Abnormal aggregation identification and early warning method and system for online ride-hailing vehicles

Abstract

The present invention relates to a method and system for identifying and warning abnormal aggregation of online-hailing vehicles, and belongs to the technical field of traffic warning. The method includes: constructing an online-hailing vehicle grid image data set in a target area based on online-hailing vehicle operation time series data, wherein the online-hailing vehicle grid image data set represents the distribution of all online-hailing vehicles at different times; determining a feature data set of a target area based on the online-hailing vehicle grid image data set, subway passenger flow time series data, bus passenger flow time series data, historical environment time series data, and preset feature extraction rules; inputting the feature data set into a preset online-hailing vehicle quantity prediction model to predict the arrival volume of online-hailing vehicles and the road congestion index at each location within the target time period; inputting the arrival volume of online-hailing vehicles and the road congestion index at each location within the target time period into an online-hailing vehicle warning model to obtain warning information. The present application can accurately identify and effectively warn of abnormal aggregation behavior of online-hailing vehicles.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic warning, and in particular to a method and system for identifying and warning of abnormal aggregation of online-hailing vehicles. Background Art

[0002] With the rapid development of mobile Internet technology, online car-hailing services, as an emerging force in the urban transportation system, have greatly enriched citizens' travel options with their convenience, flexibility and efficiency, and effectively alleviated the imbalance between supply and demand in the traditional taxi industry. However, the rapid expansion of the online car-hailing industry is also accompanied by a series of challenges and problems, especially in terms of operation supervision, information feedback and safety management. Its complexity and dynamism have put forward higher requirements for the existing urban transportation management system.

[0003] In recent years, the untimely feedback of order information, lagging industry supervision and imperfect monitoring and early warning systems in online car-hailing services have repeatedly caused operational incidents that have attracted widespread public attention. These incidents have not only disrupted the normal urban traffic order, but also caused serious traffic congestion and safety hazards.

[0004] Due to the "abnormal gathering" phenomenon and potential traffic risks that online car-hailing services may induce under certain conditions, there is an urgent need for an online car-hailing abnormal gathering warning technology that can accurately identify and effectively warn of abnormal gathering behaviors of online car-hailing services. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method and system for identifying and warning of abnormal aggregation of online-hailing vehicles, aiming to solve at least one of the above-mentioned technical problems.

[0006] The technical solution of the present invention to solve the above technical problems is as follows:

[0007] In the first aspect, the present application provides a method for identifying and warning of abnormal aggregation of online ride-hailing vehicles, which adopts the following technical solutions:

[0008] A method for identifying and warning abnormal aggregation of online ride-hailing vehicles, comprising:

[0009] Preprocessing the historical traffic data and historical environmental data of the target area in a set time period, and performing time series analysis on the preprocessed historical traffic data and historical environmental data to obtain historical traffic time series data and historical environmental time series data, wherein the historical traffic time series data includes online car-hailing operation time series data, subway passenger flow time series data, and bus passenger flow time series data;

[0010] Based on the online car-hailing operation time series data, construct an online car-hailing grid image dataset of the target area, wherein the online car-hailing grid image dataset represents the distribution of all online car-hailing vehicles at different times;

[0011] Based on the online car-hailing grid image dataset, subway passenger flow time series data and bus passenger flow time series data, historical environment time series data and preset feature extraction rules, determine the feature dataset of the target area, wherein the feature dataset characterizes the passenger flow change trend of online car-hailing, subway and bus in different environments;

[0012] Inputting the feature data set into a preset online car-hailing vehicle quantity prediction model to predict the number of online car-hailing vehicles arriving and the road congestion index at each location in the target area within a target time period; wherein the preset online car-hailing vehicle quantity prediction model is trained based on a sample set including historical feature data of different locations, corresponding actual number of online car-hailing vehicle arrivals and actual road congestion index, and the target time period represents a future time period;

[0013] The number of online-hailing vehicles arriving at various locations in the target area during the target time period and the road congestion index are input into a preset online-hailing vehicle early warning model to obtain early warning information.

[0014] The beneficial effect of the present invention is that through the combination of historical data analysis and the online car-hailing quantity prediction model, the method can predict in advance the arrival volume of online car-hailing vehicles and road congestion in the target area within a specific time period in the future. By considering the temporal and spatial characteristics of traffic data, the fusion of multi-dimensional features such as environmental factors helps to more accurately characterize the traffic operation rules and environmental impacts in the target area. By conducting real-time prediction and early warning of abnormal aggregation of online car-hailing vehicles, effective in-process supervision of online car-hailing operations is carried out, the efficiency and ability of relevant management departments to respond to emergencies in online car-hailing operations are improved, and the traffic risks of abnormal aggregation of online car-hailing vehicles under specific conditions are reduced.

[0015] Based on the above technical solution, the present invention can also be improved as follows.

[0016] Furthermore, the step of constructing a grid image dataset of online-hailing vehicles in the target area based on the online-hailing vehicle operation time series data includes:

[0017] The step of constructing a grid image dataset of online-hailing vehicles in the target area based on the online-hailing vehicle operation time series data includes:

[0018] Aggregate the online ride-hailing operation time series data according to the preset time period and preset road network grid, and count the number of online ride-hailing arrivals in each grid in each time period;

[0019] Map the arrival volume of online-hailing vehicles of each grid corresponding to each time period onto the corresponding grid image, respectively, to generate a pseudo image corresponding to each time period, wherein each of the pseudo images includes a plurality of grids, one grid corresponds to a road network grid, one road network grid corresponds to an arrival volume of online-hailing vehicles, and the pixel value of each grid in each of the pseudo images represents the arrival volume of online-hailing vehicles of the grid in the corresponding time period;

[0020] Based on all the pseudo images, a grid image dataset of online ride-hailing vehicles in the target area is constructed.

[0021] The beneficial effect of adopting the above further solution is that by constructing pseudo images corresponding to different time periods, the grid image dataset not only contains spatial information, but also retains information in the time dimension. This enables the model to learn the dynamic change characteristics of online car-hailing operation data over time, providing more comprehensive data support for subsequent predictions and early warnings.

[0022] Further, the feature data set of the target area is determined based on the online car-hailing grid image data set, subway passenger flow time series data, bus passenger flow time series data, historical environment time series data and preset feature extraction rules, including:

[0023] Based on the time series analysis method, the time feature extraction is performed on the online car-hailing grid image dataset, and based on the image processing technology, the spatial feature extraction is performed on the online car-hailing grid image dataset to obtain the online car-hailing spatiotemporal feature information;

[0024] Based on the time series analysis method, time features are extracted from the subway passenger flow time series data and the bus passenger flow time series data respectively to obtain subway passenger flow time feature information and bus passenger flow time feature information;

[0025] One-hot encoding is performed on the weather data and time data in the historical environmental time series data to obtain weather numerical feature information and time numerical feature information;

[0026] The spatiotemporal characteristic information of online car-hailing, the temporal characteristic information of subway passenger flow, the temporal characteristic information of bus passenger flow, the weather numerical characteristic information and the time numerical characteristic information are integrated to obtain a feature data set of the target area.

[0027] The beneficial effect of adopting the above further scheme is that this method realizes multi-dimensional and comprehensive feature extraction of traffic conditions and environmental factors in the target area by integrating data from multiple modes of transportation such as online car-hailing, subways, and buses, as well as historical environmental data (such as weather). This multi-dimensional feature fusion helps to more accurately characterize the traffic operation rules and environmental impacts in the target area.

[0028] By constructing a feature dataset that includes multiple modes of transportation and environmental factors, the trained model can learn more comprehensive and complex feature patterns. The prediction model can show stronger generalization and robustness when dealing with different scenarios and conditions.

[0029] Furthermore, the method for generating the preset online car-hailing warning model includes:

[0030] Based on the set time period, historical online-hailing vehicle arrival information for multiple historical periods in the target area is obtained, wherein the historical online-hailing vehicle arrival information includes the arrival volume of each sample point corresponding to each location in the target area, and each sample point represents a time point;

[0031] Based on all historical online car-hailing vehicle arrival information and a preset average trend calculation formula, calculate the average trend of online car-hailing vehicle arrivals corresponding to each location in all the historical periods;

[0032] For any position in the target area, based on the arrival amount of each sample point at the position, the average trend and a preset standard deviation calculation formula, the standard deviation of all sample points at the position is calculated;

[0033] For any location in the target area, a warning envelope is generated based on the standard deviation of each sample point at the location, wherein the warning envelope represents the normal fluctuation range of the arrival volume of online ride-hailing vehicles;

[0034] Based on each of the warning envelopes, an online car-hailing warning model is generated.

[0035] The beneficial effect of adopting the above further scheme is: by calculating the average trend and standard deviation of the arrival volume of online-hailing vehicles at each location based on the historical online-hailing vehicle arrival information of multiple historical periods, and then generating a respective warning envelope for each location in the target area, this method can more accurately reflect the normal fluctuation range of the arrival volume of online-hailing vehicles.

[0036] Since the warning envelope is dynamically generated based on historical data, it can automatically adjust as the operating conditions of online ride-hailing services change. This dynamic adaptability enables the warning model to better adapt to the complexity and variability of the urban traffic environment, improving the real-time and effectiveness of the warning.

[0037] Furthermore, the arrival volume of online-hailing vehicles and the road congestion index of the target area within the target time period are input into a preset online-hailing vehicle early warning model to obtain early warning information, including:

[0038] Based on the arrival volume of online-hailing vehicles at various locations in the target area during the target time period, a prediction curve of the arrival volume of online-hailing vehicles corresponding to the warning envelope of each location is generated;

[0039] For any location in the target area, the comparison result is determined based on the online car-hailing arrival forecast curve, road congestion index and warning envelope of the location;

[0040] Based on the comparison results of each position in the target area, warning information is obtained, and the warning information includes warning position information, warning level of each warning position and warning time information.

[0041] The beneficial effect of adopting the above further scheme is that by comprehensively considering the arrival volume of online ride-hailing vehicles and the road congestion index, the model can more comprehensively evaluate the traffic conditions at different locations in the target area, thereby generating more accurate warning information. By dividing the target area into multiple locations and calculating the prediction curve of the arrival volume of online ride-hailing vehicles and the warning envelope for each location respectively, the spatial refinement of the warning information is achieved. This allows traffic management departments to formulate more accurate traffic management strategies for specific locations and improve management efficiency. The generated warning information includes warning location information, warning level and warning time information, which provides important decision-making basis for urban traffic management departments and online ride-hailing platforms.

[0042] Furthermore, the comparison result is determined based on the prediction curve of the arrival volume of online car-hailing vehicles at the location, the road congestion index and the warning envelope, including:

[0043] Compare the predicted arrival curve of online-hailing vehicles at the location with the warning envelope. If the predicted arrival curve of online-hailing vehicles at the location exceeds the normal fluctuation range of the warning envelope, determine the warning level based on the arrival volume of online-hailing vehicles that exceeds the normal fluctuation range and the road congestion index;

[0044] Determine warning time information based on the start time and end time that are beyond the normal fluctuation range;

[0045] Generate comparison results based on warning level, warning time information and warning location.

[0046] The beneficial effect of adopting the above further scheme is that by comparing the online car-hailing arrival forecast curve and the warning envelope at each location, it is possible to accurately identify which locations' traffic conditions are about to or have exceeded the normal range, thereby determining the warning location. This precise positioning helps traffic management departments and online car-hailing platforms to concentrate resources to deal with traffic problems in key areas. When the online car-hailing arrival forecast curve exceeds the normal fluctuation range of the warning envelope, the warning level is determined based on the degree of excess and the road congestion index. By recording the start and end times that exceed the normal fluctuation range, the warning time information can be clarified. And make preparations in advance based on the warning time information, such as adjusting traffic signals, increasing police deployment, or dispatching more online car-hailing vehicles, to cope with the upcoming traffic peak or congestion.

[0047] Furthermore, the preprocessing of the historical traffic data and historical environmental data of the set time period corresponding to the target area includes:

[0048] Performing data missing processing on both the historical traffic data and the historical environment data to obtain first historical traffic data and first historical environment data respectively;

[0049] Performing data anomaly processing on both the first historical traffic data and the first historical environment data to obtain second historical traffic data and second historical environment data respectively;

[0050] Data standardization is performed on both the second historical traffic data and the second historical environment data to obtain historical traffic time series data and historical environment time series data, respectively.

[0051] The beneficial effect of adopting the above further scheme is: pre-processing the historical traffic data and historical environmental data of the target area, including data missing processing, data anomaly processing and data standardization processing, can ensure the integrity of the data and improve the accuracy of the data.

[0052] In the second aspect, the present application provides an abnormal gathering identification and early warning system for online ride-hailing vehicles, which adopts the following technical solutions:

[0053] An abnormal gathering identification and early warning system for online ride-hailing vehicles, comprising:

[0054] A preprocessing module is used to preprocess the historical traffic data and historical environmental data of the target area in a set time period, and perform time series analysis on the preprocessed historical traffic data and historical environmental data to obtain historical traffic time series data and historical environmental time series data, wherein the historical traffic time series data includes online car-hailing operation time series data, subway passenger flow time series data, and bus passenger flow time series data;

[0055] A construction module is used to construct a grid image dataset of online-hailing vehicles in the target area based on the online-hailing vehicle operation time series data, wherein the grid image dataset of online-hailing vehicles represents the distribution of all online-hailing vehicles at different times;

[0056] A feature extraction module, for determining a feature data set of the target area based on the online car-hailing grid image data set, subway passenger flow time series data and bus passenger flow time series data, historical environment time series data and preset feature extraction rules, wherein the feature data set characterizes the passenger flow change trend of online car-hailing, subway and bus in different environments;

[0057] A prediction module, for inputting the feature data set into a preset online car-hailing vehicle quantity prediction model, and predicting the online car-hailing vehicle quantity arrival and road congestion index at each location in the target area within a target time period; wherein the preset online car-hailing vehicle quantity prediction model is trained based on a sample set including historical feature data of different locations, corresponding actual online car-hailing vehicle quantity arrival and actual road congestion index, and the target time period represents a future time period;

[0058] The early warning module is used to input the arrival volume of online-hailing vehicles and the road congestion index at each location in the target area within the target time period into a preset online-hailing vehicle early warning model to obtain early warning information.

[0059] In a third aspect, the present application provides an electronic device, which adopts the following technical solution:

[0060] An electronic device comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute the method for identifying and warning of abnormal aggregation of online-hailing vehicles as described in any one of the first aspects.

[0061] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0062] A computer-readable storage medium storing a computer program that can be loaded by a processor and execute the method for identifying and warning of abnormal aggregation of online-hailing vehicles as described in any one of the first aspects.

[0063] Additional aspects and advantages of the present application will be partially given in the following description, which will become apparent from the following description, or will be understood through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A flowchart of a method for identifying and warning abnormal aggregation of online ride-hailing vehicles provided by an embodiment of the present invention;

[0065] Figure 2 A schematic diagram of an online car-hailing prediction model provided by one embodiment of the present invention;

[0066] Figure 3 A schematic diagram of an online car-hailing early warning model provided by an embodiment of the present invention;

[0067] Figure 4 A module block diagram of an abnormal aggregation identification and early warning system for online ride-hailing vehicles provided by an embodiment of the present invention;

[0068] Figure 5 A structural block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the following will be combined with the appended Figure 1 To Attachment Figure 5 , the technical solutions in the embodiments of the present application are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0070] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0071] An embodiment of the present application provides a method for identifying and warning of abnormal aggregation of online ride-hailing vehicles. The method can be executed by an electronic device, which can be a server or a mobile terminal device. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services; the mobile terminal device can be a laptop computer, a desktop computer, etc., but is not limited to this.

[0072] like Figure 1 As shown, a method for identifying and warning abnormal aggregation of online ride-hailing vehicles includes steps S1 to S5:

[0073] Step S1, preprocessing the historical traffic data and historical environmental data of the target area in a set time period, and performing time series analysis on the preprocessed historical traffic data and historical environmental data to obtain historical traffic time series data and historical environmental time series data, wherein the historical traffic time series data includes online car-hailing operation time series data, subway passenger flow time series data, and bus passenger flow time series data;

[0074] In one embodiment, historical traffic data includes, but is not limited to, online car-hailing operation data, subway passenger flow data, and bus passenger flow data. Among them, online car-hailing operation data can be obtained through the background database or data interface of the online car-hailing platform, and online car-hailing operation data includes the order volume, passenger boarding and alighting time, driving trajectory, etc. in the target area during the set time period. Subway passenger flow data and bus passenger flow data can collect the corresponding subway passenger flow and bus passenger flow through detectors at various monitoring stations in the target area. Subway passenger flow data can also be obtained from the data system of the subway operating company, and the data content includes the number of people entering and exiting each station, train running intervals, etc.; bus passenger flow data can also be obtained through the data system of the bus company or the intelligent bus dispatching platform, and the data content covers the passenger flow and schedule information of the bus line.

[0075] Environmental data includes weather conditions, special events (such as large-scale events, traffic accidents, road construction, etc.) and holidays. It is obtained from meteorological departments, environmental protection departments, traffic management departments or related news websites, but is not limited to them.

[0076] In one implementation, step S1 specifically includes the following sub-steps:

[0077] Step S11, the historical traffic data and the historical environment data are both processed for missing data, and the first historical traffic data and the first historical environment data are obtained respectively; due to the possibility of missing data due to detection equipment failure, network transmission interruption, etc., it is necessary to complete the data in the case of missing information to ensure the integrity and accuracy of the data. In the example of the present invention, the traffic data filling based on K nearest neighbors is adopted.

[0078] Step S12, data anomaly processing is performed on both the first historical traffic data and the first historical environmental data to obtain second historical traffic data and second historical environmental data respectively; in real traffic scenarios, due to system sampling errors and accidental abnormal factors, outliers that deviate from the actual situation may be generated, which will have a certain impact on time series analysis and need to be processed. In the example of the present invention, outlier detection based on statistical distribution is adopted, and data smoothing technology is used to eliminate outliers.

[0079] Step S13, performing data normalization processing on the second historical traffic data and the second historical environment data, respectively obtaining historical traffic time series data and historical environment time series data. Data normalization can reduce the impact of data size, characteristics, distribution differences, etc. on the model. In the example of the present invention, the Min-Max normalization method is adopted, which maps the original data to the interval [0,1] through linear transformation.

[0080] Step S2, constructing a grid image dataset of online-hailing vehicles in the target area based on the online-hailing vehicle operation time series data, wherein the grid image dataset of online-hailing vehicles represents the distribution of all online-hailing vehicles at different times;

[0081] In one embodiment, specifically, step S2 includes:

[0082] Aggregate the online ride-hailing operation time series data according to the preset time period and preset road network grid, and count the number of online ride-hailing arrivals in each grid in each time period;

[0083] Map the arrival volume of online-hailing vehicles of each grid corresponding to each time period onto the corresponding grid image, respectively, to generate a pseudo image corresponding to each time period, wherein each of the pseudo images includes a plurality of grids, one grid corresponds to a road network grid, one road network grid corresponds to an arrival volume of online-hailing vehicles, and the pixel value of each grid in each of the pseudo images represents the arrival volume of online-hailing vehicles of the grid in the corresponding time period;

[0084] Based on all the pseudo images, a grid image dataset of online ride-hailing vehicles in the target area is constructed.

[0085] In the above implementation, first, the length of the preset time period is determined according to the analysis requirements, and the time period will be used as the basic unit for aggregating data, for example, the data is aggregated at intervals of 15 minutes in time series.

[0086] Divide the target area into several road network grids of equal or unequal sizes. The size and shape of the grid can be adjusted according to the analysis accuracy and regional characteristics. Each road network grid should have a unique identifier. After that, traverse all online car-hailing operation data (such as online car-hailing order data), determine whether the start or end position of the order falls within a certain grid, and add it to the arrival count of the corresponding road network grid. For each road network grid, count the number of online car-hailing arrivals in each time period.

[0087] After data aggregation is completed, the online ride-hailing arrival data for each time period is mapped to the corresponding grid image. The grid image is a blank image with the same size as the road network grid. Each pixel or pixel block corresponds to a road network grid. The online ride-hailing arrival data of each grid is used as the pixel value of the corresponding pixel or pixel block to generate a pseudo image corresponding to each time period.

[0088] Finally, the pseudo images of all time periods are organized into a grid image dataset of online ride-hailing vehicles in chronological order.

[0089] Step S3, determining a feature data set of the target area based on the online car-hailing grid image data set, subway passenger flow time series data and bus passenger flow time series data, historical environment time series data and preset feature extraction rules, wherein the feature data set characterizes the passenger flow change trend of online car-hailing, subway and bus in different environments;

[0090] In one embodiment, the feature data set of the target area is determined based on the online car-hailing grid image data set, subway passenger flow time series data, bus passenger flow time series data, historical environment time series data and preset feature extraction rules, including:

[0091] Based on the time series analysis method, the time feature extraction is performed on the online car-hailing grid image dataset, and based on the image processing technology, the spatial feature extraction is performed on the online car-hailing grid image dataset to obtain the online car-hailing spatiotemporal feature information;

[0092] Based on the time series analysis method, time features are extracted from the subway passenger flow time series data and the bus passenger flow time series data respectively to obtain subway passenger flow time feature information and bus passenger flow feature time information;

[0093] One-hot encoding is performed on the weather data and time data in the historical environmental time series data to obtain weather numerical feature information and time numerical feature information;

[0094] The spatiotemporal characteristic information of online car-hailing, the temporal characteristic information of subway passenger flow, the temporal characteristic information of bus passenger flow, the weather numerical characteristic information and the time numerical characteristic information are integrated to determine the characteristic data set of the target area.

[0095] In the above implementation, time series analysis methods (such as ARIMA model, exponential smoothing method, grayscale prediction method, etc.) are applied to perform time dimension analysis on the online car-hailing grid image data set, subway passenger flow time series data, and bus passenger flow time series data, and extract the changing trend and time characteristics of the online car-hailing arrival volume of each grid in different time periods; extract the changing trend of subway passenger flow and the periodicity of subway flow at each monitoring station to obtain the time characteristic information of subway passenger flow; extract the changing trend of bus flow and the periodicity of bus flow at each monitoring station to obtain the time information of bus passenger flow.

[0096] Image processing techniques (such as convolutional neural networks (CNNs) and image segmentation) are used to analyze the spatial dimension of the online ride-hailing grid image dataset. By identifying the areas with high online ride-hailing arrival volume and low online ride-hailing arrival volume in the image and the spatial distribution relationship between them, the spatial characteristics of online ride-hailing, such as the aggregation and dispersion of online ride-hailing, are extracted.

[0097] For weather data, different weather conditions (such as sunny days, rainy days, snowy days, etc.) can be one-hot encoded to obtain weather numerical feature information.

[0098] For time data, you can divide time into different time periods (such as weekdays, weekends, holidays, etc.) as needed, and perform One-hot encoding to obtain time numerical feature information.

[0099] Step S4, inputting the feature data set into a preset online car-hailing vehicle quantity prediction model to predict the arrival volume of online car-hailing vehicles and the road congestion index at each location in the target area within a target time period; wherein the preset online car-hailing vehicle quantity prediction model is trained based on a sample set including historical feature data of different locations, corresponding actual arrival volume of online car-hailing vehicles and actual road congestion index, and the target time period represents a future time period;

[0100] In one embodiment, if Figure 2 As shown in the figure, the convolution neural network (CNN) model is trained by training a sample set containing historical feature data of different locations, the corresponding actual number of online ride-hailing arrivals and the actual road congestion index. These historical feature data cover multiple time periods and different traffic conditions to ensure that the model can learn effective features and rules.

[0101] The model combines the attention mechanism to autonomously explore the importance of input features to output predictions. In order to capture the strong spatial dependence of vehicle or pedestrian traffic between different urban areas, a multi-layer CNN stacking method is used to solve the problem of the limited receptive field of a single-layer CNN.

[0102] During the training process, the accuracy and stability of the model are verified by using methods such as leaving out samples or cross-validation, and the mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE) are used to evaluate the model performance.

[0103] Step S5, inputting the arrival volume of online-hailing vehicles and the road congestion index at each location in the target area within the target time period into a preset online-hailing vehicle early warning model to obtain early warning information.

[0104] In one embodiment, Figure 3 This is a schematic diagram of the online car-hailing warning model. Figure 3 As shown, the method for generating the online car-hailing warning model includes the following sub-steps:

[0105] Step S51, based on the set time period, obtaining historical online-hailing vehicle arrival information for multiple historical periods in the target area, wherein the historical online-hailing vehicle arrival information includes the arrival volume of each sample point corresponding to each location in the target area, and each sample point represents a time point;

[0106] Step S52, based on all historical online-hailing vehicle arrival information and a preset average trend calculation formula, calculate the average trend of the online-hailing vehicle arrival corresponding to each location in all the historical periods;

[0107] Step S53, for any position in the target area, based on the arrival amount of each sample point at the position, the average trend and a preset standard deviation calculation formula, calculate the standard deviation of all sample points at the position;

[0108] Step S54, for any location in the target area, based on the standard deviation of each of the sample points at the location, generating a warning envelope, wherein the envelope represents a normal fluctuation range of the arrival volume of online-hailing vehicles;

[0109] Step S55, generating an online car-hailing warning model based on each of the warning envelopes.

[0110] In the above implementation, the sample point is a specific time point or time period in each historical period, which may be one hour, two hours, or one day, but is not limited thereto.

[0111] The average trend calculation formula for all historical periods is:

[0112] ;

[0113] The standard deviation calculation formula is:

[0114] ;

[0115] in, represents the average trend, represents the number of online ride-hailing vehicles arriving in the region at the nth sample point in the jth week, where N represents the historical period. Represents standard deviation.

[0116] The method for generating the warning envelope is: taking the average trend line obtained in step S52 as the center, extending 95% confidence levels upward and downward according to the standard deviation calculated in step S53, to generate a warning envelope. The warning envelope includes an upper trend envelope and a lower trend envelope. When the predicted value of the arrival volume of online ride-hailing vehicles exceeds the upper trend envelope, a warning is triggered.

[0117] In one embodiment, the arrival volume of online-hailing vehicles and the road congestion index in the target area within the target time period are input into a preset online-hailing vehicle early warning model to obtain early warning information, including:

[0118] Based on the arrival volume of online-hailing vehicles at various locations in the target area during the target time period, a prediction curve of the arrival volume of online-hailing vehicles corresponding to the warning envelope of each location is generated;

[0119] For any location in the target area, a comparison result is determined based on the online car-hailing vehicle arrival prediction curve, road congestion index and warning envelope at that location; in an embodiment of the present application, the online car-hailing vehicle arrival prediction curve and the warning envelope at that location are compared. If the online car-hailing vehicle arrival prediction curve at that location exceeds the normal fluctuation range of the warning envelope, the warning level is determined based on the online car-hailing vehicle arrival volume and the road congestion index that exceed the normal fluctuation range; the warning time information is determined based on the start time and end time that exceed the normal fluctuation range; and a comparison result is generated based on the warning level, warning time information and warning location.

[0120] Based on the comparison results of each location in the target area, warning information is obtained, which includes warning location information, warning level of each warning location, and warning time information. By comprehensively considering the arrival volume of online ride-hailing vehicles and the road congestion index, the model can more comprehensively evaluate the traffic conditions at different locations in the target area, thereby generating more accurate warning information.

[0121] This method combines historical data analysis with an online car-hailing vehicle quantity prediction model. This method can predict in advance the arrival volume of online car-hailing vehicles and road congestion in the target area within a specific time period in the future. By considering the temporal and spatial characteristics of traffic data and integrating multi-dimensional features such as environmental factors, it helps to more accurately characterize the traffic operation rules and environmental impacts in the target area. By conducting real-time prediction and early warning of abnormal aggregation of online car-hailing vehicles, effective in-process supervision of online car-hailing operations can be carried out, the efficiency and ability of relevant management departments to respond to emergencies in online car-hailing operations can be improved, and the traffic risks of abnormal aggregation of online car-hailing vehicles under specific conditions can be reduced.

[0122] Figure 4 A structural schematic diagram of an abnormal aggregation identification and early warning system 200 for online ride-hailing vehicles is shown.

[0123] like Figure 4 As shown, a network-hailing vehicle abnormal aggregation identification and early warning system 200 mainly includes:

[0124] A preprocessing module 201 is used to preprocess the historical traffic data and historical environmental data of the target area in a set time period, and perform time series analysis on the preprocessed historical traffic data and historical environmental data to obtain historical traffic time series data and historical environmental time series data, wherein the historical traffic time series data includes online car-hailing operation time series data, subway passenger flow time series data, and bus passenger flow time series data;

[0125] A construction module 202 is used to construct a grid image dataset of online-hailing vehicles in the target area based on the online-hailing vehicle operation time series data, wherein the grid image dataset of online-hailing vehicles represents the distribution of all online-hailing vehicles at different times;

[0126] A feature extraction module 203 is used to determine a feature data set of the target area based on the online car-hailing grid image data set, subway passenger flow time series data and bus passenger flow time series data, historical environment time series data and preset feature extraction rules, wherein the feature data set characterizes the passenger flow change trend of online car-hailing, subway and bus in different environments;

[0127] The prediction module 204 is used to input the feature data set into a preset online car-hailing vehicle quantity prediction model to predict the arrival volume of online car-hailing vehicles and the road congestion index at each location in the target area within a target time period; wherein the preset online car-hailing vehicle quantity prediction model is trained based on a sample set including historical feature data of different locations, corresponding actual arrival volume of online car-hailing vehicles and actual road congestion index, and the target time period represents a future time period;

[0128] The early warning module 205 is used to input the arrival volume of online-hailing vehicles and the road congestion index at each location in the target area within the target time period into a preset online-hailing vehicle early warning model to obtain early warning information.

[0129] Optionally, the construction module 202 is specifically used for:

[0130] Aggregate the online ride-hailing operation time series data according to the preset time period and preset road network grid, and count the number of online ride-hailing arrivals in each grid in each time period;

[0131] Map the arrival volume of online-hailing vehicles of each grid corresponding to each time period onto the corresponding grid image, respectively, to generate a pseudo image corresponding to each time period, wherein each of the pseudo images includes a plurality of grids, one grid corresponds to a road network grid, one road network grid corresponds to an arrival volume of online-hailing vehicles, and the pixel value of each grid in each of the pseudo images represents the arrival volume of online-hailing vehicles of the grid in the corresponding time period;

[0132] Based on all the pseudo images, a grid image dataset of online ride-hailing vehicles in the target area is constructed.

[0133] Optionally, the feature extraction module 203 is specifically used for:

[0134] Based on the time series analysis method, the time feature extraction is performed on the online car-hailing grid image dataset, and based on the image processing technology, the spatial feature extraction is performed on the online car-hailing grid image dataset to obtain the online car-hailing spatiotemporal feature information;

[0135] Based on the time series analysis method, time features are extracted from the subway passenger flow time series data and the bus passenger flow time series data respectively to obtain subway passenger flow time feature information and bus passenger flow time feature information;

[0136] One-hot encoding is performed on the weather data and time data in the historical environmental time series data to obtain weather numerical feature information and time numerical feature information;

[0137] The spatiotemporal characteristic information of online car-hailing, the temporal characteristic information of subway passenger flow, the temporal characteristic information of bus passenger flow, the weather numerical characteristic information and the time numerical characteristic information are integrated to obtain a feature data set of the target area.

[0138] Optionally, the method for generating the online car-hailing warning model in the warning module 205 includes:

[0139] Based on the set time period, historical online-hailing vehicle arrival information for multiple historical periods in the target area is obtained, wherein the historical online-hailing vehicle arrival information includes the arrival volume of each sample point corresponding to each location in the target area, and each sample point represents a time point;

[0140] Based on all historical online car-hailing vehicle arrival information and a preset average trend calculation formula, calculate the average trend of online car-hailing vehicle arrivals corresponding to each location in all the historical periods;

[0141] For any position in the target area, based on the arrival amount of each sample point at the position, the average trend and a preset standard deviation calculation formula, the standard deviation of all sample points at the position is calculated;

[0142] For any location in the target area, a warning envelope is generated based on the standard deviation of each sample point at the location, wherein the warning envelope represents the normal fluctuation range of the arrival volume of online ride-hailing vehicles;

[0143] Based on each of the warning envelopes, an online car-hailing warning model is generated.

[0144] Optionally, the early warning module 205 includes:

[0145] A generation submodule, for generating a prediction curve of the arrival volume of online-hailing vehicles corresponding to the warning envelope of each location based on the arrival volume of online-hailing vehicles at each location in the target area within the target time period;

[0146] A comparison submodule is used to compare the online car-hailing arrival prediction curve and the warning envelope at any location in the target area to obtain a comparison result;

[0147] The early warning submodule is used to obtain early warning information based on the comparison results of each position in the target area, and the early warning information includes early warning position information, the early warning level of each early warning position and the early warning time information.

[0148] Optional, contrast submodule, specifically used for:

[0149] Compare the predicted arrival curve of online-hailing vehicles at the location with the warning envelope. If the predicted arrival curve of online-hailing vehicles at the location exceeds the normal fluctuation range of the warning envelope, determine the warning level based on the arrival volume of online-hailing vehicles that exceeds the normal fluctuation range and the road congestion index;

[0150] Determine warning time information based on the start time and end time that are beyond the normal fluctuation range;

[0151] Generate comparison results based on warning level, warning time information and warning location.

[0152] Optionally, the preprocessing module 201 is specifically used for:

[0153] Performing data missing processing on both the historical traffic data and the historical environment data to obtain first historical traffic data and first historical environment data respectively;

[0154] Performing data anomaly processing on both the first historical traffic data and the first historical environment data to obtain second historical traffic data and second historical environment data respectively;

[0155] Data standardization is performed on both the second historical traffic data and the second historical environment data to obtain historical traffic time series data and historical environment time series data, respectively.

[0156] In one example, the module in any of the above devices can be one or more integrated circuits configured to implement the above methods, such as: one or more application specific integrated circuits (ASICs), or, one or more digital signal processors (DSPs), or, one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0157] For another example, when the modules in the device can be implemented in the form of a processing element scheduling program, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program. For another example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0158] Various objects such as various messages / information / equipment / network elements / systems / devices / actions / operations / processes / concepts that may appear in this application are named. It can be understood that these specific names do not constitute a limitation on the relevant objects. The names assigned may change with factors such as scenarios, contexts or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from the functions and technical effects embodied / executed in the technical scheme.

[0159] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0160] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0161] Figure 5 This is a structural block diagram of an electronic device 300 according to an embodiment of the present application.

[0162] like Figure 5 As shown, the electronic device 300 includes a processor 301 and a memory 302 , and may further include one or more of an information input / information output (I / O) interface 303 and a communication component 304 .

[0163] Among them, the processor 301 is used to control the overall operation of the electronic device 300 to complete all or part of the steps in the above-mentioned network car abnormal aggregation identification and warning method; the memory 302 is used to store various types of data to support the operation of the electronic device 300, and these data may include, for example, instructions for any application or method used to operate on the electronic device 300, and application-related data. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, EPROM), programmable read-only memory (Programmable Read-Only Memory, PROM), read-only memory (Read-Only Memory, ROM), magnetic memory, flash memory, disk or optical disk. One or more.

[0164] The I / O interface 303 provides an interface between the processor 301 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 304 is used to test the wired or wireless communication between the electronic device 300 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 304 can include: Wi-Fi components, Bluetooth components, NFC components.

[0165] The communication bus 305 may include a path to transmit information between the above components. The communication bus 305 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 305 may be divided into an address bus, a data bus, a control bus, etc.

[0166] The electronic device 300 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components, and is used to execute the abnormal aggregation identification and early warning method for online car-hailing vehicles given in the above embodiment.

[0167] The computer-readable storage medium provided in the embodiments of the present application is introduced below. The computer-readable storage medium described below and the method for identifying and warning abnormal aggregation of online ride-hailing vehicles described above can be referenced to each other.

[0168] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for identifying and warning of abnormal aggregation of online ride-hailing vehicles are implemented.

[0169] The computer-readable storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0170] The terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article, or apparatus.

[0171] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of application involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned application concept. For example, the above features are replaced with (but not limited to) technical features with similar functions applied in the present application.

Claims

1. A method for identifying and warning abnormal aggregation of online ride-hailing vehicles, characterized in that: include: Preprocessing the historical traffic data and historical environmental data of the target area in a set time period, and performing time series analysis on the preprocessed historical traffic data and historical environmental data to obtain historical traffic time series data and historical environmental time series data, wherein the historical traffic time series data includes online car-hailing operation time series data, subway passenger flow time series data, and bus passenger flow time series data; Based on the online car-hailing operation time series data, construct an online car-hailing grid image dataset of the target area, wherein the online car-hailing grid image dataset represents the distribution of all online car-hailing vehicles at different times; Based on the online car-hailing grid image dataset, subway passenger flow time series data and bus passenger flow time series data, historical environment time series data and preset feature extraction rules, determine the feature dataset of the target area, wherein the feature dataset characterizes the passenger flow change trend of online car-hailing, subway and bus in different environments; Inputting the feature data set into a preset online car-hailing vehicle quantity prediction model to predict the number of online car-hailing vehicles arriving and the road congestion index at each location in the target area within a target time period; wherein the preset online car-hailing vehicle quantity prediction model is trained based on a sample set including historical feature data of different locations, corresponding actual number of online car-hailing vehicle arrivals and actual road congestion index, and the target time period represents a future time period; The number of online-hailing vehicles arriving at various locations in the target area during the target time period and the road congestion index are input into a preset online-hailing vehicle early warning model to obtain early warning information.

2. According to claim 1, a method for identifying and warning abnormal aggregation of online ride-hailing vehicles is characterized in that: The step of constructing a grid image dataset of online-hailing vehicles in the target area based on the online-hailing vehicle operation time series data includes: Aggregate the online ride-hailing operation time series data according to the preset time period and preset road network grid, and count the number of online ride-hailing arrivals in each grid in each time period; Map the arrival volume of online-hailing vehicles of each grid corresponding to each time period onto the corresponding grid image, respectively, to generate a pseudo image corresponding to each time period, wherein each of the pseudo images includes a plurality of grids, one grid corresponds to a road network grid, one road network grid corresponds to an arrival volume of online-hailing vehicles, and the pixel value of each grid in each of the pseudo images represents the arrival volume of online-hailing vehicles of the grid in the corresponding time period; Based on all the pseudo images, a grid image dataset of online ride-hailing vehicles in the target area is constructed.

3. A method for identifying and warning abnormal aggregation of online ride-hailing vehicles according to claim 1 or 2, characterized in that: The determining of the feature data set of the target area based on the online car-hailing grid image data set, subway passenger flow time series data, bus passenger flow time series data, historical environment time series data and preset feature extraction rules includes: Based on the time series analysis method, the time feature extraction is performed on the online car-hailing grid image dataset, and based on the image processing technology, the spatial feature extraction is performed on the online car-hailing grid image dataset to obtain the online car-hailing spatiotemporal feature information; Based on the time series analysis method, time features are extracted from the subway passenger flow time series data and the bus passenger flow time series data respectively to obtain subway passenger flow time feature information and bus passenger flow time feature information; One-hot encoding is performed on the weather data and time data in the historical environmental time series data to obtain weather numerical feature information and time numerical feature information; The spatiotemporal characteristic information of online car-hailing, the temporal characteristic information of subway passenger flow, the temporal characteristic information of bus passenger flow, the weather numerical characteristic information and the time numerical characteristic information are integrated to obtain a feature data set of the target area.

4. The method for identifying and warning abnormal aggregation of online ride-hailing vehicles according to claim 1, characterized in that: The method for generating the preset online car-hailing warning model includes: Based on the set time period, historical online-hailing vehicle arrival information for multiple historical periods in the target area is obtained, wherein the historical online-hailing vehicle arrival information includes the arrival volume of each sample point corresponding to each location in the target area, and each sample point represents a time point; Based on all historical online car-hailing vehicle arrival information and a preset average trend calculation formula, calculate the average trend of online car-hailing vehicle arrivals corresponding to each location in all the historical periods; For any position in the target area, based on the arrival amount of each sample point at the position, the average trend and a preset standard deviation calculation formula, the standard deviation of all sample points at the position is calculated; For any location in the target area, a warning envelope is generated based on the standard deviation of each sample point at the location, wherein the warning envelope represents the normal fluctuation range of the arrival volume of online ride-hailing vehicles; Based on each of the warning envelopes, an online car-hailing warning model is generated.

5. The method for identifying and warning abnormal aggregation of online ride-hailing vehicles according to claim 4, characterized in that: The step of inputting the arrival volume of online-hailing vehicles and the road congestion index of the target area within the target time period into a preset online-hailing vehicle early warning model to obtain early warning information includes: Based on the arrival volume of online-hailing vehicles at various locations in the target area during the target time period, a prediction curve of the arrival volume of online-hailing vehicles corresponding to the warning envelope of each location is generated; For any location in the target area, the comparison result is determined based on the online car-hailing arrival forecast curve, road congestion index and warning envelope of the location; Based on the comparison results of each position in the target area, warning information is obtained, and the warning information includes warning position information, warning level of each warning position and warning time information.

6. The method for identifying and warning abnormal aggregation of online ride-hailing vehicles according to claim 4, characterized in that: The comparison result is determined based on the prediction curve of the arrival volume of online car-hailing vehicles at the location, the road congestion index and the warning envelope, including: Compare the predicted arrival curve of online-hailing vehicles at the location with the warning envelope. If the predicted arrival curve of online-hailing vehicles at the location exceeds the normal fluctuation range of the warning envelope, determine the warning level based on the arrival volume of online-hailing vehicles that exceeds the normal fluctuation range and the road congestion index; Determine warning time information based on the start time and end time that are beyond the normal fluctuation range; Generate comparison results based on warning level, warning time information and warning location.

7. The method for identifying and warning abnormal aggregation of online ride-hailing vehicles according to claim 1, characterized in that: The preprocessing of the historical traffic data and historical environment data for the set time period corresponding to the target area includes: Performing data missing processing on both the historical traffic data and the historical environment data to obtain first historical traffic data and first historical environment data respectively; Performing data anomaly processing on both the first historical traffic data and the first historical environment data to obtain second historical traffic data and second historical environment data respectively; Data standardization is performed on both the second historical traffic data and the second historical environment data to obtain historical traffic time series data and historical environment time series data, respectively.

8. A system for identifying and warning abnormal aggregation of online ride-hailing vehicles, characterized in that: include: A preprocessing module is used to preprocess the historical traffic data and historical environmental data of the target area in a set time period, and perform time series analysis on the preprocessed historical traffic data and historical environmental data to obtain historical traffic time series data and historical environmental time series data, wherein the historical traffic time series data includes online car-hailing operation time series data, subway passenger flow time series data, and bus passenger flow time series data; A construction module is used to construct a grid image dataset of online-hailing vehicles in the target area based on the online-hailing vehicle operation time series data, wherein the grid image dataset of online-hailing vehicles represents the distribution of all online-hailing vehicles at different times; A feature extraction module, for determining a feature data set of the target area based on the online car-hailing grid image data set, subway passenger flow time series data and bus passenger flow time series data, historical environment time series data and preset feature extraction rules, wherein the feature data set characterizes the passenger flow change trend of online car-hailing, subway and bus in different environments; A prediction module, for inputting the feature data set into a preset online car-hailing vehicle quantity prediction model, and predicting the online car-hailing vehicle quantity arrival and road congestion index at each location in the target area within a target time period; wherein the preset online car-hailing vehicle quantity prediction model is trained based on a sample set including historical feature data of different locations, corresponding actual online car-hailing vehicle quantity arrival and actual road congestion index, and the target time period represents a future time period; The early warning module is used to input the arrival volume of online-hailing vehicles and the road congestion index at each location in the target area within the target time period into a preset online-hailing vehicle early warning model to obtain early warning information.

9. An electronic device, characterized in that: comprising a processor coupled to a memory; The processor is configured to execute a computer program stored in the memory, so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The method comprises a computer program or an instruction, and when the computer program or the instruction is executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 7.

Images