Freight railway dispatching and train control coordination method and system

Abstract

The application provides a freight railway dispatching and train control coordination method, which comprises the following steps: extracting features of freight railway lines including basic data, dispatching command data and train operation data, and constructing a system model; obtaining various data in a train control production system, combining an actual transportation production plan, simulating a train operation process and a dispatching command strategy in the constructed system model, and deducing an optimized solution strategy of the system model in each set scene in real time; evaluating the optimized solution strategy; and applying the optimized solution strategy meeting the requirements in the evaluation to the train control production system to guide actual transportation production. The application breaks the information barrier between a dispatching system and a train control system, and effectively improves the transportation efficiency of the freight railway.

Description

Technical Field

[0001] This invention relates to the field of rail transit communication, specifically to a method and system for coordinated scheduling and train control of freight railways. Background Technology

[0002] The daily operation of freight railways can be divided into three levels: the planning system, the dispatching system, and the control system. The planning system is responsible for compiling a planned operation schedule for a given period, and formulating work plans for train operation, locomotive utilization, construction, etc. The dispatching system coordinates, allocates, and utilizes stations, sections, trains, and locomotives within the system according to the planned operation schedule. The control system is responsible for managing routes, opening signals, and controlling train operation. The dispatching system and the train control system need to exchange information in real time. Dispatchers send operation plans, dispatching instructions, train route information, temporary speed limit orders, etc., to the train control system through the dispatching system. The train control system reports routes, signal status, section status, train position, speed, etc., to the dispatching system, cooperating to complete transportation tasks.

[0003] The actual train operation environment is complex and changeable. In addition to the general characteristics of general systems, it also has the following special characteristics: (1) High complexity. The freight railway train operation environment is complex, with many conditions and internal interference factors affecting train operation. The dispatching and train operation depend on the dispatcher and driver's control of environmental information and train performance. The ability to handle and respond to emergencies varies from person to person, resulting in poor stability and uniformity of train operation. After the operation control system developed from fixed block to moving block, the train operation density is high, the interval is small, and the coupling between trains is strong. Poor train control effect will have a great impact on the efficient and punctual operation of the system. For example, it will cause freight trains to stop unexpectedly, which will not only increase the additional time for starting and stopping, but also increase the locomotive energy consumption. (2) High dynamism. Under the influence of internal and external factors, the operation process of freight trains often deviates from the existing plan, and the dispatching and train operation status are constantly changing dynamically. At present, dispatchers cannot grasp fine-grained train status information (such as speed, position, train type, traction and braking characteristics), and the stage adjustment plan formulated based on this is not precise enough; at the same time, dispatchers have limited ways to obtain the internal status of the train control system, such as train communication failures, train degrading, etc., and cannot make effective decisions on handling methods in the first time, which is not conducive to information coordination between dispatchers and drivers at different levels of vehicles, lines, and networks, and reduces the response speed to emergencies. (3) High coupling. The freight railway network is complex, with many interference factors, strong coupling between trains, lines, and stations, and the propagation process of train delays in the network is complex, making it difficult for dispatchers to make adjustment decisions.

[0004] Passing on opposite trains is a crucial task for freight railway dispatching and train control systems. After the oncoming train arrives at the station and stops on the track, other trains can pass through the passing station. To maximize throughput, the process is typically very efficient.

[0005] Please see Figure 2 This diagram illustrates a passing maneuver between trains traveling in opposite directions. In actual transport operations, this passing maneuver requires coordination among personnel from multiple positions, as well as between the dispatching and train control systems. Dispatchers develop a passing plan through the CTC (Centralized Dispatching System) and issue it to the station's trackside train control system for route processing. The train control system calculates the train's operating authorization, and the onboard system calculates the speed curve based on the authorization and track data. The plan is then communicated to the driver, who, under the protection of the onboard system, completes the passing maneuver. Typically, dispatchers need to adjust the plan multiple times based on train operation, the train control system needs to process routes and calculate authorizations multiple times, and the driver needs to adjust their driving strategy according to the plan and forward signals. Poor coordination can lead to unnecessary speed reductions or external stops for freight trains.

[0006] Therefore, there is an urgent need to study methods and technologies for the coordinated optimization of dispatching and train operation control, in order to develop an efficient, safe, and entirely new coordinated system for railway dispatching and train operation control. Currently, there is limited research on the coordinated optimization of dispatching and train operation in freight railways. Summary of the Invention

[0007] This invention proposes a framework for the coordinated optimization of scheduling and train control systems in complex freight railway networks and time-varying environments. It addresses the modeling, decision-making, and optimization control issues in this problem. An improved dual-depth Q-network train target speed curve optimization method is designed to solve the model. Through computational experiments, various operational scenarios and emergencies are "deduced and predicted" as a reference for actual scheduling and train driving. By incorporating the computational simulation results into the analysis and utilization, coordinated optimization of management and control is achieved.

[0008] A method for coordinated freight railway scheduling and train control includes the following steps:

[0009] S1. Extract features from basic data, dispatching and command data, and train operation data of freight railway lines to construct a system model;

[0010] S2. Acquire various data from the train control production system, combine them with the actual transportation production plan, simulate the train operation process and dispatching command strategy in the constructed system model, and deduce the optimization solution strategy of the system model in each set scenario in real time.

[0011] S3. Evaluate the optimization solution strategy;

[0012] S4. Apply the optimization strategies that meet the requirements in the evaluation to the train control production system to guide actual transportation production.

[0013] Furthermore, the system model in step S1 includes: a basic data model, a scheduling system model, a train control system model, and a train driving model.

[0014] Furthermore, the various types of data in step S2 include various emergencies that may occur during train operation.

[0015] Furthermore, the optimization solution strategy in step S2 is a train speed curve and scheduling adjustment plan applicable to the given scenario.

[0016] Furthermore, step S2 includes using a Q-network algorithm to set the operating environment for passing trains and yielding trains, simulating the yielding process for passing trains and yielding trains, and establishing a Markov decision process;

[0017] The operating environment includes static data of the line, train operation plans, train location, speed and time.

[0018] Furthermore, the described operation process can be expressed as the process of the train running from station k-1 to station k and the process of the train running from station k+1 to station k.

[0019] Furthermore, the sections from station k-1 to station k and from station k+1 to station k are divided into different sub-sections according to fixed speed limits, and the maximum traction speed and maximum braking speed that will allow trains or passing trains are calculated from each sub-section.

[0020] Based on the speed at which the train will pass or the train is in its current state, the operating speed of the next state and the operating speed of the previous state are calculated. Under the premise of not exceeding the speed limit of the corresponding sub-section, the maximum traction speed is obtained based on the operating speed of the next state and the maximum braking speed is obtained based on the operating speed of the previous state. The smaller value between the maximum traction speed and the maximum braking speed is taken as the shortest running time speed value.

[0021] Furthermore, the train's arrival time at station k will be: , To ensure the train's departure time at K-1 station, To ensure the train arrives at station K at the designated time. This will allow the train to travel in sub-sections of varying lengths. w This is the number of sub-sections that the train will pass through. This is the speed value that will allow the train to travel in the shortest possible time;

[0022] The train's arrival time at station K is: , This refers to the departure time of the train at station k+1. For the arrival time of the train at station k, This represents the length of the sub-section through which the train passes. z This represents the number of sub-sections traversed by the train. This represents the speed value with the shortest travel time for the train.

[0023] Furthermore, the objective function for optimization is to always minimize the time: ,and , The time for trains to complete route procedures and open signals.

[0024] Furthermore, if a smooth process is desired, minimizing sudden accelerations and decelerations, the objective function for optimization is to minimize the cumulative rate of change in the acceleration space. ,in and These are the length of the train sub-section and the length of the train passing through the sub-section, respectively. It is the acceleration that will cause the train to reach the end of the sub-section. It will cause the train to accelerate at the starting point of the sub-section. It is measured by the acceleration of the train at the end of the sub-section. It is measured by the acceleration of the train at the starting point of the sub-section. This will affect the train's travel time within the sub-section. It is measured by the train's travel time within a sub-section.

[0025] Furthermore, step S2, which employs the Q-network algorithm to optimize the train operation process, includes the following steps:

[0026] S21. Initialize the approximation network parameters Let the target network parameters for ;

[0027] S22 will affect the train's operating status. Depend on Sure, The length of the sub-interval. To allow the train to move faster, The distance from the train to the stopping point. To allow trains to pass through at their speeds, The weighted values ​​of the objective function are... The probability of randomly selecting working conditions and actions , with 1- The probability of choosing the current optimal state ;

[0028] S23, Execute working condition actions This will allow the train's operating environment to enter the next state. and reward value ;

[0029] S24. Randomly select several from the experience replay area. Update network parameters according to gradient descent algorithm ;

[0030] S25. Update network parameters after a specified number of steps. for If the number of iterations is reached, exit; otherwise, return to S21.

[0031] This invention also proposes a freight railway dispatching and train control coordination system to implement the aforementioned dispatching and train control coordination method, comprising the following steps:

[0032] The perception learning module is used to extract features from basic data, dispatching and command data, and train operation data of freight railway lines, and to build a computable and reconfigurable system model.

[0033] The collaborative optimization module is used to acquire various types of data from the production system, combine them with the actual transportation production plan, simulate the train operation process and dispatching command strategy in the production environment, and deduce the optimization solution strategy of the system model in real time.

[0034] The strategy evaluation module tests and evaluates the optimization strategies generated by the collaborative optimization module.

[0035] If the strategy evaluation module pushes the optimization solution strategy that meets the predefined goals to the production system for execution.

[0036] Furthermore, the system model includes: a basic data model, a scheduling system model, a train control system model, and a train driving model.

[0037] Furthermore, the various types of data are derived from randomly generated unexpected events.

[0038] Furthermore, the optimization solution strategy includes: train speed curves and scheduling adjustment plans.

[0039] The present invention has the following beneficial effects:

[0040] 1. Break down information barriers between the dispatching system and the train control system to effectively improve the efficiency of freight railway transportation.

[0041] 2. A freight railway train passing model was established, providing a detailed description of the operation process and planned arrival time of through trains and passing trains. This model can accurately describe the coordination process of through trains and passing trains as they travel to the passing station, laying the foundation for optimizing passing operations.

[0042] 3. Considering the speed and position of the opposing trains, the calculation method of the dual-depth Q network is improved. The advantage is that when calculating the train curve and output conditions, the speed of the train that will give way and the distance to the passing point are taken into account. From the perspective of scheduling plan, the operation process of the two trains can be coordinated and optimized at the same time. Attached Figure Description

[0043] Figure 1 This invention provides a collaborative optimization framework for scheduling and train control.

[0044] Figure 2 This is a diagram illustrating how trains can pass each other and how trains can pass.

[0045] Figure 3 For the train to make way for the scheduling and control process;

[0046] Figure 4 The train will optimize the reinforcement learning process;

[0047] Figure 5 A schematic diagram of the speed curve considering the train's position and speed. Detailed Implementation

[0048] The scheduling and train control coordination method and system proposed in this invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of this invention will become clearer from the following description.

[0049] This invention provides a freight railway dispatching and train control collaborative system, comprising a perception learning module, a collaborative optimization module, a strategy evaluation module, and a strategy generation module. Its core is to utilize perception learning and artificial intelligence technologies to break down information barriers between the dispatching and train control systems, achieving collaborative optimization of dispatching command and train operation control. The generated strategies are then dynamically fed back to the production dispatching and train control systems in real time, guiding the train control system to ensure safe, punctual, and stable train operation. This assists dispatchers in understanding the status of train, locomotive, freight car, station routes, and section conditions, enabling unified and optimized dispatching command and improving transportation efficiency.

[0050] like Figure 1 As shown, the perception learning module is responsible for extracting features from basic data, dispatching and command data, and train operation data of freight railway lines using knowledge engineering methods, and constructing a computable and reconfigurable system model. The system model includes a basic data model, a dispatching system model, a train control system model, and a train driving model, etc.

[0051] The collaborative optimization module is responsible for using various data acquired from the production system, combined with the actual transportation production plan, to simulate the train operation process and dispatching strategies in the production environment, and to deduce the optimization solutions for the system model constructed by the perception learning module in real time. The optimization solutions include designing daily operation scenarios for freight railways, causing trains to run in the simulated environment according to a given dispatching plan, randomly generating various emergencies, automatically generating train speed curves and dispatching strategies, and obtaining driving strategies suitable for the set scenarios.

[0052] The strategy evaluation module is responsible for testing and evaluating the scheduling strategies and speed curves generated by the collaborative optimization module, based on specific operational scenarios and objectives.

[0053] If the strategy evaluation module finds that a certain strategy meets the predefined goal, the strategy generation module will push it to the scheduler and driver in the production system to guide them in transportation production.

[0054] This invention constructs a freight railway scheduling and train control collaborative system. Utilizing simulation techniques, it designs scheduling adjustments and train control experiments for complex operational scenarios. State parameters under different scenarios are extracted from actual production systems. Real-world input data is used for model training, generating a large amount of simulated data. This allows the system to continuously learn from massive amounts of data, improving its scenario awareness and problem-solving capabilities. It also effectively evaluates and predicts the decision-making effectiveness of scheduling and drivers. The learning results not only "simulate" the actual situation but also provide an effective solution for actual production.

[0055] Train passing is an important task handled by the freight railway dispatching and train control system. Figure 2 This is a timeline diagram showing how train m yields to train n at station K (the horizontal axis represents time, the vertical axis represents stations, and diagonal lines indicate the train's movement between stations). Train m must arrive at the station, connect to the track, and stop before train n can pass through the yielding station. To improve throughput, the process is usually made very compact.

[0056] Currently, train passing requires coordination among personnel from multiple positions, as well as between the dispatching and train control systems: dispatchers formulate passing plans through the centralized dispatching system and issue them to the station's trackside train control system for route processing. The train control system calculates the train's operating authorization, and the onboard system calculates the speed curve based on the authorization and track data. The plan must also be communicated to the driver, who then drives the train to complete the passing maneuver under the protection of the onboard system. Typically, dispatchers need to adjust the plan multiple times based on the train's operation, the train control system needs to process routes and calculate authorizations multiple times, and the driver needs to adjust their driving strategy accordingly based on the plan and signals ahead. Poor coordination can lead to unnecessary speed reductions or external stops for freight trains.

[0057] To achieve precise passing between two trains, the scheduling and train control coordination system proposed in this invention constructs a unified system model for the passing process. Coordinated control enables the train to accelerate into the station and stop, and then appropriately reduces the train's speed.

[0058] Based on the aforementioned scheduling and train control coordination system, this invention provides a freight railway scheduling and train control coordination method, comprising the following steps:

[0059] S1. Extract features from basic data, dispatching and command data, and train operation data of freight railway lines to construct a system model;

[0060] S2. Acquire various data from the train control production system, combine them with the actual transportation production plan, simulate the train operation process and dispatching command strategy in the constructed system model, and deduce the optimization solution strategy of the system model in each set scenario in real time.

[0061] S3. Evaluate the optimization solution strategy;

[0062] S4. Apply the optimization strategies that meet the requirements in the evaluation to the train control production system to guide actual transportation production.

[0063] Furthermore, step S2 includes using a Q-network algorithm to set the operating environment for passing trains and yielding trains, simulating the yielding process of passing trains and yielding trains, and establishing a Markov decision process; the operating environment includes static line data, the operating plans for passing trains and yielding trains, the operating position, speed, and operating time of passing trains, and the operating position, speed, and operating time of yielding trains.

[0064] The term "transfer operation process" refers to the process of the train moving from station k-1 to station k and the process of the train moving from station k+1 to station k.

[0065] Let's take the example of train m running within a section as an illustration. Let... This indicates the train's departure time from station k-1. This indicates the train's arrival time at station k. The section of track from station k-1 to station k is divided into different sub-sections based on a fixed speed limit. Each sub-section has a starting point. and the finish line and the speed limit value of the sub-section. Starting from the starting position of each sub-interval Calculate the train's maximum traction speed. Based on the train's current speed. The running speed of the next state is calculated. .like > Then the running speed of the next state is the speed limit value of the corresponding sub-interval. Otherwise, the calculation speed is... Ultimately, the maximum traction speed is obtained. Starting from the beginning position of each sub-interval Calculate the train's maximum braking speed forward. Based on the train's current speed... The running speed of the previous state was calculated. .like > Then the running speed of the previous state is the speed limit value of the corresponding sub-interval. The maximum braking speed is finally obtained. .Compare and Choose the smaller value from the two, that is To obtain the speed value of the shortest running time .

[0066] The train will arrive at station K in the following time: , This is the sub-interval number. The train is at the end of the sub-interval. The maximum traction speed is: When calculating braking speed The net force acting on the train is: For traction force, This is the traction coefficient. The braking force coefficient, For braking force, As the basic resistance, The train's acceleration is due to the added resistance: , To ensure the quality of the train, the train speed will be within a certain range. .

[0067] Similarly, a process model can be established for train n traveling between stations K+1 and k. The difference is that the train will need to stop at the passing station as planned, hence its final speed... The final speed of the train .

[0068] The departure time of train m at station k-1 and the departure time of train n at station k+1 are usually determined. The objective function to be optimized is to always minimize the time. . This indicates the time it takes for the train to arrive at station K. However, this time must meet the requirement of allowing time for route establishment and signal opening for the passing train. ,and .

[0069] The goal is for the train to run smoothly, minimizing sudden accelerations and decelerations, i.e., minimizing the cumulative rate of change of acceleration in space. ,in and These are the lengths of the m-sub-section that the train will pass through, and the lengths of the n-sub-section that the train will pass through, respectively. This will allow train m to reach the end of its sub-section. acceleration, This will allow train n to be at the starting position of its sub-section. acceleration, By passing through the end position of train n in its sub-section acceleration, It is through the starting position of the train in the sub-section acceleration, This will allow train m to travel within its sub-section for a certain period of time. It is measured by the train's travel time within its sub-section.

[0070] Furthermore, the Q-network reinforcement learning environment can be set as the actual train operation environment, including static track data, train operation plans, the position, speed, and travel time of passing trains, and the input train's position, speed, and travel time. A Markov decision process can then be established. ,in, These represent the set of operating states that will allow the train set to operate, the output operating conditions of the train set, the state transition probability, and the reward. The magnitude of the reward value is determined by the total yield time and driving smoothness. The coordination process will be illustrated using the example of train yielding.

[0071] The weighted average of the two objective functions that minimize the cumulative rate of change of time and acceleration space is taken as follows: , The coefficients represent the two objectives. Including oncoming trains in the train operation status has the advantage that when calculating train curves and output conditions, considering the speed of the train that will pass and the distance to the passing point allows for coordinated control optimization of the two trains' operation processes from a scheduling perspective. The optimization method for allowing train m to operate within the section is as follows:

[0072] S21. Initialize the approximation network parameters Let the target network parameters for ;

[0073] S22 will affect the train's operating status. Depend on Sure, The length of the sub-interval. To allow the train to move faster, The distance from the train to the stopping point. To allow trains to pass through at their speeds, The weighted values ​​of the objective function are... The probability of randomly selecting working conditions and actions , with 1- The probability of choosing the current optimal state ;

[0074] S23, Execute working condition actions This will allow the train's operating environment to enter the next state. and reward value ;

[0075] S24. Randomly select several from the experience replay area. Update network parameters according to gradient descent algorithm ;

[0076] S25. Update network parameters after a specified number of steps. for If the number of iterations is reached, exit; otherwise, return to S21.

[0077] The train and the onboard intelligent agents of passing trains will continuously interact with the reinforcement learning environment according to the above optimization process. The onboard intelligent agents will calculate the train's operating conditions and actions under the current state based on the train's position, speed, and running time, so that the train and passing trains will tend to run with the shortest possible time and the smoothest operation.

[0078] The train operation environment calculates the train's acceleration sequence and arrival time at station k by having the train and its onboard agents perform actions based on the train's operating conditions. This generates new train operating states, reward values, and action value functions. The two agents select the action value function with the highest value and feed it back to the reinforcement learning environment. The agents use this closed-loop structure to continuously update the operating conditions, ultimately obtaining the optimal sequence of actions for each condition. The speed curve corresponding to this sequence is the optimal speed curve. For example... Figure 4 As shown.

[0079] Taking a local railway line as an example, the section from JK station to MP station is 6.6 km long with a travel time of 15 minutes, and the section from MP station to BF station is 10.8 km long with a travel time of 20 minutes. The JK->MP->BF direction is the loaded direction, and the BF->MP->JK direction is the empty direction. The locomotive is DF7G, pulling a 26-car train with a total mass of 2500 tons and a maximum traction force of 270 kN. The train's initial speed is 0 km / h, and its final speed is 70 km / h. When passing through MP station on the main line, the train's initial speed will be 0 km / h, and its final speed will be 0 km / h, stopping at the MP siding track. The test line data is shown in Tables 1 and 2.

[0080] Table 1. Track conditions for test section 1

[0081]

[0082] Table 2. Track conditions for test section 2

[0083]

[0084] The target network has a fixed update step count of 300, an experience replay data area capacity of 10,000, a single experience replay data size of 64, and a deep learning rate of... Discount factor Initial greed rate Ultimately, the greed rate Each state transition .

[0085] With the optimization goals of minimizing travel time and ensuring smooth driving, the system achieves coordinated optimization of train group scheduling and operation control. It collects train operation-related information and combines it with actual timetables, basic section conditions, and basic train parameters to generate feasible optimized speed-position curves. The ultimate goal of co-optimizing train travel time and speed curves is to minimize the weighted sum of train delay time and smoothness. The core strategy is to continuously adjust the train's travel time in each section and obtain the optimal speed curve for that travel time. The travel time and acceleration set are then fed back to the train travel time adjustment model, ultimately yielding an optimal travel time adjustment strategy.

[0086] Normally, without considering yielding to oncoming trains, the train generates a speed curve with the goal of maximizing speed. In the simulation, it is considered that an oncoming train might be unable to reach the target speed during its entry into the station for some reason, causing a delay in its arrival time. The train's speed is continuously adjusted in various sections so that its arrival time precisely meets the yielding interval, avoiding situations such as external stops. Table 3 describes the corresponding position and speed optimized by the train based on the position and speed information of the train that will yield.

[0087] Table 3. Speed ​​Strategy for Passing Trains Considering Oncoming Trains

[0088]

[0089] During the simulation, the position and speed of the oncoming train are considered as the speed curve generated by the passing train, as shown in the figure. Figure 5 As shown, the train slows down approximately two kilometers from the station, based on the passing train's position, and then accelerates again to pass through the station quickly after the passing train enters the station.

[0090] Therefore, by adopting the above methods, when a passing train approaches MP station, the speed of the passing train can be automatically controlled to slow down in advance based on the position and speed of the two trains. This will allow the passing train to accelerate into the station and improve the passing efficiency of the two trains. On the one hand, it can effectively avoid the passing train stopping outside the station while waiting for the passing train to enter the station. On the other hand, it reduces the frequent communication costs between the dispatcher and the drivers of the two trains during the passing time.

[0091] Compared with the prior art, the present invention has the following beneficial effects:

[0092] 1. Break down information barriers between the dispatching system and the train control system to effectively improve the efficiency of freight railway transportation.

[0093] 2. A freight railway train passing model was established, providing a detailed description of the operation process and planned arrival time of through trains and passing trains. This model can accurately describe the coordination process of through trains and passing trains as they travel to the passing station, laying the foundation for optimizing passing operations.

[0094] 3. Considering the speed and position of the opposing trains, the calculation method of the dual-depth Q network is improved. The advantage is that when calculating the train curve and output conditions, the speed of the train that will give way and the distance to the passing point are taken into account. From the perspective of scheduling plan, the operation process of the two trains can be coordinated and optimized at the same time.

[0095] Although the present invention has been described in detail through the preferred embodiments above, it should be understood that the above description should not be considered as a limitation of the present invention. Various modifications and substitutions to the present invention will be apparent to those skilled in the art after reading the above description. Therefore, the scope of protection of the present invention should be defined by the appended claims.

Claims

1. A method for coordinated freight railway scheduling and train control, characterized in that, Include: S1. Extract features from basic data, dispatching and command data, and train operation data of freight railway lines to construct a system model; S2. Acquire various data from the train control production system, combine them with the actual transportation production plan, simulate the train operation process and dispatching command strategy in the constructed system model, and deduce the optimization solution strategy of the system model in each set scenario in real time. S3. Evaluate the optimization solution strategy; S4. Apply the optimization strategies that meet the requirements in the evaluation to the train control production system to guide actual transportation production; Step S2 includes using the Q-network algorithm to set the operating environment for passing trains and yielding trains, simulating the yielding process of passing trains and yielding trains, and establishing a Markov decision process. The operating environment includes static line data, train operation plans for passing and yielding trains, the running position, speed, and running time of passing trains, and the running position, speed, and running time of yielding trains; the yielding operation process is represented as the running process of yielding trains from station k-1 to station k and the running process of passing trains from station k+1 to station k. The section from station k-1 to station k and the section from station k+1 to station k are divided into different sub-sections according to fixed speed limits. The maximum traction speed and maximum braking speed that will allow trains or passing trains to pass through each sub-section are calculated. Based on the speed at which the train will pass or the train is in its current state, the operating speed of the next state and the operating speed of the previous state are calculated. Under the premise of not exceeding the speed limit of the corresponding sub-section, the maximum traction speed is obtained based on the operating speed of the next state and the maximum braking speed is obtained based on the operating speed of the previous state. The smaller value between the maximum traction speed and the maximum braking speed is taken as the shortest running time speed value.

2. The scheduling and train control coordination method as described in claim 1, characterized in that, The system model in step S1 includes: a basic data model, a scheduling system model, a train control system model, and a train driving model.

3. The scheduling and train control coordination method as described in claim 1, characterized in that, The various data in step S2 include various emergencies during train operation.

4. The scheduling and train control coordination method as described in claim 1, characterized in that, The optimization strategy in step S2 is a train speed curve and scheduling adjustment plan applicable to the given scenario.

5. The scheduling and train control coordination method as described in claim 1, characterized in that, The train will arrive at station K in the following time: , To ensure the train's departure time at K-1 station, To ensure the train arrives at station K at the designated time. This will allow the train to travel in sub-sections of varying lengths. w This is the number of sub-sections that the train will pass through. This is the speed value that will allow the train to travel in the shortest possible time; The train's arrival time at station K is: , This refers to the departure time of the train at station k+1. For the arrival time of the train at station k, This represents the length of the sub-section through which the train passes. z This represents the number of sub-sections traversed by the train. This represents the speed value with the shortest travel time for the train.

6. The scheduling and train control coordination method as described in claim 5, characterized in that, The objective function for optimization is to always minimize the time: ,and , The time for trains to complete route procedures and open signals.

7. The scheduling and train control coordination method as described in claim 5, characterized in that, If a smooth process is desired, minimizing sudden accelerations and decelerations, the objective function for optimization is to minimize the cumulative rate of change in the acceleration space. It is the acceleration that will cause the train to reach the end of the sub-section. It will cause the train to accelerate at the starting point of the sub-section. It is measured by the acceleration of the train at the end of the sub-section. It is measured by the acceleration of the train at the starting point of the sub-section. This will affect the train's travel time within the sub-section. It is measured by the train's travel time within a sub-section.

8. The scheduling and train control coordination method as described in claim 5, characterized in that, Step S2, which uses the Q-network algorithm to optimize the train operation process, includes the following steps: S21. Initialize the approximation network parameters Let the target network parameters for ; S22 will affect the train's operating status. Depend on Sure, To allow the train to move faster, The distance from the train to the stopping point. To allow trains to pass through at their speeds, The weighted values ​​of the objective function are... The probability of randomly selecting working conditions and actions , with 1- The probability of choosing the current optimal state ; S23, Execute working condition actions This will allow the train's operating environment to enter the next state. and reward value ; S24. Randomly select several from the experience replay area. Update network parameters according to gradient descent algorithm ; S25. Update network parameters after a specified number of steps. for If the number of iterations is reached, exit; otherwise, return to S21.

9. A freight railway dispatching and train control coordination system, used to implement the dispatching and train control coordination method according to any one of claims 1 to 8, characterized in that, Include: The perception learning module is used to extract features from basic data, dispatching and command data, and train operation data of freight railway lines, and to build a computable and reconfigurable system model. The collaborative optimization module is used to acquire various types of data from the production system, combine them with the actual transportation production plan, simulate the train operation process and dispatching command strategy in the production environment, and deduce the optimization solution strategy of the system model in real time. The strategy evaluation module tests and evaluates the optimization strategies generated by the collaborative optimization module. If the strategy evaluation module pushes the optimization solution strategy that meets the predefined goals to the production system for execution.

10. The scheduling and train control coordination system as described in claim 9, characterized in that, The system model includes: a basic data model, a scheduling system model, a train control system model, and a train driving model.

11. The scheduling and train control coordination system as described in claim 9, characterized in that, The various types of data are derived from randomly generated unexpected events.

12. The scheduling and train control coordination system as described in claim 9, characterized in that, The optimization solutions include: train speed curves and scheduling adjustment plans.

Images