Regional traffic-oriented travel portrait and congestion association analysis method and system

Abstract

The invention provides a regional traffic-oriented travel portrait and congestion association analysis method and system. The method comprises the following steps: obtaining a cross-domain multi-source traffic data set, carrying out data preprocessing to obtain a cross-domain multi-source traffic fusion data set, carrying out processing according to the cross-domain multi-source traffic fusion data set to obtain a regional travel portrait, and carrying out congestion causal association analysis according to the cross-domain multi-source traffic fusion data set and the regional travel portrait. Obtaining a causal association intensity matrix, and performing analysis processing according to the regional travel portrait and the causal association intensity matrix to obtain congestion propagation prediction data; according to the method, the cross-domain multi-source traffic data is accurately fused, the accurate regional travel portrait is constructed, and the congestion propagation prediction data is analyzed and evaluated in combination with the congestion causal association, so that the travel portrait and congestion association intelligent analysis facing regional traffic is realized.

Description

Technical Field

[0001] This application relates to the field of intelligent traffic management technology, and more specifically, to a method and system for analyzing the correlation between travel profiles and congestion in regional traffic. Background Technology

[0002] The integrated development of urban areas has led to increasingly complex relationships between cross-city travel traffic flow and congestion. Travel profiling and congestion correlation analysis have become key points for regional traffic governance. Traditional technologies suffer from insufficient cross-departmental data integration, resulting in missing travel profile features. Analysis models focusing on single-region scenarios lack generalization, and congestion analysis and governance strategies lack specificity. Furthermore, data privacy processing and cross-regional flow adaptation optimization are inadequate. Therefore, there is an urgent need for an intelligent method to achieve travel profiling and congestion correlation analysis for regional traffic.

[0003] Effective technical solutions are urgently needed to address the above problems. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for analyzing travel profiles and congestion correlations for regional transportation. This method can achieve intelligent analysis of travel profiles and congestion correlations for regional transportation by accurately integrating cross-domain multi-source traffic data, constructing accurate regional travel profiles, and combining congestion causal correlation analysis to evaluate congestion propagation prediction data.

[0005] This application also provides a method for analyzing the correlation between travel profiles and congestion in regional transportation, including the following steps: Obtain cross-domain multi-source traffic datasets and perform data preprocessing to obtain cross-domain multi-source traffic fusion datasets; The regional travel profile is obtained by processing the cross-domain multi-source traffic fusion dataset. Based on the cross-domain multi-source traffic fusion dataset and regional travel profile, a causal correlation analysis of congestion is performed to obtain a causal correlation strength matrix. Based on the regional travel profile and causal correlation strength matrix, congestion propagation prediction data are obtained through analysis and processing.

[0006] Optionally, in the method for travel profiling and congestion correlation analysis for regional traffic described in this application, the step of obtaining a cross-domain multi-source traffic dataset and performing data preprocessing to obtain a cross-domain multi-source traffic fusion dataset includes: Acquire cross-domain, multi-source traffic datasets, including traffic monitoring records, traffic sensor records, and traffic survey records; The traffic monitoring record data, traffic sensor record data, and traffic survey record data are preprocessed by data cleaning, timestamp alignment, and data standardization to obtain standard traffic monitoring record data, standard traffic sensor record data, and standard traffic survey record data. According to the traffic monitoring standard, the recorded data is processed by a preset CLIP model to obtain unstructured vehicle feature data. Based on the standard traffic sensor data and the standard traffic survey data, time alignment processing is performed using a preset Dynamic Time Warping (DTW) algorithm to obtain traffic sensor time series data. The unstructured feature data of the vehicles and the time series data of traffic sensors are fused together to obtain cross-domain multi-source traffic initial fusion data. The initial cross-domain multi-source traffic fusion data is processed with differential privacy noise addition and bias correction to obtain a cross-domain multi-source traffic fusion dataset.

[0007] Optionally, in the method for travel profiling and congestion correlation analysis for regional traffic described in this application, the step of performing differential privacy noise addition and bias correction processing on the initial cross-domain multi-source traffic fusion data to obtain a cross-domain multi-source traffic fusion dataset includes: Obtain the sensitivity value of the initial fusion data of the cross-domain multi-source traffic; A privacy budget is allocated based on the sensitivity value to obtain the privacy budget value; Based on the sensitivity value and privacy budget value, noise is added using a preset noise addition method to obtain cross-domain multi-source traffic noisy fusion data; The cross-domain multi-source traffic noise-containing fusion data is processed by deviation analysis using preset traffic conservation constraint rules to obtain data deviation values. The data deviation value is compared with a preset data deviation threshold. If the data deviation value is less than or equal to the preset data deviation threshold, then the cross-domain multi-source traffic noisy fusion data is determined to be a cross-domain multi-source traffic fusion dataset. If the data deviation value is greater than the preset data deviation threshold, then noise reduction processing is performed according to the preset gradient, and deviation correction processing is performed. If the deviation correction is satisfactory, a cross-domain multi-source traffic fusion dataset is obtained; otherwise, an early warning response is output.

[0008] Optionally, in the method for analyzing the correlation between regional traffic travel profiles and congestion described in this application, the step of processing the cross-domain multi-source traffic fusion dataset to obtain a regional travel profile includes: Data is extracted from the cross-domain multi-source traffic fusion dataset to obtain regional feature data, including cross-city travel frequency, travel distance type feature data, and travel traffic combination type feature data. The cross-city travel frequency, travel distance type feature data, and travel transportation combination type feature data are input into a preset travel group category analysis model for processing to obtain travel group category feature data. Acquire intercity ETC record data, city checkpoint record data, mobile phone signaling record data, POI data, and real-time weather record data; Based on the intercity ETC record data, city checkpoint record data, mobile phone signaling record data, POI data, real-time weather record data, cross-city travel frequency, travel distance type characteristic data, travel traffic combination type characteristic data, and travel group category characteristic data, a regional travel profile is obtained by processing the data through a preset clustering analysis algorithm.

[0009] Optionally, in the method for analyzing travel profiles and congestion correlations for regional traffic described in this application, the step of performing congestion causal correlation analysis based on the cross-domain multi-source traffic fusion dataset and regional travel profiles to obtain a causal correlation strength matrix includes: By performing correlation analysis between the cross-domain multi-source traffic fusion dataset and the regional travel profile, data on the size of the travel group and the behavioral characteristics of the travel group can be obtained. Acquire road segment operation status monitoring data and time period scene characteristic data; Based on the travel group size data and travel group behavior characteristic data, combined with the road segment operation status monitoring data and time period scene characteristic data, the data are analyzed and processed through a preset Bayesian network model to obtain a causal correlation strength matrix.

[0010] Optionally, in the method for analyzing travel profiles and congestion correlations for regional traffic described in this application, the step of analyzing and processing the regional travel profiles and causal correlation strength matrices to obtain congestion propagation prediction data includes: Data is extracted based on the causal correlation strength matrix within a preset time period to obtain travel user behavior data, road network topology data, and spatial environment data. Based on the regional travel profile, extract the congestion contribution value corresponding to the travel group category feature data; Acquire real-time traffic flow characteristic data and event type characteristic data; The user travel behavior data, road network topology data, spatial environment data, as well as the congestion contribution value, real-time traffic flow characteristic data, and event type characteristic data are input into a preset congestion prediction model for processing to obtain congestion propagation prediction data.

[0011] Optionally, the method for analyzing the correlation between travel profiles and congestion in regional transportation described in this application also includes: The initial traffic control strategy is obtained by querying the preset traffic control strategy library based on the congestion propagation prediction data. Based on the traffic strategy parameter data corresponding to the initial traffic control strategy, as well as the real-time traffic flow characteristic data and regional travel profile, the data are processed through a preset road network operation simulation model to obtain road network operation status prediction data, including congestion relief rate and vehicle delay reduction rate. The congestion mitigation rate and the vehicle delay reduction rate are weighted and summed to obtain the traffic strategy effectiveness index. The traffic strategy effectiveness indices are sorted in descending order, and the initial traffic control strategy with the largest traffic strategy effectiveness index is determined as the traffic control strategy.

[0012] Secondly, this application provides a system for analyzing travel profiles and congestion correlations for regional traffic. The system includes a memory and a processor. The memory includes a program for a method of analyzing travel profiles and congestion correlations for regional traffic. When the processor executes the program for this method, it performs the following steps: Obtain cross-domain multi-source traffic datasets and perform data preprocessing to obtain cross-domain multi-source traffic fusion datasets; The regional travel profile is obtained by processing the cross-domain multi-source traffic fusion dataset. Based on the cross-domain multi-source traffic fusion dataset and regional travel profile, a causal correlation analysis of congestion is performed to obtain a causal correlation strength matrix. Based on the regional travel profile and causal correlation strength matrix, congestion propagation prediction data are obtained through analysis and processing.

[0013] Optionally, in the travel profiling and congestion correlation analysis system for regional traffic described in this application, the step of acquiring cross-domain multi-source traffic datasets and performing data preprocessing to obtain cross-domain multi-source traffic fusion datasets includes: Acquire cross-domain, multi-source traffic datasets, including traffic monitoring records, traffic sensor records, and traffic survey records; The traffic monitoring record data, traffic sensor record data, and traffic survey record data are preprocessed by data cleaning, timestamp alignment, and data standardization to obtain standard traffic monitoring record data, standard traffic sensor record data, and standard traffic survey record data. According to the traffic monitoring standard, the recorded data is processed by a preset CLIP model to obtain unstructured vehicle feature data. Based on the standard traffic sensor data and the standard traffic survey data, time alignment processing is performed using a preset Dynamic Time Warping (DTW) algorithm to obtain traffic sensor time series data. The unstructured feature data of the vehicles and the time series data of traffic sensors are fused together to obtain cross-domain multi-source traffic initial fusion data. The initial cross-domain multi-source traffic fusion data is processed with differential privacy noise addition and bias correction to obtain a cross-domain multi-source traffic fusion dataset.

[0014] Optionally, in the travel profiling and congestion correlation analysis system for regional traffic described in this application, the step of performing differential privacy noise addition and bias correction processing on the initial cross-domain multi-source traffic fusion data to obtain a cross-domain multi-source traffic fusion dataset includes: Obtain the sensitivity value of the initial fusion data of the cross-domain multi-source traffic; A privacy budget is allocated based on the sensitivity value to obtain the privacy budget value; Based on the sensitivity value and privacy budget value, noise is added using a preset noise addition method to obtain cross-domain multi-source traffic noisy fusion data; The cross-domain multi-source traffic noise-containing fusion data is processed by deviation analysis using preset traffic conservation constraint rules to obtain data deviation values. The data deviation value is compared with a preset data deviation threshold. If the data deviation value is less than or equal to the preset data deviation threshold, then the cross-domain multi-source traffic noisy fusion data is determined to be a cross-domain multi-source traffic fusion dataset. If the data deviation value is greater than the preset data deviation threshold, then noise reduction processing is performed according to the preset gradient, and deviation correction processing is performed. If the deviation correction is satisfactory, a cross-domain multi-source traffic fusion dataset is obtained; otherwise, an early warning response is output.

[0015] As can be seen from the above, the method and system for travel profiling and congestion correlation analysis for regional traffic provided in this application achieve intelligent analysis of travel profiling and congestion correlation for regional traffic by accurately integrating cross-domain multi-source traffic data, constructing accurate regional travel profiles, and combining congestion causal correlation analysis to evaluate congestion propagation prediction data.

[0016] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating a method for analyzing the correlation between travel profiles and congestion in regional transportation, provided as an embodiment of this application; Figure 2 A flowchart illustrating the process of obtaining a cross-domain multi-source traffic fusion dataset using a method for travel profiling and congestion correlation analysis of regional traffic, as provided in this application embodiment; Figure 3 A flowchart illustrating the method for obtaining a regional travel profile and congestion correlation analysis for regional traffic, as provided in this application embodiment; Figure 4 This is a flowchart illustrating the method for obtaining congestion propagation prediction data in a regional traffic travel profiling and congestion correlation analysis provided in this application embodiment. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] Please refer to Figure 1 , Figure 1This is a flowchart illustrating a method for analyzing the correlation between travel profiles and congestion in regional traffic, as described in some embodiments of this application. This method is used in terminal devices, such as computers and mobile phones. The method includes the following steps: S11. Obtain cross-domain multi-source traffic datasets and perform data preprocessing to obtain cross-domain multi-source traffic fusion datasets; S12. Process the cross-domain multi-source traffic fusion dataset to obtain a regional travel profile; S13. Based on the cross-domain multi-source traffic fusion dataset and regional travel profile, perform congestion causal correlation analysis to obtain the causal correlation strength matrix; S14. Analyze and process the regional travel profile and causal correlation strength matrix to obtain congestion propagation prediction data.

[0022] It should be noted that, in order to achieve intelligent analysis of cross-regional traffic travel profiles and congestion correlations, cross-domain multi-source data is first collected and preprocessed to construct fused data. Then, regional travel profiles are constructed through feature extraction and typical group analysis. Next, causal correlation analysis is used to generate a causal correlation strength matrix that includes groups, road segments, and time periods. Finally, the fused data and regional travel profiles are combined to evaluate congestion propagation prediction data, and then appropriate traffic management strategies are matched based on the prediction results.

[0023] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating a method for obtaining a cross-domain multi-source traffic fusion dataset using a travel profiling and congestion correlation analysis approach for regional traffic, as described in some embodiments of this application. According to embodiments of the present invention, obtaining the cross-domain multi-source traffic dataset and performing data preprocessing to obtain the cross-domain multi-source traffic fusion dataset includes: S21. Obtain cross-domain multi-source traffic datasets, including traffic monitoring record data, traffic sensor record data, and traffic survey record data; S22. Perform data cleaning, timestamp alignment and data standardization preprocessing on the traffic monitoring record data, traffic sensor record data and traffic survey record data to obtain standard traffic monitoring record data, standard traffic sensor record data and standard traffic survey record data. S23. Based on the traffic monitoring standard, the recorded data is processed by a preset CLIP model to obtain unstructured vehicle feature data. S24. Based on the standard traffic sensor record data and the standard traffic survey record data, time alignment processing is performed using a preset dynamic time warping (DTW) algorithm to obtain traffic sensor time series data. S25. Perform data fusion processing on the unstructured feature data of the vehicles and the time series data of traffic sensors to obtain cross-domain multi-source traffic initial fusion data; S26. Perform differential privacy noise addition and bias correction processing on the initial cross-domain multi-source traffic fusion data to obtain a cross-domain multi-source traffic fusion dataset.

[0024] It should be noted that, in order to achieve the adaptation and fusion processing of cross-domain traffic data, firstly, a cross-domain multi-source traffic dataset including traffic monitoring record data, traffic sensor record data, and traffic survey record data is collected. For traffic monitoring record data, such as traffic monitoring videos, a CLIP-B / 32 pre-trained model is selected, and the model is fine-tuned through a large number of historical traffic scene datasets to obtain a preset CLIP model. Unstructured vehicle feature data is extracted, such as vehicle type coding data (cars are 1, trucks are 2) and driving status coding data (normal driving is 1, slow driving is 2). High-frequency traffic sensor record data is aggregated at a 1-minute granularity, and the average traffic flow and average vehicle speed within each minute are calculated to form a time series. Low-frequency traffic survey record data, such as the total traffic flow and average travel distance per hour, are also formed into a time series. Based on the construction of a distance matrix and the search for the optimal matching path, the data is processed using a preset DTW algorithm. The low-frequency indicators are assigned to the corresponding high-frequency time slices according to the matching path to generate unified traffic sensor time series data. Finally, data fusion and data privacy processing are performed.

[0025] According to an embodiment of the present invention, the step of performing differential privacy noise addition and bias correction processing on the initial fusion data of cross-domain multi-source traffic to obtain a cross-domain multi-source traffic fusion dataset includes: Obtain the sensitivity value of the initial fusion data of the cross-domain multi-source traffic; A privacy budget is allocated based on the sensitivity value to obtain the privacy budget value; Based on the sensitivity value and privacy budget value, noise is added using a preset noise addition method to obtain cross-domain multi-source traffic noisy fusion data; The cross-domain multi-source traffic noise-containing fusion data is processed by deviation analysis using preset traffic conservation constraint rules to obtain data deviation values. The data deviation value is compared with a preset data deviation threshold. If the data deviation value is less than or equal to the preset data deviation threshold, then the cross-domain multi-source traffic noisy fusion data is determined to be a cross-domain multi-source traffic fusion dataset. If the data deviation value is greater than the preset data deviation threshold, then noise reduction processing is performed according to the preset gradient, and deviation correction processing is performed. If the deviation correction is satisfactory, a cross-domain multi-source traffic fusion dataset is obtained; otherwise, an early warning response is output.

[0026] It should be noted that, in order to perform data privacy processing, firstly, the sensitivity value of the initial cross-domain multi-source traffic fusion data is obtained (determined by calculating the maximum change when the data undergoes a small change), and a privacy budget is allocated proportionally to obtain the privacy budget value. Then, noise is added. Next, deviation analysis is performed using preset traffic conservation constraint rules (pre-constructed based on the relationship between nodes and road segments in the road network topology, such as for any intersection, the sum of traffic flow of all inflow road segments = the sum of traffic flow of all outflow road segments) to obtain the data deviation value. Finally, the validity of the noise addition processing is determined by threshold comparison.

[0027] Please refer to Figure 3 , Figure 3 This is a flowchart illustrating a method for obtaining a regional travel profile based on a travel profile and congestion correlation analysis for regional traffic, as described in some embodiments of this application. According to an embodiment of the present invention, the step of processing the cross-domain multi-source traffic fusion dataset to obtain the regional travel profile includes: S31. Extract data from the cross-domain multi-source traffic fusion dataset to obtain regional feature data, including cross-city travel frequency, travel distance type feature data, and travel traffic combination type feature data. S32. Input the cross-city travel frequency, travel distance type feature data and travel transportation combination type feature data into a preset travel group category analysis model for processing to obtain travel group category feature data; S33. Obtain intercity ETC record data, city checkpoint record data, mobile phone signaling record data, POI data, and real-time weather record data; S34. Based on the intercity ETC record data, city checkpoint record data, mobile phone signaling record data, POI data, real-time weather record data, cross-city travel frequency, travel distance type feature data, travel traffic combination type feature data, and travel group category feature data, the data are processed by a preset clustering analysis algorithm to obtain a regional travel profile.

[0028] It should be noted that, in order to construct an accurate regional travel profile, firstly, regional feature data is extracted from the cross-domain multi-source traffic fusion dataset (cross-city travel frequency refers to the number of cross-city trips within a week; travel distance includes short or long distances; travel transportation combination type is such as driving + train; travel distance type feature data and travel transportation combination type feature data are identified by different identifiers). This data is then processed through a pre-set travel group category analysis model to obtain travel group category feature data, such as cross-city commuting category feature data (1) and tourism transit category feature data (2). Next, travel behavior, spatial scene, and environmental data, including intercity ETC record data, city checkpoint record data, mobile phone signaling record data, POI data, and real-time weather record data, are acquired. Finally, the extracted data, travel behavior, spatial scene, and environmental data, along with the identified travel group category feature data, are processed through a pre-set clustering analysis algorithm to obtain the regional travel profile. The pre-set travel group category analysis model is trained by acquiring a large number of historical samples of cross-city travel frequency, travel distance type feature data, travel transportation combination type feature data, and corresponding travel group category feature data.

[0029] According to an embodiment of the present invention, the step of performing congestion causal correlation analysis based on the cross-domain multi-source traffic fusion dataset and regional travel profile to obtain a causal correlation strength matrix includes: By performing correlation analysis between the cross-domain multi-source traffic fusion dataset and the regional travel profile, data on the size of the travel group and the behavioral characteristics of the travel group are obtained. Acquire road segment operation status monitoring data and time period scene characteristic data; Based on the travel group size data and travel group behavior characteristic data, combined with the road segment operation status monitoring data and time period scene characteristic data, the data are analyzed and processed through a preset Bayesian network model to obtain a causal correlation strength matrix.

[0030] It should be noted that, in order to construct a correlation matrix including travel groups, operating road segments, and travel time, and to assess the impact on congestion, firstly, the cross-domain multi-source traffic fusion dataset corresponding to the travel group categories identified by cluster analysis is correlated with the constructed regional travel profile, including travel group size data (such as vehicle or pedestrian traffic volume on a certain road segment during the midday peak) and travel group behavioral characteristic data (such as the average vehicle speed and average dwell time of a travel group category on a certain road segment). Then, road segment operation status monitoring data related to the operating road segments are obtained, such as congestion intensity, average delay time, number of lanes on a certain road segment, and design speed, as well as time period scenario characteristic data related to travel time, such as morning peak, off-peak, and evening peak hours. The road segment operation status monitoring data and time period scenario characteristic data are represented by different identifiers. Finally, a pre-set Bayesian network model is used to process the data based on causal constraint rules (such as the high concentration of manufacturing industries around the road segment attracting more cross-city commuters) to determine the causal correlation strength matrix for assessing congestion.

[0031] Please refer to Figure 4 , Figure 4 This is a flowchart illustrating a method for obtaining congestion propagation prediction data based on a regional traffic travel profile and congestion correlation analysis in some embodiments of this application. According to an embodiment of the present invention, the step of analyzing and processing the regional travel profile and causal correlation strength matrix to obtain congestion propagation prediction data includes: S411. Extract data based on the causal correlation strength matrix within a preset time period to obtain travel user behavior data, road network topology data, and spatial environment data. S412. Extract the congestion contribution value corresponding to the travel group category feature data based on the regional travel profile; S413. Obtain real-time traffic flow characteristic data and event type characteristic data; S42. Input the travel user behavior data, road network topology data, spatial environment data, as well as the congestion contribution value, real-time traffic flow characteristic data, and event type characteristic data into a preset congestion prediction model for processing to obtain congestion propagation prediction data.

[0032] It should be noted that, in order to assess the impact on traffic congestion in the future, the causal correlation strength matrix for a historical preset time period is first extracted. For example, over the past year, travel user behavior data, road network topology data, and spatial environment data are determined. Travel user behavior data includes combinations of travel user time, route, and mode of transportation; road network topology data includes road segment numbers, number of lanes on a given road segment, and design speed; and spatial environment data includes regional POI distribution data and meteorological data. Simultaneously, by determining the proportion of different travel group categories, the ratio of average travel speed to average group speed, and the corresponding pre-constructed preset congestion correlation coefficient, the product of these three factors is obtained. The corresponding congestion contribution value is obtained. Then, real-time traffic flow characteristic data (such as real-time traffic volume, average vehicle speed, and lane occupancy rate) and event type characteristic data are acquired. For example, if an event type is a traffic accident, it is recorded as 1 if it exists and 0 if it does not exist. Finally, the data is normalized and then processed through a preset congestion prediction model to obtain congestion propagation prediction data, which is used to evaluate the congestion propagation path and propagation intensity. The preset congestion prediction model is obtained by training a large amount of historical sample travel user behavior data, road network topology data, spatial environment data, the congestion contribution value, real-time traffic flow characteristic data, event type characteristic data, and the corresponding congestion propagation prediction data.

[0033] According to an embodiment of the present invention, it further includes: The initial traffic control strategy is obtained by querying the preset traffic control strategy library based on the congestion propagation prediction data. Based on the traffic strategy parameter data corresponding to the initial traffic control strategy, as well as the real-time traffic flow characteristic data and regional travel profile, the data are processed through a preset road network operation simulation model to obtain road network operation status prediction data, including congestion relief rate and vehicle delay reduction rate. The congestion mitigation rate and the vehicle delay reduction rate are weighted and summed to obtain the traffic strategy effectiveness index. The traffic strategy effectiveness indices are sorted in descending order, and the initial traffic control strategy with the largest traffic strategy effectiveness index is determined as the traffic control strategy.

[0034] It should be noted that the congestion propagation prediction data obtained through the evaluation is based on a pre-set traffic control strategy library constructed by those skilled in the art from a large number of historical sample cases. The corresponding traffic strategy parameter data, real-time traffic flow characteristic data, and regional travel profiles are input into a pre-set road network operation simulation model for processing to obtain road network operation status prediction data. Then, through weighted summation, the traffic strategy effectiveness index is determined. Finally, the initial traffic control strategy with the largest traffic strategy effectiveness index is determined as the traffic control strategy, thereby determining the best matching strategy. The pre-set road network operation simulation model is obtained by training with a large number of historical sample traffic strategy parameter data, real-time traffic flow characteristic data, regional travel profiles, and corresponding road network operation status prediction data.

[0035] This invention also discloses a travel profiling and congestion correlation analysis system for regional traffic, including a memory and a processor. The memory includes a method program for travel profiling and congestion correlation analysis for regional traffic. When the processor executes the method program for travel profiling and congestion correlation analysis for regional traffic, it performs the following steps: Obtain cross-domain multi-source traffic datasets and perform data preprocessing to obtain cross-domain multi-source traffic fusion datasets; The regional travel profile is obtained by processing the cross-domain multi-source traffic fusion dataset. Based on the cross-domain multi-source traffic fusion dataset and regional travel profile, a causal correlation analysis of congestion is performed to obtain a causal correlation strength matrix. Based on the regional travel profile and causal correlation strength matrix, congestion propagation prediction data are obtained through analysis and processing.

[0036] It should be noted that, in order to achieve intelligent analysis of cross-regional traffic travel profiles and congestion correlations, cross-domain multi-source data is first collected and preprocessed to construct fused data. Then, regional travel profiles are constructed through feature extraction and typical group analysis. Next, causal correlation analysis is used to generate a causal correlation strength matrix that includes groups, road segments, and time periods. Finally, the fused data and regional travel profiles are combined to evaluate congestion propagation prediction data, and then appropriate traffic management strategies are matched based on the prediction results.

[0037] According to an embodiment of the present invention, the step of acquiring a cross-domain multi-source traffic dataset and performing data preprocessing to obtain a cross-domain multi-source traffic fusion dataset includes: Acquire cross-domain, multi-source traffic datasets, including traffic monitoring records, traffic sensor records, and traffic survey records; The traffic monitoring record data, traffic sensor record data, and traffic survey record data are preprocessed by data cleaning, timestamp alignment, and data standardization to obtain standard traffic monitoring record data, standard traffic sensor record data, and standard traffic survey record data. According to the traffic monitoring standard, the recorded data is processed by a preset CLIP model to obtain unstructured vehicle feature data. Based on the standard traffic sensor data and the standard traffic survey data, time alignment processing is performed using a preset Dynamic Time Warping (DTW) algorithm to obtain traffic sensor time series data. The unstructured feature data of the vehicles and the time series data of traffic sensors are fused together to obtain cross-domain multi-source traffic initial fusion data. The initial cross-domain multi-source traffic fusion data is processed with differential privacy noise addition and bias correction to obtain a cross-domain multi-source traffic fusion dataset.

[0038] It should be noted that, in order to achieve the adaptation and fusion processing of cross-domain traffic data, firstly, a cross-domain multi-source traffic dataset including traffic monitoring record data, traffic sensor record data, and traffic survey record data is collected. For traffic monitoring record data, such as traffic monitoring videos, a CLIP-B / 32 pre-trained model is selected, and the model is fine-tuned through a large number of historical traffic scene datasets to obtain a preset CLIP model. Unstructured vehicle feature data is extracted, such as vehicle type coding data (cars are 1, trucks are 2) and driving status coding data (normal driving is 1, slow driving is 2). High-frequency traffic sensor record data is aggregated at a 1-minute granularity, and the average traffic flow and average vehicle speed within each minute are calculated to form a time series. Low-frequency traffic survey record data, such as the total traffic flow and average travel distance per hour, are also formed into a time series. Based on the construction of a distance matrix and the search for the optimal matching path, the data is processed through a preset DTW algorithm. The low-frequency indicators are assigned to the corresponding high-frequency time slices according to the matching path to generate unified traffic sensor time series data. Finally, data fusion and data privacy processing are performed.

[0039] According to an embodiment of the present invention, the step of performing differential privacy noise addition and bias correction processing on the initial cross-domain multi-source traffic fusion data to obtain a cross-domain multi-source traffic fusion dataset includes: Obtain the sensitivity value of the initial fusion data of the cross-domain multi-source traffic; A privacy budget is allocated based on the sensitivity value to obtain the privacy budget value; Based on the sensitivity value and privacy budget value, noise is added using a preset noise addition method to obtain cross-domain multi-source traffic noisy fusion data; The cross-domain multi-source traffic noise-containing fusion data is processed by deviation analysis using preset traffic conservation constraint rules to obtain data deviation values. The data deviation value is compared with a preset data deviation threshold. If the data deviation value is less than or equal to the preset data deviation threshold, then the cross-domain multi-source traffic noisy fusion data is determined to be a cross-domain multi-source traffic fusion dataset. If the data deviation value is greater than the preset data deviation threshold, then noise reduction processing is performed according to the preset gradient, and deviation correction processing is performed. If the deviation correction is satisfactory, a cross-domain multi-source traffic fusion dataset is obtained; otherwise, an early warning response is output.

[0040] It should be noted that, in order to perform data privacy processing, firstly, the sensitivity value of the initial cross-domain multi-source traffic fusion data is obtained (determined by calculating the maximum change when the data undergoes a small change), and a privacy budget is allocated proportionally to obtain the privacy budget value. Then, noise is added. Next, deviation analysis is performed using preset traffic conservation constraint rules (pre-constructed based on the relationship between nodes and road segments in the road network topology, such as for any intersection, the sum of traffic flow of all inflow road segments = the sum of traffic flow of all outflow road segments) to obtain the data deviation value. Finally, the validity of the noise addition processing is determined by threshold comparison.

[0041] According to an embodiment of the present invention, the step of processing the cross-domain multi-source traffic fusion dataset to obtain a regional travel profile includes: Data is extracted from the cross-domain multi-source traffic fusion dataset to obtain regional feature data, including cross-city travel frequency, travel distance type feature data, and travel traffic combination type feature data. The cross-city travel frequency, travel distance type feature data, and travel transportation combination type feature data are input into a preset travel group category analysis model for processing to obtain travel group category feature data. Acquire intercity ETC record data, city checkpoint record data, mobile phone signaling record data, POI data, and real-time weather record data; Based on the intercity ETC record data, city checkpoint record data, mobile phone signaling record data, POI data, real-time weather record data, cross-city travel frequency, travel distance type characteristic data, travel traffic combination type characteristic data, and travel group category characteristic data, a regional travel profile is obtained by processing the data through a preset clustering analysis algorithm.

[0042] It should be noted that, in order to construct an accurate regional travel profile, firstly, regional feature data is extracted from the cross-domain multi-source traffic fusion dataset (cross-city travel frequency refers to the number of cross-city trips within a week; travel distance includes short or long distances; travel transportation combination type is such as driving + train; travel distance type feature data and travel transportation combination type feature data are identified by different identifiers). This data is then processed through a pre-set travel group category analysis model to obtain travel group category feature data, such as cross-city commuting category feature data (1) and tourism transit category feature data (2). Next, travel behavior, spatial scene, and environmental data, including intercity ETC record data, city checkpoint record data, mobile phone signaling record data, POI data, and real-time weather record data, are acquired. Finally, the extracted data, travel behavior, spatial scene, and environmental data, along with the identified travel group category feature data, are combined and processed through a pre-set clustering analysis algorithm to obtain the regional travel profile. The pre-set travel group category analysis model is trained by acquiring a large number of historical samples of cross-city travel frequency, travel distance type feature data, travel transportation combination type feature data, and corresponding travel group category feature data.

[0043] According to an embodiment of the present invention, the step of performing congestion causal correlation analysis based on the cross-domain multi-source traffic fusion dataset and regional travel profile to obtain a causal correlation strength matrix includes: By performing correlation analysis between the cross-domain multi-source traffic fusion dataset and the regional travel profile, data on the size of the travel group and the behavioral characteristics of the travel group can be obtained. Acquire road segment operation status monitoring data and time period scene characteristic data; Based on the travel group size data and travel group behavior characteristic data, combined with the road segment operation status monitoring data and time period scene characteristic data, the data are analyzed and processed through a preset Bayesian network model to obtain a causal correlation strength matrix.

[0044] It should be noted that, in order to construct a correlation matrix including travel groups, operating road segments, and travel time, and to assess the impact on congestion, firstly, the cross-domain multi-source traffic fusion dataset corresponding to the travel group categories identified by cluster analysis is correlated with the constructed regional travel profile, including travel group size data (such as vehicle or pedestrian traffic volume on a certain road segment during the midday peak) and travel group behavioral characteristic data (such as the average vehicle speed and average dwell time of a travel group category on a certain road segment). Then, road segment operation status monitoring data related to the operating road segments are obtained, such as congestion intensity, average delay time, number of lanes on a certain road segment, and design speed, as well as time period scenario characteristic data related to travel time, such as morning peak, off-peak, and evening peak hours. The road segment operation status monitoring data and time period scenario characteristic data are represented by different identifiers. Finally, a pre-set Bayesian network model is used to process the data based on causal constraint rules (such as the high concentration of manufacturing industries around the road segment attracting more cross-city commuters) to determine the causal correlation strength matrix for assessing congestion.

[0045] According to an embodiment of the present invention, the step of analyzing and processing the regional travel profile and the causal correlation strength matrix to obtain congestion propagation prediction data includes: Data is extracted based on the causal correlation strength matrix within a preset time period to obtain travel user behavior data, road network topology data, and spatial environment data. Based on the regional travel profile, extract the congestion contribution value corresponding to the travel group category feature data; Acquire real-time traffic flow characteristic data and event type characteristic data; The user travel behavior data, road network topology data, spatial environment data, as well as the congestion contribution value, real-time traffic flow characteristic data, and event type characteristic data are input into a preset congestion prediction model for processing to obtain congestion propagation prediction data.

[0046] It should be noted that, in order to assess the impact on traffic congestion in the future, the causal correlation strength matrix for a historical preset time period is first extracted. For example, over the past year, travel user behavior data, road network topology data, and spatial environment data are determined. Travel user behavior data includes combinations of travel user time, route, and mode of transportation; road network topology data includes road segment numbers, number of lanes on a given road segment, and design speed; and spatial environment data includes regional POI distribution data and meteorological data. Simultaneously, by determining the proportion of different travel group categories, the ratio of average travel speed to average group speed, and the corresponding pre-constructed preset congestion correlation coefficient, the product of these three factors is obtained. The corresponding congestion contribution value is obtained. Then, real-time traffic flow characteristic data (such as real-time traffic volume, average vehicle speed, and lane occupancy rate) and event type characteristic data are acquired. For example, if an event type is a traffic accident, it is recorded as 1 if it exists and 0 if it does not exist. Finally, the data is normalized and then processed through a preset congestion prediction model to obtain congestion propagation prediction data, which is used to evaluate the congestion propagation path and propagation intensity. The preset congestion prediction model is obtained by training a large amount of historical sample travel user behavior data, road network topology data, spatial environment data, the congestion contribution value, real-time traffic flow characteristic data, event type characteristic data, and the corresponding congestion propagation prediction data.

[0047] According to an embodiment of the present invention, it further includes: The initial traffic control strategy is obtained by querying the preset traffic control strategy library based on the congestion propagation prediction data. Based on the traffic strategy parameter data corresponding to the initial traffic control strategy, as well as the real-time traffic flow characteristic data and regional travel profile, the data are processed through a preset road network operation simulation model to obtain road network operation status prediction data, including congestion relief rate and vehicle delay reduction rate. The congestion mitigation rate and the vehicle delay reduction rate are weighted and summed to obtain the traffic strategy effectiveness index. The traffic strategy effectiveness indices are sorted in descending order, and the initial traffic control strategy with the largest traffic strategy effectiveness index is determined as the traffic control strategy.

[0048] It should be noted that the congestion propagation prediction data obtained through the evaluation is based on a pre-set traffic control strategy library constructed by those skilled in the art from a large number of historical sample cases. The corresponding traffic strategy parameter data, real-time traffic flow characteristic data, and regional travel profiles are input into a pre-set road network operation simulation model for processing to obtain road network operation status prediction data. Then, through weighted summation, the traffic strategy effectiveness index is determined. Finally, the initial traffic control strategy with the largest traffic strategy effectiveness index is determined as the traffic control strategy, thereby determining the best matching strategy. The pre-set road network operation simulation model is obtained by training with a large number of historical sample traffic strategy parameter data, real-time traffic flow characteristic data, regional travel profiles, and corresponding road network operation status prediction data.

[0049] This invention discloses a method and system for analyzing travel profiles and congestion correlations for regional transportation. By accurately integrating cross-domain multi-source traffic data, constructing accurate regional travel profiles, and combining congestion causal correlation analysis to evaluate congestion propagation prediction data, it achieves intelligent analysis of travel profiles and congestion correlations for regional transportation.

[0050] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0051] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0052] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0053] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0054] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for analyzing the association between travel profiles and congestion for regional traffic, characterized in that, The method comprises the following steps: obtaining a cross-domain multi-source traffic data set and performing data preprocessing to obtain a cross-domain multi-source traffic fusion data set; processing the cross-domain multi-source traffic fusion data set to obtain a regional travel profile; performing congestion causal correlation analysis on the cross-domain multi-source traffic fusion data set and the regional travel profile to obtain a causal correlation strength matrix; analyzing and processing the regional travel profile and the causal correlation strength matrix to obtain congestion propagation prediction data.

2. The method of claim 1, wherein, The cross-domain multi-source traffic data set is obtained, and data preprocessing is performed to obtain a cross-domain multi-source traffic fusion data set, comprising: obtaining a cross-domain multi-source traffic data set, including traffic monitoring record data, traffic sensor record data and traffic survey record data; performing data cleaning, timestamp alignment and data standardization preprocessing on the traffic monitoring record data, traffic sensor record data and traffic survey record data to obtain traffic monitoring standard record data, traffic sensor standard record data and traffic survey standard record data; performing data extraction processing on the traffic monitoring standard record data through a preset CLIP model to obtain vehicle unstructured feature data; performing time alignment processing on the traffic sensor standard record data and the traffic survey standard record data through a preset dynamic time warping (DTW) algorithm to obtain traffic sensor time series data; performing data fusion processing on the vehicle unstructured feature data and the traffic sensor time series data to obtain cross-domain multi-source traffic initial fusion data; performing differential privacy noise addition and bias correction processing on the cross-domain multi-source traffic initial fusion data to obtain a cross-domain multi-source traffic fusion data set.

3. The zone-oriented traffic trip profiling and congestion correlation analysis method of claim 2, wherein, The cross-domain multi-source traffic initial fusion data is processed to obtain a cross-domain multi-source traffic fusion data set, comprising: obtaining a sensitivity value of the cross-domain multi-source traffic initial fusion data; allocating a privacy budget according to the sensitivity value to obtain a privacy budget value; performing noise addition processing on the sensitivity value and the privacy budget value through a preset noise addition method to obtain cross-domain multi-source traffic noisy fusion data; performing bias analysis processing on the cross-domain multi-source traffic noisy fusion data through a preset traffic conservation constraint rule to obtain a data bias value; comparing the data bias value with a preset data bias threshold value; if the data bias value is less than or equal to the preset data bias threshold value, the cross-domain multi-source traffic noisy fusion data is determined to be a cross-domain multi-source traffic fusion data set; if the data bias value is greater than the preset data bias threshold value, the noise is reduced according to a preset gradient, and bias correction processing is performed; if the bias correction is qualified, the cross-domain multi-source traffic fusion data set is obtained, otherwise, an early warning response is output.

4. The method of claim 2, wherein, The cross-domain multi-source traffic fusion data set is processed to obtain a regional travel profile, comprising: performing data extraction on the cross-domain multi-source traffic fusion data set to obtain regional feature data, including cross-city travel frequency, travel distance type feature data and travel traffic combination type feature data; The cross-city travel frequency, travel distance type feature data, and travel transportation combination type feature data are input into a preset travel group category analysis model for processing to obtain travel group category feature data. Acquire intercity ETC record data, city checkpoint record data, mobile phone signaling record data, POI data, and real-time weather record data; Based on the intercity ETC record data, city checkpoint record data, mobile phone signaling record data, POI data, real-time weather record data, cross-city travel frequency, travel distance type characteristic data, travel traffic combination type characteristic data, and travel group category characteristic data, a regional travel profile is obtained by processing the data through a preset clustering analysis algorithm.

5. The zone-oriented traffic trip profiling and congestion correlation analysis method of claim 4, wherein, The step involves performing a causal correlation analysis on congestion based on the cross-domain multi-source traffic fusion dataset and regional travel profiles to obtain a causal correlation strength matrix, including: By performing correlation analysis between the cross-domain multi-source traffic fusion dataset and the regional travel profile, data on the size of the travel group and the behavioral characteristics of the travel group are obtained. Acquire road segment operation status monitoring data and time period scene characteristic data; Based on the travel group size data and travel group behavior characteristic data, combined with the road segment operation status monitoring data and time period scene characteristic data, the data are analyzed and processed through a preset Bayesian network model to obtain a causal correlation strength matrix.

6. The zone-oriented traffic trip profiling and congestion correlation analysis method of claim 5, wherein, The step of analyzing and processing the regional travel profile and causal correlation strength matrix to obtain congestion propagation prediction data includes: Data is extracted based on the causal correlation strength matrix within a preset time period to obtain travel user behavior data, road network topology data, and spatial environment data. Based on the regional travel profile, extract the congestion contribution value corresponding to the travel group category feature data; Acquire real-time traffic flow characteristic data and event type characteristic data; The user travel behavior data, road network topology data, spatial environment data, as well as the congestion contribution value, real-time traffic flow characteristic data, and event type characteristic data are input into a preset congestion prediction model for processing to obtain congestion propagation prediction data.

7. The zone-oriented traffic trip profiling and congestion correlation analysis method of claim 6, wherein, Also includes: The initial traffic control strategy is obtained by querying the preset traffic control strategy library based on the congestion propagation prediction data. Based on the traffic strategy parameter data corresponding to the initial traffic control strategy, as well as the real-time traffic flow characteristic data and regional travel profile, the data are processed through a preset road network operation simulation model to obtain road network operation status prediction data, including congestion relief rate and vehicle delay reduction rate. The congestion mitigation rate and the vehicle delay reduction rate are weighted and summed to obtain the traffic strategy effectiveness index. The traffic strategy effectiveness indices are sorted in descending order, and the initial traffic control strategy with the largest traffic strategy effectiveness index is determined as the traffic control strategy.

8. A regional traffic-oriented travel profile and congestion correlation analysis system, characterized by, The system includes a memory and a processor. The memory contains a program for a method of analyzing the correlation between travel profiles and congestion in regional traffic. When the processor executes the program for analyzing the correlation between travel profiles and congestion in regional traffic, it performs the following steps: Obtain cross-domain multi-source traffic datasets and perform data preprocessing to obtain cross-domain multi-source traffic fusion datasets; The regional travel profile is obtained by processing the cross-domain multi-source traffic fusion dataset. Based on the cross-domain multi-source traffic fusion dataset and regional travel profile, a causal correlation analysis of congestion is performed to obtain a causal correlation strength matrix. Based on the regional travel profile and causal correlation strength matrix, congestion propagation prediction data are obtained through analysis and processing.

9. The regional traffic oriented travel profile and congestion correlation analysis system of claim 8, wherein, The process of acquiring a cross-domain multi-source traffic dataset and performing data preprocessing to obtain a cross-domain multi-source traffic fusion dataset includes: Acquire cross-domain, multi-source traffic datasets, including traffic monitoring records, traffic sensor records, and traffic survey records; The traffic monitoring record data, traffic sensor record data, and traffic survey record data are preprocessed by data cleaning, timestamp alignment, and data standardization to obtain standard traffic monitoring record data, standard traffic sensor record data, and standard traffic survey record data. According to the traffic monitoring standard, the recorded data is processed by a preset CLIP model to obtain unstructured vehicle feature data. Based on the standard traffic sensor data and the standard traffic survey data, time alignment processing is performed using a preset Dynamic Time Warping (DTW) algorithm to obtain traffic sensor time series data. The unstructured feature data of the vehicles and the time series data of traffic sensors are fused together to obtain cross-domain multi-source traffic initial fusion data. The initial cross-domain multi-source traffic fusion data is processed with differential privacy noise addition and bias correction to obtain a cross-domain multi-source traffic fusion dataset.

10. The regional traffic oriented travel profile and congestion correlation analysis system of claim 9, wherein, The process of performing differential privacy noise addition and bias correction on the initial fused cross-domain multi-source traffic data to obtain a cross-domain multi-source traffic fusion dataset includes: Obtain the sensitivity value of the initial fusion data of the cross-domain multi-source traffic; A privacy budget is allocated based on the sensitivity value to obtain the privacy budget value; Based on the sensitivity value and privacy budget value, noise is added using a preset noise addition method to obtain cross-domain multi-source traffic noisy fusion data; The cross-domain multi-source traffic noise-containing fusion data is processed by deviation analysis using preset traffic conservation constraint rules to obtain data deviation values. The data deviation value is compared with a preset data deviation threshold. If the data deviation value is less than or equal to the preset data deviation threshold, then the cross-domain multi-source traffic noisy fusion data is determined to be a cross-domain multi-source traffic fusion dataset. If the data deviation value is greater than the preset data deviation threshold, then noise reduction processing is performed according to the preset gradient, and deviation correction processing is performed. If the deviation correction is satisfactory, a cross-domain multi-source traffic fusion dataset is obtained; otherwise, an early warning response is output.

Images