A Prediction Interaction Method for the Spatiotemporal Distribution of Populations Based on an Intelligent Mining Model

Abstract

The present invention discloses an interactive method for predicting the spatio-temporal distribution of people based on an intelligent mining model in the field of urban planning, comprising the following steps: input and standardization of geospatial information; constructing a grid of 500×500 meters and coupling and calculating index values; expanding the number of changes in the number of people in each grid at each time period; calculating the extreme values of the number of changes in the number of people in each grid and the population distribution pattern; constructing a training model of the human-land relationship; calculating the spatio-temporal distribution prediction results of the population in each grid; and displaying the prediction results and providing real-time adjustment feedback. The present invention solves the problem of predicting the spatio-temporal distribution of people in a complex large-scale urban environment, solves the problems of long prediction cycle and low accuracy of the spatio-temporal distribution prediction of urban planning schemes, has more accurate spatial and temporal granularities, realizes accurate and efficient prediction of the spatio-temporal distribution of people in a complex large-scale urban built environment, and displays the prediction results and provides real-time adjustment feedback.

Description

Technical Field

[0001] The present invention belongs to the field of urban planning, and particularly relates to an interactive method for predicting the spatio-temporal distribution of crowds based on an intelligent mining model. Background Art

[0002] The development of information and communication technology has provided important data sources and technical supports for the research on urban population distribution. The accuracy of the quantitative analysis method based on the spatio-temporal distribution data of the current population in urban planning has been continuously improved, but there are problems such as long cycle, low interactivity, and difficulty in connecting with the plan. The existing prediction research on the spatio-temporal distribution of crowds mainly uses image and video recognition methods, which involve a small spatial scale and mainly focus on short-term and real-time predictions. Some research predicts through prior theories, statistics, dynamics, traditional machine learning, etc., and there are problems such as low accuracy, lack of research on the evolution law and mechanism of the spatio-temporal distribution of crowds, and small correlation with the built environment. The spatio-temporal distribution of crowds has regularity, and the urban built environment is closely related to the spatio-temporal distribution of crowds. Under the influence of the heterogeneous built environment of the city, the crowds have relatively stable distribution change characteristics. However, there is currently a lack of a method for predicting the spatio-temporal distribution of crowds in a complex large-scale urban built environment. Summary of the Invention

[0003] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide an interactive method for predicting the spatio-temporal distribution of crowds based on an intelligent mining model to solve the problems raised in the above background art.

[0004] The purpose of the present invention can be achieved by the following technical solutions:

[0005] An interactive method for predicting the spatio-temporal distribution of crowds based on an intelligent mining model includes the following steps:

[0006] Step 1, obtain the current geographical spatial information data and the current spatio-temporal distribution data of the crowds in the city, standardize the data, and import it into the database for storage;

[0007] Step 2, construct grids, couple and calculate the numerical values of various indicators of the grids with the data in Step 1, and then import it into the database for storage;

[0008] Among them, the numerical values of the indicators include the numerical values of the built environment indicators of each grid and the change quantity of the crowds in each time period;

[0009] Step 3, perform grid extraction, on-site measure the change quantity of the crowds in each time period of the extracted grids, calculate the expansion coefficient of the measured data of the extracted grids and the current spatio-temporal distribution data of the crowds, and expand the sample of the current spatio-temporal distribution data of the crowds based on the expansion coefficient;

[0010] Step 4: Import the population change quantities at different time periods of the population in the raster in Step 3, extract the extreme values of the population change quantities of each raster, calculate the population distribution change value, and cluster it into n population distribution patterns;

[0011] Step 5: Associate the population distribution patterns, the extreme values of the population change quantities in Step 4 with the built environment index values, and integrate to obtain the intelligent matching module of the human-land relationship;

[0012] Step 6: Input the geospatial information data of the urban planning and design scheme, couple and calculate the built environment characteristic index values with the raster in Step 2, import them into the human-land relationship training library, calculate the population distribution patterns and the extreme values of the population change quantities of the urban planning and design scheme, and calculate the prediction results of the spatio-temporal distribution of the population in each raster;

[0013] Step 7: Export the prediction results of the spatio-temporal distribution of the population in each raster above to a three-dimensional display sand table for display, adjust the geospatial information data through instructions, and display the feedback of the prediction results of the spatio-temporal distribution of the population in each raster after adjustment in real time.

[0014] Preferably, Step 1 specifically includes the following steps:

[0015] Step 1.1: Obtain the current urban geospatial information data, which includes urban boundary data, terrain data, building data, land use function data, road traffic data, center level data, scenic spot location data, and historical and cultural block location data;

[0016] Step 1.2: Obtain the current data of the spatio-temporal distribution of the urban population, which includes mobile phone signaling data and LBS data;

[0017] Step 1.3: Input the current urban geospatial information data and the current spatio-temporal distribution data of the population into the geographic information processing platform, and then import the standardized data into the memory.

[0018] Preferably, Step 2 specifically includes the following steps:

[0019] Step 2.1: Convert the urban boundary data in Step 1 into a 500×500-meter raster;

[0020] Step 2.2: Couple and import the geospatial information data obtained in Step 1.3 with the above raster into a computer workstation for calculation, obtain the built environment characteristic index values of each raster, and import them into the memory;

[0021] The built environment characteristic index values include three categories: spatial accessibility, functional type, and spatial capacity index;

[0022] Preferably, Step 3 specifically includes the following steps:

[0023] Step 3.1: Automatically extract 1 / 10 of the total number of grids through a computer workstation, and actually measure the number of people in each period within 24 hours for the sampled grids.

[0024] Step 3.2: Calculate the average value of the ratio of the above-mentioned measured data of the sample grids to the population spatio-temporal distribution data in Step 2 to obtain the sample expansion coefficient a. Based on the coefficient a, expand the current population spatio-temporal distribution data of all grids to obtain the expanded data of the population change quantity in each period of the grids.

[0025] Preferably, Step 4 specifically includes the following steps:

[0026] Step 4.1: Import the expanded data of the population change quantity in each period of the grids in Step 3, and extract the extreme values of the population change quantity of each grid. The extreme values include the maximum and minimum values of the population quantity in each period within 24 hours of the grid, and import them into the memory.

[0027] Step 4.2: Calculate the change value of the population distribution in each period of the grid.

[0028] Step 4.3: Cluster the above-mentioned population distribution change values, and take those with an Euclidean distance within 0.55 as the same category to obtain the population distribution pattern, which represents the type of relative change of the pedestrian flow within 24 hours in the urban grid, and import it into the memory.

[0029] Preferably, Step 5 specifically includes the following steps:

[0030] Step 5.1: Train the non-linear mapping relationship between the population distribution pattern in Step 4 and the built environment characteristic index values. The built environment characteristic index values include function type, spatial capacity, and spatial accessibility indicators.

[0031] Step 5.2: Train the non-linear mapping relationship between the extreme values of the population change quantity in Step 4 and the built environment characteristic index values. The built environment characteristic index values include function type, spatial capacity, and spatial accessibility indicators.

[0032] Step 5.3: Import the above training results into the computer workstation, and integrate the two mapping relationships to obtain the intelligent matching module of the human-land relationship.

[0033] Preferably, Step 6 specifically includes the following steps:

[0034] Step 6.1: Obtain the geospatial information data of the urban planning and design scheme, and import it into the memory. The geospatial information data is the same as that in Step 1.

[0035] Step 6.2: Import the above geospatial information data and the raster in Step 2 into a computer workstation for spatial association calculation to obtain the built environment characteristic index values of the planning and design scheme, which are the same as those in Step 2;

[0036] Step 6.3: Import the above built environment characteristic index values into the human-land relationship intelligent matching module in Step 5 to calculate the population distribution pattern and the extreme value of population change quantity of the urban planning and design scheme;

[0037] Step 6.4: Match the above extreme value of population change quantity to the vertices and lowest points of the population distribution pattern to calculate the predicted results of the spatio-temporal distribution of the population in each raster.

[0038] Preferably, Step 7 specifically includes the following steps:

[0039] Step 7.1: Import the predicted results of the spatio-temporal distribution of the population in each raster in Step 6 and the geospatial information data of the urban planning and design scheme into a three-dimensional holographic sand table platform for display;

[0040] Step 7.2: Construct a modification instruction library, which includes the instruction name and the gesture action instruction information;

[0041] Step 7.3: Set up a gesture instruction module, and set sensors at the fingertips and the palm of the five fingers of the data glove respectively;

[0042] Step 7.4: The planner wears VR glasses and adjusts the geospatial information data of the three-dimensional sand table through the data glove, compares the instruction with the instruction content in the instruction library for calculation, and determines the operation instruction of the planner;

[0043] Step 7.5: Input the modified and adjusted results of the intelligent three-dimensional sand table corresponding to the instruction into the computer workstation for calculation, adjust the geospatial information data of the urban planning and design scheme, and feedback it to Step 6.1. Repeat Step 6 and Step 7, and end after all adjustments are completed.

[0044] Advantages of the present invention:

[0045] 1. The present invention greatly shortens the population prediction time in the overall territorial space planning and greatly improves the work efficiency by intelligently mining the population distribution pattern and intelligently matching it with 33 sub-categories of the three major categories of urban built environment, and realizes the accurate and rapid prediction of the spatio-temporal distribution results of the population in the complex large-scale urban built environment;

[0046] 2. The present invention rasterizes the urban area into 500m×500m grids to obtain the predicted population distribution of each grid, and greatly improves the accuracy compared with the traditional urban planning population prediction;

[0047] 3. Based on the current urban geographical spatial information data and the current population spatio-temporal distribution data, the present invention can achieve the prediction of the dynamic changes of the urban population with an accuracy of 24 hours a day. Traditional urban planning population prediction can only predict the approximate numbers of the permanent population and the household registered population in a city with a certain accuracy in a future year, and there is a great improvement in the accuracy of the prediction time period.

[0048] 4. The present invention realizes the interactive adjustment of the urban planning scheme and the predicted population distribution, and the immediate display of the adjusted scheme in the three-dimensional sand table. The planner can immediately adjust the scheme through gesture commands, and the population prediction result is adjusted immediately, enhancing the interactivity between the planner and the three-dimensional sand table of the scheme, and immediately displaying the adjusted three-dimensional sand table model, realizing the immediate adjustment and result display of the planning scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0050] Figure 1 is the process framework diagram of the method of the present invention;

[0051] Figure 2 is the raster schematic diagram in the present invention;

[0052] Figure 3 is the schematic diagram of the principle of collecting population spatio-temporal data in the present invention;

[0053] Figure 4 is the schematic diagram of the population distribution pattern in the present invention;

[0054] Figure 5 is the schematic diagram of the three-dimensional sand table interaction in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0056] Taking the urban design of a certain city in Fujian as an example and with the attached drawings, the technical solutions of the present invention will be described in detail:

[0057] Please refer to Figure 1 As shown, the present invention discloses a method for predicting the spatio-temporal distribution of the population based on an intelligent mining model, including the following steps:

[0058] Step 1: Obtain the current geographical spatial information data and the current spatio-temporal distribution data of the population in a certain city in Fujian, standardize the data, and import it into the database for storage;

[0059] The detailed process of Step 1 is as follows:

[0060] Step 1.1: Obtain the current geographical spatial information data of a certain city in Fujian. The current geographical spatial information data of the city includes urban boundary data, terrain data, building data, land use function data, road traffic data, central level data, scenic spot location data, and historical and cultural block location data. Among them, the urban terrain data and urban building data are collected by the GeoSLAM ZEB Discovery mobile laser panoramic three-dimensional scanning system, and the boundary data, land use function data, road traffic data, central level data, scenic spot location data, and historical and cultural block location data are obtained from relevant departments;

[0061] Step 1.2: Obtain the current data of the spatio-temporal distribution of the population in a certain city in Fujian. The current data of the spatio-temporal distribution of the population in the city includes, but is not limited to, mobile phone signaling data, LBS data, and other positioning data, which are obtained from relevant departments;

[0062] Step 1.3: Input the current geographical spatial information data and the current spatio-temporal distribution data of the population into the geographic information processing platform. The shp file of the urban boundary data should contain boundary outline, location, and area information. The shp file of the terrain data should contain elevation and location information. The shp file of the building data should contain building outline, number of building floors, building height, and building location information. The shp file of the land use function data should contain land use nature, land use boundary, and location information. The line elements of the shp file of the road traffic data should contain road grade, red line width, and location information. The shp file of the central level data should contain central level zoning, grading, and location information. The shp file of the scenic spot location data should contain scenic spot range, location, and grade information. The historical and cultural block location data should contain the coordinates of the center point of the historical and cultural block, the range of the historical and cultural block, location, and grade information. The spatio-temporal distribution data of the population should contain number, time, longitude and latitude coordinate information. Import the above standardized data into the memory.

[0063] Step 2: Construct a 500×500-meter grid, couple and calculate the numerical values of various grid indicators with the various types of data in Step 1 above, and import them into the database for storage. The indicator numerical values include the numerical values of the built environment indicators of each grid and the change quantity of the population in each time period;

[0064] The detailed process of Step 2 is as follows:

[0065] Step 2.1: Through the geographic information processing platform, convert the urban boundary data in Step 1 into a 500×500-meter grid;

[0066] Step 2.2: Couple and import the geospatial information data obtained in Step 1.3 with the above grid into a computer workstation with a computing power of 5 petaFLOPS and a memory of 320 GB for calculation, obtain the built environment characteristic index values of each grid, and import them into the memory. The built environment characteristic index values include three major categories: function type, spatial capacity, and spatial accessibility indicators, which are further subdivided into 33 small categories. The specific indicators are shown in Table 1;

[0067] Table 1: Index Table of Built Environment Characteristic Values

[0068]

[0069]

[0070]

[0071] Step 2.3: Import the population spatio-temporal distribution data obtained in Step 1.3 and the above grid into a computer workstation with a computing power of 5 petaFLOPS and a memory of 320 GB for coupled calculation, obtain the change quantity of the population in each grid at each time period, and import it into the memory.

[0072] Step 3: Automatically extract grids by random sampling, as Figure 3 shown, conduct on-site measurements of the change quantity of the population in each time period of the sampled grids, calculate the expansion coefficient of the measured data of the sample grids and the current population spatio-temporal distribution data, and expand the current population spatio-temporal distribution data based on the expansion coefficient;

[0073] The detailed process of Step 3 is as follows:

[0074] Step 3.1: Automatically extract 1 / 10 of the total number of grids by random sampling using a computer workstation with a computing power of 5 petaFLOPS and a memory of 320 GB, as Figure 3 shown, use an unmanned aerial vehicle electromagnetic wave penetration imaging detector capable of counting the number of people to conduct on-site measurements of the change quantity of the population in each time period within the field range of the sampled grids, and obtain the measured number of people in each time period within 24 hours of the sampled grids;

[0075] Step 3.2: Calculate the average value of the ratio of the above measured data of the sample grids to the population spatio-temporal distribution data in Step 2 to obtain the sample expansion coefficient a, and expand the current population spatio-temporal distribution data of all grids based on the coefficient a to obtain the expanded data of the change quantity of the population in each grid at each time period. The calculation formula is:

[0076] Q it = aN it

[0077] Wherein, i is the grid number; t is a certain period within 24 hours; Q is the expanded data of the population change quantity in each period of the grid; N is the current situation data of the spatio-temporal distribution of the population in each period of the grid; a is the sample expansion coefficient.

[0078] Step 4: Import the population change quantity in each period of each grid in Step 3, and extract the extreme values of the population change quantity of each grid. For example, Figure 4 As shown, use the normalization algorithm to calculate the population distribution change value, and use the k-means algorithm to cluster it into n population distribution patterns, where those with an Euclidean distance within 0.55 belong to the same class;

[0079] The detailed process of Step 4 is as follows:

[0080] Step 4.1: Import the expanded data of the population change quantity in each period of the grid in Step 3, and extract the extreme values of the population change quantity of each grid. The extreme values include the maximum and minimum values of the population quantity in each period within 24 hours of the grid, and import them into the memory;

[0081] Step 4.2: As Figure 4 shown, use the normalization algorithm to calculate the population distribution change value of each grid. The calculation formula is:

[0082] flow pattern =(flow - flow min ) / (flow max - flow min )

[0083] Wherein, flow is the number of people in a certain period of the grid; flow min is the minimum value of the grid population; flow max is the maximum value of the grid population; flow pattern is the population distribution change value

[0084] Step 4.3: As Figure 4 shown, use the k-means clustering method to cluster the above population distribution change values, and take those with an Euclidean distance within 0.55 as the same class. Finally, obtain 8 population distribution patterns, representing the relative change situation types of the human flow within 24 hours of the urban grid, and import them into the memory.

[0085] Step 5: Use the random forest decision algorithm to associate the population distribution pattern, the extreme value of the population change quantity in Step 4 with the built environment index value respectively, and integrate to obtain the intelligent matching module of the human-land relationship;

[0086] The detailed process of Step 5 is as follows:

[0087] Step 5.1: Use the random forest decision algorithm to train the non-linear mapping relationship between the population distribution pattern and the built environment characteristic index values in Step 4. The built environment characteristic index values include function type, spatial capacity, and spatial accessibility indicators.

[0088] Step 5.2: Use the random forest decision algorithm to train the non-linear mapping relationship between the extreme value of the population change quantity and the built environment characteristic index values in Step 4. The built environment characteristic index values include function type, spatial capacity, and spatial accessibility indicators.

[0089] Step 5.3: Import the above training results into a computer workstation with a computing power of 5 petaFLOPS and a memory of 320 GB, and integrate the two mapping relationships to obtain the intelligent matching module of the human-land relationship.

[0090] Step 6: Input the geospatial information data of the urban planning and design scheme of a certain city in Fujian, couple it with the grid in Step 2 to calculate the built environment characteristic index values, import them into the human-land relationship training library, calculate the population distribution pattern and the extreme value of the population change quantity of the urban planning and design scheme, and calculate the predicted results of the spatio-temporal distribution of the population in each grid.

[0091] The detailed process of Step 6 is as follows:

[0092] Step 6.1: Obtain the geospatial information data of the urban planning and design scheme of a certain city in Fujian and import it into the memory. The geospatial information data is the same as that in Step 1.

[0093] Step 6.2: Import the above geospatial information data and the grid in Step 2 into a computer workstation with a computing power of 5 petaFLOPS and a memory of 320 GB for spatial association calculation to obtain the built environment characteristic index values of the planning and design scheme. The built environment characteristic index values are the same as those in Step 2.

[0094] Step 6.3: Import the above built environment characteristic index values into the intelligent matching module of the human-land relationship in Step 5 to calculate the population distribution pattern and the extreme value of the population change quantity of the urban planning and design scheme of a certain city in Fujian.

[0095] Step 6.4: Match the above extreme value of the population change quantity to the vertices and lowest points of the population distribution pattern to calculate the predicted results of the spatio-temporal distribution of the population in each grid.

[0096] Step 7: Export the predicted results of the spatio-temporal distribution of the population in each grid above to a three-dimensional display sand table for display, adjust the geospatial information data through instructions, and display the feedback of the predicted results of the spatio-temporal distribution of the population in each grid after adjustment in real time.

[0097] The detailed process of Step 7 is as follows:

[0098] Step 7.1: Import the spatio-temporal distribution prediction results of each grid population and the geospatial information data of the urban planning and design plan in Step 6 into the three-dimensional holographic sand table platform for display;

[0099] Step 7.2: Construct a modification instruction library, where the instruction library includes the instruction name and gesture instruction information;

[0100] Step 7.3: Set up a gesture instruction module. 5m sensors are respectively set at the fingertips and the palm of the five fingers of the data glove to ensure that the corresponding gesture instructions made by the planner wearing the data glove within the detection distance range are recognized by the sensors.

[0101] Step 7.4: The planner wears VR glasses and adjusts the geospatial information data of the three-dimensional sand table through the data glove, compares and calculates the instruction with the instruction content in the instruction library, and determines the operation instruction of the planner;

[0102] Step 7.5: Input the modified and adjusted results of the intelligent three-dimensional sand table corresponding to the instruction into a computer workstation with a computing power of 5 petaFLOPS and a memory of 320GB for calculation. After adjusting the geospatial information data of the urban planning and design plan, feedback it to Step 6.1, and repeat Steps 6 and 7.

[0103] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0104] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification only illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed.

Claims

1. A crowd spatio-temporal distribution prediction and interaction method based on an intelligent mining model, characterized in that It includes the following steps: Step 1: Obtain the current urban geographical spatial information data and the current population spatio-temporal distribution data, standardize the data, and import it into the database for storage; Step 2: Construct a grid, couple and calculate the numerical values of various grid indicators with the data in Step 1, and then import it into the database for storage; Among them, the indicator numerical values include the built environment indicator numerical values of each grid and the population change quantities in each time period; Step 3: Conduct grid extraction, on-site measure the population change quantities in each time period of the extracted grids, calculate the expansion coefficient of the measured data of the extracted grids and the current population spatio-temporal distribution data, and expand the current population spatio-temporal distribution data based on the expansion coefficient; Step 4: Import the population change quantities in each time period of the grids in Step 3, extract the extreme values of the population change quantities of each grid, calculate the population distribution change value, and cluster it into n population distribution patterns; Step 5: Associate the population distribution patterns, the extreme values of the population change quantities in Step 4 with the built environment indicator numerical values, and integrate to obtain a human-land relationship intelligent matching module; Step 6: Input the geographical spatial information data of the urban planning and design scheme, couple and calculate the built environment characteristic indicator values with the grids in Step 2, import them into the human-land relationship training library, calculate the population distribution patterns and the extreme values of the population change quantities of the urban planning and design scheme, and calculate the population spatio-temporal distribution prediction results of each grid; Step 7: Export the population spatio-temporal distribution prediction results of the above-mentioned grids to a three-dimensional display sand table for display, adjust the geographical spatial information data through instructions, and display the feedback of the population spatio-temporal distribution prediction results of each grid after real-time adjustment; The specific steps of Step 5 include the following steps: Step 5.1: Train the non-linear mapping relationship between the population distribution patterns in Step 4 and the built environment characteristic indicator values. The built environment characteristic indicator values include function type, spatial capacity, and spatial accessibility indicators; Step 5.2: Train the non-linear mapping relationship between the extreme values of the population change quantities in Step 4 and the built environment characteristic indicator values. The built environment characteristic indicator values include function type, spatial capacity, and spatial accessibility indicators; Step 5.3: Import the above training results into a computer workstation, and integrate the two mapping relationships to obtain a human-land relationship intelligent matching module; The specific steps of Step 6 include the following steps: Step 6.1: Obtain the geographical spatial information data of the urban planning and design scheme, and import it into the memory. The geographical spatial information data is the same as that in Step 1; Step 6.2: Import the above geographical spatial information data and the grids in Step 2 into the computer workstation for spatial association calculation to obtain the built environment characteristic indicator values of the planning and design scheme. The built environment characteristic indicator values are the same as those in Step 2; Step 6.3: Import the above built environment characteristic indicator values into the human-land relationship intelligent matching module in Step 5 to calculate the population distribution patterns and the extreme values of the population change quantities of the urban planning and design scheme; Step 6.4: Match the above extreme values of the population change quantities to the vertices and lowest points of the population distribution patterns, and calculate the population spatio-temporal distribution prediction results of each grid.

2. The crowd spatio-temporal distribution prediction and interaction method based on an intelligent mining model according to claim 1, wherein The specific steps of Step 1 include the following steps: Step 1.1: Obtain the current urban geospatial information data, which includes urban boundary data, topographic data, building data, land use function data, road traffic data, central hierarchy data, scenic spot location data, and historical and cultural block location data; Step 1.2: Obtain the current data on the spatio-temporal distribution of the urban population, which includes mobile signaling data and LBS data; Step 1.3: Input the current urban geospatial information data and the current spatio-temporal distribution data of the population into a geographic information processing platform, and then import the standardized data into a memory.

3. The crowd spatio-temporal distribution prediction and interaction method based on an intelligent mining model according to claim 1, wherein The specific steps of Step 2 are as follows: Step 2.1: Convert the urban boundary data in Step 1 into a 500×500-meter grid; Step 2.2: Couple and import the geospatial information data obtained in Step 1.3 with the above grid into a computer workstation for calculation to obtain the built environment characteristic index values of each grid, and import them into a memory; The built environment characteristic index values include three categories: spatial accessibility, function type, and spatial capacity index; Step 2.3: Import the spatio-temporal distribution data of the population obtained in Step 1.3 and the above grid into a computer workstation for coupled calculation to obtain the change quantity of the population in each grid at each time period, and import them into a memory.

4. An interactive method for predicting the spatio-temporal distribution of a population based on an intelligent mining model according to claim 3, characterized in that, The spatial accessibility index includes: Distance to the main center: The distance from the geometric center of each spatial analysis unit to the main center of the city; Distance to the nearest secondary center: The distance from the geometric center of each spatial analysis unit to the nearest secondary center of the city; Distance to the tertiary center: The distance from the geometric center of each spatial analysis unit to the nearest tertiary center of the city; Highest road grade: Assign a value to the highest road grade within the spatial analysis unit, expressway = 4, arterial road = 3, sub-arterial road = 2, branch road = 1; Railway length: The total length of the railway within each spatial analysis unit; Distance to the railway station hub: The distance from the geometric center within each spatial analysis unit to the nearest railway station hub of the city; Number of subway stations: The number of subway stations within the spatial unit; Distance to the airport hub: The distance from the geometric center within each spatial analysis unit to the nearest airport hub of the city; Length of the road network: The sum of the lengths of all road centerlines within the spatial analysis unit; The function type index includes: Proportion of land for public service management and service: The percentage of the area of land for public management and public service facilities in the area of the spatial analysis unit; Proportion of land for institutions of higher learning: The percentage of the area of land for universities, colleges, junior colleges and their affiliated facilities in the area of the spatial analysis unit; Proportion of land for science, education, culture and health: The percentage of the area of land for middle schools, primary schools, medical, health care, hygiene, epidemic prevention, rehabilitation and first aid facilities in the area of the spatial analysis unit; Proportion of land for commercial and service facilities: The percentage of the area of land for commercial, business, entertainment and sports facilities in the area of the spatial analysis unit; Proportion of special land: The percentage of the area of special land such as military land in the area of the spatial analysis unit; Proportion of square land: The percentage of the urban public activity sites mainly for recreation, commemoration, assembly and refuge in the area of the spatial analysis unit; Industrial land ratio: The percentage of the land area of production workshops, warehouses and their ancillary facilities of industrial and mining enterprises in the area of the spatial analysis unit; Residential land ratio: The percentage of the land area of residences and corresponding service facilities in the area of the spatial analysis unit; Transportation land ratio: The percentage of the land area of transportation facilities in the area of the spatial analysis unit; Municipal land ratio: The percentage of the land area of supply, environmental and safety facilities in the area of the spatial analysis unit; Logistics and warehousing land ratio: The percentage of the land area for material storage, transfer and distribution in the area of the spatial analysis unit; Road land ratio: The percentage of road land in the area of the spatial analysis unit; Rural homestead land: The percentage of rural homestead land in the area of the spatial analysis unit; Non-construction land ratio: The percentage of non-construction land area in the area of the spatial analysis unit; Park green space ratio: The percentage of the land area of park green spaces in the area of the spatial analysis unit; Ecological green space ratio: The percentage of the land area of ecological green spaces in the area of the spatial analysis unit; Terrain undulation degree: The difference between the altitude of the highest point and the altitude of the lowest point in each spatial analysis unit; Terrain slope: The maximum value of the ratio of the vertical height to the horizontal distance of the slope surface in each spatial analysis unit; Number of scenic and historic spots: The number of scenic and historic spots contained in each spatial analysis unit; Distance to the center point of the national historical and cultural block: The distance from each spatial analysis unit to the center point of the nearest national-level historical and cultural block in the city; Distance to the center point of the provincial historical and cultural block: The distance from each spatial analysis unit to the center point of the nearest provincial-level historical and cultural block in the city; Spatial capacity indicators include: Building density: The ratio of the total base area of buildings to the area of the spatial range within a certain spatial range; Plot ratio: The ratio of the total floor area of buildings on the ground to the area of the spatial range within a certain spatial range; Entropy of land use function: For spatial analysis unit k, the calculation formula of entropy is as follows: In the formula, landuse _qk represents the information entropy of the land use function structure in the spatial analysis unit k, and n represents the total number of land types of this category in the spatial analysis unit. represents the percentage of the p-th type of land use. Degree of land use function balance: The ratio of the entropy of the spatial analysis unit to the maximum entropy, and the calculation formula is as follows: 。 5. The crowd spatio-temporal distribution prediction and interaction method based on an intelligent mining model according to claim 1, characterized in that, Step 3 specifically includes the following steps: Step 3.1: Automatically extract 1 / 10 of the total number of grid cells through a computer workstation, and actually measure the number of people at each time period within 24 hours of the sampled grid cells; Step 3.2: Calculate the average value of the ratio of the above-mentioned measured data of the sample grid cells to the population spatio-temporal distribution data in Step 2 to obtain the sample expansion coefficient a. Based on the coefficient a, expand the current population spatio-temporal distribution data of all grid cells to obtain the expanded data of the population change quantity at each time period of the grid cells. The calculation formula is: Q it = aN it In the formula, i is the grid cell number; t is a certain time period within 24 hours; Q is the expanded data of the population change quantity at each time period of the grid cells; N is the current data of the population spatio-temporal distribution of the grid cells; a is the sample expansion coefficient.

6. The crowd spatio-temporal distribution prediction and interaction method based on an intelligent mining model according to claim 1, characterized in that, Step 4 specifically includes the following steps: Step 4.1: Import the expanded data of the population change quantity at each time period of the grid cells in Step 3, and extract the extreme values of the population change quantity of each grid cell; The extreme values include the maximum and minimum values of the population quantity at each time period within 24 hours of the grid cells, and import them into the memory; Step 4.2: Calculate the population distribution change value for each grid period. The calculation formula is as follows: flow pattern = (flow - flow min ) / (flow max - flow min ） Among them, the flow is the number of people in a grid during a certain period, flow min is the minimum value of the number of people in the grid, flow max is the maximum value of the number of people in the grid, flow pattern is the value of the change in population distribution; Step 4.3: Cluster the above population distribution change values. Those with an Euclidean distance within 0.55 are regarded as the same category to obtain the population distribution pattern, which represents the type of relative change in the human flow within 24 hours of the urban grid, and import it into the memory.

7. A method for predicting the spatio-temporal distribution of a crowd based on an intelligent mining model according to claim 1, characterized in that The specific steps of step 7 are as follows: Step 7.1: Import the spatio-temporal distribution prediction results of the population in each grid in step 6 and the geospatial information data of the urban planning and design plan into the three-dimensional holographic sand table platform for display; Step 7.2: Construct a modification instruction library, which includes the instruction name and gesture action instruction information; Step 7.3: Set up a gesture instruction module, and set sensors at the fingertips and palms of the five fingers of the data glove respectively; Step 7.4: The planner wears VR glasses and adjusts the geospatial information data of the three-dimensional sand table through the data glove, compares the instruction with the instruction content in the instruction library for calculation, and determines the operation instruction of the planner; Step 7.5: Input the modified and adjusted results of the intelligent three-dimensional sand table corresponding to the instruction into the computer workstation for calculation, adjust the geospatial information data of the urban planning and design plan, and feedback it to step 6.

1. Repeat steps 6 and 7, and end after all adjustments are completed.

Images