Non-self-powered shunting plan automatic generation method, medium and equipment

Abstract

The invention relates to a non-self-powered shunting plan automatic generation method, medium and equipment, and the method comprises the steps: firstly, collecting multi-source data, and synchronizing a train maintenance plan, so as to screen out trains with shunting demands; secondly, vehicles needing non-self-powered shunting are judged according to requirements, the optimal paths of the corresponding vehicles needing non-self-powered shunting are automatically screened, a time sequence table is established, and route conflict detection is conducted; and finally generating a standardized shunting plan. The method provided by the invention can automatically identify shunting demands, optimize path planning in multiple dimensions, intelligently detect and adjust conflicts, realize full-process automatic generation of shunting plans, improve scheduling efficiency and reliability, and have obvious advantages.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of rail transit shunting technology, in particular to a non-self-powered shunting plan automatic generation method, medium and equipment. BACKGROUND

[0002] In the daily operation of the rail transit train depot, shunting operation is one of the key links to ensure the efficient turnover of trains, maintenance and repair, and the orderly operation of operation scheduling. According to whether the train has self-walking ability, the shunting operation can be divided into two categories: self-powered shunting and non-self-powered shunting. Non-self-powered shunting mainly faces trains that cannot run independently, such as fault trains, trains waiting for maintenance, etc., which need to rely on external power equipment such as shunting locomotives for traction or pushing to complete the operation. The process of this type of operation is more complex, and the correlation between the links is strong, so the accuracy and safety of the operation plan are required.

[0003] At present, the non-self-powered shunting operation plan of the train depot mainly depends on the experience of the dispatch personnel to manually prepare, which has obvious shortcomings: on the one hand, it is difficult to identify the shunting demand, and the dispatch personnel need to spend a lot of time integrating information in a complex on-site operation environment, making it difficult to quickly and accurately respond to the shunting demand; on the other hand, the planning efficiency is low, and path planning, time allocation, and conflict checking are all completed manually, which is particularly time-consuming when multiple tasks are running in parallel, and is prone to cause unreasonable plans or safety hazards due to human negligence.

[0004] And in the prior art, although some research has focused on the automatic generation of shunting plans, most of them focus on self-powered train scheduling, and do not fully consider the special requirements of non-self-powered trains relying on external equipment for traction, complex operation process, high coupling degree of links, strict safety constraints, etc., so it is difficult to be directly applicable to the non-self-powered shunting scene.

[0005] Therefore, the present application proposes a non-self-powered shunting plan automatic generation method, medium and equipment, which aims to effectively address the above technical challenges.

[0006] It can be understood that the above statements are only background technology related to the present application and do not necessarily constitute prior art. SUMMARY

[0007] The purpose of the present application is to provide a non-self-powered shunting plan automatic generation method, medium and equipment, which automatically identifies the shunting demand, optimizes the path planning in multiple dimensions, intelligently detects and adjusts the conflicts, and realizes the automatic generation of the shunting plan in the whole process, thereby improving the scheduling efficiency and reliability.

[0008] In order to achieve the above purpose, the first aspect of the present application provides a non-self-powered shunting plan automatic generation method, comprising the following steps: Step S1, performing multi-source data collection, synchronizing train maintenance plan, to update the train maintenance plan database; wherein the multi-source data collection includes: real-time collection of train state data, target maintenance track information and engineering vehicle information; Step S2, based on the train maintenance plan database, train state data and target maintenance track information, constructing a shunting demand preliminary judgment module to screen out trains with shunting demand and output a preliminary judgment signal; Step S3, according to the preset non-self-powered shunting type judgment requirement, identifying non-self-powered shunting demand from the output signal of the shunting demand preliminary judgment module, and screening out a non-self-powered shunting demand list; Step S4, for the non-self-powered shunting demand list, for a train with non-self-powered shunting demand, screening the optimal path, including extracting basic parameters, traversing routes to generate a set of selectable paths, and selecting the optimal path through a multi-objective optimization system; Step S5, based on the optimal path, automatically creating a timing table to assign start and end times for the shunting task; Step S6, performing route conflict detection on the optimal path and timing table, and if a conflict is detected, performing adaptive adjustment until a conflict-free path is obtained; Step S7, integrating the conflict-free path and timing table to generate a standardized shunting plan; Step S8, traversing the non-self-powered shunting demand list obtained in step S3, and for each train with non-self-powered shunting demand, sequentially executing steps S4-S7.

[0009] Preferably, in step S1, the multi-source data collection includes: synchronizing train maintenance plan from enterprise management system to update train maintenance plan database, the train maintenance plan includes but is not limited to: train number to be maintained, planned maintenance type and maintenance start time; real-time collection of train state data, including but not limited to: train current parking track, traction system state and pantograph state; collecting target maintenance track information, including but not limited to: target maintenance track, contact net presence state and contact net state; collecting engineering vehicle information, including but not limited to: engineering vehicle number and engineering vehicle position.

[0010] Preferably, in step S2, the shunting demand preliminary judgment module extracts the train number to be maintained from the train maintenance plan database at regular intervals, and combines the train current parking track and target maintenance track information to screen out trains whose train current parking track and target maintenance track are inconsistent and are in the state of being maintained, that is, trains that need to be shunted, and output a preliminary judgment signal of trains that need to be shunted.

[0011] Preferably, in step S3, the non-self-powered shunting type judgment requires at least one of the following judgment conditions to be met, specifically: and / or, the current position of the train is in a catenary power-off section; and / or, the target maintenance track is without catenary; and / or, the target maintenance track is in a catenary power-off section; wherein, the non-self-powered shunting demand is identified from the preliminary judgment signal output by the shunting demand preliminary judgment module, and a non-self-powered shunting demand list is screened out; and when at least one of the judgment conditions is met, it can be judged as a non-self-powered shunting type.

[0012] Preferably, in step S4, the basic parameters of a certain train to be shunted are extracted from the non-self-powered shunting demand list obtained in step S3, specifically including: train number, current parking track, target maintenance track, maintenance start time, engineering vehicle number and engineering vehicle position; and at the same time, a multi-hook task track link composed of "engineering vehicle out of depot-pulling out line-current parking track-pulling out line-target maintenance track-pulling out line-engineering vehicle back to depot" is constructed by traversing the route.

[0013] Preferably, the multi-objective optimization system adopts a weighted scoring method, including the following evaluation indexes: switch action quantity, weight 50%, converted to 0-50 points in inverse proportion to the number; route quantity, weight 30%, converted to 0-30 points in inverse proportion to the number; path length, weight 20%, converted to 0-20 points in inverse proportion to the length; wherein, the scores of the three indexes of each selectable path in the selectable path set are weighted and summed, and the path with the highest total score is selected as the optimal path.

[0014] Preferably, in step S5, the time calculation rules of the time sequence table include: the running time of the engineering vehicle during shunting needs to be considered; the coupling and uncoupling time between the train to be shunted and the engineering vehicle needs to be considered; the completion time of the last hook task in the multi-hook task is set to 30-60 minutes before the start time of the maintenance plan.

[0015] Preferably, the time sequence table automatically created adopts a time reverse method to generate.

[0016] Preferably, in step S6, the route conflict detection is realized by a conflict detection engine; wherein, the conflict detection engine detects whether there is other operation occupation in the same track at the same time based on space-time overlap analysis.

[0017] Preferably, if there is no other operation occupation in the same track at the same time, it is proved that there is no conflict, and step S7 is turned to continue running; if there is other operation occupation in the same track at the same time, it is proved that a conflict is detected, and adaptive adjustment is needed until a conflict-free path is obtained.

[0018] Preferably, the adaptive adjustment method adopts a two-level adjustment strategy, including the following steps: First, path optimization adjustment is performed, and step S4 is called again to filter paths, select the second-best path from the set of optional paths and perform conflict detection; and only when all optional paths in the path optimization adjustment have conflicts, the timing table is adjusted, and the process returns to step S5. Under the premise of satisfying the time calculation rules of the timing table, the shunting start time is fluctuated by 5-15 minutes before and after, and the timing table is regenerated before step S6 to perform conflict detection.

[0019] Preferably, in step S7, the generated standardized shunting plan includes the following information: optimal route or adjusted route, timing table, shunted train number, engineering car number, current parking track, target maintenance track, and special safety reminders.

[0020] Preferably, the specific safety tips include: overhead contact line operation specifications and speed limits for traction of faulty trains.

[0021] A second aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for automatically generating non-self-powered shunting plans.

[0022] A third aspect of the present invention provides an electronic device including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the method for automatically generating non-self-powered shunting plans.

[0023] In summary, compared with the prior art, the method, medium, and equipment for automatically generating non-self-powered shunting plans provided by the present invention have at least the following beneficial effects: (1) By constructing a two-stage identification logic and a multi-source data fusion mechanism for the initial judgment of shunting demand and the judgment of shunting demand with non-self-powered power, the accurate and automated identification of shunting demand with non-self-powered power is realized, which effectively overcomes the defects of low efficiency and easy omission of traditional manual judgment and provides a reliable input basis for plan generation. (2) A path planning algorithm with multi-objective weighted optimization as the core was designed to comprehensively optimize key indicators such as the number of turnout actions, the number of routes and the path length, which significantly improved the work efficiency, reduced equipment wear and energy consumption, and ensured the standardization and consistency of the planning scheme. (3) An adaptive conflict resolution strategy of "prioritizing path adjustment and supplementing with time fine-tuning" was proposed, which can efficiently solve the complex conflict problems commonly encountered in non-self-powered shunting under the premise of ensuring maintenance timeliness and operational safety, thereby improving the feasibility and reliability of the plan. (4) An end-to-end automated generation system from demand identification to plan output has been constructed, which completely eliminates the reliance on manual intervention in key links, greatly improves the efficiency and standardization of plan generation, lowers the threshold for operation, and realizes the traceability of decision-making throughout the process. Attached Figure Description

[0024] Figure 1 This is a flowchart of the method for automatically generating non-self-powered shunting plans in this invention. Detailed Implementation

[0025] The present invention will be further described below with reference to the accompanying drawings and by providing a detailed description of a preferred embodiment.

[0026] It should be noted that the accompanying drawings are in a very simplified form and use non-precise proportions. They are only used to facilitate and clarify the purpose of illustrating the embodiments of the present invention, and are not intended to limit the implementation conditions of the present invention. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportional relationship, or adjustments to the size should still fall within the scope of the technical content disclosed in the present invention, provided that they do not affect the effects and objectives that the present invention can produce.

[0027] It should be noted that, in this invention, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only the expressly listed elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0028] like Figure 1 As shown, this invention proposes a method for automatically generating non-self-powered shunting plans, specifically including the following steps: Step S1: Perform multi-source data acquisition and synchronize the train maintenance plan to update the train maintenance plan database; wherein, the multi-source data acquisition includes: real-time acquisition of train status data, target maintenance track information and engineering vehicle information; Step S2: Based on the train maintenance plan database, train status data and target maintenance track information, construct a shunting demand preliminary judgment module to screen out trains with shunting needs and output a preliminary judgment signal. Step S3: Based on the preset non-self-powered shunting type determination requirements, identify non-self-powered shunting requirements from the output signal of the shunting requirement preliminary judgment module, and filter out the list of non-self-powered shunting requirements. Step S4: For the list of non-self-powered shunting needs, for a train with non-self-powered shunting needs, select the optimal path, including extracting basic parameters, traversing routes to generate a set of optional paths, and evaluating and selecting the optimal path through a multi-objective optimization system. Step S5: Based on the optimal path, automatically create a timing table and assign start and end times to the shunting task; Step S6: Perform route conflict detection on the optimal path and timing table. If a conflict is detected, perform adaptive adjustment until a conflict-free path is obtained. Step S7: Integrate conflict-free routes and timing tables to generate a standardized shunting plan; Step S8: Traverse the list of non-self-powered shunting requirements obtained in step S3. For each train with non-self-powered shunting requirements, execute steps S4-S7 in sequence.

[0029] Furthermore, in step S1, the multi-source data acquisition includes: The train maintenance plan database is updated by synchronizing the train maintenance plan from the enterprise management system. The train maintenance plan includes, but is not limited to: the train number to be maintained, the planned maintenance type, and the maintenance start time. Real-time collection of train status data, including but not limited to: the current track where the train is parked (i.e., the current position of the train), the status of the traction system and the status of the pantograph; Collect information on the target maintenance track, including but not limited to: the target maintenance track, the existence and status of the overhead contact line (i.e., whether the overhead contact line exists) and the status of the overhead contact line.

[0030] Collect information on engineering vehicles, including but not limited to: vehicle number and vehicle location.

[0031] Furthermore, in step S2, the shunting demand preliminary judgment module periodically extracts the train numbers to be inspected from the train maintenance plan database and, in conjunction with the information on the train's current parking track and the target maintenance track, filters out trains whose current parking track and target maintenance track are inconsistent and are in a state of pending maintenance. These trains are identified as needing shunting, and a preliminary judgment signal indicating the need for shunting is output. It is understood that the preliminary judgment signal includes the train number requiring shunting, the train's current parking track, and the target maintenance track.

[0032] Furthermore, in step S3, the determination of non-self-powered shunting type requires that at least one of the following determination conditions be met: 1) train's own power system failure, 2) and / or power outage of the overhead contact line at the train's current location, 3) and / or no overhead contact line at the target maintenance track, 4) and / or power outage of the overhead contact line at the target maintenance track; that is, as long as any one of the aforementioned four determination conditions is met, it can be determined that the train needs to perform non-self-powered shunting. Then, non-self-powered shunting needs are identified from the preliminary judgment signal output by the shunting demand preliminary judgment module based on the determination conditions, thereby filtering out a list of non-self-powered shunting needs, and this list includes the train number, the train's current parking track, and the target maintenance track.

[0033] Furthermore, in step S4, the basic parameters of a train to be shunted are extracted from the list of non-self-powered shunting requests obtained in step S3. These parameters include: the train number, the current parking track, the target maintenance track, the maintenance start time, the engineering car number, and the engineering car position. Simultaneously, by traversing the route, a multi-coupling task track link is constructed, consisting of "engineering car leaving the depot - lead-out track - current parking track - lead-out track - target maintenance track - lead-out track - engineering car returning to the depot." Based on this multi-coupling task track link, the engineering car and the train to be shunted must complete the following multi-coupling tasks: The engineering car first leaves the depot from its current parking position and moves to the current parking track of the train to be shunted, completing the coupling with the train. Then, it pulls the train to be shunted to the target maintenance track, subsequently detaches from the train, and returns to its parking position to return to the depot.

[0034] It should be noted that the "lead line" refers to a dedicated line or track facility used for shunting operations. Depending on the actual situation, multiple possible movement paths may still exist between different current and target locations by traversing routes and using the established lead lines. Therefore, for a particular train to be shunted, multiple feasible multi-coupling task track links may be formed, all of which are reserved as optional paths and aggregated into an optional path set. In other words, all optional paths in the optional path set are multi-coupling task track links and meet the criteria for multi-coupling task track links.

[0035] It should be noted that for multiple different optional paths, there may be differences in the number of turnout operations, the number of routes, and the path length during actual shunting operations. Therefore, it is necessary to select the optimal path from among them.

[0036] Furthermore, in step S4, the multi-objective optimization system adopts a weighted scoring method, including the following evaluation indicators: The number of turnout operations, weighted at 50%, is converted to a score of 0-50 based on an inverse ratio; that is, the fewer the number of turnout operations, the higher the score. The number of routes, weighted at 30%, is converted to a score of 0-30 inversely proportional to the number of routes; that is, the fewer the number of routes, the higher the score. Path length, with a weight of 20%, is converted to a score of 0-20 inversely proportional to the length; that is, the shorter the path length, the higher the score. Specifically, the scores of the above three indicators for each optional path in the set of optional paths are weighted and summed, and the path with the highest total score is selected as the optimal path.

[0037] Furthermore, in step S5, the automatically created time series table is generated using a time-reverse indexing method, and the time calculation rules (i.e., constraints) of the time series table include: The travel time of engineering vehicles during the shunting process needs to be considered; The coupling and uncoupling times between the trains to be moved and the engineering vehicles need to be considered; The completion time for the last task in the multi-task system (i.e., the task of returning the engineering vehicle to the depot) is set 30-60 minutes before the start time of the maintenance plan.

[0038] Furthermore, in step S6, route conflict detection is implemented through a conflict detection engine. Specifically, the conflict detection engine detects whether other operations occupy the same track simultaneously based on spatiotemporal overlap analysis. If no other operations occupy the same track simultaneously, it proves that there is no conflict, and proceeds to step S7 to continue running. If other operations occupy the same track simultaneously, it proves that a conflict has been detected, and adaptive adjustment is required until a conflict-free path is obtained.

[0039] Furthermore, the adaptive adjustment method employs a two-level adjustment strategy, specifically including: First, path optimization and adjustment are performed. Step S4 is called again to filter paths, select the second-best path from the set of available paths, and perform conflict detection. If all available paths have conflicts, the timing table is adjusted, and the process returns to step S5. Under the premise of satisfying the time calculation rules of the timing table, the shunting start time is fluctuated by 5-15 minutes before or after the time. The timing table is regenerated, and the process returns to step S6 to perform conflict detection again.

[0040] Furthermore, in step S7, the generated standardized shunting plan includes the following information: optimal path or adjusted path, timing table, shunted train number, engineering car number, currently parked track, target maintenance track, and special safety reminders; it can be understood that the special safety reminders refer to the operating specifications for tracks without overhead contact lines, as well as conditions such as the traction speed limit requirements for faulty trains.

[0041] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for automatically generating a non-self-powered shunting plan is provided.

[0042] The present invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the method for automatically generating non-self-powered shunting plans.

[0043] The present invention will be further described in detail below with reference to specific implementation examples.

[0044] The following is a scene from the daily operation of a certain rail transit depot: Maintenance plan: Train A102 needs to undergo Category A maintenance at 14:00; Train status data: Train A102's traction system is normal, and train A102 is currently parked on storage track L2, which has an overhead contact line and is currently powered normally; engineering vehicle Y01 is parked in the engineering garage and is in usable condition. Information on the target maintenance track: Track L3 meets the requirements for Class A maintenance operations, and there is no overhead contact line on this track; and the outgoing line is the shared outgoing line L1. Constraints (i.e., the time calculation rules): At least 40 minutes must be reserved before the start of the maintenance plan to complete the shunting operation (i.e., before 13:20); the operating speed of the engineering vehicle is limited to 10km / h; the time for each coupling / uncoupling is 5 minutes; it can be understood that coupling means connecting the engineering vehicle to the train to be maintained, and uncoupling means separating the engineering vehicle from the train to be maintained.

[0045] The specific implementation process of the automatic generation method for non-self-powered shunting plans provided by this invention is as follows: Step S1: Perform multi-source data acquisition The synchronous train maintenance plan updates the train maintenance plan database, indicating that train A102 needs to undergo Class A maintenance at 14:00; the real-time data collection shows that the current position of train A102 is on storage track L2, the traction system is normal, there is an overhead contact line, the pantograph is normal, and the current position of engineering vehicle Y01 is in the engineering garage; and the data collection also shows that the target maintenance track is track L3, and that track L3 has no overhead contact line.

[0046] Step S2: Preliminary assessment of shunting requirements The initial shunting demand assessment module is activated periodically, retrieving information from the database. Upon assessment, it is determined that the current position of train A102 (storage track L2) is inconsistent with its target position (maintenance track L3), and its status is pending maintenance. Therefore, the module determines that train A102 has a shunting demand and outputs an initial shunting signal.

[0047] Step S3: Determining the type of non-self-powered vehicle dispatching demand The determination is based on four criteria. Although train A102 has normal power and is currently energized, its target maintenance track, L3, lacks overhead contact lines, meeting the criterion of "target maintenance track lacking overhead contact lines." Therefore, this shunting operation is determined to be a non-self-powered shunting operation, and train A102 can be included in the list of non-self-powered shunting requests.

[0048] Step S4: Filter the optimal path Basic parameters are extracted, including: train A102 to be dispatched, current track location is L2 (storage track), target maintenance track is L3, maintenance schedule time window is before 14:00, and the engineering car is located in the engineering depot. Based on this, the multi-coupling task track link is constructed as follows: engineering car out of depot - lead-out track L1 - storage track L2 (coupled with A102) - lead-out track L1 - maintenance track L3 (decoupling) - lead-out track L1 - engineering car back to depot.

[0049] Traverse the routes and filter out the optional paths that satisfy the above track links. In this embodiment, there are two paths: Path 1: 2 turnout operations, 3 routes, and a total path length of 5000m; Path 2: 3 turnout operations, 2 routes, and a total path length of 5500m.

[0050] Multi-objective optimization evaluation employs a weighted scoring method and evaluation indicators to assess the two paths, specifically calculated as follows: For Route 1: 2 turnout operations (score: (1 / 2)*50%*100 = 25 points, standard expression is 30 points after inverse ratio conversion), 3 routes (score: (1 / 3)*30%*100 = 10 points, standard expression is 12 points), route length is 5000 meters (score: (1 / 5000)*20%*100 = 0.004 points, standard expression is 10 points after inverse ratio conversion), weighted total score = 30 + 12 + 10 = 52 points.

[0051] For Route 2: 3 turnout operations (score: (1 / 3)*50%*100 ≈ 16.7 points, standard expression is 20 points), 2 routes (score: (1 / 2)*30%*100 = 15 points, standard expression is 18 points), route length is 5500 meters (score: (1 / 5500)*20%*100 ≈ 0.0036 points, standard expression is 9 points), weighted total score = 20 + 18 + 9 = 47 points.

[0052] Therefore, comparing the weighted scores of path 1 and path 2, path 1 is the optimal path.

[0053] Step S5: Automatically create timing table Using a reverse time scheduling method, with the maintenance start time of 14:00 as the baseline, 40 minutes are pushed forward (to complete the constraint condition, i.e., 13:20), and the running time and connection / decompression time (5 minutes each) are calculated to generate the following timing table: Outbound line L1 → Engineering garage: 13:15-13:20 (running time 5 minutes); Maintenance of L3 track → L1 lead track: 13:10-13:15 (running time 5 minutes); Lead line L1 → Maintenance line L3: 13:00-13:10 (5 minutes of running time, 5 minutes of uncoiling); Parking lane L2 → Lead-out lane L1: 12:47-13:00 (running time 8 minutes, continuous coupling 5 minutes); L1 lead-out track → L2 storage track: 12:39-12:47 (running time 8 minutes); Engineering garage → Lead line L1: 12:34-12:39 (running time 5 minutes).

[0054] Step S6: Perform conflict detection If the conflict detection engine detects that no other operation is occupying the same track at the same time, it proves that there is no conflict, and proceeds to step S7 to continue running.

[0055] Step S7: Generate a standardized shunting plan The details are shown in Table 1 below: Table 1: Shunting Plan In summary, the present invention proposes an automatic generation method, medium, and equipment for non-self-powered shunting plans. Firstly, by constructing a two-stage identification logic for initial shunting demand assessment and non-self-powered shunting demand assessment, along with a multi-source data fusion mechanism, it achieves accurate and automated identification of non-self-powered shunting demands, effectively overcoming the shortcomings of traditional manual judgment, such as low efficiency and susceptibility to oversights, and providing a reliable input foundation for plan generation. Secondly, it designs a path planning algorithm based on multi-objective weighted optimization, comprehensively optimizing key indicators such as the number of turnout actions, the number of routes, and path length, significantly improving operational efficiency and reducing costs. The system addresses equipment wear and energy consumption, ensuring the standardization and consistency of planning schemes. Thirdly, it proposes an adaptive conflict resolution strategy of "prioritizing path adjustments and supplementing with time fine-tuning," which can efficiently resolve complex conflicts commonly encountered in non-self-powered shunting while ensuring maintenance timeliness and operational safety, thus improving the feasibility and reliability of the plan. Fourthly, it constructs an end-to-end fully automated generation system from demand identification to plan output, completely eliminating reliance on manual intervention in key stages, significantly improving plan generation efficiency and standardization, lowering operational barriers, and achieving full-process decision traceability.

[0056] 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 automatically generating shunting plans that are not powered by their own engines, characterized in that, Includes the following steps: Step S1: Perform multi-source data acquisition and synchronize the train maintenance plan to update the train maintenance plan database; wherein, the multi-source data acquisition includes: real-time acquisition of train status data, target maintenance track information and engineering vehicle information; Step S2: Based on the train maintenance plan database, train status data and target maintenance track information, construct a shunting demand preliminary judgment module to screen out trains with shunting needs and output a preliminary judgment signal. Step S3: Based on the preset non-self-powered shunting type determination requirements, identify non-self-powered shunting requirements from the output signal of the shunting requirement preliminary judgment module, and filter out the list of non-self-powered shunting requirements. Step S4: For the list of non-self-powered shunting needs, for a train with non-self-powered shunting needs, select the optimal path, including extracting basic parameters, traversing routes to generate a set of optional paths, and evaluating and selecting the optimal path through a multi-objective optimization system. Step S5: Based on the optimal path, automatically create a timing table and assign start and end times to the shunting task; Step S6: Perform route conflict detection on the optimal path and timing table. If a conflict is detected, perform adaptive adjustment until a conflict-free path is obtained. Step S7: Integrate conflict-free routes and timing tables to generate a standardized shunting plan; Step S8: Traverse the list of non-self-powered shunting requirements obtained in step S3. For each train with non-self-powered shunting requirements, execute steps S4-S7 in sequence.

2. The method for automatically generating non-self-powered shunting plans as described in claim 1, characterized in that, In step S1, the multi-source data acquisition includes: The train maintenance plan database is updated by synchronizing the train maintenance plan from the enterprise management system. The train maintenance plan includes, but is not limited to: the train number to be maintained, the planned maintenance type, and the maintenance start time. Real-time collection of train status data, including but not limited to: the current track where the train is parked, the status of the traction system, and the status of the pantograph; Collect information on the target maintenance track, including but not limited to: the target maintenance track, the existence status of the overhead contact line, and the status of the overhead contact line. Collect information on engineering vehicles, including but not limited to: vehicle number and vehicle location.

3. The method for automatically generating non-self-powered shunting plans as described in claim 1, characterized in that, In step S2, the shunting demand preliminary judgment module extracts the train number to be inspected from the train maintenance plan database at regular intervals, and combines the information of the train's current parking track and the target maintenance track to filter out trains that are not on the same track as the target maintenance track and are in a state of waiting for maintenance. These trains are judged to be trains that need to be shunted, and the module outputs a preliminary judgment signal that shunting is required.

4. The method for automatically generating non-self-powered shunting plans as described in claim 1, characterized in that, In step S3, the requirement for determining the non-self-powered shunting type is to satisfy at least one of the following determination conditions: And / or, the overhead contact line at the train's current location is experiencing a power outage; And / or, the target maintenance track has no overhead contact line; And / or, power outage of the overhead contact line of the target maintenance track; Specifically, non-self-powered shunting needs are identified from the preliminary judgment signals output by the shunting demand preliminary judgment module based on the judgment conditions, and a list of non-self-powered shunting needs is filtered out; and when at least one judgment condition is met, it can be determined as a non-self-powered shunting type.

5. The method for automatically generating non-self-powered shunting plans as described in claim 1, characterized in that, In step S4, the basic parameters of a train to be shunted are extracted from the list of non-self-powered shunting requirements obtained in step S3. Specifically, these parameters include: train number, current track where the train is parked, target maintenance track, maintenance start time, engineering car number, and engineering car location. Furthermore, by traversing the route, a multi-task track link is constructed, consisting of "engineering vehicle leaving the warehouse - lead-out line - currently parked track - lead-out line - target maintenance track - lead-out line - engineering vehicle returning to the warehouse".

6. The method for automatically generating non-self-powered shunting plans as described in claim 5, characterized in that, The multi-objective optimization system adopts a weighted scoring method, including the following evaluation indicators: The number of turnout operations, weighted at 50%, is converted to a score of 0-50 based on the inverse proportion of the number of operations. The number of routes, with a weight of 30%, is converted to a score of 0-30 based on the inverse proportion of the number of routes. Path length, weighted at 20%, is converted to a score of 0-20 inversely proportional to its length. Specifically, the scores of the three indicators for each optional path in the set of optional paths are weighted and summed, and the path with the highest total score is selected as the optimal path.

7. The method for automatically generating non-self-powered shunting plans as described in claim 1, characterized in that, In step S5, the time calculation rules for the time series table include: The travel time of engineering vehicles during the shunting process needs to be considered; The coupling and uncoupling times between the trains to be moved and the engineering vehicles need to be considered; The completion time for the last task in a multi-task system is set 30-60 minutes before the start time of the maintenance plan.

8. The method for automatically generating non-self-powered shunting plans as described in claim 7, characterized in that, The automatically created time series table is generated using a time-reverse indexing method.

9. The method for automatically generating non-self-powered shunting plans as described in claim 1, characterized in that, In step S6, route collision detection is implemented through a collision detection engine; The conflict detection engine detects whether other operations occupy the same track at the same time based on spatiotemporal overlap analysis.

10. The method for automatically generating non-self-powered shunting plans as described in claim 9, characterized in that, If no other work is occupying the same track at the same time, it proves that there is no conflict, proceed to step S7 and continue running; If other operations occupy the same track at the same time, it indicates that a conflict has been detected, and adaptive adjustments are required until a conflict-free path is obtained.

11. The method for automatically generating non-self-powered shunting plans as described in claim 10, characterized in that, The adaptive adjustment method employs a two-level adjustment strategy, including the following steps: First, the path is optimized and adjusted. Step S4 is called again to filter the path, select the second-best path from the set of available paths, and perform conflict detection. Furthermore, the timing table is adjusted only when all the optional paths in the path optimization adjustment have conflicts. Then, the process returns to step S5, where the shunting start time is fluctuated by 5-15 minutes before or after the timing table is satisfied, and the timing table is regenerated before proceeding to step S6 for conflict detection.

12. The method for automatically generating non-self-powered shunting plans as described in claim 1, characterized in that, In step S7, the generated standardized shunting plan includes the following information: optimal route or adjusted route, timing table, shunted train number, engineering car number, current parking track, target maintenance track, and special safety tips.

13. The method for automatically generating non-self-powered shunting plans as described in claim 12, characterized in that, The specific safety tips include: operating procedures for overhead contact line tracks and speed limits for traction of faulty trains.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for automatically generating non-self-powered shunting plans as described in any one of claims 1 to 13.

15. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the method for automatically generating non-self-powered shunting plans as described in any one of claims 1 to 13.

Images