A spatial data integration method and system for digital twin cities

Abstract

The present invention relates to the technical field of digital twin cities, and in particular to a spatial data integration method and system applied to digital cities. The method comprises the following steps: using satellite remote sensing equipment, a laser radar building scanner and a traffic sensor to obtain digital twin city spatial data and digital twin city flow data; using a spatial interpolation algorithm to interpolate and map the real-time traffic simulation data of the digital twin city to generate a traffic simulation density distribution map of the digital twin city; using a voxel modeling method based on three-dimensional modeling technology to perform three-dimensional modeling on a three-dimensional digital twin city data set; using a deep learning algorithm to perform real-time interactive visualization processing on the digital twin city traffic simulation density distribution map; performing dilation convolution and multi-scale sampling on the digital twin city building three-dimensional model and the digital twin city traffic real-time simulation interactive view to construct a digital twin smart city model; the present invention realizes efficient and orderly operation of the digital twin city.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twin cities, and in particular to a spatial data integration method and system applied to digital twin cities. Background Art

[0002] As a new urban management model and technical means, digital twin city is gradually becoming a leader in urban technology transformation. Based on digital modeling, real-time simulation, data intelligence and other technologies, digital twin city integrates urban physical space and virtual space, and provides new ideas and solutions for urban planning decision-making, resource management, environmental monitoring, emergency response and other aspects. In the traditional digital twin city system, various data have different spatial reference systems and heterogeneity of data formats, which leads to data confusion, low efficiency and inability to interconnect with the real city in real time. Therefore, a spatial data integration method and system applied to digital twin city is used to obtain real-time spatial data and traffic flow data in the real world, and combined with artificial intelligence, computer deep learning and other technologies, seamless integration and collaborative use of data are achieved. The complex pattern of data is learned by artificial neural network with multi-layer structure, and data visualization and interactive processing are realized, real-time monitoring and operation guidance are carried out, and a digital twin smart city model is established through dilated convolution and multi-scale sampling, and real-time management of virtual city and real city is realized, and full-method and multi-level urban supervision and command are carried out. Summary of the invention

[0003] In order to solve the above-mentioned technical problems, the present invention proposes a spatial data integration method and system applied to a digital twin city to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above object, the present invention provides a spatial data integration method applied to a digital twin city, comprising the following steps:

[0005] Step S1: using satellite remote sensing equipment, laser radar and traffic sensors to obtain digital twin city spatial data and digital twin city flow data; digital twin city data includes urban three-dimensional spatial data, urban geographic spatial data, urban building layout data and building structure spatial data; digital twin city flow data includes pedestrian flow data and road traffic flow data;

[0006] Step S2: Use the spatial matching method to integrate the spatial data of the digital twin city to generate a three-dimensional digital twin city data set; use the relevant information factor fusion algorithm to normalize and fuse the digital twin city traffic data to generate real-time traffic simulation data of the digital twin city;

[0007] Step S3: Use a spatial interpolation algorithm to interpolate and map the real-time traffic simulation data of the digital twin city to generate a traffic simulation density distribution map of the digital twin city;

[0008] Step S4: performing three-dimensional modeling on the three-dimensional digital twin city data set using a voxel modeling method based on three-dimensional modeling technology to generate a three-dimensional model of digital twin city buildings;

[0009] Step S5: using a deep learning algorithm to perform real-time interactive visualization processing on the digital twin city traffic simulation density distribution map to generate an interactive view of the digital twin city traffic real-time simulation;

[0010] Step S6: Use the circular convolution algorithm to perform dilated convolution and multi-scale sampling on the three-dimensional model of digital twin city buildings and the interactive view of digital twin city traffic real-time simulation to build a digital twin smart city model.

[0011] The present invention provides a spatial data integration method for a mathematical twin city. In order to obtain basic information of the city, including building structure, geographical location, passenger flow and road traffic flow, etc., these data are the basis of the digital twin city and the data source for subsequent processing and analysis. The spatial data of the digital twin city is integrated through a spatial matching method, and different types of data are integrated into a whole, which is convenient for subsequent data processing and analysis, and can also reduce repeated data parts and reduce the cost of data storage and processing. The relevant information factor fusion algorithm is used to normalize and fuse the traffic data of the digital twin city, and the traffic data from different sources are fused into a more complete and accurate data set. At the same time, the traffic simulation data can be updated in real time to ensure its accuracy and timeliness. The real-time traffic simulation data of the digital twin city is interpolated and mapped through a spatial interpolation algorithm to match the traffic flow data with the actual spatial distribution. Intuitively express the traffic conditions of the city, perform three-dimensional modeling of the digital twin city data set, and present the architectural structure of the digital twin city to users in a more intuitive and vivid way, so that users can better understand the architectural layout and structure of the city. Through deep learning algorithms, real-time interactive visualization of the digital twin city traffic simulation density distribution map is performed to generate an interactive view of the digital twin city traffic real-time simulation, allowing users to have a deeper understanding of the city’s traffic conditions and feel the city’s changes and development more intuitively. The generated interactive view allows users to interact in real time and understand the real-time situation of the city. Through the circular convolution algorithm, a digital twin smart city model is constructed, and all aspects of the city are integrated and optimized to generate a more complete and accurate digital twin city model, which will also provide a more accurate basis for urban planning and management, realize real-time management of virtual cities and real cities, and conduct full-method, multi-level urban supervision and command.

[0012] Preferably, step S1 comprises the following steps:

[0013] Step S11: Use high-precision and high-resolution satellite remote sensing equipment to cruise and scan the city to obtain digital twin city spatial data; the digital twin city multi-source data includes urban three-dimensional spatial data, urban geographic spatial data and urban building layout data;

[0014] Step S12: Use laser radar to perform laser scanning on the internal structure of urban buildings to generate building structure space data, where the building structure space data includes building shape and size data, building elevation data, building internal structure data, and building material data;

[0015] Step S13: Use traffic sensors to obtain digital twin city traffic data, where the digital twin city traffic data includes pedestrian traffic data and road traffic flow data.

[0016] The present invention uses high-precision and high-resolution satellite remote sensing equipment to perform cruise scanning, so as to obtain the actual form and layout of the city. By acquiring the spatial data of the digital twin city, it can assist in urban spatial planning and management, and improve the accuracy and scientific nature of urban planning. By using lidar to perform laser scanning of the internal structure of a building, detailed information such as the internal structure, facilities and decoration of the building can be obtained. This information helps to strengthen the management and maintenance of the building, and also provides basic data for subsequent building model construction and simulation. By acquiring the flow data of the digital twin city through traffic sensors, information such as pedestrian flow and road traffic flow can be obtained, which is very important for urban traffic management and traffic planning, and can help analyze urban traffic hotspots and bottlenecks and improve the level of urban traffic intelligence.

[0017] Preferably, step S2 comprises the following steps:

[0018] Step S21: using a vector data standardization algorithm to standardize the digital twin city spatial data to generate digital twin city vector space benchmark data;

[0019] Step S22: using a spatial matching method to perform spatial data integration on the digital twin city vector space benchmark data to generate a three-dimensional digital twin city data set;

[0020] Step S23: Use the principal component analysis method to perform principal component dimensionality reduction extraction on the digital twin city traffic data to obtain relevant factor data;

[0021] Step S24: using the correlation factor weight analysis method to analyze the correlation factor coefficients of the correlation factor data, and generating standard related information factors;

[0022] Step S25: normalizing the digital twin city flow data using the maximum-minimum normalization method to generate digital twin city flow range data;

[0023] Step S26: Based on the standard relevant information factors, the relevant information factor fusion algorithm is used to fuse the digital twin city traffic range data to generate real-time traffic simulation data for the digital twin city.

[0024] The present invention standardizes the spatial data of the digital twin city through a vector data standardization algorithm, and can process data from different sources for consistency to ensure the consistency and standardization of the data. After the digital twin city vector space benchmark data is generated, the complexity of subsequent calculations and processing can be reduced, and the accuracy and reliability of the digital twin city model can be improved. The spatial data integration of the digital twin city vector space benchmark data is performed using the spatial matching method, and data from different sources can be integrated to generate a three-dimensional digital twin city data set, which can better reflect the spatial state and spatial distribution characteristics of the city, and provide more complete and comprehensive data support for the digital twin city model and analysis. The principal component analysis method is used to perform principal component dimensionality reduction extraction on the digital twin city traffic data, and the main traffic flow factors can be extracted from a large amount of traffic flow data, which simplifies the complexity of the data and makes the data easier to process and understand. , the correlation factor weight analysis method is used to analyze the correlation factor coefficients of the correlation factor data, and standard relevant information factors are generated. The correlation between various traffic flow factors can be deeply analyzed, and the role relationship and influence degree between various factors can be better revealed, providing a more refined data basis for subsequent data processing and analysis. The maximum-minimum normalization method is used to normalize the digital twin city traffic data, which can unify the data of various traffic factors to the same scale, avoiding the dimensional difference of the data, making the comparison between different factors more accurate and reliable. Based on the standard relevant information factors, the relevant information factor fusion algorithm is used to fuse the digital twin city traffic range data, which can fuse the data of various traffic factors to generate more comprehensive and accurate digital twin city real-time traffic simulation data, providing a more scientific and effective decision-making basis for subsequent traffic management and planning.

[0025] Preferably, step S3 comprises the following steps:

[0026] Step S31: preprocessing the digital twin city real-time traffic simulation data to generate a digital twin city real-time traffic preprocessing pipeline;

[0027] Step S32: using a matrix decomposition method to perform matrix division on the real-time traffic preprocessing pipeline of the digital twin city to generate a traffic matrix diagram of the digital twin city;

[0028] Step S33: using a feature point detection algorithm to mark feature points on the digital twin city traffic matrix diagram to generate a digital twin city traffic flow coordinate point grid;

[0029] Step S34: using the point sampling interpolation of the spatial interpolation algorithm to perform weighted average calculation on the digital twin city traffic flow coordinate point grid to obtain the digital twin city traffic simulation interpolation points;

[0030] Step S35: Based on the digital twin city traffic simulation interpolation points, the urban traffic flow spatial interpolation distance weight calculation formula is used to interpolate and map the digital twin city traffic flow coordinate point grid to generate a digital twin city traffic simulation density distribution map;

[0031] The present invention performs data preprocessing on the real-time traffic simulation data of the digital twin city, thereby removing noise and redundant information in the data, ensuring the accuracy and reliability of the traffic data of the digital twin city, and generating a real-time traffic preprocessing pipeline for the digital twin city at the same time, thereby improving the efficiency and accuracy of subsequent processing and calculation, and using the matrix decomposition method to perform matrix division on the real-time traffic preprocessing pipeline for the digital twin city, thereby dividing the traffic data according to different factors, thereby simplifying the processing and calculation of the data, and generating a traffic matrix diagram for the digital twin city, thereby more intuitively displaying the traffic relationship between different road sections and regions, and providing basic data for subsequent traffic analysis and simulation, and using a feature point detection algorithm to mark feature points on the traffic matrix diagram for the digital twin city, thereby identifying and marking key points and key factors in the traffic data, thereby facilitating subsequent data analysis and processing, and generating a traffic flow coordinate point grid for the digital twin city, It can more clearly reflect the traffic flow distribution and rules in different regions and sections, and use the point sampling interpolation of the spatial interpolation algorithm to perform weighted average calculation on the traffic flow coordinate point grid of the digital twin city. The weighted average of the traffic flow in different regions can be counted to further reflect the overall characteristics of the digital twin city traffic data. After obtaining the digital twin city traffic simulation interpolation points, the traffic flow of the digital twin city can be simulated more accurately. Based on the digital twin city traffic simulation interpolation points, the urban traffic flow spatial interpolation distance weight calculation formula is used to interpolate and map the digital twin city traffic flow coordinate point grid. The digital twin city traffic flow data can be fitted from the local to the whole, and a digital twin city traffic simulation density distribution map can be generated. This can provide a more comprehensive and intuitive understanding of the traffic conditions and congestion conditions in different regions and sections, and provide more scientific and comprehensive data support for the traffic planning and management of the digital twin city.

[0032] Preferably, the urban traffic flow spatial interpolation distance weight calculation formula of step S35 is specifically:

[0033]

[0034] Among them, W i is the interpolation weight of the i-th sampling point, w pop,i is the flow of people at the i-th sampling point, w traf,i is the road traffic volume at the i-th sampling point, q pop,i is the flow weight value of the i-th sampling point, q traf,i is the road traffic flow weight value of the i-th sampling point, θ i is the azimuth between the i-th sampling point and the point to be interpolated, d i is the distance between the i-th sampling point and the point to be interpolated, p is the attenuation exponent, and λ is the attenuation rate.

[0035] The present invention is achieved by pop,i +w traf,i Calculate the data of road vehicle and pedestrian flow, through Calculate the weighted values ​​of the pedestrian flow and road traffic flow at the i-th sampling point, taking into account the weights of the population flow and road traffic flow. By incorporating the weights of the population flow and road traffic flow at the sampling point into the calculation, the traffic conditions in different areas can be better reflected. This can effectively avoid the situation of simple averaging or weighted averaging of the entire city, thereby improving the accuracy of traffic flow estimation. The cosine function and distance decay function are used to make the sampling points with a longer distance have a smaller weight on the interpolation points, and the sampling points with a shorter distance have a larger weight on the interpolation points, so that the interpolation calculation can be performed more reasonably. The azimuth and distance weights can accurately describe the spatial relationship between the points to be interpolated and the sampling points, and take into account factors such as spatial distance and direction, so that the interpolation results are more accurate and reliable. p and λ are two adjustable parameters used to adjust the influence of distance. The larger p is, the faster the influence of the sampling points on the estimated value decreases with the increase of distance. The larger λ is, the slower the influence of the sampling points on the estimated value decreases with the increase of distance. The weight function uses the cosine function and distance decay function to make the sampling points with a longer distance have a smaller weight on the interpolation points, and the sampling points with a shorter distance have a larger weight on the interpolation points. The role of the cosine function is to adjust the weight according to the direction angle, and the role of the distance attenuation function is to adjust the weight according to the distance. Through the normalization of the weights, the influence of the difference in the number of sampling points on the interpolation results is avoided, making the interpolation results more stable and reliable. Taking into account the flow of people and road traffic, it more comprehensively reflects the spatial changes in urban traffic flow, and can provide useful data support for urban traffic planning and management.

[0036] Preferably, step S33 includes the following steps:

[0037] Step S331: Detect feature points of the digital twin city traffic matrix diagram using a feature point detection algorithm to generate initial digital twin city traffic feature point data;

[0038] Step S332: Filter and screen the digital twin city traffic feature point data using the characteristic value threshold limiting method to generate benchmark digital twin city traffic feature point data;

[0039] Step S333: Marking the traffic feature point data of the benchmark digital twin city by using the spatial three-dimensional coordinate method to generate the traffic feature point coordinates of the digital twin city;

[0040] Step S334: Use the equidistant grid algorithm to grid the coordinates of the digital twin city traffic feature points to generate a digital twin city traffic flow coordinate point grid.

[0041] The present invention performs feature point detection on the digital twin city traffic matrix diagram through a feature point detection algorithm, and can automatically identify and mark key feature points in the digital twin city traffic data, such as intersections, crossroads, etc. After generating the initial digital twin city traffic feature point data, it can provide more information and features for the subsequent digital twin city traffic data analysis. The digital twin city traffic feature point data is filtered and screened by the feature value threshold limitation method, which can exclude some irrelevant or very small feature points, thereby improving the accuracy and efficiency of the digital twin city traffic analysis. After generating the benchmark digital twin city traffic feature point data, it can provide more accurate and meaningful data support for the subsequent generation of digital twin city traffic coordinate point grids. The benchmark digital twin city traffic is filtered and screened by the spatial three-dimensional coordinate method. By marking the feature point data, the location information of the digital twin city traffic feature points can be located in three-dimensional coordinates, which can more accurately express the location and distribution of the digital twin city traffic data feature points in the real world. After the digital twin city traffic feature point coordinates are generated, it can provide more accurate and meaningful data support for the generation of the digital twin city traffic coordinate point grid. The digital twin city traffic feature point coordinates are gridded using the equidistant grid algorithm, and the digital twin city traffic feature point data can be discretized into grid data, which is convenient for subsequent digital twin city traffic data analysis and processing. After the digital twin city traffic flow coordinate point grid is generated, it will provide more comprehensive, accurate and intuitive basic data support for digital twin city traffic simulation and analysis.

[0042] Preferably, step S4 comprises the following steps:

[0043] Step S41: performing stereo pixel conversion on the three-dimensional digital twin city data set to generate large three-dimensional voxels such as the digital twin city;

[0044] Step S42: gridding a large three-dimensional voxel such as a digital twin city using an inverse distance weighted method to generate a digital twin city grid model;

[0045] Step S43: Use the voxel modeling method based on the three-dimensional modeling technology to perform three-dimensional modeling on the digital twin city grid model to generate a three-dimensional model of the digital twin city buildings.

[0046] The present invention converts three-dimensional pixels of a three-dimensional digital twin city data set into three-dimensional models, which can more realistically express the spatial structure and characteristics of the digital twin city. After generating equal-sized three-dimensional voxels, each voxel has the same size, which is convenient for calculation and modeling in subsequent processing. The inverse distance weighted method is used to grid the large three-dimensional voxels of the digital twin city, and the three-dimensional data can be discretized into grid points, which is convenient for subsequent processing and modeling. The inverse distance weighted method can effectively convert the three-dimensional voxel information into a grid model, retaining the morphology and characteristics of the digital twin city as much as possible. Features, making the subsequent digital twin city building three-dimensional model more accurate and reliable. The digital twin city grid model is three-dimensionally modeled using the voxel modeling method based on three-dimensional modeling technology, which can convert the digital twin city data into a real three-dimensional building model. The voxel modeling method based on three-dimensional modeling technology can convert the digital twin city building features from discrete grid points into a real three-dimensional building model, so that the form and structure of the digital twin city can be more realistically expressed and simulated. At the same time, the voxel modeling method based on three-dimensional modeling technology can also support rapid model updates and expansions, making the modeling of digital twin cities more efficient and flexible.

[0047] Preferably, step S5 comprises the following steps:

[0048] Step S51: using a deep learning algorithm to perform trend analysis on the digital twin city traffic simulation density distribution map to generate digital twin city traffic trend data;

[0049] Step S52: performing trend calculation on the digital twin city traffic trend data using the digital twin city traffic trend prediction formula to generate a digital twin city traffic trend graph;

[0050] Step S53: using the digital twin city traffic simulation density distribution map and the digital twin city traffic trend map to perform real-time simulation analysis to generate a digital twin city traffic real-time simulation map;

[0051] Step S54: performing data visualization processing on the real-time simulation diagram of digital twin city traffic to generate a real-time simulation visualization view of digital twin city traffic;

[0052] Step S55: Use the JavaScript library to dynamically and interactively process the visualization view of the real-time simulation of digital twin city traffic to generate an interactive view of the real-time simulation of digital twin city traffic.

[0053] The present invention uses a deep learning algorithm to perform trend analysis on the digital twin city traffic simulation density distribution map, and can conduct in-depth exploration and analysis of the traffic conditions of the digital twin city. Through trend analysis, digital twin city traffic trend data can be generated, including information such as traffic bottlenecks, peak hours and congestion conditions. These data can guide digital twin city traffic operations and planning, improve urban traffic operation efficiency and safety, and use the digital twin city traffic trend prediction formula to perform trend calculation on the digital twin city traffic trend data, which can predict the future traffic trends and congestion conditions of the digital twin city. The traffic condition information in the digital twin city traffic trend map can be used to study traffic planning and deployment, predict possible future congestion conditions, prepare traffic emergency plans, etc. The digital twin city traffic simulation density distribution map and the digital twin city traffic trend map are used for real-time simulation analysis, which can monitor the traffic status of the digital twin city in real time and respond and cope with emergencies in a timely manner. The real-time traffic simulation diagram of the digital twin city can predict the traffic conditions in certain areas, help city managers optimize routes and resource allocation, and solve the problem of urban traffic congestion. Data visualization processing of the real-time traffic simulation diagram of the digital twin city can visualize the traffic conditions of the digital twin city and help users quickly obtain and understand traffic information. The real-time simulation visualization view of the digital twin city traffic makes data easier for people to understand and apply through visual charts, images, and maps, making the operation of urban traffic more efficient and safe. The real-time simulation visualization view of the digital twin city traffic is dynamically and interactively processed using the JavaScript library, which can realize the dynamic interactive presentation of the digital twin city traffic simulation. This interactive view of the real-time simulation of the digital twin city traffic can help users have a deeper understanding of the traffic conditions of the digital twin city, including traffic flow, congestion, etc., so as to optimize urban traffic planning, real-time scheduling and prediction in a targeted manner.

[0054] Preferably, the digital twin city traffic trend prediction formula in step S52 is specifically:

[0055]

[0056] Where x is the spatial position vector area value, t is the time period, f(x,t) is the traffic flow prediction trend density of the spatial position vector area value of x in the time period t, and w i (t) is the trend influence weight of the i-th sampling point in time period t, N pop,i is the flow threshold of the i-th sampling point, N traf,iis the traffic flow threshold of the i-th sampling point, u i is the mean vector of the i-th sampling point in space.

[0057] The present invention is achieved by i (t) Taking the spatial position vector area value x and time t as prediction conditions, it represents the traffic flow trend density distribution of the predicted target area and time period. Among the n sampling points in the same time period, the weight w of each sampling point is calculated. i (t), the weight reflects the impact of the current moment on traffic flow. The weight calculation can take into account factors such as historical data and current weather. Common methods include linear regression and support vector regression. pop,i +N traf,i , for each sampling point i, its flow threshold N pop,i and traffic flow threshold N traf,i It represents the density level of population and traffic flow at the location of the point, and is used to measure the impact of the point on traffic flow. For each sampling point i, its spatial position mean vector represents the spatial location attribute of the point, which can be obtained using cluster analysis and other methods. The Gaussian kernel function is used to obtain a distribution probability density function about the location coordinates, where is the relative position deviation, is the probability density function of the Gaussian distribution, and the other e -t is the attenuation coefficient in the time dimension. The influence of each sampling point is added up by weighted summation to obtain the traffic flow trend density distribution of the position vector area value in time period t. The advantage of this method is that it can take into account the influence of multiple factors such as population flow, vehicle flow and spatial distribution on traffic flow, so the prediction effect is more accurate.

[0058] Preferably, step S6 comprises the following steps:

[0059] Step S61: using a superpixel convolutional network to perform convolution preprocessing on the interactive view of the real-time simulation of digital twin city traffic, and generate a feature sample cluster of the real-time simulation of digital twin city traffic;

[0060] Step S62: using a circular convolution algorithm to perform convolution data cutting on the digital twin city traffic real-time simulation feature sample cluster to generate a digital twin city traffic real-time simulation feature network;

[0061] Step S63: using the dilated convolution algorithm to perform edge feature enhancement processing on the digital twin city traffic real-time simulation feature network to generate a digital twin city traffic real-time simulation feature map;

[0062] Step S64: using a multi-scale sampling algorithm to perform spatial pyramid pooling multi-layer sampling on the digital twin city building three-dimensional model to generate a digital twin city building feature sequence;

[0063] Step S65: Use the association rule algorithm to perform data mining modeling on the digital twin city building feature sequence and the digital twin city traffic real-time simulation feature map to build a digital twin smart city model.

[0064] The present invention performs convolution preprocessing on the interactive view of the real-time simulation of digital twin city traffic through a superpixel convolution network, which can effectively reduce data redundancy and distribute data to different pixel areas. The superpixel convolution network can further extract the feature information of the real-time simulation image of digital twin city traffic, quickly locate and identify information such as traffic objects and vehicles, thereby providing a high-quality data basis for subsequent processing. The circular convolution algorithm is used to perform convolution data cutting on the feature sample clusters of the real-time simulation of digital twin city traffic, which can effectively compress and classify the original data, and perform complex calculations based on the data features. The circular convolution algorithm can retain the spatial information and connection relationship of the original data, and can also adaptively perform data learning and model optimization to improve the accuracy and reliability of the model. The dilated convolution algorithm is used to perform edge feature enhancement processing on the feature network of the real-time simulation of digital twin city traffic, which can further extract edge features in the feature map of the real-time simulation of digital twin city traffic, such as lane markings, traffic signs, etc. Thereby further improving the accuracy and reliability of digital twin city traffic simulation. The dilated convolution algorithm can also effectively reduce shadow interference and edge blur problems, and improve the visual effect of digital twin city traffic simulation. The multi-scale sampling algorithm is used to perform spatial pyramid pooling and multi-layer sampling on the three-dimensional model of digital twin city buildings, which can retain the information of digital twin city buildings at different scales, thereby improving the high definition and accuracy of digital twin city simulation. The multi-scale sampling algorithm can also detect the spatial distribution and morphological characteristics of digital twin city buildings to optimize the spatial design and planning of buildings. The association rule algorithm is used to perform data mining modeling on the digital twin city building feature sequence and the digital twin city traffic real-time simulation feature map. Through the correlation analysis of the digital twin city building feature sequence and the digital twin city traffic real-time simulation feature map, a digital twin smart city model is established. This model can provide more accurate data support and analysis for city managers, and support the smart operation and sustainable development of digital twin cities.

[0065] Preferably, step S64 includes the following steps:

[0066] Step S641: Use a multi-scale sampling algorithm to perform spatial pyramid pooling multi-layer sampling on the digital twin city building three-dimensional model to generate digital twin city building convolution feature data;

[0067] Step S642: performing convolution feature mapping on the digital twin city building convolution feature data to generate a digital twin city building convolution feature vector;

[0068] Step S643: Use the digital twin city building convolution feature vector to perform vector splicing to generate a digital twin city building feature sequence.

[0069] The present invention performs spatial pyramid pooling multi-layer sampling on the three-dimensional model of digital twin city buildings through a multi-scale sampling algorithm, and can sample and process the data of digital twin city buildings at different scales, retain different detail information and features, thereby generating high-quality convolution feature data. The multi-scale sampling algorithm can also optimize the model accuracy and reliability of digital twin city buildings, improve the modeling and analysis efficiency of digital twin city buildings, and perform convolution feature mapping on the convolution feature data of digital twin city buildings. The data of digital twin city buildings can be preliminarily processed and filtered, and the key features and information of digital twin city building data can be extracted to generate convolution feature vectors of digital twin city buildings. These feature vectors can be used in data analysis, modeling and optimization to improve the visualization effect and application benefits of digital twin city buildings. The convolution feature vectors of digital twin city buildings are used for vector splicing, and the information of different feature vectors can be fused and integrated to form a complete digital twin city building feature sequence. This feature sequence can be used in modeling, analysis and optimization of digital twin cities, etc., to provide more accurate and comprehensive data support for urban planning and management, and improve the sustainable development and intelligent operation level of digital twin cities.

[0070] In one embodiment of the present specification, a spatial data integration method and system for a digital twin city are provided, including:

[0071] at least one processor;

[0072] a memory communicatively coupled to the at least one processor;

[0073] The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the spatial data integration method applied to the digital twin city as described above.

[0074] The present invention uses a spatial data integration system applied to digital twin cities, uses satellite remote sensing equipment, laser radar and traffic sensors to obtain digital twin city spatial data and digital twin city traffic data, integrates multi-source spatial data and traffic flow data in the digital twin city, improves the accuracy and integrity of the data, and provides more reliable and comprehensive data support for urban planning and management. The deep learning algorithm and the cyclic convolution algorithm are used to realize real-time visualization of traffic simulation data in the digital twin city and intelligent processing of building three-dimensional models, so that urban data can be presented in an intuitive manner, providing support for the intelligent management and operation of the city, integrating and merging digital twin city data in different fields, constructing a digital twin smart city model, providing support and innovation for cross-domain applications of digital twin cities, realizing integration, analysis and visualization of digital twin city data, and supporting urban planning and decision-making; at the same time, the system can also realize traffic flow prediction and emergency response in the digital twin city, and provide support for emergency event handling in the city. The system can improve the accuracy and integrity of digital twin city data, provide support for the intelligent management and operation of digital twin cities, and promote the integration and innovation of cross-domain applications of digital twin cities. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 A schematic diagram of the steps of a spatial data integration method applied to a digital twin city according to the present invention;

[0076] Figure 2 Detailed implementation flow chart of step S1;

[0077] Figure 3 Detailed implementation flow chart of step S2;

[0078] Figure 4 Detailed implementation flow chart of step S3. DETAILED DESCRIPTION

[0079] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0080] This application example provides a spatial data integration method and system for digital twin cities. The execution subjects of the spatial data integration method and system for digital twin cities include but are not limited to the following: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of this application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0081] See also Figures 1 to 3The present invention provides a spatial data integration method applied to a digital twin city, the method comprising the following steps:

[0082] Step S1: using satellite remote sensing equipment, laser radar building scanners and traffic sensors to obtain digital twin city spatial data and digital twin city flow data; digital twin city data includes urban three-dimensional spatial data, urban geographic spatial data, urban building layout data and building structure spatial data; digital twin city flow data includes pedestrian flow data and road traffic flow data;

[0083] Step S2: Use the spatial matching method to integrate the spatial data of the digital twin city to generate a three-dimensional digital twin city data set; use the relevant information factor fusion algorithm to normalize and fuse the digital twin city traffic data to generate real-time traffic simulation data of the digital twin city;

[0084] Step S3: Use a spatial interpolation algorithm to interpolate and map the real-time traffic simulation data of the digital twin city to generate a traffic simulation density distribution map of the digital twin city;

[0085] Step S4: performing three-dimensional modeling on the three-dimensional digital twin city data set using a voxel modeling method based on three-dimensional modeling technology to generate a three-dimensional model of digital twin city buildings;

[0086] Step S5: using a deep learning algorithm to perform real-time interactive visualization processing on the digital twin city traffic simulation density distribution map to generate an interactive view of the digital twin city traffic real-time simulation;

[0087] Step S6: Use the circular convolution algorithm to perform dilated convolution and multi-scale sampling on the three-dimensional model of digital twin city buildings and the interactive view of digital twin city traffic real-time simulation to build a digital twin smart city model.

[0088] The present invention provides a spatial data integration method for a mathematical twin city. In order to obtain basic information of the city, including building structure, geographical location, passenger flow and road traffic flow, etc., these data are the basis of the digital twin city and the data source for subsequent processing and analysis. The spatial data of the digital twin city is integrated through a spatial matching method, and different types of data are integrated into a whole, which is convenient for subsequent data processing and analysis, and can also reduce repeated data parts and reduce the cost of data storage and processing. The relevant information factor fusion algorithm is used to normalize and fuse the traffic data of the digital twin city, and the traffic data from different sources are fused into a more complete and accurate data set. At the same time, the traffic simulation data can be updated in real time to ensure its accuracy and timeliness. The real-time traffic simulation data of the digital twin city is interpolated and mapped through a spatial interpolation algorithm to match the traffic flow data with the actual spatial distribution. Intuitively express the traffic conditions of the city, perform three-dimensional modeling of the digital twin city data set, and present the architectural structure of the digital twin city to users in a more intuitive and vivid way, so that users can better understand the architectural layout and structure of the city. Through deep learning algorithms, real-time interactive visualization of the digital twin city traffic simulation density distribution map is performed to generate an interactive view of the digital twin city traffic real-time simulation, allowing users to have a deeper understanding of the city’s traffic conditions and feel the city’s changes and development more intuitively. The generated interactive view allows users to interact in real time and understand the real-time situation of the city. Through the circular convolution algorithm, a digital twin smart city model is constructed, and all aspects of the city are integrated and optimized to generate a more complete and accurate digital twin city model, which will also provide a more accurate basis for urban planning and management, realize real-time management of virtual cities and real cities, and conduct full-method, multi-level urban supervision and command.

[0089] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of a spatial data integration method and system applied to a digital twin city of the present invention. In this example, the steps of the spatial data integration method applied to a digital twin city include:

[0090] Step S1: Use satellite remote sensing equipment, lidar building scanners and traffic sensors to obtain digital twin city spatial data and digital twin city flow data; digital twin city data includes urban three-dimensional spatial data, urban geographic spatial data, urban building layout data and building structure spatial data, and digital twin city flow data includes pedestrian flow data and road traffic flow data.

[0091] In an embodiment of the present invention, satellite remote sensing equipment is used to obtain three-dimensional spatial data and geographic spatial data of a city, including terrain features, landforms, environment, etc.; at the same time, information such as the layout and structure of buildings within the city is obtained. These data will serve as the spatial data basis of the digital twin city. A lidar building scanner is used to scan and measure the buildings in the city to obtain structural and characteristic information such as the height, facade, wing angle, opening, roof and wall of the buildings. These data are used for modeling and updating of building simulation in the digital twin city model. Traffic sensors, as an important source of traffic data for the digital twin city, can collect urban road traffic flow data, such as the driving speed, density, and traffic volume of cars and public transportation vehicles. At the same time, they can also collect pedestrian flow data, such as population density in urban areas, pedestrian flow at tourist attractions, and other information.

[0092] Step S2: Use the spatial matching method to integrate the spatial data of the digital twin city to generate a three-dimensional digital twin city data set; use the relevant information factor fusion algorithm to normalize and fuse the digital twin city traffic data to generate real-time traffic simulation data for the digital twin city.

[0093] In an embodiment of the present invention, after satellite remote sensing data, building scanning and lidar data, traffic sensor data, etc. are sorted, quality inspection and data cleaning are carried out, the digital twin city spatial data is standardized using a vector data standardization algorithm to generate digital twin city vector space benchmark data, the digital twin city vector space benchmark data is spatially integrated using a spatial matching method to generate a three-dimensional digital twin city data set, the collected pedestrian and road traffic flow data are preprocessed, including data cleaning and missing value filling, to ensure the accuracy and completeness of the data, relevant information factors such as traffic congestion level, population density, tourist activities, weather conditions, etc. are extracted from the traffic flow and pedestrian flow management data, the extracted information factors are standardized to eliminate the problem of inconsistent numerical ranges between different information factors, the normalized information factors are weighted, summarized and calculated using statistical methods, and flow data fusion is performed to generate real-time traffic simulation data for the digital twin city.

[0094] Step S3: Use the spatial interpolation algorithm to interpolate and map the real-time traffic simulation data of the digital twin city to generate a traffic simulation density distribution map of the digital twin city.

[0095] In an embodiment of the present invention, the real-time traffic simulation data of the digital twin city is organized into grid data, the interpolation method, interpolation resolution, interpolation range and other parameters are determined, and a suitable spatial interpolation algorithm is selected, such as Kriging interpolation, inverse distance weighted interpolation, radial basis function interpolation, etc. The parameter values ​​are set according to the requirements of the interpolation algorithm. For example, Kriging interpolation requires setting parameters such as the fluctuation rate factor, sample point threshold, and Kriging type. The grid data is interpolated using the interpolation algorithm and parameters to generate a traffic simulation density distribution map of the digital twin city.

[0096] Step S4: Use the voxel modeling method based on the three-dimensional modeling technology to perform three-dimensional modeling on the three-dimensional digital twin city data set to generate a three-dimensional model of the digital twin city building.

[0097] In an embodiment of the present invention, stereo pixel conversion is performed on a three-dimensional digital twin city data set to generate large three-dimensional voxels such as a digital twin city. Voxel modeling software is used to perform modeling and design based on three-dimensional data sets such as buildings, roads, and natural landscapes, and model adjustments and detail processing are performed to ensure that the model fits the real scene as closely as possible. The established three-dimensional digital twin city building model is optimized, including model simplification, material, mapping, coloring, lighting and other processing to improve the display effect and realism of the model. Virtual reality technology is used to verify the digital twin city building model, including verification of the model's accuracy, stability, fluency and other aspects.

[0098] Step S5: Use a deep learning algorithm to perform real-time interactive visualization of the digital twin city traffic simulation density distribution map to generate an interactive view of the digital twin city traffic real-time simulation.

[0099] In an embodiment of the present invention, a deep learning algorithm is used to perform trend analysis on a digital twin city traffic simulation density distribution map to generate digital twin city traffic trend data, and then a digital twin city traffic trend prediction formula is used to perform trend calculation on the digital twin city traffic trend data to generate a digital twin city traffic trend map. The traffic density distribution map is displayed and interactively presented with a designed interactive visualization interface, including real-time response to user gesture operations, real-time interaction prompts with users, etc., and real-time traffic data is continuously received and pre-processed. A deep learning algorithm is used to perform real-time prediction and update the traffic density distribution map to maintain the real-time nature of the interactive view of the real-time simulation of digital twin city traffic.

[0100] Step S6: Use the circular convolution algorithm to perform dilated convolution and multi-scale sampling on the three-dimensional model of digital twin city buildings and the interactive view of digital twin city traffic real-time simulation to build a digital twin smart city model.

[0101] In an embodiment of the present invention, a suitable recurrent convolution algorithm is selected, such as a recurrent convolutional neural network (RCNN) based on a graph convolutional neural network (GCN), and the three-dimensional model of digital twin city buildings and the interactive view of digital twin city traffic real-time simulation are subjected to dilated convolution processing to extract local spatial properties. Sampling algorithms of different scales are used to perform multi-scale sampling processing on the three-dimensional model of digital twin city buildings and the interactive view of digital twin city traffic real-time simulation to extract global spatial properties. The local spatial properties and the global spatial properties are integrated through a fusion algorithm to obtain a digital twin smart city model.

[0102] In the embodiment of the present invention, reference Figure 2 The above is a flowchart of the detailed implementation steps of step S1. In one embodiment of the present specification, the detailed implementation steps of step S1 include:

[0103] Step S11: Use high-precision and high-resolution satellite remote sensing equipment to cruise and scan the city to obtain digital twin city spatial data; the digital twin city multi-source data includes urban three-dimensional spatial data, urban geographic spatial data and urban building layout data;

[0104] Step S12: The laser radar building scanner performs laser scanning on the internal structure of the urban building to generate building structure space data, where the building structure space data includes building shape and size data, building elevation data, building internal structure data and building material data;

[0105] Step S13: Use traffic sensors to obtain digital twin city traffic data, where the digital twin city traffic data includes pedestrian traffic data and road traffic flow data.

[0106] The present invention uses high-precision and high-resolution satellite remote sensing equipment to perform cruise scanning, so as to obtain the actual form and layout of the city. By acquiring digital twin city spatial data, it can assist in urban spatial planning and management, and improve the accuracy and scientific nature of urban planning. By using a lidar building scanner to perform laser scanning of the internal structure of a building, detailed information such as the internal structure, facilities and decoration of the building can be obtained. This information helps to strengthen the management and maintenance of the building, and also provides basic data for subsequent building model construction and simulation. By acquiring digital twin city flow data through traffic sensors, information such as pedestrian flow and road traffic flow can be obtained, which is very important for urban traffic management and traffic planning, and can help analyze urban traffic hotspots and bottlenecks and improve the level of urban traffic intelligence.

[0107] In an embodiment of the present invention, a suitable high-precision satellite remote sensing device is selected to obtain satellite data that can provide high-resolution images, a cruise scan is performed on the city range, and data processing is performed to make it available spatial data. The data is synthesized into a continuous, fast-access three-dimensional environmental model by interpolation or other methods. A suitable laser radar building scanner device is selected to perform a scanning operation and obtain laser point cloud data. The scanned data is denoised and screened to remove generated noise and useless data. An algorithm is applied to process the data to obtain the three-dimensional structural spatial data of the building (including shape and size data, elevation data, indoor / outdoor structure data, material data, etc.). Sensor equipment is arranged or installed to collect traffic data in the city. The sensor can be an intersection monitoring camera, a GPS tracking device, an application traffic monitoring, etc. The data is processed to extract the required pedestrian flow data and road traffic flow data.

[0108] In the embodiment of the present invention, reference Figure 3 The above is a flowchart of the detailed implementation steps of step S2. In one embodiment of the present specification, the detailed implementation steps of step S2 include:

[0109] Step S21: using a vector data standardization algorithm to standardize the digital twin city spatial data to generate digital twin city vector space benchmark data;

[0110] Step S22: using a spatial matching method to perform spatial data integration on the digital twin city vector space benchmark data to generate a three-dimensional digital twin city data set;

[0111] Step S23: Use the principal component analysis method to perform principal component dimensionality reduction extraction on the digital twin city traffic data to obtain relevant factor data;

[0112] Step S24: using the correlation factor weight analysis method to analyze the correlation factor coefficients of the correlation factor data, and generating standard related information factors;

[0113] Step S25: normalizing the digital twin city flow data using the maximum-minimum normalization method to generate digital twin city flow range data;

[0114] Step S26: Based on the standard relevant information factors, the relevant information factor fusion algorithm is used to fuse the digital twin city traffic range data to generate real-time traffic simulation data for the digital twin city.

[0115] The present invention standardizes the spatial data of the digital twin city through a vector data standardization algorithm, and can process data from different sources for consistency to ensure the consistency and standardization of the data. After the digital twin city vector space benchmark data is generated, the complexity of subsequent calculations and processing can be reduced, and the accuracy and reliability of the digital twin city model can be improved. The spatial data integration of the digital twin city vector space benchmark data is performed using the spatial matching method, and data from different sources can be integrated to generate a three-dimensional digital twin city data set, which can better reflect the spatial state and spatial distribution characteristics of the city, and provide more complete and comprehensive data support for the digital twin city model and analysis. The principal component analysis method is used to perform principal component dimensionality reduction extraction on the digital twin city traffic data, and the main traffic flow factors can be extracted from a large amount of traffic flow data, which simplifies the complexity of the data and makes the data easier to process and understand. , the correlation factor weight analysis method is used to analyze the correlation factor coefficients of the correlation factor data, and standard relevant information factors are generated. The correlation between various traffic flow factors can be deeply analyzed, and the role relationship and influence degree between various factors can be better revealed, providing a more refined data basis for subsequent data processing and analysis. The maximum-minimum normalization method is used to normalize the digital twin city traffic data, which can unify the data of various traffic factors to the same scale, avoiding the dimensional difference of the data, making the comparison between different factors more accurate and reliable. Based on the standard relevant information factors, the relevant information factor fusion algorithm is used to fuse the digital twin city traffic range data, which can fuse the data of various traffic factors to generate more comprehensive and accurate digital twin city real-time traffic simulation data, providing a more scientific and effective decision-making basis for subsequent traffic management and planning.

[0116] In an embodiment of the present invention, a suitable vector data standardization algorithm is selected, such as linear transformation, standard score, etc., to process the spatial data of the digital twin city to make it conform to the standard vector data format, generate the vector space benchmark data of the digital twin city, process the vector space benchmark data of the digital twin city to make it conform to the requirements of the three-dimensional digital twin city data set, integrate the data through the spatial matching method, generate the three-dimensional data set of the digital twin city, use the principal component analysis algorithm (PCA) to process the data, perform principal component analysis on the digital twin city traffic data, extract a small number of principal components, reduce the dimension of the data, and obtain the correlation factor data, process the correlation factor data using the correlation factor weight analysis method, select a suitable analysis method, such as the hierarchical analysis method (AHP) and the measure function method, generate standard related information factors to reflect the relative importance of different factors, and use the maximum-minimum normalization method (Max-Min Normalization) is used to normalize the traffic data of the digital twin city so that the traffic data range is between 0 and 1. The traffic data from different sources are processed with a unified standard so that they can be compared and referenced. Appropriate relevant information factor fusion algorithms are selected, such as comprehensive evaluation method, grey correlation method, etc. Standard relevant information factors and normalized digital twin city traffic range data are used to generate real-time traffic simulation data of the digital twin city.

[0117] In the embodiment of the present invention, reference Figure 4 The above is a flowchart of the detailed implementation steps of step S3. In one embodiment of the present specification, the detailed implementation steps of step S3 include:

[0118] Step S31: preprocessing the digital twin city real-time traffic simulation data to generate a digital twin city real-time traffic preprocessing pipeline;

[0119] Step S32: using a matrix decomposition method to perform matrix division on the real-time traffic preprocessing pipeline of the digital twin city to generate a traffic matrix diagram of the digital twin city;

[0120] Step S33: using a feature point detection algorithm to mark feature points on the digital twin city traffic matrix diagram to generate a digital twin city traffic flow coordinate point grid;

[0121] Step S34: using the point sampling interpolation of the spatial interpolation algorithm to perform weighted average calculation on the digital twin city traffic flow coordinate point grid to obtain the digital twin city traffic simulation interpolation points;

[0122] Step S35: Based on the digital twin city traffic simulation interpolation points, the urban traffic flow spatial interpolation distance weight calculation formula is used to interpolate and map the digital twin city traffic flow coordinate point grid to generate a digital twin city traffic simulation density distribution map.

[0123] The present invention performs data preprocessing on the real-time traffic simulation data of the digital twin city, thereby removing noise and redundant information in the data, ensuring the accuracy and reliability of the traffic data of the digital twin city, and generating a real-time traffic preprocessing pipeline for the digital twin city at the same time, thereby improving the efficiency and accuracy of subsequent processing and calculation, and using the matrix decomposition method to perform matrix division on the real-time traffic preprocessing pipeline for the digital twin city, thereby dividing the traffic data according to different factors, thereby simplifying the processing and calculation of the data, and generating a traffic matrix diagram for the digital twin city, thereby more intuitively displaying the traffic relationship between different road sections and regions, and providing basic data for subsequent traffic analysis and simulation, and using a feature point detection algorithm to mark feature points on the traffic matrix diagram for the digital twin city, thereby identifying and marking key points and key factors in the traffic data, thereby facilitating subsequent data analysis and processing, and generating a traffic flow coordinate point grid for the digital twin city, It can more clearly reflect the traffic flow distribution and rules in different regions and sections, and use the point sampling interpolation of the spatial interpolation algorithm to perform weighted average calculation on the traffic flow coordinate point grid of the digital twin city. The weighted average of the traffic flow in different regions can be counted to further reflect the overall characteristics of the digital twin city traffic data. After obtaining the digital twin city traffic simulation interpolation points, the traffic flow of the digital twin city can be simulated more accurately. Based on the digital twin city traffic simulation interpolation points, the urban traffic flow spatial interpolation distance weight calculation formula is used to interpolate and map the digital twin city traffic flow coordinate point grid. The digital twin city traffic flow data can be fitted from the local to the whole, and a digital twin city traffic simulation density distribution map can be generated. This can provide a more comprehensive and intuitive understanding of the traffic conditions and congestion conditions in different regions and sections, and provide more scientific and comprehensive data support for the traffic planning and management of the digital twin city.

[0124] In an embodiment of the present invention, data preprocessing is performed on the digital twin city real-time traffic simulation data, including data cleaning, deduplication, format conversion, etc., and a digital twin city real-time traffic preprocessing pipeline is designed. According to actual needs, including data cleaning, preprocessing, feature extraction and other processing steps, a complete set of digital twin city real-time traffic preprocessing processes is formed, and the digital twin city real-time traffic preprocessing pipeline is executed to generate a preprocessed digital twin city real-time traffic simulation data set, and matrix decomposition processing is performed on the digital twin city real-time traffic simulation data set generated by the digital twin city real-time traffic preprocessing pipeline. The digital twin city real-time traffic simulation data set is matrix-partitioned using the matrix decomposition method, and each traffic element is decomposed into matrices of multiple dimensions. According to the decomposed matrix, a digital twin city traffic matrix diagram is generated, and feature point detection is performed on the digital twin city traffic matrix diagram to mark the For locations where traffic flow is concentrated, such as road intersections and main roads, the feature point marking information is used to generate a digital twin city traffic flow coordinate point grid, and a point sampling interpolation algorithm is performed on the digital twin city traffic flow coordinate point grid. The traffic flow information of the surrounding points is used to perform a weighted average calculation on each point to obtain the traffic flow information of the point. The digital twin city traffic simulation interpolation points are used to interpolate and map the digital twin city traffic flow coordinate point grid based on the urban traffic flow spatial interpolation distance weight calculation formula. On the digital twin city traffic flow coordinate point grid, according to the interpolation mapping results, a digital twin city traffic simulation density distribution map is generated to reflect the urban traffic flow distribution.

[0125] In one embodiment of the present specification, the urban traffic flow spatial interpolation distance weight calculation formula in step S35 is specifically:

[0126]

[0127] Among them, W i is the interpolation weight of the i-th sampling point, w pop,i is the flow of people at the i-th sampling point, w traf,i is the road traffic volume at the i-th sampling point, q pop,i is the flow weight value of the i-th sampling point, q traf,i is the road traffic flow weight value of the i-th sampling point, θ i is the azimuth between the i-th sampling point and the point to be interpolated, d i is the distance between the i-th sampling point and the point to be interpolated, p is the attenuation exponent, and λ is the attenuation rate.

[0128] The present invention is achieved by pop,i +wtraf,i Calculate the data of road vehicle and pedestrian flow through Calculate the weighted values ​​of the pedestrian flow and road traffic flow at the i-th sampling point, taking into account the weights of the population flow and road traffic flow: By incorporating the weights of the population flow and road traffic flow at the sampling point into the calculation, the traffic conditions in different areas can be better reflected. This can effectively avoid the situation of simple averaging or weighted averaging of the entire city, thereby improving the accuracy of traffic flow estimation. The cosine function and distance decay function are used to make the weight of the sampling points with a longer distance smaller and the weight of the sampling points with a shorter distance larger, so as to make the interpolation calculation more reasonable. The azimuth and distance weights can accurately describe the spatial relationship between the interpolation points and the sampling points, and take into account factors such as distance and direction in space, so as to make the interpolation results more accurate and reliable. p and λ are two adjustable parameters used to adjust the influence of distance. The larger p is, the faster the influence of the sampling points on the estimated value decreases as the distance increases; the larger λ is, the slower the influence of the sampling points on the estimated value decreases as the distance increases. The weight function uses the cosine function and distance decay function to make the weight of the sampling points with a longer distance smaller and the weight of the sampling points with a shorter distance larger. The role of the cosine function is to adjust the weight according to the direction angle, and the role of the distance attenuation function is to adjust the weight according to the distance. Through the normalization of the weights, the influence of the difference in the number of sampling points on the interpolation results is avoided, making the interpolation results more stable and reliable. Taking into account the flow of people and road traffic, it more comprehensively reflects the spatial changes in urban traffic flow, and can provide useful data support for urban traffic planning and management.

[0129] In one embodiment of this specification, the specific steps of step S33 are:

[0130] Step S331: Detect feature points of the digital twin city traffic matrix diagram using a feature point detection algorithm to generate initial digital twin city traffic feature point data;

[0131] Step S332: Filter and screen the digital twin city traffic feature point data using the characteristic value threshold limiting method to generate benchmark digital twin city traffic feature point data;

[0132] Step S333: Marking the traffic feature point data of the benchmark digital twin city by using the spatial three-dimensional coordinate method to generate the traffic feature point coordinates of the digital twin city;

[0133] Step S334: Use the equidistant grid algorithm to grid the coordinates of the digital twin city traffic feature points to generate a digital twin city traffic flow coordinate point grid.

[0134] The present invention performs feature point detection on the digital twin city traffic matrix diagram through a feature point detection algorithm, and can automatically identify and mark key feature points in the digital twin city traffic data, such as intersections, crossroads, etc. After generating the initial digital twin city traffic feature point data, it can provide more information and features for the subsequent digital twin city traffic data analysis. The digital twin city traffic feature point data is filtered and screened by the feature value threshold limitation method, which can exclude some irrelevant or very small feature points, thereby improving the accuracy and efficiency of the digital twin city traffic analysis. After generating the benchmark digital twin city traffic feature point data, it can provide more accurate and meaningful data support for the subsequent generation of digital twin city traffic coordinate point grids. The benchmark digital twin city traffic is filtered and screened by the spatial three-dimensional coordinate method. By marking the feature point data, the location information of the digital twin city traffic feature points can be located in three-dimensional coordinates, which can more accurately express the location and distribution of the digital twin city traffic data feature points in the real world. After the digital twin city traffic feature point coordinates are generated, it can provide more accurate and meaningful data support for the generation of the digital twin city traffic coordinate point grid. The digital twin city traffic feature point coordinates are gridded using the equidistant grid algorithm, and the digital twin city traffic feature point data can be discretized into grid data, which is convenient for subsequent digital twin city traffic data analysis and processing. After the digital twin city traffic flow coordinate point grid is generated, it will provide more comprehensive, accurate and intuitive basic data support for digital twin city traffic simulation and analysis.

[0135] In an embodiment of the present invention, a suitable feature point detection algorithm is selected, such as SIFT (Scale-Invariant Feature Transform), and feature point detection is performed on the digital twin city traffic matrix diagram to obtain initial digital twin city traffic feature point data. The initial digital twin city traffic feature point data is filtered and screened using the eigenvalue threshold limitation method to remove feature points of low quality to obtain benchmark digital twin city traffic feature point data. A suitable eigenvalue threshold is selected, and the size and quality of the screening result are controlled by adjusting the threshold. The benchmark digital twin city traffic feature point data is marked with feature points using the spatial three-dimensional coordinate method so that each feature point can determine its position information in three-dimensional space. The digital twin city traffic feature point coordinate data is generated according to the coordinates of the feature points in the matrix diagram and the spatial three-dimensional coordinates of the digital twin city. The digital twin city traffic feature point coordinate data is gridded using the equidistant grid algorithm to map the feature point coordinates to a regular grid. The traffic flow size in each grid is determined according to the number of feature points on the grid, and is used as the attribute of the digital twin city traffic flow coordinate point grid.

[0136] In one embodiment of this specification, the specific steps of step S4 are:

[0137] Step S41: performing stereo pixel conversion on the three-dimensional digital twin city data set to generate large three-dimensional voxels such as the digital twin city;

[0138] Step S42: gridding a large three-dimensional voxel such as a digital twin city using an inverse distance weighted method to generate a digital twin city grid model;

[0139] Step S43: Use the voxel modeling method based on the three-dimensional modeling technology to perform three-dimensional modeling on the digital twin city grid model to generate a three-dimensional model of the digital twin city buildings.

[0140] The present invention converts three-dimensional pixels of a three-dimensional digital twin city data set into three-dimensional models, which can more realistically express the spatial structure and characteristics of the digital twin city. After generating equal-sized three-dimensional voxels, each voxel has the same size, which is convenient for calculation and modeling in subsequent processing. The inverse distance weighted method is used to grid the large three-dimensional voxels of the digital twin city, and the three-dimensional data can be discretized into grid points, which is convenient for subsequent processing and modeling. The inverse distance weighted method can effectively convert the three-dimensional voxel information into a grid model, retaining the morphology and characteristics of the digital twin city as much as possible. Features, making the subsequent digital twin city building three-dimensional model more accurate and reliable. The digital twin city grid model is three-dimensionally modeled using the voxel modeling method based on three-dimensional modeling technology, which can convert the digital twin city data into a real three-dimensional building model. The voxel modeling method based on three-dimensional modeling technology can convert the digital twin city building features from discrete grid points into a real three-dimensional building model, so that the form and structure of the digital twin city can be more realistically expressed and simulated. At the same time, the voxel modeling method based on three-dimensional modeling technology can also support rapid model updates and expansions, making the modeling of digital twin cities more efficient and flexible.

[0141] In an embodiment of the present invention, a three-dimensional digital twin city data set is processed, and the three-dimensional entity is divided into equal-sized cubic blocks to generate large three-dimensional voxels such as the digital twin city. The size of the voxels is adjusted as needed to adapt to different modeling accuracy and calculation requirements. The large three-dimensional voxels such as the digital twin city are gridded using the inverse distance weighted method and converted into a three-dimensional grid model. The digital twin city grid model is three-dimensionally modeled using a voxel modeling method based on three-dimensional modeling technology to generate a three-dimensional model of a digital twin city building. Various model editing software, such as 3ds Max, SketchUp, etc., can be used during modeling to establish, edit and optimize the model. When modeling, the structure, appearance, texture and other aspects of the digital twin city buildings need to be considered to reproduce the architectural appearance of the digital twin city as realistically as possible. After the modeling is completed, the model can be rendered and displayed to show different perspectives and characteristics of the digital twin city buildings.

[0142] In one embodiment of this specification, the specific steps of step S5 are:

[0143] Step S51: using a deep learning algorithm to perform trend analysis on the digital twin city traffic simulation density distribution map to generate digital twin city traffic trend data;

[0144] Step S52: performing trend calculation on the digital twin city traffic trend data using the digital twin city traffic trend prediction formula to generate a digital twin city traffic trend graph;

[0145] Step S53: using the digital twin city traffic simulation density distribution map and the digital twin city traffic trend map to perform real-time simulation analysis to generate a digital twin city traffic real-time simulation map;

[0146] Step S54: performing data visualization processing on the real-time simulation diagram of digital twin city traffic to generate a real-time simulation visualization view of digital twin city traffic;

[0147] Step S55: using a JavaScript library to dynamically and interactively process the visualization view of the real-time simulation of traffic in the digital twin city, and generate an interactive view of the real-time simulation of traffic in the digital twin city;

[0148] The present invention uses a deep learning algorithm to perform trend analysis on the digital twin city traffic simulation density distribution map, and can conduct in-depth exploration and analysis of the traffic conditions of the digital twin city. Through trend analysis, digital twin city traffic trend data can be generated, including information such as traffic bottlenecks, peak hours and congestion conditions. These data can guide digital twin city traffic operations and planning, improve urban traffic operation efficiency and safety, and use the digital twin city traffic trend prediction formula to perform trend calculation on the digital twin city traffic trend data, which can predict the future traffic trends and congestion conditions of the digital twin city. The traffic condition information in the digital twin city traffic trend map can be used to study traffic planning and deployment, predict possible future congestion conditions, prepare traffic emergency plans, etc. The digital twin city traffic simulation density distribution map and the digital twin city traffic trend map are used for real-time simulation analysis, which can monitor the traffic status of the digital twin city in real time and respond and cope with emergencies in a timely manner. The real-time traffic simulation diagram of the digital twin city can predict the traffic conditions in certain areas, help city managers optimize routes and resource allocation, and solve the problem of urban traffic congestion. Data visualization processing of the real-time traffic simulation diagram of the digital twin city can visualize the traffic conditions of the digital twin city and help users quickly obtain and understand traffic information. The real-time simulation visualization view of the digital twin city traffic makes data easier for people to understand and apply through visual charts, images, and maps, making the operation of urban traffic more efficient and safe. The real-time simulation visualization view of the digital twin city traffic is dynamically and interactively processed using the JavaScript library, which can realize the dynamic interactive presentation of the digital twin city traffic simulation. This interactive view of the real-time simulation of the digital twin city traffic can help users have a deeper understanding of the traffic conditions of the digital twin city, including traffic flow, congestion, etc., so as to optimize urban traffic planning, real-time scheduling and prediction in a targeted manner.

[0149] In an embodiment of the present invention, a trend analysis is performed on a digital twin city traffic simulation density distribution map, and a deep learning algorithm is used to explore the rules and trend information therein. Data sorting and cleaning are performed based on the digital twin city traffic trend data obtained by the analysis to ensure the accuracy and completeness of the trend data. A digital twin city traffic trend prediction formula is used to perform trend calculation on the digital twin city traffic trend data to generate a digital twin city traffic trend map. Based on the digital twin city traffic trend map, the development trend of digital twin city traffic is understood. The digital twin city traffic simulation density distribution map and the digital twin city traffic trend map are used for real-time simulation, and analysis is performed in combination with real-time traffic data. The results of real-time simulation can reflect the traffic conditions and changes in traffic flow in the digital twin city. The real-time simulation diagram of the digital twin city traffic is visualized, and the real-time simulation results are presented as a visualization view to facilitate user interactive operations. The visualization method can adopt a variety of technologies, such as computer graphics, virtual reality technology, etc., to present a more realistic digital twin city traffic simulation scene. The JavaScript library is used to dynamically and interactively process the real-time simulation visualization view of the digital twin city traffic to achieve multiple functions, including scene switching, traffic simulation parameter setting, interactive query, etc., so that users can feel the traffic conditions of the digital twin city more intuitively.

[0150] In one embodiment of the present specification, the digital twin city traffic trend prediction formula in step S52 is specifically:

[0151]

[0152] Where x is the spatial position vector area value, t is the time period, f(x,t) is the traffic flow prediction trend density of the spatial position vector area value of x in the time period t, and w i (t) is the trend influence weight of the i-th sampling point in time period t, N pop,i is the flow threshold of the i-th sampling point, N traf,i is the traffic flow threshold of the i-th sampling point, u i is the mean vector of the i-th sampling point in space.

[0153] The present invention is achieved by i (t) Taking the spatial position vector area value x and time t as prediction conditions, it represents the traffic flow trend density distribution of the predicted target area and time period. Among the n sampling points in the same time period, the weight w of each sampling point is calculated. i (t), the weight reflects the impact of the current moment on traffic flow. The weight calculation can take into account factors such as historical data and current weather. Common methods include linear regression and support vector regression. pop,i +Ntraf,i , for each sampling point i, its flow threshold N pop,i and traffic flow threshold N traf,i It represents the density level of population and traffic flow at the location of the point, and is used to measure the impact of the point on traffic flow. For each sampling point i, its spatial position mean vector represents the spatial location attribute of the point, which can be obtained using cluster analysis and other methods. The Gaussian kernel function is used to obtain a distribution probability density function about the location coordinates, where is the relative position deviation, is the probability density function of the Gaussian distribution, and the other e -t is the attenuation coefficient in the time dimension. The influence of each sampling point is added up by weighted summation to obtain the traffic flow trend density distribution of the position vector area value in time period t. The advantage of this method is that it can take into account the influence of multiple factors such as population flow, vehicle flow and spatial distribution on traffic flow, so the prediction effect is more accurate.

[0154] In one embodiment of this specification, the specific steps of step S6 are:

[0155] Step S61: using a superpixel convolutional network to perform convolution preprocessing on the interactive view of the real-time simulation of digital twin city traffic to generate a feature sample cluster of the real-time simulation of digital twin city traffic;

[0156] Step S62: using a circular convolution algorithm to perform convolution data cutting on the digital twin city traffic real-time simulation feature sample cluster to generate a digital twin city traffic real-time simulation feature network;

[0157] Step S63: using the dilated convolution algorithm to perform edge feature enhancement processing on the digital twin city traffic real-time simulation feature network to generate a digital twin city traffic real-time simulation feature map;

[0158] Step S64: using a multi-scale sampling algorithm to perform spatial pyramid pooling multi-layer sampling on the digital twin city building three-dimensional model to generate a digital twin city building feature sequence;

[0159] Step S65: Use the association rule algorithm to perform data mining modeling on the digital twin city building feature sequence and the digital twin city traffic real-time simulation feature map to build a digital twin smart city model.

[0160] The present invention performs convolution preprocessing on the interactive view of the real-time simulation of digital twin city traffic through a superpixel convolution network, which can effectively reduce data redundancy and distribute data to different pixel areas. The superpixel convolution network can further extract the feature information of the real-time simulation image of digital twin city traffic, quickly locate and identify information such as traffic objects and vehicles, thereby providing a high-quality data basis for subsequent processing. The circular convolution algorithm is used to perform convolution data cutting on the feature sample clusters of the real-time simulation of digital twin city traffic, which can effectively compress and classify the original data, and perform complex calculations based on the data features. The circular convolution algorithm can retain the spatial information and connection relationship of the original data, and can also adaptively perform data learning and model optimization to improve the accuracy and reliability of the model. The dilated convolution algorithm is used to perform edge feature enhancement processing on the feature network of the real-time simulation of digital twin city traffic, which can further extract edge features in the feature map of the real-time simulation of digital twin city traffic, such as lane markings, traffic signs, etc. Thereby further improving the accuracy and reliability of digital twin city traffic simulation. The dilated convolution algorithm can also effectively reduce shadow interference and edge blur problems, and improve the visual effect of digital twin city traffic simulation. The multi-scale sampling algorithm is used to perform spatial pyramid pooling and multi-layer sampling on the three-dimensional model of digital twin city buildings, which can retain the information of digital twin city buildings at different scales, thereby improving the high definition and accuracy of digital twin city simulation. The multi-scale sampling algorithm can also detect the spatial distribution and morphological characteristics of digital twin city buildings to optimize the spatial design and planning of buildings. The association rule algorithm is used to perform data mining modeling on the digital twin city building feature sequence and the digital twin city traffic real-time simulation feature map. Through the correlation analysis of the digital twin city building feature sequence and the digital twin city traffic real-time simulation feature map, a digital twin smart city model is established. This model can provide more accurate data support and analysis for city managers, and support the smart operation and sustainable development of digital twin cities.

[0161] In an embodiment of the present invention, a superpixel convolution network is used to process an interactive view of a real-time simulation of traffic in a digital twin city to generate a feature sample cluster of a real-time simulation of traffic in a digital twin city. The superpixel convolution network can better process complex structures and irregular shapes in traffic scenes, thereby improving the accuracy and robustness of the algorithm. The edge feature of the feature network of the real-time simulation of traffic in a digital twin city is enhanced by using an expansion convolution algorithm to generate a feature map of real-time simulation of traffic in a digital twin city. The edge feature of the feature network is enhanced by using an expansion convolution algorithm to better capture edge information and improve the edge detection effect of the algorithm. The three-dimensional model of a building in a digital twin city is subjected to spatial pyramid pooling multi-layer sampling by using a multi-scale sampling algorithm. Through the multi-scale sampling algorithm, multiple features of the building in a digital twin city can be extracted at different levels to generate a feature sequence of a building in a digital twin city. The feature sequence of the building in a digital twin city and the feature map of real-time simulation of traffic in a digital twin city are subjected to data mining modeling by using an association rule algorithm. Through the association rule algorithm, the correlation between the feature sequence of the building in a digital twin city and the feature map of real-time simulation of traffic can be modeled to construct a digital twin smart city model.

[0162] In one embodiment of this specification, the specific steps of step S64 are:

[0163] Step S641: Use a multi-scale sampling algorithm to perform spatial pyramid pooling multi-layer sampling on the digital twin city building three-dimensional model to generate digital twin city building convolution feature data;

[0164] Step S642: performing convolution feature mapping on the digital twin city building convolution feature data to generate a digital twin city building convolution feature vector;

[0165] Step S643: Use the digital twin city building convolution feature vector to perform vector splicing to generate a digital twin city building feature sequence.

[0166] The present invention performs spatial pyramid pooling multi-layer sampling on the three-dimensional model of digital twin city buildings through a multi-scale sampling algorithm, and can sample and process the data of digital twin city buildings at different scales, retain different detail information and features, thereby generating high-quality convolution feature data. The multi-scale sampling algorithm can also optimize the model accuracy and reliability of digital twin city buildings, improve the modeling and analysis efficiency of digital twin city buildings, and perform convolution feature mapping on the convolution feature data of digital twin city buildings. The data of digital twin city buildings can be preliminarily processed and filtered, and the key features and information of digital twin city building data can be extracted to generate convolution feature vectors of digital twin city buildings. These feature vectors can be used in data analysis, modeling and optimization to improve the visualization effect and application benefits of digital twin city buildings. The convolution feature vectors of digital twin city buildings are used for vector splicing, and the information of different feature vectors can be fused and integrated to form a complete digital twin city building feature sequence. This feature sequence can be used in modeling, analysis and optimization of digital twin cities, etc., to provide more accurate and comprehensive data support for urban planning and management, and improve the sustainable development and intelligent operation level of digital twin cities.

[0167] In one embodiment of the present specification, a spatial data integration method and system for a digital twin city are provided, comprising:

[0168] at least one processor;

[0169] a memory communicatively coupled to the at least one processor;

[0170] The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the spatial data integration method applied to the digital twin city as described above.

[0171] The present invention uses a spatial data integration system applied to digital twin cities, uses satellite remote sensing equipment, laser radar and traffic sensors to obtain digital twin city spatial data and digital twin city traffic data, integrates multi-source spatial data and traffic flow data in the digital twin city, improves the accuracy and integrity of the data, and provides more reliable and comprehensive data support for urban planning and management. The deep learning algorithm and the cyclic convolution algorithm are used to realize real-time visualization of traffic simulation data in the digital twin city and intelligent processing of building three-dimensional models, so that urban data can be presented in an intuitive manner, providing support for the intelligent management and operation of the city, integrating and merging digital twin city data in different fields, constructing a digital twin smart city model, providing support and innovation for cross-domain applications of digital twin cities, realizing integration, analysis and visualization of digital twin city data, and supporting urban planning and decision-making; at the same time, the system can also realize traffic flow prediction and emergency response in the digital twin city, and provide support for emergency event handling in the city. The system can improve the accuracy and integrity of digital twin city data, provide support for the intelligent management and operation of digital twin cities, and promote the integration and innovation of cross-domain applications of digital twin cities.

[0172] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0173] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0174] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A spatial data integration method applied to digital twin cities, characterized in that: The following steps are involved: Step S1: using satellite remote sensing equipment, laser radar and traffic sensors to obtain digital twin city spatial data and digital twin city traffic data; Digital twin city data includes urban three-dimensional spatial data, urban geographic spatial data, urban building layout data and building structure spatial data; digital twin city traffic data includes pedestrian flow data and road traffic flow data; Step S2: Using the spatial matching method to integrate the spatial data of the digital twin city to generate a three-dimensional digital twin city data set; The digital twin city traffic data is normalized and fused using the relevant information factor fusion algorithm to generate real-time traffic simulation data of the digital twin city, wherein the relevant information factor fusion algorithm includes a comprehensive evaluation method or a grey correlation method; Step S3: Use a spatial interpolation algorithm to interpolate and map the real-time traffic simulation data of the digital twin city to generate a traffic simulation density distribution map of the digital twin city; Step S4: performing three-dimensional modeling on the three-dimensional digital twin city data set using a voxel modeling method based on three-dimensional modeling technology to generate a three-dimensional model of digital twin city buildings; Step S5: Use a deep learning algorithm to perform real-time interactive visualization processing on the digital twin city traffic simulation density distribution map to generate a real-time interactive view of the digital twin city traffic simulation; the specific steps of step S5 are: Step S51: using a deep learning algorithm to perform trend analysis on the digital twin city traffic simulation density distribution map to generate digital twin city traffic trend data; Step S52: performing trend calculation on the digital twin city traffic trend data using the digital twin city traffic trend prediction formula to generate a digital twin city traffic trend graph; Step S53: using the digital twin city traffic simulation density distribution map and the digital twin city traffic trend map to perform real-time simulation analysis to generate a digital twin city traffic real-time simulation map; Step S54: performing data visualization processing on the real-time simulation diagram of digital twin city traffic to generate a real-time simulation visualization view of digital twin city traffic; Step S55: using a JavaScript library to dynamically and interactively process the visualization view of the real-time simulation of traffic in the digital twin city, and generate an interactive view of the real-time simulation of traffic in the digital twin city; The digital twin city traffic trend prediction formula in step S52 is specifically: Where x is the spatial position vector area value, t is the time period, f(x,t) is the traffic flow prediction trend density of the spatial position vector area value of x in the time period t, and w i (t) is the trend influence weight of the i-th sampling point in time period t, N pop,i is the flow threshold of the i-th sampling point, N traf,i is the traffic flow threshold of the i-th sampling point, u i is the mean vector of the spatial position of the i-th sampling point; Step S6: Use the circular convolution algorithm to perform dilated convolution and multi-scale sampling on the three-dimensional model of digital twin city buildings and the interactive view of real-time simulation of digital twin city traffic to build a digital twin smart city model; the specific steps of step S6 are: Step S61: using a superpixel convolutional network to perform convolution preprocessing on the interactive view of the real-time simulation of digital twin city traffic, and generate a feature sample cluster of the real-time simulation of digital twin city traffic; Step S62: using a circular convolution algorithm to perform convolution data cutting on the digital twin city traffic real-time simulation feature sample cluster to generate a digital twin city traffic real-time simulation feature network; Step S63: using the dilated convolution algorithm to perform edge feature enhancement processing on the digital twin city traffic real-time simulation feature network to generate a digital twin city traffic real-time simulation feature map; Step S64: using a multi-scale sampling algorithm to perform spatial pyramid pooling multi-layer sampling on the digital twin city building three-dimensional model to generate a digital twin city building feature sequence; Step S65: Use the association rule algorithm to perform data mining modeling on the digital twin city building feature sequence and the digital twin city traffic real-time simulation feature map to build a digital twin smart city model.

2. The method according to claim 1, characterized in that The specific steps of step S1 are: Step S11: Use high-precision and high-resolution satellite remote sensing equipment to cruise and scan the city to obtain digital twin city spatial data; the digital twin city multi-source data includes urban three-dimensional spatial data, urban geographic spatial data and urban building layout data; Step S12: using laser radar to perform laser scanning on the internal structure of urban buildings to generate building structure space data, the building structure space data includes building shape and size data, building elevation data, building internal structure data and building material data; Step S13: Use traffic sensors to obtain digital twin city traffic data, where the digital twin city traffic data includes pedestrian traffic data and road traffic flow data.

3. The method according to claim 2, characterized in that The specific steps of step S2 are: Step S21: using a vector data standardization algorithm to standardize the digital twin city spatial data to generate digital twin city vector space benchmark data; Step S22: using a spatial matching method to perform spatial data integration on the digital twin city vector space benchmark data to generate a three-dimensional digital twin city data set; Step S23: Use the principal component analysis method to perform principal component dimensionality reduction extraction on the digital twin city traffic data to obtain relevant factor data; Step S24: using the correlation factor weight analysis method to analyze the correlation factor coefficients of the correlation factor data, and generating standard related information factors; Step S25: normalizing the digital twin city flow data using the maximum-minimum normalization method to generate digital twin city flow range data; Step S26: Based on standard relevant information factors, the digital twin city traffic range data is fused using the relevant information factor fusion algorithm to generate real-time traffic simulation data for the digital twin city.

4. The method according to claim 3, characterized in that The specific steps of step S3 are: Step S31: preprocessing the digital twin city real-time traffic simulation data to generate a digital twin city real-time traffic preprocessing pipeline; Step S32: using a matrix decomposition method to perform matrix division on the real-time traffic preprocessing pipeline of the digital twin city to generate a traffic matrix diagram of the digital twin city; Step S33: using a feature point detection algorithm to mark feature points on the digital twin city traffic matrix diagram to generate a digital twin city traffic flow coordinate point grid; Step S34: using the point sampling interpolation of the spatial interpolation algorithm to perform weighted average calculation on the digital twin city traffic flow coordinate point grid to obtain the digital twin city traffic simulation interpolation points; Step S35: Based on the digital twin city traffic simulation interpolation points, the urban traffic flow spatial interpolation distance weight calculation formula is used to interpolate and map the digital twin city traffic flow coordinate point grid to generate a digital twin city traffic simulation density distribution map; The specific calculation formula of the urban traffic flow spatial interpolation distance weight in step S35 is: Among them, W i is the interpolation weight of the i-th sampling point, w pop,i is the flow of people at the i-th sampling point, w traf,i is the road traffic volume at the i-th sampling point, q pop,i is the weight value of the flow of people at the i-th sampling point, q traf,i is the road traffic flow weight value of the i-th sampling point, θ i is the azimuth between the i-th sampling point and the point to be interpolated, d i is the distance between the i-th sampling point and the point to be interpolated, p is the attenuation exponent, and λ is the attenuation rate.

5. The method according to claim 4, characterized in that The specific steps of step S33 are: Step S331: Detect feature points of the digital twin city traffic matrix diagram using a feature point detection algorithm to generate initial digital twin city traffic feature point data; Step S332: Filter and screen the digital twin city traffic feature point data using the characteristic value threshold limiting method to generate benchmark digital twin city traffic feature point data; Step S333: Marking the traffic feature point data of the benchmark digital twin city by using the spatial three-dimensional coordinate method to generate the traffic feature point coordinates of the digital twin city; Step S334: Use the equidistant grid algorithm to grid the coordinates of the digital twin city traffic feature points to generate a digital twin city traffic flow coordinate point grid.

6. The method according to claim 5, characterized in that The specific steps of step S4 are: Step S41: performing stereo pixel conversion on the three-dimensional digital twin city data set to generate large three-dimensional voxels such as the digital twin city; Step S42: gridding a large three-dimensional voxel such as a digital twin city using an inverse distance weighted method to generate a digital twin city grid model; Step S43: Use the voxel modeling method based on the three-dimensional modeling technology to perform three-dimensional modeling on the digital twin city grid model to generate a three-dimensional model of the digital twin city buildings.

7. The method according to claim 1, characterized in that The specific steps of step S64 are: Step S641: Use a multi-scale sampling algorithm to perform spatial pyramid pooling multi-layer sampling on the digital twin city building three-dimensional model to generate digital twin city building convolution feature data; Step S642: performing convolution feature mapping on the digital twin city building convolution feature data to generate a digital twin city building convolution feature vector; Step S643: Use the digital twin city building convolution feature vector to perform vector splicing to generate a digital twin city building feature sequence.

8. A spatial data integration system applied to a digital twin city, characterized in that: include: at least one processor; at least one memory communicatively coupled to the processor; The memory stores a computer program executable by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the spatial data integration method applied to a digital twin city as described in any one of claims 1 to 7.

Images