Train operation adjusting method and device in rain and snow mode

Abstract

The invention provides a train operation adjusting method and device in a rain and snow mode. The method comprises the steps that a target running gear of a next whole-line train in the rain and snow mode is obtained; wherein the rain and snow mode comprises a plurality of driving gears, and different driving gears correspond to different train operation data; predicting an operation plan of the train according to the target driving gear based on a train operation adjustment model to obtain a target operation plan so as to guide the train to operate in a rain and snow mode; the train operation adjustment model is constructed by taking the arrival time of each train, the departure time of each train and the running gear of each train as state variables and taking the minimum passenger riding waiting time and the minimum train operation cost as objective functions; the train operation cost comprises train energy cost and train maintenance cost. According to the method, the driving efficiency and safety of the whole-line train in the rain and snow mode are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of train operation control, and in particular to a method and device for adjusting train operation in a rain and snow mode. Background Art

[0002] At present, urban rail transit operation control systems usually include a rain and snow mode function, that is, the system will automatically enter the rain and snow mode under rainy and snowy conditions, limiting the train's traction, maximum operating speed and emergency braking rate to ensure the safety of train operation.

[0003] Speed ​​limit operation of trains will cause train delays and even lead to the failure of subsequent vehicle turnover plans, which will lead to the cancellation of planned trains. Therefore, in order to maintain the line capacity as much as possible under rainy and snowy weather conditions, train dispatchers need to respond quickly, make full use of line resources and vehicle resources, and take adjustment measures to organize driving. After the speed limit is lifted, train dispatchers also need to adjust the operation diagram in time so that subsequent trains can run according to plan. At the same time, train dispatchers also need to collect weather conditions on the line and take into account the issuance of dispatching orders. If a driver has unclear vision or finds that the train is idling and slipping, he needs to command the driver to slow down and organize other subsequent trains to run at the corresponding speed limit. In the above-mentioned handling process, a large amount of information interaction is usually required between dispatchers and equipment and personnel, and the adjustment of the train operation diagram is completed based on this.

[0004] Since rain and snow weather have a wide range of impacts on line operations, large scale and long duration, the train dispatcher often needs to make large-scale adjustments to the planned operation diagram when adjusting the train operation diagram. In addition, the train dispatcher needs to flexibly consider various adjustment methods during the adjustment process, which greatly increases the difficulty and work intensity of the train dispatcher when adjusting the train operation diagram. In the related technology, the train operation diagram remains unchanged after entering the rain and snow mode. If the train speed and emergency braking rate are reduced, the trains on the entire line will be delayed compared to the previous operation diagram, and may even fail to implement the subsequent vehicle turnover plan, leading to the cancellation of the planned trains. At this time, the train dispatcher needs to collect weather conditions on the line and take into account the issuance of dispatching orders. In addition, a large amount of information interaction is required between the dispatcher and the equipment and personnel, resulting in low train operation efficiency and possible safety problems caused by improper human operation. Summary of the invention

[0005] The present invention provides a method and device for adjusting train operation in a rain and snow mode, so as to solve the problem that in the prior art, in the process of maintaining the normal operation of the train in the rain and snow mode, the train dispatcher needs to collect relevant information and convey instructions on the line, and a large amount of information interaction is required between equipment and personnel, resulting in low train operation efficiency and low driving safety, thereby improving the driving efficiency and safety of the entire line of trains in the rain and snow mode.

[0006] The present invention provides a method for adjusting train operation in a rain and snow mode, comprising: Obtain the target driving gear of the trains on the entire line under the rain and snow mode; wherein the rain and snow mode includes multiple driving gears, and different driving gears correspond to different train operation data; the train operation data includes train traction, maximum operating speed and emergency braking rate; Based on the train operation adjustment model, the operation plan of the train is predicted according to the target driving gear to obtain a target operation plan to guide the train to operate in the rain and snow mode; wherein, the target operation plan includes at least two of the train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; the train operation adjustment model is constructed by taking the arrival time of each train, the departure time of each train, and the driving gear of each train as state variables, and minimizing the passenger waiting time and minimizing the train operating cost as the objective function; the train operating cost includes the train energy cost and the train maintenance cost.

[0007] According to a train operation adjustment method in a rain and snow mode provided by the present invention, the train operation adjustment model is constructed by the following steps: Collect arrival sample train operation data of multiple trains on the running line; The state variables composed of the sample train operation data are used as inputs of the target neural network, and the train operation loss is used as the loss function for iterative training to obtain the train operation adjustment model; The train operation loss is based on the waiting time, the train operation cost and Q Estimated value determined; Q The estimate represents the total expected future cost starting from the mth stage given the current state and current decision, where m is a positive integer greater than 1.

[0008] According to a train operation adjustment method in a rain and snow mode provided by the present invention, the target neural network is a deep reinforcement learning model, and the deep reinforcement learning model includes a first layer network and a second layer network; The state variables composed of the sample train operation data are used as the input of the target neural network, and the train operation loss is used as the loss function for iterative training to obtain the train operation adjustment model, which includes: The state variables are used as the input of the first layer network, and the decision actions are used as the output of the first layer network to construct the strategy network of the train operation adjustment model; the state variables and the decision actions output by the first layer network are used as the input of the second layer network, and the decision actions output by the first layer network are used as the output of the second layer network. Q The estimated value is used to construct a value network of the train operation adjustment model for the output of the second layer network; Based on the train operation loss according to the output of the second layer network Q The estimated value calculates the network loss, and updates the weight of the deep reinforcement learning model according to the network loss, and obtains the train operation adjustment model when the deep reinforcement learning model converges.

[0009] According to a train operation adjustment method in a rain and snow mode provided by the present invention, the train operation loss is expressed by the following formula: L = αC 1+ βC 2+ γ ( Q target − Q ( s , a )) 2 ; in, α and β is the weight coefficient, γ is the discount factor in Q learning, Q target is the target Q value, Q ( s , a ) is the Q Estimated value, s Represents the state quantity, a Indicates decision-making action. s and a Related to the current driving gear.

[0010] According to a train operation adjustment method under rain and snow mode provided by the present invention, the target operation plan includes train stop time, section operation time, train offline information, train number, train departure station, train terminal station and station information; the station information includes station name, station location and distance between stations; After obtaining the target operation plan, the method further includes: The train route is determined according to the train stop time, section running time, train offline information, train number, train departure station and train arrival station, and the station name, station location and distance between stations are matched with the train route information to obtain a train operation diagram.

[0011] The present invention also provides a train operation adjustment device in a rain and snow mode, comprising: The gear acquisition module is used to obtain the target driving gear of the trains on the entire line in the rain and snow mode; wherein the rain and snow mode includes multiple driving gears, and different driving gears correspond to different train operation data; the train operation data includes train traction, maximum operating speed and emergency braking rate; A prediction module is used to predict the train operation plan according to the target driving gear based on the train operation adjustment model to obtain a target operation plan to guide the train to operate in the rain and snow mode; wherein the target operation plan includes at least two of the train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; the train operation adjustment model is constructed by taking the arrival time of each train, the departure time of each train, and the driving gear of each train as state variables, and minimizing the passenger waiting time and minimizing the train operating cost as the objective function; the train operating cost includes the train energy cost and the train maintenance cost.

[0012] According to a train operation adjustment device in a rain and snow mode provided by the present invention, the target operation plan includes train stop time, section operation time, train offline information, train number, train departure station, train terminal station and station information; the station information includes station name, station location and distance between stations; The device also includes: The operation diagram generation module is used to determine the train route according to the train stop time, section operation time, train offline information, train number, train departure station and train terminal after obtaining the target operation plan, and match the station name, station location and distance between stations with the train route information to obtain the train operation diagram.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a train operation adjustment method in the rain and snow mode as described above is implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the train operation adjustment method in the rain and snow mode as described in any one of the above is implemented.

[0015] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for adjusting train operation in the rain and snow mode as described above is implemented.

[0016] The train operation adjustment method and device under the rain and snow mode provided by the present invention obtain the target driving gear of the trains of the whole line under the rain and snow mode, and then based on the arrival time of each train, the departure time of each train, and the driving gear of each train as state variables, minimize the waiting time of passengers and minimize the operating cost of the train as the objective function to construct a train operation adjustment model, and predict the operation plan of the train according to the target driving gear to guide the operation of the train under the rain and snow mode, thereby improving the driving efficiency and safety of the trains of the whole line under the rain and snow mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is one of the flow charts of the train operation adjustment method in the rain and snow mode provided by the present invention.

[0019] Figure 2 This is the second flow chart of the train operation adjustment method in the rain and snow mode provided by the present invention.

[0020] Figure 3 It is a structural schematic diagram of the train operation adjustment device in the rain and snow mode provided by the present invention.

[0021] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] Combine the following Figure 1-Figure 3 The present invention describes a method and device for adjusting train operation in a rain and snow mode.

[0024] Figure 1 FIG. 1 is one of the flow charts of the train operation adjustment method in the rain and snow mode provided by the present invention, such as Figure 1 As shown, the train operation adjustment method in the rain and snow mode includes the following steps: Step 110: Obtain the target driving gear of all trains on the entire line in the rain and snow mode; wherein the rain and snow mode includes multiple driving gears, and different driving gears correspond to different train operation data; the train operation data includes train traction, maximum operating speed and emergency braking rate.

[0025] In this step, all trains on the line include high-speed trains, intercity trains or subways on the designated train operation line.

[0026] In this step, the rain and snow mode of the train is a train operation mode designed to cope with severe weather conditions. In this mode, the train will automatically or manually adjust its operating parameters according to weather conditions to ensure driving safety.

[0027] In this step, the driving gear includes multiple levels, each level corresponding to a different train traction, maximum operating speed and emergency braking rate.

[0028] In this embodiment, the driving gear position includes a low gear position (or a safety gear position), a middle gear position or a high gear position.

[0029] Specifically, when the train is in low gear, the traction of the train is reduced to reduce the friction between the wheels and the track and reduce the risk of slipping; the maximum operating speed of the train is reduced. For example, in rainy and snowy weather, the train may run at 40 km / h or less; the emergency braking rate of the train is increased to shorten the braking distance and improve safety in emergency situations; when the train is in medium gear, the traction of the train is moderate, and the maximum operating speed is higher than that in low gear, which can be dynamically adjusted according to weather conditions and track conditions, and the emergency braking rate is maintained at a reasonable level to ensure that the train can stop quickly when needed; when the train is in high gear, the traction of the train is close to the level in normal weather; the maximum operating speed is maintained at or slightly lower than the maximum speed in normal weather to ensure safe driving in bad weather, and the emergency braking rate is similar to that in normal weather, but may be slightly adjusted due to weather conditions.

[0030] Step 120: predict the train operation plan according to the target driving gear based on the train operation adjustment model to obtain the target operation plan to guide the train to operate in the rain and snow mode; wherein the target operation plan includes at least two of the train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; the train operation adjustment model is constructed by taking the arrival time of each train, the departure time of each train, and the driving gear of each train as state variables, and minimizing the passenger waiting time and minimizing the train operating cost as the objective function; the train operating cost includes the train energy cost and the train maintenance cost.

[0031] In this step, the passenger waiting cost is quantitatively expressed by the classical queuing theory to obtain the passenger waiting time, and the train energy cost and train maintenance cost are quantified to obtain C2 In this step, the objective function is designed based on the following two factors: (1) From the perspective of passengers, the overall waiting cost must be minimized; (2) From the perspective of the number of trains on the entire line, that is, after entering the rain and snow mode, as few trains as possible need to be retained in the depot so that there are still enough trains on the entire line to undertake transportation tasks.

[0032] In this step, the train operation adjustment model is a mathematical model or a dynamic operation diagram editing agent, the input is the target gear of the train in the rain and snow mode, and the output is the lowest-cost train transportation strategy corresponding to the target gear, that is, the target operation plan.

[0033] In this embodiment, the train operation adjustment model realizes a series of parameter adjustments in the strategy function through a large number of scheduling adjustment learning in the actual operating environment, and the strategy function is fitted by a neural network, thereby obtaining mapping rules for train operation strategies after different vehicles in different areas enter different levels of rain and snow modes. The mapping rules can realize the operation diagram after entering the rain and snow mode under the optimal condition of the objective function.

[0034] Specifically, when the train on the line slips multiple times, the control system prompts the train dispatcher to enter the rain and snow mode. At this time, the traction force, maximum operating speed and emergency braking rate of the trains on the entire line can be obtained according to the rain and snow mode gear set by the train dispatcher (different gears correspond to different restrictions). The subsequent operation plan of the train (stop time, section operation time, whether to go offline) can be inferred based on the above train operation data through the trained train operation adjustment model.

[0035] In this embodiment, if there are currently M trains operating the entire line, after entering the rain and snow mode, the entire line needs to be reduced to N trains to undertake operating tasks due to the reduction in train running speed and the increase in traction braking time.

[0036] The train operation adjustment method under the rain and snow mode provided by the embodiment of the present invention obtains the target driving gear of the trains of the whole line under the rain and snow mode, and then based on the arrival time of each train, the departure time of each train, and the driving gear of each train as the state variables, minimizes the waiting time of passengers and minimizes the operating cost of the train as the objective function to construct a train operation adjustment model, and predicts the operation plan of the train according to the target driving gear to guide the operation of the train under the rain and snow mode, thereby improving the driving efficiency and safety of the trains of the whole line under the rain and snow mode.

[0037] In some embodiments, the train operation adjustment model is constructed by the following steps: collecting arrival sample train operation data of multiple trains on the running line; using the state variables composed of the sample train operation data as the input of the target neural network, and iteratively training with the train operation loss as the loss function to obtain the train operation adjustment model; wherein the train operation loss is based on the waiting time, the train operation cost and Q Estimates are determined; Q The estimate represents the total expected future cost starting from the mth stage given the current state and current decision, where m is a positive integer greater than 1.

[0038] In this embodiment, the target neural network includes but is not limited to a deep reinforcement learning model.

[0039] In this embodiment, setting For the J Train Arrival i Standing time, For the J Train from i Station departure time, For the J Different rain and snow mode levels of the train correspond to different maximum speed limits and train traction braking rates; the action decision made by the vehicle controller according to the instructions of the train dispatcher can be expressed as { , , }.

[0040] In this embodiment, the feasible solution set for each stage decision is determined according to the train operation time status. , then the decision actions made by the agent when the state changes according to the rain and snow pattern Should meet .

[0041] Specifically, the target neural network is a deep reinforcement learning model, which includes a first-layer network and a second-layer network; the state variables composed of sample train operation data are used as the input of the target neural network, and the train operation loss is used as the loss function for iterative training, so as to obtain a train operation adjustment model, which includes: using the state variables as the input of the first-layer network and the decision action as the output of the first-layer network to construct a strategy network for the train operation adjustment model; using the state variables and the decision action output by the first-layer network as the input of the second-layer network, and using the decision action output by the first-layer network as the output of the second-layer network. Q The estimated value is used to construct the value network of the train operation adjustment model based on the output of the second layer network; Q The estimated value calculates the network loss, and updates the weight of the deep reinforcement learning model according to the network loss. When the deep reinforcement learning model converges, the train operation adjustment model is obtained.

[0042] In this embodiment, a two-layer neural network is constructed to solve the problem of the agent falling into the curse of dimensionality, and the state variables { , , }Input into the first layer of the neural network to generate a corresponding (embedded) decision action, whose goal is to minimize the total cost of the operation; input the decision action into the second layer of the neural network to obtain the Q estimate.

[0043] In this embodiment, the Q estimation value can be calculated by the following formula: ; in, It is k The state variables of the phase, It is k Decision making at this stage, is a discount factor that weighs the importance of current costs against future costs.

[0044] In this embodiment, during the deep reinforcement learning model training process, the two-layer neural network is first initialized, and the parameters in the strategy function are set to default values, that is, the default strategy is set; then the scheduling decision information is extracted , generate state parameters; in any state, according to the strategy function, apply the corresponding action to the current environment state, and generate a new network environment state, that is, state transition; then according to the change of state quantity, calculate the corresponding reward function value, and evaluate the action taken to indicate the implementation effect of the selected action; finally, according to the evaluation result, modify the parameters in the strategy function and adjust the mapping criteria from state to action; repeat the above training process until the lowest cost train transportation strategy is selected.

[0045] In this embodiment, the train operation loss is expressed by the following formula: L = αC 1+ βC 2+ γ ( Q target − Q ( s , a )) 2 ; in, α and β is the weight coefficient, γ is the discount factor in Q learning, Q target is the target Q value (which can be calculated using the Bellman equation or the Monte Carlo method), Q ( s , a )forQ Estimated value, s Represents the state quantity, a Indicates decision-making action. s and a Related to the current driving gear; the weights of the neural network can be updated through optimization algorithms such as gradient descent.

[0046] The train operation adjustment method under the rain and snow mode provided by the embodiment of the present invention constructs a neural network decision-making model, constructs a dynamic operation diagram editing intelligent body with the minimum passenger waiting time and the maximum number of trains on the line as the objective function, and realizes a series of parameter adjustments in the strategy function through a large number of scheduling adjustment learning in the actual operation environment. The strategy function is fitted by the neural network, thereby obtaining the mapping rules for the train operation strategy after different vehicles in different areas enter different levels of rain and snow modes, so as to obtain the optimal operation plan of the train under the rain and snow mode.

[0047] In some embodiments, the target operation plan includes train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; the station information includes station name, station location and distance between stations; after obtaining the target operation plan, the method also includes: determining the train route according to the train stop time, section operation time, train offline information, train number, train departure station and train terminal, matching the station name, station location and distance between stations with the train route information to obtain a train operation diagram.

[0048] In this embodiment, the target operation plan can be obtained from the official website of the railway department, official applications, social media platforms or news reports; station information includes the name, location, line, type of trains that stop at the station, etc. This information can be obtained through the official website of the railway department, map applications or field visits.

[0049] In this embodiment, the relevant program is designed through the following steps to realize automatic drawing of the train operation diagram: (1) Determine the time axis and station axis; determine a time axis in the drawing software to represent the running time of the train; (2) Determine the station axis; indicate the stations that the train passes through; draw the train route. According to the target operation plan, mark the departure time, arrival time and stops of each train on the time axis and station axis; use lines to connect these marked points to form the train route; use different colors or marks to distinguish different train numbers and types.

[0050] (3) Add detailed information: Add necessary detailed information on the train timetable, such as train number, running direction, stop time, etc.; you can also add detailed information on the station on the timetable, such as the station's facilities, service hours, etc.

[0051] The train operation adjustment method under the rain and snow mode provided in the embodiment of the present invention obtains a train operation diagram by matching the target operation plan with the train route information. Through the analysis and research of the train operation diagram, the operation status and existing problems of the railway network can be understood, and data support can be provided for the planning of railway development to ensure that the trains running on the entire line maintain normal and safe operation under the rain and snow mode.

[0052] Figure 2 This is a second flow chart of the train operation adjustment method in the rain and snow mode provided by the present invention. Figure 2 In the illustrated embodiment, a method for adjusting train operation under a rain and snow mode is implemented by the following steps: (1) establishing a neural network model, which includes determining environmental variables of the process under the rain and snow mode and determining an objective function; (2) constructing a deep reinforcement learning architecture, specifically calculating different strategies to adjust the train operation status and cost based on inputs such as the rain and snow mode level and train data, and making a final scheduling decision based on the prediction results; (3) repeating the above step (2) to train the model; (4) completing the fitting of the strategy function by the neural network, thereby obtaining a mapping rule for the train operation strategy after different vehicles in different areas enter different levels of rain and snow modes, and implementing the operation diagram after entering the rain and snow mode under the optimal objective function according to the mapping rule.

[0053] The train operation adjustment device in the rain and snow mode provided by the present invention is described below. The train operation adjustment device in the rain and snow mode described below and the train operation adjustment method in the rain and snow mode described above can be referenced to each other.

[0054] Figure 3 : is a schematic diagram of the structure of the train operation adjustment device in the rain and snow mode provided by the present invention, such as Figure 3 As shown, the train operation adjustment device in the rain and snow mode includes: a gear acquisition module 310 and a prediction module 320.

[0055] The gear acquisition module 310 is used to obtain the target driving gear of the trains on the entire line in the rain and snow mode; wherein the rain and snow mode includes multiple driving gears, and different driving gears correspond to different train operation data; the train operation data includes train traction, maximum operating speed and emergency braking rate; The prediction module 320 is used to predict the train operation plan according to the target driving gear based on the train operation adjustment model, and obtain the target operation plan to guide the train to operate in the rain and snow mode; wherein the target operation plan includes at least two of the train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; the train operation adjustment model is constructed by taking the arrival time of each train, the departure time of each train, and the driving gear of each train as state variables, and minimizing the passenger waiting time and minimizing the train operating cost as the objective function; the train operating cost includes the train energy cost and the train maintenance cost.

[0056] The train operation adjustment device under the rain and snow mode provided by the embodiment of the present invention obtains the target driving gear of the trains of the whole line under the rain and snow mode, and then based on the arrival time of each train, the departure time of each train, and the driving gear of each train as state variables, minimizes the waiting time of passengers and minimizes the operating cost of the train as the objective function to obtain a train operation adjustment model, and predicts the operation plan of the train according to the target driving gear to guide the operation of the train under the rain and snow mode, thereby improving the driving efficiency and safety of the trains of the whole line under the rain and snow mode.

[0057] In some embodiments, the target operation plan includes train stop time, section operation time, train offline information, train number, train departure station, train destination station and station information; station information includes station name, station location and distance between stations; the train operation adjustment device under the rain and snow mode also includes: an operation diagram generation module 330.

[0058] The operation diagram generation module 330 is used to determine the train routes according to the train stop time, section operation time, train offline information, train number, train departure station and train arrival station after obtaining the target operation plan, and match the station name, station location and distance between stations with the train route information to obtain the train operation diagram.

[0059] The train operation adjustment device in the rain and snow mode provided in the embodiment of the present invention obtains a train operation diagram by matching the target operation plan with the train route information. Through the analysis and research of the train operation diagram, the operation status and existing problems of the railway network can be understood, and data support can be provided for the planning of railway development to ensure that the trains running on the entire line maintain normal and safe operation in the rain and snow mode.

[0060] Figure 4 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 4As shown, the electronic device may include: a processor (processor) 410 , a communication interface (Communications Interface) 420 , a memory (memory) 430 and a communication bus 440 , wherein the processor 410 , the communication interface 420 , and the memory 430 communicate with each other through the communication bus 440 . The processor 410 can call the logic instructions in the memory 430 to execute the train operation adjustment method in the rain and snow mode, the method comprising: obtaining the target driving gear of the trains on the entire line in the rain and snow mode; wherein the rain and snow mode includes multiple driving gears, and different driving gears correspond to different train operation data; the train operation data includes train traction, maximum operating speed and emergency braking rate; based on the train operation adjustment model, the train operation plan is predicted according to the target driving gear to obtain a target operation plan to guide the train to operate in the rain and snow mode; wherein the target operation plan includes at least two of the train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; the train operation adjustment model is constructed by taking the arrival time of each train, the departure time of each train, and the driving gear of each train as state variables, and minimizing the passenger waiting time and minimizing the train operation cost as the objective function; the train operation cost includes the train energy cost and the train maintenance cost.

[0061] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0062] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the train operation adjustment method in the rain and snow mode provided by the above methods, the method comprising: obtaining the target driving gear of the entire line train in the rain and snow mode; wherein the rain and snow mode includes multiple driving gears, and different driving gears correspond to different train operation data; the train operation data includes train traction, maximum operating speed and emergency braking rate; based on the train operation adjustment model, the train operation plan is predicted according to the target driving gear to obtain a target operation plan to guide the train to operate in the rain and snow mode; wherein the target operation plan includes at least two of the train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; the train operation adjustment model is constructed by taking the arrival time of each train, the departure time of each train, and the driving gear of each train as state variables, and minimizing the passenger waiting time and minimizing the train operation cost as the objective function; the train operation cost includes the train energy cost and the train maintenance cost.

[0063] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the train operation adjustment method in the rain and snow mode provided by the above-mentioned methods, the method comprising: obtaining a target driving gear of all trains on the entire line in the rain and snow mode; wherein the rain and snow mode includes multiple driving gears, and different driving gears correspond to different train operation data; the train operation data includes train traction, maximum operating speed and emergency braking rate; based on the train operation adjustment model, the operation plan of the train is predicted according to the target driving gear to obtain a target operation plan to guide the train to operate in the rain and snow mode; wherein the target operation plan includes at least two of the train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; the train operation adjustment model is constructed by taking the arrival time of each train, the departure time of each train and the driving gear of each train as state variables, and minimizing the passenger waiting time and minimizing the train operation cost as the objective function; the train operation cost includes the train energy cost and the train maintenance cost.

[0064] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0065] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for adjusting train operation in rain and snow mode, characterized in that: include: Obtain the target driving gear of the trains on the entire line under the rain and snow mode; wherein the rain and snow mode includes multiple driving gears, and different driving gears correspond to different train operation data; the train operation data includes train traction, maximum operating speed and emergency braking rate; Based on the train operation adjustment model, the operation plan of the train is predicted according to the target driving gear to obtain a target operation plan to guide the train to operate in the rain and snow mode; wherein, the target operation plan includes at least two of the train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; the train operation adjustment model is constructed by taking the arrival time of each train, the departure time of each train, and the driving gear of each train as state variables, and minimizing the passenger waiting time and minimizing the train operating cost as the objective function; the train operating cost includes the train energy cost and the train maintenance cost.

2. The train operation adjustment method in rain and snow mode according to claim 1, characterized in that: The train operation adjustment model is constructed by the following steps: Collect arrival sample train operation data of multiple trains on the running line; The state variables composed of the sample train operation data are used as inputs of the target neural network, and the train operation loss is used as the loss function for iterative training to obtain the train operation adjustment model; The train operation loss is based on the waiting time, the train operation cost and Q Estimated value determined; Q The estimate represents the total expected future cost starting from the mth stage given the current state and current decision, where m is a positive integer greater than 1.

3. The train operation adjustment method in rain and snow mode according to claim 2, characterized in that: The target neural network is a deep reinforcement learning model, and the deep reinforcement learning model includes a first layer network and a second layer network; The state variables composed of the sample train operation data are used as the input of the target neural network, and the train operation loss is used as the loss function for iterative training to obtain the train operation adjustment model, which includes: The state variables are used as the input of the first layer network, and the decision actions are used as the output of the first layer network to construct the strategy network of the train operation adjustment model; the state variables and the decision actions output by the first layer network are used as the input of the second layer network, and the decision actions output by the first layer network are used as the output of the second layer network. Q The estimated value is used to construct a value network of the train operation adjustment model for the output of the second layer network; Based on the train operation loss according to the output of the second layer network Q The estimated value calculates the network loss, and updates the weight of the deep reinforcement learning model according to the network loss, and obtains the train operation adjustment model when the deep reinforcement learning model converges.

4. The train operation adjustment method in rain and snow mode according to claim 2 or 3, characterized in that: The train operation loss is expressed by the following formula: L = αC 1+ βC 2+ γ ( Q target − Q ( s , a )) 2 ； in, α and β is the weight coefficient, γ is the discount factor in Q learning, Q target is the target Q value, Q ( s , a ) is the Q Estimated value, s Represents the state quantity, a Indicates decision-making action. s and a Related to the current driving gear.

5. The train operation adjustment method in rain and snow mode according to claim 1, characterized in that: The target operation plan includes train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; The station information includes station name, station location and distance between stations; After obtaining the target operation plan, the method further includes: The train route is determined according to the train stop time, section running time, train offline information, train number, train departure station and train arrival station, and the station name, station location and distance between stations are matched with the train route information to obtain a train operation diagram.

6. A train operation adjustment device in rain and snow mode, characterized in that: include: The gear acquisition module is used to obtain the target driving gear of the trains on the entire line in the rain and snow mode; wherein the rain and snow mode includes multiple driving gears, and different driving gears correspond to different train operation data; the train operation data includes train traction, maximum operating speed and emergency braking rate; A prediction module is used to predict the train operation plan according to the target driving gear based on the train operation adjustment model to obtain a target operation plan to guide the train to operate in the rain and snow mode; wherein the target operation plan includes at least two of the train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; the train operation adjustment model is constructed by taking the arrival time of each train, the departure time of each train, and the driving gear of each train as state variables, and minimizing the passenger waiting time and minimizing the train operating cost as the objective function; the train operating cost includes the train energy cost and the train maintenance cost.

7. The train operation adjustment device in rain and snow mode according to claim 6, characterized in that: The target operation plan includes train stop time, section operation time, train offline information, train number, train departure station, train terminal and station information; The station information includes station name, station location and distance between stations; The device also includes: The operation diagram generation module is used to determine the train route according to the train stop time, section operation time, train offline information, train number, train departure station and train terminal after obtaining the target operation plan, and match the station name, station location and distance between stations with the train route information to obtain the train operation diagram.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the train operation adjustment method in the rain and snow mode as described in any one of claims 1 to 5 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the train operation adjustment method in the rain and snow mode as described in any one of claims 1 to 5 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the train operation adjustment method in the rain and snow mode as described in any one of claims 1 to 5 is implemented.

Images