Travel-service-oriented running situation prediction system for high-speed traffic emergencies

Abstract

The invention, which relates to the field of high-speed traffic operation situation prediction, discloses a travel-service-oriented operation situation prediction system for high-speed traffic emergencies, comprising a severe weather traffic influence prediction module, a traffic time duration prediction module, and a traffic flow operation situation prediction module. The system is characterized in that the severe weather traffic influence prediction module carries out quantitative processing on the influence of the meteorological data and the traffic data on the highway running speed and the accident rate based on the collected meteorological data and the traffic data. According to the travel service-oriented running situation prediction system for the high-speed traffic emergencies, through integration of multi-source data, comprehensive consideration of various influence factors and application of a data analysis method and an advanced model, accurate prediction of the running situation of the high-speed traffic emergencies can be effectively realized, and at the same time, the running situation of the high-speed traffic emergencies is guided by public behaviors, so that the running situation of the high-speed traffic emergencies is predicted. Personalized travel information and suggestions are provided, the public is helped to reasonably plan the travel, congested and dangerous road sections are avoided, and the travel safety and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the field of high-speed traffic operation situation prediction, and in particular to an operation situation prediction system for high-speed traffic emergencies oriented to travel services. Background Art

[0002] The impact of severe weather on highway traffic safety and traffic efficiency is profound and extensive. Under severe weather conditions, the road environment is complex and changeable, which can easily cause a series of traffic problems, such as slippery road surface, reduced visibility, vehicle loss of control, etc., which directly lead to an increase in the incidence of traffic accidents and a decrease in traffic efficiency.

[0003] Therefore, it is necessary to propose an operation status prediction system for high-speed traffic emergencies oriented to travel services to solve the above problems. Summary of the invention

[0004] The main purpose of the present invention is to provide an operation status prediction system for high-speed traffic emergencies for travel services, which can effectively solve the problems in the background technology.

[0005] To achieve the above object, the technical solution adopted by the present invention is:

[0006] The operation status prediction system of high-speed traffic emergencies for travel services includes a severe weather traffic impact prediction module, a traffic time duration prediction module, and a traffic flow operation status prediction module. The severe weather traffic impact prediction module quantifies the impact of the severe weather traffic data and traffic data on the highway operation speed and accident rate based on the collected meteorological data and traffic data;

[0007] The traffic time duration prediction module determines the fluctuation range of the duration of the traffic blockage event based on the acquired historical traffic time data, and provides the public with the estimated duration information of the traffic event based on this;

[0008] The traffic flow operation situation prediction module predicts the traffic flow operation situation after a traffic event occurs based on the historical data collected from the traffic event management system and the traffic flow monitoring system.

[0009] Preferably, the severe weather traffic impact prediction module includes a data collection and integration module, a data cleaning and preprocessing module, a correlation analysis and model building module, and a prediction and display module, wherein the data collection and integration module is used to obtain meteorological data including weather conditions, precipitation intensity, visibility, etc. from the meteorological department, and collect traffic data such as traffic flow, vehicle speed, and accident type from the traffic event management system and the toll collection system, accurately match the meteorological data with the highway section pile number, and clarify the specific location, time, and type of severe weather;

[0010] The data cleaning and preprocessing module is used to comprehensively clean the collected data, fill in missing values ​​based on a specific algorithm, remove outliers based on a set threshold, and use a standardized method to unify the format of data from different sources. It performs a stationarity test and trend decomposition on time series data, removes seasonal fluctuations and trend influences, and constructs a unified data set based on the correlation analysis data in the spatiotemporal dimensions.

[0011] Preferably, the association analysis and model building module adopts descriptive statistical analysis, draws histograms and box plots of operating speed and accident rate under different weather conditions, intuitively displays data distribution, uses Pearson and Spearman correlation coefficients for correlation analysis, draws scatter plots to quantify the correlation between bad weather and operating speed and accident rate, establishes a model based on multiple linear regression, quantifies the impact of bad weather, screens significantly influencing variables, uses time series analysis to reveal the dynamic impact of bad weather on traffic operation, constructs a correlation model between bad weather and traffic accident rate and travel speed, and an accident risk assessment model that comprehensively considers meteorological, traffic and accident factors;

[0012] The prediction and display module predicts the average operating speed and accident probability of each vehicle type under different severe weather conditions based on the constructed model, uses a line graph to display the changing trend of the operating speed of each vehicle type under different precipitation intensities, and uses a heat map to present the distribution of accident risks in different road sections, so as to provide the public with travel speed references and accident risk warnings, and help the public plan their travel in advance.

[0013] Preferably, the traffic time duration prediction module includes a data collection and preprocessing module, an in-depth analysis module of influencing factors, a model building and prediction module, and an information release and application module, wherein the data collection and preprocessing module is used to collect historical traffic event data from the local highway traffic event management system, covering the time, location, type, severity and duration of the accident, and at the same time obtain the vehicle flow, average speed, and occupancy rate from the traffic flow monitoring system, and combine the meteorological data to eliminate weather-related influencing factors, clean the collected data, remove outliers and missing values, and perform standardization and feature engineering processing to extract features related to the duration of traffic events.

[0014] Preferably, the in-depth analysis module of influencing factors is used to deeply analyze the composition of the duration of traffic accidents, including accident discovery time, response time, clearance time and traffic restoration time, to clarify that the accident discovery time is mainly affected by the alarm response delay, wherein the alarm response delay includes the difficulty of accident discovery, the reporting ability of the reporter, and the patrol ability of the patrol car, wherein the accident response time is affected by the preparation time and the time to arrive at the scene, the preparation time depends on the availability of rescue personnel and materials, the time to arrive at the scene is related to the geographical location of the rescue department, the speed of the rescue vehicle, the traffic conditions and the weather; the accident clearance time is related to the rescue workload and efficiency, and is affected by the type of accident, the severity, and the business proficiency of the rescue personnel; the traffic restoration time depends on the length of the queue upstream of the accident point and the speed at which the queue dissipates.

[0015] Preferably, the model building and prediction module is used to predict the preparation delay time using a decision tree model, and to select key factors from candidate influencing factors such as the geographical location of the rescue department, the speed of the rescue vehicle, traffic congestion, and weather conditions through variance analysis, and to build a decision tree model based on the key factors;

[0016] For the duration of the event, a sensitivity analysis prediction model is constructed. Considering the prediction accuracy requirements for the duration of highway traffic blockage events, a regression model is used to analyze the delay time classification model of traffic blockage events based on sensitivity analysis to determine the relationship between influencing factors and duration, and calculate the degree of influence of the fluctuation of uncertain factors on the duration, so as to determine the time fluctuation range and obtain the prediction result with intervals. The formula is:

[0017]

[0018] where t + is the upper limit of the classification model; t - is the lower limit of the classification model; L is the calculated value of the regression model; T i is the fluctuation value of the uncertain factor i;

[0019] The information release and application module releases the estimated duration of traffic events to the public through multiple channels, including mobile phone APP push, real-time website updates, traffic radio broadcasts, etc. The public can reasonably plan their travel time based on this information, choose to avoid affected sections of the road or adjust their travel plans to reduce waiting time.

[0020] Preferably, the traffic flow operation situation prediction module includes a data collection and processing module, a model building and analysis module, and a situation prediction and display module, wherein the data collection and processing module is used to collect historical traffic event data and traffic flow data from the traffic event management system, the traffic flow monitoring system, and the highway ETC system, and at the same time, for the abnormal data collected by the highway ETC system, it mines the spatiotemporal evolution law of the traffic flow, characterizes the internal evolution mechanism of the traffic flow, and constructs a spatiotemporal characteristic vector model of highway traffic flow, which is used to extract the spatiotemporal characteristic information of potential traffic flow.

[0021] Preferably, the model building and analysis module predicts highway traffic flow based on the traffic perspective using a random forest prediction model, builds a highway traffic speed prediction model through a deep learning combined prediction model, fits the spatiotemporal evolution of traffic speed, and predicts traffic speed change trends, which is used to deeply analyze the impact of typical traffic events such as traffic accidents, vehicle failures, and road construction on the operation status of traffic flow.

[0022] Preferably, the situation prediction and display module is used to comprehensively consider the type, severity, location, time, weather conditions, and road characteristics of traffic events, use the constructed model to predict the operation situation of traffic flow after the traffic event occurs, and construct a traffic congestion heat map to intuitively display the degree of congestion in different sections of road. The road section speed change curve is used to present the real-time change of vehicle speed, which is used to provide real-time road condition information to help the public choose the best travel route, avoid congested sections, and improve travel efficiency.

[0023] Compared with the prior art, the present invention provides an operation situation prediction system for high-speed traffic emergencies for travel services, which has the following beneficial effects:

[0024] 1. The operation status prediction system for high-speed traffic emergencies for travel services can effectively realize accurate prediction of the operation status of high-speed traffic emergencies by integrating multi-source data, comprehensively considering various influencing factors, and using data analysis methods and advanced models. At the same time, it is guided by public travel behavior and provides personalized travel information and suggestions to help the public plan their trips reasonably, avoid congestion and dangerous sections, and improve travel safety and efficiency.

[0025] 2. The operation status prediction system of high-speed traffic emergencies for travel services can collect and process data in real time, promptly reflect changes in traffic conditions, provide the public with the latest traffic information, and enable the public to adjust their travel plans in a timely manner. It can also continuously optimize system models and functions based on public feedback and actual traffic conditions, improve the system's adaptability and accuracy, and provide better services to the public. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1is a system block diagram of the present invention;

[0027] Figure 2 It is a precipitation intensity-average running speed scatter plot diagram of the present invention;

[0028] Figure 3 It is a scatter plot and a function curve diagram of the running speed of a small car and the rainfall intensity of the present invention;

[0029] Figure 4 It is a traffic accident duration composition diagram of the present invention;

[0030] Figure 5 is an accident delay time diagram of the present invention;

[0031] Figure 6 is an analysis diagram of factors affecting preparation time of the present invention;

[0032] Figure 7 is a diagram of factors influencing the time to arrive at the site of the present invention;

[0033] Figure 8 is a diagram of factors affecting the accident clearance time of the present invention;

[0034] Fig. 9 It is a diagram of candidate influencing factors for constructing a decision tree model for time to arrive at the scene of the present invention;

[0035] Fig.10 is a diagram of candidate influencing factors for the construction preparation time decision tree model of the present invention;

[0036] Fig.11 It is a diagram of alternative influencing factors of the construction accident clearance time decision tree model of the present invention;

[0037] Fig.12 It is a top-level structural diagram of the decision tree for initial splitting using accident types according to the present invention. DETAILED DESCRIPTION

[0038] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0039] Embodiment 1:

[0040] like Figure 1-Figure 12 As shown, the operation situation prediction system of high-speed traffic emergencies for travel services includes a severe weather traffic impact prediction module, a traffic time duration prediction module, and a traffic flow operation situation prediction module. The severe weather traffic impact prediction module quantifies the impact of the severe weather traffic data and traffic data on the highway operation speed and accident rate based on the collected meteorological data and traffic data;

[0041] The severe weather traffic impact prediction module includes data collection and integration module, data cleaning and preprocessing module, correlation analysis and model building module, and prediction and display module. The data collection and integration module is used to obtain meteorological data such as weather conditions, precipitation intensity, and visibility from the meteorological department, and collect traffic data such as traffic flow, vehicle speed, and accident type from the traffic event management system and toll collection system, accurately match the meteorological data with the highway section pile number, and clarify the specific location, time, and type of severe weather.

[0042] The data cleaning and preprocessing module is used to comprehensively clean the collected data, fill in missing values ​​based on specific algorithms, eliminate outliers based on set thresholds, and use standardized methods to unify the formats of data from different sources. It performs stationarity tests and trend decomposition on time series data, removes seasonal fluctuations and trend influences, and constructs a unified data set based on data correlation analysis in the spatiotemporal dimensions.

[0043] The correlation analysis and model building module uses descriptive statistical analysis to draw histograms and box plots of operating speed and accident rate under different weather conditions, intuitively display data distribution, use Pearson and Spearman correlation coefficients for correlation analysis, draw scatter plots to quantify the correlation between bad weather and operating speed and accident rate, build models based on multiple linear regression, quantify the impact of bad weather, screen significant influencing variables, use time series analysis to reveal the dynamic impact of bad weather on traffic operation, build bad weather and traffic accident rate, travel speed correlation model, and accident risk assessment model that comprehensively considers meteorological, traffic and accident factors;

[0044] The prediction and display module is based on the constructed model to predict the average operating speed and accident probability of each type of vehicle under different severe weather conditions. It uses a line graph to show the changing trend of the operating speed of each type of vehicle under different precipitation intensities, and uses a heat map to present the distribution of accident risks in different road sections. It is used to provide the public with travel speed references and accident risk warnings, helping the public to plan their travel in advance.

[0045] The traffic time duration prediction module determines the fluctuation range of the duration of traffic blockage events based on the acquired historical traffic time data, and provides the public with information on the estimated duration of traffic events based on this;

[0046] The traffic time duration prediction module includes a data collection and preprocessing module, an in-depth analysis module of influencing factors, a model building and prediction module, and an information release and application module. The data collection and preprocessing module is used to collect historical traffic event data from the local highway traffic event management system, covering the time, location, type, severity and duration of the accident. At the same time, it obtains vehicle flow, average vehicle speed, and occupancy rate from the traffic flow monitoring system, and combines meteorological data to eliminate weather-related influencing factors. The collected data is cleaned to remove outliers and missing values, and standardized and feature-engineered to extract features related to the duration of traffic events.

[0047] The in-depth analysis module of influencing factors is used to deeply analyze the composition of traffic accident duration, including accident discovery time, response time, clearance time and traffic restoration time. It is clear that the accident discovery time is mainly affected by the alarm response delay, among which the alarm response delay includes the difficulty of accident discovery, the reporting ability of the reporter, and the patrol ability of the patrol car. The accident response time is affected by the preparation time and the time to arrive at the scene. The preparation time depends on the availability of rescue personnel and materials. The time to arrive at the scene is related to the geographical location of the rescue department, the speed of the rescue vehicle, the traffic conditions and the weather; the accident clearance time is related to the rescue workload and efficiency, and is affected by the type of accident, severity, and the business proficiency of the rescue personnel; the traffic restoration time depends on the length of the queue upstream of the accident point and the speed at which the queue dissipates.

[0048] The model building and prediction module is used to predict the preparation delay time using a decision tree model. The key factors are selected from alternative influencing factors such as the geographical location of the rescue department, the speed of the rescue vehicle, traffic congestion, and weather conditions through variance analysis, and a decision tree model is built based on this.

[0049] For the duration of the event, a sensitivity analysis prediction model is constructed. Considering the prediction accuracy requirements for the duration of highway traffic blockage events, a regression model is used to analyze the delay time classification model of traffic blockage events based on sensitivity analysis to determine the relationship between influencing factors and duration, and calculate the degree of influence of the fluctuation of uncertain factors on the duration, so as to determine the time fluctuation range and obtain the prediction result with intervals. The formula is:

[0050]

[0051] where t + is the upper limit of the classification model; t - is the lower limit of the classification model; L is the calculated value of the regression model; T i is the fluctuation value of the uncertain factor i;

[0052] The information release and application module releases the estimated duration of traffic incidents to the public through multiple channels, including mobile phone APP push, real-time website updates, traffic broadcasts, etc. The public can plan their travel time reasonably based on this information, choose to avoid affected sections of the road or adjust their travel plans to reduce waiting time.

[0053] The traffic flow operation situation prediction module collects historical data from the traffic event management system and the traffic flow monitoring system to predict the traffic flow operation situation after a traffic event occurs.

[0054] The traffic flow operation situation prediction module includes data collection and processing module, model construction and analysis module, situation prediction and display module. The data collection and processing module is used to collect historical traffic event data and traffic flow data from traffic event management system, traffic flow monitoring system and highway ETC system. At the same time, for the abnormal data collected by highway ETC system, it mines the spatiotemporal evolution law of traffic flow, characterizes the internal evolution mechanism of traffic flow, and constructs the spatiotemporal characteristic vector model of highway traffic flow to extract the spatiotemporal characteristic information of potential traffic flow.

[0055] The model building and analysis module uses the random forest prediction model to predict highway traffic flow from a traffic perspective, and builds a highway traffic speed prediction model through a deep learning combined prediction model to fit the spatiotemporal evolution of traffic speed and predict the trend of traffic speed changes. This module is used to deeply analyze the impact of typical traffic events such as traffic accidents, vehicle failures, and road construction on the operation of traffic flow.

[0056] Traffic accidents will lead to lane closures, reduce road capacity, and reduce the average speed of the affected road section by 30%-50%, and the traffic volume by 20%-40%, with the impact lasting for several hours; vehicle breakdowns usually occupy one lane, reducing the traffic volume of the affected lane by 10%-20%, and the speed by 15%-30%, and the impact time is generally within 30 minutes to 1 hour; road construction occupies part of the lane or closes the road section, resulting in a 15%-25% reduction in traffic volume and a 20%-35% reduction in speed on the affected road section, and the impact lasts for several days to weeks, and will also increase traffic pressure on surrounding roads. Consider the superposition effect of bad weather and traffic events, such as traffic accidents in foggy weather, which will cause a sharp deterioration in the traffic flow situation.

[0057] The situation prediction and display module is used to comprehensively consider the type, severity, location, time, weather conditions, and road characteristics of traffic incidents, and use the constructed model to predict the operation situation of traffic flow after a traffic incident occurs. It constructs a traffic congestion heat map to intuitively display the degree of congestion in different sections of road, and uses the section speed change curve to present the real-time changes in vehicle speed. It is used to provide real-time road condition information to help the public choose the best travel route, avoid congested sections, and improve travel efficiency.

[0058] Embodiment 2:

[0059] By statistically calculating the road traffic volume data on sunny and rainy days, the changes in road traffic volume on sunny and rainy days are obtained, as shown in Table 1.

[0060] Average hourly traffic volume (veh / h) of each vehicle type on a two-way four-lane expressway under different weather conditions:

[0061] Small car Large car Trailer Equivalent traffic volume sunny 596 320 173 1755 rain 462 221 114 1246

[0062] It can be seen that the average hourly road traffic volume statistics on sunny and rainy days are significantly different, and the road traffic volume on rainy days is smaller than that on sunny days. The traffic volume of small vehicles decreased by 22.5%, the traffic volume of large vehicles decreased by 30.9%, the traffic volume of trailers decreased by 34.1%, and the equivalent traffic volume of all vehicles decreased by 29.0%, indicating that weather changes will have a relatively large impact on road traffic volume.

[0063] By statistically calculating the traffic speed data of road traffic flow on sunny days and rainy days, the comparison of traffic speed of road traffic flow on sunny days and rainy days, as well as the changes of traffic speed of road traffic flow under different precipitation intensities are obtained, as shown in Table 2, Table 3 and Figure 2 As shown,

[0064] Table 2 Average operating speed of various types of vehicles on two-way four-lane expressways under different weather conditions (km / h):

[0065]

[0066]

[0067] Table 3 Average operating speed of various types of vehicles on two-way four-lane expressways under different precipitation intensities (km / h):

[0068]

[0069] From the above analysis, we can see that when the rainfall intensity is 1, 1.5, 2, 2.5, 3, 3.5, 4 mm / 5 min, the total average speed of highway traffic is 7.42%, 7.78%, 10.59%, 10.79%, 15.33%, 15.89%, 14.95% lower than the average speed of 87.3 km / h on sunny days. When the rainfall intensity is 1, 1.5, 2, 2.5, 3, 3.5, 4 mm / 5 min, the average speed of small cars on highway is 9.45%, 10.74%, 13.47%, 14.10%, 18.48%, 19.16%, 20.13% lower than the average speed of 98.3 km / h on sunny days. When the rainfall intensity was 1, 1.5, 2, 2.5, 3, 3.5, 4 mm / 5 min, the average speed of large vehicles on the highway decreased by 11.38%, 12.13%, 13.49%, 14.30%, 15.59%, 15.33%, 13.24% respectively compared with the average speed of 81.1 km / h on sunny days. When the rainfall intensity was 1, 1.5, 2, 2.5, 3, 3.5, 4 mm / 5 min, the average speed of trailer vehicles on the highway decreased by 3.13%, 4.58%, 3.65%, 5.82%, 12.34%, 15.96%, 15.23% respectively compared with the average speed of 69.0 km / h on sunny days.

[0070] The observed small car running speed and rainfall intensity are fitted with functions in the 99% confidence interval to find the objective law between them. The fitted scatter plot and function curve are shown in Figure 3 As shown, the corresponding small car running speed function expression is as follows:

[0071] V 平均 =9.19e -4.572R +89.55e -0.1651R ;

[0072] Where V 平均 is the running speed (km / h); R is the rainfall intensity (mm / min);

[0073] From the figure, we can see that the distribution of the observed data points of the vehicle speed on the example highway is not very discrete. The correlation coefficient of the fitting function is 0.9833, the multiple correlation coefficient is 0.9799, the fitting error is 10.59, and the mean square error is 0.8404. This function is a typical exponential function, which better reflects the changes in vehicle speed with changes in rainfall intensity.

[0074] Embodiment three:

[0075] like Figure 1As shown in the figure, the operation situation prediction system of high-speed traffic emergencies for travel services is implemented.

[0076] It should be noted that the present invention is an operation status prediction system for high-speed traffic emergencies for travel services. When in use, it collects data from various data sources at regular intervals to ensure the real-time and accuracy of the data. When severe weather occurs or traffic incidents occur, relevant data is obtained in a timely manner and the system is updated. The collected data is cleaned, preprocessed and feature extracted, and corresponding analysis methods and models are used for calculation and prediction to obtain prediction results of the traffic operation status. The prediction results are released to the public through various channels, such as personalized push of mobile phone APP, real-time display of websites, scheduled broadcasts of traffic radio, etc. According to the travel preferences and location information of the public, personalized travel suggestions are provided to them, feedback information on the use of the system by the public is collected, the differences between the prediction results and the actual situation are analyzed, the model and system are optimized and improved, and the accuracy of the prediction and the service quality of the system are continuously improved.

[0077] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. The operation status prediction system of high-speed traffic emergencies for travel services includes a severe weather traffic impact prediction module, a traffic time duration prediction module, and a traffic flow operation status prediction module, which is characterized by: The severe weather traffic impact prediction module quantifies the impact of the severe weather traffic impact on the highway operating speed and accident rate based on the collected meteorological data and traffic data; The traffic time duration prediction module determines the fluctuation range of the duration of the traffic blockage event based on the acquired historical traffic time data, and provides the public with the estimated duration information of the traffic event based on this; The traffic flow operation situation prediction module predicts the traffic flow operation situation after a traffic event occurs based on the historical data collected from the traffic event management system and the traffic flow monitoring system.

2. The operation situation prediction system for high-speed traffic emergencies for travel services according to claim 1 is characterized by: The severe weather traffic impact prediction module includes a data collection and integration module, a data cleaning and preprocessing module, a correlation analysis and model building module, and a prediction and display module, wherein the data collection and integration module is used to obtain meteorological data including weather conditions, precipitation intensity, visibility, etc. from the meteorological department, and at the same time collect traffic data such as traffic flow, vehicle speed, and accident type from the traffic event management system and the toll collection system, accurately match the meteorological data with the highway section pile number, and clarify the specific location, time, and type of severe weather; The data cleaning and preprocessing module is used to comprehensively clean the collected data, fill in missing values ​​based on a specific algorithm, remove outliers based on a set threshold, and use a standardized method to unify the format of data from different sources. It performs a stationarity test and trend decomposition on time series data, removes seasonal fluctuations and trend influences, and constructs a unified data set based on the correlation analysis data in the spatiotemporal dimensions.

3. The operation situation prediction system for high-speed traffic emergencies for travel services according to claim 2 is characterized by: The association analysis and model building module adopts descriptive statistical analysis to draw histograms and box plots of operating speed and accident rate under different weather conditions, intuitively display data distribution, use Pearson and Spearman correlation coefficients for correlation analysis, draw scatter plots to quantify the correlation between bad weather and operating speed and accident rate, establish a model based on multiple linear regression, quantify the impact of bad weather, screen significant influencing variables, use time series analysis to reveal the dynamic impact of bad weather on traffic operation, build a correlation model between bad weather and traffic accident rate and travel speed, and a risk assessment model that comprehensively considers meteorological, traffic and accident factors; The prediction and display module predicts the average operating speed and accident probability of each vehicle type under different severe weather conditions based on the constructed model, uses a line graph to display the changing trend of the operating speed of each vehicle type under different precipitation intensities, and uses a heat map to present the distribution of accident risks in different road sections, so as to provide the public with travel speed references and accident risk warnings, and help the public plan their travel in advance.

4. The operation situation prediction system for high-speed traffic emergencies for travel services according to claim 1 is characterized by: The traffic time duration prediction module includes a data collection and preprocessing module, an in-depth analysis module of influencing factors, a model building and prediction module, and an information release and application module. The data collection and preprocessing module is used to collect historical traffic event data from the local highway traffic event management system, covering the time, location, type, severity and duration of the accident, and at the same time obtain the vehicle flow, average speed, and occupancy rate from the traffic flow monitoring system, and combine meteorological data to eliminate weather-related influencing factors, clean the collected data, remove outliers and missing values, and perform standardization and feature engineering processing to extract features related to the duration of traffic events.

5. The operation situation prediction system for high-speed traffic emergencies for travel services according to claim 4 is characterized by: The in-depth analysis module of influencing factors is used to deeply analyze the composition of the duration of traffic accidents, including accident discovery time, response time, clearance time and traffic restoration time, and clarify that the accident discovery time is mainly affected by the alarm response delay, wherein the alarm response delay includes the difficulty of accident discovery, the reporting ability of the reporter, and the patrol ability of the patrol car, wherein the accident response time is affected by the preparation time and the time to rush to the scene, wherein the preparation time depends on the availability of rescue personnel and materials, and the time to rush to the scene is related to the geographical location of the rescue department, the speed of the rescue vehicle, the traffic conditions and the weather; the accident clearance time is related to the rescue workload and efficiency, and is affected by the type of accident, the severity, and the business proficiency of the rescue personnel; Traffic restoration time depends on the length of the queue upstream of the accident point and the speed at which the queue dissipates.

6. The operation situation prediction system for high-speed traffic emergencies for travel services according to claim 4 is characterized by: The model building and prediction module is used to predict the preparation delay time using a decision tree model, and to select key factors from alternative influencing factors such as the geographical location of the rescue department, the speed of the rescue vehicle, traffic congestion, and weather conditions through variance analysis, and to build a decision tree model based on this; For the duration of the event, a sensitivity analysis prediction model is constructed. Considering the prediction accuracy requirements for the duration of highway traffic blockage events, a regression model is used to analyze the delay time classification model of traffic blockage events based on sensitivity analysis to determine the relationship between influencing factors and duration, and calculate the degree of influence of the fluctuation of uncertain factors on the duration, so as to determine the time fluctuation range and obtain the prediction result with intervals. The formula is: where t + is the upper limit of the classification model; t - is the lower limit of the classification model; L is the calculated value of the regression model; T i is the fluctuation value of the uncertain factor i; The information release and application module releases the estimated duration of traffic events to the public through multiple channels, including mobile phone APP push, real-time website updates, traffic radio broadcasts, etc. The public can reasonably plan their travel time based on this information, choose to avoid affected sections of the road or adjust their travel plans to reduce waiting time.

7. The operation situation prediction system for high-speed traffic emergencies for travel services according to claim 1 is characterized by: The traffic flow operation situation prediction module includes a data collection and processing module, a model construction and analysis module, and a situation prediction and display module, wherein the data collection and processing module is used to collect historical traffic event data and traffic flow data from the traffic event management system, the traffic flow monitoring system, and the highway ETC system. At the same time, for the abnormal data collected by the highway ETC system, the spatiotemporal evolution law of the traffic flow is mined, the internal evolution mechanism of the traffic flow is characterized, and the spatiotemporal characteristic vector model of the highway traffic flow is constructed to extract the spatiotemporal characteristic information of the potential traffic flow.

8. The high-speed traffic emergency operation situation prediction system for travel services according to claim 7 is characterized by: The model building and analysis module uses a random forest prediction model to predict highway traffic flow based on a traffic perspective, builds a highway traffic speed prediction model through a deep learning combined prediction model, fits the spatiotemporal evolution of traffic speed, and predicts the trend of traffic speed changes. It is used to deeply analyze the impact of typical traffic events such as traffic accidents, vehicle failures, and road construction on the operation status of traffic flow.

9. The high-speed traffic emergency operation situation prediction system for travel services according to claim 7, characterized in that: The situation prediction and display module is used to comprehensively consider the type, severity, location, time, weather conditions, and road characteristics of traffic events, use the constructed model to predict the operation situation of traffic flow after a traffic event occurs, and construct a traffic congestion heat map to intuitively display the congestion level of different road sections. The road section vehicle speed change curve is used to present the real-time change of vehicle speed, and is used to provide real-time road condition information to help the public choose the best travel route, avoid congested sections, and improve travel efficiency.

Images