters, the tunnel location parameters, the slope parameters, the curve radius parameters and the signal blocking area parameters are marked on the electronic map to obtain the electronic map of the train test line.

[0007] In a possible implementation, the sensor array is matched according to the scene characteristics of different operating sections in the electronic map of the train test line to obtain a real-time matching sensor, and the following processing is performed: the train test line is divided into multiple operating sections according to the annotation information of the electronic map of the train test line; multiple sensing accuracy requirements are obtained according to the operating scene characteristics of the multiple operating sections; the sensor array is matched according to the multiple sensing accuracy requirements, and after cost optimization, the real-time matching sensor is obtained.

[0008] In a possible implementation, the following processing is performed: the sensor array includes a Beidou positioning sensor, a laser ranging sensor, a Doppler radar speed sensor, an inertial navigation sensor, an accelerometer, and an axle speed sensor.

[0009] In a possible implementation, the ground dispatch center fuses and corrects the multi-source sensor data to obtain a corrected position and a corrected speed, and performs the following processing: the ground dispatch center unifies the data format, filters out noise, and aligns the data of the multi-source sensor data to obtain preprocessed multi-source data; performs dynamic weighted fusion on the position information and speed information in the preprocessed multi-source data to obtain a fused position and a fused speed; and corrects the fused position and the fused speed to obtain a corrected position and a corrected speed.

[0010] In a possible implementation, the position information and speed information in the pre-processed multi-source data are dynamically weighted and fused respectively to obtain a fused position and a fused speed, and the following processing is performed: a dynamic weighted Kalman filtering algorithm is used to dynamically adjust the weight of each sensor according to the current running scene information and real-time environmental information of the train, and the position information provided by each sensor in the pre-processed multi-source data is fused to obtain a fused position; a dynamic weighted averaging algorithm is used to dynamically adjust the weight of each sensor according to the current running scene information and real-time environmental information of the train, and the speed information provided by each sensor in the pre-processed multi-source data is fused to obtain a fused speed.

[0011] In a possible implementation, the weight of each sensor is dynamically adjusted according to the current operating scenario information and real-time environmental information of the train, and the following processing is performed: the position of the train is monitored in real time, the operating section in which the train is located is determined according to the current position of the train, and the operating scenario characteristics of the section are identified; the operating environment of the train is monitored in real time to obtain real-time environmental information; based on the operating scenario characteristics and the real-time environmental information, sensor reliability analysis is performed to obtain a reliability coefficient; based on the reliability coefficient, the weight of each sensor is adjusted.

[0012] In a possible implementation, the fused position and fused speed are respectively corrected to obtain a corrected position and a corrected speed, and the following processing is performed: the fused position is corrected using the auxiliary positioning information provided by the trackside equipment to obtain a corrected position; the fused speed is compared with the historical operating data, and the fused speed is corrected according to the statistical characteristics of the historical data to obtain a corrected speed.

[0013] The present application also provides a rail train speed monitoring and early warning system with multi-source data fusion, including: a train test line electronic map construction module, used to obtain the line key parameters of the train test line, and construct the train test line electronic map according to the line key parameters; a sensor array matching module, used to match the sensor array according to the scene characteristics of different operating sections in the train test line electronic map, and obtain real-time matching sensors; a multi-source sensor data acquisition module, used to obtain the real-time position and real-time speed of the train through the real-time matching sensors during the train operation, and obtain multi-source sensor data; a data fusion correction module, used for the ground dispatching center to fuse and correct the multi-source sensor data to obtain the corrected position and corrected speed; a prediction module, used to predict according to the corrected position and corrected speed, and obtain the predicted speed value and predicted position at the next moment; an early warning module, used to match and extract the scene speed limit parameters according to the train operation scene corresponding to the predicted position, and if the predicted speed value does not meet the scene speed limit parameters, trigger the vehicle-ground collaborative early warning mechanism to issue an early warning reminder.

[0014] The present application also provides an electronic device, comprising: a memory for storing executable instructions; and a processor for implementing a rail train speed monitoring and early warning method using multi-source data fusion when executing the executable instructions stored in the memory.

[0015] The method, system and equipment for monitoring and warning of rail train speed through multi-source data fusion proposed in this application first obtains the key parameters of the train test line, and constructs an electronic map of the train test line based on the key parameters. Then, according to the scene characteristics of different operating sections in the electronic map of the train test line, the sensor array is matched to obtain a real-time matching sensor. Then, during the train operation, the real-time position and real-time speed of the train are obtained through the real-time matching sensor to obtain multi-source sensor data. Then, the ground dispatching center fuses and corrects the multi-source sensor data to obtain a corrected position and corrected speed. Then, a prediction is made based on the corrected position and corrected speed to obtain the predicted speed value and predicted position at the next moment. Finally, according to the train operation scene corresponding to the predicted position, the scene speed limit parameters are matched and extracted. If the predicted speed value does not meet the scene speed limit parameters, the vehicle-ground collaborative warning mechanism is triggered to issue a warning reminder. The technical effect of improving the accuracy and timeliness of rail train speed monitoring and warning is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0017] Figure 1 A flow chart of a rail train speed monitoring and early warning method using multi-source data fusion provided in an embodiment of the present application.

[0018] Figure 2 A schematic diagram of the structure of a rail train speed monitoring and early warning system with multi-source data fusion provided in an embodiment of the present application.

[0019] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0020] Explanation of the accompanying symbols: train test line electronic map construction module 10, sensor array matching module 20, multi-source sensor data acquisition module 30, data fusion correction module 40, prediction module 50, early warning module 60, input device 301, processor 302, memory 303, output device 304. DETAILED DESCRIPTION

[0021] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0022] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0023] In the following description, reference is made to “some embodiments” which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict. The terms “including” and “having” and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0024] The present application embodiment provides a multi-source data fusion rail train speed monitoring and early warning method, such as Figure 1 As shown, the method includes:

[0025] Step S100: Acquire key line parameters of the train test line, and construct an electronic map of the train test line based on the key line parameters.

[0026] Specifically, high-precision Geographic Information System (GIS) equipment and surveying and mapping technologies are used to obtain key parameters of the train test line, such as its geographic coordinates, slope, curve radius, and track type. These key parameters are then processed using GIS software to construct an electronic map of the train test line. This digital map contains the line's geometry, topology, and related attribute information. The map is stored on a server or cloud platform for real-time access and updating.

[0027] For example, drones equipped with LiDAR equipment can be used to scan train test lines, acquiring highly accurate terrain and track information. This data can then be imported into GIS software to generate detailed electronic maps. Alternatively, key points along the train test lines can be located and measured using GPS and inertial measurement units (IMUs). This data can then be combined with GIS software to create electronic maps that include slope and curve information.

[0028] In one possible implementation, the key parameters of the train test line are obtained, and an electronic map of the train test line is constructed based on the key parameters of the line. Step S100 further includes step S110, which obtains the speed limit section parameters of the track line according to preset safety requirements. Specifically, by parsing the safety specifications and standards of the rail transit industry, the specific requirements of the speed limit section are extracted. These specifications exist in the form of documents, and key information can be extracted through natural language processing (NLP) technology. Using a geographic information system (GIS) device, combined with the geographic coordinates of the track line, the starting point and end point of the speed limit section are marked, and the corresponding speed limit value (such as 60km / h) is recorded. The speed limit section parameters are stored in a database and associated with the geographic information of the electronic map.

[0029] Step S120: Tunnel location parameters are acquired based on the tunnel's starting and ending points and the tunnel's internal environment. Specifically, a laser radar (LiDAR) or ultrasonic sensor is used to detect the tunnel's starting and ending points, and the tunnel's length and internal environment information, such as ventilation and lighting conditions, are recorded. The geographic coordinates of the tunnel's starting and ending points are recorded using a GIS device, and the tunnel's internal environment information is associated with the geographic coordinates. The tunnel location parameters are stored in a database for annotation on an electronic map.

[0030] Step S130: Obtaining slope parameters based on track gradient changes. Specifically, using a high-precision gradient sensor (e.g., a tilt sensor) or a total station, the gradient changes along the track are measured. The geographic coordinates of the gradient change points are recorded using a GIS device, and the gradient value (e.g., 3%) is associated with the geographic coordinates. The gradient parameters are stored in a database for annotation on electronic maps.

[0031] Step S140: Obtain the curve radius parameter based on the curvature of the track. Specifically, a high-precision total station or laser scanner is used to measure the curvature of the track and calculate the curve radius (e.g., 300 meters). The geographic coordinates of the curve radius change points are recorded using a GIS device, and the curve radius values are associated with the geographic coordinates. The curve radius parameters are stored in a database for annotation on electronic maps.

[0032] Step S150, obtain the signal blocking area parameters based on the regional signal strength. Specifically, use a signal strength tester, such as a Wi-Fi signal tester or a 4G / 5G signal tester, to measure the signal strength along the track. Use GIS equipment to record the geographic coordinates of the signal blocking area (such as an area with a signal strength below -80dBm), and associate the signal strength value with the geographic coordinates. Store the signal blocking area parameters in a database for marking on an electronic map.

[0033] Step S160: Mark the speed limit section parameters, tunnel location parameters, slope parameters, curve radius parameters, and signal obstruction area parameters on an electronic map to obtain an electronic map of the train test line. Specifically, GIS software (such as ArcGIS or QGIS) is used to mark the above parameters on the electronic map. Each parameter is distinguished by a different color or symbol, and the parameter data stored in the database is integrated with the geographic information of the electronic map to generate a complete electronic map of the train test line. The marked electronic map is stored on a server or cloud platform for real-time access and updating. For example, the speed limit section is marked with a yellow line to display the speed limit value; the tunnel location is marked with a blue area to display the tunnel name and length; the slope change is marked with a green arrow to display the slope value (uphill is an upward arrow, downhill is a downward arrow); the curve radius is marked with a red arc to display the curve radius value; the signal obstruction area is marked with a gray area to display the signal strength range. This marked information is stored on the server for use by the ground dispatch center. Table 1 is a specific example of an electronic map of a train test line, where the train test line is a circular line with a total length of 10 kilometers and contains multiple scene features.

[0034] Table 1: Example of electronic map of train test line

[0035]

[0036]

[0037]

[0038] Step S200 , matching the sensor array according to the scene features of different running sections in the electronic map of the train test line to obtain real-time matching sensors.

[0039] Specifically, the electronic map data of the train test line is read from the server database, obtaining the starting and ending mileage, geographic coordinates, and relevant parameter values (such as speed limit, tunnel length, and slope) for each section. Based on the train's real-time location, the scenario characteristics of the current operating section are directly searched from the electronic map. The train's current section is determined using geographic coordinates or mileage range, and detailed parameters for that section are obtained. For example, if the train is currently operating between 4.0 km and 5.5 km, the electronic map data can directly identify this section as "Tunnel A," and detailed tunnel parameters (such as 1.5 km length, good ventilation, and good lighting) can be obtained. Pre-set sensor matching rules are used to select the most appropriate sensor combination based on the scenario characteristics. These rules can be developed based on expert experience and actual test results and stored in the system. Based on the matching results, sensors are selected from the pre-set sensor array and configured for real-time operation. The sensor operating mode and parameters are dynamically adjusted through the sensor management system (SMS).

[0040] In a possible implementation, the sensor array includes a Beidou positioning sensor, a laser ranging sensor, a Doppler radar speed sensor, an inertial navigation sensor, an accelerometer, and an axle speed sensor.

[0041] Specifically, Beidou positioning sensors use the Beidou satellite navigation system to provide positioning information for real-time train position monitoring. Laser ranging sensors transmit and receive laser signals to measure the distance between the train and surrounding objects (such as tunnel walls), assisting in positioning and displacement calculation. In practical applications, a high-precision linear array laser sensor scans the longitudinal direction of the track, detecting sleeper edges based on changes in reflection intensity. This method utilizes a fixed sleeper spacing along the track as a reference, providing reliable positioning information even in areas with weak or blocked GPS signals (such as tunnels). The mileage calculation method using laser ranging sensors is as follows: The laser ranging sensor scans the track and detects changes in reflection intensity at the sleeper edges to determine the sleeper's position. Based on the standard sleeper spacing (typically 0.6 meters for Chinese railways) and combined with train displacement data (such as axle encoder pulses), the number of sleepers passed is calculated in real time. The cumulative mileage calculation formula is: Cumulative mileage = Number of sleepers counted by laser × Standard sleeper spacing + Distance within the current sleeper segment. Actual sleeper spacing may vary slightly (e.g., ±2 cm) due to construction errors or deformation. To improve the accuracy of mileage calculations, a sliding window statistical model for sleeper spacing was established, calculating the average spacing of the nearest few sleepers in real time. If the average spacing deviates from the standard spacing by more than a threshold (e.g., 1 cm), subsequent mileage calculations are dynamically corrected. The correction formula is: Corrected cumulative mileage = Number of sleepers counted by laser × Corrected average spacing + Displacement within the current sleeper segment.

[0042] Doppler radar speed sensors use the Doppler effect to measure train speed. Inertial navigation sensors provide dynamic attitude information by measuring the train's acceleration and angular velocity. Accelerometers measure changes in the train's acceleration and are used to monitor its dynamic behavior. Axle speed sensors calculate train speed by measuring the rotational speed of the train's axles. Table 2 shows an example of a matching rule for a sensor array.

[0043] Table 2: Example of sensor array matching rules

[0044]

[0045]

[0046] In one possible implementation, a sensor array is matched based on the scene characteristics of different operating sections in the electronic map of the train test line to obtain real-time matching sensors. Step S200 further includes step S210, which divides the train test line into multiple operating sections based on the annotation information of the electronic map of the train test line. Specifically, GIS software (such as ArcGIS or QGIS) is used to read the annotation information of the electronic map, including parameters such as speed limit sections, tunnel locations, slope changes, curve radius, and signal obstruction areas. Based on the annotation information and combined with preset operating scene characteristics, the train test line is divided into multiple operating sections. Each section corresponds to one or more specific operating scene characteristics. The divided operating section information is stored in a database for subsequent sensor matching and real-time monitoring. For example, ArcGIS software is used to read the annotation information of the electronic map and divide the train test line into multiple sections based on the speed limit section parameters. The starting and ending mileage of each section are recorded in the database. For example, based on the tunnel location parameters, the tunnel area is separately divided into an operating section, and the geographic coordinates of the tunnel's starting and ending points are recorded.

[0047] Step S220, obtain multiple sensor accuracy requirements based on the operating scenario characteristics of the multiple operating sections. Specifically, the scenario characteristics of each operating section are analyzed to determine the specific requirements of the section for sensor accuracy. For example, the tunnel section requires high-precision positioning and speed measurement, and the straight section has relatively low requirements for speed measurement accuracy. Based on the actual test results, a demand rule base is established to store the sensor accuracy requirements corresponding to different scenario characteristics. Through the rule matching algorithm, the corresponding sensor accuracy requirements are obtained from the demand rule base based on the scenario characteristics of each operating section. For example, for the tunnel section, the demand rule base stipulates that high-precision positioning (such as ±0.5 meters) and speed measurement (such as ±0.1km / h) are required. For the straight section, the positioning accuracy requirement is relaxed to ±1 meter, and the speed measurement accuracy is ±0.5km / h. For the curve section with a smaller curve radius, the demand rule base stipulates that high-precision dynamic monitoring is required to ensure the safe operation of the train in the curve.

[0048] In step S230, sensor array matching is performed based on the multiple sensor accuracy requirements, and after cost optimization, real-time matching sensors are obtained. Specifically, a database containing various sensor performance parameters (such as accuracy, range, sampling rate, etc.) and cost information is established. Based on the sensor accuracy requirements for each operating section and in combination with the sensor performance database, a matching algorithm is used to select the most suitable sensor combination. While meeting the accuracy requirements, a cost optimization algorithm, such as a linear programming or genetic algorithm, is used to select the sensor combination with the lowest cost. The matched sensors are configured for real-time operation and dynamically adjusted through the sensor management system (SMS). For example, for the tunnel section, based on the accuracy requirements, the matching algorithm selects a Beidou positioning sensor (high-precision positioning), a Doppler radar velocity sensor (high-precision speed measurement), and a laser ranging sensor (assisted positioning) from the sensor performance database. Then, a cost optimization algorithm is used to select the combination with the lowest cost. This implementation method dynamically matches the most suitable sensor combination based on the scenario characteristics of different operating sections of the train test line, while also performing cost optimization. Through these steps, high-precision position and velocity information can be obtained in each operating section, while controlling the overall cost of the system.

[0049] Step S300: During the operation of the train, the real-time position and real-time speed of the train are acquired through the real-time matching sensor to obtain multi-source sensor data.

[0050] Specifically, real-time matching sensors collect the train's real-time position and speed data through their respective technical means (such as radar speed measurement, GPS positioning, etc.). The data collected by the sensors is transmitted in real time to the ground dispatching center via wireless communication technologies (such as 4G / 5G, Wi-Fi). At the ground dispatching center, the collected multi-source sensor data undergoes preliminary processing, such as filtering and denoising, to improve the accuracy and reliability of the data. For example, a GPS receiver is used to obtain the train's real-time position information, while a radar sensor is used to measure the train's speed. This data is transmitted to the ground dispatching center in real time via the 5G network.

[0051] In step S400, the ground dispatching center performs fusion correction on the multi-source sensor data to obtain a corrected position and a corrected speed.

[0052] Specifically, the ground dispatch center uses multi-source data fusion algorithms, such as Kalman filtering and particle filtering, to fuse data from different sensors to eliminate differences and errors between the data and improve data accuracy and reliability. Based on the fusion results, the ground dispatch center uses correction data to correct the train's real-time position and speed. The corrected data is stored in a database and updated in real time to facilitate subsequent prediction and early warning processing. For example, the Kalman filter algorithm is used to fuse GPS positioning data and radar speed measurement data. GPS provides high-precision position information, and radar provides real-time speed information. The Kalman filter algorithm can eliminate the errors between the two types of data, resulting in more accurate train position and speed information.

[0053] In one possible implementation, the ground dispatch center performs fusion correction on the multi-source sensor data to obtain a corrected position and a corrected speed. Step S400 further includes step S410, in which the ground dispatch center performs data format unification, noise filtering, and data alignment on the multi-source sensor data to obtain preprocessed multi-source data. Specifically, a data conversion tool (such as an ETL tool, Extract-Transform-Load) is used to convert the data formats of different sensors into a unified format. For example, the timestamp format of all sensors is unified to the ISO8601 standard, the position data is unified to the latitude and longitude format, and the speed data is unified to km / h. For example, the timestamp format provided by the Beidou positioning sensor is "YYYY-MM-DD HH:MM:SS", while the timestamp format provided by the laser ranging sensor is a Unix timestamp. Use the ETL tool to convert all timestamps to the ISO8601 standard format.

[0054] Digital filters, such as low-pass and high-pass filters, are used to filter sensor data and remove noise. For example, a Doppler radar speed sensor can be affected by ambient noise. A Kalman filter can be used to smooth the speed data, remove noise, and improve data reliability.

[0055] Use time synchronization algorithms, such as timestamp alignment algorithms, to align data from different sensors to the same timeline. For example, interpolation methods can be used to align data with inconsistent timestamps to ensure data consistency.

[0056] In step S420, the position information and speed information in the preprocessed multi-source data are dynamically weighted and fused to obtain a fused position and a fused speed. Specifically, a dynamic weighted fusion algorithm, such as an adaptive weighted fusion algorithm, is used to fuse the preprocessed multi-source data. The weight of each sensor is dynamically adjusted based on its accuracy and reliability. For example, for position information, the Beidou positioning sensor has a higher weight, but in a tunnel, the weight of the laser ranging sensor will increase. For speed information, the Doppler radar speed sensor has a higher weight, but in a straight section, the weight of the axle speed sensor will also increase.

[0057] Fusion algorithms (such as Kalman filtering and particle filtering) are used to fuse position and velocity data. These algorithms dynamically adjust the fusion results based on the accuracy and reliability of the sensors. For example, using the Kalman filter algorithm to fuse the position data of a Beidou positioning sensor and a laser ranging sensor produces more accurate fused position information. Using the particle filter algorithm to fuse the velocity data of a Doppler radar velocity sensor and an axle speed sensor produces more accurate fused velocity information.

[0058] Step S430 corrects the fused position and fused velocity, respectively, to obtain a corrected position and corrected velocity. Specifically, correction algorithms, such as a deviation correction algorithm or auxiliary correction data, are used to correct the fused position and fused velocity. The correction algorithm can adjust the fusion results based on historical data and known error models. For example, if historical data reveals that the Beidou positioning sensor has systematic deviations in certain areas, the deviation correction algorithm can be used to correct the fused position. The fused velocity is also corrected based on known error models to ensure data accuracy.

[0059] In one possible implementation, the position information and velocity information in the preprocessed multi-source data are dynamically weighted and fused to obtain a fused position and fused velocity. Step S420 further includes step S421, employing a dynamic weighted Kalman filter algorithm to dynamically adjust the weights of each sensor based on the train's current operating scenario and real-time environmental information, and fuse the position information provided by each sensor in the preprocessed multi-source data to obtain a fused position. Specifically, the preprocessed multi-source position data is fused using the Kalman filter algorithm. Kalman filtering is a recursive algorithm that dynamically adjusts weights based on sensor accuracy and reliability, thereby obtaining a more accurate fused position. The weights of each sensor are dynamically adjusted based on the train's current operating scenario (e.g., tunnels, curves, slope changes, etc.) and real-time environmental information (e.g., signal strength, weather conditions, etc.). For example, in tunnels, the weight of laser ranging sensors is increased; on straight sections, the weight of Beidou positioning sensors is increased. The preprocessed multi-source position data is input into the Kalman filter, and the fused position is obtained through state estimation and update steps.

[0060] For example, in tunnels, due to the weak Beidou positioning signal, the laser ranging sensor's weight is dynamically adjusted to 0.7, while the Beidou positioning sensor's weight is adjusted to 0.3. Using the Kalman filter algorithm, the position data from the laser ranging sensor and the Beidou positioning sensor are fused to obtain a more accurate fused position.

[0061] For example, on straight sections of road, the Beidou positioning sensor's weight is dynamically adjusted to 0.8, while the laser ranging sensor's weight is adjusted to 0.2. Using the Kalman filter algorithm, the position data from the Beidou positioning sensor and the laser ranging sensor are fused to obtain a more accurate fused position.

[0062] Step S422, using a dynamic weighted average algorithm, dynamically adjusts the weight of each sensor according to the current train operation scene information and real-time environmental information, and fuses the speed information provided by each sensor in the pre-processed multi-source data to obtain a fused speed. Specifically, the pre-processed multi-source speed data is fused using a weighted average algorithm. According to the accuracy and reliability of the sensor, the weight of each sensor is dynamically adjusted to obtain a more accurate fused speed. According to the current train operation scene information (such as tunnels, curves, slope changes, etc.) and real-time environmental information (such as signal strength, weather conditions, etc.), the weight of each sensor is dynamically adjusted. For example, in a curved area, the weight of the axle speed sensor will increase; in a straight section, the weight of the Doppler radar speed sensor will increase. The pre-processed multi-source speed data is weighted averaged to obtain a fused speed.

[0063] For example, in a curve, the weight of the axle speed sensor is dynamically adjusted to 0.7, while the weight of the Doppler radar speed sensor is adjusted to 0.3. The speed data from the axle speed sensor and the Doppler radar speed sensor are fused using a weighted averaging algorithm to produce a more accurate fused speed.

[0064] For example, on straight sections of road, the weight of the Doppler radar speed sensor is dynamically adjusted to 0.8, while the weight of the axle speed sensor is adjusted to 0.2. Using a weighted averaging algorithm, the speed data from the Doppler radar and axle speed sensors are fused to produce a more accurate fused speed.

[0065] In one possible implementation, the weight of each sensor is dynamically adjusted according to the current operating scenario information and real-time environmental information of the train. Step S421 and step S422 further include: real-time monitoring of the position of the train, determining the operating section in which the train is located according to the current position of the train, and identifying the operating scenario characteristics of the section; real-time monitoring of the train operating environment to obtain real-time environmental information; performing sensor reliability analysis based on the operating scenario characteristics and the real-time environmental information to obtain a reliability coefficient; and adjusting the weight of each sensor based on the reliability coefficient.

[0066] Specifically, Beidou positioning sensors are used to obtain the train's precise location information in real time. Using Geographic Information System (GIS) technology, the train's current position is matched with the operating sections on an electronic map to determine the train's operating section. For example, the Beidou positioning sensors transmit location data once a second. After receiving this data, the ground dispatch center uses the GIS system to compare the train's current position with the operating sections on the electronic map to determine that the train is currently located in Tunnel A.

[0067] Based on the annotation information on the electronic map, the scene characteristics of the train's operating section are extracted, such as speed limit, tunnel environment, slope, curve radius, etc. For example, the train is currently in Tunnel A, and the electronic map annotation shows that Tunnel A has a speed limit of 60 km / h, a length of 1.5 km, and good ventilation and lighting.

[0068] Environmental monitoring sensors (such as meteorological sensors and signal strength sensors) are used to obtain real-time information about the train's operating environment, such as weather conditions and signal strength. For example, meteorological sensors monitor the temperature, humidity, and wind speed in tunnels in real time, while signal strength sensors monitor the communication signal strength between the train and the ground dispatching center.

[0069] According to the operating scenario characteristics and real-time environmental information, the reliability of each sensor is analyzed and the reliability coefficient of each sensor is calculated. The reliability coefficient can be calculated based on the sensor's historical data, environmental conditions and known error models. For example, the reliability coefficient calculation formula of the Beidou positioning sensor can be: 北斗 =α×signal strength factor+β×environmental factor, where α and β are weight coefficients, and α+β=1. The signal strength factor and environmental factor are calculated as follows: Assuming the signal strength error in the tunnel is 30%, the maximum signal strength error is 50%, the environmental error is 20%, and the maximum environmental error is 30%, then: Assuming α = 0.6, β = 0.4, then: Reliability coefficient 北斗 =0.6×0.4+0.4×0.33=0.372.

[0070] The weight of each sensor is dynamically adjusted based on its reliability coefficient. This weight adjustment can be achieved through weighted averaging or a Kalman filter algorithm. For example, in Tunnel A, the reliability coefficient of the Beidou positioning sensor is 0.372, and the reliability coefficient of the laser ranging sensor is 0.449. When fusing the position information, the weight of the Beidou positioning sensor is adjusted to 0.453, and the weight of the laser ranging sensor is adjusted to 0.547.

[0071] In one possible implementation, the fused position and fused speed are corrected respectively to obtain a corrected position and a corrected speed, and step S430 further includes step S431, using the auxiliary positioning information provided by the trackside equipment to correct the fused position to obtain a corrected position. Specifically, positioning equipment installed beside the track, such as transponders, wireless communication base stations, etc., are used to provide auxiliary positioning information. These devices can send positioning information to the train control system through wireless communication technologies (such as Wi-Fi, 4G / 5G). For example, the trackside transponder sends precise geographic location information when the train passes, and this information is obtained through the receiver on the train.

[0072] The auxiliary positioning information provided by the trackside equipment is compared with the fused position information, and the fused position is corrected using a difference correction algorithm. The difference correction algorithm can dynamically adjust the correction amount based on the positioning accuracy and reliability of the trackside equipment. For example, the positioning accuracy provided by the trackside transponder is ±0.1 meters, while the accuracy of the fused position is ±0.5 meters. Using the difference correction algorithm, the positioning information of the trackside equipment is compared with the fused position to correct the error in the fused position. The fused position is corrected using a correction formula. The correction formula can be designed based on the positioning accuracy and reliability of the trackside equipment. For example: Corrected position = fused position + γ(trackside equipment positioning information - fused position), where γ is an adjustment factor that represents the reliability of the trackside equipment information. If the trackside equipment is very reliable, γ can be close to 1; if the trackside equipment coverage is insufficient or the reliability is low, γ can be appropriately reduced. For example, when a train passes a trackside transponder, the transponder provides positioning information of 100.0 meters, while the fused position is 100.3 meters. Assuming that the reliability of the trackside transponder is high, γ = 0.8, the corrected fusion position is: corrected position = 100.3 + 0.8 × (100.0 - 100.3) = 100.06 meters.

[0073] Step S432 compares the fused speed with historical operating data and modifies the fused speed based on the statistical characteristics of the historical data to obtain a modified speed. Specifically, historical operating data for trains operating in the same or similar scenarios is obtained from a database. This data includes information such as speed, acceleration, and position. For example, historical speed data for trains operating in Tunnel A is obtained from the database, including speed values and timestamps.

[0074] The fused speed is compared with historical operating data, and a correction is calculated using statistical analysis methods (such as mean and standard deviation). The correction can be dynamically adjusted based on the statistical characteristics of the historical data. The fused speed is corrected using a correction formula designed based on the statistical characteristics of the historical data. For example, corrected speed = fused speed + δ × (historical average speed - fused speed), where δ is an adjustment factor representing the reliability of the historical data. If the historical data is highly reliable, δ can be close to 1; if the current operating environment is significantly volatile, δ can be appropriately reduced. For example, historical operating data shows that the average speed of trains in Tunnel A is 58 km / h, and the fused speed is 60 km / h. Assuming the reliability of the historical data is high, δ = 0.7, the corrected fused speed is: Corrected speed = 60 + 0.7 × (58 - 60) = 58.6 km / h.

[0075] Step S500 , performing prediction based on the corrected position and corrected speed to obtain a predicted speed value and predicted position at the next moment.

[0076] Specifically, a prediction model is constructed using machine learning or deep learning algorithms (such as long short-term memory (LSTM) networks) to predict the train's next position and speed based on the corrected train position and speed data. The prediction model is trained and optimized using historical data to improve prediction accuracy and reliability. During train operation, the prediction model is invoked in real time to predict the train's next position and speed based on the current corrected position and speed data.

[0077] For example, the corrected position and velocity data are normalized to the range of 0-1 to construct time series data, allowing the model to better learn patterns in the data. A long short-term memory (LSTM) network was selected as the prediction model, as it effectively handles long-term dependencies in time series data. Train operation data from the past year was used as the training set, with the position and velocity at each time point as input and the position and velocity at the next moment as output for supervised training of the LSTM model. Mean squared error (MSE) was used as the loss function, and the Adam optimizer was used for gradient descent. Cross-validation was used to select the optimal hyperparameters, such as a learning rate of 0.001 and a hidden layer size of 128. During the train operation, the LSTM model was called once per second, taking the corrected position and velocity at the current moment as input and outputting the predicted position and velocity at the next moment.

[0078] Step S600: According to the train operation scenario corresponding to the predicted position, the scenario speed limit parameters are matched and extracted. If the predicted speed value does not meet the scenario speed limit parameters, the vehicle-ground collaborative warning mechanism is triggered to issue a warning reminder.

[0079] Specifically, based on the predicted location, the corresponding scenario speed limit parameters are extracted from the electronic map. The speed limit parameters are stored in a database and associated with the scenario features of the electronic map. The predicted speed value is compared with the scenario speed limit parameters. If the predicted speed value exceeds the speed limit parameters, an early warning mechanism is triggered. The early warning information is sent to the train driver and ground dispatchers via the train-to-ground communication system (such as 4G / 5G, Wi-Fi), and measures such as automatic deceleration or stopping can be taken. For example, if the predicted location is in a tunnel, the tunnel speed limit parameters (such as 60km / h) are extracted from the electronic map. If the predicted speed value exceeds 60km / h, an early warning information is sent to the train driver and ground dispatchers via the train-to-ground communication system, and the automatic deceleration device is triggered. For example, if the predicted location is on a curve, the corresponding speed limit parameters (such as 40km / h) are extracted based on the curve's radius and curvature. If the predicted speed value exceeds 40km / h, an early warning information is sent to the train driver and ground dispatchers via the train-to-ground communication system, and the emergency braking procedure is initiated. The embodiment of the present application adopts the method of obtaining key line parameters to construct an electronic map of the train test line, matching the sensor array according to the scene features of different sections of the electronic map, determining the real-time matching sensor, and when the train is running, the real-time matching sensor obtains the real-time position and speed of the train to form multi-source sensor data. The ground dispatching center fuses and corrects the multi-source sensor data to obtain the corrected position and speed, and makes a prediction based on the corrected position and speed to obtain the predicted speed value and position at the next moment, and matches the speed limit parameters of the scene corresponding to the predicted position. If the predicted speed value does not meet the requirements, the vehicle-ground collaborative early warning mechanism is triggered to provide an early warning. These technical means achieve the technical effect of improving the accuracy and timeliness of rail train speed monitoring and early warning.

[0080] In the above, refer to Figure 1 The rail train speed monitoring and early warning method based on multi-source data fusion according to an embodiment of the present invention is described in detail. Figure 2 A rail train speed monitoring and early warning system using multi-source data fusion according to an embodiment of the present invention is described.

[0081] The multi-source data fusion rail train speed monitoring and early warning system according to an embodiment of the present invention is designed to address the technical issues of inaccurate monitoring and untimely warnings in existing rail train speed monitoring and early warning systems, thereby improving the accuracy and timeliness of rail train speed monitoring and early warning. The multi-source data fusion rail train speed monitoring and early warning system includes a train test line electronic map construction module 10, a sensor array matching module 20, a multi-source sensor data acquisition module 30, a data fusion correction module 40, a prediction module 50, and an early warning module 60.

[0082] The train test line electronic map construction module 10 is used to obtain the key line parameters of the train test line and construct the train test line electronic map based on the key line parameters; the sensor array matching module 20 is used to match the sensor array according to the scene characteristics of different operating sections in the train test line electronic map to obtain real-time matching sensors; the multi-source sensor data acquisition module 30 is used to obtain the real-time position and real-time speed of the train through the real-time matching sensors during the train operation to obtain multi-source sensor data; the data fusion and correction module 40 is used for the ground dispatching center to fuse and correct the multi-source sensor data to obtain the corrected position and corrected speed; the prediction module 50 is used to make a prediction based on the corrected position and corrected speed to obtain the predicted speed value and predicted position at the next moment; the early warning module 60 is used to match and extract the scene speed limit parameters according to the train operation scene corresponding to the predicted position. If the predicted speed value does not meet the scene speed limit parameters, the vehicle-ground collaborative early warning mechanism is triggered to issue an early warning reminder.

[0083] The specific configuration of the train test line electronic map construction module 10 will be described in detail below. As described above, the key parameters of the train test line are obtained, and the train test line electronic map is constructed based on the key parameters. The train test line electronic map construction module 10 can further include: a speed limit section parameter acquisition unit for acquiring the speed limit section parameters of the track line according to preset safety requirements; a tunnel location parameter acquisition unit for acquiring the tunnel location parameters based on the starting point, end point and tunnel environment information of the tunnel; a slope parameter acquisition unit for acquiring the slope parameter based on the slope change of the track; a curve radius parameter acquisition unit for acquiring the curve radius parameter based on the curvature of the track; a signal obstruction area parameter acquisition unit for acquiring the signal obstruction area parameter based on the regional signal strength; and a marking unit for marking the speed limit section parameter, the tunnel location parameter, the slope parameter, the curve radius parameter and the signal obstruction area parameter on the electronic map to obtain the train test line electronic map.

[0084] The specific configuration of the sensor array matching module 20 will be described in detail below. As described above, the sensor array is matched based on the scenario characteristics of different operating sections in the electronic map of the train test line to obtain real-time matching sensors. The sensor array matching module 20 may further include: a train test line division unit for dividing the train test line into multiple operating sections based on the annotation information of the electronic map of the train test line; a sensing accuracy requirement acquisition unit for acquiring multiple sensing accuracy requirements based on the operating scenario characteristics of the multiple operating sections; and a sensor array matching unit for performing sensor array matching based on the multiple sensing accuracy requirements, performing cost optimization, and obtaining real-time matching sensors.

[0085] The sensor array matching module 20 may further include: a sensor array construction unit for constructing a sensor array, wherein the sensor array includes a Beidou positioning sensor, a laser ranging sensor, a Doppler radar speed sensor, an inertial navigation sensor, an accelerometer and an axle speed sensor.

[0086] The specific configuration of the data fusion and correction module 40 will be described in detail below. As described above, the ground dispatch center performs fusion and correction on the multi-source sensor data to obtain a corrected position and corrected velocity. The data fusion and correction module 40 may further include: a data preprocessing unit configured to enable the ground dispatch center to unify the data format, filter noise, and align the multi-source sensor data to obtain preprocessed multi-source data; a dynamic weighted fusion unit configured to perform dynamic weighted fusion on the position information and velocity information in the preprocessed multi-source data to obtain a fused position and a fused velocity; and a correction unit configured to correct the fused position and fused velocity to obtain a corrected position and a corrected velocity.

[0087] Among them, the position information and speed information in the pre-processed multi-source data are dynamically weighted and fused respectively to obtain a fused position and a fused speed. The dynamic weighted fusion unit may further include: a position fusion subunit for adopting a dynamic weighted Kalman filtering algorithm, dynamically adjusting the weight of each sensor according to the current running scene information and real-time environmental information of the train, and fusing the position information provided by each sensor in the pre-processed multi-source data to obtain a fused position; a speed fusion subunit for adopting a dynamic weighted averaging algorithm, dynamically adjusting the weight of each sensor according to the current running scene information and real-time environmental information of the train, and fusing the speed information provided by each sensor in the pre-processed multi-source data to obtain a fused speed.

[0088] Among them, the weight of each sensor is dynamically adjusted according to the current operation scene information and real-time environmental information of the train. The position fusion subunit and the speed fusion subunit may further include: a section operation scene feature recognition component for monitoring the position of the train in real time, determining the operation section in which the train is located according to the current position of the train, and identifying the operation scene features of the section; a real-time environmental information acquisition component for monitoring the train operation environment in real time and obtaining real-time environmental information; a sensor reliability analysis component for performing sensor reliability analysis based on the operation scene features and the real-time environmental information to obtain a reliability coefficient; a weight adjustment component for adjusting the weight of each sensor according to the reliability coefficient.

[0089] Wherein, the fused position and fused speed are corrected respectively to obtain a corrected position and a corrected speed. The correction unit may further include: a position correction subunit for correcting the fused position using the auxiliary positioning information provided by the trackside equipment to obtain a corrected position; a speed correction subunit for comparing the fused speed with the historical operating data, and correcting the fused speed according to the statistical characteristics of the historical data to obtain a corrected speed.

[0090] The multi-source data fusion rail train speed monitoring and early warning system provided in an embodiment of the present invention can execute the multi-source data fusion rail train speed monitoring and early warning method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0091] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0092] Based on the foregoing embodiments, an embodiment of the present application further provides an electronic device. Figure 3 1 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing an embodiment of the present invention. Figure 3 The electronic device shown is merely an example and should not limit the functionality and scope of use of the embodiments of the present invention. The electronic device is in the form of a general-purpose computing device, and its components may include, but are not limited to, an input device 301, a processor 302, a memory 303, and an output device 304. The processor 302 may be one or more; the memory 303 may include a computer-readable medium and at least one program product, which has a set (at least one) of program modules configured to perform the functions of the various embodiments of the present application.

[0093] The memory 303 shown in the embodiment of the present invention may adopt any combination of one or more computer-readable media; the computer-readable storage medium may be, but is not limited to, an infrared, semiconductor system, device or component, or any combination of the above, for storing software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the multi-source data fusion rail train speed monitoring and early warning method in the embodiment of the present invention. The processor 302 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 303, thereby realizing the above-mentioned multi-source data fusion rail train speed monitoring and early warning method.

[0094] The above specific embodiments do not constitute a limitation to the scope of protection of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of this application should be included in the scope of protection of this application. In some cases, the actions or steps recorded in this application can be performed in an order different from that in the embodiments and can still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A multi-source data fusion rail train speed monitoring and early warning method, characterized in that: The method comprises: Acquiring key line parameters of the train test line, and constructing an electronic map of the train test line based on the key line parameters; Matching the sensor array according to the scene features of different running sections in the electronic map of the train test line to obtain real-time matching sensors; During the operation of the train, the real-time position and real-time speed of the train are obtained through the real-time matching sensor to obtain multi-source sensor data; The ground dispatching center fuses and corrects the multi-source sensor data to obtain a corrected position and a corrected speed; Predicting the corrected position and corrected speed to obtain a predicted speed value and predicted position at the next moment; According to the train operation scenario corresponding to the predicted position, the scenario speed limit parameters are matched and extracted. If the predicted speed value does not meet the scenario speed limit parameters, the vehicle-ground collaborative early warning mechanism is triggered to issue an early warning reminder.

2. The multi-source data fusion rail train speed monitoring and early warning method according to claim 1 is characterized in that: Acquiring key line parameters of the train test line and constructing an electronic map of the train test line based on the key line parameters, including: Obtain the speed limit section parameters of the track line according to preset safety requirements; Obtain tunnel location parameters based on the tunnel's starting point, end point, and tunnel environment information; According to the slope change of the track, the slope parameter is obtained; According to the curvature of the track, obtain the curve radius parameter; According to the regional signal strength, obtain the signal blocking area parameters; The speed limit section parameters, the tunnel location parameters, the slope parameters, the curve radius parameters and the signal blocking area parameters are marked on an electronic map to obtain an electronic map of the train test line.

3. The multi-source data fusion rail train speed monitoring and early warning method according to claim 2 is characterized in that: According to the scene characteristics of different running sections in the electronic map of the train test line, the sensor array is matched to obtain a real-time matching sensor, including: Dividing the train test line into a plurality of operating sections according to the annotation information of the train test line electronic map; Acquiring multiple sensing accuracy requirements based on the operating scenario characteristics of the multiple operating sections; Sensor array matching is performed according to the multiple sensing accuracy requirements, and after cost optimization, a real-time matching sensor is obtained.

4. The multi-source data fusion rail train speed monitoring and early warning method according to claim 1 is characterized in that: The sensor array includes a Beidou positioning sensor, a laser ranging sensor, a Doppler radar speed sensor, an inertial navigation sensor, an accelerometer and an axle speed sensor.

5. The multi-source data fusion rail train speed monitoring and early warning method according to claim 1 is characterized in that: The ground dispatch center fuses and corrects the multi-source sensor data to obtain the corrected position and corrected speed, including: The ground dispatching center performs data format unification, noise filtering, and data alignment on the multi-source sensor data to obtain pre-processed multi-source data; Performing dynamic weighted fusion on the position information and the speed information in the pre-processed multi-source data to obtain a fused position and a fused speed; The fusion position and fusion speed are corrected respectively to obtain a corrected position and a corrected speed.

6. The multi-source data fusion rail train speed monitoring and early warning method according to claim 5 is characterized in that: Performing dynamic weighted fusion on the position information and speed information in the pre-processed multi-source data to obtain a fused position and a fused speed, including: A dynamic weighted Kalman filter algorithm is used to dynamically adjust the weight of each sensor according to the current train operation scene information and real-time environmental information, and the position information provided by each sensor in the pre-processed multi-source data is fused to obtain a fused position; A dynamic weighted average algorithm is used to dynamically adjust the weight of each sensor according to the current train operation scene information and real-time environmental information, and the speed information provided by each sensor in the pre-processed multi-source data is fused to obtain the fused speed.

7. The multi-source data fusion rail train speed monitoring and early warning method according to claim 6, characterized in that: Dynamically adjust the weight of each sensor based on the train's current operating scenario and real-time environmental information, including: Monitor the train's location in real time, determine the operating section based on the train's current location, and identify the operating scenario characteristics of the section; Real-time monitoring of the train operating environment to obtain real-time environmental information; Performing sensor reliability analysis based on the operating scenario characteristics and the real-time environmental information to obtain a reliability coefficient; The weight of each sensor is adjusted according to the reliability coefficient.

8. The multi-source data fusion rail train speed monitoring and early warning method according to claim 5, characterized in that: Correcting the fusion position and the fusion speed respectively to obtain a corrected position and a corrected speed, including: Using the auxiliary positioning information provided by the trackside equipment, the fusion position is corrected to obtain the corrected position; The fusion speed is compared with the historical operation data, and the fusion speed is corrected according to the statistical characteristics of the historical data to obtain the corrected speed.

9. The multi-source data fusion rail train speed monitoring and early warning system is characterized by: The system is used to implement the rail train speed monitoring and early warning method based on multi-source data fusion according to any one of claims 1 to 8, and the system includes: A train test line electronic map construction module is used to obtain key line parameters of the train test line and construct an electronic map of the train test line based on the key line parameters; A sensor array matching module is used to match the sensor array according to the scene characteristics of different running sections in the electronic map of the train test line to obtain real-time matching sensors; A multi-source sensor data acquisition module is used to obtain the real-time position and real-time speed of the train through the real-time matching sensor during the train operation to obtain multi-source sensor data; A data fusion correction module is used by the ground dispatch center to perform fusion correction on the multi-source sensor data to obtain a corrected position and corrected speed; A prediction module, configured to make a prediction based on the corrected position and corrected speed to obtain a predicted speed value and a predicted position at the next moment; The early warning module is used to match and extract the scene speed limit parameters according to the train operation scene corresponding to the predicted position. If the predicted speed value does not meet the scene speed limit parameters, the vehicle-ground collaborative early warning mechanism is triggered to issue an early warning reminder.

10. An electronic device, characterized in that: The electronic device comprises: a memory for storing executable instructions; The processor is configured to implement the rail train speed monitoring and early warning method based on multi-source data fusion as described in any one of claims 1 to 8 when executing the executable instructions stored in the memory.

Images