Railway passenger transport timetable generation method and system based on collaborative scheduling

Abstract

The invention provides a railway passenger transport timetable generation method and system based on collaborative scheduling. The method comprises the following steps: constructing a local heat matrix according to passenger flow heat of other stations corresponding to a local area where a current station is located; an initial passenger transport timetable of the current station is generated through popularity learning features of the local popularity matrix; train operation scheduling is carried out based on the initial passenger transport time table, the in-station scheduling coupling degree of the current station is determined through the passenger flow popularity and the passenger transport adjustment proportion of the current station, and the inter-station scheduling coupling degree is determined based on the local popularity matrix and the popularity contribution ratio; and determining a dynamic scheduling sensitive factor corresponding to the current station according to the in-station scheduling coupling degree and the inter-station scheduling coupling degree, and carrying out dynamic feedback updating on the initial passenger transport time table through the dynamic scheduling sensitive factor, so that the train time table can be dynamically fed back and updated according to the passenger flow coupling condition of the local area where the station is located after the station is scheduled. And the generation efficiency of the passenger transport adjustment scheme and the passenger load factor matching degree are improved.

Description

Technical Field

[0001] The present application relates to the field of passenger transport and dispatching technology, and more specifically, to a method and system for generating a railway passenger timetable based on collaborative dispatching. Background Art

[0002] The deployment of passenger train capacity needs to adapt to changes in railway market demand. During peak seasons such as the Spring Festival, summer travel, and holidays, the market demand is abundant, and railway companies deploy capacity under the constraints of line capacity, vehicle resources, and other conditions to meet passenger flow demand; during the off-season of passenger flow demand, railway companies need to formulate accurate capacity deployment strategies based on future passenger flow conditions in order to adjust capacity in a timely, fast, and convenient manner. Generally, strategies such as adding / stopping trains and adding / reducing trains are adopted to adapt to off-season market demand and reduce operating costs. With the continuous development of the scale of my country's high-speed railway network, higher requirements are placed on the efficiency and benefits of daily passenger adjustment plans.

[0003] In the existing passenger adjustment plans, the adjustment of daily capacity plans mainly relies on manual experience, which is not scientific and innovative enough, lacks intelligent decision-making support, and cannot pre-evaluate and evaluate multiple possible plans; the ticket quota adjustment and capacity plan are adjusted separately at different stages, and the two lack mutual feedback and linkage. In actual operation, the plans are mostly transmitted in electronic documents and paper documents. A complete system has not yet been established to support the efficient development of daily operation organization adjustment business. Therefore, how to improve the generation efficiency of passenger adjustment plans and the matching degree of passenger load factor has become a problem that needs to be solved urgently. Summary of the invention

[0004] The present application provides a method and system for generating a railway passenger timetable based on collaborative scheduling, which can dynamically feedback and update the train timetable according to the passenger flow coupling situation in the local area where the station is located after scheduling, thereby improving the generation efficiency of passenger adjustment plans and the occupancy rate matching degree.

[0005] In a first aspect, the present application provides a method for generating a railway passenger timetable based on collaborative scheduling. The method can be executed by a network device, or can also be executed by a chip configured in the network device, and the present application does not limit this.

[0006] Specifically, the method includes: Obtain the passenger flow information of the current station, conduct passenger flow heat learning based on the passenger flow information of the current station, obtain the passenger flow heat of the current station, determine the local area where the current station is located, and construct a local heat matrix based on the passenger flow heat of other stations corresponding to the local area where the current station is located; Performing eigenvalue decomposition on the local heat matrix to obtain heat learning features corresponding to the local area where the current station is located, and generating an initial passenger timetable for the current station through the heat learning features; Perform train operation dispatching based on the initial passenger transport schedule, determine the in-station dispatching coupling degree of the current station through the passenger flow heat degree and the passenger transport adjustment ratio of the current station, and determine the inter-station dispatching coupling degree based on the local heat matrix and the heat contribution ratio corresponding to the local area where the current station is located; Determine the dynamic dispatching sensitivity factor corresponding to the current station according to the in-station dispatching coupling degree and the inter-station dispatching coupling degree, and perform dynamic feedback update on the initial passenger transport schedule through the dynamic dispatching sensitivity factor.

[0007] Combined with the first aspect, in some implementation manners of the first aspect, generating the initial passenger transport schedule of the current station through the heat learning feature specifically includes: Obtain a plurality of heat learning feature samples and the passenger transport time adjustment table templates respectively corresponding to them, and the passenger transport time adjustment table templates are formulated through manual experience; Perform mean clustering on the plurality of heat learning feature samples and the heat learning feature, and generate the initial passenger transport schedule of the current station based on the passenger transport time adjustment table templates respectively corresponding to the heat learning features in the clustering result.

[0008] Combined with the first aspect, in some implementation manners of the first aspect, determining the inter-station dispatching coupling degree based on the local heat matrix and the heat contribution ratio corresponding to the local area where the current station is located specifically includes: Obtain a preset windowing period, extract the local heat matrices corresponding to the local area where the current station is located in different windowing periods after train operation dispatching, and perform coupling feature extraction on the local heat matrices respectively corresponding to each windowing period to obtain the heat coupling features respectively corresponding to each windowing period; Obtain the heat contribution ratios corresponding to the current station in different windowing periods, and determine the inter-station dispatching coupling degree corresponding to the current station based on the heat contribution ratios and the heat coupling features of the current station in different windowing periods.

[0009] Combined with the first aspect, in some implementation manners of the first aspect, determining the dynamic dispatching sensitivity factor corresponding to the current station according to the in-station dispatching coupling degree and the inter-station dispatching coupling degree specifically includes: Obtain the dynamic sensitivity prediction interval, and obtain the in-station prediction weight and the inter-station prediction weight; Based on the in-station prediction weight and the inter-station prediction weight, perform weighted fusion on the in-station dispatching coupling degree and the inter-station dispatching coupling degree within each dynamic sensitivity prediction interval to obtain the station passenger flow coupling degrees corresponding to different interval times; Perform forward correlation prediction on the station passenger flow coupling degrees corresponding to different interval times, and use the station passenger flow coupling degree within the next interval time as the dynamic dispatching sensitivity factor corresponding to the current station.

[0010] Combined with the first aspect, in some implementations of the first aspect, the dynamic feedback update of the initial passenger train timetable by the dynamic scheduling sensitivity factor specifically includes: based on the threshold interval where the dynamic scheduling sensitivity factor is located, determining the corresponding dynamic feedback period, collecting in real time the heat learning features corresponding to the local area where the current station is located by the dynamic feedback period, and generating an updated passenger train timetable according to the heat learning features and the dynamic scheduling sensitivity factor.

[0011] Combined with the first aspect, in some implementations of the first aspect, the passenger flow information of the current station specifically includes: historical ticket sales data, entrance and exit gate records, and train operation records.

[0012] Combined with the first aspect, in some implementations of the first aspect, the passenger flow heat learning through the passenger flow information of the current station to obtain the passenger flow heat of the current station specifically includes: using a single-hidden-layer neural network to learn and cluster the passenger flow information of the current station, and taking the heat value corresponding to the clustering result as the passenger flow heat corresponding to the current station.

[0013] In a second aspect, the present application provides a railway passenger train timetable generation system based on collaborative scheduling, which includes a timetable generation unit, and the timetable generation unit includes: A station data recording unit, configured to obtain the passenger flow information of the current station, perform passenger flow heat learning through the passenger flow information of the current station to obtain the passenger flow heat of the current station, determine the local area where the current station is located, and construct a local heat matrix according to the passenger flow heat of other stations corresponding to the local area where the current station is located; A first timetable generation module, configured to perform eigenvalue decomposition on the local heat matrix to obtain the heat learning features corresponding to the local area where the current station is located, and generate an initial passenger train timetable for the current station through the heat learning features; A second timetable generation module, configured to perform train operation scheduling based on the initial passenger train timetable, determine the in-station scheduling coupling degree of the current station through the passenger flow heat and the passenger transport adjustment ratio of the current station, and determine the inter-station scheduling coupling degree based on the local heat matrix and the heat contribution ratio corresponding to the local area where the current station is located; A second timetable generation module, configured to determine the dynamic scheduling sensitivity factor corresponding to the current station according to the in-station scheduling coupling degree and the inter-station scheduling coupling degree, and perform dynamic feedback update on the initial passenger train timetable through the dynamic scheduling sensitivity factor.

[0014] In a third aspect, the present application provides a computer terminal device, which includes a memory and a processor. The memory stores code, and the processor is configured to obtain the code and execute the above-mentioned method for generating a railway passenger transport timetable based on collaborative scheduling.

[0015] In a fourth aspect, the present application provides a computer-readable storage medium that stores at least one computer program, and the computer program is loaded and executed by a processor to implement the operations performed by the above-mentioned method for generating a railway passenger transport timetable based on collaborative scheduling.

[0016] The technical solutions provided by the disclosed embodiments of the present application have the following beneficial effects: In the method and system for generating a railway passenger transport timetable based on collaborative scheduling provided by the present application, first, the passenger flow information of the current station is obtained, the passenger flow heat is learned through the passenger flow information of the current station to obtain the passenger flow heat of the current station, the local area where the current station is located is determined, and a local heat matrix is constructed according to the passenger flow heat of other stations corresponding to the local area where the current station is located; the local heat matrix is subjected to eigenvalue decomposition to obtain the heat learning feature corresponding to the local area where the current station is located, and an initial passenger transport timetable for the current station is generated through the heat learning feature; train operation scheduling is performed based on the initial passenger transport timetable, and the in-station scheduling coupling degree of the current station is determined through the passenger flow heat and the passenger transport adjustment ratio of the current station, and the inter-station scheduling coupling degree is determined based on the local heat matrix and the heat contribution ratio corresponding to the local area where the current station is located; the dynamic scheduling sensitivity factor corresponding to the current station is determined according to the in-station scheduling coupling degree and the inter-station scheduling coupling degree, and the initial passenger transport timetable is dynamically feedback updated through the dynamic scheduling sensitivity factor.

[0017] Therefore, it can be seen that in the present application, by constructing the heat matrix of the local area where the current station is located and performing eigenvalue decomposition, the main component mode of the passenger flow change in the area is extracted as the heat learning feature, so that the initial timetable generation stage has regional characteristics, avoiding the fragmentation of scheduling between stations, and measuring the impact of scheduling on the passenger flow heat of the station itself through the in-station scheduling coupling degree, and measuring the linkage impact of the station scheduling on other stations in the area through the inter-station scheduling coupling degree, realizing the real-time closed-loop of scheduling behavior → passenger flow change → model perception, improving the local and global response effects of the dynamic perception scheduling strategy, and being able to generate an optimizable scheme faster than the manual experience-based static adjustment method, improving the generation efficiency of the passenger transport adjustment plan.

[0018] Thus, it is possible to dynamically feedback update the train timetable according to the passenger flow coupling situation of the local area where the station is located after scheduling, improving the generation efficiency of the passenger transport adjustment plan and the matching degree of the passenger occupancy rate. Description of the Drawings

[0019] Figure 1 It is an exemplary flowchart of a method for generating a railway passenger transport timetable based on collaborative scheduling shown in some embodiments of the present application; Figure 2 It is a schematic structural diagram of a timetable generation unit shown in some embodiments of the present application; Figure 3 It is a schematic structural diagram of a computer terminal device for implementing a method for generating a railway passenger transport timetable based on collaborative scheduling shown in some embodiments of the present application. Detailed implementation manners

[0020] The present application obtains the passenger flow information of the current station, conducts passenger flow heat learning through the passenger flow information of the current station to obtain the passenger flow heat of the current station, determines the local area where the current station is located, and constructs a local heat matrix according to the passenger flow heat of other stations corresponding to the local area where the current station is located; performs eigenvalue decomposition on the local heat matrix to obtain the heat learning features corresponding to the local area where the current station is located, and generates an initial passenger transport timetable for the current station through the heat learning features; conducts train operation scheduling based on the initial passenger transport timetable, and determines the in-station scheduling coupling degree of the current station through the passenger flow heat and the passenger transport adjustment ratio of the current station, and determines the inter-station scheduling coupling degree based on the local heat matrix and the heat contribution ratio corresponding to the local area where the current station is located; determines the dynamic scheduling sensitivity factor corresponding to the current station according to the in-station scheduling coupling degree and the inter-station scheduling coupling degree, and dynamically feedback updates the initial passenger transport timetable through the dynamic scheduling sensitivity factor, so as to be able to dynamically feedback update the train timetable according to the passenger flow coupling situation of the local area where the station is located after scheduling, improving the generation efficiency of the passenger transport adjustment plan and the matching degree of the passenger occupancy rate.

[0021] To better understand the above technical solution, the above technical solution will be described in detail below in combination with the accompanying drawings of the specification and specific implementation manners. Refer to Figure 1 , this figure is an exemplary flowchart of a method for generating a railway passenger transport timetable based on collaborative scheduling shown in some embodiments of the present application. The method 100 for generating a railway passenger transport timetable based on collaborative scheduling mainly includes the following steps: In step S101, obtain the passenger flow information of the current station, conduct passenger flow heat learning through the passenger flow information of the current station to obtain the passenger flow heat of the current station, determine the local area where the current station is located, and construct a local heat matrix according to the passenger flow heat of other stations corresponding to the local area where the current station is located.

[0022] Optionally, in some embodiments, the passenger flow information of the current station specifically includes: historical ticket sales data, in-out gate records, train operation records, holiday, weather, event tags, authorized real-time face recognition monitoring data.

[0023] It should be noted that the passenger flow heat described in this application is used to quantify the density of the total number of passengers entering and leaving the station per unit time and the intensity of the density change trend relative to the same type of riding environment, and is used to reflect the overall passenger flow busy situation of the station. Preferably, in some embodiments, the passenger flow heat of the current station is learned through the passenger flow information of the current station. The specific process of obtaining the passenger flow heat of the current station includes: using a single-hidden-layer neural network to learn and cluster the passenger flow information of the current station, and taking the heat value corresponding to the clustering result as the passenger flow heat corresponding to the current station.

[0024] When specifically implemented, during the process of using a single-hidden-layer neural network to learn and cluster the passenger flow information of the current station, the passenger flow information of the current station can be input in the form of a data vector. The hidden layer of the single-hidden-layer neural network contains multiple activation function nodes for classification training, and the classification result is output through the output layer of the single-hidden-layer neural network. According to the clustering result corresponding to the passenger flow information of the current station, a passenger flow heat score mapping is performed to obtain the corresponding heat value as the passenger flow heat corresponding to the current station. Among them, different dimensions of the data vector respectively include: the number of passengers entering the station on the current day, the number of passengers leaving the station on the current day, the current time period, the number of train trips, the special event flag, the weather information, and the historical same-period comparison ratio. Among them, a plurality of pre-prepared passenger flow information samples and the corresponding artificial passenger flow heat scores are used as the training set. The plurality of passenger flow information samples are input into the single-hidden-layer neural network for classification training. The hidden layer of the single-hidden-layer neural network contains multiple activation function nodes for classifying and training the input samples, and the classification result is output through the output layer of the single-hidden-layer neural network. Then, the classification result is compared with the corresponding artificial passenger flow heat score result. When the correlation between the clustering result and the artificial passenger flow heat score result is lower than the preset threshold, the activation parameters in the activation function in the hidden layer of the single-hidden-layer neural network are adjusted until the correlation between the clustering result and the artificial passenger flow heat score result reaches the preset standard, and the mapping relationship between the clustering result and the artificial passenger flow heat score is obtained.

[0025] It should be noted that in the railway passenger transport dispatching optimization system, the local area where the current station is located generally refers to the set of adjacent stations selected within a certain range with the current station as the core, and is used for in-region passenger flow heat analysis, resource coordination, and dispatching optimization. Preferably, in some embodiments, the provincial area where the current station is located can be used as the local area where the current station is located. When constructing a local heat matrix according to the passenger flow heat of other stations corresponding to the local area where the current station is located, the passenger flow heat of each station corresponding to the local area where the current station is located is formed into a column vector according to the label order, and the column vectors corresponding to different time periods within the preset historical period form the local heat matrix.

[0026] In step S102, eigenvalue decomposition is performed on the local heat matrix to obtain the heat learning features corresponding to the local area where the current station is located, and an initial passenger train schedule for the current station is generated through the heat learning features.

[0027] Optionally, in some embodiments, the principal component analysis method can be used to perform eigenvalue decomposition on the local heat matrix, and the eigenvector composed of the principal component eigenvalues corresponding to the local heat matrix is used as the heat learning features corresponding to the local area where the current station is located.

[0028] Optionally, in some embodiments, the heat learning features can be input into a pre-trained pressure scoring model, the passenger flow pressure score is output according to the scoring model, the daily train operation quantity of the current station is determined based on the passenger flow pressure score, multiple operable time periods are equally spaced according to the daily train operation quantity, and an initial passenger train schedule is generated according to all the operable time periods. During the process of generating the initial passenger train schedule according to the daily train operation quantity, the train operation time needs to satisfy the principle of non-conflicting train operations.

[0029] Optionally, in some embodiments, the initial passenger train schedule for the current station generated through the heat learning features can also be determined in the following manner: Obtain multiple heat learning feature samples and their respective corresponding passenger train time adjustment table templates, and the passenger train time adjustment table templates are formulated based on manual experience; Perform mean clustering on the multiple heat learning feature samples and the heat learning features, and generate an initial passenger train schedule for the current station based on the passenger train time adjustment table templates corresponding to the respective heat learning features in the clustering result.

[0030] It should be noted that through the research and analysis of the historical transport capacity deployment situation and the actual passenger departure and arrival data, it is found that during the transport capacity adjustment process within the research period, especially during the off-peak passenger flow period, there is a certain deviation between the transport capacity deployment and the expected passenger flow growth. There is an urgent need to analyze and study the transport capacity deployment strategies and implementation effects in different time periods. By statistically analyzing indicators such as the actual deployed transport capacity, the operation situation of EMUs, the occupancy rate of trains, and the operation benefits, etc., analyze and summarize the transport capacity deployment strategies, deployment effects, and existing problems adopted by different railway administrations. The analysis results show that the current transport capacity adjustment strategies mainly focus on the transport capacity deployment strategy at the train level and the batch deployment strategy at the ticket quota level. And through clustering to identify similar popularity characteristics, this application can automatically match the corresponding manual experience adjustment timetable template to generate the initial passenger transport timetable of the current station, improving the degree of automation and retaining the experience of dispatching experts. Optionally, in some embodiments, K-means mean clustering can be used to perform mean clustering on multiple popularity learning feature samples and the popularity learning feature. Each passenger transport time adjustment table template corresponds to multiple popularity learning feature samples, and then the passenger transport time adjustment table template with the largest corresponding number of all popularity learning feature samples in the clustering center where the popularity learning feature is located is used as the initial passenger transport timetable.

[0031] In step S103, based on the initial passenger transport timetable, train operation scheduling is performed, and the in-station scheduling coupling degree of the current station is determined through the passenger flow popularity and the passenger transport adjustment ratio of the current station. The inter-station scheduling coupling degree is determined based on the local popularity matrix and the popularity contribution ratio corresponding to the local area where the current station is located.

[0032] It should be noted that the in-station scheduling coupling degree in this application is used to quantify the sensitivity of the passenger flow popularity in the station to the change of the passenger transport plan scheduling. The in-station scheduling coupling degree is determined based on the change trends of the passenger transport adjustment ratio and the passenger flow popularity after train operation scheduling. Optionally, in some embodiments, determining the in-station scheduling coupling degree of the current station through the passenger flow popularity and the passenger transport adjustment ratio of the current station specifically includes: After train operation scheduling, obtain the change trend of the passenger flow popularity of the current station; Obtain the daily passenger transport adjustment ratio, and extract the passenger transport adjustment trend of the current station based on the daily passenger transport adjustment ratio; Extract the trend coupling degree from the passenger flow popularity change trend and the passenger transport adjustment trend to determine the in-station scheduling coupling degree of the current station.

[0033] It should be noted that the passenger transport adjustment ratio is the ratio between the total number of passenger trains departing from the current station today and the total number of passenger trains departing from the current station yesterday.

[0034] Optionally, in some embodiments, obtaining the change trend of the passenger flow popularity of the current station specifically includes: Record the passenger flow heat of the current station on different dates, and form a passenger flow heat sequence according to the time series; Extract the trend of the passenger flow heat sequence to obtain the passenger flow heat change trend corresponding to the current station; specifically, when implementing, the empirical mode decomposition method can be used to extract multiple intrinsic mode functions corresponding to the passenger flow heat sequence, and the mean value of the intrinsic mode functions is used as the passenger flow heat change trend corresponding to the current station.

[0035] Optionally, in some embodiments, extracting the passenger transport adjustment trend of the current station based on the daily passenger transport adjustment ratio specifically includes: Record the daily passenger transport adjustment ratio of the current station on different dates, and form a passenger transport adjustment ratio sequence according to the time series; Extract the trend of the passenger transport adjustment ratio sequence to obtain the passenger transport adjustment trend corresponding to the current station. Specifically, when implementing the process of extracting the trend of the passenger transport adjustment ratio sequence, the same extraction method as that of the passenger flow heat change trend can be used, and this application will not elaborate on this.

[0036] Preferably, in some embodiments, extracting the trend coupling degree from the passenger flow heat change trend and the passenger transport adjustment trend to determine the in-station dispatching coupling degree of the current station specifically includes: Based on a preset coupling degree extraction period, collect trend values of the passenger flow heat change trend and the passenger transport adjustment trend respectively to obtain a passenger flow heat trend value sequence and a passenger transport adjustment trend value sequence; After normalizing the passenger flow heat trend value sequence and the passenger transport adjustment trend value sequence, perform coupling degree analysis according to the passenger flow heat trend value sequence and the passenger transport adjustment trend value sequence to obtain the in-station dispatching coupling degree of the current station.

[0037] Specifically, when implementing, the coupling degree extraction period can be mapped based on the current passenger flow density. In some embodiments, the coupling degree extraction period can be fixed as a constant value of 6h, and then the Pearson correlation coefficient between the passenger flow heat trend value sequence and the passenger transport adjustment trend value sequence can be used as the in-station dispatching coupling degree.

[0038] It should be noted that in the process of determining the inter-station dispatching coupling degree based on the local heat matrix and heat contribution ratio corresponding to the local area where the current station is located, the local area heat is the sum of the passenger flow heats of all stations in the local area where the current station is located, and the heat contribution ratio corresponding to the current station is the ratio between the passenger flow heat of the current station and the local area heat.

[0039] It should be noted that the inter-station scheduling coupling degree described in this application is used to reflect the degree of coupling impact of the current station scheduling change on the passenger flow heat in the local area where the station is located. The inter-station scheduling coupling degree is determined based on the local heat matrix and heat contribution ratio in the local area after train operation scheduling. Optionally, in some embodiments, determining the inter-station scheduling coupling degree based on the local heat matrix and heat contribution ratio corresponding to the local area where the current station is located specifically includes: Obtain the preset windowing period, extract the local heat matrices corresponding to the local area where the current station is located in different windowing periods after train operation scheduling, and perform coupling feature extraction on the local heat matrices corresponding to each windowing period to obtain the heat coupling features corresponding to each windowing period; Obtain the heat contribution ratios corresponding to the current station in different windowing periods, and determine the inter-station scheduling coupling degree corresponding to the current station based on the heat contribution ratios and heat coupling features of the current station in different windowing periods.

[0040] In specific implementation, the windowing period is preset to 24h. Furthermore, the local heat matrices corresponding to the local area where the current station is located in different windowing periods can be respectively subjected to singular value decomposition, and the mean of the matrix eigenvalues corresponding to the local heat matrices corresponding to the local area where the current station is located in different windowing periods is used as the heat coupling feature corresponding to each windowing period. Preferably, in some embodiments, after normalizing the heat contribution ratios and heat coupling features of the current station in different windowing periods, the standard deviation of the heat contribution ratios and the heat coupling features is used as the inter-station scheduling coupling degree.

[0041] In step S104, determine the dynamic scheduling sensitivity factor corresponding to the current station according to the in-station scheduling coupling degree and the inter-station scheduling coupling degree, and perform dynamic feedback update on the initial passenger train timetable through the dynamic scheduling sensitivity factor.

[0042] It should be noted that the dynamic scheduling sensitivity factor described in this application is used to quantify the sensitivity of the passenger flow in the station and the local area to scheduling changes. When the dynamic scheduling sensitivity factor is relatively large, since the scheduling change is relatively large, the feedback for dynamically updating the passenger train timetable can be appropriately reduced, that is, the update speed of the passenger train timetable is reduced, so as to avoid the passenger flow disorder caused by the frequent update of the passenger train timetable, and improve the safety and comfort of train passengers during boarding. Preferably, in some embodiments, determining the dynamic scheduling sensitivity factor corresponding to the current station according to the in-station scheduling coupling degree and the inter-station scheduling coupling degree specifically includes: Obtain the dynamic sensitivity prediction interval, and obtain the in-station prediction weight and the inter-station prediction weight; Based on the in-station prediction weight and the inter-station prediction weight, the in-station scheduling coupling degree and the inter-station scheduling coupling degree within each dynamic sensitivity prediction interval are weighted and fused to obtain the station passenger flow coupling degrees corresponding to different interval times. Perform forward correlation prediction on the station passenger flow coupling degrees corresponding to different interval times, and use the station passenger flow coupling degree within the next interval time as the dynamic scheduling sensitivity factor corresponding to the current station.

[0043] In specific implementation, the in-station prediction weight and the inter-station prediction weight are calibrated as constants based on experience, and the station passenger flow coupling degree = the in-station prediction weight × the in-station scheduling coupling degree + the inter-station prediction weight × the inter-station scheduling coupling degree. In some embodiments, a moving average autoregressive model can be used to perform forward correlation prediction on the station passenger flow coupling degrees corresponding to different interval times, and the station passenger flow coupling degree within the next interval time is used as the dynamic scheduling sensitivity factor corresponding to the current station. Preferably, a preferred embodiment for obtaining the dynamic scheduling sensitivity factor in this application is given below: First, a prediction sample period of 15 days is preset, and the dynamic sensitivity prediction interval is 1 h. At this time, the station passenger flow coupling degrees within the past 360 h of the station can be recorded respectively to obtain a station passenger flow coupling degree sequence. In some other embodiments, the station passenger flow coupling degree can also be preset to other time lengths; furthermore, a time series graph of this station passenger flow coupling degree sequence can be drawn, and the abscissa of the time series graph corresponds to different moments. Furthermore, the time series graph of the station passenger flow coupling degree sequence set can be exponentially transformed to eliminate the trend of the variance changing with time in the time series graph.

[0044] Secondly, according to the time series graph of the station passenger flow coupling degree sequence, draw the autocorrelation coefficient graph of the station passenger flow coupling degree, where the horizontal axis of the autocorrelation coefficient graph is the lag period number, and the vertical axis is the value of the autocorrelation coefficient. Draw the partial autocorrelation coefficient graph of the station passenger flow coupling degree, where the horizontal axis of the partial autocorrelation coefficient graph is the lag period number, and the vertical axis is the value of the partial autocorrelation coefficient.

[0045] According to the characteristics of the autocorrelation coefficient graph and the partial autocorrelation coefficient graph, the order of the model and the value range of the coefficients can be initially determined. For example, the autocorrelation coefficient graph can be drawn to observe whether the autocorrelation coefficient shows a truncated feature after a certain order. If the autocorrelation coefficient drops sharply and remains near 0 after a certain order, the order of the autoregressive model can be initially determined; draw the partial autocorrelation coefficient graph to observe whether the partial autocorrelation coefficient shows a truncated feature after a certain order. If the partial autocorrelation coefficient drops sharply and remains near 0 after a certain order, the order of the moving average model can be initially determined.

[0046] In specific implementation, the last significant autocorrelation coefficient can be found based on the autocorrelation coefficient graph first, which is the order of the autocorrelation model. For example, if the last significant autocorrelation coefficient in the autocorrelation coefficient graph is at the 3rd order, the order of the autocorrelation model is 3. Then, based on the partial autocorrelation coefficient graph, the last significant partial autocorrelation coefficient is found, which is the order of the moving average model. For example, if the last significant partial autocorrelation coefficient in the partial autocorrelation coefficient graph is at the 2nd order, the order of the moving average model is 2. Finally, based on the autocorrelation coefficient graph and the partial autocorrelation coefficient graph, the order (p, q) of the autoregressive moving average model is determined. For example, if both the autocorrelation coefficient graph and the partial autocorrelation coefficient graph decay to zero after the 3rd order, the order of the autoregressive moving average model is (3, 3). Then, according to the order of the autoregressive moving average model, appropriate parameters are selected to establish the autoregressive moving average model of the station passenger flow coupling degree sequence. By bringing the station passenger flow coupling degree sequence into the autoregressive moving average model, the subsequent station passenger flow coupling degree can be predicted. Among them, the station passenger flow coupling degree at the end of the next maintenance time period is used as the dynamic scheduling sensitive factor.

[0047] In specific implementation, for example, the least squares method can be used to estimate the parameters and conduct significance tests for the autoregressive and moving average processes of the model. In some embodiments, the significance level of the significance test is 0.04. Finally, the most suitable autoregressive moving average model parameters are selected according to the Schwarz Bayesian criterion, so as to determine the final autoregressive moving average model of the station passenger flow coupling degree.

[0048] It should be noted that according to the response sensitivity of the station to the scheduling change (i.e., the dynamic scheduling sensitive factor), the present application adaptively adjusts the update frequency and adjustment amplitude of the timetable, realizing the dynamic closed-loop optimization of the passenger transport timetable. Optionally, in some embodiments, the dynamic feedback update of the initial passenger transport timetable by the dynamic scheduling sensitive factor specifically includes: determining the corresponding dynamic feedback period based on the threshold interval where the dynamic scheduling sensitive factor is located, collecting the heat learning features corresponding to the local area where the current station is located in real time by the dynamic feedback period, and generating the updated passenger transport timetable according to the heat learning features and the dynamic scheduling sensitive factor.

[0049] In specific implementation, the dynamic scheduling sensitive factor can be mapped in intervals according to the threshold interval where the dynamic scheduling sensitive factor is located and a preset linear mapping table to obtain the corresponding dynamic feedback period, and the passenger transport timetable is updated periodically based on the dynamic feedback period. Generally speaking, the larger the dynamic scheduling sensitive factor, the larger the dynamic feedback period, so that when the coupling degree of the passenger train scheduling is relatively high, the change frequency of the train timetable can be reduced, thus avoiding the chaos of the station operation caused by the frequent change of the train timetable, and improving the comfort and safety of the passengers during the train ride.

[0050] Preferably, in some embodiments, generating an updated passenger train schedule according to the popularity learning feature and the dynamic scheduling sensitivity factor specifically includes: Obtain multiple popularity learning feature samples and their corresponding passenger train schedule adjustment form templates respectively, where the passenger train schedule adjustment form templates are formulated based on manual experience; Obtain the dynamic scheduling sensitivity factor, correct the clustering threshold of the initial clustering according to the dynamic scheduling sensitivity factor, perform mean clustering on multiple popularity learning feature samples and the popularity learning feature according to the corrected clustering threshold, and generate an updated passenger train schedule based on the passenger train schedule adjustment form templates corresponding to each popularity learning feature in the clustering result.

[0051] Among them, the K-means mean clustering method is used to perform mean clustering on multiple popularity learning feature samples and the popularity learning feature. The K-means mean clustering forms a clustering center by assigning each data point to the centroid closest to it. The condition for satisfying the clustering result is that the distance difference between each clustering center and the centroid is lower than a preset clustering threshold. Among them, the corrected clustering threshold = the initial clustering threshold / the dynamic scheduling sensitivity factor.

[0052] It should be noted that a collaborative technology for formulating a daily passenger transport adjustment plan and generating a full-course passenger train schedule provided by this application belongs to the field of railway passenger transport operation optimization; aiming at the problems of insufficient dynamic response, difficult multi-professional collaboration, and low calculation efficiency caused by the independent operation of the passenger flow prediction, train adjustment, seat allocation, and schedule compilation links in the traditional method, this technology proposes a phased collaborative optimization and closed-loop feedback mechanism; by constructing a multi-professional modular collaborative system for passenger transport, transportation, dispatching, and vehicles, establishing a closed-loop process of "adjustment plan generation → schedule dynamic generation → operation evaluation → feedback optimization", and realizing lightweight data interaction through a standardized interface; the core technologies include: a dynamic hierarchical response mechanism to realize the linkage adjustment of seat control and train operation; a six-stage operation line generation algorithm for the transportation system based on space-time conflict detection; a three-dimensional conflict detection system for the dispatching profession that integrates car body turnover, space-time resources, and station capabilities; in implementation, through a four-stage collaborative execution process, realize the cross-stage parallel call of car body resources and the closed-loop feedback of the full-link data; this technology significantly improves the generation efficiency of the adjustment plan, the matching degree of the passenger occupancy rate, and the car body turnover efficiency, and effectively solves the problem of dynamic adaptation of railway daily transport capacity under a large-scale road network.

[0053] In addition, on the other hand of this application, in some embodiments, this application provides a railway passenger train schedule generation system based on collaborative dispatching. The device includes a schedule generation unit, refer to Figure 2, The figure is a schematic structural diagram of the exemplary hardware and / or software of the timetable generation unit shown in some embodiments of the present application. The timetable generation unit 200 includes a station data recording unit 201, a first timetable generation module 202, and a second timetable generation module 203, which are described as follows: The station data recording unit 201 is configured to obtain the passenger flow information of the current station, perform passenger flow heat learning through the passenger flow information of the current station to obtain the passenger flow heat of the current station, determine the local area where the current station is located, and construct a local heat matrix according to the passenger flow heat of other stations corresponding to the local area where the current station is located; The first timetable generation module 202 is configured to perform eigenvalue decomposition on the local heat matrix to obtain the heat learning features corresponding to the local area where the current station is located, and generate an initial passenger transport timetable for the current station through the heat learning features; The second timetable generation module 203 is configured to perform train operation scheduling based on the initial passenger transport timetable, determine the in-station scheduling coupling degree of the current station through the passenger flow heat and the passenger transport adjustment ratio of the current station, and determine the inter-station scheduling coupling degree based on the local heat matrix and the heat contribution ratio corresponding to the local area where the current station is located; The second timetable generation module 203 is further configured to determine the dynamic scheduling sensitivity factor corresponding to the current station according to the in-station scheduling coupling degree and the inter-station scheduling coupling degree, and perform dynamic feedback update on the initial passenger transport timetable through the dynamic scheduling sensitivity factor.

[0054] The above text details an example of a method and system for generating a railway passenger transport timetable based on collaborative scheduling provided by the embodiments of the present application. It can be understood that, in order to implement the above functions, the corresponding device includes the corresponding hardware structure and / or software module for executing each function.

[0055] Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed in this article, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function in the application is executed in the way of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Therefore, professional technicians can use different methods to implement the described functions for each specific application, but this implementation should not be considered to exceed the scope of the present application.

[0056] In addition, the present application also provides a computer terminal device, which includes a memory and a processor. The memory stores code, and the processor is configured to obtain the code and execute the above-mentioned method for generating a railway passenger transport timetable based on collaborative scheduling.

[0057] In some embodiments, refer to Figure 3 , which is a schematic structural diagram of a computer terminal device for implementing a method for generating a railway passenger transport timetable based on collaborative scheduling according to some embodiments of the present application. The method for generating a railway passenger transport timetable based on collaborative scheduling in the above embodiments can be implemented by Figure 3 The computer terminal device shown. The computer terminal device 300 includes at least one communication bus 301, a communication interface 302, a processor 303, and a memory 304.

[0058] The processor 303 can be a general-purpose central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more for controlling the execution of a method for generating a railway passenger transport timetable based on collaborative scheduling in the present application.

[0059] The communication bus 301 may include a path for transmitting information between the above components.

[0060] The memory 304 can be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM), or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disks, or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 304 can exist independently and be connected to the processor 303 through the communication bus 301. The memory 304 can also be integrated with the processor 303.

[0061] Among them, the memory 304 is used to store the program code for executing the solution of the present application and is controlled by the processor 303 for execution. The processor 303 is used to execute the program code stored in the memory 304. The program code may include one or more software modules. The determination of the heat learning feature in the above embodiments can be implemented by one or more software modules in the program code in the processor 303 and the memory 304.

[0062] A communication interface 302, using any device such as a transceiver, is used to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0063] Optionally, the above computer terminal device 300 may further include a power supply 305 for supplying power to various components or circuits in the real-time computer terminal device.

[0064] In a specific implementation, as an embodiment, the computer terminal device may include multiple processors, and each of these processors may be a single-CPU processor or a multi-CPU processor. Here, the processor may refer to one or more devices, circuits, and / or processing cores for processing data (such as computer program instructions).

[0065] The above computer terminal device may be a general-purpose computer terminal device or a special-purpose computer terminal device. In a specific implementation, the computer terminal device may be a desktop computer, a laptop computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiments of the present application do not limit the type of the computer terminal device.

[0066] In addition, in other aspects of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores at least one computer program, and the computer program is loaded and executed by a processor to implement the operations performed by the above method for generating a railway passenger transport timetable based on collaborative scheduling.

[0067] In summary, in a method and system for generating a railway passenger transport timetable based on collaborative scheduling disclosed in an embodiment of the present application, first, passenger flow information of the current station is obtained, passenger flow heat learning is performed through the passenger flow information of the current station to obtain the passenger flow heat of the current station, the local area where the current station is located is determined, and a local heat matrix is constructed according to the passenger flow heat of other stations corresponding to the local area where the current station is located; eigenvalue decomposition is performed on the local heat matrix to obtain heat learning features corresponding to the local area where the current station is located, and an initial passenger transport timetable for the current station is generated through the heat learning features; train operation scheduling is performed based on the initial passenger transport timetable, and the in-station scheduling coupling degree of the current station is determined through the passenger flow heat and passenger transport adjustment ratio of the current station, and the inter-station scheduling coupling degree is determined based on the local heat matrix and heat contribution ratio corresponding to the local area where the current station is located; the dynamic scheduling sensitivity factor corresponding to the current station is determined according to the in-station scheduling coupling degree and the inter-station scheduling coupling degree, and the initial passenger transport timetable is dynamically feedback updated through the dynamic scheduling sensitivity factor, so that the train timetable can be dynamically feedback updated according to the passenger flow coupling situation of the local area where the station is located after scheduling, and the generation efficiency of the passenger transport adjustment plan and the seat occupancy matching degree are improved.

[0068] The above are only embodiments of the present application. Specific technical solutions or common knowledge such as characteristics that are well known in the art are not described in detail herein. It should be noted that for those skilled in the art, without departing from the technical solution of the present application, several modifications and improvements can still be made, and these should also be regarded as the protection scope of the present application, and these will not affect the implementation effect of the present application and the practicality of the patent.

[0069] The protection scope required by the present application should be based on the content of its claims. The specific implementation manners and the like described in the specification can be used to explain the content of the claims. 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 invention. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these changes and modifications.

Claims

1. A method for generating a railway passenger timetable based on collaborative scheduling, characterized in that: include: Obtain the passenger flow information of the current station, conduct passenger flow heat learning based on the passenger flow information of the current station, obtain the passenger flow heat of the current station, determine the local area where the current station is located, and construct a local heat matrix based on the passenger flow heat of other stations corresponding to the local area where the current station is located; Performing eigenvalue decomposition on the local heat matrix to obtain heat learning features corresponding to the local area where the current station is located, and generating an initial passenger timetable for the current station through the heat learning features; Train operation scheduling is performed based on the initial passenger transport timetable, and the intra-station scheduling coupling degree of the current station is determined by the passenger flow heat and passenger adjustment ratio of the current station, and the inter-station scheduling coupling degree is determined based on the local heat matrix and heat contribution ratio corresponding to the local area where the current station is located; The dynamic scheduling sensitivity factor corresponding to the current station is determined according to the intra-station scheduling coupling degree and the inter-station scheduling coupling degree, and the initial passenger transport timetable is dynamically fed back and updated through the dynamic scheduling sensitivity factor.

2. The method according to claim 1, characterized in that Generating the initial passenger timetable of the current station through the heat learning feature specifically includes: Acquire multiple heat learning feature samples and their corresponding passenger transport timetable adjustment table templates, wherein the passenger transport timetable adjustment table templates are formulated through manual experience; Mean clustering is performed on multiple thermal learning feature samples and the thermal learning features, and an initial passenger timetable for the current station is generated based on the passenger timetable adjustment table templates corresponding to each thermal learning feature in the clustering result.

3. The method according to claim 1, characterized in that The inter-station scheduling coupling degree is determined based on the local heat matrix and heat contribution ratio corresponding to the local area where the current station is located, including: Obtain a preset windowing period, extract the local heat matrices corresponding to the local area where the current station is located in different windowing periods after train operation scheduling, and perform coupling feature extraction on the local heat matrices corresponding to each windowing period to obtain the heat coupling features corresponding to each windowing period; Obtain the heat contribution ratios corresponding to the current station in different windowing periods, and determine the inter-station scheduling coupling degree corresponding to the current station based on the heat contribution ratios and heat coupling characteristics of the current station in different windowing periods.

4. The method according to claim 1, characterized in that Determining the dynamic scheduling sensitivity factor corresponding to the current station according to the intra-station scheduling coupling degree and the inter-station scheduling coupling degree specifically includes: Get the dynamic sensitivity prediction interval, get the intra-station prediction weight and the inter-station prediction weight; Based on the intra-station prediction weight and the inter-station prediction weight, the intra-station scheduling coupling degree and the inter-station scheduling coupling degree within each dynamic sensitivity prediction interval are weightedly integrated to obtain the station passenger flow coupling degree corresponding to different interval times; The passenger flow coupling degree of stations corresponding to different interval times is forward predicted, and the passenger flow coupling degree of the station in the next interval time is used as the dynamic scheduling sensitive factor corresponding to the current station.

5. The method according to claim 1, characterized in that The dynamic feedback update of the initial passenger timetable through the dynamic scheduling sensitive factor specifically includes: determining the corresponding dynamic feedback cycle based on the threshold interval of the dynamic scheduling sensitive factor, using the dynamic feedback cycle to collect the heat learning features corresponding to the local area where the current station is located in real time, and generating an updated passenger timetable based on the heat learning features and the dynamic scheduling sensitive factor.

6. The method according to claim 1, characterized in that The passenger flow information of the current station specifically includes: historical ticket sales data, entry and exit gate records and train operation records.

7. The method according to claim 1, characterized in that Passenger flow heat learning is performed through the passenger flow information of the current station to obtain the passenger flow heat of the current station, which specifically includes: using a single hidden layer neural network to learn and cluster the passenger flow information of the current station, and using the heat value corresponding to the clustering result as the passenger flow heat corresponding to the current station.

8. A railway passenger timetable generation system based on collaborative scheduling, comprising a timetable generation unit, characterized in that: The timetable generating unit comprises: The station data recording unit is used to obtain the passenger flow information of the current station, perform passenger flow heat learning based on the passenger flow information of the current station, obtain the passenger flow heat of the current station, determine the local area where the current station is located, and construct a local heat matrix based on the passenger flow heat of other stations corresponding to the local area where the current station is located; A first timetable generation module is used to perform eigenvalue decomposition on the local heat matrix to obtain heat learning features corresponding to the local area where the current station is located, and generate an initial passenger timetable for the current station through the heat learning features; The second timetable generation module is used to perform train operation scheduling based on the initial passenger timetable, and determine the station scheduling coupling degree of the current station based on the passenger flow heat and passenger adjustment ratio of the current station, and determine the inter-station scheduling coupling degree based on the local heat matrix and heat contribution ratio corresponding to the local area where the current station is located; The second timetable generating module is further used to determine the dynamic scheduling sensitivity factor corresponding to the current station according to the intra-station scheduling coupling degree and the inter-station scheduling coupling degree, and dynamically feedback and update the initial passenger timetable through the dynamic scheduling sensitivity factor.

9. A computer terminal device, characterized in that: The computer terminal device includes a memory and a processor, the memory stores codes, and the processor is configured to obtain the codes and execute a method for generating a railway passenger timetable based on collaborative scheduling as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing at least one computer program, characterized in that: The computer program is loaded and executed by a processor to implement the operations performed by a method for generating a railway passenger timetable based on collaborative scheduling as described in any one of claims 1 to 7.

Images