A method and device for predicting traffic flow in a highway construction zone and a storage medium

Abstract

The application provides a highway construction zone traffic flow prediction method and device and a storage medium, including: inputting a target construction operation position, operation time and operation duration on a highway into a first network layer of a pre-trained traffic flow prediction model to determine a first traffic flow prediction time period of an upstream adjacent toll gate of the target construction operation position; if the type of the target construction operation position is a first type, inputting the first traffic flow prediction time period into a second network layer of the traffic flow prediction model to predict a first traffic flow sequence corresponding to the traffic flow prediction time period; if the type of the target construction operation position is a second type, determining a second traffic flow prediction time period of an upstream adjacent toll station of the target construction operation position to predict a second traffic flow sequence. The traffic flow of the target construction operation position is adaptively predicted, and the traffic flow of the target construction operation position in a future time period is quickly and accurately predicted.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus and storage medium for predicting traffic flow in highway construction areas. Background Technology

[0002] Many highways are now undergoing increasingly heavy maintenance and expansion work. Given a clear work plan, it's necessary to close off certain sections of the road within a certain area to provide sufficient space for construction machinery and personnel. However, the reduced road space due to construction increases the braking effort required by drivers, potentially causing psychological stress and resistance, thus reducing driving safety. Ensuring smooth traffic flow and safety in construction sections while meeting normal construction requirements, minimizing delays and fuel consumption, and improving driver and passenger comfort are key considerations in current traffic management plans for maintenance and construction. Therefore, effectively predicting future traffic flow in construction areas has become a significant technical challenge. Summary of the Invention

[0003] In view of this, the purpose of this application is to provide a traffic flow prediction method, device and storage medium for highway construction areas. The target construction location, the operation time of the target construction location and the operation duration of the target construction location are input into a traffic flow prediction model obtained by fusing a periodic trend model, a real-time model and a disturbance model of time series. The model adaptively predicts the traffic flow at the target construction location and quickly and accurately predicts the traffic flow at the target construction location in the future time period.

[0004] This application provides a traffic flow prediction method for highway construction areas, the traffic flow prediction method comprising:

[0005] The target construction work location on the highway, the operation time of the target construction work location, and the operation duration of the target construction work location are input into the first network layer of the pre-trained traffic flow prediction model to determine the first traffic flow prediction time period of the upstream adjacent checkpoint of the target construction work location; wherein, the traffic flow prediction model is obtained by fusing a periodic trend model of time series, a real-time model of time series, and a disturbance model.

[0006] If the type of the target construction location is the first type, then the first traffic flow prediction time period is input into the second network layer of the traffic flow prediction model to predict the first traffic flow sequence corresponding to the traffic flow prediction time period; wherein, the first traffic flow sequence is the traffic flow sequence of the upstream adjacent checkpoint of the target construction location;

[0007] If the target construction location is of type two, then the second traffic flow prediction time period of the upstream adjacent toll station of the target construction location is determined, and the second traffic flow prediction time period is input into the second network layer of the traffic flow prediction model to predict the second traffic flow sequence corresponding to the second traffic flow prediction time period; wherein, the second traffic flow sequence is the inbound traffic flow, outbound traffic flow and upstream adjacent checkpoint of the upstream adjacent toll station.

[0008] In one possible implementation, the first traffic flow prediction time period of the upstream adjacent checkpoint of the target construction operation location is determined by the following steps:

[0009] The first distance between the target construction location and the upstream adjacent checkpoint was determined;

[0010] Determine the first target speed among multiple historical vehicle speeds near the upstream checkpoint;

[0011] Based on the ratio of the first distance to the first target speed, the first driving time for the vehicle to reach the target construction work location from the upstream adjacent checkpoint is determined;

[0012] Based on the first driving time, the operation time, the first operation duration, and the current operation time, a first traffic flow prediction time period is determined.

[0013] In one possible implementation, after detecting whether the time difference between the operation time and the current operation time is greater than or equal to a preset time threshold, the traffic flow prediction method further includes:

[0014] Based on multiple recent traffic flow characteristics of the upstream adjacent checkpoints, real-time traffic flow is predicted to determine the real-time traffic flow sequence.

[0015] Based on the sum of the time-based traffic flow sequence, the disturbance traffic flow sequence, and the real-time traffic flow sequence, the first traffic flow sequence of the upstream adjacent checkpoint during the traffic flow prediction time period is predicted.

[0016] In one possible implementation, the second traffic flow prediction time of the upstream adjacent toll station of the target construction site is determined by the following steps:

[0017] A second target speed is determined based on the historical vehicle speeds of multiple upstream adjacent toll stations;

[0018] The second distance between the target construction site and the upstream adjacent toll station was determined;

[0019] Based on the ratio of the second distance to the second target speed, the second driving time for the vehicle to reach the target construction work location from the upstream adjacent toll station is determined;

[0020] The second traffic flow prediction time period is determined based on the second driving time, the operation time, the second operation duration, and the current operation time.

[0021] In one possible implementation, the second traffic flow sequence corresponding to the second traffic flow prediction time period is predicted through the following steps:

[0022] Based on the traffic flow prediction model, the inbound traffic flow time series and the outbound traffic flow sequence of the nearby toll station are predicted during the second traffic flow prediction time period.

[0023] The first sequence is determined by summing the first traffic flow sequence with the inbound traffic flow time sequence;

[0024] Based on the difference between the first sequence and the outbound traffic flow time sequence, the second traffic flow sequence corresponding to the second traffic flow prediction time period is predicted.

[0025] In one possible implementation, the traffic prediction model is determined through the following steps:

[0026] The sample traffic flow data is input into the time series model to train the time training model and output undisturbed prediction data. The undisturbed prediction data and the time features of the undisturbed data in the sample traffic flow data are input into a fully connected neural network to generate the periodic trend model of the time series and determine the first prediction error value of the periodic trend model of the time series.

[0027] The neural network model is trained based on recent sample traffic flow data and the first prediction error value to generate a real-time model of the time series, and a second prediction error value of the real-time model of the time series is determined.

[0028] The neural network model is trained based on historical sample perturbation data and the second prediction error value to generate the perturbation model;

[0029] The periodic trend model, the real-time model, and the disturbance model of the time series are fused to generate the traffic prediction model.

[0030] This application embodiment also provides a traffic flow prediction device for highway construction areas, the traffic flow prediction device comprising:

[0031] The first determining module is used to input the target construction operation location on the highway, the operation time of the target construction operation location, and the operation duration of the target construction operation location into the first network layer of the pre-trained traffic flow prediction model to determine the first traffic flow prediction time period of the upstream adjacent checkpoint of the target construction operation location; wherein, the traffic flow prediction model is obtained by fusing a periodic trend model of time series, a real-time model of time series, and a disturbance model.

[0032] The second determining module is used to input the first traffic flow prediction time period into the second network layer of the traffic flow prediction model if the type of the target construction operation location is the first type, and predict the first traffic flow sequence corresponding to the traffic flow prediction time period; wherein, the first traffic flow sequence is the traffic flow sequence of the upstream adjacent checkpoint of the target construction operation location;

[0033] The third determining module is used to determine the second traffic flow prediction time period of the upstream adjacent toll station of the target construction operation location if the type of the target construction operation location is the second type, and input the second traffic flow prediction time period into the second network layer of the traffic flow prediction model to predict the second traffic flow sequence corresponding to the second traffic flow prediction time period; wherein, the second traffic flow sequence is the inbound traffic flow, outbound traffic flow and upstream adjacent checkpoint of the upstream adjacent toll station.

[0034] This application also provides an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the traffic flow prediction method for highway construction areas described above are performed.

[0035] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of the traffic flow prediction method for highway construction areas as described above.

[0036] This application provides a traffic flow prediction method, device, and storage medium for highway construction areas. The traffic flow prediction method includes: inputting the target construction operation location on the highway, the operation time of the target construction operation location, and the operation duration of the target construction operation location into the first network layer of a pre-trained traffic flow prediction model to determine a first traffic flow prediction time period for the upstream adjacent checkpoint of the target construction operation location; wherein, the traffic flow prediction model is obtained by fusing a time series periodic trend model, a time series real-time model, and a disturbance model; if the type of the target construction operation location is a first type, then the first traffic flow prediction time period is input into the... In the second network layer of the traffic flow prediction model, a first traffic flow sequence corresponding to the traffic flow prediction time period is predicted; wherein, the first traffic flow sequence is the traffic flow sequence of the upstream adjacent checkpoint of the target construction operation location; if the type of the target construction operation location is the second type, then the second traffic flow prediction time period of the upstream adjacent toll station of the target construction operation location is determined, and the second traffic flow prediction time period is input into the second network layer of the traffic flow prediction model to predict the second traffic flow sequence corresponding to the second traffic flow prediction time period; wherein, the second traffic flow sequence is the inbound traffic flow, outbound traffic flow, and upstream adjacent checkpoint of the upstream adjacent toll station. The target construction operation location, the operation time of the target construction operation location, and the operation duration of the target construction operation location are input into the traffic flow prediction model obtained by fusing the periodic trend model, the real-time model, and the disturbance model of the time series, to adaptively predict the traffic flow of the target construction operation location, and quickly and accurately predict the traffic flow of the target construction operation location in the future time period.

[0037] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0038] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 A flowchart illustrating the traffic flow prediction method for highway construction zones provided in this application embodiment;

[0040] Figure 2This is one of the structural schematic diagrams of a traffic flow prediction device for a highway construction area provided in an embodiment of this application;

[0041] Figure 3 A second schematic diagram of a traffic flow prediction device for a highway construction area provided in an embodiment of this application;

[0042] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0044] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0045] To enable those skilled in the art to use the content of this application in conjunction with the specific application scenario of "predicting traffic flow in highway construction areas", the following implementation methods are provided. For those skilled in the art, the general principles defined herein can be applied to other embodiments and application scenarios without departing from the spirit and scope of this application.

[0046] The methods, apparatus, electronic devices, or computer-readable storage media described in this application can be applied to any scenario that requires traffic flow prediction in highway construction areas. This application does not limit the specific application scenario. Any scheme using a traffic flow prediction method, apparatus, and storage medium for highway construction areas provided in this application is within the protection scope of this application.

[0047] First, the applicable scenarios for this application will be introduced. This application can be applied to the field of data processing technology.

[0048] Research has revealed that many highways are entering a period of increasingly heavy maintenance and expansion work. Given a clear construction plan, it is necessary to close off certain sections of the road within a certain spatial area to provide sufficient space for construction machinery and personnel. However, the reduced road space caused by construction increases the braking effort required by drivers, potentially leading to psychological stress and resistance, thus reducing driving safety. Ensuring smooth traffic flow and safety in construction sections while meeting normal construction requirements, minimizing delays and fuel consumption, and improving driver and passenger comfort are key considerations in developing current traffic management plans for maintenance and construction. Therefore, effectively predicting future traffic flow in construction areas has become a significant technical challenge.

[0049] Based on this, this application provides a traffic flow prediction method for highway construction areas. The method inputs the target construction location, the operation time of the target construction location, and the operation duration of the target construction location into a traffic flow prediction model obtained by fusing a periodic trend model, a real-time model, and a disturbance model of time series. The method adaptively predicts the traffic flow at the target construction location, and quickly and accurately predicts the traffic flow at the target construction location in the future time period.

[0050] Please see Figure 1 , Figure 1 A flowchart illustrating the traffic flow prediction method for highway construction areas provided in this application embodiment. Figure 1 As shown in the embodiments of this application, the traffic flow prediction method includes:

[0051] S101: Input the target construction operation location on the highway, the operation time of the target construction operation location, and the operation duration of the target construction operation location into the first network layer of the pre-trained traffic flow prediction model to determine the first traffic flow prediction time period of the upstream adjacent checkpoint of the target construction operation location.

[0052] In this step, the target construction work location on the highway, the operation time of the target construction work location, and the operation duration of the target construction work location are input into the first network layer of the traffic flow prediction model to determine the first traffic flow prediction time period of the upstream adjacent checkpoint of the target construction work location.

[0053] Here, the traffic prediction model is obtained by fusing a periodic trend model, a real-time model, and a disturbance model of the time series.

[0054] The target construction location can be divided into two types. If the upstream of the target construction location is a toll station, the target construction location is of the second type. If the upstream of the target construction location is a checkpoint (traffic flow detection), the target construction location is of the first type.

[0055] In one possible implementation, the first traffic flow prediction time period of the upstream adjacent checkpoint of the target construction operation location is determined by the following steps:

[0056] A: Determine the first distance between the target construction location and the upstream adjacent checkpoint.

[0057] Here, the first distance between the target construction location and the upstream adjacent checkpoint is determined based on GPS data.

[0058] B: Determine the first target speed among multiple historical vehicle speeds near the upstream checkpoint.

[0059] Here, the mode speed among multiple historical vehicle speeds near the upstream checkpoint is determined as the first target speed.

[0060] C: Based on the ratio of the first distance to the first target speed, determine the first driving time for the vehicle to reach the target construction work location from the upstream adjacent checkpoint.

[0061] Here, the ratio of the first distance to the first target speed determines the time required for the vehicle to reach the target construction site from the upstream adjacent checkpoint, and this required time is defined as the first driving time.

[0062] D: Based on the first driving time, the operation time, the first operation duration, and the current operation time, a first traffic flow prediction time period is determined.

[0063] Here, the first traffic flow prediction time period is determined based on the first driving time, the operation time, the first operation duration, and the current operation time.

[0064] Here, the first traffic flow prediction period is the time period during which future traffic flow is predicted for the upstream adjacent checkpoints of the target construction site.

[0065] The task time refers to the start time of the task.

[0066] The first traffic flow prediction time period is [T] start -T now +T1,T start -Tnow +T1+T work ], where T start For the assignment time, T now T is the current working time, T1 is the first driving time, T work The duration of the first task.

[0067] S102: If the type of the target construction operation location is the first type, then the first traffic flow prediction time period is input into the second network layer of the traffic flow prediction model to predict the first traffic flow sequence corresponding to the traffic flow prediction time period.

[0068] In this step, if the type of the target construction location is the first type, the first traffic flow prediction time period is input into the second network layer of the traffic flow prediction model to predict the first traffic flow sequence corresponding to the traffic flow prediction time period.

[0069] The first traffic flow sequence is the traffic flow sequence of the upstream adjacent checkpoint of the target construction operation location.

[0070] Here, the first traffic flow sequence is the traffic flow sequence of the upstream adjacent checkpoint predicted under the first traffic flow prediction time period. Here, the first traffic flow sequence is used as the predicted traffic flow sequence of the target construction operation location of the first type.

[0071] In one possible implementation, if the type of the target construction location is a first type, then inputting the first traffic flow prediction time period into the second network layer of the traffic flow prediction model to predict the first traffic flow sequence corresponding to the traffic flow prediction time period includes:

[0072] a: Detect whether the time difference between the operation time and the current operation time is greater than or equal to a preset time threshold.

[0073] Here, the time difference between the current task time and the task time is checked to see if it is greater than a preset time threshold.

[0074] b: If so, then based on the time characteristics of the first traffic flow prediction time period, the periodic and trend-based traffic flows are predicted to determine the time-based traffic flow sequence; based on the disturbance characteristics of the upstream adjacent checkpoint, the traffic flow corresponding to the disturbance characteristics is predicted to determine the disturbance traffic flow sequence.

[0075] Here, if so, then based on the time characteristics of the first traffic flow prediction period, the periodic and trend-based traffic flows are predicted to determine the time-based traffic flow sequence. Based on the disturbance characteristics of the upstream adjacent checkpoints, the traffic flows corresponding to the disturbance characteristics are predicted to determine the disturbance traffic flow sequence.

[0076] Here, the process of predicting and determining the time-based traffic flow sequence for periodic and trend-based traffic flow is obtained by calling a periodic trend model of the time series to predict traffic flow.

[0077] The disturbance characteristics include adverse weather, construction, sudden traffic incidents, traffic control measures, and traffic flow data during the epidemic.

[0078] Here, the process of predicting traffic flow corresponding to disturbance characteristics and determining the disturbance traffic flow sequence is obtained by calling the disturbance model to predict traffic flow.

[0079] c: Based on the sum of the time traffic flow sequence and the disturbance traffic flow sequence, predict the first traffic flow sequence of the upstream adjacent checkpoint during the traffic flow prediction time period.

[0080] Here, based on the sum of the time traffic flow sequence and the disturbance traffic flow sequence, the first traffic flow sequence of the upstream adjacent checkpoint during the traffic flow prediction time period is predicted, and the first traffic flow sequence is determined as the traffic flow sequence of the target construction operation location of the first type.

[0081] In a specific embodiment, the time difference between the start time of the construction operation and the current operation time is calculated. If T start -T now ≥p, extract time features, call the potential model to predict traffic flow, and obtain T. t Extract disturbance features, call the disturbance model to predict traffic flow, and obtain D. t X predict t =T t +D t , where D t For the disturbance traffic flow sequence, T t This is a time-based traffic flow sequence.

[0082] In one possible implementation, after detecting whether the time difference between the operation time and the current operation time is greater than or equal to a preset time threshold, the traffic flow prediction method further includes:

[0083] (1): Based on the recent traffic flow characteristics of the upstream adjacent checkpoints, the real-time traffic flow is predicted and the real-time traffic flow sequence is determined.

[0084] Here, real-time traffic flow is predicted based on multiple recent traffic flow characteristics of upstream adjacent checkpoints, and the real-time traffic flow sequence is determined.

[0085] The prediction of real-time traffic flow sequences is obtained in advance using a real-time model.

[0086] (2): Based on the sum of the time traffic flow sequence, the disturbance traffic flow sequence and the real-time traffic flow sequence, the first traffic flow sequence of the upstream adjacent checkpoint during the traffic flow prediction time period is predicted.

[0087] Here, based on the sum of the time-based traffic flow sequence, the disturbance traffic flow sequence, and the real-time traffic flow sequence, the first traffic flow sequence of the upstream adjacent checkpoint during the traffic flow prediction period is predicted.

[0088] In a specific embodiment, the time difference between the start time of the construction operation and the current operation time is calculated. If T start -T now <p, extract time features, call the potential model to predict traffic flow, and obtain T. t Extract disturbance features, call the disturbance model to predict traffic flow, and obtain D. t By extracting recent data features and calling a real-time model to predict traffic, R is obtained. t X predict t =T t +D t +R t Among them, D t For the disturbance traffic flow sequence, T t For time-based traffic flow sequences, R t This is a real-time traffic flow sequence.

[0089] S103: If the type of the target construction operation location is the second type, then determine the second traffic flow prediction time period of the upstream adjacent toll station of the target construction operation location, input the second traffic flow prediction time period into the second network layer of the traffic flow prediction model, and predict the second traffic flow sequence corresponding to the second traffic flow prediction time period.

[0090] In this step, if the type of the target construction site is the second type, the second traffic flow prediction time period of the upstream adjacent toll station of the target construction site is determined, and the second traffic flow prediction time period is input into the second network layer of the traffic flow prediction model to predict the second traffic flow sequence corresponding to the second traffic flow prediction time period.

[0091] The second traffic flow sequence consists of the inbound traffic flow, outbound traffic flow, and traffic flow at the upstream adjacent toll station and the upstream adjacent checkpoint.

[0092] Here, the second traffic flow sequence is the predicted inbound traffic flow, outbound traffic flow, and upstream checkpoint traffic flow of the upstream adjacent toll station during the second traffic flow prediction time period. The second traffic flow sequence is determined as the target construction operation location of the second type.

[0093] In one possible implementation, the second traffic flow prediction time of the upstream adjacent toll station of the target construction site is determined by the following steps:

[0094] I: Determine the second target speed based on the historical vehicle speeds of multiple upstream adjacent toll stations.

[0095] Here, the mode speed is found among multiple historical vehicle speeds near the upstream toll station, and the mode speed is determined as the second target speed.

[0096] II: Determine the second distance between the target construction site and the upstream adjacent toll station.

[0097] Here, the second distance between the target construction site and the nearest toll station upstream is calculated. The second distance is calculated based on GPS data.

[0098] III: Based on the ratio of the second distance to the second target speed, determine the second driving time for the vehicle to travel from the upstream adjacent toll station to the target construction work location.

[0099] Here, the second driving time for the vehicle to reach the target construction site from the upstream adjacent toll station is determined based on the ratio of the second distance to the second target speed.

[0100] IV: Based on the second driving time, the operation time, the second operation duration, and the current operation time, determine the second traffic flow prediction time period.

[0101] Here, the second traffic flow prediction time period is determined based on the second driving time, the operation time, the second operation duration, and the current operation time.

[0102] The second traffic flow prediction time period is [T] start -T now +T 21 T start -T now +T2+T work ], where T start For the assignment time, T now T is the current working time, T2 is the first driving time, T work The duration of the second task.

[0103] In one possible implementation, the second traffic flow sequence corresponding to the second traffic flow prediction time period is predicted through the following steps:

[0104] (i): Based on the traffic flow prediction model, the time series of inbound traffic flow and the time series of outbound traffic flow of the adjacent toll station are predicted during the second traffic flow prediction period.

[0105] Here, based on the traffic flow prediction model, the time series of inbound traffic flow and the time series of outbound traffic flow of the nearby toll station are predicted during the second traffic flow prediction period.

[0106] Here, the prediction methods for the time series of inbound traffic flow and the time series of outbound traffic flow near the nearby toll station are consistent with the prediction methods for the traffic flow of the upstream nearby checkpoints mentioned above, and will not be repeated here.

[0107] (ii): The first sequence is determined by summing the first traffic flow sequence and the inbound traffic flow time sequence.

[0108] Here, the first sequence is determined by summing the first traffic flow sequence and the inbound traffic flow time sequence.

[0109] (iii): Based on the difference between the first sequence and the outbound traffic flow time sequence, predict the second traffic flow sequence corresponding to the second traffic flow prediction time period.

[0110] Here, based on the difference between the first sequence and the outbound traffic flow time sequence, the second traffic flow sequence corresponding to the second traffic flow prediction time period is predicted.

[0111] Here, the second traffic flow sequence corresponding to the second traffic flow prediction time period is the traffic flow of the second type of target construction operation location under the second traffic flow prediction time period.

[0112] In one possible implementation, the traffic prediction model is determined through the following steps:

[0113] i: Input the sample traffic flow data into the time series model, train the time training model, output undisturbed prediction data, input the undisturbed prediction data and the time characteristics of the undisturbed data in the sample traffic flow data into the fully connected neural network, generate the periodic trend model of the time series, and determine the first prediction error value of the periodic trend model of the time series.

[0114] Here, sample traffic flow data is input into a time series model to train the time training model and output undisturbed prediction data. The undisturbed prediction data and the time characteristics of the undisturbed data in the sample traffic flow data are input into a fully connected neural network to generate a periodic trend model of the time series and determine the first prediction error value of the periodic trend model of the time series.

[0115] ii: The neural network model is trained based on recent sample traffic flow data and the first prediction error value to generate a real-time model of the time series, and a second prediction error value of the real-time model of the time series is determined.

[0116] Here, the neural network model is trained using recent sample traffic flow data and the first prediction error value to generate a real-time model of the time series, and the second prediction error value of the real-time model of the time series is determined.

[0117] iii: The neural network model is trained based on historical sample perturbation data and the second prediction error value to generate the perturbation model.

[0118] Here, the neural network model is trained based on historical sample perturbation data and the second prediction error value to generate a dynamic model.

[0119] iv: The periodic trend model, the real-time model, and the disturbance model of the time series are fused to generate the traffic prediction model.

[0120] Here, the results of the periodic trend model, the real-time model, and the disturbance model of the time series are fused to generate a traffic prediction model.

[0121] Specifically, modeling is performed on the trend, periodicity, real-time characteristics, and disturbances in the time-series dataset. Data for trend, periodicity, real-time characteristics, and disturbances are extracted from historical traffic data. X period+trend For datasets used for trend and periodic modeling, X realtime For datasets used in real-time modeling, X disturbance For the dataset used in perturbation modeling, the prediction results of each model are fused to obtain a traffic prediction sequence.

[0122] Here, the model combines the start time of the construction operation with the current time:

[0123] T start -T now ≥p, the traffic prediction model is: X predict t =T t +D t ;

[0124] T start -T now <p, the traffic prediction model is: X predict t =T t +R t +D t ;

[0125] Where p is the preset time interval, X period+trend ={X t-p X t-2 , ..., X t-k}, where k is in minutes and represents the amount of historical data required. X realtime ={X t-1 X t-2 , ..., X t-p} indicates that a recent amount of data is needed. X disturbance Traffic flow data during historical periods of adverse weather, construction, accidents, and pandemics.

[0126] Here, the cyclical trend model is established in the following way:

[0127] Extracting X from sample traffic flow data period+trend ={X t-p X t-2 , ..., X t-k A traffic flow prediction model is established to predict traffic flow and obtain prediction data. Time-related features of the undisturbed (undisturbed means without the influence of weather, construction, sudden traffic events, epidemics, etc.) data are extracted. The extracted features are as follows: daily days are converted to categorical variables by weekday, with feature values ​​of 1-7 for Monday to Sunday; holidays are converted to 0-1 variables, with a feature value of 1 for holidays and 0 for non-holidays; major holidays are distinguished by m days before and m days after, with feature values ​​of 1, 2, 3, ... for National Day, Spring Festival, Labor Day, etc.; the beginning, middle, and end of the month are distinguished by feature values ​​of 1, 2, and 3; weekdays and non-working days are distinguished by feature values ​​of 1 for weekdays and 0 for non-working days; and the four seasons are distinguished by feature values ​​of 1, 2, 3, and 4. Using n minutes as the time window, a 24-hour day corresponds to {0, 24 / (n / 60)}. The undisturbed time features and the undisturbed Prophet traffic flow prediction model prediction data are combined to form feature data. Using the feature data as input and traffic flow as output, a fully connected neural network prediction model is established.

[0128] Here, the real-time model is established as follows: Real-time performance indicates that recent data has a greater impact on predicted traffic than historical data; the real-time model is a model of the relationship between recent data and predicted traffic. The modeling process is as follows: Extract X from historical traffic time-series data. realtime ={X t-1 X t-2 , ..., X t-p}, where p is in minutes, represents the amount of recent data needed, and p can be less than k and greater than or equal to 0. X realtimeThe data contains only undisturbed data (undisturbed means free from the influence of weather, construction, sudden traffic events, traffic control measures, epidemics, etc.). Feature extraction is performed on the following features: Statistical features for the prediction period, including maximum, minimum, quantiles, skewness, and kurtosis; daily statistical features, including maximum, minimum, quantiles, skewness, and kurtosis; statistical features for peak periods, including maximum, minimum, quantiles, skewness, and kurtosis; statistical features for the first 30 minutes of the prediction period, including maximum, minimum, quantiles, skewness, and kurtosis. Using the extracted features as input and the prediction errors of trend and periodic models as output, a real-time model is built using XGBoost. The above steps are repeated iteratively with different values ​​of p, outputting the p-value with the smallest error and the real-time model.

[0129] Here, the perturbation model is established in the following way:

[0130] In reality, traffic flow is affected by weather, construction, sudden traffic incidents, traffic control measures, and the pandemic. The disturbance model is a model of the relationship between these disturbance events and traffic flow prediction errors. The modeling process is as follows: Historical disturbance data samples are extracted from historical traffic flow time-series data. Disturbance factor data is extracted, including: construction information (construction type, location, and time); accident information (event type, location, time, and duration); pandemic information (location of prohibited vehicles); and weather information (rain, snow, and fog). Feature extraction is then performed on the disturbance factors. The extracted features are as follows: Epidemic features: average traffic flow during peak hours in the epidemic area, maximum traffic flow during peak hours in the epidemic area; Weather features: feature values ​​of 1, 2, and 3 to distinguish between heavy rain, moderate rain, and light rain; feature values ​​of 1, 2, and 3 to distinguish between heavy snow, moderate snow, and light snow; feature values ​​of 1, 2, and 3 to distinguish between dense fog, moderate fog, and light fog; Accident features: accident type, accident start time, accident end time, and distance between accident location and predicted location; Construction features: construction type, construction start time, construction end time, and distance between construction location and predicted location. Using the extracted features as input and the prediction bias of the trend-periodic model + real-time model as output, the XGBoost method is used to establish a perturbation model.

[0131] Data suitable for trend, real-time, and disturbance modeling are extracted from the original data respectively, and time series trend, real-time, and disturbance models are established respectively. Then, the results of each model are merged to obtain the traffic flow prediction model. Among them, the periodic trend model integrates time series forecasting and machine learning methods to improve the accuracy of prediction; the real-time model strengthens the impact of recent data on predicted traffic flow; and the disturbance model comprehensively considers the interference of various influencing factors on traffic flow. The construction location and construction period prediction model proposed in this scheme can adaptively predict traffic flow based on the input construction operation location, operation time, and operation duration.

[0132] This application provides a traffic flow prediction method for highway construction areas. The traffic flow prediction method includes: inputting the target construction operation location on the highway, the operation time of the target construction operation location, and the operation duration of the target construction operation location into the first network layer of a pre-trained traffic flow prediction model to determine a first traffic flow prediction time period of the upstream adjacent checkpoint of the target construction operation location; wherein, the traffic flow prediction model is obtained by fusing a time series periodic trend model, a time series real-time model, and a disturbance model; if the type of the target construction operation location is a first type, then the first traffic flow prediction time period is input into the traffic flow prediction model. In the second network layer of the traffic flow prediction model, a first traffic flow sequence corresponding to the traffic flow prediction time period is predicted. This first traffic flow sequence is the traffic flow sequence of the upstream adjacent checkpoint of the target construction work location. If the type of the target construction work location is a second type, a second traffic flow prediction time period for the upstream adjacent toll station is determined. This second traffic flow prediction time period is input into the second network layer of the traffic flow prediction model to predict the second traffic flow sequence corresponding to the second traffic flow prediction time period. This second traffic flow sequence includes the inbound traffic flow, outbound traffic flow, and traffic flow of the upstream adjacent toll station, as well as the traffic flow of the upstream adjacent checkpoint. The target construction work location, the operation time of the target construction work location, and the operation duration of the target construction work location are input into a traffic flow prediction model obtained by fusing a time series periodic trend model, a time series real-time model, and a disturbance model. This adaptively predicts the traffic flow at the target construction work location, quickly and accurately predicting the traffic flow at the target construction work location in future time periods.

[0133] Please see Figure 2 , Figure 3 , Figure 2 This is one of the structural schematic diagrams of a traffic flow prediction device for a highway construction area provided in an embodiment of this application; Figure 3 This is a second schematic diagram of a traffic flow prediction device for a highway construction area, provided as an embodiment of this application. Figure 2 As shown, the traffic flow prediction device 200 for the highway construction area includes:

[0134] The first determining module 210 is used to input the target construction operation location on the highway, the operation time of the target construction operation location, and the operation duration of the target construction operation location into the first network layer of the pre-trained traffic flow prediction model to determine the first traffic flow prediction time period of the upstream adjacent checkpoint of the target construction operation location; wherein, the traffic flow prediction model is obtained by fusing a periodic trend model of time series, a real-time model of time series, and a disturbance model.

[0135] The second determining module 220 is used to input the first traffic flow prediction time period into the second network layer of the traffic flow prediction model if the type of the target construction operation location is the first type, and predict the first traffic flow sequence corresponding to the traffic flow prediction time period; wherein, the first traffic flow sequence is the traffic flow sequence of the upstream adjacent checkpoint of the target construction operation location;

[0136] The third determining module 230 is used to determine the second traffic flow prediction time period of the upstream adjacent toll station of the target construction operation location if the type of the target construction operation location is the second type, input the second traffic flow prediction time period into the second network layer of the traffic flow prediction model, and predict the second traffic flow sequence corresponding to the second traffic flow prediction time period; wherein, the second traffic flow sequence is the inbound traffic flow, outbound traffic flow and upstream adjacent checkpoint of the upstream adjacent toll station.

[0137] Furthermore, the first determining module 210 determines the first traffic flow prediction time period of the upstream adjacent checkpoint of the target construction operation location through the following steps:

[0138] The first distance between the target construction location and the upstream adjacent checkpoint was determined;

[0139] Determine the first target speed among multiple historical vehicle speeds near the upstream checkpoint;

[0140] Based on the ratio of the first distance to the first target speed, the first driving time for the vehicle to reach the target construction work location from the upstream adjacent checkpoint is determined;

[0141] Based on the first driving time, the operation time, the first operation duration, and the current operation time, a first traffic flow prediction time period is determined.

[0142] Furthermore, when the second determining module 220 is used to input the first traffic flow prediction time period into the second network layer of the traffic flow prediction model if the type of the target construction operation location is the first type, and to predict the first traffic flow sequence corresponding to the traffic flow prediction time period, the second determining module 220 is specifically used for:

[0143] Detect whether the time difference between the current operation time and the operation time is greater than or equal to a preset time threshold;

[0144] If so, based on the time characteristics of the first traffic flow prediction time period, the periodic and trend-based traffic flows are predicted to determine the time-based traffic flow sequence; based on the disturbance characteristics of the upstream adjacent checkpoint, the traffic flow corresponding to the disturbance characteristics is predicted to determine the disturbance traffic flow sequence.

[0145] Based on the sum of the time-based traffic flow sequence and the disturbance-based traffic flow sequence, the first traffic flow sequence of the upstream adjacent checkpoint during the traffic flow prediction time period is predicted.

[0146] Furthermore, after detecting whether the time difference between the operation time and the current operation time is greater than or equal to a preset time threshold, the second determining module 220 is further configured to:

[0147] Based on multiple recent traffic flow characteristics of the upstream adjacent checkpoints, real-time traffic flow is predicted to determine the real-time traffic flow sequence.

[0148] Based on the sum of the time-based traffic flow sequence, the disturbance traffic flow sequence, and the real-time traffic flow sequence, the first traffic flow sequence of the upstream adjacent checkpoint during the traffic flow prediction time period is predicted.

[0149] Furthermore, the third determining module 230 determines the second traffic flow prediction time of the upstream adjacent toll station of the target construction operation location through the following steps:

[0150] A second target speed is determined based on the historical vehicle speeds of multiple upstream adjacent toll stations;

[0151] The second distance between the target construction site and the upstream adjacent toll station was determined;

[0152] Based on the ratio of the second distance to the second target speed, the second driving time for the vehicle to reach the target construction work location from the upstream adjacent toll station is determined;

[0153] The second traffic flow prediction time period is determined based on the second driving time, the operation time, the second operation duration, and the current operation time.

[0154] Furthermore, the third determining module 230 predicts the second traffic flow sequence corresponding to the second traffic flow prediction time period through the following steps:

[0155] Based on the traffic flow prediction model, the inbound traffic flow time series and the outbound traffic flow sequence of the nearby toll station are predicted during the second traffic flow prediction time period.

[0156] The first sequence is determined by summing the first traffic flow sequence with the inbound traffic flow time sequence;

[0157] Based on the difference between the first sequence and the outbound traffic flow time sequence, the second traffic flow sequence corresponding to the second traffic flow prediction time period is predicted. Further, as... Figure 3 As shown, the traffic flow prediction device 200 for highway construction areas also includes a training module 240, which is used to determine the traffic flow prediction model through the following steps:

[0158] The sample traffic flow data is input into the time series model to train the time training model and output undisturbed prediction data. The undisturbed prediction data and the time features of the undisturbed data in the sample traffic flow data are input into a fully connected neural network to generate the periodic trend model of the time series and determine the first prediction error value of the periodic trend model of the time series.

[0159] The neural network model is trained based on recent sample traffic flow data and the first prediction error value to generate a real-time model of the time series, and a second prediction error value of the real-time model of the time series is determined.

[0160] The neural network model is trained based on historical sample perturbation data and the second prediction error value to generate the perturbation model;

[0161] The periodic trend model, the real-time model, and the disturbance model of the time series are fused to generate the traffic prediction model.

[0162] This application provides a traffic flow prediction device for a highway construction area. The traffic flow prediction device includes: a first determining module, used to input the target construction operation location on the highway, the operation time of the target construction operation location, and the operation duration of the target construction operation location into the first network layer of a pre-trained traffic flow prediction model to determine a first traffic flow prediction time period of the upstream adjacent checkpoint of the target construction operation location; wherein, the traffic flow prediction model is obtained by fusing a time series periodic trend model, a time series real-time model, and a disturbance model; a second determining module, used to input the first traffic flow prediction time period into the first network layer of the pre-trained traffic flow prediction model if the type of the target construction operation location is a first type. In the second network layer of the traffic flow prediction model, a first traffic flow sequence corresponding to the traffic flow prediction time period is predicted; wherein, the first traffic flow sequence is the traffic flow sequence of the upstream adjacent checkpoint of the target construction operation location; a third determining module is used to determine the second traffic flow prediction time period of the upstream adjacent toll station if the type of the target construction operation location is the second type, and input the second traffic flow prediction time period into the second network layer of the traffic flow prediction model to predict the second traffic flow sequence corresponding to the second traffic flow prediction time period; wherein, the second traffic flow sequence is the inbound traffic flow, outbound traffic flow, and upstream adjacent checkpoint of the upstream adjacent toll station. The target construction operation location, the operation time of the target construction operation location, and the operation duration of the target construction operation location are input into the traffic flow prediction model obtained by fusing a time series periodic trend model, a time series real-time model, and a disturbance model, to adaptively predict the traffic flow of the target construction operation location, and quickly and accurately predict the traffic flow of the target construction operation location in the future time period.

[0163] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 As shown, the electronic device 400 includes a processor 410, a memory 420, and a bus 430.

[0164] The memory 420 stores machine-readable instructions executable by the processor 410. When the electronic device 400 is running, the processor 410 communicates with the memory 420 via the bus 430. When the machine-readable instructions are executed by the processor 410, they can perform the operations described above. Figure 1 The steps of the traffic flow prediction method for highway construction areas in the method embodiment shown are described in detail in the method embodiment, and will not be repeated here.

[0165] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described actions. Figure 1 The steps of the traffic flow prediction method for highway construction areas in the method embodiment shown are described in detail in the method embodiment, and will not be repeated here.

[0166] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0167] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0168] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0169] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0170] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0171] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method of traffic flow prediction for a highway work zone, characterized by, The traffic flow prediction method comprises: inputting a target construction work position on a highway, a work time of the target construction work position, and a work duration of the target construction work position into a first network layer of a pre-trained traffic flow prediction model to determine a first traffic flow prediction time period of an upstream adjacent toll gate of the target construction work position; wherein the traffic flow prediction model is obtained by fusing a periodic trend model of a time series, a real-time model of a time series, and a disturbance model; if the type of the target construction work position is a first type, inputting the first traffic flow prediction time period into a second network layer of the traffic flow prediction model to predict a first traffic flow sequence corresponding to the traffic flow prediction time period; wherein the first traffic flow sequence is a traffic flow sequence of the upstream adjacent toll gate of the target construction work position; if the type of the target construction work position is a second type, determining a second traffic flow prediction time period of an upstream adjacent toll station of the target construction work position, inputting the second traffic flow prediction time period into the second network layer of the traffic flow prediction model to predict a second traffic flow sequence corresponding to the second traffic flow prediction time period; wherein the second traffic flow sequence is an inbound traffic flow, an outbound traffic flow, and a traffic flow of an upstream adjacent toll gate of the upstream adjacent toll station; if the type of the target construction work position is the first type, inputting the first traffic flow prediction time period into the second network layer of the traffic flow prediction model to predict the first traffic flow sequence corresponding to the traffic flow prediction time period, comprising: detecting whether a time difference between the work time and a current work time is greater than or equal to a preset time threshold; if yes, predicting a periodic and trend traffic flow based on a time feature of the first traffic flow prediction time period to determine a time traffic flow sequence; predicting a traffic flow corresponding to a disturbance feature based on the disturbance feature of the upstream adjacent toll gate to determine a disturbance traffic flow sequence; predicting the first traffic flow sequence of the upstream adjacent toll gate in the traffic flow prediction time period based on a sum of the time traffic flow sequence and the disturbance traffic flow sequence.

2. The traffic flow prediction method according to claim 1, characterized in that, The first traffic flow prediction time period of the upstream adjacent toll gate of the target construction work position is determined by the following steps: determining a first distance between the target construction work position and the upstream adjacent toll gate; determining a first target speed in a plurality of historical vehicle speeds of the upstream adjacent toll gate; determining a first driving time of a vehicle from the upstream adjacent toll gate to the target construction work position based on a ratio of the first distance and the first target speed; determining the first traffic flow prediction time period based on the first driving time, the work time, a first work duration, and a current work time.

3. The traffic flow prediction method according to claim 1, characterized in that, After the detection of whether the time difference between the work time and the current work time is greater than or equal to the preset time threshold, the traffic flow prediction method further comprises: predict a real-time traffic flow sequence based on the plurality of recent traffic flow features of the upstream adjacent toll gate; predict a first traffic flow sequence of the upstream adjacent toll gate in the traffic flow prediction time period based on a sum of the time traffic flow sequence, the disturbance traffic flow sequence and the real-time traffic flow sequence.

4. The traffic flow prediction method according to claim 1, characterized in that, determine a second traffic flow prediction time of an upstream adjacent toll gate of the target construction operation position by the following steps: determine a second target speed of a plurality of historical vehicle speeds of the upstream adjacent toll gate; determine a second distance between the target construction operation position and the upstream adjacent toll gate; determine a second driving time of a vehicle from the upstream adjacent toll gate to the target construction operation position based on a ratio of the second distance and the second target speed; determine the second traffic flow prediction time period based on the second driving time, the operation time, a second operation duration and a current operation time.

5. The traffic flow prediction method according to claim 1, characterized in that, predict a second traffic flow sequence corresponding to the second traffic flow prediction time period by the following steps: predict an entering traffic flow time sequence of the adjacent toll gate and an exiting traffic flow sequence of the adjacent toll gate in the second traffic flow prediction time period based on the traffic flow prediction model; determine a first sequence based on a sum of the first traffic flow sequence and the entering traffic flow time sequence; predict the second traffic flow sequence corresponding to the second traffic flow prediction time period based on a difference between the first sequence and the exiting traffic flow sequence.

6. The traffic flow prediction method of claim 1, wherein, determine the traffic flow prediction model by the following steps: input sample traffic flow data into a time sequence model to train the time training model, output undisturbed prediction data, input time features of the undisturbed data in the sample traffic flow data and the undisturbed prediction data into a fully connected neural network to generate a periodic trend model of the time sequence and determine a first prediction error value of the periodic trend model of the time sequence; train the neural network model based on recent sample traffic flow data and the first prediction error value to generate a real-time model of the time sequence and determine a second prediction error value of the real-time model of the time sequence; train the neural network model based on historical sample disturbance data and the second prediction error value to generate a disturbance model; fuse the periodic trend model of the time sequence, the real-time model of the time sequence and the disturbance model to generate the traffic flow prediction model.

7. A traffic flow prediction device for a highway work zone, characterized by comprising: The traffic flow prediction device comprises: a first determining module configured to input a target construction operation position on an expressway, an operation time of the target construction operation position and an operation duration of the target construction operation position into a first network layer of a pre-trained traffic flow prediction model to determine a first traffic flow prediction time period of an upstream adjacent toll gate of the target construction operation position; wherein the traffic flow prediction model is obtained by fusing a periodic trend model of a time sequence, a real-time model of a time sequence and a disturbance model. The second determining module is configured to, if the type of the target construction operation position is the first type, input the first traffic flow prediction time period into a second network layer of the traffic flow prediction model to predict a first traffic flow sequence corresponding to the traffic flow prediction time period, wherein the first traffic flow sequence is a traffic flow sequence of a nearby upstream toll station of the target construction operation position. The third determining module is configured to, if the type of the target construction operation position is the second type, determine a second traffic flow prediction time period of a nearby upstream toll station of the target construction operation position, input the second traffic flow prediction time period into the second network layer of the traffic flow prediction model to predict a second traffic flow sequence corresponding to the second traffic flow prediction time period, wherein the second traffic flow sequence is an inbound traffic flow, an outbound traffic flow and a traffic flow of a nearby upstream toll station of the nearby upstream toll station. The second determining module is configured to, if the type of the target construction operation position is the first type, input the first traffic flow prediction time period into a second network layer of the traffic flow prediction model to predict a first traffic flow sequence corresponding to the traffic flow prediction time period: determine whether a time difference between the operation time and a current operation time is greater than or equal to a preset time threshold; if yes, predict periodic and trend traffic flows based on time characteristics of the first traffic flow prediction time period to determine a time traffic flow sequence; predict traffic flows corresponding to the disturbance characteristics based on the disturbance characteristics of the nearby upstream toll station to determine a disturbance traffic flow sequence; predict the first traffic flow sequence of the nearby upstream toll station in the traffic flow prediction time period based on a sum of the time traffic flow sequence and the disturbance traffic flow sequence.

8. An electronic device, comprising: The electronic device comprises a processor, a memory and a bus, the memory stores machine readable instructions executable by the processor, when the electronic device is running, the processor and the memory communicate through the bus, and the machine readable instructions are executed by the processor to perform the steps of the expressway construction zone traffic flow prediction method according to any one of claims 1 to 6. The computer readable storage medium stores a computer program, and the computer program is executed by the processor to perform the steps of the expressway construction zone traffic flow prediction method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, ​

Images