A bus line passenger flow prediction method, system and terminal considering space-time characteristics

Abstract

The application discloses a kind of bus route passenger flow prediction method, system and terminal considering space-time characteristics, the method includes: preprocessing to bus card data, line passenger flow data is counted and the time granularity is determined;Qualitative analysis is carried out on line passenger flow data from time dimension and space dimension respectively, and the bus route passenger flow characteristics on time dimension and the bus route passenger flow characteristics on space dimension are obtained respectively;Quantify the similarity relationship between line passenger flows and the flow-through relationship between line stations, and filter out high correlation space-time characteristics;Bus route passenger flow prediction combination model is constructed, and bus route passenger flow prediction combination model is trained, tested and tuned;The bus route data to be predicted is input into the trained bus route passenger flow prediction combination model, and the bus route passenger flow prediction result is output.The application provides more accurate and effective line passenger flow prediction data, helps system to formulate more practical daily operation scheduling scheme, reduces resource waste, and improves transportation efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of bus big data processing technology, and particularly relates to a bus line passenger flow prediction method and system considering space-time characteristics, a terminal and a computer readable storage medium. BACKGROUND

[0002] With the rapid advancement of urbanization, urban transportation problems are becoming increasingly prominent. Public transportation systems play an important role in urban transportation, providing convenient and efficient travel options for urban residents, reducing traffic congestion, improving environmental quality, and promoting sustainable urban development.

[0003] Traditional buses, as an important part of public transportation systems, have their irreplaceable role in urban transportation due to their inclusiveness and accessibility. With the influence of various factors, the public's travel habits have changed, leading to a sharp decline in passenger flow on urban buses and trams. For example, some bus routes have low load factors during certain time periods and sections, and the passenger flow is unevenly distributed. This uneven phenomenon leads to resource waste during off-peak hours and passenger congestion during peak hours, affecting the efficiency and service quality of the public transportation system.

[0004] With the continuous development of intelligent transportation systems, its construction has become an effective means to solve urban transportation problems and an important indicator of the construction of "smart cities". As an important part of intelligent transportation systems, the bus intelligent scheduling system adheres to the policy direction of promoting sustainable urban development, improving public transportation service levels, reducing traffic congestion, and reducing carbon emissions. This system helps optimize passenger travel experience and promotes the development of urban transportation towards intelligence, efficiency, and environmental protection, becoming an important trend in future bus development. As one of the major components of intelligent transportation systems, the bus intelligent scheduling system can develop efficient vehicle scheduling plans based on passenger flow demand during different time periods, thereby helping public transportation enterprises reduce operating costs. Accurate passenger flow prediction can effectively obtain the dynamic changes of passenger flow, providing key support for intelligent optimization algorithms and ensuring the effectiveness and practicality of scheduling plans; however, existing technologies cannot provide more accurate and effective line passenger flow prediction data.

[0005] Therefore, the prior art still needs to be improved and developed. SUMMARY

[0006] The main purpose of the present application is to provide a bus line passenger flow prediction method and system considering space-time characteristics, a terminal and a computer readable storage medium, which aims to solve the problem of the prior art that cannot provide more accurate and effective line passenger flow prediction data, leading to resource waste and low transportation efficiency.

[0007] To achieve the above object, the application provides a bus line passenger flow prediction method considering space-time characteristics, which comprises the following steps:

[0008] Obtaining bus card swiping data, preprocessing the bus card swiping data, and determining line passenger flow data and time granularity according to the preprocessed bus card swiping data;

[0009] Qualitatively analyzing the line passenger flow data from the time dimension and the space dimension respectively to obtain bus line passenger flow characteristics in the time dimension and the space dimension respectively;

[0010] Quantifying the passenger flow similarity relationship between lines and the passenger flow circulation relationship between line stations according to the bus line passenger flow characteristics in the time dimension and the space dimension, and screening out high correlation space-time characteristics;

[0011] Constructing a bus line passenger flow prediction combination model, training, testing and optimizing the bus line passenger flow prediction combination model using the high correlation space-time characteristics, and obtaining a trained bus line passenger flow prediction combination model;

[0012] Obtaining to-be-predicted bus line data, inputting the to-be-predicted bus line data into the trained bus line passenger flow prediction combination model, and outputting bus line passenger flow prediction results.

[0013] The bus line passenger flow prediction method considering space-time characteristics, wherein the obtaining bus card swiping data, preprocessing the bus card swiping data, and determining line passenger flow data and time granularity according to the preprocessed bus card swiping data specifically comprises:

[0014] Obtaining all bus card swiping data of a prediction target line in a historical period, preprocessing all bus card swiping data of the prediction target line in a selected historical period, wherein the preprocessing comprises deduplication and error data deletion;

[0015] According to the time granularity used for short-term prediction, the line passenger flow data is counted, and the optimal prediction time granularity is determined through autocorrelation measurement and stationarity measurement.

[0016] The bus line passenger flow prediction method considering space-time characteristics, wherein the qualitatively analyzing the line passenger flow data from the time dimension and the space dimension respectively to obtain bus line passenger flow characteristics in the time dimension and the space dimension respectively specifically comprises:

[0017] Qualitatively analyzing the line passenger flow data from the time dimension to determine passenger flow time characteristics and bus line passenger flow relationship, and obtaining bus line passenger flow characteristics in the time dimension;

[0018] The passenger flow data of the bus routes are qualitatively analyzed from a spatial dimension to determine the spatial characteristics of passenger flow and the relationship between passenger flow of bus routes, thereby obtaining passenger flow characteristics of bus routes in a spatial dimension.

[0019] The bus route passenger flow prediction method considering spatiotemporal characteristics, wherein the method quantifies the passenger flow similarity relationship between routes and the passenger flow circulation relationship between route stations based on the bus route passenger flow characteristics in the temporal dimension and the bus route passenger flow characteristics in the spatial dimension, and screens out highly correlated spatiotemporal characteristics, specifically includes:

[0020] Analyze the similarity of overall and local passenger flows on the route, construct overall and local passenger flow similarity graphs, and use graph convolutional neural networks for self-learning to obtain the similarity features of overall and local passenger flows on the route respectively;

[0021] Calculate the ratio of passengers traveling back and forth between stations, obtain the station with the largest circulation intensity, and construct the circulation characteristics between stations according to the average circulation ratio of each station based on the station with the largest circulation intensity corresponding to all stations on the line;

[0022] Construct time features, which include: features of the previous 6 adjacent moments, features of the moment in the previous 4 weeks, and features of whether it is a weekday.

[0023] The bus route passenger flow prediction method considering spatiotemporal characteristics, wherein the calculation of the round-trip ratio of passengers between stations, obtaining the station with the largest circulation intensity corresponding to the station, and constructing the circulation characteristics between the station lines according to the average circulation ratio of each station based on the maximum circulation station corresponding to all stations on the line, specifically includes:

[0024] Passengers swipe their ID cards at the starting station O to the getting-off station D in the passenger flow OD data. i The number of card swipes of the passenger from station O to station D is calculated, and the round-trip ratio of the passenger from station O to station D is calculated. i Add up the round trip ratios of the two stations to get O and D i The fixed interaction strength between sites is calculated as:

[0025]

[0026] in, Indicates O and D i fixed interaction strength between sites; Indicates that the same passenger travels from O to D i Number of sites; Indicates that the same passenger from D i The number of times you go to station O; N means you get on the bus from station O i The number of passengers getting off at the station;

[0027] The station with the largest fixed interaction intensity corresponding to all stations of the line is taken as the maximum interaction station D of the station m The passenger flow of the maximum interaction station is multiplied by the average outflow proportion from the maximum interaction station to the station to obtain the passenger flow interaction feature of the station, and the line station flow feature is the sum of the passenger flow features of all stations in the line.

[0028]

[0029] Wherein, L i represents the line station flow feature of the i line; represents the passenger flow of the maximum interaction station of the jth station in the i line at the t time; represents the average travel proportion from the maximum interaction station of the jth station in the i line to the station.

[0030] The bus line passenger flow prediction method considering the space-time feature, wherein the bus line passenger flow prediction combination model is a GCN-XGBoost bus line passenger flow prediction combination model.

[0031] The bus line passenger flow prediction method considering the space-time feature, wherein the GCN-XGBoost bus line passenger flow prediction combination model comprises a GCN feature extraction layer, a feature fusion layer and an XGBoost layer.

[0032] The GCN feature extraction layer is used to capture the passenger flow similarity relationship between lines by using GCN, and extract line passenger flow similarity features.

[0033] The feature fusion layer is used to fuse all selected features after the GCN extracts the line passenger flow similarity features to form a seven-dimensional feature.

[0034] The XGBoost layer is used to input the seven-dimensional feature obtained by the feature fusion layer into XGBoost for passenger flow prediction.

[0035] In addition, in order to achieve the above purpose, the application also provides a bus line passenger flow prediction system considering space-time features, wherein the bus line passenger flow prediction system considering space-time features comprises:

[0036] The data processing module is used for acquiring bus card swiping data, preprocessing the bus card swiping data, and determining the line passenger flow data and the time granularity according to the preprocessed bus card swiping data.

[0037] The data analysis module is configured to perform qualitative analysis on the line passenger flow data from the time dimension and the space dimension respectively, and obtain line passenger flow characteristics in the time dimension and line passenger flow characteristics in the space dimension respectively.

[0038] The feature screening module is configured to quantify passenger flow similarity relationship between lines and passenger flow circulation relationship between line stations according to the line passenger flow characteristics in the time dimension and the line passenger flow characteristics in the space dimension, and screen out high correlation spatiotemporal characteristics.

[0039] The model training module is configured to construct a line passenger flow prediction combination model, train, test and optimize the line passenger flow prediction combination model using the high correlation spatiotemporal characteristics, and obtain a trained line passenger flow prediction combination model.

[0040] The passenger flow prediction module is configured to obtain to-be-predicted line data, input the to-be-predicted line data into the trained line passenger flow prediction combination model, and output a line passenger flow prediction result.

[0041] In addition, to achieve the above object, the application further provides a terminal, wherein the terminal comprises a memory, a processor, and a bus for connecting the memory and the processor; and the terminal further comprises a space-time feature considering bus line passenger flow prediction program stored in the memory and executable on the processor, and the space-time feature considering bus line passenger flow prediction program is executed by the processor to implement the steps of the space-time feature considering bus line passenger flow prediction method.

[0042] In addition, to achieve the above object, the application further provides a computer readable storage medium, wherein the computer readable storage medium stores a space-time feature considering bus line passenger flow prediction program, and the space-time feature considering bus line passenger flow prediction program is executed by a processor to implement the steps of the space-time feature considering bus line passenger flow prediction method.

[0043] In the present application, bus card data is obtained, the bus card data is preprocessed, line passenger flow data is counted and time granularity is determined according to the preprocessed bus card data; the line passenger flow data is qualitatively analyzed from the time dimension and the space dimension respectively, and the bus line passenger flow characteristics in the time dimension and the bus line passenger flow characteristics in the space dimension are obtained respectively; the line passenger flow similarity relationship and the line station passenger flow circulation relationship are quantified according to the bus line passenger flow characteristics in the time dimension and the bus line passenger flow characteristics in the space dimension, and high correlation space-time characteristics are screened out; a bus line passenger flow prediction combination model is constructed, the high correlation space-time characteristics are used to train, test and optimize the bus line passenger flow prediction combination model, and a trained bus line passenger flow prediction combination model is obtained; the to-be-predicted bus line data is obtained, the to-be-predicted bus line data is input into the trained bus line passenger flow prediction combination model, and a bus line passenger flow prediction result is output. The present application provides more accurate and effective line passenger flow prediction data for the bus intelligent scheduling system, helps the system to formulate a more practical daily operation scheduling scheme, reduces resource waste, and improves transportation efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 is a flow chart of a preferred embodiment of the bus line passenger flow prediction method considering space-time characteristics of the present application;

[0045] Figure 2 is a framework diagram of a bus line passenger flow prediction combination model of GCN-XGBoost in a preferred embodiment of the bus line passenger flow prediction method considering space-time characteristics of the present application;

[0046] Figure 3 is a comparison diagram of the accuracy of the bus line passenger flow prediction combination model of GCN-XGBoost and other models in a preferred embodiment of the bus line passenger flow prediction method considering space-time characteristics of the present application;

[0047] Figure 4 is a principle schematic diagram of a preferred embodiment of the bus line passenger flow prediction system considering space-time characteristics of the present application;

[0048] Figure 5 is a running environment schematic diagram of a preferred embodiment of the terminal of the present application. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical scheme and advantages of the present application more clear and definite, the present application is further described in detail below with reference to the drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0050] In the current public transport daily operation, there are often problems such as resource waste and service imbalance. In the scheduling operation of public transport lines, relying on the public transport intelligent scheduling system can better schedule public transport resources according to real-time passenger conditions, and reduce resource waste. However, in the daily public transport passenger flow prediction, the spatial characteristics between lines are often ignored, while only the time characteristics of the public transport passenger flow are considered. Fully considering the time and space characteristics can better cope with passenger flow fluctuations and improve prediction accuracy.

[0051] To solve the above technical problems, the application provides a bus line passenger flow prediction method considering time and space characteristics. In the embodiment of the application, bus card data is counted by half-hour time granularity, and then the passenger flow characteristics are mined from the time and space dimensions, the relationship between the above dimension characteristic factors and the bus line passenger flow is determined, and then a construction method of similar characteristics between lines and flow intensity characteristics between line stations is proposed according to the relationship between the characteristic factors and the bus line passenger flow, and high correlation time and space characteristics are selected. Then, a GCN-XGBoost bus line passenger flow prediction combination model is constructed, and 28 days of historical passenger flow data are used for model training, testing and optimization to verify the superiority of the model.

[0052] As shown in the bus line passenger flow prediction method considering time and space characteristics in the preferred embodiment of the application, Figure 1 The bus line passenger flow prediction method considering time and space characteristics comprises the following steps:

[0053] Step S10, obtaining bus card data, preprocessing the bus card data, counting line passenger flow data according to the preprocessed bus card data and determining the time granularity.

[0054] Specifically, all bus card data of the prediction target line in the historical period is obtained, and the all bus card data of the prediction target line in the selected historical period is preprocessed by Python, wherein the preprocessing includes de-duplication and error data deletion; the line passenger flow data is counted according to the time granularity used for short-term prediction target, and the optimal prediction time granularity is determined through autocorrelation measurement and stationarity measurement, so as to better mine passenger flow characteristics and achieve better prediction effect.

[0055] In this embodiment, the card data can be de-duplicated, error data deleted and the like, and then the passenger flow data of 15min, 30min and 60min (the time granularity usually used for short-term prediction) is counted for autocorrelation measurement and stationarity measurement to determine the optimal time granularity for prediction. The autocorrelation can reflect the periodicity, trend and other characteristic strengths of the time series; the stationarity of the time series is helpful to simplify the model construction and improve the prediction accuracy, and the statistical characteristics (such as mean and variance) of the stationary sequence do not change with time, so that the prediction is more stable and reliable.

[0056] Step S20, respectively from the time dimension and the spatial dimension on the line passenger flow data qualitative analysis, respectively get time dimension on the bus line passenger flow characteristics and spatial dimension on the bus line passenger flow characteristics.

[0057] Specifically, from the time dimension and the spatial dimension on the line passenger flow characteristics, the relationship between the above dimension characteristic factors and the bus line passenger flow is clear, the time dimension on the line passenger flow data is qualitatively analyzed, the passenger flow time characteristics and the relationship between the bus line passenger flow are determined, and the time dimension on the bus line passenger flow characteristics is obtained. First, the total passenger flow of weekdays, non-working days and each period is visually analyzed, aiming to find the passenger flow law under different time attributes, such as weekday rush hour, passenger flow periodicity, etc. From the spatial dimension on the line passenger flow data qualitative analysis, the passenger flow space characteristics and the relationship between the bus line passenger flow are determined, and the spatial dimension on the bus line passenger flow characteristics is obtained.

[0058] The average percentage difference between each line from the overall passenger flow and the local passenger flow (morning peak / early peak / late peak / late peak) is calculated, aiming to verify the similar relationship between the passenger flow between lines; The same passenger return ratio between stations is calculated, and the correlation coefficient between the return ratio between stations and the correlation between the passenger flow between stations is calculated to verify whether there is a relationship between the return ratio and the correlation between the passenger flow between stations. The results show that there is indeed a similar relationship between the passenger flow between lines, and there is a high coupling between the return ratio and the correlation between the passenger flow between stations, so it can be preliminarily determined that these two spatial factors have an impact on passenger flow, and the factors can be quantified as characteristics for consideration.

[0059] Step S30, according to the time dimension on the bus line passenger flow characteristics and the spatial dimension on the bus line passenger flow characteristics, quantifying the passenger flow similarity relationship between lines and the passenger flow flow-through relationship between line stations, and screening out high correlation space-time characteristics.

[0060] Specifically, the constructed spatial characteristics include: overall line passenger flow similarity characteristics, local line passenger flow similarity characteristics, line station flow-through characteristics; The time characteristics included in the screening together include: the characteristics of the previous 6 time points, the characteristics of the previous 4 weeks at this time point, and the characteristics of whether it is a working day. Correlation analysis is performed on these characteristics and the real passenger flow, and the influence factor characteristics with a correlation greater than 0.3 are screened out (usually a correlation coefficient greater than 0.3 is considered to have a correlation between the two).

[0061] Step S30 includes:

[0062] Step S301, constructing the overall line passenger flow similarity feature and the local line passenger flow similarity feature: analyzing the overall and local passenger flow similarity of the line, constructing the overall and local passenger flow similarity graph, and using a graph convolutional neural network (GCN) for self-learning to obtain the overall line passenger flow similarity feature and the local line passenger flow similarity feature.

[0063] Step S302, constructing the line station flow feature: calculating the round-trip ratio of passengers between stations, obtaining the station corresponding to the maximum flow intensity, and constructing the line station flow feature according to the maximum flow station corresponding to all stations in the line according to the average flow ratio of each station.

[0064] Further, step S302 includes the following steps:

[0065] Step S302a, calculating the fixed interaction intensity between stations.

[0066] Calculation method: through the passenger card ID in the passenger flow OD data, the number of times of card swiping from the starting station O to the alighting station D i and the number of times of card swiping from the alighting station D i to the starting station O, the round-trip ratio of the passenger between the stations O and D i is calculated, the round-trip ratio of all passengers between the stations O and D i is added, and the fixed interaction intensity between the stations O and D i is obtained. The fixed interaction intensity is calculated as:

[0067]

[0068] wherein, represents the fixed interaction intensity between the stations O and D i ; represents the number of times of the same passenger from the station O to the station D i ; represents the number of times of the same passenger from the station D i to the station O; and N represents the number of passengers boarding the station O and alighting the station D i .

[0069] Step S302b, calculating the line station flow feature.

[0070] Calculation method: taking the station corresponding to the maximum fixed interaction intensity of all stations in the line as the maximum interaction station D m of the station, multiplying the passenger flow of the maximum interaction station by the average outflow ratio from the maximum interaction station to the station to obtain the station passenger flow interaction feature, and calculating the line station flow feature by adding the station passenger flow interaction features of all stations in the line. The line station flow feature is calculated as:

[0071]

[0072] wherein, Li represents the line station flow characteristics of line i; represents the passenger flow of the maximum interactive station at the jth station in line i at time t; It represents the average travel ratio from the maximum interactive station of the jth station in the i-th line to the j-th station.

[0073] Step S303: construct time features.

[0074] The time features included in the screening include: the features of the previous 6 consecutive moments, the features of the moment in the previous 4 weeks, and whether it is a weekday. The specific construction method is as follows:

[0075] Features of the previous t moments: predict the passenger flow t moments before the target moment;

[0076] Features of the time in the previous t weeks: predict the passenger flow at the time t weeks before the target time;

[0077] Whether it is a working day: 0 (non-working day) / 1 (working day).

[0078] Step S40: construct a bus route passenger flow prediction combination model, and use the highly correlated spatiotemporal features to train, test, and optimize the bus route passenger flow prediction combination model to obtain a trained bus route passenger flow prediction combination model.

[0079] Specifically, the bus route passenger flow prediction combination model is a GCN-XGBoost bus route passenger flow prediction combination model.

[0080] Step S40 includes the following steps:

[0081] Step S401, model architecture design, such as Figure 2 As shown:

[0082] The GCN-XGBoost bus route passenger flow prediction combination model includes: a GCN feature extraction layer, a feature fusion layer and an XGBoost layer; the GCN feature extraction layer is used to use GCN (graph convolutional neural network) to capture the passenger flow similarity relationship between routes and extract the route passenger flow similarity features; the feature fusion layer is used to fuse all selected features after GCN extracts the route passenger flow similarity features to form a seven-dimensional feature (each dimension is: passenger flow at the adjacent previous moment, passenger flow at the adjacent previous two moments, passenger flow at the adjacent previous week, whether it is a weekday, overall passenger flow similarity features of the route, local passenger flow similarity features of the route, and route station circulation features); the XGBoost layer is used to input the seven-dimensional features obtained by the feature fusion layer into XGBoost for passenger flow prediction.

[0083] Step S402: model training and testing.

[0084] The existing bus card swiping data is used as the training test data set, and is divided into a training set, a validation set and a test set according to a ratio of 7:1.5:1.5.

[0085] Step S403, model parameter tuning.

[0086] Hyperparameter tuning: in the GCN feature extraction layer, the hidden layer dimension and batch size are tuned using grid search; in the XGBoost layer, six key parameters such as learning rate and maximum tree depth are tuned using Bayesian optimization.

[0087] Step S404, model performance, feature contribution and applicability analysis.

[0088] Model performance analysis: using MAPE (mean absolute percentage error), RMSE (root mean square error) and other evaluation indicators to quantify the prediction accuracy of the model on the validation set and the test set, and using accuracy to compare with common models, such as Figure 3 As shown, the common prediction models include single models XGBoost, SVM, random forest, LSTM, GCN and common combination models GCN-LSTM.

[0089] Feature contribution analysis: by outputting the contribution of each feature in the XGBoost model, the coupling degree between feature contribution and feature correlation is calculated.

[0090] Applicability analysis: from the time dimension, passenger flow dimension and line attribute dimension, the applicability of the model is analyzed.

[0091] Step S50, obtaining the to-be-predicted bus line data, inputting the to-be-predicted bus line data into the trained bus line passenger flow prediction combination model, and outputting the bus line passenger flow prediction result.

[0092] Specifically, after reaching the trained bus line passenger flow prediction combination model, the trained bus line passenger flow prediction combination model can be used to quickly predict the bus line passenger flow prediction line. Then, after obtaining the to-be-predicted bus line data, the to-be-predicted bus line data can be directly input into the trained bus line passenger flow prediction combination model to quickly output the bus line passenger flow prediction result.

[0093] The application extracts the time characteristics and space characteristics of the passenger flow of the bus line, analyzes the relationship between the passenger flow characteristics and the rail transit passenger flow by combining the bus card data, and then screens out high correlation factors for the input of the passenger flow prediction model based on the correlation of the two, so that the application uses high correlation space-time characteristics to further explore the relationship between the high correlation space-time characteristics and the line passenger flow, which is more in line with the passenger travel mode and provides data support for bus operation.

[0094] The application fully considers the high impact time factor and combines the space factor between line stations to realize short-time prediction of the bus line passenger flow by processing historical passenger flow data through Python, so as to provide more accurate passenger flow basis for the intelligent bus system and help realize efficient scheduling and balance of supply and demand relationship in bus daily operation.

[0095] The application realizes bus line passenger flow prediction considering more comprehensive space-time characteristics to provide more accurate passenger flow data support for the intelligent bus scheduling system and help more reasonably realize bus resource scheduling arrangement.

[0096] Further, as shown in Figure 4 The application also correspondingly provides a bus line passenger flow prediction system considering space-time characteristics based on the above bus line passenger flow prediction method considering space-time characteristics, wherein the bus line passenger flow prediction system considering space-time characteristics comprises:

[0097] A data processing module is configured to acquire bus card data, pre-process the bus card data, and determine line passenger flow data and time granularity according to the pre-processed bus card data.

[0098] A data analysis module is configured to qualitatively analyze the line passenger flow data from the time dimension and the space dimension respectively, and obtain bus line passenger flow characteristics in the time dimension and bus line passenger flow characteristics in the space dimension respectively.

[0099] A feature screening module is configured to quantize the passenger flow similarity relationship between lines and the passenger flow flow-through relationship between line stations according to the bus line passenger flow characteristics in the time dimension and the bus line passenger flow characteristics in the space dimension, and screen out high correlation space-time characteristics.

[0100] A model training module is configured to construct a bus line passenger flow prediction combination model, train, test and optimize the bus line passenger flow prediction combination model using the high correlation space-time characteristics, and obtain a trained bus line passenger flow prediction combination model.

[0101] The passenger flow prediction module is configured to acquire to-be-predicted bus line data, input the to-be-predicted bus line data into the trained bus line passenger flow prediction combination model, and output a bus line passenger flow prediction result.

[0102] Further, as shown in Figure 5 Based on the above-mentioned bus line passenger flow prediction method and system considering the space-time characteristics, the application further provides a terminal, which comprises a processor 10, a memory 20 and a display 30. Figure 5 Only some components of the terminal are shown, but it should be understood that all the shown components are not required, and more or less components can be alternatively implemented.

[0103] The memory 20 can be an internal storage unit of the terminal in some embodiments, such as a hard disk or a memory of the terminal. The memory 20 can also be an external storage device of the terminal in other embodiments, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory 20 can include both the internal storage unit and the external storage device of the terminal. The memory 20 is configured to store application software and various data installed on the terminal, such as program codes of the terminal, etc. The memory 20 can also be configured to temporarily store data that has been output or will be output. In an embodiment, the memory 20 stores a bus line passenger flow prediction program 40 considering the space-time characteristics, which can be executed by the processor 10, so as to implement the bus line passenger flow prediction method considering the space-time characteristics.

[0104] The processor 10 can be a central processing unit (CPU), a microprocessor or other data processing chip in some embodiments, which is configured to run program codes or process data stored in the memory 20, such as to execute the bus line passenger flow prediction method considering the space-time characteristics, etc.

[0105] The display 30 can be an LED display, a liquid crystal display, a touch liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, etc. in some embodiments. The display 30 is configured to display information of the terminal and to display a visualized user interface. The components 10-30 of the terminal communicate with each other through a system bus.

[0106] In an embodiment, the following steps are implemented when the processor 10 executes the bus route passenger flow prediction program 40 considering the space-time features in the memory 20:

[0107] Obtain bus card swiping data, pre-process the bus card swiping data, and count route passenger flow data and determine the time granularity according to the pre-processed bus card swiping data;

[0108] Qualitatively analyze the route passenger flow data from the time dimension and the space dimension respectively, and obtain bus route passenger flow features in the time dimension and bus route passenger flow features in the space dimension respectively;

[0109] Quantify the passenger flow similarity relationship between routes and the passenger flow circulation relationship between route stations according to the bus route passenger flow features in the time dimension and the bus route passenger flow features in the space dimension, and screen out high-correlation space-time features;

[0110] Construct a bus route passenger flow prediction combination model, train, test, and optimize the bus route passenger flow prediction combination model using the high-correlation space-time features, and obtain a trained bus route passenger flow prediction combination model;

[0111] Obtain to-be-predicted bus route data, input the to-be-predicted bus route data into the trained bus route passenger flow prediction combination model, and output bus route passenger flow prediction results.

[0112] The bus route passenger flow prediction method considering space-time features, wherein the obtaining bus card swiping data, pre-processing the bus card swiping data, and counting route passenger flow data and determining the time granularity according to the pre-processed bus card swiping data specifically comprises:

[0113] Obtain all bus card swiping data of a prediction target route in a historical period, and pre-process all bus card swiping data of the prediction target route in a selected historical period, wherein the pre-processing includes de-duplication and error data deletion;

[0114] Count route passenger flow data according to the time granularity used for short-term prediction, and determine the optimal prediction time granularity through autocorrelation measurement and stationarity measurement.

[0115] The bus route passenger flow prediction method considering space-time features, wherein the qualitatively analyzing the route passenger flow data from the time dimension and the space dimension respectively, and obtaining bus route passenger flow features in the time dimension and bus route passenger flow features in the space dimension specifically comprises:

[0116] Qualitatively analyze the route passenger flow data from the time dimension, determine passenger flow time features and bus route passenger flow relationships, and obtain bus route passenger flow features in the time dimension;

[0117] The passenger flow data of the bus routes are qualitatively analyzed from a spatial dimension to determine the spatial characteristics of passenger flow and the relationship between passenger flow of bus routes, thereby obtaining passenger flow characteristics of bus routes in a spatial dimension.

[0118] The bus route passenger flow prediction method considering spatiotemporal characteristics, wherein the method quantifies the passenger flow similarity relationship between routes and the passenger flow circulation relationship between route stations based on the bus route passenger flow characteristics in the temporal dimension and the bus route passenger flow characteristics in the spatial dimension, and screens out highly correlated spatiotemporal characteristics, specifically includes:

[0119] Analyze the similarity of overall and local passenger flows on the route, construct overall and local passenger flow similarity graphs, and use graph convolutional neural networks for self-learning to obtain the similarity features of overall and local passenger flows on the route respectively;

[0120] Calculate the ratio of passengers traveling back and forth between stations, obtain the station with the largest circulation intensity, and construct the circulation characteristics between stations according to the average circulation ratio of each station based on the station with the largest circulation intensity corresponding to all stations on the line;

[0121] Construct time features, which include: features of the previous 6 adjacent moments, features of the moment in the previous 4 weeks, and features of whether it is a weekday.

[0122] The bus route passenger flow prediction method considering spatiotemporal characteristics, wherein the calculation of the round-trip ratio of passengers between stations, obtaining the station with the largest circulation intensity corresponding to the station, and constructing the circulation characteristics between the station lines according to the average circulation ratio of each station based on the maximum circulation station corresponding to all stations on the line, specifically includes:

[0123] Passengers swipe their ID cards at the starting station O to the getting-off station D in the passenger flow OD data. i The number of card swipes of the passenger from station O to station D is calculated, and the round-trip ratio of the passenger from station O to station D is calculated. i Add up the round trip ratios of the two stations to get O and D i The fixed interaction strength between sites is calculated as:

[0124]

[0125] in, Indicates O and D i fixed interaction strength between sites; Indicates that the same passenger travels from O to D i Number of sites; Indicates that the same passenger from D i The number of times you go to station O; N means you get on the bus from station O i The number of passengers getting off at the station;

[0126] The station with the largest fixed interaction intensity corresponding to all stations of the line is taken as the maximum interaction station D of the station m The passenger flow of the maximum interaction station is multiplied by the average outflow proportion from the maximum interaction station to the station to obtain the passenger flow interaction feature of the station, the line station flow circulation feature is the sum of the station passenger flow circulation features of all stations in the line, and the line station flow circulation feature is calculated as:

[0127]

[0128] Wherein, L i represents the line station flow circulation feature of the i line; represents the passenger flow of the maximum interaction station of the jth station in the i line at the t time; represents the average travel proportion from the maximum interaction station of the jth station in the i line to the station.

[0129] The bus line passenger flow prediction method considering the space-time features, wherein the bus line passenger flow prediction combination model is a bus line passenger flow prediction combination model of GCN-XGBoost.

[0130] The bus line passenger flow prediction method considering the space-time features, wherein the bus line passenger flow prediction combination model of GCN-XGBoost comprises a GCN feature extraction layer, a feature fusion layer and an XGBoost layer;

[0131] The GCN feature extraction layer is used to capture the passenger flow similarity relationship between lines by GCN, and extract line passenger flow similarity features;

[0132] The feature fusion layer is used to fuse all selected features after the line passenger flow similarity features are extracted by GCN, and form a seven-dimensional feature;

[0133] The XGBoost layer is used to input the seven-dimensional feature obtained by the feature fusion layer into XGBoost for passenger flow prediction.

[0134] The application also provides a computer readable storage medium, wherein the computer readable storage medium stores a bus line passenger flow prediction program considering space-time features, and the bus line passenger flow prediction program considering space-time features realizes the steps of the bus line passenger flow prediction method considering space-time features when executed by a processor.

[0135] In summary, the application provides a bus line passenger flow prediction method, system, terminal and computer readable storage medium considering space-time characteristics, the method comprising: obtaining bus card swiping data, preprocessing the bus card swiping data, and determining line passenger flow data and time granularity according to the preprocessed bus card swiping data; qualitatively analyzing the line passenger flow data from the time dimension and the space dimension respectively, and obtaining bus line passenger flow characteristics in the time dimension and bus line passenger flow characteristics in the space dimension respectively; quantifying the passenger flow similarity relationship between lines and the passenger flow circulation relationship between line stations according to the bus line passenger flow characteristics in the time dimension and the bus line passenger flow characteristics in the space dimension, and screening out high correlation space-time characteristics; constructing a bus line passenger flow prediction combination model, training, testing and tuning the bus line passenger flow prediction combination model using the high correlation space-time characteristics, and obtaining a trained bus line passenger flow prediction combination model; obtaining to-be-predicted bus line data, inputting the to-be-predicted bus line data into the trained bus line passenger flow prediction combination model, and outputting a bus line passenger flow prediction result. The application provides more accurate and effective line passenger flow prediction data for a bus intelligent scheduling system, helps the system to formulate a more practical daily operation scheduling scheme, reduces resource waste, and improves transportation efficiency.

[0136] It should be noted that in this document, the terms "comprising", "containing", or any other variant thereof are intended to cover non-exclusive inclusion, so that processes, methods, articles, or terminals including a series of elements not only include those elements, but also include other elements not explicitly listed, or include elements inherent to such processes, methods, articles, or terminals. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of another identical element in the process, method, article, or terminal including the element.

[0137] Of course, those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing relevant hardware (such as a processor, a controller, etc.) through a computer program, and the program can be stored in a computer readable storage medium readable by a computer, and the program can include the processes of the above-mentioned method embodiments when executed. The computer readable storage medium can be a memory, a disk, an optical disk, etc.

[0138] It should be understood that the application is not limited to the above examples, and those skilled in the art can make improvements or changes according to the above description, and all these improvements and changes should be within the protection scope of the appended claims of the application.

Claims

1. A bus route passenger flow prediction method considering spatiotemporal characteristics, characterized in that: The bus route passenger flow prediction method considering spatiotemporal characteristics includes: Obtaining bus card swiping data, preprocessing the bus card swiping data, and calculating route passenger flow data based on the preprocessed bus card swiping data and determining a time granularity; Qualitative analysis is performed on the passenger flow data of the bus route from the time dimension and the space dimension respectively, to obtain the bus route passenger flow characteristics in the time dimension and the bus route passenger flow characteristics in the space dimension respectively; Based on the bus route passenger flow characteristics in the temporal dimension and the bus route passenger flow characteristics in the spatial dimension, the passenger flow similarity relationship between routes and the passenger flow circulation relationship between route stations are quantified, and highly correlated spatiotemporal features are screened, specifically including: Analyze the similarity of overall and local passenger flows on the route, construct overall and local passenger flow similarity graphs, and use graph convolutional neural networks for self-learning to obtain the similarity features of overall and local passenger flows on the route respectively; Calculate the ratio of passengers traveling back and forth between stations, obtain the station with the largest circulation intensity, and construct the circulation characteristics between stations according to the average circulation ratio of each station based on the station with the largest circulation intensity corresponding to all stations on the line; Constructing time features, the time features include: features of the previous 6 adjacent moments, features of the moment in the previous 4 weeks, and features of whether it is a weekday; Passengers swipe their ID cards at the starting station O to the getting-off station in the passenger flow OD data. The number of card swipes is related to the passenger's The number of card swipes to the O station is calculated for the passenger O station and the station The round trip ratio will be used to transfer all passengers between O and Add the round trip ratios of the two stations to get O and fixed interaction strength between sites; The station with the largest fixed interaction intensity corresponding to all stations on the line is taken as the maximum interaction station of the station , the maximum interactive site passenger flow of the site multiplied by the average outflow ratio from the maximum interactive site to the site is taken as the site passenger flow interaction feature, and the line site flow feature is the sum of the site passenger flow flow features of all sites in the line; Constructing a bus route passenger flow prediction combination model, and using the highly correlated spatiotemporal features to train, test, and optimize the bus route passenger flow prediction combination model to obtain a trained bus route passenger flow prediction combination model; Obtain bus route data to be predicted, input the bus route data to a trained bus route passenger flow prediction combination model, and output a bus route passenger flow prediction result.

2. The bus route passenger flow prediction method considering spatiotemporal characteristics according to claim 1 is characterized in that: The obtaining of bus card swiping data, preprocessing the bus card swiping data, and counting route passenger flow data based on the preprocessed bus card swiping data and determining a time granularity specifically include: Acquire all bus card swiping data for the historical period of the predicted target route, and pre-process all bus card swiping data for the historical period selected by the predicted target route, wherein the pre-processing includes deduplication and deletion of erroneous data; The passenger flow data of the route is counted according to the time granularity used for short-term prediction, and the optimal prediction time granularity is determined through autocorrelation measurement and stationarity measurement.

3. The bus route passenger flow prediction method considering spatiotemporal characteristics according to claim 1 is characterized in that: The qualitative analysis of the bus route passenger flow data from the time dimension and the space dimension is respectively performed to obtain bus route passenger flow characteristics in the time dimension and bus route passenger flow characteristics in the space dimension, specifically including: Conducting a qualitative analysis on the passenger flow data of the bus routes from a time dimension, determining the relationship between the passenger flow time characteristics and the bus route passenger flow, and obtaining the bus route passenger flow characteristics in the time dimension; The passenger flow data of the bus routes are qualitatively analyzed from a spatial dimension to determine the spatial characteristics of passenger flow and the relationship between passenger flow of bus routes, thereby obtaining passenger flow characteristics of bus routes in a spatial dimension.

4. The bus route passenger flow prediction method considering spatiotemporal characteristics according to claim 1, characterized in that: The fixed interaction strength is calculated as: ； in, Indicates O and fixed interaction strength between sites; Indicates that the same passenger travels from O to Number of sites; Indicates that the same passenger Number of visits to site O; Indicates boarding at station O The number of passengers getting off at the station; The circulation characteristics of the line station are calculated as: ； in, represents the line station flow characteristics of line i; represents the passenger flow of the maximum interactive station at the jth station in line i at time t; It represents the average travel ratio from the maximum interactive station of the jth station in the i-th line to the j-th station.

5. The bus route passenger flow prediction method considering spatiotemporal characteristics according to claim 1 is characterized in that: The bus route passenger flow prediction combination model is a GCN-XGBoost bus route passenger flow prediction combination model.

6. The bus route passenger flow prediction method considering spatiotemporal characteristics according to claim 5 is characterized in that: The GCN-XGBoost bus route passenger flow prediction combined model includes: a GCN feature extraction layer, a feature fusion layer and an XGBoost layer; The GCN feature extraction layer is used to capture the similarity relationship between passenger flows between routes using GCN and extract similar features of passenger flows between routes; The feature fusion layer is used to fuse all selected features to form a seven-dimensional feature after GCN extracts similar features of route passenger flow; The XGBoost layer is used to input the seven-dimensional features obtained by the feature fusion layer into XGBoost to perform passenger flow prediction.

7. A bus route passenger flow prediction system considering spatiotemporal characteristics, characterized in that: The bus route passenger flow prediction system considering spatiotemporal characteristics is applied to the bus route passenger flow prediction method considering spatiotemporal characteristics according to any one of claims 1 to 6, and the bus route passenger flow prediction system considering spatiotemporal characteristics includes: A data processing module is used to obtain bus card swiping data, pre-process the bus card swiping data, calculate route passenger flow data based on the pre-processed bus card swiping data, and determine time granularity; A data analysis module is used to perform qualitative analysis on the bus route passenger flow data from the time dimension and the space dimension, respectively, to obtain bus route passenger flow characteristics in the time dimension and the space dimension; A feature screening module is used to quantify the passenger flow similarity relationship between routes and the passenger flow circulation relationship between route stations based on the bus route passenger flow characteristics in the time dimension and the bus route passenger flow characteristics in the space dimension, and screen out highly correlated spatiotemporal features; A model training module is used to construct a bus route passenger flow prediction combination model, and use the highly correlated spatiotemporal features to train, test and optimize the bus route passenger flow prediction combination model to obtain a trained bus route passenger flow prediction combination model; The passenger flow prediction module is used to obtain bus route data to be predicted, input the bus route data to be predicted into the trained bus route passenger flow prediction combination model, and output the bus route passenger flow prediction result.

8. A terminal, characterized in that: The terminal includes: a memory, a processor, and a bus route passenger flow prediction program that considers spatiotemporal characteristics, which is stored in the memory and can be run on the processor. When the bus route passenger flow prediction program that considers spatiotemporal characteristics is executed by the processor, the steps of the bus route passenger flow prediction method that considers spatiotemporal characteristics as described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a bus route passenger flow prediction program that considers spatiotemporal characteristics. When the bus route passenger flow prediction program that considers spatiotemporal characteristics is executed by a processor, the steps of the bus route passenger flow prediction method that considers spatiotemporal characteristics as described in any one of claims 1 to 6 are implemented.

Images