Marshalling station time consumption prediction method and system based on large model

Abstract

The invention provides a marshalling station time consumption prediction method and system based on a large model, and the method comprises the following steps: obtaining the data of each operation stage of the disassembly and marshalling operation of a marshalling station, and obtaining the data of the operation stages; training by using the operation stage data to obtain a marshalling station disintegration and marshalling operation time consumption prediction model; and using the marshalling station disintegration and marshalling operation time consumption prediction model to predict time consumption information corresponding to each stage of marshalling station disintegration and marshalling operation. According to the technical scheme, the efficiency and accuracy of marshalling station operation management are improved.

Description

Background Art

[0002] In the operation and management of railway marshalling yards, the accurate prediction of the time-consuming of disintegration operations and marshalling operations is crucial for improving transport efficiency. Traditional prediction methods mainly rely on manual train following and manual time recording, followed by manual statistical analysis. This manual operation mode has many defects. First, a large amount of manual input is required, and a large number of staff need to be arranged to follow the train, record and statistically analyze data. This not only increases the labor cost, but also easily leads to data recording errors or omissions due to human factors. Second, the cycle is long. It often takes a long time from data collection to the final prediction result, which is difficult to meet the requirements of high efficiency and rapidity of modern railway transportation. Third, the efficiency is low. The speed of manual data processing is limited, and real-time prediction of operation time-consuming cannot be achieved. Finally, the accuracy is poor. Manual recording and statistics are difficult to accurately grasp various complex situations in the operation process, resulting in a large deviation between the prediction result and the actual situation, and the timeliness is poor. It cannot provide effective support for operation decision-making in a timely manner, seriously restricting the improvement of the operation management level of marshalling yards and the improvement of railway transport efficiency. Summary of the Invention

[0003] This application provides a method and system for predicting the time-consuming of marshalling yards based on large models to improve the efficiency and accuracy of marshalling yard operation management.

[0004] In the first aspect, a method for predicting the time-consuming of marshalling yards based on large models is provided, including the following steps: Obtain data of each operation stage of disintegration and marshalling operations in the marshalling yard to obtain operation stage data; Use the operation stage data to train a prediction model for the time-consuming of disintegration and marshalling operations in the marshalling yard; Use the prediction model for the time-consuming of disintegration and marshalling operations in the marshalling yard to predict the time-consuming information corresponding to each stage of disintegration and marshalling operations in the marshalling yard.

[0005] In the above technical solution, by obtaining data of each operation stage of disintegration and marshalling operations in the marshalling yard to obtain operation stage data; using the operation stage data to train a prediction model for the time-consuming of disintegration and marshalling operations in the marshalling yard; using the prediction model for the time-consuming of disintegration and marshalling operations in the marshalling yard to predict the time-consuming information corresponding to each stage of disintegration and marshalling operations in the marshalling yard; the efficiency and accuracy of marshalling yard operation management are improved.

[0006] In a specific feasible implementation scheme, the step of using the operation stage data to train a prediction model for the time-consuming of disintegration and marshalling operations in the marshalling yard specifically includes: Use the operation stage data to obtain a training sample set; Use the training sample set to train a prediction model for the time-consuming of disintegration and marshalling operations in the marshalling yard.

[0007] In a specific feasible implementation, the LightGBM algorithm is used to train the prediction model for the time consumption of marshalling yard disintegration and marshalling operations.

[0008] In a specific feasible implementation, normalization processing is performed on the first feature dataset and the second feature dataset of the training sample set, and the normalized data is used as the preset time consumption prediction model for marshalling yard disintegration and marshalling operations, thereby obtaining the prediction model for the time consumption of marshalling yard disintegration and marshalling operations.

[0009] Secondly, a marshalling yard time consumption prediction system based on a large model is provided, including: An operation stage data module, configured to obtain the data of each operation stage of marshalling yard disintegration and marshalling operations, so as to obtain the operation stage data; A prediction model module, configured to use the operation stage data to train and obtain a prediction model for the time consumption of marshalling yard disintegration and marshalling operations; A time consumption prediction module, configured to use the prediction model for the time consumption of marshalling yard disintegration and marshalling operations to predict the time consumption information corresponding to each stage of marshalling yard disintegration and marshalling operations.

[0010] In the above technical solution, by obtaining the data of each operation stage of marshalling yard disintegration and marshalling operations to obtain the operation stage data; using the operation stage data to train and obtain a prediction model for the time consumption of marshalling yard disintegration and marshalling operations; using the prediction model for the time consumption of marshalling yard disintegration and marshalling operations to predict the time consumption information corresponding to each stage of marshalling yard disintegration and marshalling operations; the efficiency and accuracy of marshalling yard operation management are improved.

[0011] In a specific feasible implementation, using the operation stage data to train and obtain a prediction model for the time consumption of marshalling yard disintegration and marshalling operations specifically includes: Using the operation stage data to obtain a training sample set; Using the training sample set to train and obtain a prediction model for the time consumption of marshalling yard disintegration and marshalling operations.

[0012] In a specific feasible implementation, the LightGBM algorithm is used to train the prediction model for the time consumption of marshalling yard disintegration and marshalling operations.

[0013] In a specific feasible implementation, normalization processing is performed on the first feature dataset and the second feature dataset of the training sample set, and the normalized data is used as the preset time consumption prediction model for marshalling yard disintegration and marshalling operations, thereby obtaining the prediction model for the time consumption of marshalling yard disintegration and marshalling operations.

[0014] In a third aspect, an electronic device is provided. The electronic device includes a processor, the processor is coupled to a memory, and at least one computer program is stored in the memory. The at least one computer program is loaded and executed by the processor to enable the electronic device to implement any one of the big model-based marshalling station time-consuming prediction methods.

[0015] In the above technical solution, by obtaining the data of each operation stage of the marshalling station's disintegration and marshalling operations, the operation stage data is obtained; using the operation stage data, a time-consuming prediction model for the marshalling station's disintegration and marshalling operations is trained; using the time-consuming prediction model for the marshalling station's disintegration and marshalling operations to predict the time-consuming information corresponding to each stage of the marshalling station's disintegration and marshalling operations; the efficiency and accuracy of the marshalling station operation management are improved.

[0016] In a fourth aspect, a computer-readable storage medium is provided. At least one computer program is stored in the computer-readable storage medium. The at least one computer program is loaded and executed by a processor to enable the computer-readable storage medium to implement any one of the big model-based marshalling station time-consuming prediction methods.

[0017] In the above technical solution, by obtaining the data of each operation stage of the marshalling station's disintegration and marshalling operations, the operation stage data is obtained; using the operation stage data, a time-consuming prediction model for the marshalling station's disintegration and marshalling operations is trained; using the time-consuming prediction model for the marshalling station's disintegration and marshalling operations to predict the time-consuming information corresponding to each stage of the marshalling station's disintegration and marshalling operations; the efficiency and accuracy of the marshalling station operation management are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of the big model-based marshalling station time-consuming prediction method provided by an embodiment of the present application; Figure 2 It is a structural block diagram of the big model-based marshalling station time-consuming prediction system provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The present application will be further described in detail below with reference to the drawings and embodiments. Through these descriptions, the features and advantages of the present application will become more clearly defined.

[0020] The special term "exemplary" here means "serving as an example, embodiment or illustration". Any embodiment described as "exemplary" here does not have to be construed as superior or better than other embodiments. Although various aspects of the embodiments are shown in the drawings, unless otherwise specified, the drawings do not have to be drawn to scale.

[0021] In addition, the technical features involved in different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0022] To facilitate the understanding of the marshalling station time-consuming prediction method and system based on a large model provided by the embodiments of the present application, the application scenario will be described first. The marshalling station time-consuming prediction method and system based on a large model provided by the embodiments of the present application are used to improve the efficiency and accuracy of marshalling station operation management. In the operation management of railway marshalling stations, the accurate prediction of the time-consuming of disintegration operations and marshalling operations is crucial for improving transportation efficiency. The traditional prediction methods mainly rely on manual vehicle following and manual time recording, and then manual statistical analysis. This manual operation mode has many defects. First, a large amount of manual input is required, and a large number of staff need to be arranged to follow the vehicle, record and statistically analyze data. This not only increases the labor cost, but also easily leads to data recording errors or omissions due to human factors. Second, the cycle is long. It often takes a long time from data collection to the final prediction result, which is difficult to meet the requirements of high-efficiency and fast modern railway transportation. Third, the efficiency is low. The speed of manual data processing is limited, and real-time operation time-consuming prediction cannot be achieved. Finally, the accuracy is poor. Manual recording and statistics are difficult to accurately grasp various complex situations in the operation process, resulting in a large deviation between the prediction result and the actual situation, and poor timeliness. It cannot provide effective support for operation decisions in a timely manner, seriously restricting the improvement of marshalling station operation management level and railway transportation efficiency. Therefore, the embodiments of the present application provide a marshalling station time-consuming prediction method and system based on a large model to improve the efficiency and accuracy of marshalling station operation management. The following will be described in detail with specific drawings by way of examples.

[0023] Reference Figure 1 and Figure 2 , Figure 1 is the flowchart of the marshalling station time-consuming prediction method based on a large model provided by the embodiments of the present application; Figure 2 is the structural block diagram of the marshalling station time-consuming prediction system based on a large model provided by the embodiments of the present application.

[0024] In Figure 1 ,the embodiments of the present application provide a marshalling station time-consuming prediction method based on a large model, including the following steps: Obtain the data of each operation stage of the marshalling station disintegration and marshalling operations to obtain the operation stage data; Use the operation stage data to train a marshalling station disintegration and marshalling operation time-consuming prediction model; Use the marshalling station disintegration and marshalling operation time-consuming prediction model to predict the time-consuming information corresponding to each stage of the marshalling station disintegration and marshalling operations.

[0025] In the above technical solution, by obtaining the data of each operation stage of the marshalling station's disintegration and marshalling operations, the operation stage data is obtained; using the operation stage data, a prediction model for the time consumption of the marshalling station's disintegration and marshalling operations is trained; using the prediction model for the time consumption of the marshalling station's disintegration and marshalling operations to predict the time consumption information corresponding to each stage of the marshalling station's disintegration and marshalling operations; the efficiency and accuracy of the marshalling station operation management are improved.

[0026] In a specific and feasible implementation, the step of training a prediction model for the time consumption of the marshalling station's disintegration and marshalling operations by using the operation stage data specifically includes: Using the operation stage data, a training sample set is obtained; Using the training sample set to train a prediction model for the time consumption of the marshalling station's disintegration and marshalling operations.

[0027] In a specific and feasible implementation, the LightGBM algorithm is used to train the prediction model for the time consumption of the marshalling station's disintegration and marshalling operations.

[0028] It should be noted that LightGBM is a machine learning algorithm based on Gradient Boosting Decision Tree (GBDT), including: Histogram-based decision tree algorithm: During the training process, LightGBM uses a histogram-based algorithm to split nodes instead of the traditional sorting-based decision tree algorithm. This method divides continuous feature values into discrete bins, constructs a histogram, and then traverses these bins to find the optimal splitting point. This method reduces memory usage and improves training speed, especially when dealing with large-scale data.

[0029] Leaf-wise growth strategy: Different from the traditional Level-wise growth strategy (i.e., all leaves at each layer are split simultaneously), LightGBM adopts the Leaf-wise growth strategy, that is, each time the leaf with the largest gain is selected from the current all leaves for splitting. This strategy can improve the accuracy of the model, but it is also prone to overfitting. Therefore, LightGBM introduces the max-depth and min-data-in-leaf parameters to control the depth of the tree and the minimum amount of data in the leaf nodes.

[0030] Compared with traditional gradient boosting algorithms (such as XGBoost), LightGBM has the following advantages: Faster training speed: By using histogram acceleration and Leaf-wise growth strategy, LightGBM is faster than traditional methods when dealing with large-scale data.

[0031] Lower memory consumption: The histogram algorithm and sparse optimization technology significantly reduce the use of memory, enabling larger datasets to be processed.

[0032] Better accuracy: The leaf-wise growth strategy and histogram optimization can improve the accuracy of the model, especially performing better on complex datasets and high-dimensional data.

[0033] Support for parallel training: LightGBM supports parallel training and can effectively utilize multi-core CPUs for acceleration.

[0034] Preferably, the marshalling yard disintegration and marshalling operation time-consuming prediction model adopts the LightGBM (Light Gradient Boosting Machine) large model.

[0035] In a specific feasible implementation, normalization processing is performed on the first feature dataset and the second feature dataset based on the training sample set, and the normalized data is used as the preset marshalling yard disintegration and marshalling operation time-consuming prediction model to obtain the marshalling yard disintegration and marshalling operation time-consuming prediction model.

[0036] Specifically, the marshalling yard time-consuming prediction method based on the large model includes: obtaining data for each link of the current disintegration and marshalling operations; among them, the disintegration operations include, but are not limited to, data for 6 operation links such as waiting at the peak, empty journey, returning to the turnout, coupling, pre-pushing, and pushing the hump, and the marshalling operations include, but are not limited to, data for 6 operation links such as waiting for coupling, marshalling, pulling out, uncoupling and coupling, returning to the turnout, and empty journey, as well as external data. The data constituting the operation links are the basic information data of the current operation, and the external data is the data other than the basic information data at the start of the current operation link; inputting the operation link data and the data other than it into the preset disintegration and marshalling time-consuming prediction model to predict the time-consuming information corresponding to the current marshalling and disintegration operation links.

[0037] Further, the step of obtaining the current marshalling yard disintegration and marshalling operations and their peripheral data includes: obtaining the first feature data and the second feature data corresponding to the current marshalling yard disintegration and marshalling operations and their peripheral data respectively; among them, the first feature data includes, but is not limited to, waiting at the peak, empty journey, returning to the turnout, coupling, pre-pushing, pushing the hump, and the marshalling operations include, but are not limited to, waiting for coupling, marshalling, pulling out, uncoupling and coupling, returning to the turnout, and empty journey, and the second feature data includes, but is not limited to, the data of the storage yard, the storage track data, the hump type, the shunting locomotive number data, the shift information data of the operating personnel, the marshalling yard data, the pulling yard data, the weather information data, and the temperature information data; the step of inputting the marshalling yard disintegration and marshalling operations and their peripheral data into the preset marshalling yard marshalling and disintegration time-consuming prediction model to predict the time-consuming information corresponding to each operation link of the current marshalling yard marshalling and disintegration includes: inputting the first feature data and the second feature data into the preset marshalling yard disintegration and marshalling time-consuming prediction model to predict the time-consuming information corresponding to each operation link of the current marshalling yard disintegration and marshalling.

[0038] Preferably, the steps of training the time-consuming prediction model for the disintegration and formation operations of a marshalling station based on a training sample set to obtain the trained time-consuming prediction model for the disintegration and formation of a marshalling station include: using the first feature and the second feature data in the training sample set as the input data of the LightGBM algorithm, and using the operation time-consuming data as the output data of the LightGBM algorithm to train the time-consuming prediction model for the disintegration and formation operations of a marshalling station.

[0039] In this embodiment, the current disintegration and formation operation data information of the marshalling station is obtained; wherein, the operation data includes but is not limited to: operation data at each stage, operation-related train formation information data, formation plan information data, operation shunting locomotive information data, operation personnel team and weather information data, and the operation data at each operation stage of the current disintegration and formation of the marshalling station includes the operation start time and the operation end time as the basic information data of the data; the above operation data is respectively input into the preset time-consuming prediction model for the disintegration and formation operations of the marshalling station to predict the time-consuming information corresponding to each stage of the current formation or disintegration operation.

[0040] Preferably, the steps of obtaining the current disintegration and formation operation data of the marshalling station include the first feature data and the second feature data; wherein, the first feature data includes operation stage execution data, operation stage splitting dependency data, data of the start execution time and the end time of the stage operation, car type and vehicle type data, train formation information data, shunting operation plan data, operation personnel team data, and the second feature data includes yard data, track information data, hump information data, shunting locomotive information data, operation personnel team information data, weather information data, etc.; the disintegration and formation operation data of the marshalling station is respectively input into the preset time-consuming prediction model for the disintegration and formation operations of the marshalling station to predict the time-consuming information corresponding to each stage of the current disintegration and formation operations, including: inputting the first feature data and the second feature data into the preset time-consuming prediction model for the disintegration and formation operations of the marshalling station to predict the time-consuming information corresponding to each stage of the current disintegration and formation operations of the marshalling station.

[0041] Preferably, the marshalling station disintegration and marshalling operation prediction model is trained through the following steps: obtaining a training sample set; wherein, the training sample set includes a plurality of training samples, and the training samples are operation data of each stage of the historical disintegration and marshalling operations of the marshalling station. The historical disintegration and marshalling operation data of the marshalling station include execution data of each stage, car formation, marshalling plan, shunting locomotive information, and work team information data. The operation stage data includes operation stage composition data and operation stage duration data. The operation stage composition data is the historical basic data information constituting the disintegration and marshalling operations of the marshalling station, and the external data is the data other than the basic information data when the historical disintegration and marshalling operations of the marshalling station start to be executed; training the marshalling station disintegration and marshalling operation duration prediction model based on the training sample set to obtain the trained marshalling station disintegration and marshalling operation duration prediction model.

[0042] Preferably, the steps of obtaining the first feature data, second feature data corresponding to the operation stage composition data, car formation, marshalling plan, shunting locomotive information, work team information data, and operation stage duration data respectively include: yard information data, track information data, hump information data, shunting locomotive information data, work shift information data, and weather information data as the second feature data.

[0043] Preferably, based on the first feature data set and the second feature data set of the training sample set, reasonable normalization processing is performed, and the normalized data is used as the preset marshalling station disintegration and marshalling operation duration prediction model to obtain the trained marshalling station disintegration and marshalling operation stage duration prediction model.

[0044] Preferably, the steps of training the marshalling station disintegration and marshalling operation duration prediction model based on the training sample set to obtain the trained marshalling station disintegration and marshalling operation duration prediction model include: using the operation stage composition data, yard information data, track information data, hump information data, shunting locomotive information data, work shift information data, and weather information data in the training sample set as the input data of the LightGBM algorithm, and using the operation stage duration data as the output data of the LightGBM algorithm to train the marshalling station disintegration and marshalling operation duration prediction model.

[0045] In the above technical solution, the influence of the data of each operation link itself and external data on the duration of the disintegration operation and marshalling operation is fully considered, and the duration of the disintegration operation and marshalling operation can be accurately predicted, so as to facilitate the reasonable formulation of operation plans, optimize operation management work, and discover potential bottleneck problems in advance.

[0046] In Figure 2In this case, the embodiment of the present application provides a marshalling station time-consuming prediction system based on a large model, including: An operation stage data module, configured to obtain data of each operation stage of marshalling station disintegration and marshalling operations, and obtain operation stage data; A prediction model module, configured to use the operation stage data to train a marshalling station disintegration and marshalling operation time-consuming prediction model; A time-consuming prediction module, configured to use the marshalling station disintegration and marshalling operation time-consuming prediction model to predict the time-consuming information corresponding to each stage of marshalling station disintegration and marshalling operations.

[0047] In the above technical solution, by obtaining data of each operation stage of marshalling station disintegration and marshalling operations to obtain operation stage data; using the operation stage data to train a marshalling station disintegration and marshalling operation time-consuming prediction model; using the marshalling station disintegration and marshalling operation time-consuming prediction model to predict the time-consuming information corresponding to each stage of marshalling station disintegration and marshalling operations; the efficiency and accuracy of marshalling station operation management are improved.

[0048] In a specific feasible implementation, using the operation stage data to train a marshalling station disintegration and marshalling operation time-consuming prediction model specifically includes: Using the operation stage data to obtain a training sample set; Using the training sample set to train a marshalling station disintegration and marshalling operation time-consuming prediction model.

[0049] In a specific feasible implementation, the LightGBM algorithm is used to train the marshalling station disintegration and marshalling operation time-consuming prediction model.

[0050] In a specific feasible implementation, based on the first feature data set and the second feature data set of the training sample set, normalization processing is performed, and the normalized data is used as a preset time-consuming prediction model for marshalling station disintegration and marshalling operations to obtain a marshalling station disintegration and marshalling operation time-consuming prediction model.

[0051] Specifically, the marshalling station time-consuming prediction system based on the large model includes: An acquisition module, configured to acquire current marshalling station disintegration and marshalling operation data; wherein, the marshalling station disintegration and marshalling operation data includes operation link composition data and car train composition, marshalling plan, locomotive information, and operation team information data, the operation link composition data is basic information data constituting the current marshalling station disintegration and marshalling operations, and the yard information data, track information data, hump information data, locomotive information data, operation team information data, weather information data are data other than the basic information data for the start and end time information of the current marshalling station disintegration and marshalling operations; A prediction module, configured to input data other than the marshalling station disintegration and marshalling operation data and the basic information data into a preset marshalling station disintegration and marshalling operation time consumption prediction model, so as to predict the time consumption information corresponding to the current marshalling station disintegration and marshalling operation.

[0052] An embodiment of the present application further provides an electronic device, which includes a processor. The processor is coupled with a memory, and at least one computer program is stored in the memory. The at least one computer program is loaded and executed by the processor, so that the electronic device implements any one of the marshalling station time consumption prediction methods based on a large model.

[0053] In the above technical solution, by obtaining the data of each operation stage of the marshalling station disintegration and marshalling operation, operation stage data is obtained; using the operation stage data, a marshalling station disintegration and marshalling operation time consumption prediction model is trained; using the marshalling station disintegration and marshalling operation time consumption prediction model to predict the time consumption information corresponding to each stage of the marshalling station disintegration and marshalling operation; the efficiency and accuracy of marshalling station operation management are improved.

[0054] An embodiment of the present application further provides a computer-readable storage medium, in which at least one computer program is stored. The at least one computer program is loaded and executed by a processor, so that the computer-readable storage medium implements any one of the marshalling station time consumption prediction methods based on a large model.

[0055] In the above technical solution, by obtaining the data of each operation stage of the marshalling station disintegration and marshalling operation, operation stage data is obtained; using the operation stage data, a marshalling station disintegration and marshalling operation time consumption prediction model is trained; using the marshalling station disintegration and marshalling operation time consumption prediction model to predict the time consumption information corresponding to each stage of the marshalling station disintegration and marshalling operation; the efficiency and accuracy of marshalling station operation management are improved.

[0056] Those skilled in the art of the technical field to which the present application belongs know that the present application can be implemented as a system, a method, or a computer program product.

[0057] Therefore, the present disclosure can be specifically implemented in the following forms: it can be completely hardware, can be completely software (including firmware, resident software, microcode, etc.), or can be a combination of hardware and software. Generally, it is referred to as "circuit", "module" or "system" in this article. In addition, in some embodiments, the present application can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable media contains computer-readable program codes.

[0058] Any combination of one or more computer-readable media may be employed. The computer-readable media may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example - but not limited to - an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In this document, the computer-readable storage medium may be any tangible medium that contains or stores a program which can be used by or in connection with an instruction execution system, apparatus, or device.

[0059] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application. On this basis, various substitutions and improvements can be made to the present application, and all of these fall within the protection scope of the present application.

Claims

1. A method for predicting the time consumption of a marshalling station based on a large model, characterized in that, It includes the following steps: Obtain the data of each operation stage of the marshalling station's disintegration and marshalling operations to obtain the operation stage data; Use the operation stage data to train a prediction model for the time consumption of the marshalling station's disintegration and marshalling operations; Use the prediction model for the time consumption of the marshalling station's disintegration and marshalling operations to predict the time consumption information corresponding to each stage of the marshalling station's disintegration and marshalling operations.

2. The method for predicting the time consumption of a marshalling station based on a large model according to claim 1, wherein The step of using the operation stage data to train a prediction model for the time consumption of the marshalling station's disintegration and marshalling operations specifically includes: Use the operation stage data to obtain a training sample set; Use the training sample set to train a prediction model for the time consumption of the marshalling station's disintegration and marshalling operations.

3. The method for predicting the time consumption of a marshalling station based on a large model according to claim 2, wherein Adopt the LightGBM algorithm to train the prediction model for the time consumption of the marshalling station's disintegration and marshalling operations.

4. The method for predicting the time consumption of a marshalling station based on a large model according to claim 3, wherein Perform normalization processing on the first feature data set and the second feature data set of the training sample set, and use the normalized data as the preset time consumption prediction model for the marshalling station's disintegration and marshalling operations to obtain the prediction model for the time consumption of the marshalling station's disintegration and marshalling operations.

5. A marshalling station time-consuming prediction system based on a large model, characterized in that, It includes: An operation stage data module for obtaining the data of each operation stage of the marshalling station's disintegration and marshalling operations to obtain the operation stage data; A prediction model module for using the operation stage data to train a prediction model for the time consumption of the marshalling station's disintegration and marshalling operations; A time consumption prediction module for using the prediction model for the time consumption of the marshalling station's disintegration and marshalling operations to predict the time consumption information corresponding to each stage of the marshalling station's disintegration and marshalling operations.

6. The marshalling station time-consuming prediction system based on a large model according to claim 5, characterized in that, The step of using the operation stage data to train a prediction model for the time consumption of the marshalling station's disintegration and marshalling operations specifically includes: Use the operation stage data to obtain a training sample set; Use the training sample set to train a prediction model for the time consumption of the marshalling station's disintegration and marshalling operations.

7. The marshalling station time-consuming prediction system based on a large model according to claim 6, characterized in that, Adopt the LightGBM algorithm to train the prediction model for the time consumption of the marshalling station's disintegration and marshalling operations.

8. The marshalling station time-consuming prediction system based on a large model according to claim 7, characterized in that, Perform normalization processing on the first feature data set and the second feature data set of the training sample set, and use the normalized data as the preset time consumption prediction model for the marshalling station's disintegration and marshalling operations to obtain the prediction model for the time consumption of the marshalling station's disintegration and marshalling operations.

9. An electronic device, characterized in that, The electronic device includes a processor, the processor is coupled with a memory, and at least one computer program is stored in the memory. The at least one computer program is loaded and executed by the processor so that the electronic device implements the method for predicting the time consumption of a marshalling station based on a large model according to any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, At least one computer program is stored in the computer-readable storage medium. The at least one computer program is loaded and executed by a processor so that the computer-readable storage medium implements the method for predicting the time consumption of a marshalling station based on a large model according to any one of claims 1 to 4.

Images