Urban planning and construction management system based on space-time big data

Abstract

The invention discloses an urban planning and construction management system based on space-time big data, and relates to the technical field of urban planning and management. Analyzing the multi-source heterogeneous data of the urban area to obtain a plurality of ground feature type areas of the urban area; according to the building model rule and the multi-source heterogeneous data, constructing a spatio-temporal data fusion model in the urban area; performing spatial analysis on each three-dimensional model in the spatio-temporal data fusion model to obtain each type of spatial adaptability coefficient of each three-dimensional model; constructing a trend prediction model, and obtaining predicted social statistical data of each surface feature type region; and according to each type of spatial adaptability coefficient of each three-dimensional model and the predicted social statistical data, the recommendation function attribute of each three-dimensional model is output, and the scientificity and efficiency of urban planning and construction management are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban planning and management, and in particular to an urban planning, construction and management system based on spatiotemporal big data. Background Art

[0002] The prior art CN2023114044076 "A method and system for simulating the construction of urban infrastructure based on digital twins" obtains the layout and direction of various infrastructures in the city according to the urban planning and design information, and then processes the data of various influencing factors of urban planning, and renders the data of various influencing factors in the digital twin model of urban planning to obtain the future direction of urban planning and design visualization, so as to enable urban planning and design and management departments to better understand the future operation of urban infrastructure, so as to provide a prerequisite reference basis for urban planning and design renovation, and make more wise decisions, so as to make the urban infrastructure design plan more in line with the actual urban planning. That is, it effectively solves the shortcomings of the prior art that it cannot well visualize the future direction of urban infrastructure.

[0003] Prior art CN2023109587373 "City digital twin platform and method based on large-scale GIS lightweight engine" builds a digital twin model of the large-scale city based on GIS data and other data sources collected by the large-scale GIS lightweight engine; processes and analyzes city data based on the digital twin model; monitors various data of the large-scale city in real time and makes real-time predictions; and provides data visualization and analysis tools to help users make decisions and plans. In this way, various indicators and trends of the city can be obtained to help users understand the operating status and development trends of the city.

[0004] With the acceleration of urbanization and the continuous expansion of urban scale, urban planning, construction and management are facing many challenges. Traditional urban planning, construction and management methods mainly rely on manual experience and limited data, which makes it difficult to fully and accurately grasp the dynamic changes and development needs of cities. In terms of urban planning, it is impossible to fully consider the factors of urban space and time dimensions, resulting in a lack of scientificity and foresight in planning schemes.

[0005] With the development of information technology, big data technology has been gradually applied to various fields. Spatiotemporal big data contains rich geographic spatial information and time series information, which can provide comprehensive and accurate data support for urban planning and construction management. However, there is still a lack of an urban planning and construction management system that can efficiently integrate and utilize spatiotemporal big data. Summary of the invention

[0006] In order to solve the above technical problems, the object of the present invention is to provide an urban planning and construction management system based on spatiotemporal big data, including a cloud, wherein the cloud is communicatively connected with a data acquisition module, a data analysis module, a model building module, a spatial correlation analysis module, a trend prediction module and an attribute matching module; The data collection module is used to obtain multi-source heterogeneous data in urban areas, mark the collection time, and set the collection cycle; The data analysis module is used to analyze multi-source heterogeneous data in urban areas and obtain several land feature type areas in urban areas; The model building module is used to build a spatiotemporal data fusion model in the urban area based on the building model rules and multi-source heterogeneous data; The spatial correlation analysis module is used to perform spatial analysis on each three-dimensional model in the spatiotemporal data fusion model to obtain various types of spatial adaptability coefficients of each three-dimensional model; The trend prediction module is used to build a trend prediction model to obtain predicted social statistics for each land feature type area; The attribute matching module is used to output the recommended functional attributes of each three-dimensional model based on the spatial adaptability coefficients of each type of three-dimensional model and the predicted social statistical data.

[0007] Furthermore, the multi-source heterogeneous data include satellite remote sensing data, point cloud data, Internet of Things data (including traffic flow sensors, real-time dynamic data obtained by environmental monitoring stations) and social statistical data (population size, economic data).

[0008] Furthermore, the data analysis module analyzes the multi-source heterogeneous data of the urban area, and the process of obtaining several land feature type areas in the urban area includes: Data preprocessing of satellite remote sensing data of urban areas begins with radiation correction to eliminate the influence of the sensor itself and the atmosphere on the radiation brightness of remote sensing images. For example, for optical remote sensing images, the interference of atmospheric scattering and absorption on the images is removed by the 6S model based on the physical model. Then, geometric correction is performed to correct the remote sensing images to a unified geographic coordinate system to match the actual geographic spatial position. For example, ground control points (GCPs) are used for geometric correction. By selecting the same-name points on the map of the image and the urban area, correction is performed using methods such as polynomial transformation, and Gaussian filtering is used to filter the remote sensing images to reduce the influence of noise on texture analysis. The types of objects to be studied (including buildings, vegetation, bare land, and water bodies) are preset, and the band combination to be studied is determined according to the types of objects to be studied. Spectral features are extracted from the satellite remote sensing data after data preprocessing to obtain the spectral reflectance data of each band in the band combination to be studied. According to the spectral reflectance data of each band in the band combination to be studied, several object type areas in the urban area are obtained.

[0009] The visible light band (such as blue, green, and red light bands) can reflect the color and material differences of the building surface. The near-infrared band can assist in distinguishing between buildings and vegetation (vegetation usually has a higher reflectivity in the near-infrared band, while buildings have a relatively low reflectivity). The short-wave infrared band has a certain ability to identify the roof material of the building (such as insulation and waterproof coatings of different materials). Generally speaking, the reflectivity of buildings in the visible light band varies depending on the building materials (such as cement, masonry, glass, etc.), and the overall reflectivity is lower than that of vegetation in the near-infrared band. The process of obtaining building areas and other landform types in urban areas includes: Draw the spectral curves of buildings and other land object types (with the band as the horizontal axis and the reflectivity as the vertical axis), and obtain the building areas and other land object types in the urban area based on the spectral reflectance data of each band in the band combination to be studied in different areas in the urban area and the spectral curves of buildings and other land object types.

[0010] Furthermore, the model building module constructs a spatiotemporal data fusion model in an urban area according to the building model rules and multi-source heterogeneous data, and the process includes: Preset the building model rules for different land feature type areas, segment the point cloud data according to the land feature type areas, distinguish the building point cloud data, ground point cloud data, vegetation point cloud data, etc., and construct the 3D model of each land feature type area according to the building model rules and point cloud data of different land feature type areas; for example: For residential buildings, the building model rules stipulate that it is a multi-layer rectangular structure with a height of 3 meters for each floor. After extracting the point cloud belonging to the residential building from the point cloud data, the model framework is constructed according to this rule, and the segmented building point cloud data is fitted to the constructed model framework. Satellite remote sensing images are used to add texture to the building model; For ground point cloud data, a ground model is constructed based on the segmented ground point cloud data. The building model rules stipulate that a triangulated network (TIN) algorithm is used to connect the ground point cloud into a triangular mesh to generate a terrain surface. Based on the elevation information of the point cloud, the terrain undulations are accurately reflected. For vegetation point cloud data, a model is constructed based on the type and growth characteristics of the vegetation. For example, for trees, a parametric method is used to determine the height, diameter, shape and size of the trunk based on the point cloud data, and then a corresponding tree model is created. For other features such as roads and water systems, they are also constructed based on their point cloud data features and corresponding building model rules. The direction and width of roads are determined according to the point cloud data, and they are constructed as rectangular blocks or models with a certain curvature. The water system is constructed as a water body model with a certain shape and transparency based on the water boundary and depth information reflected by the point cloud. Obtain the matching relationship between the IoT data in the urban area and each three-dimensional model, and match the IoT data in the urban area with the three-dimensional model according to the matching relationship to obtain a spatiotemporal data fusion model; Collect real-time data in urban areas from various IoT devices (such as smart sensors, cameras, smart meters, water meters, etc.). These data cover traffic flow, environmental parameters (temperature, humidity, air quality, etc.), energy consumption, equipment operation status, etc. Use the installation location information of IoT devices (usually obtained through GPS or other positioning technologies) to match them with the geographic coordinates of the three-dimensional model. For example, for a traffic flow sensor installed on a certain road, obtain its longitude and latitude coordinates through a spatial index algorithm (such as R-tree) to find the corresponding location point in the three-dimensional road model.

[0011] Furthermore, the spatial correlation analysis module performs spatial analysis on each three-dimensional model in the spatiotemporal data fusion model, and the process of obtaining each type of spatial adaptability coefficient of each three-dimensional model includes: According to the IoT data of each three-dimensional model in the spatiotemporal data fusion model, the numerical time series sequence corresponding to each type of environmental indicator of each three-dimensional model is obtained, the numerical time series sequence corresponding to each type of environmental indicator is used as the evaluation indicator, the indicator weights and different environmental quality levels of the evaluation indicators are preset, and the membership matrix of the evaluation indicators for different environmental quality levels is obtained through fuzzy comprehensive evaluation, and the environmental quality level of each three-dimensional model is obtained according to the membership matrix and the indicator weights; Perform spatial analysis on the IoT data of each three-dimensional model to obtain the residential traffic accessibility coefficient, commercial traffic accessibility coefficient, and industrial traffic accessibility coefficient of each three-dimensional model; According to the environmental quality level and residential traffic accessibility coefficient of each three-dimensional model, the residential space adaptability coefficient qv of each three-dimensional model is obtained. ,in, It represents the weight factor corresponding to the environmental quality level in the calculation process of the adaptability coefficient of the living space. Indicates the environmental quality level, represents the residential traffic accessibility coefficient; according to the environmental quality level and commercial traffic accessibility coefficient of each three-dimensional model, the commercial space adaptability coefficient bv of each three-dimensional model is obtained. ,in, It represents the weight factor corresponding to the environmental quality level in the calculation process of the commercial space adaptability coefficient. represents the adaptability coefficient of commercial space; according to the environmental quality level and industrial traffic accessibility coefficient of each three-dimensional model, the industrial space adaptability coefficient ov of each three-dimensional model is obtained. ,in, It represents the weight factor corresponding to the environmental quality level in the calculation process of the industrial space adaptability coefficient. Represents the industrial transportation accessibility coefficient.

[0012] Furthermore, the process of obtaining the residential traffic accessibility coefficient, the commercial traffic accessibility coefficient and the industrial traffic accessibility coefficient of each three-dimensional model includes: Acquire the functional attributes of different three-dimensional models (such as shopping malls, transportation hubs, hospitals, etc.), preset the residential importance coefficients of different functional attributes for the residential transportation accessibility coefficients, the commercial importance coefficients of different functional attributes for the commercial transportation accessibility coefficients, and the industrial importance coefficients of different functional attributes for the industrial transportation accessibility coefficients, and acquire the residential importance coefficients, commercial importance coefficients, and industrial importance coefficients of different three-dimensional models according to the functional attributes of different three-dimensional models; The shortest path algorithm (such as Dijkstra algorithm) is used to obtain the shortest path distance from each 3D model to other 3D models. Meanwhile, the traffic flow data in the IoT data is obtained, and the traffic flow data is statistically analyzed to obtain the average traffic flow of the shortest path from each 3D model to other 3D models. According to the shortest path distance between the three-dimensional model and other three-dimensional models, the average traffic flow of the shortest path, and the residential importance coefficient, commercial importance coefficient and industrial importance coefficient of other three-dimensional models, the residential traffic accessibility coefficient, commercial traffic accessibility coefficient and industrial traffic accessibility coefficient of the three-dimensional model are obtained.

[0013] The calculation process for obtaining the residential transportation accessibility coefficient of the three-dimensional model is as follows: ; in, represents the residential transportation accessibility coefficient of three-dimensional model i, k represents traversing all other three-dimensional models, represents the shortest path distance from 3D model i to other 3D models k, represents the average traffic flow of the shortest path from 3D model i to other 3D models k, Represents the conversion coefficient. The calculation process of the commercial traffic accessibility coefficient and the industrial traffic accessibility coefficient of the three-dimensional model is consistent with the calculation process of the above-mentioned residential traffic accessibility coefficient and will not be repeated. The above-mentioned formulas are all calculated by removing the dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data.

[0014] Furthermore, the trend prediction module constructs a trend prediction model, and the process of obtaining predicted social statistical data for each land feature type area includes: According to the social statistical data of the urban area, the population and economic data (such as GDP value) of each land feature type area in several historical collection periods are obtained, and a trend prediction model is constructed based on deep learning. The population data and economic data of each land feature type area in several historical collection periods are used as training sets and test sets, and the training sets are input into the trend prediction model for training until the loss function training is stable, and the model parameters are saved. The trend prediction model is tested by the test set until it meets the preset requirements, and the trend prediction model is output; The predicted population and economic data of each land feature type area are output based on the trend prediction model.

[0015] Building a trend prediction model based on deep learning is a complex process, which involves multiple steps such as model selection, training, verification and testing. The following is a detailed supplementary explanation of this process: The convolutional neural network (CNN) suitable for time series analysis is selected as the deep learning architecture. After the model architecture is determined, the Huber loss function is selected as the optimization target, and then the prepared training set is input into the selected deep learning model to start training. During the training process, the weights are continuously updated through the back propagation algorithm, so that the loss function gradually decreases until it reaches a stable state. During this period, techniques such as Early Stopping are used to avoid overfitting. In addition to the basic training process, various parameters of the model are tuned through Grid Search, including learning rate, batch size, regularization coefficient, etc.

[0016] After the model training is completed and the parameters are adjusted, the final evaluation is performed through the test set to obtain the evaluation results of the model, which include classification indicators such as accuracy, recall, F1 score, etc. Based on the evaluation results on the test set, it is judged whether the model has reached the expected standards. If the requirements are met, the model parameters are saved and ready for deployment; if not ideal, it is necessary to return to a previous stage to re-examine issues such as data quality, model structure or training strategy.

[0017] Further, the attribute matching module outputs the recommended functional attributes of each three-dimensional model according to the spatial adaptability coefficients of each type of each three-dimensional model and the predicted social statistical data, and the process of outputting the recommended functional attributes of each three-dimensional model includes: According to the adaptability coefficient of the residential space of each three-dimensional model and the predicted population and predicted economic data of the area where each three-dimensional model is located, the residential recommendation coefficient qvk of each three-dimensional model is obtained. ,in, It represents the weight factor corresponding to the predicted population in the calculation process of the residential recommendation coefficient. 5 represents the weight factor corresponding to the predicted economic data in the calculation process of the residential recommendation coefficient. represents the predicted population size, Represents predicted economic data. According to the commercial space adaptability coefficient of each 3D model and the predicted population and predicted economic data of the area where each 3D model is located, the commercial recommendation coefficient bvk of each 3D model is obtained. ,in, It represents the weight factor corresponding to the predicted population in the calculation process of the commercial recommendation coefficient. 7 represents the weight factor corresponding to the predicted economic data in the process of calculating the commercial recommendation coefficient. The industrial recommendation coefficient ovk of each three-dimensional model is obtained according to the industrial space adaptability coefficient of each three-dimensional model and the predicted population and predicted economic data of the feature type area where each three-dimensional model is located. ,in, It represents the weight factor corresponding to the predicted population in the calculation process of the industrial recommendation coefficient. 9 represents the weight factor corresponding to the predicted economic data in the calculation process of the industrial recommendation coefficient; The residential recommendation coefficient, commercial recommendation coefficient and industrial recommendation coefficient of the three-dimensional model are compared, and the recommended functional attributes of the three-dimensional model are obtained according to the comparison results. For example, if the residential recommendation coefficient of the three-dimensional model is greater than the commercial recommendation coefficient and the industrial recommendation coefficient, then the recommended functional attribute of the three-dimensional model is the residential attribute.

[0018] Compared with the prior art, the present invention has the following beneficial effects: 1. This system integrates multi-source heterogeneous data, including satellite remote sensing data, point cloud data, IoT data and social statistics, breaking the limitation of single traditional data source and enabling the system to comprehensively and comprehensively reflect various conditions of the city. For example, satellite remote sensing data can be used to grasp the city's land use and topography, while IoT data can present the city's dynamic operation in real time, such as traffic flow and environmental parameters. The integration of multiple data provides a richer and more accurate information basis for subsequent analysis, greatly improving the breadth and depth of data processing.

[0019] 2. The data analysis module conducts in-depth mining based on multi-source heterogeneous data. By extracting and analyzing the spectral features of satellite remote sensing data, it can accurately identify different types of land objects in urban areas, such as buildings, vegetation, bare land, and water bodies. This precise classification of land objects helps to understand the spatial layout of the city more clearly and provide reliable data support for urban planning.

[0020] 3. The model building module builds a spatiotemporal data fusion model based on building model rules and multi-source data. By segmenting the point cloud data by object type, building a three-dimensional model in combination with the corresponding model rules, and then matching the IoT data with it, a deep fusion of urban spatial and temporal dimension information is achieved. This model can simulate the actual situation of the city more realistically. For example, in urban construction projects, the impact of construction projects on the surrounding environment and infrastructure can be accurately evaluated based on the model, thereby optimizing the construction plan.

[0021] 4. The spatial correlation analysis module performs spatial analysis based on the spatiotemporal data fusion model to obtain various spatial adaptability coefficients of each three-dimensional model. These coefficients comprehensively consider factors such as environmental quality and traffic accessibility, and provide a scientific basis for the rational layout of urban functional areas. For example, when planning residential areas, areas with good environmental quality and convenient transportation can be selected based on the adaptability coefficient of residential space to improve the quality of life of residents.

[0022] 5. The trend prediction module uses deep learning to build a trend prediction model, and predicts the population and economic data of each type of area based on historical social statistics. This prediction capability enables urban planners to understand the development trend of the city in advance. For example, if they predict that the population of a certain area will grow rapidly in the future, they can plan the construction of public service facilities such as schools and hospitals in advance to avoid resource shortages or waste.

[0023] 6. The attribute matching module outputs recommended functional attributes for each 3D model based on the spatial adaptability coefficient and predicted social statistical data. By comparing the residential, commercial and industrial recommendation coefficients, the optimal functional positioning of each area can be reasonably determined. For example, areas with high commercial recommendation coefficients can be planned as commercial areas to promote the prosperity and development of the urban economy. This function provides direct decision-making support for urban planning, construction and management, and improves the scientificity and rationality of decision-making.

[0024] Combining the above functions, this system can analyze and predict cities from multiple perspectives, providing comprehensive and accurate information and scientific decision-making basis for urban planning, construction and management. Compared with traditional urban planning management methods, it reduces the subjectivity and limitations of human experience and judgment, and greatly improves the scientific nature of planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic diagram of an urban planning and construction management system based on spatiotemporal big data according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0027] like Figure 1 As shown, an urban planning and construction management system based on spatiotemporal big data includes a cloud, wherein the cloud is connected to a data acquisition module, a data analysis module, a model building module, a spatial correlation analysis module, a trend prediction module and an attribute matching module; The data collection module is used to obtain multi-source heterogeneous data in urban areas, mark the collection time, and set the collection cycle; The data analysis module is used to analyze multi-source heterogeneous data in urban areas and obtain several land feature type areas in urban areas; The model building module is used to build a spatiotemporal data fusion model in the urban area based on the building model rules and multi-source heterogeneous data; The spatial correlation analysis module is used to perform spatial analysis on each three-dimensional model in the spatiotemporal data fusion model to obtain various types of spatial adaptability coefficients of each three-dimensional model; The trend prediction module is used to build a trend prediction model to obtain predicted social statistics for each land feature type area; The attribute matching module is used to output the recommended functional attributes of each three-dimensional model based on the spatial adaptability coefficients of each type of three-dimensional model and the predicted social statistical data.

[0028] It should be further explained that, in the specific implementation process, the multi-source heterogeneous data include satellite remote sensing data, point cloud data, Internet of Things data (including traffic flow sensors, real-time dynamic data obtained by environmental monitoring stations) and social statistical data (population size, economic data).

[0029] It should be further explained that, in the specific implementation process, the data analysis module analyzes the multi-source heterogeneous data of the urban area, and the process of obtaining several land feature type areas in the urban area includes: Data preprocessing of satellite remote sensing data of urban areas begins with radiation correction to eliminate the influence of the sensor itself and the atmosphere on the radiation brightness of remote sensing images. For example, for optical remote sensing images, the interference of atmospheric scattering and absorption on the images is removed by the 6S model based on the physical model. Then, geometric correction is performed to correct the remote sensing images to a unified geographic coordinate system to match the actual geographic spatial position. For example, ground control points (GCPs) are used for geometric correction. By selecting the same-name points on the map of the image and the urban area, correction is performed using methods such as polynomial transformation, and Gaussian filtering is used to filter the remote sensing images to reduce the influence of noise on texture analysis. The types of objects to be studied (including buildings, vegetation, bare land, and water bodies) are preset, and the band combination to be studied is determined according to the types of objects to be studied. Spectral features are extracted from the satellite remote sensing data after data preprocessing to obtain the spectral reflectance data of each band in the band combination to be studied. According to the spectral reflectance data of each band in the band combination to be studied, several object type areas in the urban area are obtained.

[0030] The visible light band (such as blue, green, and red light bands) can reflect the color and material differences of the building surface. The near-infrared band can assist in distinguishing between buildings and vegetation (vegetation usually has a higher reflectivity in the near-infrared band, while buildings have a relatively low reflectivity). The short-wave infrared band has a certain ability to identify the roof material of the building (such as insulation and waterproof coatings of different materials). Generally speaking, the reflectivity of buildings in the visible light band varies depending on the building materials (such as cement, masonry, glass, etc.), and the overall reflectivity is lower than that of vegetation in the near-infrared band. The process of obtaining building areas and other landform types in urban areas includes: Draw the spectral curves of buildings and other land object types (with the band as the horizontal axis and the reflectivity as the vertical axis), and obtain the building areas and other land object types in the urban area based on the spectral reflectance data of each band in the band combination to be studied in different areas in the urban area and the spectral curves of buildings and other land object types.

[0031] It should be further explained that, in the specific implementation process, the model building module constructs the spatiotemporal data fusion model in the urban area according to the building model rules and multi-source heterogeneous data, and the process includes: Preset the building model rules for different land feature type areas, segment the point cloud data according to the land feature type areas, distinguish the building point cloud data, ground point cloud data, vegetation point cloud data, etc., and construct the 3D model of each land feature type area according to the building model rules and point cloud data of different land feature type areas; for example: For residential buildings, the building model rules stipulate that it is a multi-layer rectangular structure with a height of 3 meters for each floor. After extracting the point cloud belonging to the residential building from the point cloud data, the model framework is constructed according to this rule, and the segmented building point cloud data is fitted to the constructed model framework. Satellite remote sensing images are used to add texture to the building model; For ground point cloud data, a ground model is constructed based on the segmented ground point cloud data. The building model rules stipulate that a triangulated network (TIN) algorithm is used to connect the ground point cloud into a triangular mesh to generate a terrain surface. Based on the elevation information of the point cloud, the terrain undulations are accurately reflected. For vegetation point cloud data, a model is constructed based on the type and growth characteristics of the vegetation. For example, for trees, a parametric method is used to determine the height, diameter, shape and size of the trunk based on the point cloud data, and then a corresponding tree model is created. For other features such as roads and water systems, they are also constructed based on their point cloud data features and corresponding building model rules. The direction and width of roads are determined according to the point cloud data, and they are constructed as rectangular blocks or models with a certain curvature. The water system is constructed as a water body model with a certain shape and transparency based on the water boundary and depth information reflected by the point cloud. Obtain the matching relationship between the IoT data in the urban area and each three-dimensional model, and match the IoT data in the urban area with the three-dimensional model according to the matching relationship to obtain a spatiotemporal data fusion model; Collect real-time data in urban areas from various IoT devices (such as smart sensors, cameras, smart meters, water meters, etc.). These data cover traffic flow, environmental parameters (temperature, humidity, air quality, etc.), energy consumption, equipment operation status, etc. Use the installation location information of IoT devices (usually obtained through GPS or other positioning technologies) to match them with the geographic coordinates of the three-dimensional model. For example, for a traffic flow sensor installed on a certain road, obtain its longitude and latitude coordinates through a spatial index algorithm (such as R-tree) to find the corresponding location point in the three-dimensional road model.

[0032] It should be further explained that, in the specific implementation process, the spatial correlation analysis module performs spatial analysis on each three-dimensional model in the spatiotemporal data fusion model, and the process of obtaining each type of spatial adaptability coefficient of each three-dimensional model includes: According to the IoT data of each three-dimensional model in the spatiotemporal data fusion model, the numerical time series sequence corresponding to each type of environmental indicator of each three-dimensional model is obtained, the numerical time series sequence corresponding to each type of environmental indicator is used as the evaluation indicator, the indicator weights and different environmental quality levels of the evaluation indicators are preset, and the membership matrix of the evaluation indicators for different environmental quality levels is obtained through fuzzy comprehensive evaluation, and the environmental quality level of each three-dimensional model is obtained according to the membership matrix and the indicator weights; Perform spatial analysis on the IoT data of each three-dimensional model to obtain the residential traffic accessibility coefficient, commercial traffic accessibility coefficient, and industrial traffic accessibility coefficient of each three-dimensional model; According to the environmental quality level and residential traffic accessibility coefficient of each three-dimensional model, the residential space adaptability coefficient qv of each three-dimensional model is obtained. ,in, It represents the weight factor corresponding to the environmental quality level in the calculation process of the adaptability coefficient of the living space. Indicates the environmental quality level, represents the residential traffic accessibility coefficient; according to the environmental quality level and commercial traffic accessibility coefficient of each three-dimensional model, the commercial space adaptability coefficient bv of each three-dimensional model is obtained. ,in, It represents the weight factor corresponding to the environmental quality level in the calculation process of the commercial space adaptability coefficient. represents the adaptability coefficient of commercial space; according to the environmental quality level and industrial traffic accessibility coefficient of each three-dimensional model, the industrial space adaptability coefficient ov of each three-dimensional model is obtained. ,in, It represents the weight factor corresponding to the environmental quality level in the calculation process of the industrial space adaptability coefficient. Represents the industrial transportation accessibility coefficient.

[0033] It should be further explained that, in the specific implementation process, the process of obtaining the residential traffic accessibility coefficient, commercial traffic accessibility coefficient and industrial traffic accessibility coefficient of each three-dimensional model includes: Acquire the functional attributes of different three-dimensional models (such as shopping malls, transportation hubs, hospitals, etc.), preset the residential importance coefficients of different functional attributes for the residential transportation accessibility coefficients, the commercial importance coefficients of different functional attributes for the commercial transportation accessibility coefficients, and the industrial importance coefficients of different functional attributes for the industrial transportation accessibility coefficients, and acquire the residential importance coefficients, commercial importance coefficients, and industrial importance coefficients of different three-dimensional models according to the functional attributes of different three-dimensional models; The shortest path algorithm (such as Dijkstra algorithm) is used to obtain the shortest path distance from each 3D model to other 3D models. Meanwhile, the traffic flow data in the IoT data is obtained, and the traffic flow data is statistically analyzed to obtain the average traffic flow of the shortest path from each 3D model to other 3D models. According to the shortest path distance between the three-dimensional model and other three-dimensional models, the average traffic flow of the shortest path, and the residential importance coefficient, commercial importance coefficient and industrial importance coefficient of other three-dimensional models, the residential traffic accessibility coefficient, commercial traffic accessibility coefficient and industrial traffic accessibility coefficient of the three-dimensional model are obtained.

[0034] The calculation process for obtaining the residential transportation accessibility coefficient of the three-dimensional model is as follows: ; in, represents the residential transportation accessibility coefficient of three-dimensional model i, k represents traversing all other three-dimensional models, represents the shortest path distance from 3D model i to other 3D models k, represents the average traffic flow of the shortest path from 3D model i to other 3D models k, Represents the conversion coefficient. The calculation process of the commercial traffic accessibility coefficient and the industrial traffic accessibility coefficient of the three-dimensional model is consistent with the calculation process of the above-mentioned residential traffic accessibility coefficient and will not be repeated. The above-mentioned formulas are all calculated by removing the dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data.

[0035] It should be further explained that, in the specific implementation process, the trend prediction module constructs a trend prediction model and obtains the predicted social statistical data of each land feature type area, including: According to the social statistical data of the urban area, the population and economic data (such as GDP value) of each land feature type area in several historical collection periods are obtained, and a trend prediction model is constructed based on deep learning. The population data and economic data of each land feature type area in several historical collection periods are used as training sets and test sets, and the training sets are input into the trend prediction model for training until the loss function training is stable, and the model parameters are saved. The trend prediction model is tested by the test set until it meets the preset requirements, and the trend prediction model is output; The predicted population and economic data of each land feature type area are output based on the trend prediction model.

[0036] Building a trend prediction model based on deep learning is a complex process, which involves multiple steps such as model selection, training, verification and testing. The following is a detailed supplementary explanation of this process: The convolutional neural network (CNN) suitable for time series analysis is selected as the deep learning architecture. After the model architecture is determined, the Huber loss function is selected as the optimization target, and then the prepared training set is input into the selected deep learning model to start training. During the training process, the weights are continuously updated through the back propagation algorithm, so that the loss function gradually decreases until it reaches a stable state. During this period, techniques such as Early Stopping are used to avoid overfitting. In addition to the basic training process, various parameters of the model are tuned through Grid Search, including learning rate, batch size, regularization coefficient, etc.

[0037] After the model training is completed and the parameters are adjusted, the final evaluation is performed through the test set to obtain the evaluation results of the model, which include classification indicators such as accuracy, recall, F1 score, etc. Based on the evaluation results on the test set, it is judged whether the model has reached the expected standards. If the requirements are met, the model parameters are saved and ready for deployment; if not ideal, it is necessary to return to a previous stage to re-examine issues such as data quality, model structure or training strategy.

[0038] It should be further explained that, in the specific implementation process, the attribute matching module outputs the recommended functional attributes of each 3D model according to the spatial adaptability coefficients of each type of 3D model and the predicted social statistical data, including: According to the adaptability coefficient of the residential space of each three-dimensional model and the predicted population and predicted economic data of the area where each three-dimensional model is located, the residential recommendation coefficient qvk of each three-dimensional model is obtained. ,in, It represents the weight factor corresponding to the predicted population in the calculation process of the residential recommendation coefficient. 5 represents the weight factor corresponding to the predicted economic data in the calculation process of the residential recommendation coefficient. represents the predicted population size, Represents predicted economic data. According to the commercial space adaptability coefficient of each 3D model and the predicted population and predicted economic data of the area where each 3D model is located, the commercial recommendation coefficient bvk of each 3D model is obtained. ,in, It represents the weight factor corresponding to the predicted population in the calculation process of the commercial recommendation coefficient. 7 represents the weight factor corresponding to the predicted economic data in the process of calculating the commercial recommendation coefficient. The industrial recommendation coefficient ovk of each three-dimensional model is obtained according to the industrial space adaptability coefficient of each three-dimensional model and the predicted population and predicted economic data of the feature type area where each three-dimensional model is located. ,in, It represents the weight factor corresponding to the predicted population in the calculation process of the industrial recommendation coefficient. 9 represents the weight factor corresponding to the predicted economic data in the calculation process of the industrial recommendation coefficient; The residential recommendation coefficient, commercial recommendation coefficient and industrial recommendation coefficient of the three-dimensional model are compared, and the recommended functional attributes of the three-dimensional model are obtained according to the comparison results. For example, if the residential recommendation coefficient of the three-dimensional model is greater than the commercial recommendation coefficient and the industrial recommendation coefficient, then the recommended functional attribute of the three-dimensional model is the residential attribute.

[0039] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An urban planning and construction management system based on spatiotemporal big data, characterized in that: It includes a cloud, wherein the cloud is communicatively connected with a data acquisition module, a data analysis module, a model building module, a spatial correlation analysis module, a trend prediction module and an attribute matching module; The data collection module is used to obtain multi-source heterogeneous data in urban areas, mark the collection time, and set the collection cycle; The data analysis module is used to analyze multi-source heterogeneous data in urban areas and obtain several land feature type areas in urban areas; The model building module is used to build a spatiotemporal data fusion model in the urban area based on the building model rules and multi-source heterogeneous data; The spatial correlation analysis module is used to perform spatial analysis on each three-dimensional model in the spatiotemporal data fusion model to obtain various types of spatial adaptability coefficients of each three-dimensional model; The trend prediction module is used to build a trend prediction model to obtain predicted social statistics for each land feature type area; The attribute matching module is used to output the recommended functional attributes of each three-dimensional model based on the spatial adaptability coefficients of each type of three-dimensional model and the predicted social statistical data.

2. The urban planning and construction management system based on spatiotemporal big data according to claim 1 is characterized in that: The multi-source heterogeneous data include satellite remote sensing data, point cloud data, Internet of Things data and social statistical data.

3. The urban planning and construction management system based on spatiotemporal big data according to claim 2 is characterized in that: The data analysis module analyzes the multi-source heterogeneous data of the urban area, and the process of obtaining several land feature type areas in the urban area includes: The satellite remote sensing data of the urban area is preprocessed, the type of land object to be studied is preset, the band combination to be studied is determined according to the type of land object to be studied, and the spectral feature extraction is performed on the satellite remote sensing data after data preprocessing to obtain the spectral reflectance data of each band in the band combination to be studied. According to the spectral reflectance data of each band in the band combination to be studied, several land object type areas in the urban area are obtained.

4. The urban planning and construction management system based on spatiotemporal big data according to claim 3 is characterized in that: The model building module builds a spatiotemporal data fusion model in an urban area based on building model rules and multi-source heterogeneous data. The process includes: Preset building model rules for different land feature type areas, segment point cloud data according to land feature type areas, and construct 3D models of each land feature type area based on the building model rules and point cloud data for different land feature type areas; The matching relationship between the IoT data in the urban area and each three-dimensional model is obtained, and the IoT data in the urban area is matched with the three-dimensional model according to the matching relationship to obtain a spatiotemporal data fusion model.

5. The urban planning and construction management system based on spatiotemporal big data according to claim 4 is characterized in that: The spatial correlation analysis module performs spatial analysis on each 3D model in the spatiotemporal data fusion model, and the process of obtaining various types of spatial adaptability coefficients of each 3D model includes: According to the IoT data of each three-dimensional model in the spatiotemporal data fusion model, the numerical time series sequence corresponding to each type of environmental indicator of each three-dimensional model is obtained, the numerical time series sequence corresponding to each type of environmental indicator is used as the evaluation indicator, the indicator weights and different environmental quality levels of the evaluation indicators are preset, and the membership matrix of the evaluation indicators for different environmental quality levels is obtained through fuzzy comprehensive evaluation, and the environmental quality level of each three-dimensional model is obtained according to the membership matrix and the indicator weights; Perform spatial analysis on the IoT data of each three-dimensional model to obtain the residential traffic accessibility coefficient, commercial traffic accessibility coefficient, and industrial traffic accessibility coefficient of each three-dimensional model; According to the environmental quality level and residential traffic accessibility coefficient of each three-dimensional model, the residential space adaptability coefficient of each three-dimensional model is obtained; according to the environmental quality level and commercial traffic accessibility coefficient of each three-dimensional model, the commercial space adaptability coefficient of each three-dimensional model is obtained; according to the environmental quality level and industrial traffic accessibility coefficient of each three-dimensional model, the industrial space adaptability coefficient of each three-dimensional model is obtained.

6. The urban planning and construction management system based on spatiotemporal big data according to claim 5 is characterized in that: The process of obtaining the residential traffic accessibility coefficient, commercial traffic accessibility coefficient and industrial traffic accessibility coefficient of each three-dimensional model includes: Acquire functional attributes of different three-dimensional models, preset residential importance coefficients of different functional attributes for residential traffic accessibility coefficients, commercial importance coefficients of different functional attributes for commercial traffic accessibility coefficients, and industrial importance coefficients of different functional attributes for industrial traffic accessibility coefficients, and acquire residential importance coefficients, commercial importance coefficients, and industrial importance coefficients of different three-dimensional models according to the functional attributes of different three-dimensional models; The shortest path algorithm is used to obtain the shortest path distance from each 3D model to other 3D models. At the same time, traffic flow data in the IoT data is obtained, and statistical analysis is performed on the traffic flow data to obtain the average traffic flow of the shortest path from each 3D model to other 3D models. According to the shortest path distance between the three-dimensional model and other three-dimensional models, the average traffic flow of the shortest path, and the residential importance coefficient, commercial importance coefficient and industrial importance coefficient of other three-dimensional models, the residential traffic accessibility coefficient, commercial traffic accessibility coefficient and industrial traffic accessibility coefficient of the three-dimensional model are obtained.

7. The urban planning and construction management system based on spatiotemporal big data according to claim 6 is characterized in that: The trend prediction module builds a trend prediction model and obtains the predicted social statistical data of each land feature type area. The process includes: According to the social statistical data of the urban area, the population and economic data of each land feature type area in several historical collection periods are obtained, and a trend prediction model is constructed based on deep learning. The population and economic data of each land feature type area in several historical collection periods are used as training sets and test sets, and the training sets are input into the trend prediction model for training until the loss function training is stable, and the model parameters are saved, and the trend prediction model is tested by the test set until it meets the preset requirements, and the trend prediction model is output; The predicted population and economic data of each land feature type area are output based on the trend prediction model.

8. The urban planning and construction management system based on spatiotemporal big data according to claim 7 is characterized in that: The attribute matching module outputs the recommended functional attributes of each 3D model based on the spatial adaptability coefficients of each type of 3D model and the predicted social statistical data, including: According to the adaptability coefficient of the residential space of each three-dimensional model and the predicted population and predicted economic data of the property type area where each three-dimensional model is located, the residential recommendation coefficient of each three-dimensional model is obtained; according to the adaptability coefficient of the commercial space of each three-dimensional model and the predicted population and predicted economic data of the property type area where each three-dimensional model is located, the commercial recommendation coefficient of each three-dimensional model is obtained; according to the adaptability coefficient of the industrial space of each three-dimensional model and the predicted population and predicted economic data of the property type area where each three-dimensional model is located, the industrial recommendation coefficient of each three-dimensional model is obtained; The residential recommendation coefficient, commercial recommendation coefficient and industrial recommendation coefficient of the three-dimensional model are compared, and the recommended functional attributes of the three-dimensional model are obtained according to the comparison results.

Images