Emergency response coordination system based on big data

Abstract

The invention relates to the technical field of high-speed information management, in particular to an emergency response coordination system based on big data, and the system comprises a data processing and event recognition module which obtains real-time data based on high-speed traffic monitoring equipment, carries out the clustering analysis, divides the accident types in an accident report according to time features and space features, and carries out the calculation of the accident types; and recognizing the similarity between the high-speed traffic congestion degree and the accident type, and generating traffic congestion and accident mode recognition data. According to the method, the real-time data of the expressway emergencies are subjected to preprocessing and clustering analysis, and the accident types are deeply divided based on different time and space characteristics, so that the system can more accurately identify the accident types and the potential influence of the accident types on traffic congestion. By analyzing the difference between the time characteristics and the space characteristics, the system can identify the accident position, the time period and the road specific area more accurately. And performing time sequence prediction on the accident occurrence possibility by using an accumulation analysis mode.

Description

Technical Field

[0001] The present invention relates to the technical field of high-speed information management, and in particular to an emergency response coordination system based on big data. Background Art

[0002] Highway information management technology mainly includes the collection, transmission, processing and analysis of highway operation data to achieve comprehensive monitoring and real-time management of traffic operation conditions. By building an intelligent information system and combining technologies such as the Internet of Things, big data, and cloud computing, the field brings together various types of traffic information to provide data support and decision-making basis for highway operation management, emergency handling, and road network optimization.

[0003] Among them, the emergency response coordination system is a system that uses data analysis and processing technology to provide emergency response support for highway emergencies. Through the collection and analysis of real-time data, the accident location can be quickly located, the impact range of the accident can be assessed, and relevant emergency resources (such as traffic management, rescue departments, etc.) can be coordinated to respond quickly to reduce the impact of the incident on traffic.

[0004] When dealing with highway emergencies, existing technologies have difficulty distinguishing accident types in depth based on temporal and spatial characteristics, and fail to adequately classify and identify accident characteristics in different time periods or geographical locations. For example, nighttime accidents and daytime peak-hour accidents have different impacts, but existing technologies do not make detailed analysis of these factors during processing, resulting in low accuracy in accident identification. In addition, during the risk assessment process, it is difficult for the system to effectively assess potential high-risk periods, including holiday traffic peaks and areas in advance. For example, this affects the implementation of pre-intervention measures. At the same time, it is difficult for the system to achieve accurate diffusion simulation of high- and low-risk areas, resulting in insufficient in-depth impact assessment of high-risk areas, which is not conducive to timely response to potential risks in the region. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose an emergency response coordination system based on big data.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: An emergency response coordination system based on big data includes: The data processing and event recognition module is based on the real-time data obtained by the high-speed traffic monitoring equipment and performs cluster analysis to classify the accident types in the accident report according to time characteristics and spatial characteristics, identify the similarity between the degree of high-speed traffic congestion and the accident type, and generate traffic congestion and accident pattern recognition data; The event dynamic prediction and risk assessment module performs time series prediction on the possibility of an accident based on the traffic congestion and accident pattern recognition data and according to the time characteristics and spatial characteristics, obtains accident probability prediction data, divides the accident probability prediction data into multiple risk levels, and generates accident risk level assessment data; The emergency response scope assessment module obtains the current high-speed traffic flow and vehicle density information based on the accident risk level assessment data, simulates the diffusion scope of the accident under various conditions through a diffusion model, obtains the accident diffusion scope data, divides the accident impact area into levels of distance and degree of impact according to the accident diffusion scope data, assesses the potential impact of the accident diffusion, and generates emergency response scope assessment data; The event linkage priority sorting and resource scheduling module conducts linkage impact analysis on all high-speed traffic emergencies based on the emergency response scope assessment data, evaluates the interrelationships between events, prioritizes all events by impact coefficients, obtains event priority sorting results, dynamically allocates emergency resources based on the event priority sorting results, and generates emergency resource scheduling management results.

[0007] As a further solution of the present invention, the step of obtaining the accident types in the accident report according to the time characteristics and the spatial characteristics is specifically as follows: Collect real-time data obtained by high-speed traffic monitoring equipment, including traffic flow, accident reports, event locations and vehicle density information, and pre-process them. Classify and group the processed data according to time and space characteristics, extract information on accident-prone areas and congested areas, and generate traffic congestion and accident pattern data sets; Based on the traffic congestion and accident pattern dataset, the data in the accident report is classified according to time characteristics and spatial characteristics. By analyzing the distribution of accident types in risk areas and specific time periods, the distribution patterns of accident types in multiple areas and time periods are identified to obtain accident characteristic data.

[0008] As a further solution of the present invention, the steps for acquiring the traffic congestion and accident pattern recognition data are specifically as follows: Based on the accident characteristic data, the formula is adopted: ; Calculate the similarity score of the incident , used to evaluate the similarity of accidents in time and space characteristics, and obtain accident characteristics and similarity score data; in, It is The frequency of accidents with time characteristics, is the weight of the time feature, It is The frequency of accidents with spatial characteristics, is the weight of the spatial feature, is the total number of event logs; Based on the accident characteristics and similarity score data, by comparing the relationship between traffic flow and accident frequency, congestion patterns in areas with multiple accident frequencies and risk periods are identified, and traffic congestion and accident pattern recognition data are generated.

[0009] As a further solution of the present invention, the steps for obtaining the accident risk level assessment data are specifically as follows: Based on the traffic congestion and accident pattern recognition data, the formula is adopted: ; Calculate the predicted value of the accident probability in the next time period , get the accident probability prediction data; in, Is the current time period The probability of an accident occurring, is the adjustment factor, Is the current time period frequency of accidents; Based on the accident probability prediction data, the predicted values ​​are divided into multiple risk levels, the corresponding risk time periods and areas are marked, and range annotation is performed to generate accident risk level assessment data.

[0010] As a further solution of the present invention, the steps for acquiring the accident diffusion range data are specifically as follows: Based on the accident risk level assessment data, the current high-speed traffic flow and vehicle density information is obtained, the corresponding diffusion intensity parameters are set according to the risk level of the area, and the diffusion radius at each time point in the diffusion model is recorded to obtain preliminary diffusion simulation data; Based on the preliminary diffusion simulation data, the diffusion radius of the risk area is sorted and labeled, and the risk level and diffusion range in the area are displayed in multiple colors to generate accident diffusion range data.

[0011] As a further solution of the present invention, the step of acquiring the emergency response scope assessment data is specifically as follows: Based on the accident diffusion range data, the accident impact on the risk area is hierarchically divided into distance and degree of impact, the accident impact area is divided into multiple circles according to distance, the impact degree of each circle is marked, and a hierarchical division data table is established; Based on the hierarchical division data table, the formula is used: ; Calculate the predicted diffusion time and the predicted diffusion intensity , generate emergency response scope assessment data; in, is the predicted diffusion distance, is the propagation speed in the model, is the model adjustment coefficient, is the initial accident impact, It is a simulated environmental impact factor.

[0012] As a further solution of the present invention, the step of obtaining the correlation between the evaluation events is specifically: Based on the emergency response scope assessment data, the linkage impact of the high-speed traffic emergency events is analyzed, including analyzing the time characteristics, location and diffusion range data of the events, classifying the time period, spatial location and diffusion characteristics of the events, and establishing a basic data set for event linkage analysis; Based on the event linkage analysis basic data set, the formula is used: ; Calculate the impact coefficient of an event , get the impact weight of the event on the overall emergency response; in, is the time feature weight coefficient, is the time characteristic factor, is the spatial feature weight coefficient, is the spatial characteristic factor, is the diffusion range weight coefficient, is the diffusion range factor.

[0013] As a further solution of the present invention, the steps for obtaining the emergency resource scheduling management result are specifically: Based on the impact weight of the event on the overall emergency response, the impact coefficient data of all events are summarized and sorted, and the event priorities are arranged in order from high to low to obtain the event priority ranking result; According to the event priority ranking result, the parameters for dynamic allocation of emergency resources are adjusted, and the resource allocation ratio is adjusted by the influence coefficient, so that high-priority events obtain more concentrated resource allocation and obtain emergency resource scheduling management results.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are: In the present invention, by preprocessing and clustering the real-time data of highway emergencies, the accident types are deeply divided based on different time and space characteristics, and the system can more accurately identify the types of accidents and their potential impact on traffic congestion. By analyzing the differences in time characteristics and space characteristics, the system can more accurately identify the accident location, time period and specific road areas. The possibility of an accident is predicted in time series using a cumulative analysis method, which effectively provides accurate risk levels and time and space positioning information, allowing the system to perceive high-risk areas in advance and provide support for pre-emergency measures. In addition, the hierarchical division of accident-affected areas is achieved through diffusion models of different risk levels, and the simulation of accident diffusion in high-risk areas is strengthened, thereby making the assessment of the emergency response scope more accurate. Through the characteristic influence coefficients of multiple events, dynamic sorting of event priorities is achieved. For emergency resource allocation, it can be more concentrated on high-priority accident sites, realizing fast and efficient resource scheduling and event processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a system flow chart of the present invention; Figure 2 A flow chart of the present invention for classifying accident types in accident reports according to time characteristics and spatial characteristics; Figure 3 Obtaining traffic congestion and accident pattern recognition data flow chart for the present invention; Figure 4 A flow chart for obtaining accident risk level assessment data for the present invention; Figure 5 A flow chart for obtaining accident diffusion range data for the present invention; Figure 6 A flow chart for obtaining emergency response scope assessment data for the present invention; Figure 7 A flow chart for evaluating the inter-relationships between events for the present invention; Figure 8 The present invention is a flow chart for obtaining emergency resource scheduling management results. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0017] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0018] See also Figure 1 , an emergency response coordination system based on big data includes: The data processing and event recognition module performs data preprocessing based on the real-time data obtained by the highway traffic monitoring equipment, including highway traffic flow, accident reports, event locations, and vehicle density information. By performing cluster analysis on the preprocessed data, the accident types in the accident reports are divided according to time characteristics and spatial characteristics. The time characteristics include the time period when the accident occurred, the difference characteristics between weekdays and weekends and holidays, and the spatial characteristics include the geographical location of the accident, as well as the specific regional characteristics of variable lanes, exit ramps, long downhill slopes or tunnel entrances. The module identifies the similarity between the degree of highway traffic congestion and the type of accident, and generates traffic congestion and accident pattern recognition data; The event dynamic prediction and risk assessment module analyzes the identified high-speed traffic congestion degree and accident type based on traffic congestion and accident pattern recognition data, and uses a step-by-step accumulation method to perform time series prediction on the possibility of accidents according to time and space characteristics, and obtains accident probability prediction data. The accident probability prediction data is divided into different risk levels, and risk time periods and areas are marked to generate accident risk level assessment data; The emergency response scope assessment module obtains the current high-speed traffic flow and vehicle density information based on the accident risk level assessment data, and simulates the diffusion scope of the accident under various conditions through the diffusion model, including simulating the diffusion scope of accidents under different conditions for accidents of different risk levels. For areas marked as high-risk, higher diffusion intensity parameters are used for simulation, while for medium and low-risk areas, correspondingly lower diffusion parameters are used to obtain the accident diffusion scope data, and the accident-affected areas are divided into levels of distance and degree of impact according to the accident diffusion scope data, and the diffusion time and diffusion intensity are calculated. By evaluating the potential impact of the accident diffusion, the emergency response scope assessment data is generated; The event linkage priority sorting and resource scheduling module conducts linkage impact analysis on all high-speed traffic emergencies based on the emergency response scope assessment data. By analyzing the time characteristics, spatial characteristics and diffusion range characteristics of multiple events, the impact coefficient of the event is calculated, and the correlation between events is evaluated. All events are prioritized by the impact coefficient, and the event priority sorting results are obtained. According to the event priority sorting results, emergency resources are dynamically allocated, resources are preferentially dispatched to the location of high-priority events, and emergency resource scheduling management results are generated; Traffic congestion and accident pattern recognition data include traffic flow analysis, accident location distribution and time period classification; accident risk level assessment data include accident probability prediction for each time period, spatial distribution risk level and regional marking; emergency response scope assessment data include diffusion intensity, diffusion radius and potential impact area division; emergency resource scheduling results include resource allocation plans for high-priority events, resource standby arrangements and linkage deployment for medium and low priority events.

[0019] See also Figure 2 , the specific steps for obtaining the accident types in the accident report according to the time characteristics and spatial characteristics are as follows: Collect real-time data obtained by high-speed traffic monitoring equipment, including traffic flow, accident reports, event locations and vehicle density information, and pre-process them. Classify and group the processed data according to time and space characteristics, extract information on accident-prone areas and congested areas, and generate traffic congestion and accident pattern data sets; The real-time data obtained by high-speed traffic monitoring equipment, including traffic flow, accident reports, event locations and vehicle density information, is first preprocessed. The data preprocessing module uses anomaly detection and missing data completion methods to clean the data. For example, the K nearest neighbor filling method is used to complete the missing traffic flow data, and the zscore method is used to remove abnormal peak values ​​to ensure the accuracy and continuity of the data. Then, cluster analysis is performed on the preprocessed traffic data. For example, the density-based DBSCAN algorithm is used to group the data by vehicle density and accident location, and identify multiple high-density areas and congestion locations where accidents occur. The data is further classified according to specific time characteristics and spatial characteristics through clustering results. Examples of time feature classification include dividing the data into morning and evening peak hours, off-peak hours, and different traffic patterns on weekdays, weekends, and holidays. Spatial feature classification can be distinguished according to geographical location, such as focusing on marking service areas, toll stations, exit ramps and other areas prone to accidents. Through the above classification and clustering operations, traffic congestion and accident pattern recognition data are generated, which is convenient for further analysis and prediction in subsequent modules.

[0020] Based on the traffic congestion and accident pattern dataset, the data in the accident report is classified according to time and space characteristics. By analyzing the distribution of accident types in risk areas and specific time periods, the distribution patterns of accident types in various areas and time periods are identified to obtain accident characteristic data. The types of accidents in the accident reports are classified according to time characteristics and spatial characteristics. Taking time characteristics classification as an example, the accident frequency within each hour can be counted and further distinguished according to the morning and evening peak hours on weekdays and weekends, so as to identify specific periods with high accident frequencies. The spatial characteristics acquisition method uses the geographic information system (GIS) to accurately mark the geographical location information of the accident. For example, the buffer analysis of GIS is used to define the scope of high-risk areas such as service areas, toll stations and exit ramps as a buffer zone with a radius of 200 meters. By analyzing the accident report data in these areas, high-accident risk areas and their typical accident types can be identified, such as rear-end collisions common in service areas and scratch accidents common in toll stations. Through this composite analysis of time and space characteristics, the distribution patterns of different accident types in different time periods and specific areas are obtained, and accident spatiotemporal characteristic data are generated to provide accurate data support for subsequent risk assessment and emergency response.

[0021] See also Figure 3 ,The specific steps for obtaining traffic congestion and accident pattern recognition data are: Based on the accident characteristic data, the formula is used: ; Calculate the similarity score of the incident , used to evaluate the similarity of accidents in time and space characteristics, and obtain accident characteristics and similarity score data; in, It is The frequency of accidents with time characteristics is obtained by summarizing the data recorded by surveillance cameras and traffic data. is the weight of the time feature, which indicates the influence of the time feature in the similarity calculation. The weight value is obtained by statistically analyzing the accident rate in different time periods. It can be determined based on the frequency analysis of traffic flow and accident history data. For example, more accidents during holiday peak hours have a greater impact on accident similarity and are given a higher weight. It is The accident frequency of spatial characteristics represents the accident frequency in a specific spatial location, which is obtained through the geographical location information of traffic events. is the weight of the spatial feature, which indicates the influence of the spatial feature in the similarity calculation. It is obtained based on the accident frequency data analysis of each spatial feature area. For example, a higher weight is given when there are more accidents on the exit ramp. is the total number of event logits, indicating the number of temporal and spatial features involved in similarity calculation.

[0022] For example, suppose the frequency of accidents during the morning rush hour is 4 and the weight is 0.6; the frequency of accidents in the variable lane is 3 and the weight is 0.4.

[0023] Calculate the time feature impact value: ; Calculate the spatial feature influence value: ; Calculate the similarity score: ; The result shows that the similarity score after weighting the temporal and spatial features is 1.8.

[0024] Based on the accident characteristics and similarity score data, by comparing the relationship between traffic flow and accident frequency, the congestion patterns in the accident frequency areas and risk periods are identified, and traffic congestion and accident pattern recognition data are generated; Use a similarity score of 1.8 to evaluate and identify the association between traffic congestion and accident types in specific areas and time periods. Assume that two key areas (service area A and toll station B) and two time periods (morning peak and evening peak) are selected for analysis. During the morning peak, the historical accident frequency of service area A is 5 times / week, and the corresponding similarity score is 1.8. During the same time period, the historical accident frequency of toll station B is 3 times / week, and the similarity score is also 1.8. During the evening peak, the accident frequency of service area A is 2 times / week, while toll station B has an accident frequency of 4 times / week, assuming similarity scores of 1.2 and 1.5 respectively. For the morning peak period, both service area A and toll station B have a high similarity score (1.8), indicating that the accident patterns of these two areas are similar and the accident frequency is high during the morning peak. On the traffic flow data of these two locations, during the morning peak, the vehicle flow of service area A is 3000 vehicles per hour, while the flow of toll station B is 2500 vehicles per hour. During the evening peak, a difference was formed between service area A with a lower similarity score and toll station B with a higher similarity score. Service area A's accident similarity score dropped to 1.2, and the evening peak vehicle flow dropped to 1,800 vehicles per hour, while toll station B's accident similarity score was 1.5, with a flow of 2,300 vehicles per hour. By comparison, it can be seen that during the morning peak, service area A and toll station B have higher similarity scores, which, combined with high vehicle flow and accident frequency, indicate that these locations are more prone to congestion during the morning peak and need to be paid special attention. During the evening peak, toll station B has relatively high similarity and accident frequency, and vehicle flow remains at a relatively high level, which also shows a certain congestion risk. Finally, the generated traffic congestion and accident pattern recognition data includes the peak period accident frequency and similarity score of each area, and the congestion risk of service area A during the morning peak and toll station B during the morning and evening peaks is marked as a high-risk area, providing reference information for resource scheduling for emergency response.

[0025] See also Figure 4 ,The specific steps for obtaining accident risk level assessment data are as follows: Based on traffic congestion and accident pattern recognition data, the formula is used: ; Calculate the predicted value of the accident probability in the next time period , get the accident probability prediction data; in, is the probability of an accident in the current time period, indicating the possibility of an accident in the current time period. It is generally obtained through statistical analysis of historical data and reflects the baseline risk of accidents in the current time period. For example, it is obtained through the statistics of the frequency of traffic accidents in the past hour. For example, if the monitoring data records that the accident frequency in the past hour is 0.2, then set As the current probability of an accident, It is an adjustment coefficient used to control the response degree of the prediction results to the current accident frequency. This coefficient is determined by experiments or historical data backtesting analysis, and generally floats between 0 and 1. For example, when backtesting historical data, the changes in the prediction results are observed by adjusting different coefficient values, and the value suitable for the prediction accuracy is selected. It is the accident frequency in the current time period, indicating the actual observed accident frequency in the current time period, which is directly obtained through real-time monitoring or traffic record data.

[0026] For example, if the probability of an accident occurring in the current time period is , the adjustment factor is , the current accident frequency is .

[0027] Calculate the predicted probability: ; calculate : ; Calculate adjustments: ; Calculate the forecast probability for the next period: ; The result shows that the predicted value of the accident probability in the next period is 0.235.

[0028] Based on the accident probability prediction data, the predicted values ​​are divided into multiple risk levels, the corresponding risk periods and areas are marked, and the ranges are marked to generate accident risk level assessment data; Based on the predicted value of accident probability , bring the probability value into the risk level classification to identify and mark time periods and areas with different risk levels. Assume that the risk level is divided into three intervals according to the probability value: low risk (0-0.1), medium risk (0.1-0.2), and high risk (0.2 and above). According to this classification rule, the predicted value of 0.235 falls in the high-risk interval, which means that the risk of accidents in this area in the next period is high. According to the prediction results, the high-risk level is marked to a specific time period (such as the morning peak period) and area (such as a service area or toll station). Assume that this probability prediction value comes from the accident monitoring data of service area A. By comparing the risk levels of different areas, service area A can be marked as a high-risk area during the morning peak. Combined with other high-risk areas (such as the morning and evening peak periods of toll station B), the high-risk areas are marked on the map, and a data table containing the accident risk level assessment of each time period and area is formed. For example, the morning peak period of service area A is marked as a high-risk period, while the morning and evening peak periods of toll station B are marked as high-risk areas.

[0029] See also Figure 5 , the specific steps for obtaining the accident diffusion range data are: Based on the accident risk level assessment data, the current high-speed traffic flow and vehicle density information is obtained, the corresponding diffusion intensity parameters are set according to the risk level of the area, and the diffusion radius at each time point in the diffusion model is recorded to obtain preliminary diffusion simulation data; The current high-speed traffic flow and vehicle density information is obtained, and the diffusion range of accidents at different risk levels is simulated through the diffusion model. Diffusion simulation is performed using software such as MATLAB or ArcGIS, and different diffusion intensity parameters are set to reflect the different diffusion conditions of accidents in high, medium and low risk areas. For example, in high-risk areas (such as highway service areas), the diffusion intensity parameter is set to 0.9 to reflect the rapid diffusion of accidents in this dense area. During the simulation process, the diffusion radius is recorded in steps of 5 minutes, so that the software obtains the diffusion radius every 5 minutes. For example, the diffusion radius reaches 1 km at 10 minutes and expands to 1.5 km at 15 minutes; for medium-risk areas (such as highway exit ramps), the diffusion intensity parameter is set to 0.6 to reflect the slower diffusion speed in medium-dense areas. Under the same time step, the diffusion radius data obtained by the software is 0.7 km at 10 minutes and 1 km at 15 minutes. The diffusion intensity parameter in low-risk areas (such as ordinary lanes) is set to 0.3, reflecting the slower diffusion speed in low-density areas. The diffusion range obtained by simulation results is only 0.4 kilometers in 10 minutes and 0.6 kilometers in 15 minutes.

[0030] Based on the preliminary diffusion simulation data, the diffusion radius of the risk area is sorted and marked, and the risk level and diffusion range in the area are displayed in multiple colors to generate accident diffusion range data; Based on the above simulation results, the accident diffusion range data is obtained, and the diffusion range data of high, medium and low risk areas are sorted and marked. The diffusion radius data of high-risk areas (such as service areas) are marked as 1 km (10 minutes) and 1.5 km (15 minutes), and are marked in red on the map to highlight their risk of rapid diffusion; medium-risk areas (such as highway exit ramps) are marked as 0.7 km (10 minutes) and 1 km (15 minutes), and are marked in orange; low-risk areas (such as ordinary lanes) are marked in green, and the diffusion radius is only 0.4 km (10 minutes) and 0.6 km (15 minutes). The final generated diffusion range data table will include the diffusion intensity, radius and time step information of each risk area as a reference for emergency response range assessment.

[0031] See also Figure 6 ,The specific steps for obtaining emergency response scope assessment data are as follows: Based on the accident diffusion range data, the accident impact in the risk area is divided into different levels based on distance and degree of impact. The accident impact area is divided into multiple circles based on distance, the impact degree of each circle is marked, and a hierarchical division data table is established. First, the location of the accident is taken as the core point, and the outer boundary of the impact area, that is, the maximum radius of the accident impact, is determined based on the accident diffusion range data. For example, the diffusion model shows that the impact range of this accident is 5 kilometers, so the maximum circle radius is set to 5 kilometers. With the core point as the center, multiple impact circles are divided according to fixed distances. For example, the maximum radius of 5 kilometers can be divided into three circles of 1 kilometer, 3 kilometers and 5 kilometers according to the accident data, and these circles correspond to different levels of accident impact. Specifically, the inner circle of 1 kilometer represents the most serious impact area, followed by 3 kilometers, and 5 kilometers is a slightly affected area. According to the distance information in the accident diffusion range data, the impact level is set from the innermost circle to the outside. For example: the 1-kilometer range is set as the first-level impact area (highest impact), which is usually the priority range for accident emergency response. The 3-kilometer range is the second-level impact area, indicating a medium impact and requiring a secondary response. The 5-kilometer range is the third-level impact area, that is, a lighter impact area, and the lowest priority response is applicable. Create corresponding records for each circle, including the radius of the circle, the physical distance from the core point, the circle mark (level 1, level 2, level 3), and the corresponding description of the degree of impact. For example, the 1 km circle is marked as "high impact", the 3 km circle is marked as "medium impact", and the 5 km circle is marked as "low impact". Organize the above circle distances and marks into a data table for use in subsequent emergency responses. The data table can contain the following fields: circle radius, impact level, physical distance from the core point, description of impact, etc. This table is used in the emergency command system to provide a clear division of impact areas for decision-making, so that resources can be allocated to different circles according to priority, such as service areas, toll stations, exit ramps, etc., to ensure priority response in each level of area.

[0032] Based on the hierarchical division of the data table, the formula is used: ; Calculate the predicted diffusion time and the predicted diffusion intensity , generate emergency response scope assessment data; in, It is the predicted diffusion distance, which indicates the predicted range of the modeled accident spreading from the center point to the set impact circle, and changes according to the setting of simulation conditions. It is the propagation speed in the model, which is used to reflect the assumed speed of accident spread. High, medium and low propagation speeds are set according to different risk levels. It is the model adjustment coefficient, which is used to adjust the calculation of diffusion intensity according to the accident risk level, and helps to distinguish the different impact ranges of high, medium and low risks in the simulation. is the initial accident impact, which indicates the basic impact of the accident severity. It is set to a fixed value in the model, for example, a high-severity accident is set to 5, and a minor accident is set to as low as 2 or 3. It is a simulated environmental impact factor, which represents the environmental conditions set in the model, such as the impact of weather and road conditions on accident diffusion. It is adjusted according to the simulation scenario, for example, a lower value is taken in bad weather and a higher value is taken in good weather.

[0033] For example, let the diffusion distance The model propagation speed is 3 km. For 60 km / h: ; Calculate the predicted diffusion intensity For example, suppose the initial accident impact 5, environmental impact factor is 0.8, the model adjustment coefficient For 1.5: ; The results show that under the modeling conditions, the predicted diffusion time of the accident within a 3-kilometer range is 3 minutes, and the predicted diffusion intensity is 6. This indicates that the impact of the accident will quickly spread to the surrounding areas and has a large impact range within 3 kilometers. The value of diffusion intensity 6 indicates that the dense traffic flow near the accident site is prone to further traffic congestion or secondary accident risks. According to the diffusion intensity 6, the 3-kilometer range around the accident site is set as a high-priority response area. Due to the particularity of the highway, priority attention should be paid to traffic congestion in service areas, toll stations, entrance and exit ramps, and near the accident site. For these key areas, rescue vehicles, tow trucks and warning signs are arranged in advance to quickly divert traffic and clear accidents in the diffusion area.

[0034] See also Figure 7 ,The steps to obtain the correlation between the evaluation events are as follows: Based on the emergency response scope assessment data, the linkage impact of the high-speed traffic emergencies is analyzed, including the analysis of the time characteristics, location and diffusion range data of the events, the classification of the time period, spatial location and diffusion characteristics of the events, and the establishment of a basic data set for event linkage analysis; The data features of each emergency event are decomposed and sorted item by item, including the time characteristics of the event, the specific location of the event, and the diffusion range of the event at different time steps. First, the time characteristics are subdivided according to specific time periods, divided into morning peak, evening peak and non-peak periods, and the frequency of events in these different time periods is recorded. Secondly, through spatial feature data, according to the geographical location of the event, the event data is associated with specific highway nodes, including key locations such as service areas, toll stations, and exit ramps, and the surrounding conditions at the geographical location are further identified. Then, the diffusion model is used to calculate the impact range of each event at a specific time step, record the diffusion distance and impact intensity value, and classify these data according to specific time periods and spatial locations. Events with similar characteristics are merged to form a complete linkage data set. This data set contains time characteristics, spatial characteristics, and diffusion range characteristics, which serve as the basic input for the subsequent calculation of event impact coefficients and provide necessary pre-information for priority sorting and resource allocation.

[0035] Based on the event linkage analysis of the basic data set, the formula is used: ; Calculate the impact coefficient of an event , get the impact weight of the event on the overall emergency response; in, It is the time characteristic weight coefficient, which is used to adjust the contribution of the time characteristic factor to the overall impact coefficient. It is usually obtained by analyzing historical data to determine the weight of the difference between peak and non-peak time periods on the impact of events. By observing the relationship between traffic flow and event incidence in different time periods, the time weight is refined. It is a time characteristic factor, which is used to measure the impact of the time period of the event on the occurrence of the accident. It is divided into peak period and non-peak period, and a higher weight is given to the peak period to reflect the diffusion rate and impact range of the event in high-density traffic flow. This factor is directly obtained through the time data of the event. For example, according to whether the event occurs during the morning peak or evening peak, the corresponding value is set. It is the spatial feature weight coefficient, which indicates the weight adjustment of the event location. It is used to reflect the weight of the comprehensive impact of different spatial features on the event. It is determined through historical data or field monitoring. Different areas can be marked in the GIS system. The spatial feature weights are adjusted according to the sensitivity of different areas (such as service areas, toll stations, etc.). It is a spatial characteristic factor that reflects the sensitivity of the location of the event and has a greater impact on traffic flow or resource allocation. It is obtained by analyzing the GIS data of the event location. For example, high-density node areas such as toll stations and service areas are assigned higher values, while ordinary lanes are assigned lower values. It is the diffusion range weight coefficient, which is used to adjust the intensity and scope of event diffusion, reflecting the weight of diffusion characteristics in the influence coefficient. This value is adjusted according to the impact of event diffusion on traffic flow, congestion, etc. in the surrounding area. For example, the weight of the diffusion range is determined by analyzing the affected area and duration of historical events. It is the diffusion range factor, which indicates the diffusion intensity and coverage of the event, and reflects the degree of spread of the accident impact in physical space. It is obtained through the simulation calculation results of the diffusion model, combining the diffusion radius and the affected range. For example, it is set to a higher value in the high diffusion area to quantify the large-scale impact.

[0036] For example, if the time characteristic factor of the current event (Morning peak), spatial characteristic factors (located in the service area), diffusion range factor (High diffusion range). Set the weight coefficient , , .

[0037] Enter the formula for calculation: ; The result shows that the comprehensive impact coefficient of the event is 1.32.

[0038] See also Figure 8 ,The specific steps for obtaining the emergency resource scheduling management results are: Based on the impact weight of the event on the overall emergency response, the impact coefficient data of all events are summarized and sorted, and the event priorities are arranged in order from high to low to obtain the event priority ranking results; Taking the impact coefficient as the core basis for sorting, all high-speed traffic emergencies are prioritized to determine the response order of each event. The specific steps are as follows: First, summarize the impact coefficient data of each event, organize it into a table or list format, and arrange it in order from high to low according to the impact coefficient value. The higher the impact coefficient value, the greater the pressure and priority response demand of the event on the overall emergency system, so the higher the ranking. For example, for event A, event B, and event C, assume that their impact coefficients are 1.45, 1.32, and 1.20 respectively. Arrange these three events in order from large to small according to the impact coefficient, then event A has the highest priority, followed by event B, and event C has the lowest priority. Next, sort the events: event A ranks the highest priority, followed by event B, and finally event C. Display the sorting results in a list or table for subsequent resource allocation. For example, the final priority sorting result can be expressed as follows: Event A: impact coefficient 1.45, highest priority, Event B: impact coefficient 1.32, medium priority, Event C: impact coefficient 1.20, lowest priority. This priority sorting result provides the emergency management system with a clear event response order to ensure that high-priority events are handled in a timely manner and emergency resources are arranged.

[0039] According to the event priority ranking results, the parameters for dynamic allocation of emergency resources are adjusted. By adjusting the resource allocation ratio through the influence coefficient, high-priority events receive more concentrated resource allocation, and the emergency resource scheduling management results are obtained; First, in the priority sorting list, the resource allocation ratio is set according to the impact coefficient. For event A with the highest impact coefficient, the allocation parameters of the main resources are set to a higher value, for example, 80% of the rescue vehicles and personnel are allocated to ensure that the resource needs of high-priority events are met first. At the same time, according to the diffusion intensity factor of event A, the parameters for allocating resources are appropriately increased to improve the response intensity to the area. For secondary priority event B, the resource allocation parameters are set to a lower ratio according to its impact coefficient and priority, for example, 50% of the remaining resources are allocated to ensure sufficient emergency response. By adjusting the resource allocation ratio by the impact coefficient, high-priority events obtain more concentrated resource allocation. The system automatically dynamically allocates the ratio parameters according to the sorting results to form the final resource scheduling allocation parameter table, which provides a dynamic scheduling strategy based on event priority for the emergency system.

[0040] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. An emergency response coordination system based on big data, characterized in that: The system comprises: The data processing and event recognition module is based on the real-time data obtained by the high-speed traffic monitoring equipment and performs cluster analysis to classify the accident types in the accident report according to time characteristics and spatial characteristics, identify the similarity between the degree of high-speed traffic congestion and the accident type, and generate traffic congestion and accident pattern recognition data; The event dynamic prediction and risk assessment module performs time series prediction on the possibility of an accident based on the traffic congestion and accident pattern recognition data and according to the time characteristics and spatial characteristics, obtains accident probability prediction data, divides the accident probability prediction data into multiple risk levels, and generates accident risk level assessment data; The emergency response scope assessment module obtains the current high-speed traffic flow and vehicle density information based on the accident risk level assessment data, simulates the diffusion scope of the accident under various conditions through a diffusion model, obtains the accident diffusion scope data, divides the accident impact area into levels of distance and degree of impact according to the accident diffusion scope data, assesses the potential impact of the accident diffusion, and generates emergency response scope assessment data; The event linkage priority sorting and resource scheduling module conducts linkage impact analysis on all high-speed traffic emergencies based on the emergency response scope assessment data, evaluates the interrelationships between events, prioritizes all events by impact coefficients, obtains event priority sorting results, dynamically allocates emergency resources based on the event priority sorting results, and generates emergency resource scheduling management results.

2. The big data-based emergency response coordination system according to claim 1, characterized in that: The steps of obtaining the accident types in the accident report according to the time characteristics and the space characteristics are specifically as follows: Collect real-time data obtained by high-speed traffic monitoring equipment, including traffic flow, accident reports, event locations and vehicle density information, and pre-process them. Classify and group the processed data according to time and space characteristics, extract information on accident-prone areas and congested areas, and generate traffic congestion and accident pattern data sets; Based on the traffic congestion and accident pattern dataset, the data in the accident report is classified according to time characteristics and spatial characteristics. By analyzing the distribution of accident types in risk areas and specific time periods, the distribution patterns of accident types in multiple areas and time periods are identified to obtain accident characteristic data.

3. The big data-based emergency response coordination system according to claim 2, characterized in that: The steps for acquiring the traffic congestion and accident pattern recognition data are specifically as follows: Based on the accident characteristic data, the formula is adopted: ； Calculate the similarity score of the incident , used to evaluate the similarity of accidents in time and space characteristics, and obtain accident characteristics and similarity score data; in, It is The frequency of accidents with time characteristics, is the weight of the time feature, It is The frequency of accidents with spatial characteristics, is the weight of the spatial feature, is the total number of event logs; Based on the accident characteristics and similarity score data, by comparing the relationship between traffic flow and accident frequency, congestion patterns in areas with multiple accident frequencies and risk periods are identified, and traffic congestion and accident pattern recognition data are generated.

4. The big data-based emergency response coordination system according to claim 3, characterized in that: The steps for obtaining the accident risk level assessment data are specifically as follows: Based on the traffic congestion and accident pattern recognition data, the formula is adopted: ； Calculate the predicted value of the accident probability in the next time period , get the accident probability prediction data; in, Is the current time period The probability of an accident occurring, is the adjustment factor, Is the current time period frequency of accidents; Based on the accident probability prediction data, the predicted values ​​are divided into multiple risk levels, the corresponding risk time periods and areas are marked, and range annotation is performed to generate accident risk level assessment data.

5. The big data-based emergency response coordination system according to claim 4, characterized in that: The steps for obtaining the accident diffusion range data are specifically as follows: Based on the accident risk level assessment data, the current high-speed traffic flow and vehicle density information is obtained, the corresponding diffusion intensity parameters are set according to the risk level of the area, and the diffusion radius at each time point in the diffusion model is recorded to obtain preliminary diffusion simulation data; Based on the preliminary diffusion simulation data, the diffusion radius of the risk area is sorted and labeled, and the risk level and diffusion range in the area are displayed in multiple colors to generate accident diffusion range data.

6. The big data-based emergency response coordination system according to claim 5, characterized in that: The steps for obtaining the emergency response scope assessment data are specifically as follows: Based on the accident diffusion range data, the accident impact on the risk area is hierarchically divided into distance and degree of impact, the accident impact area is divided into multiple circles according to distance, the impact degree of each circle is marked, and a hierarchical division data table is established; Based on the hierarchical division data table, the formula is used: ； Calculate the predicted diffusion time and the predicted diffusion intensity , generate emergency response scope assessment data; in, is the predicted diffusion distance, is the propagation speed in the model, is the model adjustment coefficient, is the initial accident impact, It is a simulated environmental impact factor.

7. The big data-based emergency response coordination system according to claim 6, characterized in that: The steps for obtaining the correlation between the evaluation events are specifically as follows: Based on the emergency response scope assessment data, the linkage impact of the high-speed traffic emergency events is analyzed, including analyzing the time characteristics, location and diffusion range data of the events, classifying the time period, spatial location and diffusion characteristics of the events, and establishing a basic data set for event linkage analysis; Based on the event linkage analysis basic data set, the formula is used: ； Calculate the impact coefficient of an event , get the impact weight of the event on the overall emergency response; in, is the time feature weight coefficient, is the time characteristic factor, is the spatial feature weight coefficient, is the spatial characteristic factor, is the diffusion range weight coefficient, is the diffusion range factor.

8. The big data-based emergency response coordination system according to claim 7, characterized in that: The steps for obtaining the emergency resource scheduling management result are specifically as follows: Based on the impact weight of the event on the overall emergency response, the impact coefficient data of all events are summarized and sorted, and the event priorities are arranged in order from high to low to obtain the event priority ranking result; According to the event priority ranking result, the parameters for dynamic allocation of emergency resources are adjusted, and the resource allocation ratio is adjusted by the influence coefficient, so that high-priority events obtain more concentrated resource allocation and obtain emergency resource scheduling management results.

Images