Macro-regional division method based on urban traffic demand

Abstract

The present invention relates to the fields of urban traffic management and intelligent transportation, and discloses a method for macro-region division based on urban traffic demand, comprising the following steps: obtaining traffic demand of traffic nodes in an urban dynamic road network, constructing a traffic demand adjacency matrix, a degree matrix, and a Laplace matrix, and extracting the aggregation degree characteristics between nodes using a spectral decomposition method; using a K-means clustering algorithm to classify node feature vectors and preliminarily divide macro-regions; introducing GPS normalized coordinate features, using a two-dimensional Gaussian distribution to model the spatial distribution of nodes, and combining spectral clustering feature distance calculation to optimize the regional division results based on a comprehensive distance; and finally, iteratively optimizing category allocation based on the comprehensive distance to obtain a macro-region division scheme with minimal segmentation loss. The present invention can improve the rationality of traffic region division, optimize the coverage of macro-regions, enhance the scientific nature of traffic management, and improve the efficiency of urban traffic resource allocation.

Description

Technical Field

[0001] The present invention relates to the field of urban traffic management and intelligent traffic technology, and in particular to a macro-region division method based on urban traffic demand. Background Art

[0002] With the acceleration of urbanization, the complexity of transportation networks continues to increase. Reasonable traffic zoning is crucial for optimizing urban traffic management and resource allocation. Traditional traffic zoning methods rely primarily on administrative divisions or empirical methods, lacking sufficient consideration of dynamic factors such as traffic flow and the strength of inter-node connections. This makes the zoning results difficult to adapt to actual traffic needs. In recent years, data-driven clustering analysis methods have been introduced into the field of traffic zoning. Methods such as K-means and hierarchical clustering have improved the rationality of zoning to a certain extent. However, these methods often calculate only based on traffic flow or geographic location, ignoring the topological structure of the traffic network. As a result, traffic connections within the same category may be weak, while traffic interactions between different categories remain significant, reducing the scientific nature and effectiveness of the zoning.

[0003] Spectral clustering, on the other hand, is widely used for network partitioning because it fully leverages graph structural information. However, in urban transportation applications, relying solely on spectral decomposition features for clustering can result in overly broad coverage of some areas, or even overlap between areas, making practical management more difficult. Furthermore, the complexity of urban road networks leads to significant imbalances in traffic demand, a factor that traditional spectral clustering methods fail to fully account for. This results in some heavily weighted nodes being aggregated into large single areas, while smaller node clusters are isolated and fragmented, compromising the overall partitioning effect.

[0004] In practical applications, the rational division of macro-regions requires not only considering the topological relationships of the transportation network but also incorporating geographic spatial characteristics to ensure that the regions are moderately sized and have clear boundaries for easy management and regulation. However, existing technologies struggle to effectively constrain the spatial extent of regions while ensuring reasonable transportation connections, making the division results less practical for practical traffic management. Summary of the Invention

[0005] In response to the shortcomings of existing technologies, the present invention provides a macro-region division method based on urban traffic demand, which solves the problem that existing traffic region division methods fail to fully combine traffic topology relationships and spatial distribution characteristics, resulting in unreasonable region division, blurred boundaries, and high management difficulty.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A macro-region division method based on urban traffic demand includes the following steps:

[0007] Obtain real-time traffic demand data for multiple traffic nodes in the urban transportation network;

[0008] Based on the traffic demand data, construct an adjacency matrix, a degree matrix and a Laplace matrix of the traffic network;

[0009] Performing spectral decomposition on the Laplace matrix to extract spectral feature vectors of traffic nodes;

[0010] Based on the spectral feature vector, the traffic nodes are preliminarily divided into multiple macro-regions using clustering algorithm;

[0011] The divided macro-regions are optimized in combination with the spatial location information of the traffic nodes to obtain the final region division result.

[0012] Preferably, the traffic node is an urban road intersection or road section, and the traffic demand data includes at least one of Internet travel platform data, traffic detection equipment data, and bus route data.

[0013] Preferably, the adjacency matrix is ​​constructed as follows:

[0014] If the traffic node v i With v j If there is a traffic demand correlation between them, then the adjacency matrix element w ij =T(i→j)+T(j→i), otherwise w ij =0;

[0015] Where T(i→j) represents the node v i to v j traffic flow.

[0016] Preferably, the Laplace matrix is ​​a normalized Laplace matrix, which is constructed as follows:

[0017] L norm =D -1 / 2 (DW)D -1 / 2

[0018] Among them, W is the adjacency matrix, D is the degree matrix, and D -1 / 2 is the inverse square root of the degree matrix.

[0019] Preferably, the step of performing spectral decomposition on the Laplace matrix to extract characteristic vectors of traffic nodes includes:

[0020] Perform eigenvalue decomposition on the normalized Laplace matrix, extract the eigenvectors corresponding to the 2k smallest eigenvalues, and construct a characteristic matrix.

[0021] Preferably, the clustering algorithm is a K-means algorithm, comprising:

[0022] Initialize cluster centers;

[0023] Calculate the Euclidean distance from each node to the cluster center;

[0024] Iteratively update the node category until the clustering is stable.

[0025] Preferably, the step of optimizing the divided macro-areas in combination with the spatial location information of the traffic nodes includes:

[0026] Initialize geospatial region division based on GPS coordinates of traffic nodes;

[0027] Calculate the mean and variance of GPS coordinates of nodes in each category and fit a two-dimensional Gaussian distribution;

[0028] Calculate the comprehensive distance between the node and each category based on the spectral features and GPS coordinate features;

[0029] Update the node category according to the comprehensive distance until the category division is stable.

[0030] Preferably, the GPS spatial probability distance is a probability density value of a two-dimensional Gaussian distribution, and the calculation formula is:

[0031]

[0032] Among them, (x, y) is the GPS coordinate of the node, (μ x ,μ y ) represents the mean of GPS coordinates of all nodes in the current category, σ x ,σ y Indicates the standard deviation of the GPS coordinates of all nodes in the current category.

[0033] Preferably, the calculation formula of the comprehensive distance is:

[0034] d final =d spectral ×P(x,y)

[0035] Where: d spectral is the Euclidean distance between spectral clustering feature vectors; P(x,y) is the GPS spatial probability distance.

[0036] The macro-region division device based on urban traffic demand of the present invention includes:

[0037] Data acquisition module, used to obtain real-time traffic demand data of traffic nodes;

[0038] Matrix building module, used to generate adjacency matrix, degree matrix and normalized Laplacian matrix;

[0039] Spectral decomposition module, used to perform eigenvalue decomposition on the normalized Laplace matrix and extract spectral feature vectors;

[0040] Clustering module, used to divide preliminary regions using the K-means algorithm;

[0041] The optimization module is used to spatially optimize the regional division results in combination with GPS coordinates.

[0042] The present invention provides a macro-region division method based on urban traffic demand, which has the following beneficial effects:

[0043] 1. The present invention extracts the structural characteristics of the traffic network through spectral decomposition and classifies the nodes in combination with K-means clustering, so that nodes within the same category have a high traffic connection, and the connection between nodes of different categories is weak, ensuring that the regional division conforms to the actual flow pattern of urban traffic demand.

[0044] 2. Traditional clustering-based division methods may result in some areas being too large or overlapping, affecting management effectiveness. This invention introduces GPS normalized coordinate features into the clustering process to ensure the spatial scope of the macro-region is reasonable, avoid cross-regional overlap, and improve management efficiency.

[0045] 3. By calculating the segmentation loss between traffic nodes and combining it with geographic spatial distribution information, the present invention achieves accurate zoning of the traffic network, provides clear regional boundaries for traffic managers, helps to formulate more scientific traffic planning and flow control strategies, and improves the operability of traffic management.

[0046] 4. The rational zoning of traffic demand facilitates more precise layout and optimization of transportation infrastructure. This invention can provide a scientific basis for public transportation route planning, signal control optimization, road resource scheduling, etc. based on traffic flow characteristics and regional distribution, thereby improving resource utilization and alleviating traffic congestion.

[0047] 5. By using a two-dimensional Gaussian distribution to fit the regional GPS spatial distribution, the present invention can intuitively display the traffic distribution in different regions, enabling managers to more clearly understand the urban traffic flow pattern and provide support for the subsequent construction of intelligent transportation systems and regional development planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 Schematic diagram of the method flow of the present invention;

[0049] Figure 2 Schematic diagram of the device structure of the present invention;

[0050] Figure 3 Schematic diagram of a method flow in an embodiment of the present invention;

[0051] Figure 4 Schematic diagram of traffic node demand distribution during the morning and evening rush hours in a city according to an embodiment of the present invention; (a) is the morning rush hour, and (b) is the evening rush hour;

[0052] Figure 5 Schematic diagram of preliminary regional division results according to an embodiment of the present invention; (a) represents the morning peak, and (b) represents the evening peak;

[0053] Figure 6 Schematic diagram of the regional division result after adding GPS normalization features in an embodiment of the present invention; (a) is the morning peak hour, and (b) is the evening peak hour.

[0054] Among them, 10, data acquisition module; 20, matrix construction module; 30, spectral decomposition module; 40, clustering module; 50, optimization module. DETAILED DESCRIPTION

[0055] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0056] Please see the attached Figure 1 The present invention provides a macro-region division method based on urban traffic demand. By combining the topological relationship, spatial distribution characteristics and spectral clustering technology of urban traffic demand, it realizes the dynamic and reasonable division of traffic areas, thereby optimizing urban traffic management and improving the efficiency of traffic scheduling.

[0057] like Figure 1 As shown, the macro-region division method based on urban traffic demand may include the following steps:

[0058] S1. Obtain real-time traffic demand data of multiple traffic nodes in the urban transportation network;

[0059] S2. Based on the traffic demand data, construct the adjacency matrix, degree matrix and Laplace matrix of the traffic network;

[0060] S3, perform spectral decomposition on the Laplace matrix to extract the characteristic vector of the traffic node;

[0061] S4, based on the feature vector, use clustering algorithm to divide the traffic node into multiple macro areas;

[0062] S5. Optimize the divided macro-area based on the spatial location information of the traffic nodes.

[0063] The following is a detailed description of each step in the method of the present invention, which comprehensively explains the specific implementation principles, technical details and processes of each step.

[0064] Step S1 primarily involves acquiring real-time traffic demand data from multiple nodes within the urban transportation network and preprocessing this data to provide reliable data support for subsequent steps. This process encompasses four key aspects: data source, data type, data preprocessing, and data storage. This ensures that traffic demand data accurately reflects current traffic conditions and provides effective input for subsequent steps.

[0065] First, traffic demand data can be obtained in the following ways:

[0066] 1. Internet travel platform data

[0067] Through common online ride-hailing and shared bicycle platforms in the city, various travel information is collected in real time. The data provided by these platforms include but are not limited to:

[0068] Departure point and destination coordinates;

[0069] The start and end times of the trip;

[0070] Trip type (e.g., ride-hailing, shared bikes, taxis, etc.);

[0071] Number of people traveling and frequency of travel.

[0072] With this information, the traffic intensity between traffic nodes can be estimated, and the traffic demand of each node can be further calculated.

[0073] 2. Traffic detection equipment data

[0074] Traditional traffic detection equipment (such as geomagnetic detectors, traffic cameras, radar speed guns, etc.) provides real-time traffic flow information, including:

[0075] Traffic volume per unit time;

[0076] traffic speeds and vehicle types;

[0077] Traffic light status and road congestion.

[0078] This information can reflect the congestion level of traffic nodes in real time and assist in traffic demand analysis.

[0079] 3. Bus route data

[0080] Through floating car technology (FCD), the trajectory data of buses moving in the city can be collected in real time. This data includes: GPS coordinates of the vehicles;

[0081] The vehicle's travel time and path;

[0082] The length of time the vehicle stayed, where it stayed, and the number of passengers.

[0083] By processing these data, we can estimate the traffic flow between different traffic nodes and supplement the deficiencies of other data sources.

[0084] Data acquisition relies on multiple channels and sensors, and data from different sources have different characteristics and accuracy. Therefore, data fusion and processing are required to ensure data integrity and consistency.

[0085] Data preprocessing includes the following aspects:

[0086] 1. Denoising

[0087] After receiving traffic data, the first step is to remove noise from the data. Common noise sources include GPS signal loss and traffic sensor failures. This data needs to be cleaned using a filtering algorithm. Preferably, a Kalman filter is used to process the trajectory data to reduce errors.

[0088] 2. Time synchronization

[0089] Since data sources may be collected at different times, data from different sources needs to be synchronized. This is done by using an interpolation algorithm to match data from different time points to ensure that all data is consistent at the same time point.

[0090] 3. Spatial matching

[0091] Because different data sources (such as GPS trajectory data and road sensor data) reflect inconsistent spatial locations, it is necessary to map all data to a unified traffic node coordinate system. This process is achieved through spatial interpolation technology, which maps traffic data from different sources to specific traffic nodes.

[0092] In terms of data storage, pre-processed data needs to be stored according to traffic nodes. The data of each traffic node includes:

[0093] Node ID;

[0094] Traffic demand intensity of the node;

[0095] Corresponding traffic flow, travel frequency and other values.

[0096] Data can be stored in a database management system (such as MySQL, PostgreSQL) for quick retrieval during subsequent calculations.

[0097] Through the above steps, accurate urban traffic demand data can be obtained and preprocessed, providing effective input for subsequent steps such as traffic network modeling, spectral decomposition, and regional division.

[0098] As for step S2, this step constructs the adjacency matrix, degree matrix and Laplace matrix of the traffic network through traffic demand data, providing basic data support for subsequent spectral decomposition and cluster analysis.

[0099] First, based on the traffic demand data, the adjacency matrix, degree matrix and Laplace matrix of the transportation network are constructed through the following process.

[0100] The adjacency matrix W in the transportation network is a matrix that represents the interconnectedness between transportation nodes, where the elements of the matrix w are ij Represents node v i With node v j The traffic demand intensity between them. The elements of the adjacency matrix can be calculated by the following formula:

[0101] w ij =T(i→j)+T(j→i)

[0102] Among them, T(i→j) represents the distance from the traffic node v i to v j Traffic flow or demand intensity, T(j→i) represents the traffic flow from traffic node v j to v i If there is no traffic flow between two nodes, then w ij =0.

[0103] The degree matrix D is a diagonal matrix with the elements D on the diagonal ii Represents node v i The degree of node v i The total traffic demand intensity connected to other nodes. The elements of the degree matrix can be calculated by the following formula:

[0104]

[0105] That is, the diagonal elements of the degree matrix are the sum of the traffic demand intensities of the node and all other nodes.

[0106] The Laplace matrix L is a commonly used matrix in graph theory and is widely used in graph cluster analysis. The Laplace matrix reflects the connection strength and topological structure between traffic nodes. In this invention, the normalized Laplace matrix L is used. norm , which is defined as:

[0107] L norm =D -1 / 2 (DW)D -1 / 2

[0108] Among them, D is the degree matrix, W is the adjacency matrix, and D -1 / 2 is the inverse square root of the degree matrix D.

[0109] The normalized Laplace matrix introduces the degree matrix D to normalize the data, so that the influence of each node is no longer directly affected by the node degree, ensuring that the connection relationship between nodes is fairly reflected in the matrix. This matrix provides the basis for spectral clustering analysis.

[0110] Through the above steps, we can construct an adjacency matrix, degree matrix, and Laplace matrix that reflect the structure and traffic demand of the urban transportation network. This can accurately reflect the connection relationships and traffic distribution between nodes in the urban transportation network, providing basic data for subsequent spectral decomposition and clustering analysis. At the same time, it ensures the normalization of node degrees during the calculation process, which helps to improve the accuracy and effectiveness of spectral clustering.

[0111] In step S3, the Laplace matrix is ​​spectrally decomposed to extract information about the solution that minimizes the partitioning loss between nodes, providing a basis for subsequent regional division. Specifically, the spectral decomposition process can divide traffic nodes according to the structural characteristics of the traffic network, so that nodes with large connection weights are grouped together, thereby minimizing the partitioning loss.

[0112] First, a spectral decomposition of the Laplacian matrix is ​​performed to extract information about the solution that minimizes the partitioning loss between nodes, providing a basis for subsequent regional division. The spectral decomposition process divides traffic nodes based on the structural characteristics of the traffic network, grouping nodes with large connection weights into the same category to minimize the partitioning loss.

[0113] Given the Laplace matrix L of the transportation network norm , which describes the connection relationship and traffic intensity between traffic nodes. The goal of the spectral decomposition process is to find a segmentation scheme that minimizes the graph segmentation loss through eigenvalue decomposition. Let the row vector c = [c1,c2,...,c n ] T Represents the category of each traffic node, where c i Indicates the category of node i, and the connection weight between node i and node j is recorded as w ij The objective function is expressed as:

[0114]

[0115] If node i and node j belong to the same category, then c i =c j, in which case the term in the loss function is zero. If nodes i and j belong to different categories, the value of this term increases. Therefore, by minimizing the split loss, this objective function classifies nodes with large connection weights that need to be separated into different categories, while keeping nodes with smaller connection weights in the same category.

[0116] To solve this optimization problem, we need to normalize the Laplace matrix L norm Perform eigenvalue decomposition and get:

[0117] L norm v i =λ i v i

[0118] Among them, λ i is the i-th eigenvalue of the Laplace matrix, v i is the corresponding eigenvector. After eigenvalue decomposition, we obtain n eigenvalues ​​and their corresponding n-dimensional eigenvectors. The relationship between eigenvalues ​​and eigenvectors can reveal the partitioning relationship between nodes. The partitioning scheme corresponding to the eigenvector with the smaller eigenvalue indicates the smallest partitioning loss.

[0119] Eigenvectors are continuous values. Therefore, to facilitate subsequent clustering, we select the eigenvectors corresponding to the smallest k to 2k eigenvalues ​​as valid features, where k is the predefined number of cluster categories. These eigenvectors correspond to the smallest segmentation loss and effectively reflect the cluster structure of the transportation network. These selected eigenvectors form the feature matrix F, which serves as input to the K-means clustering algorithm to achieve the final node classification.

[0120] By spectral decomposition of the Laplace matrix and selecting eigenvectors, the graph partitioning loss can be minimized, allowing nodes with larger connection weights to be grouped together, while nodes with smaller connection weights are grouped together. This method effectively extracts the partitioning relationships between nodes through eigenvalue decomposition, providing reliable data support for subsequent clustering analysis. By selecting the eigenvector corresponding to the minimum eigenvalue, the accuracy of the clustering results is improved, ensuring that the final traffic node division is more consistent with actual traffic needs and regional characteristics, providing an optimized solution for urban traffic management.

[0121] In step S4, based on the spectral decomposition results from the previous step, the K-means algorithm is used to cluster the feature vectors to complete the regional division of traffic nodes. The goal of this step is to map the feature vectors to a finite number of categories based on the segmentation relationship between nodes, so that similar nodes are grouped together in the same region, achieving reasonable regional division based on traffic demand.

[0122] First, the feature matrix F extracted in step S3 is used as the input for K-means clustering. F consists of k to 2k feature vectors, each representing a low-dimensional feature of a traffic node. The goal of the K-means clustering algorithm is to classify traffic nodes into k categories through iterative optimization, so that nodes within the same category have strong traffic connections, while nodes in different categories have weak traffic connections.

[0123] Define the cluster center matrix C, where C = [c1,c2,...,c k ] represents the center point of k categories. The clustering process first randomly initializes k cluster centers, and then calculates the distance between each node x and i Distance to each cluster center:

[0124] d(x i ,c j )=∥x i -c j ∥

[0125] Among them, x i is the i-th row data in the feature matrix F, representing the feature representation of node i, c j is the cluster center of the current category j. Assign each node to the cluster center c with the smallest distance j Corresponding categories:

[0126] cluster(x i )=argmin j d(x i ,c j )

[0127] After completing a node assignment, recalculate the new cluster center of each category, that is, calculate the mean of all nodes in the category:

[0128]

[0129] Among them, |C j | is the number of nodes in category j, x i is the feature representation of all nodes in this category. After this step is completed, the cluster center is updated and the above calculation is repeated until the cluster center no longer changes significantly or the maximum number of iterations is reached.

[0130] K-means clustering terminates when all cluster centers converge (i.e., the difference between the newly calculated center and the previously calculated center is within a set threshold), or when the maximum number of iterations is reached. Ultimately, the K-means clustering results will indicate the classification of each node, representing the regional division of the transportation network.

[0131] In step S5, based on the spectral decomposition and K-means clustering results from the previous steps, the spatial extent of the regions is optimized by introducing the normalized GPS coordinate features of the nodes. This makes the macro-regions more reasonable, avoids excessive coverage and overlap, and improves management feasibility. The goal of this step is to comprehensively utilize the topological relationships of node traffic demand and spatial distribution characteristics to establish a visual traffic region model and achieve more accurate regional division.

[0132] First, we use the correspondence between the GPS coordinates of the nodes and the Geohash grid to perform an initial classification of n nodes into k categories to form a preliminary spatial region division scheme. Geohash encoding is used to map geographic coordinates to a specific spatial grid, preserving the spatial relationship between adjacent nodes. For each category, we calculate the mean and variance of the longitude and latitude of all nodes within it, and use a two-dimensional Gaussian distribution to fit the GPS spatial distribution of that category:

[0133]

[0134] Among them, μ x ,μ y Represents the mean longitude and latitude of the nodes in this category, σ x ,σ y Represents the standard deviation of longitude and latitude respectively. This step is used to visualize the spatial distribution of the region on the map and stop the iteration when the spatial distribution of the GPS category tends to be stable.

[0135] Then, the mean center of the spectral clustering feature of each category is calculated as the spectral clustering feature center of the category. Define the mean center vector v of the spectral clustering feature c :

[0136]

[0137] Among them, C is the node set of the current category, v i is the spectral decomposition feature vector of node i. This center point is used to calculate the clustering distance from the node to the category.

[0138] Next, the comprehensive distance from each node to each category center is calculated, taking into account the influence of GPS coordinate distance and spectral clustering feature distance. Spectral clustering feature distance calculates the Euclidean distance between the spectral decomposition feature of each node and the spectral feature of the category center:

[0139] d spectral (v i ,v c )=∥v i -v c ∥

[0140] Finally, calculate node i to category C jThe comprehensive distance is expressed as the product of the spectral clustering feature distance and the GPS spatial probability distance:

[0141] d final =d spectral ×P(x,y)

[0142] Among them, d spectral Reflecting the topological feature differences of nodes, P(x,y) represents the spatial belonging probability of a node in a category. This calculation method comprehensively considers the topological structure and spatial position of the node.

[0143] Each node is assigned to the category with the smallest distance, the node category label is updated, and the process returns to the category spatial distribution calculation step for the next iteration until the category GPS spatial distribution stabilizes. This method makes the spatial coverage of the macro region more reasonable and improves the operability of urban traffic management.

[0144] This method makes comprehensive use of traffic topological relationships and spatial distribution characteristics to make regional division more in line with actual traffic needs, optimize traffic flow analysis, improve urban traffic management, and provide a scientific basis for traffic planning and resource allocation.

[0145] The apparatus for macro-region division based on urban traffic demand described below and the method for macro-region division based on urban traffic demand described above can refer to each other.

[0146] Please see the attached Figure 2 The present invention also provides a macro-region division device based on urban traffic demand, including:

[0147] The data acquisition module 10 is used to obtain real-time traffic demand data of traffic nodes;

[0148] A matrix construction module 20 is used to generate an adjacency matrix, a degree matrix and a normalized Laplace matrix;

[0149] Spectral decomposition module 30, for performing eigenvalue decomposition on the normalized Laplace matrix and extracting eigenvectors;

[0150] A clustering module 40 is used to divide the preliminary regions by using a K-means algorithm;

[0151] The optimization module 50 is used to perform spatial optimization on the area division result in combination with the GPS coordinates.

[0152] The device of this embodiment can be used to execute the above method embodiment, and its principles and technical effects are similar, so they will not be repeated here.

[0153] Example:

[0154] like Figure 3As shown, the embodiment of the present invention provides a method for urban traffic zone division based on the topological relationship of traffic demand and spatial distribution characteristics, with the goal of minimizing zone segmentation losses and achieving reasonable management of traffic nodes in the urban dynamic road network. The specific implementation steps are described below with reference to the diagram.

[0155] Step 1: Obtain traffic demand at traffic nodes in the urban dynamic road network

[0156] The traffic demand of urban traffic nodes is obtained through traffic surveys, road network traffic status data from Internet travel platforms, and flow data collected by traditional traffic detection equipment.

[0157] The granularity of traffic nodes can be selected as intersections, sections or roads, and the node granularity is determined based on data accuracy and regional division requirements.

[0158] Figure 4 (a) and Figure 4 (b) shows the traffic node demand distribution of a certain city during the morning and evening peak hours.

[0159] Step 2: Calculate the traffic demand matrix

[0160] 1. Construct the adjacency matrix W:

[0161] Calculate the node pairs connected by traffic demand, and define the connection weight as the sum of the two-way traffic demand between the nodes: w ij =Demand from transportation node i to j + demand from transportation node j to i.

[0162] 2. Calculate the degree matrix D:

[0163] The total demand weight of each node is calculated by the adjacency matrix W:

[0164]

[0165] 3. Calculate the Laplace matrix L:

[0166] Used to measure the segmentation loss of the graph:

[0167]

[0168] 4. Calculate the normalized Laplace matrix L norm :

[0169] Due to the uneven distribution of traffic demand caused by differences in road grades and locations, it is necessary to normalize the Laplace matrix:

[0170]

[0171] Step 3: Spectral decomposition

[0172] Compute the eigenvalue decomposition:

[0173] For the normalized Laplace matrix L norm Perform eigenvalue decomposition to obtain several eigenvectors with minimum segmentation loss.

[0174] Let L norm There are n eigenvalues ​​and corresponding eigenvectors, and k to 2k eigenvectors whose values ​​are closest to zero are selected as clustering features.

[0175] Through spectral decomposition, the aggregation characteristics of traffic nodes are obtained. Subsequently, the nodes are preliminarily divided into regions through K-means clustering.

[0176] Step 4: K-means clustering

[0177] Preliminary clustering based on spectral decomposition features

[0178] The K-means algorithm is used to cluster the spectral decomposition feature vectors to obtain k preliminary regions.

[0179] The preliminary regional division results are as follows: Figure 5 (a) and Figure 5 As shown in (b), different colors and shapes represent different types of traffic areas.

[0180] Since some macro-regions have a wide coverage area and overlap, they are difficult to manage.

[0181] It is necessary to introduce GPS normalized coordinate features to optimize the regional scope.

[0182] Step 5: Introduce GPS normalized coordinate features to optimize regional division

[0183] 1. Initialize classification

[0184] Through Geohash coding, n nodes are divided into k categories according to GPS coordinates as the initial clustering result.

[0185] 2. Calculate GPS statistical characteristics

[0186] Calculate the latitude and longitude mean and variance of each category, and use a two-dimensional Gaussian distribution to fit the spatial distribution of the category:

[0187]

[0188] 3. Calculate the spectral clustering feature center

[0189] Compute the mean center of the spectral clustering features for each class:

[0190]

[0191] 4. Calculate the distance from the node to the category center

[0192] Calculate the spectral clustering feature distance:

[0193] d spectral (v i ,v c )=∥v i -v c ∥

[0194] Calculate the comprehensive distance, combining the spectral clustering feature distance and GPS probability distribution:

[0195] d final =d spectral ×P(x,y)

[0196] 5. Update node category

[0197] Based on the minimum comprehensive distance principle, the node categories are updated and iterative calculations are performed until the GPS spatial distribution of the categories is stable.

[0198] Figure 6 This shows the regional division results after adding GPS normalization features. Figure 5 ,The optimized regional boundaries are clearer, the macro-region span is limited, and the manageability is improved.

[0199] In summary, based on the topological relationship of traffic demand and GPS normalization characteristics, the present invention adopts spectral decomposition and K-means clustering methods to achieve efficient urban traffic area division and improve the rationality and scientificity of traffic management.

[0200] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A macro-regional division method based on urban traffic demand, characterized by: The following steps are involved: Obtain real-time traffic demand data for multiple traffic nodes in the urban transportation network; Based on the traffic demand data, construct an adjacency matrix, a degree matrix and a Laplace matrix of the traffic network; Performing spectral decomposition on the Laplace matrix to extract spectral feature vectors of traffic nodes; Based on the spectral feature vector, the traffic nodes are preliminarily divided into multiple macro-regions using clustering algorithm; The macro-regions after preliminary division are optimized based on the spatial location information of the traffic nodes to obtain the final regional division results, which specifically include: Initialize geospatial region division based on GPS coordinates of traffic nodes; Calculate the mean and variance of GPS coordinates of nodes in each category and fit a two-dimensional Gaussian distribution; Calculate the comprehensive distance between the node and each category based on the spectral features and GPS coordinate features; Update the node category according to the comprehensive distance until the category division is stable; The GPS spatial probability distance is the probability density value of the two-dimensional Gaussian distribution, and the calculation formula is: Where (x, y) is the GPS coordinate of the node, μ x ,μ y represents the mean of GPS coordinates of all nodes in the current category, σ x ,σ y Indicates the standard deviation of GPS coordinates of all nodes in the current category; The calculation formula of the comprehensive distance is: d final =d spectral ×P(x,y) Where: d spectral is the Euclidean distance between spectral clustering feature vectors; P(x,y) is the GPS spatial probability distance.

2. The macro-region division method based on urban traffic demand according to claim 1 is characterized in that: The traffic node is an urban road intersection or road section, and the traffic demand data includes at least one of Internet travel platform data, traffic detection equipment data, and bus route data.

3. The macro-region division method based on urban traffic demand according to claim 1 is characterized in that: The adjacency matrix is ​​constructed as follows: If the traffic node v i With v j If there is a traffic demand correlation between them, then the adjacency matrix element w ij =T(i→j)+T(j→i), otherwise w ij =0; Where T(i→j) represents the node v i to v j traffic flow.

4. The macro-region division method based on urban traffic demand according to claim 3 is characterized in that: The Laplace matrix is ​​a normalized Laplace matrix, which is constructed as follows: L norm =D -1 / 2 (D-W)D -1 / 2 Among them, W is the adjacency matrix, D is the degree matrix, and D -1 / 2 is the inverse square root of the degree matrix.

5. The macro-region division method based on urban traffic demand according to claim 4 is characterized in that: The step of performing spectral decomposition on the Laplace matrix to extract the characteristic vector of the traffic node includes: Perform eigenvalue decomposition on the normalized Laplace matrix and extract the eigenvectors corresponding to the 2k smallest eigenvalues ​​to form the K-means clustering features.

6. The macro-region division method based on urban traffic demand according to claim 1 is characterized in that: The clustering algorithm is a K-means algorithm, including: Initialize cluster centers; Calculate the Euclidean distance from each node to the cluster center; Iteratively update the node category until the clustering is stable.

7. A macro-region division device based on urban traffic demand, configured to execute the macro-region division method based on urban traffic demand according to any one of claims 1 to 6, characterized in that: include: Data acquisition module, used to obtain real-time traffic demand data of traffic nodes; Matrix building module, used to generate adjacency matrix, degree matrix and normalized Laplacian matrix; Spectral decomposition module, used to perform eigenvalue decomposition on the normalized Laplace matrix and extract eigenvectors; Clustering module, used to divide preliminary regions using the K-means algorithm; The optimization module is used to spatially optimize the regional division results in combination with GPS coordinates.

Images