Expressway data management method based on data center

Abstract

The invention discloses an expressway data management method based on a data center, and belongs to the technical field of monitoring data processing, and the method specifically comprises the steps: forming an expressway infrastructure list directory through multi-dimensional classification, and achieving the dynamic updating through the Internet of Things technology; coding and preprocessing the infrastructure attribute data, constructing a digitalized expression attribute data set, and storing the digitalized expression attribute data set in a data table; performing feature extraction and rule analysis on a vehicle driving route by using a space-time analysis strategy, and performing short-term and long-term prediction on the driving route by using a prediction model in combination with real-time traffic data; and an optimal traffic planning scheme is automatically generated based on a prediction result, and virtual verification and optimization are performed through a simulation technology, so that the highway traffic management efficiency and the intelligent level are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of monitoring data processing, and specifically is a highway data management method based on a data middle station. Background Art

[0002] With the rapid development of highways, the amount of data generated on highways is becoming increasingly large, including toll data, monitoring data, vehicle data, etc. The traditional data management method has problems such as data dispersion, low processing efficiency, and difficulty in forming an effective data infrastructure. Therefore, an efficient data management method is needed to achieve comprehensive management and efficient use of highway data.

[0003] For example, a Chinese patent with the authorization announcement number CN114566058B discloses a method and system for managing highway operation and maintenance monitoring data, including: receiving monitoring data uploaded by monitoring equipment; automatically screening and processing the monitoring data to obtain normal monitoring data, abnormal monitoring data, congestion monitoring data and blank monitoring data; deleting the blank monitoring data; storing the normal monitoring data, abnormal monitoring data and congestion monitoring data in the normal database, abnormal database and congestion database respectively. The blank monitoring data in this technical solution has basically no reference value and will not be called for evidence later. The blank monitoring data is directly deleted, which reduces the space occupied by useless data, and the normal monitoring data, abnormal monitoring data and congestion monitoring data are classified and stored. It is quick and convenient to retrieve the monitoring data as evidence later, which improves the management efficiency of highway operation and maintenance.

[0004] For example, a Chinese patent with the authorization announcement number CN112948639B discloses a unified storage management method and system for highway data middle platform data, including: constructing a description code sequence according to highway data, determining the first cutoff condition, the second cutoff condition, the step length, and the number of compressions according to the importance of the description code sequence; setting a first sliding window and a second sliding window to slide on the description code sequence, and the two windows are adjacent and non-overlapping; according to the determined conditions, the number of compressions of the description code sequence reaches the conditions, and the obtained description code compression and encryption sequence is stored in the unified storage unit of the highway data middle platform. This technical solution realizes the unified storage of highway data, improves the transmission efficiency of highway data, the space utilization rate of the storage system, and the privacy protection performance of data.

[0005] The above existing technologies all have the following problems: although the monitoring data is classified and stored, it does not involve in-depth analysis or prediction of the data. This means that it is impossible to proactively discover potential traffic problems or bottlenecks; it mainly focuses on data processing and storage, and does not involve applications in traffic planning and management; it lacks real-time and dynamic performance. Summary of the invention

[0006] In view of the shortcomings of the existing technology, the present invention proposes a highway data management method based on a data middle platform, which forms a list of highway infrastructure through multi-dimensional classification, and realizes dynamic update by using the Internet of Things technology; encodes and preprocesses the infrastructure attribute data, constructs a digital expression attribute data set, and stores it in the data middle platform; uses spatiotemporal analysis strategies to extract features and analyze the rules of vehicle routes, combines real-time traffic data, and uses prediction models to make short-term and long-term predictions of routes; automatically generates the optimal traffic planning scheme based on the prediction results, and performs virtual verification and optimization through simulation technology, thereby improving the efficiency and intelligence level of highway traffic management.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A highway data management method based on a data middle platform, comprising:

[0009] Step S1: Using IoT sensors to intelligently collect highway infrastructure data to form an initial facility list, multi-dimensionally classify highway infrastructure according to different application requirements and facility value manifestation forms, form a highway infrastructure list directory, and dynamically update the highway infrastructure list directory through the IoT;

[0010] Step S2: according to the different types of characteristics of highway infrastructure, each attribute data of each type of highway infrastructure is encoded to form a digital expression attribute data set of highway infrastructure, and at the same time, the digital expression attribute data of highway infrastructure is preprocessed, and the preprocessed digital expression attribute data of highway infrastructure is stored in the data warehouse of the data middle station;

[0011] Step S3: Use the spatiotemporal analysis strategy to analyze the vehicle driving routes of the pre-processed highway infrastructure digital expression attribute data stored in the data warehouse, obtain the characteristics and rules of the driving routes, combine the real-time traffic data flow, and use the prediction model to make short-term and long-term predictions of the driving routes;

[0012] Step S4: Based on the characteristics and regularity analysis of the driving route, combined with the short-term and long-term prediction results, the traffic planning scheme is automatically generated and recommended, and simulation technology is introduced to virtually verify the recommended traffic planning scheme. According to the verification results, the optimal traffic planning scheme is optimized, and at the same time, the optimized traffic planning scheme is deployed and implemented.

[0013] Specifically, the specific steps of encoding in step S2 include:

[0014] S2.1: Obtain a list of highway infrastructure, obtain infrastructure classification results based on the list of highway infrastructure using IoT sensors, and identify the attributes of each type of infrastructure;

[0015] S2.2: Based on the infrastructure classification and attribute identification results, use the relative proportion method to calculate the relative proportion of each attribute in all attributes. And according to the calculated attribute ratio P i , assign weights W to the attributes under each category i =P i / 100, based on the infrastructure classification and attribute weight allocation results, dynamic coding rules are formulated, where P i represents the proportion of the i-th attribute, s i Indicates the importance of the i-th attribute, L i represents the measurability of the i-th attribute, B i represents the variability of the i-th attribute, R i represents the relevance of the i-th attribute, W i Represents the weight value of the i-th attribute;

[0016] S2.3: According to the dynamic coding rules, use the coding algorithm to encode each attribute data of each type of highway infrastructure code = title + location + b1 + b2 + ... + check. If there is data that does not conform to the coding rules, it is corrected by the automatic correction algorithm or marked as abnormal data, where code represents the coding result, title represents the prefix, location represents the location identifier, b1 and b2 represent the attribute identifiers, and check represents the check code;

[0017] S2.4: Integrate the encoded attribute data into a data set to form a digital expression attribute data set of highway infrastructure, and store it in the data warehouse of the data middle platform.

[0018] Specifically, the specific steps of the automatic correction algorithm in S2.3 include:

[0019] S2.31: Obtain infrastructure classification results and dynamic coding rules, and filter out infrastructure classification results that do not comply with dynamic coding rules through string matching, format verification, and data type checking operations;

[0020] S2.32: Design a correction strategy based on the type and degree of non-compliance with the rules, the correction strategy including filling in default values ​​for missing data, reformatting for erroneous data, and truncating or rounding off for out-of-range data;

[0021] S2.33: Apply correction strategies to correct data that does not conform to the rules, including string replacement, numerical calculation, and type conversion;

[0022] S2.34: After correction, check again whether the infrastructure classification results comply with the coding rules. If they still do not comply, mark them as abnormal data or perform manual intervention.

[0023] Specifically, the specific steps of step S3 include:

[0024] S3.1: Extracting the digitally expressed attribute data of highway infrastructure from the data warehouse of the data center, and collecting real-time traffic data streams, wherein the extracted digitally expressed attribute data of highway infrastructure includes road network data and vehicle driving records, and the real-time traffic data streams include vehicle position, speed, and direction;

[0025] S3.2: Use the Euclidean distance calculation method to map the location information in the vehicle driving record to the road network data, and connect the points in the vehicle driving record into continuous trajectories according to the timestamp and location information to obtain trajectory data, wherein each trajectory contains the location information of the vehicle in the time series;

[0026] S3.3: Use trajectory clustering method to analyze the driving route of trajectory data, identify the driving route, and extract its characteristics and rules;

[0027] S3.4: Load the pre-built time series analysis prediction model, and train and verify the pre-built time series analysis prediction model using real-time traffic data stream data according to the characteristics and rules of the driving route;

[0028] S3.5: Use the trained time series analysis prediction model to make short-term and long-term predictions for the driving route, and evaluate the time series analysis prediction model based on the prediction results. At the same time, draw a driving volume trend chart and a driving route map based on the prediction results.

[0029] Specifically, the specific steps of S3.3 include:

[0030] S3.31: Obtain trajectory data and perform preprocessing;

[0031] S3.32: Set the distance threshold h and the time period Δt, and initialize a variable to record the starting point of the current segment;

[0032] S3.33: Traverse the continuous trajectory data in time order, and for each trajectory point, calculate the Euclidean distance d between it and the end point of the current segment within the time period Δt;

[0033] If d≤h, add the trajectory point to the current segment;

[0034] If d>h, the current segment is ended and the trajectory point is used as the starting point of the new segment. At the same time, each segment of the trajectory is stored in a pre-created array to obtain the segmented trajectory data;

[0035] S3.34: Use a clustering strategy to cluster the segmented trajectories, and analyze the clustered driving routes to extract their characteristics and rules.

[0036] Specifically, the specific steps of the clustering strategy in S3.34 include:

[0037] A1: Get the segmented trajectory data X = {x1,…,x n}, set the parameter ε and the minimum number of points q, and initialize an empty set C, where x j ,j∈[1,n] represents the jth segment of trajectory data, n represents the number of segments of trajectory data;

[0038] A2: For each object x j , check object x j Has it been visited?

[0039] If the object x j If it has not been visited, it is marked as visited and a new cluster is created.

[0040] A3: Find object x j All objects in the ε-neighborhood of ;

[0041] If N ε ≥q, then x j Marked as a core object, where N ε Represents object x j The number of objects in the ε-neighborhood of ;

[0042] And for object x j For each object in the ε-neighborhood of x, if j If it has not been assigned to any cluster, add it to the cluster middle;

[0043] A4: Traverse the clusters. For each object in the cluster, find the objects in its ε-neighborhood and add the objects that meet the conditions to the cluster until no new objects can be added to the cluster until;

[0044] A5: Cluster Add it to the clustering result set C, repeat iteratively until all objects in X have been visited, and output the clustering result set C.

[0045] Specifically, the specific steps of step S4 include:

[0046] S4.1: Obtain the characteristics and regularity analysis results of the driving route, and the short-term and long-term prediction results of the driving route;

[0047] S4.2: Based on the short-term and long-term prediction results of the driving route, identify potential traffic bottlenecks and congestion points, and formulate a traffic planning scheme, which includes intersection control measures and diversion strategies;

[0048] S4.3: Use MATSim traffic simulation software to virtually verify the generated traffic planning scheme, and optimize and adjust the traffic planning scheme based on the simulation verification results;

[0049] S4.4: Recommend the best traffic planning scheme to the traffic management department.

[0050] Specifically, the features extracted in S3.34 include the starting point, the end point, and the key areas of the driving route, and the extracted rules include the time pattern and frequency of driving.

[0051] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of a highway data management method based on a data middle station when executing the computer program.

[0052] A computer-readable storage medium stores computer instructions, which, when executed, execute the steps of a highway data management method based on a data middle station.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] 1. The present invention proposes a highway data management method based on a data middle platform, which intelligently collects highway infrastructure data through Internet of Things sensors, and performs multi-dimensional classification and dynamic update of the infrastructure, thereby realizing digital management and real-time monitoring of highway infrastructure and improving the efficiency and accuracy of infrastructure management.

[0055] 2. The present invention proposes a highway data management method based on a data middle platform, which uses spatiotemporal analysis strategies and prediction models to make short-term and long-term predictions on driving routes, combines the characteristics and regularity analysis of driving routes, automatically generates and recommends traffic planning schemes, and performs virtual verification and optimization through simulation technology. This process not only improves the pertinence and scientific nature of traffic planning, but also effectively alleviates traffic congestion and improves road traffic efficiency. At the same time, the introduction of simulation technology also makes the verification and optimization of traffic planning schemes more convenient and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A schematic diagram of a highway data management method based on a data middle station according to the present invention;

[0057] Figure 2 This is a principle flow chart of a highway data management method based on a data middle station of the present invention;

[0058] Figure 3 A schematic diagram of a vehicle driving route in a highway data management method based on a data middle station of the present invention;

[0059] Figure 4 This is a flow chart of the vehicle driving route prediction process of a highway data management method based on a data middle station of the present invention;

[0060] Figure 5 This is an architecture diagram of a highway data management system based on a data middle platform in the present invention. DETAILED DESCRIPTION

[0061] Example 1

[0062] See also Figure 1 and Figure 2 , an embodiment provided by the present invention: a highway data management method based on a data middle station, comprising the following steps:

[0063] Step S1: Using IoT sensors to intelligently collect highway infrastructure data to form an initial facility list, multi-dimensionally classify highway infrastructure according to different application requirements and facility value manifestation forms, form a highway infrastructure list directory, and dynamically update the highway infrastructure list directory through the IoT;

[0064] Furthermore, the specific steps of step S1 include:

[0065] (1) Deploy IoT sensors, such as temperature sensors, humidity sensors, pressure sensors, and RFID tag readers, along highways and at key facilities to ensure that the sensors can accurately and in real time collect highway infrastructure data;

[0066] (2) Using the Internet of Things technology, the highway infrastructure data collected by sensors is transmitted to the data center. Using data analysis algorithms and intelligent recognition technology, the collected highway infrastructure data is processed and analyzed to form an initial facility list;

[0067] Among them, highway infrastructure data includes but is not limited to road structure data, bridge and tunnel data, traffic facility data, and monitoring equipment data.

[0068] (3) Investigate the facilities and equipment along the expressway, clarify their application needs and the form in which the facilities’ value is reflected, and conduct a preliminary assessment of the facilities and equipment to determine key indicators such as their importance, frequency of use, and maintenance costs;

[0069] (4) Construct a multi-dimensional classification system based on the type, function, location and other characteristics of facilities and equipment, set classification standards, and ensure the accuracy and consistency of classification;

[0070] Among them, the classification standards include infrastructure types, such as road facilities, bridge facilities, tunnel facilities, etc.; functional uses, such as access services, safety services, maintenance services, etc.; geographical locations, such as road sections, regions, etc.

[0071] At the same time, the classification standards should clarify the specific division methods and basis for each classification dimension to ensure the accuracy and consistency of the classification. For example, in terms of infrastructure type, it can be classified according to the components of the highway, such as roads, bridges, and tunnels; in terms of functional use, it can be classified according to the role of the infrastructure in highway operations, such as access, service, and safety.

[0072] (5) Based on the multi-dimensional classification system and in accordance with the classification standards, a list of highway infrastructure should be compiled. The list should include detailed information such as the name, model, quantity, location, and value of the facilities and equipment, and should clearly display the infrastructure categories and specific contents under each classification dimension;

[0073] (6) Deploy IoT sensors, RFID tags and other equipment at key facilities and equipment along the highway, establish an IoT data transmission network, and ensure that data can be transmitted to the data center in real time and accurately;

[0074] (7) Design a dynamic update mechanism for the infrastructure inventory directory, including data update frequency and update method;

[0075] (8) Set trigger conditions, such as changes in facility and equipment status, updates to maintenance records, etc., to automatically trigger the update process.

[0076] Step S2: according to the different types of characteristics of highway infrastructure, each attribute data of each type of highway infrastructure is encoded to form a digital expression attribute data set of highway infrastructure, and at the same time, the digital expression attribute data of highway infrastructure is preprocessed, and the preprocessed digital expression attribute data of highway infrastructure is stored in the data warehouse of the data middle station;

[0077] Step S3: Use the spatiotemporal analysis strategy to analyze the vehicle driving routes of the pre-processed highway infrastructure digital expression attribute data stored in the data warehouse, obtain the characteristics and rules of the driving routes, combine the real-time traffic data flow, and use the prediction model to make short-term and long-term predictions of the driving routes;

[0078] Step S4: Based on the characteristics and regularity analysis of the driving route, combined with the short-term and long-term prediction results, the traffic planning scheme is automatically generated and recommended, and simulation technology is introduced to virtually verify the recommended traffic planning scheme. According to the verification results, the optimal traffic planning scheme is optimized, and at the same time, the optimized traffic planning scheme is deployed and implemented.

[0079] To sum up, it needs to be understood that the entire process managed in a highway data management method based on a data middle platform includes the acquisition, encoding, judgment, analysis, optimization and implementation of highway data.

[0080] The specific steps of encoding in step S2 include:

[0081] S2.1: Obtain a list of highway infrastructure, and based on the list of highway infrastructure, use IoT sensors to obtain infrastructure classification results and identify the attributes of each type of infrastructure, such as size, material, structure, service life, maintenance record, and economic value;

[0082] S2.2: Based on the infrastructure classification and attribute identification results, use the relative proportion method to calculate the relative proportion of each attribute in all attributes. And according to the calculated attribute ratio P i , assign weights W to the attributes under each category i =P i / 100, based on the infrastructure classification and attribute weight allocation results, dynamic coding rules are formulated, where P i represents the proportion of the i-th attribute, s i Indicates the importance of the i-th attribute, L i represents the measurability of the i-th attribute, B i represents the variability of the i-th attribute, R i represents the relevance of the i-th attribute, W i Represents the weight value of the i-th attribute;

[0083] S2.3: According to the dynamic coding rules, use the coding algorithm to encode each attribute data of each type of highway infrastructure code = title + location + b1 + b2 + ... + check. If there is data that does not conform to the coding rules, it is corrected by the automatic correction algorithm or marked as abnormal data, where code represents the coding result, title represents the prefix, location represents the location identifier, b1 and b2 represent the attribute identifiers, and check represents the check code;

[0084] S2.4: Integrate the encoded attribute data into a data set to form a digital expression attribute data set of highway infrastructure, and store it in the data warehouse of the data middle platform.

[0085] The specific steps of the automatic correction algorithm in S2.3 include:

[0086] S2.31: Obtain infrastructure classification results and dynamic coding rules, and filter out infrastructure classification results that do not comply with dynamic coding rules through string matching, format verification, and data type checking operations;

[0087] S2.32: Design a correction strategy based on the type and degree of non-compliance with the rules, the correction strategy including filling in default values ​​for missing data, reformatting for erroneous data, and truncating or rounding off for out-of-range data;

[0088] S2.33: Apply correction strategies to correct data that does not conform to the rules, including string replacement, numerical calculation, and type conversion;

[0089] S2.34: After correction, check again whether the infrastructure classification results comply with the coding rules. If they still do not comply, mark them as abnormal data or perform manual intervention.

[0090] Example 2

[0091] See also Figure 3-Figure 4 In this embodiment, the specific steps of step S3 include:

[0092] S3.1: Extracting the digitally expressed attribute data of highway infrastructure from the data warehouse of the data center, and collecting real-time traffic data streams, wherein the extracted digitally expressed attribute data of highway infrastructure includes road network data and vehicle driving records, and the real-time traffic data streams include vehicle position, speed, and direction;

[0093] S3.2: Use the Euclidean distance calculation method to map the location information in the vehicle driving record to the road network data, and connect the points in the vehicle driving record into continuous trajectories according to the timestamp and location information to obtain trajectory data, wherein each trajectory contains the location information of the vehicle in the time series;

[0094] Furthermore, the specific steps of mapping the location information in the vehicle driving record to the road network data using the Euclidean distance calculation method include:

[0095] (1) Obtain road network data and vehicle driving records;

[0096] (2) using the Euclidean distance calculation method to calculate the distance between each position point in the vehicle's driving trajectory and each point on the road network data, wherein the Euclidean distance calculation method is a prior art in the art and is not an inventive solution of the present application, and is not described in detail herein;

[0097] (3) mapping each position point in the vehicle's driving trajectory to the closest road network data point based on the distance calculation result, wherein the distance between each position point and each point on the road network data is compared, and the point with the smallest distance is selected as the mapping result;

[0098] (4) Post-process the mapping results, such as smoothing and removing redundant points, to improve the accuracy and reliability of the mapping results.

[0099] S3.3: Use trajectory clustering method to analyze the driving route of trajectory data, identify the driving route, and extract its characteristics and rules;

[0100] S3.4: Load a pre-built time series analysis prediction model, and train and verify the pre-built time series analysis prediction model using real-time traffic data flow data according to the characteristics and rules of the driving route, wherein the time series analysis prediction model is the prior art content in this field, is not the inventive solution of the present application, and is not described in detail here;

[0101] S3.5: Use the trained time series analysis prediction model to make short-term and long-term predictions for the driving route, and evaluate the time series analysis prediction model based on the prediction results. At the same time, draw a driving volume trend chart and a driving route map based on the prediction results.

[0102] Among them, in the present invention, the short-term prediction is set to 1 week and the long-term prediction is set to 2 years.

[0103] Furthermore, the specific steps of S3.5 include:

[0104] (1) Collect driving history data, including driving volume, time, starting point, and end point, and pre-process the data, such as cleaning, deduplication, and normalization;

[0105] (2) Load the pre-trained time series analysis and prediction model;

[0106] (3) Use the trained model to make short-term and long-term predictions of driving routes and output the prediction results, including driving volume, driving time, and driving routes;

[0107] (4) Based on the comparison between the prediction results and the actual driving data, the prediction accuracy of the model is evaluated using the mean square error, wherein the mean square error is a prior art in the art and is not an inventive solution of the present application, and is not described in detail here;

[0108] (5) Use the visualization tool Matplotlib to draw a travel volume trend graph to show the change trend of travel volume over time, wherein Matplotlib graphics drawing is a prior art content in this field and is not an inventive solution of the present application, and is not described in detail here;

[0109] (6) Draw a driving route map based on the prediction results, showing the starting point, end point and main path of the trip.

[0110] The specific steps of S3.3 include:

[0111] S3.31: Obtain trajectory data and perform preprocessing;

[0112] S3.32: Set the distance threshold h and the time period Δt, and initialize a variable to record the starting point of the current segment;

[0113] S3.33: Traverse the continuous trajectory data in time order, and for each trajectory point, calculate the Euclidean distance d between it and the end point of the current segment within the time period Δt;

[0114] If d≤h, add the trajectory point to the current segment;

[0115] If d>h, the current segment is ended and the trajectory point is used as the starting point of the new segment. At the same time, each segment of the trajectory is stored in a pre-created array to obtain the segmented trajectory data;

[0116] S3.34: Use a clustering strategy to cluster the segmented trajectories, and analyze the clustered driving routes to extract their characteristics and rules.

[0117] The features extracted in S3.34 include the starting point, end point, and key areas of the driving route, and the extracted rules include the time pattern and frequency of driving.

[0118] The specific steps of the clustering strategy in S3.34 include:

[0119] A1: Get the segmented trajectory data X = {x1,…,x n}, set the parameter ε and the minimum number of points q, and initialize an empty set C, where x j ,j∈[1,n] represents the jth segment of trajectory data, n represents the number of segments of trajectory data;

[0120] A2: For each object x j , check object x j Has it been visited?

[0121] If the object x j If it has not been visited, it is marked as visited and a new cluster is created.

[0122] A3: Find object x j All objects in the ε-neighborhood of ;

[0123] If N ε ≥q, then x j Marked as a core object, where N ε Represents object x j The number of objects in the ε-neighborhood of ;

[0124] And for object x j For each object in the ε-neighborhood of x, if j If it has not been assigned to any cluster, add it to the cluster middle;

[0125] A4: Traverse the clusters. For each object in the cluster, find the objects in its ε-neighborhood and add the objects that meet the conditions to the cluster until no new objects can be added to the cluster until;

[0126] A5: Cluster Add it to the clustering result set C, repeat iteratively until all objects in X have been visited, and output the clustering result set C.

[0127] The specific steps of step S4 include:

[0128] S4.1: Obtain the characteristics and regularity analysis results of the driving route, and the short-term and long-term prediction results of the driving route;

[0129] S4.2: Based on the short-term and long-term prediction results of the driving route, identify potential traffic bottlenecks and congestion points, such as key intersections and highway entrances, and formulate a traffic planning scheme, which includes intersection control measures and diversion strategies. The intersection control measures include signal light control and lane adjustment, and the diversion strategies include setting up temporary lanes and guiding vehicles to detour;

[0130] Among them, when identifying potential traffic bottlenecks and congestion points, we mainly focus on outliers in the prediction results, such as sudden increase in traffic flow, reduced vehicle speed, etc., which may be potential traffic bottlenecks and congestion points. Then, combined with the characteristics and rules of the driving route and real-time traffic data, potential traffic bottlenecks and congestion points can be determined. Among them, we focus on key intersections, highway entrances, transportation hubs and other areas on the driving route, which are more prone to traffic congestion.

[0131] S4.3: Use MATSim traffic simulation software to virtually verify the generated traffic planning scheme, and optimize and adjust the traffic planning scheme according to the simulation verification results, including adjusting the signal light cycle and optimizing the lane layout. MATSim traffic simulation is the prior art content in this field and is not the inventive solution of this application, and will not be elaborated here;

[0132] S4.4: Recommend the best traffic planning scheme to the traffic management department.

[0133] Example 3

[0134] See also Figure 5 , a highway data management system based on a data middle platform, comprising:

[0135] Infrastructure inventory management module, data processing module, prediction module, and solution generation module;

[0136] The infrastructure inventory management module is used to classify the facilities and equipment along the expressway in multiple dimensions, and to compile and dynamically update the expressway infrastructure inventory;

[0137] The data processing module is used to encode different types of characteristics of highway infrastructure, form digital expression attribute data sets, and perform preprocessing;

[0138] A forecasting module for analyzing vehicle routes on highway infrastructure data using spatiotemporal analysis strategies and making short-term and long-term forecasts;

[0139] The solution generation module is used to automatically generate and recommend the optimal traffic planning solution based on the characteristics and regularity analysis of the driving route, and perform virtual verification and optimization.

[0140] The infrastructure inventory management module includes: classification unit, inventory compilation unit, and dynamic update unit;

[0141] Classification unit, used to classify facilities and equipment along the highway in multiple dimensions according to different application requirements and forms of facility value expression;

[0142] Inventory compilation unit, used to compile the inventory catalog of highway infrastructure;

[0143] A dynamic update unit is used to update the highway infrastructure inventory directory in real time or periodically through the Internet of Things.

[0144] The data processing module includes: an encoding unit, a preprocessing unit, and a data storage unit;

[0145] A coding unit, used for coding each attribute data of each type of highway infrastructure;

[0146] A preprocessing unit is used to preprocess the digital expression attribute data of highway infrastructure, such as cleaning, formatting, conversion, etc.;

[0147] The data storage unit is used to store the preprocessed data in the data warehouse of the data center.

[0148] The prediction module includes: a driving route analysis unit, a prediction model unit, and a result output unit;

[0149] A driving route analysis unit is used to analyze the driving routes of the data stored in the data warehouse and extract features and rules;

[0150] A prediction model unit, which is used to combine real-time traffic data streams and use machine learning models to make short-term and long-term predictions of driving routes;

[0151] The result output unit is used to output the prediction results in a visual way for use by subsequent modules.

[0152] The solution generation module includes: solution generation unit, recommendation unit, simulation verification unit, and solution optimization unit;

[0153] A scheme generation unit is used to automatically generate an optimal traffic planning scheme according to the characteristics and rules of the driving route;

[0154] A recommendation unit, used to recommend the generated optimal traffic planning scheme to decision makers or users;

[0155] The simulation verification unit introduces simulation technology to virtually verify the recommended optimal traffic planning scheme;

[0156] The scheme optimization unit is used to adjust and optimize the optimal traffic planning scheme according to the verification results.

[0157] In summary, the infrastructure inventory management module, data processing module, prediction module, scheme generation module and the units they contain together constitute a complete and efficient highway data management system, which can realize the comprehensive management of highway infrastructure, data coding and processing, spatiotemporal analysis and prediction, and the generation and optimization of traffic planning schemes.

[0158] Example 4

[0159] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a highway data management method based on a data middle station are implemented. For details, please refer to the above method embodiment, which will not be repeated here.

[0160] A computer-readable storage medium stores computer instructions, which, when executed, execute the steps of a highway data management method based on a data middle station, wherein the storage medium may be a volatile or non-volatile computer-readable storage medium.

[0161] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the purpose and scope of protection of the present invention, and all of these are within the protection of the present invention.

Claims

1. A highway data management method based on a data middle station, characterized in that: include: Step S1: Using IoT sensors to intelligently collect highway infrastructure data to form an initial facility list, multi-dimensionally classify highway infrastructure according to different application requirements and facility value manifestation forms, form a highway infrastructure list directory, and dynamically update the highway infrastructure list directory through the IoT; Step S2: according to the different types of characteristics of highway infrastructure, each attribute data of each type of highway infrastructure is encoded to form a digital expression attribute data set of highway infrastructure, and at the same time, the digital expression attribute data of highway infrastructure is preprocessed, and the preprocessed digital expression attribute data of highway infrastructure is stored in the data warehouse of the data middle station; Step S3: Use the spatiotemporal analysis strategy to analyze the vehicle driving routes of the pre-processed highway infrastructure digital expression attribute data stored in the data warehouse, obtain the characteristics and rules of the driving routes, combine the real-time traffic data flow, and use the prediction model to make short-term and long-term predictions of the driving routes; Step S4: Based on the characteristics and regularity analysis of the driving route, combined with the short-term and long-term prediction results, the traffic planning scheme is automatically generated and recommended, and simulation technology is introduced to virtually verify the recommended traffic planning scheme. According to the verification results, the traffic planning scheme is optimized, and at the same time, the optimized traffic planning scheme is deployed and implemented.

2. A highway data management method based on a data middle station as claimed in claim 1, characterized in that: The specific steps of encoding in step S2 include: S2.1: Obtain a list of highway infrastructure, obtain infrastructure classification results based on the list of highway infrastructure using IoT sensors, and identify the attributes of each type of infrastructure; S2.2: Based on the infrastructure classification and attribute identification results, use the relative proportion method to calculate the relative proportion of each attribute in all attributes. And according to the calculated attribute ratio P i , assign weights W to the attributes under each category i =P i / 100, based on the infrastructure classification and attribute weight allocation results, dynamic coding rules are formulated, where P i represents the proportion of the i-th attribute, s i Indicates the importance of the i-th attribute, L i represents the measurability of the i-th attribute, B i represents the variability of the i-th attribute, R i represents the relevance of the i-th attribute, W i Represents the weight value of the i-th attribute; S2.3: According to the dynamic coding rules, use the coding algorithm to encode each attribute data of each type of highway infrastructure code = title + location + b1 + b2 + ... + check. If there is data that does not conform to the coding rules, it is corrected by the automatic correction algorithm or marked as abnormal data, where code represents the coding result, title represents the prefix, location represents the location identifier, b1 and b2 represent the attribute identifiers, and check represents the check code; S2.4: Integrate the encoded attribute data into a data set to form a digital expression attribute data set of highway infrastructure, and store it in the data warehouse of the data middle platform.

3. A highway data management method based on a data middle station as claimed in claim 2, characterized in that: The specific steps of the automatic correction algorithm in S2.3 include: S2.31: Obtain infrastructure classification results and dynamic coding rules, and filter out infrastructure classification results that do not comply with dynamic coding rules through string matching, format verification, and data type checking operations; S2.32: Design a correction strategy based on the type and degree of non-compliance with the rules, the correction strategy including filling in default values ​​for missing data, reformatting for erroneous data, and truncating or rounding off for out-of-range data; S2.33: Apply correction strategies to correct data that does not conform to the rules, including string replacement, numerical calculation, and type conversion; S2.34: After correction, check again whether the infrastructure classification results comply with the coding rules. If they still do not comply, mark them as abnormal data or perform manual intervention.

4. A highway data management method based on a data middle station as claimed in claim 3, characterized in that: The specific steps of step S3 include: S3.1: Extracting the digitally expressed attribute data of highway infrastructure from the data warehouse of the data center, and collecting real-time traffic data streams, wherein the extracted digitally expressed attribute data of highway infrastructure includes road network data and vehicle driving records, and the real-time traffic data streams include vehicle position, speed, and direction; S3.2: Use the Euclidean distance calculation method to map the location information in the vehicle driving record to the road network data, and connect the points in the vehicle driving record into continuous trajectories according to the timestamp and location information to obtain trajectory data, wherein each trajectory contains the location information of the vehicle in the time series; S3.3: Use trajectory clustering method to analyze the driving route of trajectory data, identify the driving route, and extract its characteristics and rules; S3.4: Load the pre-built time series analysis prediction model, and train and verify the pre-built time series analysis prediction model using real-time traffic data stream data according to the characteristics and rules of the driving route; S3.5: Use the trained time series analysis prediction model to make short-term and long-term predictions for the driving route, and evaluate the time series analysis prediction model based on the prediction results. At the same time, draw a driving volume trend chart and a driving route map based on the prediction results.

5. A highway data management method based on a data middle station as claimed in claim 4, characterized in that: The specific steps of S3.3 include: S3.31: Obtain trajectory data and perform preprocessing; S3.32: Set the distance threshold h and the time period Δt, and initialize a variable to record the starting point of the current segment; S3.33: Traverse the continuous trajectory data in time order, and for each trajectory point, calculate the Euclidean distance d between it and the end point of the current segment within the time period Δt; If d≤h, add the trajectory point to the current segment; If d>h, the current segment is ended and the trajectory point is used as the starting point of the new segment. At the same time, each segment of the trajectory is stored in a pre-created array to obtain the segmented trajectory data; S3.34: Use a clustering strategy to cluster the segmented trajectories, and analyze the clustered driving routes to extract their characteristics and rules.

6. A highway data management method based on a data middle station as claimed in claim 5, characterized in that: The specific steps of the clustering strategy in S3.34 include: A1: Get the segmented trajectory data X = {x1,…,x n }, set the parameter ε and the minimum number of points q, and initialize an empty set C, where x j ,j∈[1,n] represents the jth segment of trajectory data, n represents the number of segments of trajectory data; A2: For each object x j , check object x j Has it been visited? If the object x j If it has not been visited, it is marked as visited and a new cluster is created. A3: Find object x j All objects in the ε-neighborhood of ; If N ε ≥q, then x j Marked as a core object, where N ε Represents object x j The number of objects in the ε-neighborhood of ; And for object x j For each object in the ε-neighborhood of x, if j If it has not been assigned to any cluster, add it to the cluster middle; A4: Traverse the clusters. For each object in the cluster, find the objects in its ε-neighborhood and add the objects that meet the conditions to the cluster until no new objects can be added to the cluster until; A5: Cluster Add it to the clustering result set C, repeat iteratively until all objects in X have been visited, and output the clustering result set C.

7. A highway data management method based on a data middle station as claimed in claim 6, characterized in that: The specific steps of step S4 include: S4.1: Obtain the characteristics and regularity analysis results of the driving route, and the short-term and long-term prediction results of the driving route; S4.2: Based on the short-term and long-term prediction results of the driving route, identify potential traffic bottlenecks and congestion points, and formulate a traffic planning scheme, which includes intersection control measures and diversion strategies; S4.3: Use MATSim traffic simulation software to virtually verify the generated traffic planning scheme, and optimize and adjust the traffic planning scheme based on the simulation verification results; S4.4: Recommend the best traffic planning scheme to the traffic management department.

8. A highway data management method based on a data middle station as claimed in claim 7, characterized in that: The features extracted in S3.34 include the starting point, the end point, and the key areas of the driving route, and the extracted rules include the time pattern and frequency of driving.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a highway data management method based on a data middle platform as described in any one of claims 1-8 are implemented.

10. A computer-readable storage medium, characterized in that: Computer instructions are stored thereon, and when the computer instructions are executed, the steps of a highway data management method based on a data middle station as described in any one of claims 1 to 8 are executed.

Images