Train return capability improvement method, electronic equipment and storage medium

Abstract

The present application provides a method for improving the train turnaround capacity, an electronic device and a storage medium, the method comprising: building a business model based on the train turnaround business; based on the business model, using dimensional modeling to construct a data bus matrix and a physical model; according to the physical model, performing extraction-conversion-loading development to obtain the final train turnaround data; analyzing the final train turnaround data to obtain factors affecting the train turnaround; mining the factors to determine a plan to improve the train turnaround capacity. The method provided in the present application is based on the train turnaround business, constructing a physical model, and based on the physical model, performing extraction-conversion-loading development to obtain the final train turnaround data, and then based on the final train turnaround data, obtaining factors affecting the train turnaround, mining the factors, determining a plan to improve the train turnaround capacity, and realizing the improvement of the urban rail operation capacity.

Description

Technical Field

[0001] The present application relates to the field of rail transit technology, and in particular to a method for improving the return capability of a train, an electronic device and a storage medium. Background Art

[0002] With the continuous growth of urban rail construction and operating mileage, and under the background of the country's smart transportation strategy, how to achieve the development goals of "ensuring operational safety, improving operational efficiency, lean enterprise management, and improving passenger experience" is a difficult problem facing operating companies. The train turnaround capability is an important factor in improving the operational efficiency of urban rail transit, so improving the train turnaround capability is of great significance to the efficient operation of urban rail and improving passenger experience.

[0003] During the operation of the train, a large amount of data is generated, including operation log data, various signal data, equipment data, etc. From the perspective of data structure, it can be divided into structured, semi-structured, and unstructured data. These data are not only messy but also very tedious and complicated to sort out. If these data are directly analyzed and mined, due to the differences in data structure, it will be impossible to accurately associate data of different structures, affecting the accuracy of analysis and mining, and making the final train's return ability have certain limitations due to inaccurate analysis and mining. Summary of the invention

[0004] In order to improve the turning capability of a train, the present application provides a method for improving the turning capability of a train, an electronic device and a storage medium.

[0005] In a first aspect of the present application, a method for improving the turnaround capacity of a train is provided, the method comprising:

[0006] Establish a business model based on the train return business;

[0007] Based on the business model, dimensional modeling is used to construct a data bus matrix and a physical model;

[0008] According to the physical model, extract-convert-load development is performed to obtain the final train return data;

[0009] Analyze the final train reversal data to obtain factors affecting the train reversal;

[0010] The factors are explored to determine the plan for improving the train's turnaround capacity.

[0011] Optionally, constructing a data bus matrix based on the business model by using dimensional modeling includes:

[0012] Select the service model based on the priority of the train return service;

[0013] Declare the granularity of the fact table for the chosen business model;

[0014] Determine dimensions and facts based on the selected business model and granularity;

[0015] Build a data bus matrix based on the business model, dimensions, and facts selected under each data domain.

[0016] Optionally, the physical model includes: a source data layer, a dimension layer, a data standard layer, a unified data warehouse layer, a label data layer and an application layer;

[0017] The source data layer is used to store original data;

[0018] The dimension layer is used to store dimensions;

[0019] The data standard layer is used to store the processed data of the source data layer;

[0020] The unified data warehouse layer is used to store the final data;

[0021] The label data layer is used to store the label of the final data;

[0022] The application layer is used to provide application services.

[0023] Optionally, performing extraction-transformation-loading development according to the physical model includes:

[0024] Extracting initial train return data, and storing the initial train return data in the source data layer;

[0025] Convert the initial train return data and store the conversion result in the data standard layer;

[0026] Load the dimension layer;

[0027] According to the dimension layer, the data in the data standard layer is processed to obtain the final train return data;

[0028] The final train return data is stored in the unified data warehouse layer, and the label data layer and application layer are constructed based on the final train return data.

[0029] Optionally, analyzing the final train return data includes:

[0030] According to the final train reversal data, the average speed of the train when turning in and out of the reversal track is calculated, the historical reversal data is compared, and the reversal situations in different time periods are compared.

[0031] Optionally, the factors affecting the train's reversal are one or more of the following: the stop time at the starting platform, the time for handling the turning route, the time for traveling in the turning section, the average speed in the turning section, the time for changing ends of the train, the time for handling the turning route, the time for traveling in the turning section, the average speed in the turning section, and the stop time at the ending platform.

[0032] Optionally, mining the factors to determine a solution to improve the train turnaround capacity includes:

[0033] Mining the factor based on a pre-trained linear regression model to obtain an adjusted value of the factor;

[0034] A plan for improving the train's turnaround capacity is determined based on the adjustment value.

[0035] Optionally, the linear regression model is:

[0036]

[0037] Among them, i is the factor identifier, n is the total number of factors, and x i is the data corresponding to the i-th factor, θ i is the adjustment value corresponding to the i-th factor.

[0038] In a second aspect of the present application, an electronic device is provided, comprising:

[0039] Memory;

[0040] Processor; and

[0041] Computer programs;

[0042] The computer program is stored in the memory and is configured to be executed by the processor to implement the method as described in the first aspect above.

[0043] According to a third aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored; the computer program is executed by a processor to implement the method described in the first aspect above.

[0044] The present application provides a method for improving train reversal capacity, an electronic device and a storage medium, the method comprising: establishing a business model based on the train reversal business; based on the business model, using dimensional modeling to construct a data bus matrix and a physical model; performing extraction-conversion-loading development according to the physical model to obtain the final train reversal data; analyzing the final train reversal data to obtain factors affecting the train reversal; mining the factors to determine a plan to improve the train reversal capacity.

[0045] Among them, the method provided in this application is based on the train return business, builds a physical model, and according to the physical model, performs extraction-conversion-loading development to obtain the final train return data, and then based on the final train return data, obtains the factors affecting the train return, mines the factors, determines the train return capacity improvement plan, and realizes the improvement of the urban rail operation capacity.

[0046] In addition, in one implementation, specific limitations are placed on the construction of the data bus matrix, clarifying the implementation process of constructing the data bus matrix based on the business model and using dimensional modeling, ensuring the correlation between the data bus matrix and the business, improving the matching degree between the data bus matrix and the train return business, ensuring the feasibility and improvement effect of the train return capacity improvement plan, and realizing the improvement of the urban rail operation capacity.

[0047] In addition, in one implementation, the architecture of the physical model is specifically limited. Through this physical model, only a set of unified standards needs to be built based on the original layer data. The standards for data production downstream remain unchanged, which is used to solve the problem of inconsistent data formats of different manufacturers, thereby enhancing the universal capabilities of this architecture.

[0048] In addition, in one implementation, the scheme for extracting, converting and loading according to the physical model is specifically limited, so that the final train return data format obtained after development is consistent, making subsequent mining and analysis more meaningful, ensuring the feasibility and improvement effect of the train return capacity improvement scheme, and realizing the improvement of urban rail operation capacity.

[0049] In addition, in one implementation, the final train turnaround data analysis process is specifically limited, and the relationship between the factors affecting the train turnaround and the business model is analyzed, which ensures the feasibility and improvement effect of the train turnaround capacity improvement plan and achieves the improvement of the urban rail operation capacity.

[0050] In addition, in one implementation, the factors of train reversal are specifically defined, and the relationship between the factors affecting train reversal and the business model is clarified, thereby ensuring the feasibility and improvement effect of the plan to improve the train reversal capacity and achieving an improvement in the urban rail operating capacity.

[0051] In addition, in one implementation, by mining factors to determine a plan to improve the train's turnaround capacity, it is more conducive to improving the train's turnaround capacity, increasing the train's carrying capacity, and thus improving the operational efficiency of rail transit.

[0052] In addition, in one implementation, by limiting the linear regression model, the factor mining plan is clarified, which ensures the improvement of the train's return capacity, thereby improving the operational efficiency of rail transit.

[0053] The electronic device provided by the present application has a computer program executed by a processor to construct a physical model based on the train return business, and according to the physical model, extract-convert-load development is performed to obtain the final train return data, and then based on the final train return data, the factors affecting the train return are obtained, the factors are mined, and a plan to improve the train return capacity is determined to achieve an improvement in the urban rail operation capacity.

[0054] The computer-readable storage medium provided in the present application has a computer program thereon which is executed by a processor to construct a physical model based on the train return business, and according to the physical model, extract-convert-load development is performed to obtain the final train return data, and then based on the final train return data, the factors affecting the train return are obtained, the factors are mined, and a plan to improve the train return capacity is determined to achieve an improvement in the urban rail operation capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0056] Figure 1 A flow chart of a method for improving the train turnaround capability provided in an embodiment of the present application;

[0057] Figure 2 A schematic diagram of a train turnaround provided in an embodiment of the present application;

[0058] Figure 3 A flowchart of a dimensional modeling method provided in an embodiment of the present application;

[0059] Figure 4 A schematic diagram of a physical model architecture provided in an embodiment of the present application;

[0060] Figure 5 A schematic diagram of the relationship between data input and output in a linear regression model provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the technical solutions and advantages in the embodiments of the present application more clearly understood, the exemplary embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than an exhaustive list of all the embodiments. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0062] In the process of implementing this application, the inventor found that a large amount of data is generated during the operation of the train, including operation log data, various signal data, equipment data, etc., which are divided into structured, semi-structured and unstructured data in terms of data structure. These data are not only messy but also very tedious and complicated to sort out. If these data are directly analyzed and mined, there are certain limitations.

[0063] In response to the above problems, an embodiment of the present application provides a method for improving the train turnaround capacity, an electronic device, and a storage medium, the method comprising: establishing a business model based on the train turnaround business; based on the business model, using dimensional modeling to construct a data bus matrix and a physical model; performing ETL development according to the physical model to obtain the final train turnaround data; analyzing the final train turnaround data to obtain the factors that affect the train turnaround; mining the factors to determine a plan to improve the train turnaround capacity. The method provided in the present application is based on the train turnaround business, constructing a physical model, and performing extraction-conversion-loading development according to the physical model to obtain the final train turnaround data, and then based on the final train turnaround data, obtaining the factors that affect the train turnaround, mining the factors, determining a plan to improve the train turnaround capacity, and realizing the improvement of the urban rail operation capacity.

[0064] See also Figure 1 , a method for improving the train turnaround capacity provided in this embodiment, the implementation process is as follows:

[0065] 101. Establish a business model based on the train return business.

[0066] Figure 2 A schematic diagram of a train turnaround based on Figure 2 , the specific train return business can be obtained as follows:

[0067] Operation 1: The train enters the starting platform of the terminal station and stops.

[0068] Service 2: After the train stops, passengers get off the train and turn into the route at the same time.

[0069] Service 3: After all passengers have gotten off the train and the reversing route has been completed, the train enters the reversing track and travels to the reversing stop.

[0070] Service 4: The train enters the turning point, stops completely, changes ends, and simultaneously handles the exit route.

[0071] Service 5: The train has completed the end change and the folding route, and the train is heading to the terminal platform.

[0072] Service 6: The train enters the terminal platform and stops steadily. Passengers board the train and the route is processed at the same time.

[0073] Service 7: Passengers have boarded the train and the entry procedures have been completed. At this time, the train leaves the terminal platform.

[0074] During the actual turning process of the vehicle, the above seven tasks are performed in sequence.

[0075] The above are the detailed steps of the whole process of train turnaround from entering the starting platform of the terminal station to leaving the terminating platform of the terminal station. Through the above business, business modeling can be realized to obtain the business model.

[0076] The review of the train turnaround business clarified the specific content of the train turnaround business, improved the matching degree between the physical model and the train turnaround business, ensured the accuracy of the factors affecting the train turnaround, ensured the feasibility and improvement effect of the train turnaround capacity improvement plan, and realized the improvement of the urban rail operation capacity.

[0077] 102. Based on the business model, dimensional modeling is used to build the data bus matrix and the physical model.

[0078] 1. Based on the business model, the implementation of building a data bus matrix using dimensional modeling can adopt the following solutions:

[0079] 1) Select a business model based on the priority of the train return business.

[0080] 2) Declare the granularity of the fact table for the selected business model.

[0081] 3) Determine dimensions and facts based on the selected business model and granularity.

[0082] The dimensions include but are not limited to one or more of the following: time, platform, route, train, link, station, logical section, axle counting section, etc.

[0083] Since the dimensions are all related to the business, the dimensions are used to improve the correlation between the data bus matrix and the business, as well as the matching degree between the data bus matrix and the train return business, thereby ensuring the feasibility and improvement effect of the train return capacity improvement plan and realizing the improvement of the urban rail operation capacity.

[0084] 4) Build a data bus matrix based on the business models, dimensions, and facts selected under each data domain.

[0085] Through the above process, the construction of the data bus matrix is ​​completed, the correlation between the data bus matrix and the business is ensured, the matching degree between the data bus matrix and the train return business is improved, the feasibility and improvement effect of the train return capacity improvement plan are ensured, and the improvement of the urban rail operation capacity is achieved.

[0086] This step uses dimensional modeling to construct the data bus matrix based on the characteristics of easy understanding, high query performance, strong symmetry and scalability.

[0087] Dimensional modeling is a logical design method for structuring data, which divides the objective world into measures and contexts. Measures are often in the form of numerical values, and facts are surrounded by contexts. This context is intuitively divided into independent logical blocks, called dimensions.

[0088] In the specific implementation, the process of dimensional modeling method is as follows Figure 3 As shown,

[0089] 1) Business selection process: This process is the process of selecting the business according to the priority of the business process, that is, the business model will be selected according to the priority of the train return business.

[0090] 2) Declaration granularity: mainly declares the granularity of the fact table, that is, declares the granularity of the fact table for the selected business model.

[0091] Generally, the particle size is selected to be fine enough. In specific implementations, atomic-level particle size is more appropriate.

[0092] 3) Determine the dimensions: Based on the business process and granularity, you need to select relevant dimensions, that is, determine the dimensions based on the selected business model and granularity.

[0093] Among them, dimensions include time, platform, route, train, link, station, logical section, axle counting section, etc.

[0094] 4) Determine the facts: Select the facts applicable to the business process, that is, determine the facts based on the selected business model and granularity.

[0095] Among them, facts include train operation facts, operation plan facts, etc.

[0096] After clarifying the business processes under the data domain and determining the dimensions, you can then construct the bus matrix, that is, construct the data bus matrix based on the business models, dimensions, and facts selected under each data domain.

[0097] When building, first clarify which dimensions the business process is related to, and define the business processes and dimensions under each data domain.

[0098] Table 1 shows the data bus matrix of the train operation section, including dimensions such as purchase time (year, quarter, month, day, hour, minute, second), location (city, line, centralized station, station, platform, axle counting section, logical section), equipment type (signal, turnout, approach, screen door, emergency closing button), and related personnel. Facts such as train departure, train arrival, train departure, and train return.

[0099] Table 1

[0100]

[0101]

[0102] 2. The constructed physical model includes: operational data layer (ODS), dimension layer (DIM), data standard layer (DSL), unified data warehouse layer (DW), label data layer (LABLE) and application layer (ADS).

[0103] Unified data warehouse layer, including: detail layer (DWD, Data Warehouse Detail) and light-grained summary layer (DWS, Data Warehouse Service).

[0104] The source data layer is used to store the original data.

[0105] Dimension layer, used to store dimensions.

[0106] The data standard layer is used to store the data processed by the source data layer.

[0107] Unified data warehouse layer, used to store final data.

[0108] The label data layer is used to store the labels of the final data.

[0109] The application layer is used to provide application services.

[0110] Through the above physical model, we only need to build a unified set of standards based on the original layer data. The standards for data production downstream remain unchanged, which is used to solve the problem of inconsistent data formats of different manufacturers, thereby enhancing the universal capabilities of this architecture.

[0111] In specific applications, the physical model constructed must include the above 6 layers. If you want to store temporary data, the physical model can also include a temporary data layer (TMP, Temporary). Figure 4 The physical model shown is divided into 7 layers as a whole, including the source data layer (mainly used to store original data tables), the temporary data layer (used to store temporary tables), the dimension layer (created dimension tables), the data standard layer, the unified data warehouse layer (including the detail layer and the light-grained summary layer), the label data layer, and the application layer.

[0112] Among them, the dimension layer and temporary data layer run through the entire vertical data warehouse layering, while other model layers are built from the bottom up. At the same time, data is loaded from the bottom up, and top-down references are not allowed.

[0113] Figure 4 The physical model architecture shown is different from the traditional data warehouse layered architecture and introduces a data standard layer and a label data layer.

[0114] Considering the inconsistency of original data formats generated by different manufacturers, a data standard layer is introduced based on the original data layer. The significance of the data standard layer is that in this physical model architecture, only a set of unified standards need to be built based on the data in the source data layer. The standards for data production downstream remain unchanged. Only unified processing is performed at this layer to solve the problem of inconsistent data formats of different manufacturers, thereby enhancing the universal capabilities of this architecture.

[0115] The significance of the label data layer is that you only need to establish a unified label system in advance. For the data required by the application layer, you only need to combine labels to meet most business needs. However, due to the fast-changing and diverse characteristics of traditional data warehouses, in order to quickly meet the needs, you have to continuously expand the table and constantly change the model to meet the business needs. The application of the label data layer can solve this problem well.

[0116] An exemplary physical table structure entering the light-grained summary layer is shown in Table 2, where each return in the table is a record.

[0117] Table 2

[0118]

[0119]

[0120] An exemplary physical table structure after the light-grained summary layer is summarized is shown in Table 3, where each return in the table is a record.

[0121] Table 3

[0122]

[0123]

[0124] 103. Based on the physical model, perform ETL (Extract-Transform-Load) development to obtain the final train return data.

[0125] In this step, the specific implementation process of ETL development based on the physical model is as follows:

[0126] 1. Extract the initial train return data and store it in the source data layer.

[0127] The initial train return data may be data in the driving log.

[0128] 2. Convert the initial train return data and store the conversion results in the data standard layer.

[0129] 3. Load the dimension layer.

[0130] 4. According to the dimension layer, the data in the data standard layer is processed to obtain the final train return data.

[0131] 5. Store the final train return data in the unified data warehouse layer, and build the label data layer and application layer based on the final train return data.

[0132] In the process of building the physical model, the multi-dimensional model building, the model layering building, and the entity table building have been completed. Next, ETL development can be carried out. First, load the original layer data, then process the original data to obtain the data standard layer, then load the dimension layer, and then build the unified data warehouse layer -> label data layer -> application layer step by step. During the loading process, you need to pay attention to the order of data loading. If there is a reverse pointing, it violates the bottom-up data loading process, and the model needs to be adjusted. The optimized model can meet the online conditions until there is no reverse pointing problem, and then the task online scheduling can be carried out.

[0133] Through the above-mentioned ETL development solution based on the physical model, the final train return data format obtained after development can be consistent, making subsequent mining and analysis more meaningful, ensuring the feasibility and improvement effect of the train return capacity improvement plan, and realizing the improvement of urban rail operation capacity.

[0134] 104. Analyze the final train turnaround data to obtain factors that affect the train turnaround.

[0135] In this step, based on the final train reversal data, the average speed of the train on the reversing track is calculated, the historical reversal data is compared, and the reversal situations in different time periods are compared. The factors affecting the train reversal are: the stop time at the starting platform, the time for turning in the route, the time for running in the turning section, the average speed of the turning section, the time for changing ends of the train, the time for turning out the route, the time for running in the turning section, the average speed of the turning section and the stop time at the ending platform.

[0136] This step can be realized by ETL job processing to analyze and calculate the data of the vehicle reversing process, including the calculation of the average speed of the train when turning in and out of the reversing track, the comparison of historical reversing data, the comparison of reversing conditions in different periods, etc., and then obtain the following key factors affecting the train reversing:

[0137] 1. Stop time at the starting platform (arrival time and departure time);

[0138] 2. The time taken to process the route is included (the time from the start of route processing to the opening of the route);

[0139] 3. Travel time in the turnaround section (time from departure at the starting station to arrival at the turnaround track);

[0140] 4. Average speed in the reversing section (the average speed of the train from the platform to the stop point on the reversing track);

[0141] 5. Train end-changing time (the time from the start of the train end-changing to the completion of the end-changing);

[0142] 6. The time for processing the access route is calculated (the time from the start of processing the access route to the time when the access route is opened);

[0143] 7. Travel time in the turnaround section (from departure on the turnaround track to arrival at the terminal station);

[0144] 8. Average speed of the turnaround section (average speed from the turnaround track stop to the terminal platform);

[0145] 9. End the platform stop time (from arrival time to departure time).

[0146] During specific implementation, other factors that may affect the train reversal may also be obtained. This embodiment does not limit the specific factors that affect the train reversal.

[0147] The above factors affecting train turnaround are obtained based on the final train turnaround data, and the final train turnaround data is pre-processed based on the business model. Therefore, the factors affecting train turnaround are strongly related to the business model. Through the relationship between the factors affecting train turnaround and the business model, the feasibility and improvement effect of the train turnaround capacity improvement plan are guaranteed, thereby achieving an improvement in the urban rail operation capacity.

[0148] 105. Mining factors to determine the solution to improve the train turnaround capacity.

[0149] The linear regression algorithm is a common algorithm in machine learning and is widely used in practical applications. The linear regression algorithm can be used to train the model for train reversal, and then simulate and deduce the relationship between the train reversal capacity and various factors of the train reversal. The train reversal capacity can be improved by adjusting the train reversal factors.

[0150] The deduction (i.e. prediction) of train reversal capability is a regression problem. Analysis shows that there are multiple factors (such as 9) that affect train reversal, so it is a multivariate regression problem.

[0151] Therefore, in this step, the factors can be mined based on the pre-trained linear regression model to obtain the adjusted values ​​of the factors, and then the train turnaround capacity improvement plan can be determined based on the adjusted values.

[0152] If the adjustment values ​​are all used as a solution to improve the train's turnaround capacity

[0153] Among them, the relationship between data input and output in the linear regression model can be as follows Figure 5 shown.

[0154] The linear regression model is:

[0155]

[0156] i is the factor identifier, n is the total number of factors, x i is the data corresponding to the i-th factor, θ i is the adjustment value corresponding to the i-th factor.

[0157] The train turnaround simulation is actually to train the model through a large amount of training data, and finally get the prediction function y=h θ (x). The train return factor is used as a variable for data input, and then the prediction function h is obtained through learning θ (x) The output variable is the result.

[0158] x 0 ,x 1 ,x 2 ,…,x n is the input, h θ (x) is the output, and the parameter (i.e., the adjustment value corresponding to the factor) θ can be obtained by calculation. 0 ,θ 1 ,θ 2 ,…,θ n The value of is the solution of the prediction function. According to the function curve, we can guide the appropriate adjustment of relevant parameters, conduct simulation verification, and then propose improvement plans as a reference.

[0159] By mining the factors and determining the plan to improve the train's turning capacity, it is more conducive to improving the train's turning capacity, increasing the train's carrying capacity and thus improving the operational efficiency of rail transit.

[0160] The method provided in this embodiment first analyzes the entire link of the train return business, and then designs the model based on dimensional modeling. It is more meaningful to analyze and mine the data after ETL processing.

[0161] This method processes the train's driving log data based on data modeling analysis and data mining technology to improve the train's turnaround capability. This method can improve the train's turnaround capability, thereby improving the urban rail operating capacity.

[0162] This method can solve the following problems:

[0163] 1. Solve the problem of classifying heterogeneous data in a specified manner, using multiple collection methods, and finally storing them uniformly in a data warehouse to form complete standard original layer data.

[0164] 2. First, complete business modeling by analyzing the entire process of train turnaround, design consistent facts and dimensions based on dimensional modeling, build a data bus matrix, and finally build a physical model.

[0165] 3. Through the constructed physical model, ETL development is carried out to obtain the required result data. According to the obtained data, the key factors affecting the train reversal are analyzed. By mining and analyzing the train reversal factor data, a train reversal improvement plan is proposed to improve the train reversal capacity.

[0166] The method provided in this embodiment is based on the train return business, builds a physical model, and according to the physical model, performs extraction-conversion-loading development to obtain the final train return data, and then based on the final train return data, obtains the factors affecting the train return, mines the factors, determines the train return capacity improvement plan, and realizes the improvement of the urban rail operation capacity.

[0167] Based on the same inventive concept of the method for improving the train return capability, this embodiment provides an electronic device, which includes: a memory, a processor, and a computer program.

[0168] The computer program is stored in the memory and is configured to be executed by the processor to implement the above-mentioned method for improving the train return capacity.

[0169] Specifically,

[0170] Establish a business model based on the train return business;

[0171] Based on the business model, dimensional modeling is used to build a data bus matrix and a physical model;

[0172] According to the physical model, extract-convert-load development is carried out to obtain the final train return data;

[0173] Analyze the final train turnaround data to obtain the factors that affect the train turnaround;

[0174] Mining factors and determining solutions to improve train turnaround capacity.

[0175] Optionally, based on the business model, dimensional modeling is used to construct a data bus matrix, including:

[0176] Select the service model based on the priority of the train return service;

[0177] Declare the granularity of the fact table for the chosen business model;

[0178] Determine dimensions and facts based on the selected business model and granularity;

[0179] Build a data bus matrix based on the business model, dimensions, and facts selected under each data domain.

[0180] Optionally, the physical model includes: a source data layer, a dimension layer, a data standard layer, a unified data warehouse layer, a label data layer and an application layer;

[0181] The source data layer is used to store the original data;

[0182] Dimension layer, used to store dimensions;

[0183] The data standard layer is used to store the data processed by the source data layer;

[0184] Unified data warehouse layer, used to store final data;

[0185] The label data layer is used to store the labels of the final data;

[0186] The application layer is used to provide application services.

[0187] Optionally, perform extract-transform-load development based on the physical model, including:

[0188] Extract the initial train return data and store the initial train return data in the source data layer;

[0189] Convert the initial train return data and store the conversion result in the data standard layer;

[0190] Load the dimension layer;

[0191] According to the dimension layer, the data in the data standard layer is processed to obtain the final train return data;

[0192] The final train return data is stored in the unified data warehouse layer, and the label data layer and application layer are constructed based on the final train return data.

[0193] Optionally, analyze the final train turnaround data, including:

[0194] Based on the final train reversal data, calculate the average speed of the train when turning in and out of the reversal track, compare historical reversal data, and compare reversal situations in different periods;

[0195] Optionally, the factors affecting the train's reversal are one or more of the following: the stop time at the starting platform, the time for handling the turning route, the time for traveling in the turning section, the average speed in the turning section, the time for changing ends of the train, the time for handling the turning route, the time for traveling in the turning section, the average speed in the turning section, and the stop time at the ending platform.

[0196] Optionally, factors are mined to determine solutions to improve train turnaround capacity, including:

[0197] Mining the factors based on the pre-trained linear regression model to obtain the adjusted values ​​of the factors;

[0198] Determine the plan to improve the train's turnaround capacity based on the adjustment value.

[0199] Optionally, the linear regression model is:

[0200]

[0201] Among them, i is the factor identifier, n is the total number of factors, and x i is the data corresponding to the i-th factor, θ i is the adjustment value corresponding to the i-th factor.

[0202] The electronic device provided in this embodiment has a computer program executed by a processor to construct a physical model based on the train return business, and extract-convert-load development according to the physical model to obtain the final train return data, and then obtain the factors affecting the train return based on the final train return data, and mine the factors to determine the train return capacity improvement plan, so as to achieve the improvement of the urban rail operation capacity.

[0203] Based on the same inventive concept of the method for improving the train reversing capability, this embodiment provides a computer on which a computer program can be stored. The computer program is executed by a processor to implement the method for improving the train reversing capability.

[0204] Specifically,

[0205] Establish a business model based on the train return business;

[0206] Based on the business model, dimensional modeling is used to build a data bus matrix and a physical model;

[0207] According to the physical model, extract-convert-load development is carried out to obtain the final train return data;

[0208] Analyze the final train turnaround data to obtain the factors that affect the train turnaround;

[0209] Mining factors and determining solutions to improve train turnaround capacity.

[0210] Optionally, based on the business model, dimensional modeling is used to construct a data bus matrix, including:

[0211] Select the service model based on the priority of the train return service;

[0212] Declare the granularity of the fact table for the chosen business model;

[0213] Determine dimensions and facts based on the selected business model and granularity;

[0214] Build a data bus matrix based on the business model, dimensions, and facts selected under each data domain.

[0215] Optionally, the physical model includes: a source data layer, a dimension layer, a data standard layer, a unified data warehouse layer, a label data layer and an application layer;

[0216] The source data layer is used to store the original data;

[0217] Dimension layer, used to store dimensions;

[0218] The data standard layer is used to store the data processed by the source data layer;

[0219] Unified data warehouse layer, used to store final data;

[0220] The label data layer is used to store the labels of the final data;

[0221] The application layer is used to provide application services.

[0222] Optionally, perform Extract-Transform-Load (ETL) development based on the physical model, including:

[0223] Extract the initial train return data and store the initial train return data in the source data layer;

[0224] Convert the initial train return data and store the conversion result in the data standard layer;

[0225] Load the dimension layer;

[0226] According to the dimension layer, the data in the data standard layer is processed to obtain the final train return data;

[0227] The final train return data is stored in the unified data warehouse layer, and the label data layer and application layer are constructed based on the final train return data.

[0228] Optionally, the final train reversal data is analyzed to obtain factors affecting the train reversal, including:

[0229] Based on the final train reversal data, calculate the average speed of the train when turning in and out of the reversal track, compare historical reversal data, and compare reversal situations in different periods;

[0230] Optionally, the factors affecting the train's reversal are one or more of the following: the stop time at the starting platform, the time for handling the turning route, the time for traveling in the turning section, the average speed in the turning section, the time for changing ends of the train, the time for handling the turning route, the time for traveling in the turning section, the average speed in the turning section, and the stop time at the ending platform.

[0231] Optionally, factors are mined to determine solutions to improve train turnaround capacity, including:

[0232] Mining the factors based on the pre-trained linear regression model to obtain the adjusted values ​​of the factors;

[0233] Determine the plan to improve the train's turnaround capacity based on the adjustment value.

[0234] Optionally, the linear regression model is:

[0235]

[0236] Among them, i is the factor identifier, n is the total number of factors, and x i is the data corresponding to the i-th factor, θ i is the adjustment value corresponding to the i-th factor.

[0237] The computer-readable storage medium provided in this embodiment has a computer program executed by a processor to construct a physical model based on the train return business, and perform extraction-conversion-loading development according to the physical model to obtain the final train return data, and then obtain the factors affecting the train return based on the final train return data, and mine the factors to determine the train return capacity improvement plan, so as to improve the urban rail operation capacity.

[0238] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0239] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0240] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0241] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0242] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0243] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for improving train turnaround capacity, characterized in that: The method comprises: Establish a business model based on the train return business; Based on the business model, dimensional modeling is used to construct a data bus matrix and a physical model; According to the physical model, extract-convert-load development is performed to obtain the final train return data; Analyze the final train reversal data to obtain factors affecting the train reversal; Mining the factors and determining the plan to improve the train turnaround capacity; The method of constructing a data bus matrix based on the business model by using dimensional modeling includes: Select the service model based on the priority of the train return service; Declare the granularity of the fact table for the chosen business model; Determine dimensions and facts based on the selected business model and granularity; Build a data bus matrix based on the business model, dimensions and facts selected under each data domain; The physical model includes: a source data layer, a dimension layer, a data standard layer, a unified data warehouse layer, a label data layer and an application layer; The source data layer is used to store original data; The dimension layer is used to store dimensions; The data standard layer is used to store the processed data of the source data layer; The unified data warehouse layer is used to store the final data; The label data layer is used to store the label of the final data; The application layer is used to provide application services; The extract-transform-load development is performed according to the physical model, including: Extracting initial train return data, and storing the initial train return data in the source data layer; Convert the initial train return data and store the conversion result in the data standard layer; Load the dimension layer; According to the dimension layer, the data in the data standard layer is processed to obtain the final train return data; The final train return data is stored in the unified data warehouse layer, and the label data layer and application layer are constructed based on the final train return data; The mining of the factors to determine a solution for improving the train turnaround capacity includes: Mining the factor based on a pre-trained linear regression model to obtain an adjusted value of the factor; Determine a plan to improve the train's turnaround capacity according to the adjustment value; The linear regression model is: Among them, i is the factor identifier, n is the total number of factors, and x i is the data corresponding to the i-th factor, θ i is the adjustment value corresponding to the i-th factor.

2. The method according to claim 1, characterized in that The analyzing the final train return data includes: According to the final train reversal data, the average speed of the train when turning in and out of the reversal track is calculated, the historical reversal data is compared, and the reversal situations in different time periods are compared.

3. The method according to claim 1, characterized in that The factors that affect the train's reversal are one or more of the following: the stop time at the starting platform, the time for handling the turn-in route, the time for traveling in the turn-in section, the average speed in the turn-in section, the time for train changeover, the time for handling the turn-out route, the time for traveling in the turn-out section, the average speed in the turn-out section, and the stop time at the terminating platform.

4. An electronic device, characterized in that: include: Memory; processor; as well as Computer programs; The computer program is stored in the memory and is configured to be executed by the processor to implement the method according to any one of claims 1 to 3.

5. A computer-readable storage medium, characterized in that: A computer program is stored thereon; the computer program is executed by a processor to implement the method according to any one of claims 1 to 3.

Images