Scheduling and control integrated method for virtual marshalling of single-line bidirectional railway heavy haul train
By constructing a two-layer optimization framework for train operation control and scheduling planning, and combining acceleration, position update, and safety distance constraints, the problem of disconnect between scheduling and control of heavy-haul trains on single-track bidirectional railways was solved, achieving efficient scheduling and control integration and improving transportation efficiency and safety.
Patent Information
- Application Number
- CN202511740912.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-25
- Publication Date
- 2026-01-02
AI Technical Summary
The existing scheduling mode for heavy-haul trains on single-track bidirectional railways suffers from limited transport capacity and a disconnect between scheduling and control. In particular, the need to dismantle trains when they meet is cumbersome and cannot adapt to the dynamic requirements of virtual train formation, resulting in low transport efficiency.
A two-layer optimization framework for train operation control and scheduling is constructed. By combining constraints such as acceleration, position update, and safety distance, an efficient scheduling plan is generated through a multi-stage iterative algorithm to ensure the optimization of train formation/deformation time and running trajectory, and to achieve deep integration of scheduling and control.
It improves the transport density and vehicle turnover efficiency of single-track two-way railways, reduces calculation time, meets the real-time scheduling needs of heavy-haul railways, and ensures traffic safety.
Smart Images
Figure CN121246892A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of single-track bidirectional railway transportation scheduling technology, specifically to an integrated scheduling and control method for virtual train formation of heavy-haul trains on single-track bidirectional railways. Background Technology
[0002] Heavy-haul railways, as the core carrier for long-distance transportation of bulk commodities (such as coal, ore, and steel), play an irreplaceable role in ensuring energy security and supporting industrial production due to their advantages of large capacity, low energy consumption, and low cost. Among them, single-track bidirectional (STB) lines are widely distributed between resource-producing and consumption areas due to their low construction costs and strong terrain adaptability, and are an important part of the heavy-haul railway network. Their transportation efficiency directly affects the smoothness of the entire logistics chain.
[0003] In existing technologies, heavy-haul trains on STB lines are typically coupled mechanically to form combined trains of 20,000 tons or more. To avoid head-on collisions between trains in both directions, trains must pass each other on the siding at intermediate stations—that is, a train in one direction stops on the siding and waits for the oncoming train to pass before continuing its journey. Dispatching plans are mostly based on the running time of fixed block sections, only considering the independent operation characteristics of a single train, without taking into account the dynamic operational needs of trains.
[0004] However, the existing operation and scheduling model has significant drawbacks: on the one hand, the length of mechanically coupled trains often exceeds the length of the intermediate station siding, requiring trains to be disassembled when passing each other, which is cumbersome and time-consuming. This means that STB lines can only allow trains to pass in one direction at a time, limiting transport capacity. On the other hand, after the application of virtual formation technology, trains need to dynamically complete formation / deformation operations. The existing scheduling plan based on fixed running time cannot adapt to its dynamic speed and distance characteristics, and the scheduling and control links are disconnected, which can easily lead to the plan becoming unexecutable. At the same time, traditional solution methods have low computational efficiency and cannot meet the real-time scheduling requirements, which seriously restricts the realization of the heavy-haul transport potential of STB lines. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention provides an integrated scheduling and control method for virtual train formation of heavy-haul trains on single-track bidirectional railways, solving the problem of balancing transportation efficiency and train operation safety caused by limited track resources on single-track railways.
[0006] To achieve the above objectives, the present invention provides the following technical solution: an integrated scheduling and control method for virtual train formation of heavy-haul trains on single-track bidirectional railways, comprising the following steps: Step 1: Obtain the line parameters, unit train parameters, and transportation demand parameters for a single-track, two-way railway; Step 2: Input the line parameters, unit train parameters and transportation demand parameters output in Step 1 into the train operation control module, construct the formation time optimization model and the unforming time optimization model, and calculate the formation time, unforming time and section running time; Step 3: Input the train formation time, unforming time, and interval running time output in Step 2, as well as the transportation demand parameters output in Step 1, into the train scheduling module to construct a total travel time optimization model and calculate the arrival time and departure time of the unit train. Step 4: Input the arrival and departure times of the unit trains output in Step 3 into the multi-stage iterative algorithm module, solve the total travel time optimization model in stages based on heuristic rules, and output the comprehensive scheduling plan for heavy-haul trains on a single-track bidirectional railway.
[0007] Preferably, in step 1: The line parameters include station locations, section length, turnout locations, and maximum speed limit per section. The unit train parameters include train length, maximum speed, and braking acceleration; The transportation demand parameters include the number of trains and the direction of travel.
[0008] The preferred grouping time optimization model is based on For the goal, among which and These are the times when the two trains reach their maximum formation position, respectively. and All figures refer to train length.
[0009] Preferably, the model constraints in step 2 include acceleration constraints, position update constraints, and safety distance constraints; Acceleration constraints: ,in Let k be the speed of unit train k at time t. For unit train At any moment speed, The maximum acceleration threshold for a single train. The minimum time interval; Position update constraints: ,in For unit train At any moment Location, For unit train At any moment Location; Safety distance constraints also include RBD mode and ABD mode; RBD mode: ,in =50m; ABD mode: ,in This is the absolute braking margin.
[0010] Preferably, the total travel time optimization model in step 3 is based on For the goal, among which This represents the total time for trains to arrive at the station. Let the total time of the departing train be the total time, and satisfy the following conditions: Train interval constraints: ,in Minimum departure interval; Virtual grouping constraints: ,in This is the baseline value for the continuous hanging time interval; Meeting constraints: and ,in For interval occupancy priority, It is a positive number.
[0011] Preferably, the multi-stage iterative algorithm in step 4 includes heuristic rules, iterative logic, and termination conditions; Heuristic rules: and ,in For interval occupancy priority, This indicates a return / turnaround status during the delivery phase. Iterative logic: Divide the model into sub-models based on the number of turnovers n, and base it on the previous stage. Update the meeting constraints of the current sub-model; Termination condition: The number of iterations m ≥ n, where m is the current iteration number and n is the number of unit train turns.
[0012] This invention provides an integrated scheduling and control method for virtual train formation of heavy-haul trains on single-track bidirectional railways. It offers the following advantages: 1. This invention constructs a two-layer optimization framework for train operation control and scheduling planning. The upper layer optimizes the formation / deformation time and running trajectory of virtual train formations, while the lower layer generates scheduling plans by combining constraints such as train intervals, meeting trains, and virtual formations. This achieves deep integration of scheduling and control. The framework solves the problem of disconnect between scheduling and control of virtual train formations on STB lines, effectively improving the line's transport density and vehicle turnover efficiency.
[0013] 2. This invention designs a multi-stage iterative algorithm, introduces heuristic rules to narrow the solution space, and uses the priority variable of the previous stage interval occupancy to update the constraints of the next stage, avoiding repeated calculations. Compared with the commercial solver CPLEX and the traditional two-stage method, this algorithm can generate high-quality scheduling plans more efficiently, significantly reducing the calculation time while ensuring train operation safety, and meeting the real-time scheduling needs of heavy-haul railways. Attached Figure Description
[0014] Figure 1 This is a system flowchart of the present invention. Detailed Implementation
[0015] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0016] Please see the appendix Figure 1 This invention provides an integrated scheduling and control method for virtual train formation of heavy-haul trains on single-track bidirectional railways, comprising the following steps: Step 1: Obtain the line parameters, unit train parameters, and transportation demand parameters for a single-track bidirectional railway. The line parameters include station location, section length, turnout location, and maximum speed limit per section. The unit train parameters include train length, maximum speed, and braking acceleration. The transportation demand parameters include the number of trains and their direction of travel. Output the line parameters, unit train parameters, and transportation demand parameters. The station location is the coordinate value distributed along the main line (unit: m), the section length is the difference between the center coordinates of two adjacent stations, and the turnout location is the coordinate of the track node used for train turning within the station. The maximum speed limit for each section is set separately (unit: m / s). In the unit train parameters, the train length is the fixed formation length of a single heavy-duty train, the maximum speed is the highest operating speed allowed by the train design (unit: m / s), and the braking acceleration is the maximum deceleration of the train during emergency braking (taken as a negative value, unit: m / s²). In the transportation demand parameters, the number of trains is counted separately according to the direction of entry (e.g., from the end point to the starting point) and the direction of exit (e.g., from the starting point to the end point). The directions of travel are distinguished by preset up / down indicators.
[0017] Step 2: Input the line parameters, unit train parameters, and transportation demand parameters output from Step 1 into the train operation control module to construct the train formation time optimization model and the train unforming time optimization model. The train formation time optimization model uses the objective function... Minimize as the objective, where These are the times when the two train units arrive at their maximum formation position, respectively. , The lengths of the two train units are respectively; the decoupling time optimization model uses the objective function , Minimize as the objective, where These are the times when the two train units arrive at the decoupling position, respectively; the train formation time, decoupling time, and section travel time are calculated through the model, and the formation time, decoupling time, and section travel time are output. Maximum group position For the pre-defined coordinates of specific stations or sections where train formation is permitted (such as the throat area of an intermediate station), the train formation time optimization model determines the time by solving the train motion equations. , Ensure that the time difference between the arrival of the two trains at this position meets the requirements for train formation operation (e.g., ≤60s), and the train formation position is as follows. , The coordinates of the siding for the terminal or intermediate station are used. The decoupling time optimization model calculates the time from train formation to decoupling and stopping at the siding, ensuring that the decoupling process does not affect the passage of subsequent trains. The section running time is the travel time of the train from the start to the end of the section, which is determined by both position update constraints and speed limit constraints.
[0018] Step 3: Input the train formation time, unforming time, and section travel time output from Step 2, along with the transportation demand parameters output from Step 1, into the train scheduling module to construct a total travel time optimization model. The total travel time optimization model uses the objective function... , Minimize as the objective, where This represents the total travel time of the train entering the station. The total travel time of the departing unit train is given. The model satisfies the headway constraint, bidirectional operation constraint, meeting constraint, and virtual formation constraint. The arrival and departure times of the unit trains are calculated through the model and output. Total travel time This refers to the total time it takes for a train to travel from its departure station to its arrival station. The objective function is to calculate the total travel time of a train from its origin station to its destination station. It achieves global optimization by accumulating the total travel time of all trains entering and leaving the station. The headway constraint ensures that trains traveling in the same direction do not collide. The bidirectional operation constraint ensures that the arrival and departure logic of trains at the station is consistent (e.g., the departure time is not earlier than the arrival time). The meeting constraint avoids bidirectional train conflicts by prioritizing the use of sections. The virtual formation constraint ensures the synchronization of train formations. The model is solved using a linear programming method, and the output arrival and departure times are accurate to the second.
[0019] Step 4: Input the arrival and departure times of the unit trains output in Step 3 into the multi-stage iterative algorithm module. Solve the total travel time optimization model in stages based on heuristic rules. The heuristic rules include... and ,in For trains departing from station unit within section b With the station-entry unit train Priority variables of occupancy For the turnaround state variables of the unit train during the delivery phase, The state variables are the turnaround states of unit trains during the collection and transportation phase; the solution is iteratively solved until the termination condition is met, and the comprehensive scheduling plan for heavy-haul trains on a single-track bidirectional railway is output.
[0020] The multi-stage iterative algorithm divides the complete scheduling cycle into multiple sub-stages based on the number of train turnarounds. Each sub-stage corresponds to one round trip. In the heuristic rules, =1 indicates that a train is departing from the station. The vehicles providing the service need to turn back to serve the trains entering the station. Service, at this time =1 (Exit trains have priority to occupy the section). Indicates the arrival of trains at the station. The service vehicles need to turn back to perform the outbound train h+1 service, at which time =0 (inbound trains have priority to occupy the section), each iteration passes the previous stage. Adjust the passing constraints and gradually optimize the scheduling plan.
[0021] The stepwise optimization of the scheduling plan specifically involves: in each iteration, using the solution obtained from the (m-1)th turnaround... Substitute the values into the passing constraints of the current sub-model, update the constraint boundaries, and then solve the sub-model. The integrated dispatching plan includes the arrival and departure times of each train at each station, the timing of marshalling / demarshalling operations, and the speed curve of the section.
[0022] In this embodiment of the invention, step 2, the train formation time optimization model further includes basic train operation constraints, which include acceleration constraints and position update constraints; the acceleration constraints satisfy... ,in Let k be the speed of unit train k at time t. For unit train At any moment speed, The maximum acceleration threshold for a single train. The minimum time interval is given; the position update constraint is satisfied. ,in For unit train At any moment Location, For unit train At any moment The location.
[0023] The minimum time interval δ is the system sampling period (e.g., 1 second), and the acceleration constraint limits the change in train speed between adjacent sampling times (e.g., ...). When the velocity is 0.5 m / s², the velocity change is ≤0.5 m / s. To ensure the smooth operation of the train, the position update constraint is based on the uniform motion approximation (the velocity change within a short time interval is negligible). The current position is calculated by the velocity and time interval of the previous moment, and the accumulated error is eliminated through periodic calibration.
[0024] In this embodiment of the invention, step 2, the decoupling time optimization model further includes a safety distance constraint. The safety distance constraint includes a first safety distance constraint based on relative braking distance in moving block mode, and a second safety distance constraint based on absolute braking distance in moving block mode; the first safety distance constraint satisfies... ,in Forward unit train At any moment Location, For the following unit train At any moment Location, For relative braking safety margin, The relative braking distance is based on the speed k of the preceding unit train; The second safety distance constraint is satisfied. ,in For absolute braking safety margin, For rear-running unit trains Absolute braking distance of speed.
[0025] Relative braking distance This indicates the distance a train travels from emergency braking to a complete stop; absolute braking distance. This indicates the distance from which a following train comes to a complete stop after emergency braking. , These are empirical values (e.g., 50m, 30m) used to compensate for uncertainties such as braking delay. The two constraints are activated during the train formation (relative distance must be maintained) and train unforming (absolute distance must be maintained) stages, respectively.
[0026] In this embodiment of the invention, in step 3, the vehicle interval constraint is satisfied. ,in This refers to the arrival time of the latter train in one of two adjacent train units traveling in the same direction. This refers to the arrival time of the preceding train in one of two adjacent train units traveling in the same direction. This is the minimum arrival time interval for trains departing from the station.
[0027] In this embodiment of the invention, in step 3, the vehicle interval constraint is satisfied. ,in This refers to the arrival time of the latter train in one of two adjacent train units traveling in the same direction. This refers to the arrival time of the preceding train in one of two adjacent train units traveling in the same direction. This is the minimum arrival time interval for trains departing from the station.
[0028] Minimum arrival time interval The interval constraint is determined based on the section length and train speed (e.g., 300s), ensuring that the following train can only reach the start of the section after the preceding train has completely left the section, thus avoiding rear-end collisions in the same direction. The same logic applies to the headway constraint for trains entering the station. .
[0029] In this embodiment of the invention, in step 3, the meeting constraint is satisfied. and in For departure unit trains Drive out of the section Time, For station entry unit trains Entering the section Time, For trains departing from station unit within section b With the station-entry unit train Priority variables of occupancy It is a sufficiently large positive number.
[0030] M is set to 10 times the maximum running time of the interval (e.g., 3600s) to ensure the constraint logic is valid: when When =1 (outbound priority), the previous formula simplifies to: ≥ (Only after the departing train has left the station can the arriving train enter); When When =0 (entry priority), the latter equation simplifies to: ≥ (After an inbound train has departed, an outbound train may enter), to avoid conflicts between trains in both directions within the section.
[0031] In this embodiment of the invention, in step 3, the virtual grouping constraint satisfies and ,in This refers to the arrival time of the next train entering the station within the same virtual train formation. This refers to the arrival time of the previous train entering the station within the same virtual train formation. This is the reference value for the train formation time interval of the station entry unit. For station entry unit trains and Grouping state variables, For sufficiently large positive numbers, where When =1, the constraint takes effect (the two trains must meet the formation time interval requirement). When =0, the constraint is invalid (the two trains do not need to be grouped together).
[0032] Grouping state variables =1 indicates that two trains need to be combined into one train; in this case, the constraint simplifies to... (like =60s), ensuring synchronized arrival times; =0 indicates that no grouping is required, and the constraint is automatically invalidated (because M is large enough). The virtual grouping constraint of departing trains adopts the same logic.
[0033] In this embodiment of the invention, in step 4, the termination condition of the multi-stage iterative algorithm is that the number of iterations m ≥ n, where m is the current iteration number and n is the number of unit train turns; in each iteration, based on the solution obtained in the previous stage... The constraints of the current stage sub-model are updated, and then the current stage's value is obtained by solving the sub-model. value.
[0034] The number of turns, n, represents the number of times the train completes one round trip (e.g., 2 times). Each iteration corresponds to one turn. In the initial stage of the iteration... The default value is used (e.g., 0, priority for entering the station). In subsequent stages, the priority of meeting constraints is adjusted based on the previous results, gradually narrowing the solution space. When m=n, the output is... The value covers the entire turnover cycle, ensuring the consistency of the scheduling plan.
[0035] In this embodiment of the invention, the maximum speed limit of the line parameters in step 1 is denoted as... In step 2, the unit train's running speed within section b is full. ,in For unit train In the interval Inner Time Operating speed, maximum speed limit within the section Speed constraints are set differently based on route conditions (e.g., 60 km / h on curved sections, 80 km / h on straight sections) and converted to m / s units (e.g., 60 km / h ≈ 16.7 m / s). Speed constraints are implemented through real-time monitoring. This function triggers a deceleration command when the speed limit is approached, ensuring that the speed is not exceeded.
[0036] Maximum speed limit in the section Speed limits are set differently based on route conditions (e.g., 60 km / h on curved sections, 80 km / h on straight sections) and converted to m / s units (e.g., 60 km / h ≈ 16.7 m / s). Speed constraints are monitored in real time. This function triggers a deceleration command when the speed limit is approached, ensuring that the speed is not exceeded.
[0037] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A method for integrated scheduling and control of virtual train formation for heavy-haul trains on single-track bidirectional railways, characterized in that, Includes the following steps: Step 1: Obtain the line parameters, unit train parameters, and transportation demand parameters for a single-track, two-way railway; Step 2: Input the line parameters, unit train parameters and transportation demand parameters output in Step 1 into the train operation control module to construct the formation time optimization model and the unforming time optimization model, and calculate the formation time, unforming time and section running time. Step 3: Input the train formation time, unforming time, and interval running time output in Step 2, as well as the transportation demand parameters output in Step 1, into the train scheduling module to construct a total travel time optimization model and calculate the arrival time and departure time of the unit train. Step 4: Input the arrival and departure times of the unit trains output in Step 3 into the multi-stage iterative algorithm module, solve the total travel time optimization model in stages based on heuristic rules, and output the comprehensive scheduling plan for heavy-haul trains on a single-track bidirectional railway.
2. The integrated scheduling and control method for virtual train formation of heavy-haul trains on a single-track bidirectional railway according to claim 1, characterized in that, In step 1: The line parameters include station locations, section length, turnout locations, and maximum speed limit per section. The unit train parameters include train length, maximum speed, and braking acceleration; The transportation demand parameters include the number of trains and the direction of travel.
3. The integrated scheduling and control method for virtual train formation of heavy-haul trains on a single-track bidirectional railway according to claim 1, characterized in that, Grouping time optimization model For the goal, among which and These are the times when the two trains reach their maximum formation position, respectively. and All figures refer to train length.
4. The integrated scheduling and control method for virtual train formation of heavy-haul trains on a single-track bidirectional railway according to claim 1, characterized in that, The model constraints in step 2 include acceleration constraints, position update constraints, and safety distance constraints; Acceleration constraints: ,in Let k be the speed of unit train k at time t. For unit train At any moment speed, The maximum acceleration threshold for a single train. The minimum time interval; Position update constraints: ,in For unit train At any moment Location, For unit train At any moment Location; Safety distance constraints also include RBD mode and ABD mode; RBD mode: ,in =50m; ABD mode: ,in This is the absolute braking margin.
5. The integrated scheduling and control method for virtual train formation of heavy-haul trains on a single-track bidirectional railway according to claim 1, characterized in that, The total travel time optimization model in step 3 is based on For the goal, among which This represents the total time for trains to arrive at the station. Let the total time of the departing train be the total time, and satisfy the following conditions: Train interval constraints: ,in Minimum departure interval; Virtual grouping constraints: ,in This serves as the baseline value for the grouping time interval; Meeting constraints: and ,in For interval occupancy priority, It is a positive number.
6. The integrated scheduling and control method for virtual train formation of heavy-haul trains on a single-track bidirectional railway according to claim 1, characterized in that, The multi-stage iterative algorithm in step 4 includes heuristic rules, iterative logic, and termination conditions; Heuristic rules: and ,in For interval occupancy priority, This indicates a return / turnaround status during the delivery phase. Iterative logic: Divide the model into sub-models based on the number of turnovers n, and base it on the previous stage. Update the meeting constraints of the current sub-model; Termination condition: The number of iterations m ≥ n, where m is the current iteration number and n is the number of unit train turns.