Railway freight train adaptive asynchronous control method based on artificial intelligence

Abstract

The present application relates to the field of artificial intelligence, and more particularly to a railway freight train adaptive asynchronous control method based on artificial intelligence. The method comprises the following steps: collecting train operation data through train sensors, and performing train operation and track linkage processing and train operation real-time positioning detection to generate train operation-track linkage data and train operation real-time positioning data; obtaining train operation planning data; performing train operation scheduling design based on the train operation planning data and the train operation real-time positioning data to generate train operation scheduling data; performing train each car operation state and train traction force linkage analysis and train adaptive asynchronous control instruction setting based on the train operation real-time positioning data, the train operation-track linkage data and the train operation scheduling data, and performing train adaptive asynchronous control operation. The present application realizes train adaptive asynchronous control.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to an adaptive asynchronous control method for railway freight trains based on artificial intelligence. Background Technology

[0002] With the development of intelligent railways, higher demands are placed on traction force distribution, carriage status linkage, train positioning accuracy, and scheduling adaptability during railway freight train operation. In existing technologies, railway freight train control methods mostly rely on control logic driven by fixed rules or centralized scheduling methods based on preset paths. When freight trains face complex operating conditions such as changes in track curve radius, gradient undulations, and communication blind spots in tunnels, they struggle to adaptively and asynchronously adjust their operating status and control strategies. This leads to problems such as uneven carriage status distribution, delayed traction response, and accumulation of positioning errors during train operation. Furthermore, this results in problems such as imbalance in traction force coordination among different carriages, excessive stress fluctuations in carriages, and loss of control over train operation rhythm. Summary of the Invention

[0003] Based on this, the present invention provides an adaptive asynchronous control method for railway freight trains based on artificial intelligence to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, an artificial intelligence-based adaptive asynchronous control method for railway freight trains includes the following steps:

[0005] Step S1: Collect train operation data through train sensors, perform train operation and track linkage processing based on the train operation data, and generate train operation-track linkage data;

[0006] Step S2: Based on train operation data and train operation-track linkage data, perform real-time train positioning detection and generate real-time train positioning data;

[0007] Step S3: Obtain train operation planning data; Design train operation scheduling based on train operation planning data and real-time train operation positioning data, and generate train operation scheduling data;

[0008] Step S4: Based on the real-time positioning data of train operation and the train operation-track linkage data, perform a linkage analysis of the operating status of each carriage of the train and the train traction force, and generate linkage data of the operating status of each carriage of the train and the train traction force.

[0009] Step S5: Based on the operating status of each carriage of the train, the train traction linkage data, and the train operation scheduling data, set the train adaptive asynchronous control command, generate the train adaptive asynchronous control command data, and execute the train adaptive asynchronous control operation based on the train adaptive asynchronous control command data.

[0010] Furthermore, step S1 includes the following steps:

[0011] Step S11: Collect train operation data through train sensors, analyze train operation characteristics based on train operation data, and generate train operation characteristic data;

[0012] Step S12: Perform real-time track detection based on train operation characteristic data to generate real-time track data for the train;

[0013] Step S13: Analyze the train track structure change information based on the real-time train track data to generate train track structure change information data;

[0014] Step S14: Based on train operation characteristic data and train track structure change information data, perform train operation and track linkage processing to generate train operation-track linkage data.

[0015] Furthermore, step S2 includes the following steps:

[0016] Step S21: Based on the train operation data, perform train-tower base station communication and train speed detection respectively, and generate train-tower base station communication data and train speed data respectively;

[0017] Step S22: Monitor changes in the train's operating environment based on the train operation-track linkage data, and generate data on changes in the train's operating environment;

[0018] Step S23: Based on train-tower base station communication data, train speed data, and train driving environment change data, perform train driving and tunnel driving positioning analysis to generate train driving and tunnel driving positioning data;

[0019] Step S24: Based on the train-tower base station communication data, train speed data, and train-tunnel travel positioning data, perform real-time train positioning detection and generate real-time train positioning data.

[0020] Furthermore, step S23 includes the following steps:

[0021] Step S231: Based on the train's operating environment change data, perform tunnel travel sequence and time difference analysis for each train carriage to generate tunnel travel sequence-time difference data for each train carriage.

[0022] Step S232: Based on the tunnel travel sequence-time difference data of each train carriage and the communication data between the train and the tower base station, perform communication switching and recovery detection of each train carriage, and generate communication switching-recovery data of each train carriage;

[0023] Step S233: Calculate the train positioning delay based on the communication switching-recovery data of each carriage and the train speed data, and generate train positioning delay data;

[0024] Step S234: Perform train travel and tunnel travel positioning analysis based on train travel positioning delay data and train speed data to generate train travel-tunnel travel positioning data.

[0025] Furthermore, step S3 includes the following steps:

[0026] Step S31: Obtain train operation planning data, and divide the train running track occupancy section according to the train operation planning data to generate train running track occupancy section data;

[0027] Step S32: Analyze the intersecting track sections of each train based on the data of the track occupancy of the train and the real-time positioning data of the train, and generate data of each intersecting track section;

[0028] Step S33: Analyze the train freight load configuration based on the train operation data and generate train freight load configuration data;

[0029] Step S34: Based on the train freight load configuration data, the data of each train intersecting track section, and the data of the track occupancy of the train, perform load adaptability matching of each train in the intersecting track section to generate load adaptability data of each train in the intersecting track section.

[0030] Step S35: Based on the data of the occupied sections of the train's running track, the load adaptability data of each train in the intersecting track section, and the data of each train in the intersecting track section, design the train operation scheduling and generate train operation scheduling data.

[0031] Furthermore, step S4 includes the following steps:

[0032] Step S41: Based on the real-time positioning data of train operation and the train operation-track linkage data, the train track curve and the train track slope are detected respectively, and the train track curve data and the train track slope data are generated respectively.

[0033] Step S42: Perform dynamic analysis of train curve track travel response based on train travel curve data to generate train curve track travel response data;

[0034] Step S43: Perform dynamic analysis of train track gradient response based on train track gradient data to generate train track gradient response dynamic data;

[0035] Step S44: Based on the train's curved track travel response dynamic data and the train track gradient response dynamic data, perform a linkage analysis of the operating status of each train car and the train's traction force, and generate linkage data of the operating status of each train car and the train's traction force.

[0036] Furthermore, step S42 includes the following steps:

[0037] Step S421: Based on the train travel track curve data, perform train travel contact axial center of gravity detection on the train travel-track linkage data to generate train travel contact axial center of gravity data;

[0038] Step S422: Analyze the train track deflection displacement based on the train travel contact axis center of gravity data to generate train track deflection displacement data;

[0039] Step S423: Analyze the contact area and resistance of the train curve track based on the train track deflection displacement data, and generate train curve track contact area-resistance data;

[0040] Step S424: Perform dynamic analysis of train curve track travel response based on the train curve track contact area-resistance data to generate train curve track travel response dynamic data.

[0041] Furthermore, step S43 includes the following steps:

[0042] Step S431: Based on the track gradient data, detect the traction spacing of each car in the train operation-track linkage data and generate the traction spacing data of each car.

[0043] Step S432: Analyze the push-pull stress of each car based on the traction spacing data of each car, and generate push-pull stress data for each car;

[0044] Step S433: Based on the push-pull stress data of each carriage, perform axial offset detection on the train operation-track linkage data to generate axial offset data for each train;

[0045] Step S434: Based on the axial offset data of each train and the push-pull stress data of each carriage, perform dynamic analysis of the train track gradient response to generate dynamic data of the train track gradient response.

[0046] Furthermore, step S44 includes the following steps:

[0047] Step S441: Based on the train track gradient response dynamic data, analyze the longitudinal force coordination and speed difference of each train on the gradient section, and generate longitudinal force coordination-speed difference data of each train on the gradient section.

[0048] Step S442: Analyze the longitudinal traction deflection force of each train on the curved track based on the train's dynamic response data on the curved track, and generate longitudinal traction deflection force data for each train on the curved track.

[0049] Step S443: Based on the longitudinal traction deflection force data of each train on the curved track and the longitudinal force coordination-speed difference data of each train on the slope section, perform a linkage analysis of the operating status of each carriage and the train traction force, and generate linkage data of the operating status of each carriage and the train traction force.

[0050] Furthermore, step S5 includes the following steps:

[0051] Step S51: Calculate the freight load configuration of each carriage of the train based on the train operation scheduling data, and generate the freight load configuration data of each carriage of the train.

[0052] Step S52: Based on the operating status of each carriage of the train - the train traction force linkage data and the freight load configuration data of each carriage of the train, perform traction power and braking analysis of each carriage of the train to generate traction power and braking data of each carriage of the train.

[0053] Step S53: Set the train adaptive asynchronous control command based on the traction power-braking data of each carriage, generate the train adaptive asynchronous control command data, and execute the train adaptive asynchronous control operation based on the train adaptive asynchronous control command data.

[0054] The beneficial effects of this invention are:

[0055] This invention proposes an AI-based adaptive asynchronous control method for railway freight trains. It collects train operation data through train sensors and performs train operation and track linkage processing. The sensors can capture multi-dimensional data in real time, including train speed, acceleration, carriage vibration, track smoothness, curve radius, and gradient, ensuring data coverage of the entire train operation scenario. The linkage processing deeply correlates the train's operating state with the physical characteristics of the track. Real-time positioning detection based on train operation data and linkage data offers advantages in improved positioning accuracy and timely dynamic response. Combining train operation data with linkage data enables cross-validation of multi-source data, reducing the impact of factors such as obstruction and signal delay. Real-time dynamic updates of positioning ensure immediate capture of sudden changes in train position. Train scheduling design is performed by acquiring train operation planning data and combining it with real-time positioning data, optimizing the dynamic adaptability of train scheduling and the efficiency of resource allocation. Scheduling design combined with real-time positioning data can accurately match the deviation between the actual train position and the planned path, avoiding track resource conflicts and improving the overall transportation efficiency of the railway network. Based on real-time train positioning data and train operation and linkage data, the system performs linkage analysis on the carriage operation status and traction force. This enables refined management of train operation status and optimized configuration of traction force output. The linkage analysis will identify which carriages require higher traction support, while lightly loaded carriages or those on level tracks can appropriately reduce traction force output, reducing uneven tension between carriages caused by uneven traction force distribution. Based on train operation and linkage data as well as scheduling data, the system sets and executes adaptive asynchronous control commands for the train. It can formulate differentiated control strategies for each carriage according to its different operation status, traction force requirements, and scheduling requirements. Asynchronous control enables the train to operate more smoothly in complex track environments.

[0056] This invention discloses an artificial intelligence-based adaptive asynchronous control method for railway freight trains. By real-time positioning of the train's running location and tunnel communication blind spots, and simultaneously detecting and analyzing changes in the radius of the track curve and gradient undulations of each train carriage, it can monitor in real time issues such as uneven distribution of carriage status, traction response delay, and accumulation of positioning errors during train operation. This allows for the formulation of adaptive asynchronous adjustment control commands, resolving problems such as unbalanced traction force coordination among train carriages, excessive stress fluctuations in carriages, and loss of control over train rhythm. Furthermore, it coordinates the track usage of multiple trains in intersecting track sections, avoiding scheduling conflicts in these sections, thereby improving freight train transportation efficiency and track resource utilization. Attached Figure Description

[0057] Figure 1 This is a flowchart illustrating the steps of an artificial intelligence-based adaptive asynchronous control method for railway freight trains according to the present invention.

[0058] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S1.

[0059] Figure 3 for Figure 1 A detailed flowchart illustrating the implementation steps of step S2.

[0060] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0061] The technical method of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0062] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0063] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0064] To achieve the above objectives, please refer to Figures 1 to 3 This invention provides an adaptive asynchronous control method for railway freight trains based on artificial intelligence, comprising the following steps:

[0065] Step S1: Collect train operation data through train sensors, perform train operation and track linkage processing based on the train operation data, and generate train operation-track linkage data;

[0066] Step S2: Based on train operation data and train operation-track linkage data, perform real-time train positioning detection and generate real-time train positioning data;

[0067] Step S3: Obtain train operation planning data; Design train operation scheduling based on train operation planning data and real-time train operation positioning data, and generate train operation scheduling data;

[0068] Step S4: Based on the real-time positioning data of train operation and the train operation-track linkage data, perform a linkage analysis of the operating status of each carriage of the train and the train traction force, and generate linkage data of the operating status of each carriage of the train and the train traction force.

[0069] Step S5: Based on the operating status of each carriage of the train, the train traction linkage data, and the train operation scheduling data, set the train adaptive asynchronous control command, generate the train adaptive asynchronous control command data, and execute the train adaptive asynchronous control operation based on the train adaptive asynchronous control command data.

[0070] In the embodiments of this invention, please refer to Figure 1 The diagram shown is a flowchart illustrating the steps of an artificial intelligence-based adaptive asynchronous control method for railway freight trains according to the present invention. In this example, the artificial intelligence-based adaptive asynchronous control method for railway freight trains includes the following steps:

[0071] Step S1: Collect train operation data through train sensors, perform train operation and track linkage processing based on the train operation data, and generate train operation-track linkage data;

[0072] In this embodiment of the invention, millimeter-wave radar, infrared temperature sensors, and acceleration sensors installed at the train's front, carriage bottom, and wheelsets, along with fiber optic strain gauges and displacement sensors deployed along the track, are used to collect data. The millimeter-wave radar captures the distance and relative speed between the train and obstacles in front every second; the infrared temperature sensors monitor the friction temperature of the wheelsets in real time; the acceleration sensors record the vertical and lateral vibration acceleration of the carriages; the fiber optic strain gauges transmit the strain values ​​of the track fasteners every second; and the displacement sensors measure the track settlement. The collected speed, temperature, vibration, strain, and settlement data are input into a deep learning-based convolutional neural network model. This model, trained with multiple sets of historical data, can identify different track conditions (e.g., a curve radius of 300 meters and a gradient of 2‰) and train operating parameters (e.g., a speed of 80 km / h and an acceleration of 0.5 m / s²). 2 The system establishes the correlation between track smoothness level (levels 1-5), wheel-rail friction coefficient (0.2-0.6), and train dynamic response coefficient, and outputs linked data.

[0073] Step S2: Based on train operation data and train operation-track linkage data, perform real-time train positioning detection and generate real-time train positioning data;

[0074] In this embodiment of the invention, track curve parameters and gradient values ​​from the linked data are combined with train GPS positioning data and wheel diameter pulse counter data for fusion calculation. A Kalman filter algorithm is used to process the multi-source data. This algorithm sets the process noise covariance matrix Q to 0.01I and the measurement noise covariance matrix R to 0.1I, updating the positioning result every 0.2 seconds. Specifically, when the train travels to a curved section of the track, the curve radius value (e.g., 500 meters) from the linked data is used to correct GPS drift caused by signal obstruction. Simultaneously, the pulse count is compensated based on wheel diameter wear (preset standard wheel diameter 1.2 meters, actual measurement 1.18 meters), calculating the train's precise position in the track mileage coordinate system (error ≤ 0.5 meters). Real-time positioning data is generated, including latitude and longitude, track section number (e.g., K123+450), and distance to the next turnout (e.g., 300 meters). For example, even when GPS signals are lost in a tunnel, positioning accuracy can still be maintained through wheel diameter counting and track gradient data.

[0075] Step S3: Obtain train operation planning data; Design train operation scheduling based on train operation planning data and real-time train operation positioning data, and generate train operation scheduling data;

[0076] In this embodiment of the invention, train operation planning data is obtained from the railway dispatch center database, including the planned arrival time of each station (e.g., arriving at station A at 10:00), the speed limit for each section (e.g., 100 km / h for a certain section), and the passing point (e.g., at K150). Real-time train location data (e.g., current location K110+200, actual time 9:40) and the planning data are input into a scheduling model based on a genetic algorithm. This model aims to minimize the total delay time, setting the population size to 50, the crossover probability to 0.8, and the mutation probability to 0.05, and performing multiple iterations. For example, when real-time location shows that the train arrives at section K110 5 minutes later than planned, the model automatically adjusts the operating parameters for subsequent sections, increasing the target speed to 110 km / h in the K120-K140 section with a speed limit of 120 km / h, while simultaneously reducing the stop time at station B by 3 minutes. It also generates scheduling data containing the estimated arrival time for each section, speed curves, and passing adjustment schemes, ensuring that the final arrival time deviation at the destination station is controlled within a certain range.

[0077] Step S4: Based on the real-time positioning data of train operation and the train operation-track linkage data, perform a linkage analysis of the operating status of each carriage of the train and the train traction force, and generate linkage data of the operating status of each carriage of the train and the train traction force.

[0078] In this embodiment of the invention, the current track conditions of the train (e.g., uphill slope of 1.5‰, straight section) are determined based on real-time positioning data. Combined with the track resistance coefficient (e.g., 0.005) from train operation and linkage data, data from pressure sensors (measuring load), speed sensors (measuring car speed), and current sensors (measuring traction motor current) installed in each car are analyzed. A multiple linear regression model is used, with car load (50-80 tons), track gradient (-2‰ to 3‰), and operating speed (60-120 km / h) as independent variables, and traction force (100-500 kN) as the dependent variable. The regression equation (R0) is obtained by fitting 1000 sets of sample data. 2 =0.95). For example, when the positioning data shows that the train is on a 2‰ uphill section and the pressure sensor of carriage 3 shows a load of 75 tons, the model calculates that carriage needs 320kN of traction force, while carriage 5, with a load of 50 tons, only needs 210kN, and generates linkage data that includes the traction force requirement value of each carriage and the load-traction force matching degree (0-100%).

[0079] Step S5: Based on the operating status of each carriage of the train, the train traction linkage data, and the train operation scheduling data, set the train adaptive asynchronous control command, generate the train adaptive asynchronous control command data, and execute the train adaptive asynchronous control operation based on the train adaptive asynchronous control command data.

[0080] In this embodiment of the invention, train operation and linkage data (e.g., 320kN required for carriage 3, matching degree 92%) and scheduling data (e.g., acceleration to 110km / h required for section K130-K150) are input into a fuzzy control model. This model includes three input variables (traction demand deviation, scheduling time deviation, and carriage state coefficient) and two output variables (traction motor voltage adjustment and current adjustment). It is calculated using a membership function (triangular distribution) and 49 control rules. For example, when the actual traction output of carriage 3 is 300kN (deviation -20kN), and the scheduling requires this section to be completed 2 minutes ahead of schedule, the model outputs a control command to increase the voltage by 5% and the current by 8%. Simultaneously, for carriage 5, which has a lighter load, the model outputs a command to decrease the voltage by 3% and the current by 2%, forming differentiated control. Control commands are sent to the traction converters of each carriage via the train's Ethernet or the train's built-in local area network. The converters adjust the IGBT switching frequency (1-10kHz) in real time according to the commands to achieve asynchronous adjustment of traction force.

[0081] Furthermore, step S1 includes the following steps:

[0082] Step S11: Collect train operation data through train sensors, analyze train operation characteristics based on train operation data, and generate train operation characteristic data;

[0083] Step S12: Perform real-time track detection based on train operation characteristic data to generate real-time track data for the train;

[0084] Step S13: Analyze the train track structure change information based on the real-time train track data to generate train track structure change information data;

[0085] Step S14: Based on train operation characteristic data and train track structure change information data, perform train operation and track linkage processing to generate train operation-track linkage data.

[0086] As an embodiment of the present invention, reference is made to... Figure 2 As shown, Figure 1 A detailed flowchart of step S1 is shown below. In this embodiment, step S1 includes the following steps:

[0087] Step S11: Collect train operation data through train sensors, analyze train operation characteristics based on train operation data, and generate train operation characteristic data;

[0088] In this embodiment of the invention, a laser Doppler velocimeter is installed on the top of the train's front, a three-dimensional acceleration sensor is installed on the bogie of the carriage, and a temperature sensor is installed on the wheelset axle box. These sensors collect real-time data on the train's speed, longitudinal / lateral / vertical acceleration, and wheelset temperature. The collected 60-minute continuous data is divided into time windows of 10 seconds each and input into a random forest-based classification model. This model is trained on multiple sets of operating data with different loads (3000-5000 tons) and different track conditions, and can identify the train at constant speed (80 km / h) and acceleration (0.3 m / s²). 2 Deceleration (-0.4m / s) 2 The system outputs dynamic response characteristics under the following conditions, including the train's acceleration curve (0-60km / h acceleration time 120 seconds), braking deceleration curve (800-0km / h braking distance 800 meters), wheelset temperature change rate (temperature rise of 5℃ per hour), and vibration frequency distribution (mainly concentrated in 5-15Hz). For example, when the lateral acceleration of the train exceeds 0.6g and lasts for 10 seconds during acceleration, the model automatically marks it as an abnormal operating characteristic.

[0089] Step S12: Perform real-time track detection based on train operation characteristic data to generate real-time track data for the train;

[0090] In this embodiment of the invention, vibration frequency and wheelset temperature change rate from train operation characteristic data are retrieved and processed in conjunction with track image data collected by a track flaw detection radar and a high-definition line array camera installed at the bottom of the train. The vibration frequency data is decomposed using a wavelet transform algorithm to extract vibration energy values ​​in the 4-8Hz frequency band, which corresponds to track bed loosening characteristics. Simultaneously, the Canny edge detection algorithm from machine vision is used to process the track images and identify the geometric shape of the track fasteners. The vibration energy value and fastener morphology parameters are input into the support vector machine model. The model is set with a radial basis kernel function (parameter σ = 0.8). Through training with multiple sets of track state samples, it can output real-time running track data including track bed compaction (70%-95%), fastener integrity rate (85%-100%), rail wear (0-3mm), and track gauge deviation (-2mm to +2mm). For example, when the vibration energy value is detected to exceed 80dB and the fastener integrity rate is less than 90%, the model determines it to be a loose track bed area and marks the specific location (K210+350 to K210+450).

[0091] Step S13: Analyze the train track structure change information based on the real-time train track data to generate train track structure change information data;

[0092] In this embodiment of the invention, the real-time track data of the train is differentially calculated according to the daily, weekly, and monthly time dimensions to extract the changes in track parameters. A sliding window algorithm (window size of 7 days, step size of 1 day) is used to process the rail wear data to calculate the daily wear increment (0.01-0.05 mm / day); the gauge deviation data is converted to the frequency domain through Fourier transform to identify the periodic variation components (period of 30 days ± 5 days); and the monthly change rate of track bed compaction (-2% to 1% / month) is analyzed using trend line fitting (using the least squares method). These changes are input into an LSTM-based time-series prediction model, which contains three hidden layers (64 neurons per layer). After training with one year of track change data, the model can output track structure change information data, including rail wear trend (expected wear of 0.8 mm in the next 30 days), track gauge deviation periodicity (deviation increases from the 5th to the 10th of each month), track bed settlement rate (0.5 mm / month), and fastener loosening spread rate (expanding by 10 meters per day). For example, when the track bed settlement rate in a certain section is calculated to exceed 0.8 mm / month for three consecutive months, the model automatically generates track structure warning information and marks the starting mileage (K180+200) and the affected range (200 meters) of the change area.

[0093] Step S14: Based on train operation characteristic data and train track structure change information data, perform train operation and track linkage processing to generate train operation-track linkage data.

[0094] In this embodiment of the invention, the braking distance and acceleration performance parameters in the train operation characteristic data are correlated with the rail wear and track bed settlement rate in the track structure change information data. The correlation between braking distance and rail wear (r = 0.75) and the correlation between acceleration performance and track bed settlement rate (r = -0.68) are determined by calculating the Pearson correlation coefficient (range -1 to 1). The relevant parameters are input into a graph neural network model, which contains 128-dimensional node features. The edge weights are set based on the parameter influence (40% for rail wear and 30% for track bed settlement). After training with 50,000 sets of linked samples, the model can output operation-track linkage data, including the correlation matrix between the train running resistance coefficient (0.012-0.018) and track structure changes, the curve of wheel-rail contact stress distribution (200-300MPa) with wear, and the correspondence between train dynamic response lag time (0.5-2 seconds) and track bed settlement. For example, when the rail wear reaches 2.5mm and the running resistance coefficient rises to 0.017, the model outputs a linkage suggestion that the train should reduce its maximum speed by 5km / h and marks the corresponding track section (K230+100 to K235+000).

[0095] Furthermore, step S2 includes the following steps:

[0096] Step S21: Based on the train operation data, perform train-tower base station communication and train speed detection respectively, and generate train-tower base station communication data and train speed data respectively;

[0097] Step S22: Monitor changes in the train's operating environment based on the train operation-track linkage data, and generate data on changes in the train's operating environment;

[0098] Step S23: Based on train-tower base station communication data, train speed data, and train driving environment change data, perform train driving and tunnel driving positioning analysis to generate train driving and tunnel driving positioning data;

[0099] Step S24: Based on the train-tower base station communication data, train speed data, and train-tunnel travel positioning data, perform real-time train positioning detection and generate real-time train positioning data.

[0100] As an embodiment of the present invention, reference is made to... Figure 3 As shown, Figure 1 A detailed flowchart of step S2 is shown below. In this embodiment, step S2 includes the following steps:

[0101] Step S21: Based on the train operation data, perform train-tower base station communication and train speed detection respectively, and generate train-tower base station communication data and train speed data respectively;

[0102] In this embodiment of the invention, a 5G communication module installed on the train roof establishes a communication connection with a control tower base station deployed every 5 kilometers along the line. Every 2 seconds, it sends information such as the train identification code, current latitude and longitude, and number of carriages. Simultaneously, it receives data from the control tower base station regarding track occupancy status 3 kilometers ahead and signal display status (red / green / yellow light). This generates train-control tower base station communication data including communication signal strength (-50 to -100 dBm), data transmission success rate (99.9%), and information exchange frequency. Simultaneously, a Hall effect speed sensor is installed at the train's wheelset axle end, outputting a pulse signal every 0.1 seconds. This pulse signal is converted by a counter to obtain the train's instantaneous speed (0-120 km / h). Combined with wheel diameter (1.2 m ± 0.005 m) compensation calculations, train speed data including speed change curves and acceleration values ​​is generated. For example, when the communication signal strength is below -90 dBm for 5 seconds, the train speed is simultaneously recorded as 85 km / h.

[0103] Step S22: Monitor changes in the train's operating environment based on the train operation-track linkage data, and generate data on changes in the train's operating environment;

[0104] In this embodiment of the invention, the track smoothness level and wheel-rail friction coefficient from the train operation-track linkage data are retrieved and processed in conjunction with data collected by meteorological sensors and infrared thermal imagers installed on both sides of the train. Using a threshold judgment method, when the temperature is below 0°C and the humidity is above 85%, it is marked as an environment prone to icing; at this time, the baseline value of the track adhesion coefficient (0.3) needs to be multiplied by a correction factor of 0.8; using real-time wind sensor data combined with the track alignment (30°E-60°E), the impact force is calculated using the formula F = 0.5 × ρ × v. 2 ×S×sinθ (where ρ is the air density, 1.2 kg / m³) 3 (where v is wind speed, S is the train side area, and θ is the angle between wind direction and track). Environmental parameters and linkage data are input into a Bayesian network model (containing 5 layers of nodes, with prior probabilities set based on multiple sets of historical data). The output includes environmental risk levels (levels 1-5), track adhesion coefficient correction values ​​(0.8-1.2 times the baseline value), and train running resistance increments (50-300 N / t). For example, when a wind speed of 15 m / s is detected and the track smoothness level is 4, the model determines the environmental risk level to be 3 and marks the affected area (K320+100 to K320+800).

[0105] Step S23: Based on train-tower base station communication data, train speed data, and train driving environment change data, perform train driving and tunnel driving positioning analysis to generate train driving and tunnel driving positioning data;

[0106] In this embodiment of the invention, train positioning analysis for both train operation and tunnel operation is performed based on signal status and track occupancy information from train-tower base station communication data, instantaneous speed and acceleration from train speed data, and environmental risk level from train operating environment change data. In non-tunnel sections, the base station triangulation method is used to calculate the train position by combining the signal reception times of three adjacent tower base stations with the base station coordinates (latitude and longitude known). When a train enters a tunnel (500-3000 meters long), the RFID tags installed every 100 meters on the tunnel wall (store tunnel mileage information) and the RFID reader at the front of the train are activated to read the tag information and obtain the relative position inside the tunnel. At the same time, the mileage accumulation in the train speed data (error ≤ 0.5 meters per 100 meters) is combined with the data through a weighted average method (base station positioning weight 0.7, RFID positioning weight 0.3) to generate train-tunnel travel positioning data that includes the location switching point inside and outside the tunnel (accurate to 1 meter), the distance from the entrance / exit inside the tunnel (0-3000 meters), and the positioning update frequency (1 time / 0.5 seconds). For example, when the train enters the 1000-meter-long K350 tunnel, at a distance of 300 meters from the entrance, the RFID positioning shows 298 meters, the speed accumulation shows 302 meters, and the merged positioning is 300 meters.

[0107] Step S24: Based on the train-tower base station communication data, train speed data, and train-tunnel travel positioning data, perform real-time train positioning detection and generate real-time train positioning data.

[0108] In this embodiment of the invention, train-to-tower base station communication data (including the status of the track ahead), train speed data (including instantaneous speed and acceleration), and train-tunnel positioning data (including precise position) are input into an extended Kalman filter algorithm for fusion processing. The algorithm's state variables include position (x, y), velocity v, and acceleration a; the observation variables are base station positioning, RFID positioning, and speed sensor data. The process noise variance is set to 0.001, and the measurement noise variance is dynamically adjusted according to different scenarios (0.01 inside the tunnel, 0.005 outside the tunnel). The positioning results are updated every 0.2 seconds. Outliers are removed through residual analysis (residual ≤ 0.5 meters), and calibration is performed using track mileage markers (one every 100 meters) from the real-time train track data. This generates real-time train positioning data containing absolute position (error ≤ 0.3 meters), relative track mileage (accurate to 0.1 meters), and positioning confidence (95%-99.9%). For example, when the train reaches K400+500, the base station positioning shows K400+500.2, the RFID positioning (if in a tunnel) shows K400+499.8, and the speed accumulation shows K400+500.1. After algorithm fusion and calibration, the final positioning is K400+500.0, with a confidence level of 99.5%.

[0109] Furthermore, step S23 includes the following steps:

[0110] Step S231: Based on the train's operating environment change data, perform tunnel travel sequence and time difference analysis for each train carriage to generate tunnel travel sequence-time difference data for each train carriage.

[0111] In this embodiment of the invention, the environmental risk level and tunnel section information from the train's operating environment change data are retrieved and analyzed in conjunction with the car spacing data collected by laser ranging sensors installed at the heads of each car. The laser ranging sensors output the distance values ​​of adjacent cars every 0.5 seconds, and the total length from the front of the train to the nth car is calculated by accumulating these values ​​(each car is 14 meters long, and the total length of 15 cars is 210 meters). When the train enters the tunnel entrance (based on the tunnel entrance kilometer marker), the entry time of the front of the train is recorded. Combined with the instantaneous speed in the train speed data (e.g., 80 km / h, converted to 22.22 m / s), the theoretical time for each car to enter the tunnel is calculated (the second car enters 0.675 seconds later than the front, the third car enters 0.675 seconds later than the second, and so on). Simultaneously, the time difference is corrected according to the environmental risk level (e.g., level 3). When the wind force is ≥10 m / s, the time difference between each car increases by 0.05 seconds. Generate sequence-time difference data that includes the entry time of each carriage into the tunnel (accurate to 0.01 seconds), the exit time from the tunnel, the entry time difference between carriages (0.675-0.725 seconds), and the total length adapted to the tunnel length. For example, when 15 carriages enter a 1000-meter tunnel, the entry time of the locomotive is 10:00:00.00, and the entry time of the 15th carriage is 10:00:10.12, which deviates from the theoretically calculated 10:00:10.13 by 0.01 seconds.

[0112] Step S232: Based on the tunnel travel sequence-time difference data of each train carriage and the communication data between the train and the tower base station, perform communication switching and recovery detection of each train carriage, and generate communication switching-recovery data of each train carriage;

[0113] In this embodiment of the invention, the communication status of each carriage is monitored based on the carriage travel sequence-time difference data in the tunnel, combined with the signal strength and transmission success rate in the train-tower base station communication data. An independent 5G communication sub-module (the same model as the main module at the front of the train) is installed on the top of each carriage to record the communication signal strength in real time. When the front of the train enters the tunnel, the time when the second carriage will enter the tunnel is predicted based on the sequence-time difference data, and the switching procedure of the communication sub-module of that carriage is initiated 0.5 seconds in advance (switching from tower base station communication to emergency communication in the tunnel). By comparing the actual switching time with the predicted time (deviation ≤ 0.1 seconds), and combining the transmission success rate within 30 seconds after the switching (≥ 99.5%), the success of the switching is determined. When the carriage exits the tunnel, the communication recovery procedure is initiated with a time difference lag, based on the time of the front of the train's exit, and the time it takes for the signal strength to recover to -80dBm after recovery (≤ 2 seconds) is recorded. Generate communication handover-recovery data that includes the start time of communication handover for each carriage, the handover success determination result (success / failure), the recovery completion time, and the data packet loss rate (≤0.1%) during the handover. For example, the handover for carriage 5 was started at the predicted time of 10:00:03.20 and was actually completed at 10:00:03.28, with a deviation of 0.08 seconds. The transmission success rate after the handover was 99.8%, and the handover was judged to be successful.

[0114] Step S233: Calculate the train positioning delay based on the communication switching-recovery data of each carriage and the train speed data, and generate train positioning delay data;

[0115] In this embodiment of the invention, the positioning delay is calculated by combining the switching delay time (the time difference between the actual switching and the predicted switching) and the recovery completion time in the communication switching-recovery data with the acceleration and instantaneous speed in the train speed data. When communication switching fails (e.g., transmission success rate < 99.5%), the failure duration (≤ 5 seconds) is recorded, and the train travel distance during this period (≤ 111.1 meters) is calculated based on the speed data (e.g., 22.22 m / s). This distance is the positioning lag caused by the communication interruption. For carriages that successfully switch, the switching delay time (0.08 seconds) is multiplied by the speed to obtain the positioning delay distance (0.08 × 22.22 ≈ 1.78 meters). Simultaneously, when the carriage exits the tunnel and communication is restored, the train travel distance (44.44 meters) within the recovery completion time (2 seconds) is calculated as the positioning delay during the recovery phase. The arithmetic mean method is used to calculate the positioning delay of the entire train (the average delay distance of each carriage), generating positioning delay data that includes the delay distance of each carriage during the switching phase (0-1.78 meters), the delay distance of the recovery phase (0-44.44 meters), the total delay distance (0-46.22 meters), and the duration of the delay. For example, the switching delay of carriage 5 is 1.78 meters, the recovery delay is 44.44 meters, and the total delay is 46.22 meters. Carriage 8 has no switching failure and a total delay of 1.80 meters. The average delay of the entire train is 24.01 meters.

[0116] Step S234: Perform train travel and tunnel travel positioning analysis based on train travel positioning delay data and train speed data to generate train travel-tunnel travel positioning data.

[0117] In this embodiment of the invention, the positioning delay data is retrieved and combined with the accumulated mileage value in the train speed data to correct the positioning of the train and the train traveling in tunnels. The positioning delay distance is superimposed on the original positioning data (such as base station triangulation or RFID positioning results) in the speed direction (the forward direction is positive). When the delay distance is 46.22 meters, the original positioning K350+300 is corrected to K350+346.22. Simultaneously, based on the delay duration (e.g., 5 seconds) and the rate of change of speed (acceleration 0.2 m / s²), the positioning is also corrected. 2The delay correction per second is calculated using linear interpolation (9.0 meters in the first second, 9.4 meters in the second, and so on until the correction is complete after 5 seconds). Inside the tunnel, the corrected positioning data is calibrated every 100 meters, based on the absolute position read from the RFID tag (e.g., K350+500), to ensure that the cumulative error is ≤1 meter. In non-tunnel sections, the base station triangulation results (error ≤3 meters) are fused with the corrected data, and a moving average method (window size 5 data points) is used to smooth out positioning fluctuations. Generate driving-tunnel driving positioning data that includes corrected tunnel positioning (accurate to 0.1 meters), non-tunnel positioning, delay correction curve (correction per second), and calibration point error (≤0.5 meters). For example, the original positioning of the train at K350+300 in the tunnel is displayed as K350+300 due to delay. After correction, it becomes K350+346.22. When the train reaches the RFID tag at K350+500, the calibration shows the actual position as K350+500.3, with an error of 0.3 meters.

[0118] Furthermore, step S3 includes the following steps:

[0119] Step S31: Obtain train operation planning data, and divide the train running track occupancy section according to the train operation planning data to generate train running track occupancy section data;

[0120] In this embodiment of the invention, train operation planning data is retrieved from the railway dispatch center database. This data includes the planned routes for each train (e.g., the K100-K500 section), planned departure times (e.g., 7:00, 8:30), planned arrival times, and stops (one every 50 kilometers). This data is then combined with electronic mileage markers installed along the track (one every 1 kilometer, storing precise mileage information) to divide the track occupancy sections. A section segmentation algorithm is used, with two adjacent stations as the basic unit (e.g., K100 to K150 as one unit), and then further divided into sub-sections every 10 kilometers. Each sub-section is marked with a planned occupancy time period (e.g., train A occupies K100-K110 from 8:00-8:10). Simultaneously, occupancy rules are set according to track type (single-track / double-track). Only one train is allowed to occupy a single-track section at the same time, while double-track sections have separate occupancy for each direction. Generate track occupancy data that includes the section number (e.g., Q100-1 represents K100-K110), the planned train number, the start time of occupation (accurate to the minute), the end time of occupation, the section length (10 kilometers), and the track type. For example, train B plans to occupy the double-track section Q150-2 (K150-K170) from 9:00 to 9:20, which does not conflict with train C occupying the opposite section Q150-3 from 9:10 to 9:30.

[0121] Step S32: Analyze the intersecting track sections of each train based on the data of the track occupancy of the train and the real-time positioning data of the train, and generate data of each intersecting track section;

[0122] In this embodiment of the invention, train intersection situations are analyzed based on the data of the occupied sections of the train's running track, combined with real-time train positioning data (including the current mileage and speed of each train). A train position-time matrix is ​​established, with time (in minutes) on the horizontal axis and track mileage on the vertical axis. The occupied section data of each train is transformed into rectangular areas in the matrix (e.g., train A at 8:00-8:10 corresponds to K100-K110). Real-time positioning data is used to mark the coordinates of each train's current position in the matrix (e.g., train D at 8:05 is located at K120). A collision detection algorithm is used to calculate the overlap of the rectangular areas of different trains. Areas with an overlap greater than 0 are potential intersection sections. Further analysis of potential sections reveals that when two trains have speeds of 80km / h and 90km / h respectively, and the distance between them is less than 5 kilometers, these are identified as intersection sections requiring close monitoring. Generate intersection track section data that includes the intersection section mileage (e.g., K130-K135), the estimated intersection time (e.g., 8:20), the train numbers involved (2 trains), the planned speed at the intersection (80km / h and 70km / h), and whether the track is a passing section (yes / no). For example, train E and train F are expected to intersect at 8:20 in the single-track section K130-K135, which has a passing platform (500 meters long).

[0123] Step S33: Analyze the train freight load configuration based on the train operation data and generate train freight load configuration data;

[0124] In this embodiment of the invention, freight load configuration analysis is performed based on real-time load data collected by carriage load sensors (installed at the bottom of each carriage, measuring range 0-100 tons, accuracy ±0.5 tons) from the train operation data. The load sensors in each carriage upload data every 10 minutes (e.g., 50 tons for carriage 1, 60 tons for carriage 2), and the total train load is calculated by summing the data (total load of 15 carriages: 800 tons ± 5 tons). Combined with cargo type information (e.g., coal, steel), the load distribution uniformity of each carriage (standard deviation ≤ 5 tons) is calculated. Uniformity < 3 tons is excellent, 3-5 tons is good, and > 5 tons is poor. Simultaneously, based on track gradient data, the impact of load on climbing ability is analyzed. When the gradient > 3‰, the maximum load limit for each carriage is 60 tons. Generate freight load configuration data that includes car number, single car load (tons), total load (tons), load uniformity level, cargo type suitability (e.g., steel suitability 90%), and load compliance (compliant / non-compliant) under gradient restrictions. For example, if the 8th car has a load of 65 tons, it is determined to be non-compliant on a 3‰ gradient section and needs to be marked for adjustment.

[0125] Step S34: Based on the train freight load configuration data, the data of each train intersecting track section, and the data of the track occupancy of the train, perform load adaptability matching of each train in the intersecting track section to generate load adaptability data of each train in the intersecting track section.

[0126] In this embodiment of the invention, freight load configuration data (including total load and uniformity of each train), intersecting track section data (including gradient and length of the intersecting section), and train track occupancy data (including speed limits) are used to perform load adaptability matching. Using a load-gradient matching algorithm, when the gradient of the intersecting section is 2‰, trains with a total load ≤1000 tons can maintain a speed of 80 km / h, while those with 1000-1200 tons need to reduce their speed to 70 km / h. Considering the load uniformity of each train, trains with better uniformity can increase their speed by 5 km / h when intersecting on curves (curve radius 300 meters), while those with poor uniformity need to decrease their speed by 5 km / h. Simultaneously, based on the time overlap of the track occupancy sections, a safe distance (≥500 meters) between the two trains in the intersecting section is calculated; the heavily loaded train needs to decelerate 10 seconds in advance. Generate load adaptability data for each train in the junction section, including the recommended speed (60-80km / h), deceleration start distance (e.g., K129.5), acceleration end distance (e.g., K135.5), load and section adaptability level (excellent / good / poor), and safety distance margin (500-800 meters). For example, train G has a total load of 900 tons (excellent uniformity), a recommended speed of 75km / h in the junction section with a 2‰ gradient, and a safety distance margin of 600 meters.

[0127] Step S35: Based on the data of the occupied sections of the train's running track, the load adaptability data of each train in the intersecting track section, and the data of each train in the intersecting track section, design the train operation scheduling and generate train operation scheduling data.

[0128] In this embodiment of the invention, train scheduling is designed based on data on the occupied sections of the train's running track, the load adaptability data of each train in the intersecting track sections, and the data of each train's intersecting track sections. A genetic algorithm is used to optimize the scheduling, with the fitness function being the minimization of the total delay time (target value ≤ 5 minutes). The population size is set to 100, the crossover probability is 0.7, the mutation probability is 0.05, and the iteration is 50 times. The suggested speed and safety distance in the intersecting sections are transformed into constraints. When two trains intersect on a single track, the train with the heavier load (e.g., 1200 tons) is given priority to pass, and the other train stops at the passing station (stopping time 5 minutes). Combining real-time train positioning data, the speed of trains deviating from the plan (e.g., a 2-minute delay) in subsequent sections is adjusted (increasing by 5 km / h within the speed limit). Generate train scheduling data that includes adjusted departure times (accurate to the minute), travel speeds in each section (60-90 km / h), passing and avoidance schemes (who passes first), estimated arrival times (deviation ≤ 1 minute), and scheduling adjustment criteria (such as load priority). For example, if train H is delayed by 2 minutes, its speed will be increased from 80 km / h to 85 km / h in the K200-K250 section to ensure it arrives at K250 station as scheduled at 10:00, and the passing time with subsequent trains will be adjusted to 10:10.

[0129] Furthermore, step S4 includes the following steps:

[0130] Step S41: Based on the real-time positioning data of train operation and the train operation-track linkage data, the train track curve and the train track slope are detected respectively, and the train track curve data and the train track slope data are generated respectively.

[0131] In this embodiment of the invention, track curve and gradient detection are performed based on real-time train positioning data (including precise mileage and position coordinates) and train operation-track linkage data (including track structure parameters). A laser profile scanner is installed at the bottom of the train's front end to acquire track cross-sectional profile data every second. Combined with mileage information from the real-time positioning data (such as K500+200), the track curve radius is calculated using a curve fitting algorithm (using the least squares method). When the profile within a continuous 50-meter range conforms to the characteristics of a circular arc, it is determined to be a curve segment, and the running track curve data, including curve radius (300-1000 meters), curve length (50-500 meters), and curve start / end mileage, is output. Meanwhile, the tilt sensor (measurement range -15° to 15°, accuracy ±0.01°) installed on the train bogie is used to collect the track tilt angle. Combined with the altitude information in the real-time positioning data (error ±0.5 meters), the slope value is calculated by the slope calculation formula. A slope data (-5‰ to 5‰) is output every 10 meters, generating running track slope data that includes the slope start / end mileage, slope value, slope length (100-1000 meters), and slope type (uphill / downhill). For example, in the section from K510+300 to K511+300, a curve radius of 500 meters is detected, and there is also a 2‰ uphill slope. Curve data and slope data are generated accordingly.

[0132] Step S42: Perform dynamic analysis of train curve track travel response based on train travel curve data to generate train curve track travel response data;

[0133] In this embodiment of the invention, dynamic analysis of the travel response on curved tracks is performed based on the train's track curve data (including curve radius and length). Using a curve radius of 300 meters and a curve length of 200 meters from the curve data, combined with the train's speed (80 km / h) and wheelbase (15 meters) from the train's travel data, a multibody dynamics simulation model (containing 15 rigid car bodies and a spring-damping system) is employed for analysis. In the model, the wheel-rail contact force is calculated using Hertzian contact theory. A superelevation value of 150 mm (corresponding to a 300-meter radius) is set for the track curve segment. The centrifugal force (train's total mass 800 tons) is calculated by inputting the train speed, yielding the lateral impact force (50-150 kN). Simultaneously, actual lateral vibration data is collected using lateral acceleration sensors (range ±2g, accuracy ±0.01g) installed on both sides of the car, and compared with the simulation results to correct the dynamic response parameters. Generate dynamic data for curve track travel response, including peak lateral acceleration (0.5-1.2g), wheel-rail lateral force (80-200kN), centripetal force required for curve passage (100-300kN), carriage tilt angle (1-3°), and curve travel dynamic safety factor (≥1.2). For example, when a train passes through a curve with a radius of 300 meters at 80km / h, the peak lateral acceleration is 0.8g, the wheel-rail lateral force is 150kN, and the safety factor is 1.5, which meets the safety standards.

[0134] Step S43: Perform dynamic analysis of train track gradient response based on train track gradient data to generate train track gradient response dynamic data;

[0135] In this embodiment of the invention, track gradient response dynamic analysis is performed based on the train's track gradient data (including gradient value and length). For a 2‰ uphill gradient with a length of 1000 meters, combined with a total train load of 800 tons and a current speed of 80 km / h, dynamic analysis software is used to calculate the traction force required for driving on the slope (the calculation formula is F = mgsinθ + fmgcosθ, where θ is the gradient angle, f is the rolling friction coefficient of 0.002, m is the total train load, and g is the gravitational acceleration), yielding the theoretical traction force (200-500 kN). The actual output torque is collected by a torque sensor (range 0-10000 N·m, accuracy ±10 N·m) installed on the traction motor and converted into an actual traction force (220 kN), which is then compared to the theoretical value with a deviation of 10%. Simultaneously, the train's speed change on the slope is analyzed; when the actual traction force is greater than the theoretical value, the speed increase rate is 0.1 m / s. 2 Conversely, the rate of descent is 0.05 m / s. 2 The system generates parameters including the theoretical traction force required for the ramp, the actual traction force, the traction force deviation rate (5%-15%), and the speed change rate (±0.1m / s). 2The track gradient response dynamic data includes the gradient driving power reserve coefficient (actual traction force / theoretical traction force, ≥1.1). For example, on a 2‰ uphill section, the theoretical traction force required is 200kN, the actual output is 220kN, the power reserve coefficient is 1.1, and the speed is 0.1m / s. 2 The elevation is improved to meet the requirements for driving on a slope.

[0136] Step S44: Based on the train's curved track travel response dynamic data and the train track gradient response dynamic data, perform a linkage analysis of the operating status of each train car and the train's traction force, and generate linkage data of the operating status of each train car and the train's traction force.

[0137] In this embodiment of the invention, a linkage analysis of the carriage's operating state and traction force is performed based on the train's curved track running response dynamic data (including lateral force and centripetal force) and the train track gradient response dynamic data (including traction force and rate of change of speed). The analysis utilizes the lateral force of 150 kN and centripetal force of 200 kN from the curved track dynamic data, and the actual traction force of 220 kN and rate of change of speed of 0.1 m / s from the gradient dynamic data. 2 Based on the load sensor data of each carriage (50 tons for carriage 1, 60 tons for carriage 2, ... 55 tons for carriage 15), a traction force distribution algorithm is used for calculation. The algorithm allocates the basic traction force according to the carriage load ratio (16% for carriages with a load of 60 tons and 13% for carriages with a load of 50 tons), and then adds a lateral correction coefficient based on the curve power data (5% increase for carriages on the inside of the curve and 5% decrease for carriages on the outside), and adds a slope correction coefficient based on the slope data (10% increase for all carriages when going uphill). The actual current is collected by the current sensor (range 0-500A, accuracy ±1A) installed on the traction motor of each carriage, converted into the actual traction force of each carriage (15-25kN), and compared with the calculated value. If the deviation exceeds 5%, a secondary adjustment is made. Generate car body operation status-traction linkage data that includes the theoretical traction force, actual traction force, traction force distribution ratio (5%-18%), traction force and lateral force matching degree (≥80%), and speed synchronization rate of each car body (≥95%). For example, the second car body with a load capacity of 60 tons has a theoretical traction force of 24kN on the curve and uphill section, and an actual output of 23.5kN, with a deviation of 2% and a matching degree of 92%, which meets the linkage requirements.

[0138] Furthermore, step S42 includes the following steps:

[0139] Step S421: Based on the train travel track curve data, perform train travel contact axial center of gravity detection on the train travel-track linkage data to generate train travel contact axial center of gravity data;

[0140] In this embodiment of the invention, based on the train's track curve data (including a curve radius of 300 meters and a curve length of 200 meters) and combined with the wheel-rail friction coefficient of 0.35 in the train operation-track linkage data, the center of gravity of the train's contact axis is detected. Three displacement sensors (measuring range -50 to 50 mm, accuracy ±0.1 mm) are installed on the inner side of each pair of wheelsets to collect the lateral distance between the wheelset and the inner side of the rail (130 mm on the left and 128 mm on the right), recording data every 0.05 seconds. Using a center of gravity calculation model, with the wheelset center as the origin, the lateral distance data is converted into a wheelset center of gravity offset (left offset +2 mm, right offset -2 mm), and the center of gravity offset trend under centrifugal force is calculated based on the curve radius (offset increases by 5 mm on the outer side of the curve). Simultaneously, the wheel-rail contact point distribution in the linkage data is used to determine the reference axis of the contact axis (coinciding with the track centerline), and the deviation value (±3 mm) between the actual center of gravity and the reference axis is calculated. Generate contact axial center of gravity data that includes the three-dimensional coordinates of the wheelset center of gravity (X-axis along the track direction, Y-axis laterally, Z-axis vertically, accurate to 0.1mm), the center of gravity offset of each wheelset pair (-5 to +5mm), the axial center of gravity offset rate (0.2mm / s), and the maximum center of gravity offset of the curve segment (+6mm). For example, at the midpoint of a 300-meter radius curve, the center of gravity coordinates of the 5th wheelset pair are (K510+400,+3mm,0mm), which is +4mm offset from the straight segment.

[0141] Step S422: Analyze the train track deflection displacement based on the train travel contact axis center of gravity data to generate train track deflection displacement data;

[0142] In this embodiment of the invention, track deflection displacement analysis is performed based on the train's contact axial center of gravity data (including center of gravity offset and offset rate). A finite element analysis model is used, dividing the track structure into 1-meter-long units. The lateral force of each wheelset in the center of gravity offset data is input (calculated based on the offset; an offset of +5mm corresponds to a lateral force of 80kN) to simulate the elastic deformation of the rail under lateral force. Laser displacement sensors (measurement range 0-100mm, accuracy ±0.05mm) are installed on both sides of the track to collect the lateral displacement of the top of the rail in real time (outer rail displacement +3mm, inner rail -2mm), outputting data every 0.1 seconds. By comparing the theoretical displacement values ​​(outer +3.2mm, inner -1.8mm) from the finite element model with the actual measured values, the displacement deviation rate (6.25%) is calculated, and the track displacement change trend (increasing by 0.5mm every 10 meters) is extrapolated based on the center of gravity offset rate. Generate track deflection displacement data that includes the lateral displacement value of the rail (-3 to +4 mm), displacement deviation rate (5%-8%), track elastic deformation coefficient (0.02 mm / kN), and displacement distribution gradient of the curve segment (the displacement of the outer side increases by 2 mm from the start point to the midpoint of the curve). For example, at K510+350 of a 300-meter radius curve, the outer rail displacement is +3 mm, which deviates from the theoretical value by 0.2 mm and is within the allowable range.

[0143] Step S423: Analyze the contact area and resistance of the train curve track based on the train track deflection displacement data, and generate train curve track contact area-resistance data;

[0144] In this embodiment of the invention, the contact area and resistance of the curved track are analyzed based on the train's track deflection displacement data (including rail lateral displacement and deformation coefficient). Pressure-sensitive film (accuracy 0.1 N / mm) is adhered to the wheelset surface. 2 Using a high-speed camera (1000fps), the pressure marks on the film were photographed to identify the major axis (80mm±5mm) and minor axis (30mm±2mm) of the wheel-rail contact area. The contact area (S0=a×b, where a is the major axis contact length and b is the minor axis contact length, 2400mm) was calculated. 2 ±200mm 2 Considering the rail deformation in the track deflection displacement, when the lateral displacement is +3mm, the contact area decreases by 10% (2160mm). 2 Simultaneously, based on the contact area and the wheel-rail friction coefficient of 0.35, the rolling resistance (F = μN, where N is the vertical load on the wheelset (80kN) and μ is the wheel-rail friction coefficient of 0.35) is calculated using Coulomb's law of friction. The resulting resistance value for the curved section (28kN) is 12% higher than that for the straight section (25kN). A model incorporating the wheel-rail contact area (2000-2600mm²) is generated. 2), contact area change rate (-15% to +5%), contact pressure per unit area (30-40 N / mm²) 2 Rolling resistance during curved travel (25-30kN), and the relationship between the increase in resistance and the curve radius (300-meter radius corresponds to an increase of 3kN) are calculated using the contact area-resistance data for the curved track. For example, at a curve with a radius of 300 meters, the contact area is 2200mm². 2 The rolling resistance is 28kN, which is 3kN higher than that of the straight section, consistent with the theoretical calculation.

[0145] Step S424: Perform dynamic analysis of train curve track travel response based on the train curve track contact area-resistance data to generate train curve track travel response dynamic data.

[0146] In this embodiment of the invention, dynamic analysis of the train's curved track travel response is performed based on the contact area-resistance data (including contact area and rolling resistance). The contact pressure per unit area of ​​35 N / mm² is used from the contact area data. 2 By combining the elastic deformation coefficient in the track deflection displacement data, the power transmission efficiency of wheel-rail contact (90%) was calculated. Using a dynamic response model with an input rolling resistance of 28 kN and a train speed of 80 km / h, the additional power required for curve travel (overcoming resistance + providing centripetal force, totaling 150 kN) was calculated. Longitudinal (0.1 g), lateral (0.8 g), and vertical (0.3 g) accelerations during curve travel were collected by triaxial acceleration sensors (range ±2 g, accuracy ±0.01 g) installed at the bottom of the carriage. These values ​​were compared with the acceleration values ​​calculated by the model (longitudinal 0.09 g, lateral 0.82 g) to correct the dynamic response parameters. Generate curve track travel response dynamic data including total power demand for curve travel (150kN), wheel-rail contact power transmission efficiency (88%-92%), peak acceleration in each direction (lateral 0.8g), dynamic response lag time (0.1 seconds), and curve travel dynamic stability coefficient (≥1.2). For example, when a train passes through a curve with a radius of 300 meters at 80km / h, the total power demand is 150kN, the peak lateral acceleration is 0.8g, and the stability coefficient is 1.4, which meets the dynamic response safety standard.

[0147] Furthermore, step S43 includes the following steps:

[0148] Step S431: Based on the track gradient data, detect the traction spacing of each car in the train operation-track linkage data and generate the traction spacing data of each car.

[0149] In this embodiment of the invention, based on the train track gradient data (including a 2‰ uphill slope, 1000 meters in length) and the total train length of 210 meters in the train operation-track linkage data, the traction spacing of each carriage is detected. Laser distance sensors (measuring range 0-50 meters, accuracy ±0.05 meters) are installed at the connection points of the first and last carriages, collecting the distance between adjacent carriages every 0.1 seconds (standard spacing on straight sections is 1.5 meters ±0.05 meters). When the train enters an uphill section, the spacing change is recorded in real time. The spacing between the first and second carriages increases to 1.55 meters, the spacing between the second and third carriages increases to 1.53 meters, and so on, with the smallest change in spacing at the rear carriage (1.51 meters). Simultaneously, combined with the train speed of 80 km / h in the linkage data, the spacing change rate (0.02 meters / second) is calculated, and a spacing-slope relationship model is established by associating it with the gradient value (for every 1‰ increase in gradient, the spacing increases by an average of 0.03 meters). The system generates traction spacing data for each car, including real-time spacing between each car (1.5-1.6 meters), spacing change rate (0.01-0.03 meters / second), maximum spacing on the uphill section (1.6 meters), deviation of each spacing from the standard value (0.01-0.1 meters), and spacing distribution gradient (gradually decreasing from the front to the rear of the car). For example, at the midpoint K510+800 of the 2‰ uphill section, the spacing between the 5th and 6th cars is 1.54 meters, which deviates from the straight section by 0.04 meters, conforming to the spacing change law of the slope section.

[0150] Step S432: Analyze the push-pull stress of each car based on the traction spacing data of each car, and generate push-pull stress data for each car;

[0151] In this embodiment of the invention, push-pull stress analysis is performed on each car based on the traction spacing data of each car (including real-time spacing and rate of change). A spring-damping mechanical model is used to simulate the car connection device, converting the spacing change of each car in the spacing data into the tensile amount of the connection device (a spacing of 1.55 meters corresponds to a tensile amount of 0.05 meters). The tensile force is calculated according to Hooke's Law, yielding a tensile force of 52 kN between the first and second cars and 48 kN between the second and third cars. Simultaneously, strain gauge sensors (measuring range -2000 to 2000 με, accuracy ±5 με) are installed at the connection device to collect stress-strain values ​​in real time (strain at the connection of the first and second cars is 1000 με, corresponding to a stress of 50 kN). The calculated values ​​are compared with the measured values, and the deviation is controlled within 5%. Combining the slope length of 1000 meters in the slope data, the stress duration is analyzed. When the spacing is stable at 1.55 meters, the stress is maintained at 52 kN for 10 seconds. The system generates push-pull stress data for each car, including the push-pull stress values ​​between each car (40-60kN), the stress change rate (2kN / second), the conversion factor between stress and spacing (1 meter spacing corresponds to 1000kN), the maximum stress value on the uphill section (60kN), and the stress distribution trend (gradually decreasing from the front to the rear of the car). For example, on a 2‰ uphill section, the stress at the connection between the first and second cars is 52kN, which deviates from the calculated value by 2kN and is within the allowable error range. The stress distribution is consistent with the characteristic that the front of the car bears the main traction force.

[0152] Step S433: Based on the push-pull stress data of each carriage, perform axial offset detection on the train operation-track linkage data to generate axial offset data for each train;

[0153] In this embodiment of the invention, the axial offset of each train is detected based on the push-pull stress data (including stress value and rate of change) of each carriage, combined with the wheel-rail contact point coordinates in the train operation-track linkage data. A two-dimensional laser displacement sensor (measuring range -100 to 100 mm, accuracy ±0.1 mm) is installed at the bottom of each carriage to collect the offset of the carriage relative to the track centerline along the longitudinal and transverse directions of the track (longitudinal offset ±0.5 mm, transverse offset ±2 mm). Based on the stress difference in the push-pull stress data (the tension of the first carriage relative to the second carriage is 52 kN, and the tension of the second carriage relative to the third carriage is 48 kN, resulting in a stress difference of 4 kN), the axial deflection angle (0.1°) caused by uneven force distribution in the carriage is calculated and converted into a longitudinal offset (0.3 mm per carriage). The track centerline coordinates in the linkage data are used to determine the axial offset baseline, and the measured offset is compared with the calculated offset, with the deviation controlled within 0.2 mm. Generate axial offset data for each train, including longitudinal offset (-0.6 to +0.6 mm), lateral offset (-2 to +2 mm), axial deflection angle (0-0.2°), correlation between offset and stress difference (0.8), and the position of maximum offset (3rd car +0.6 mm). For example, on a 2‰ uphill section, the 3rd car has a longitudinal offset of +0.5 mm and a lateral offset of +1 mm due to a stress difference of 5 kN between the front and rear, which deviates from the calculated value by 0.1 mm, consistent with the stress offset law.

[0154] Step S434: Based on the axial offset data of each train and the push-pull stress data of each carriage, perform dynamic analysis of the train track gradient response to generate dynamic data of the train track gradient response.

[0155] In this embodiment of the invention, a dynamic analysis of the train track gradient response is performed based on the axial offset data (including offset amount and deflection angle) and the push-pull stress data (including stress value) of each train and each car. The longitudinal offset in the axial offset data is converted into additional resistance (10N increase per millimeter of offset), and the lateral offset is converted into wheel-rail lateral friction resistance (5N increase per millimeter of offset). The third car generates additional resistance of 5N + 5N = 10N due to the offset. Combining the total tension in the push-pull stress data (800kN sum of stress in each car), the effective traction force (total tension minus additional resistance, 799.9kN) is calculated using the dynamic balance equation. This is compared with the theoretical traction force of 200kN in step S43 to obtain the power utilization rate (799.9kN / 800kN = 99.99%). The actual output power is collected by the power sensor installed on the traction motor and converted into actual power output (800kW). The deviation from the calculated value is 0.1%. Generate track gradient response dynamic data that includes effective traction force on the gradient section, total additional resistance (50-100N), power utilization rate (99.5%-100%), power distribution ratio of each car (15% at the front, decreasing to 5% at the rear), and gradient response dynamic stability coefficient (effective traction force / theoretical traction force ≥ 0.95). For example, on a 2‰ uphill section, the total additional resistance is 80N, the effective traction force is 799.92kN, the power utilization rate is 99.99%, and the stability coefficient is 1.0, which meets the dynamic response requirements on the gradient section, and the power distribution of each car conforms to the stress distribution law.

[0156] Furthermore, step S44 includes the following steps:

[0157] Step S441: Based on the train track gradient response dynamic data, analyze the longitudinal force coordination and speed difference of each train on the gradient section, and generate longitudinal force coordination-speed difference data of each train on the gradient section.

[0158] In this embodiment of the invention, the longitudinal force coordination and speed difference of the train in the slope section are analyzed based on the train track gradient response dynamic data (including effective traction force and power distribution ratio). Torque sensors and speed sensors are installed at the traction motor of each train car, collecting data once per second. For the 2‰ uphill section, the deviation between the actual traction force and the theoretical value of the power distribution ratio of each car is calculated. For example, the theoretical traction force of the first car is 120kN, and the actual traction force is 118kN, with a deviation of -1.7%; the theoretical traction force of the fifth car is 80kN, and the actual traction force is 82kN, with a deviation of +2.5%. The deviation of all cars is controlled within ±3%, which is considered to be good coordination. At the same time, the speed difference between adjacent cars is calculated. The speed of the first car is 80km / h, and the speed of the second car is 79.8km / h, with a difference of 0.2km / h. The maximum speed difference of the entire train is 0.3km / h. The force coordination coefficient (1 - average absolute value of traction force deviation of each carriage) is calculated to be 0.98, and the speed difference coefficient (1 - maximum speed difference / average speed) is 0.996. This generates longitudinal force coordination-speed difference data, including the traction force deviation rate of each carriage (-3% to +3%), the speed difference between adjacent carriages (0.1-0.3 km / h), the longitudinal force coordination coefficient (≥0.95), the speed difference coefficient (≥0.99), and the force coordination stability of the slope section (no deviation for 5 minutes). For example, on a 2‰ uphill section, the overall train force coordination coefficient is 0.98, and the speed difference coefficient is 0.996, meeting the requirements for slope operation.

[0159] Step S442: Analyze the longitudinal traction deflection force of each train on the curved track based on the train's dynamic response data on the curved track, and generate longitudinal traction deflection force data for each train on the curved track.

[0160] In this embodiment of the invention, the longitudinal traction deflection force of a train traveling on a curved track is analyzed based on the train's dynamic response data (including lateral force and centripetal force). Three-dimensional force sensors (range -500 to 500 kN, accuracy ±1 kN) are installed at the wheelsets of each car to collect the longitudinal, lateral, and vertical forces between the wheels and rails during curved track travel, recording data every 0.05 seconds. For a 300-meter radius curve, the longitudinal traction deflection force is calculated using the lateral acceleration of 0.8g from the curved track dynamic response data (the formula for calculating the longitudinal traction deflection force is F = F...). h ×tanθ,F hLet θ be the lateral force and θ be the curve turning angle (a 300-meter radius corresponds to a turning angle of 1.91°). The calculated deflection forces are 15 kN for the first car, 20 kN for the fifth car, and 10 kN for the rear car. By comparing the actual longitudinal forces (14.8 kN for the first car and 20.2 kN for the fifth car) collected by the three-dimensional force sensors with the calculated values, the deviation is controlled within ±2%. Furthermore, the relationship between the deflection force and the curve position is analyzed: the deflection force gradually increases from the curve's starting point to the midpoint, and gradually decreases from the midpoint to the end point, with the maximum deflection force occurring at the midpoint. Generate longitudinal traction deflection force data for curved tracks, including the longitudinal traction deflection force values ​​(10-25kN) of each car, the ratio of deflection force to lateral force (0.03-0.05), the deviation rate between actual and calculated deflection force (-2% to +2%), the deflection force distribution curve of the curved section, and the location of the maximum deflection force (midpoint of the curve). For example, at the midpoint of a curve with a radius of 300 meters, the longitudinal traction deflection force of the 5th car is 20kN, which deviates from the calculated value by 0.2kN, and is within the allowable range, consistent with the characteristic that the force is greatest in the middle of the curved section.

[0161] Step S443: Based on the longitudinal traction deflection force data of each train on the curved track and the longitudinal force coordination-speed difference data of each train on the slope section, perform a linkage analysis of the operating status of each carriage and the train traction force, and generate linkage data of the operating status of each carriage and the train traction force.

[0162] In this embodiment of the invention, based on the longitudinal traction deflection force data of curved tracks and the longitudinal force coordination-speed difference data of slope sections, a linkage analysis of train car operation status and traction force is performed. The deflection force values ​​of each car in the deflection force data are retrieved, and combined with the traction force deviation rate in the coordination data, a dynamic traction force allocation model is used to correct the basic traction force (80kN for a 50-ton car and 96kN for a 60-ton car) according to the load distribution: for cars with deflection forces >15kN on curved sections, the traction force is increased by 5% (from 96kN to 100.8kN for the 5th car); for cars with speed differences >0.2km / h on slope sections, the traction force is finely adjusted by ±2% (from 88kN to 89.76kN for the 2nd car). Traction force actuator feedback sensors (range 0-200kN, accuracy ±0.5kN) are installed in each car to collect the adjusted actual traction force in real time, comparing it with the model output value; the deviation is ≤1kN. Simultaneously, the adjusted carriage operation status was analyzed, showing a 10% reduction in lateral vibration amplitude and a speed difference narrowing to within 0.1 km / h. Carriage operation status-traction linkage data was generated, including the adjusted traction force (80-120 kN), traction force adjustment range (-2% to +5%), adjusted lateral vibration value (0.4-0.7 g), adjusted speed difference (0.1-0.2 km / h), and traction force-operation status matching degree (≥0.95). For example, in the section where a 300-meter radius curve overlaps with a 2‰ uphill slope, the adjusted traction force of the 5th carriage is 100.8 kN, with a lateral vibration of 0.6 g, a speed difference of 0.1 km / h with the preceding carriage, and a matching degree of 0.98, achieving precise linkage between traction force and operating status.

[0163] Furthermore, step S5 includes the following steps:

[0164] Step S51: Calculate the freight load configuration of each carriage of the train based on the train operation scheduling data, and generate the freight load configuration data of each carriage of the train.

[0165] In this embodiment of the invention, the freight load configuration of each carriage is calculated based on train scheduling data (including planned speeds and arrival times for each section). The section speed limits (e.g., 90 km / h for the K500-K600 section), gradient information (2‰ uphill), and curve parameters (300-meter radius) from the scheduling data are imported into the train dispatching system, combined with the rated load (60 tons) and cargo type (density 2.5 tons / m³) of each carriage. 3The system employs a load balancing algorithm to distribute cargo. The algorithm aims to reduce the axle load deviation of each car by ≤5%, adjusting the load based on the power demand in different sections of the schedule: the load on the first three cars on uphill sections is reduced by 5% (57 tons), and the load on the cars in the middle of curve sections is reduced by 3% (58.2 tons). The actual load is collected in real time by weighing sensors (range 0-100 tons, accuracy ±0.1 tons) installed at the bottom of the cars, and compared with the calculated value; the deviation is controlled within ±0.5 tons. Generate freight load configuration data that includes the actual load of each car (57-60 tons), axle load distribution (20-22 tons / axle), load and section characteristics matching coefficient (≥0.9), maximum load car number (10th car, 60 tons), and load adjustment basis (such as the K550-K560 curve section). For example, in the K500-K600 section, the 5th car needs to pass through a 300-meter radius curve, so the load configuration is 58.2 tons, which deviates from the calculated value by 0.1 tons, and the axle load is 21 tons, which meets the load balance requirements.

[0166] Step S52: Based on the operating status of each carriage of the train - the train traction force linkage data and the freight load configuration data of each carriage of the train, perform traction power and braking analysis of each carriage of the train to generate traction power and braking data of each carriage of the train.

[0167] In this embodiment of the invention, traction power and braking analysis are performed based on the train's operating status and traction force linkage data (including adjusted traction force and vibration values) and freight load configuration data (including the load of each carriage). The traction force values ​​of each carriage in the linkage data (e.g., 100.8 kN for carriage 5) and the load data (58.2 tons) are used to calculate the traction power (P = FV, where F is the traction force and V is the dispatch speed of 25 m / s) using a power demand model, resulting in a requirement of 2520 kW for carriage 5. A power sensor (range 0-5000 kW, accuracy ±10 kW) is installed at the traction motor to collect the actual output power (2515 kW), with a deviation ≤5 kW. Simultaneously, based on the deceleration requirements at the station in the schedule (e.g., braking begins 1 kilometer before station K600), the braking distance (800 meters) is calculated using load data. The brake cylinder pressure (3 MPa corresponds to a deceleration of 0.5 m / s) is monitored using a brake pressure sensor (range 0-10 MPa, accuracy ±0.05 MPa). 2 To ensure the actual braking distance deviates from the calculated value by ≤10 meters, the system generates data including the traction power (2000-3000kW), braking pressure (2-5MPa), power reserve coefficient (actual power / demand power ≥1.1), and braking deceleration (0.3-0.6m / s²) for each carriage. 2Traction-braking data, including traction-braking switching response time (≤0.5 seconds). For example, the fifth car requires 2520kW of traction power at K550, but the actual output is 2515kW. The braking pressure is 3.2MPa, and the deceleration is 0.52m / s². 2 This meets the deceleration requirements in the scheduling.

[0168] Step S53: Set the train adaptive asynchronous control command based on the traction power-braking data of each carriage, generate the train adaptive asynchronous control command data, and execute the train adaptive asynchronous control operation based on the train adaptive asynchronous control command data.

[0169] In this embodiment of the invention, adaptive asynchronous control commands are set and control operations are executed based on the traction power-braking data (including power and pressure) of each train car. The traction power and braking pressure of each car are retrieved from the power data, and combined with the vibration value (0.6g) and speed difference (0.1km / h) from the linkage data in step S443, a fuzzy control algorithm is used to generate control commands: for cars with traction power lower than the required value by 5% (e.g., the 3rd car, 2400kW, requires 2500kW), the motor voltage is increased by 2% (from 380V to 387.6V); for cars with braking pressure deviation > 0.1MPa (e.g., the 7th car, 3.1MPa, requires 3.0MPa), the brake valve opening is reduced by 1%. Control commands are sent to the controllers of each car via the train bus (transmission rate 100Mbps, delay ≤10ms). The controllers drive the traction converters and brake valves to perform adjustments. After adjustment, the results are verified by feedback sensors (accuracy ±0.1%): the power of the 3rd car increases to 2490kW, and the pressure of the 7th car decreases to 3.02MPa, with deviations meeting the requirements. Adaptive asynchronous control command data is generated, including the voltage adjustment amount of each car (±1%-3%), brake valve opening (20%-30%), command execution response time (≤0.2 seconds), post-control parameter deviation (≤2%), and control mode (traction / braking). For example, at K550+500, due to a vibration of 0.6g, the commanded traction force of the 5th car is reduced by 1% (from 100.8kN to 99.8kN). After execution, the actual traction force is 99.7kN, with a deviation of 0.1kN, achieving precise asynchronous control.

[0170] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0171] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. An artificial intelligence-based adaptive asynchronous control method for railway freight trains, characterized by, The method comprises the following steps: Step S1: collecting train operation data through train sensors, performing train operation and track linkage processing according to the train operation data, and generating train operation-track linkage data; Step S1 comprises the following steps: Step S11: collecting train operation data through train sensors, performing train operation characteristic analysis according to the train operation data, and generating train operation characteristic data; Step S12: performing real-time train operation track detection according to the train operation characteristic data, and generating real-time train operation track data; Step S13: performing train track structure change information analysis according to the real-time train operation track data, and generating train track structure change information data; Step S14: performing train operation and track linkage processing based on the train operation characteristic data and the train track structure change information data, and generating train operation-track linkage data; Step S2: performing real-time train positioning detection based on the train operation data and the train operation-track linkage data, and generating real-time train positioning data; Step S3: obtaining train operation planning data; performing train operation scheduling design based on the train operation planning data and the real-time train positioning data, and generating train operation scheduling data; Step S3 comprises the following steps: Step S31: obtaining train operation planning data, and performing train track occupancy interval division according to the train operation planning data, and generating train track occupancy interval data; Step S32: performing each train intersection track section analysis based on the train track occupancy interval data and the real-time train positioning data, and generating each train intersection track section data; Step S33: performing train freight load configuration analysis according to the train operation data, and generating train freight load configuration data; Step S34: performing load adaptability matching of each train in the intersection track section based on the train freight load configuration data, the each train intersection track section data, and the train track occupancy interval data, and generating load adaptability data of each train in the intersection track section; Step S35: performing train operation scheduling design based on the train track occupancy interval data, the load adaptability data of each train in the intersection track section, and the each train intersection track section data, and generating train operation scheduling data; Step S4: performing train each car operating state and train traction force linkage analysis based on the real-time train positioning data and the train operation-track linkage data, and generating train each car operating state-train traction force linkage data; Step S5: setting train adaptive asynchronous control instructions based on the train each car operating state-train traction force linkage data and the train operation scheduling data, generating train adaptive asynchronous control instruction data, and performing train adaptive asynchronous control operation based on the train adaptive asynchronous control instruction data.

2. The artificial intelligence based adaptive asynchronous control method for railway freight trains according to claim 1, characterized in that, Step S2 comprises the following steps: Step S21: performing train and tower base station communication and train speed detection according to the train operation data, respectively, and generating train-tower base station communication data and train speed data, respectively; Step S22: Train running environment change monitoring is performed according to the train operation-track linkage data, and train running environment change data is generated; Step S23: Train running and tunnel running positioning analysis is performed based on the train-tower base station communication data, train speed data, and train running environment change data, and train running-tunnel running positioning data is generated; Step S24: Train running real-time positioning detection is performed based on the train-tower base station communication data, train speed data, and train running-tunnel running positioning data, and train running real-time positioning data is generated.

3. The artificial intelligence based adaptive asynchronous control method for railway freight trains of claim 2, wherein, Step S23 includes the following steps: Step S231: Train each car running tunnel sequence and time difference analysis is performed according to the train running environment change data, and train each car running tunnel sequence-time difference data is generated; Step S232: Train each car communication switching and recovery detection is performed based on the train each car running tunnel sequence-time difference data and the train-tower base station communication data, and train each car communication switching-recovery data is generated; Step S233: Train running positioning delay calculation is performed based on the train each car communication switching-recovery data and the train speed data, and train running positioning delay data is generated; Step S234: Train running and tunnel running positioning analysis is performed based on the train running positioning delay data and the train speed data, and train running-tunnel running positioning data is generated.

4. The artificial intelligence based adaptive asynchronous control method for railway freight trains of claim 1, wherein, Step S4 includes the following steps: Step S41: Train running track curve and train running track slope detection is respectively performed based on the train running real-time positioning data and the train operation-track linkage data, and train running track curve data and train running track slope data are respectively generated; Step S42: Train curve track running response power analysis is performed according to the train running track curve data, and train curve track running response power data is generated; Step S43: Train track slope response power analysis is performed according to the train running track slope data, and train track slope response power data is generated; Step S44: Train each car running state and train traction force linkage analysis is performed based on the train curve track running response power data and the train track slope response power data, and train each car running state-train traction force linkage data is generated.

5. The artificial intelligence based adaptive asynchronous control method for railway freight trains according to claim 4, characterized in that, Step S42 includes the following steps: Step S421: Train running contact axial center of gravity detection is performed on the train operation-track linkage data based on the train running track curve data, and train running contact axial center of gravity data is generated; Step S422: Train running track deflection displacement analysis is performed according to the train running contact axial center of gravity data, and train running track deflection displacement data is generated; Step S423: Train curve track contact area and resistance analysis is performed according to the train running track deflection displacement data, and train curve track contact area-resistance data is generated; Step S424: Train curve track running response power analysis is performed according to the train curve track contact area-resistance data, and train curve track running response power data is generated.

6. The artificial intelligence based adaptive asynchronous control method of railway freight trains according to claim 4, characterized in that, Step S43 includes the following steps: Step S431: detecting each car traction interval of the train operation-track linkage data according to the train track slope data, and generating each car traction interval data; Step S432: analyzing each car push-pull stress according to each car traction interval data, and generating each car push-pull stress data; Step S433: detecting each train axial offset of the train operation-track linkage data based on each car push-pull stress data, and generating each train axial offset data; Step S434: analyzing the train track slope response dynamics based on each train axial offset data and each car push-pull stress data, and generating the train track slope response dynamics data.

7. The artificial intelligence based adaptive asynchronous control method of railway freight trains according to claim 4, characterized in that, Step S44 includes the following steps: Step S441: analyzing the longitudinal force coordination and speed difference of each train in the slope section based on the train track slope response dynamics data, and generating the longitudinal force coordination-speed difference data of each train in the slope section; Step S442: analyzing the longitudinal traction deflection force of each train in the curved track based on the train curved track driving response dynamics data, and generating the longitudinal traction deflection force data of each train in the curved track; Step S443: analyzing the train each car operation state and train traction force linkage based on the longitudinal traction deflection force data of each train in the curved track and the longitudinal force coordination-speed difference data of each train in the slope section, and generating the train each car operation state-train traction force linkage data.

8. The artificial intelligence based adaptive asynchronous control method for railway freight trains of claim 1, wherein, Step S5 includes the following steps: Step S51: calculating the freight load configuration of each car of the train based on the train operation scheduling data, and generating the freight load configuration data of each car of the train; Step S52: analyzing the traction power and braking of each car of the train based on the train each car operation state-train traction force linkage data and the freight load configuration data of each car of the train, and generating the traction power-braking data of each car of the train; Step S53: setting the train adaptive asynchronous control instruction based on the traction power-braking data of each car of the train, generating the train adaptive asynchronous control instruction data, and performing the train adaptive asynchronous control operation based on the train adaptive asynchronous control instruction data.

Images