Urban plot height adaptive generation and demonstration platform based on artificial intelligence algorithm

Abstract

The application discloses a city plot height self-adaptive generation and demonstration platform based on an artificial intelligence algorithm, and comprises the following steps: site data collection and calculation; case data collection and conversion; site space structure generation and structure potential value division; spatial structure holographic display and selection; plot height self-adaptive generation; human-computer interaction adjustment of plot height; scheme judgment and output; the city plot height self-adaptive generation and demonstration platform based on the artificial intelligence algorithm solves the problems of complicated traditional manual scheme design process, long working period, poor interactive design, high labor cost and the like, and further improves the efficiency of plot height generation; integrates the designer thinking and computer logic, realizes a top-down and bottom-up combined height form scheme generation method, and responds to city design height scheme generation in more scenes.

Description

Technical Field

[0001] This invention belongs to the field of urban planning technology, and in particular relates to an adaptive generation and demonstration platform for urban plot height based on artificial intelligence algorithms. Background Technology

[0002] In the process of urban design, it is necessary to combine the higher-level planning requirements of the design site with the complex surrounding environment to conduct multiple rounds of deliberation and design of the height form.

[0003] In the traditional design process, urban planners adjust the urban height form through processes such as site analysis, translation of higher-level structures, implementation of indicators, collection of feedback, and multiple rounds of height adjustments. This process involves a large human resource investment and a long design cycle. With the development of computer and information technology, parametric methods of multi-factor overlay analysis have been gradually applied to the field of urban height form generation. However, the generated height form scheme lacks structural control, can only be applied to plots smaller than 50 hectares, and is difficult to adjust through interactive design. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an urban land parcel height adaptive generation and demonstration platform based on artificial intelligence algorithms, thus solving the aforementioned problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an urban land parcel height adaptive generation and demonstration platform based on artificial intelligence algorithms, comprising the following steps:

[0006] S1. Collect site geospatial information data and convert it into raster information, and calculate the characteristic values ​​of each plot;

[0007] S2. Collect geospatial information data of the case studies, convert it into raster information, and import it into the urban design case study library;

[0008] S3. The conditional adversarial generative network model is trained through the urban design scheme case library and imported into the site grid information of step S1. The spatial structure grid information of the site is generated and converted into vector data. The height potential value of the site structure is divided according to the centroid of the plot and the buffer distance of the spatial structure.

[0009] S4. Import the height potential value distribution map of the plot structure into the holographic sand table display device, and select the design structure scheme;

[0010] S5. Based on the height potential value of the land structure in S4, divide the land into key land parcels and general land parcels and generate heights;

[0011] S6, output the height result generated in S5 to the holographic sand table device operation display, generate a three-dimensional height model, select the site plot in S5 and adjust the height of the site plot, and obtain a final scheme;

[0012] S7, import the upper planning index library, and judge whether the height of the site plot meets the index requirement, if yes, output a plot height index table, and if not, return to S6 to adjust and output again.

[0013] On the basis of the above technical scheme, the application also provides the following optional technical schemes:

[0014] Further technical scheme: the specific steps of S1 are:

[0015] S101, using a computer workstation, acquiring geographic spatial information data of a site through a public space information platform, saving and importing the site spatial information into geographic information processing software;

[0016] S102, calculating characteristic indexes of each plot through the geographic information processing software, the characteristic indexes including a land use area, a perimeter, and a height potential value of a plot function, the height potential value of the plot function being valued through a plot land use function;

[0017] S103, converting the geographic spatial information data of the site into raster information.

[0018] Further technical scheme: the specific steps of S2 are:

[0019] S201, acquiring geographic spatial information data of a city design case through the method of S101, acquiring land use function and spatial structure raster information through city design text drawings, the spatial structure information including core point element raster information and axis element raster information;

[0020] S202, acquiring building height information of the above city design case through a surveying and mapping unmanned aerial vehicle and inputting the building height information into geographic information processing software, calculating the highest building height of each plot as a plot height, generating a height distribution frequency histogram through a data statistical command, an x-axis being a plot height and a y-axis being height frequency changes, wherein the plot height unit of the x-axis is meters;

[0021] S203, extracting core points and axis elements in the spatial structure raster information of the above city design scheme through a Hog image feature extraction and recognition algorithm and converting them into pictures, wherein the core points and the axis lines are all converted into raster information;

[0022] S204, importing the geographic spatial information data and the raster information of the city design case into a city design case library and storing them.

[0023] Further technical solutions: the specific steps of S3 are:

[0024] S301, machine learning training is performed on the mapping features of the land function grid information and the spatial structure grid information in S2, a mapping relationship between the two is generated, and a conditional adversarial generation network model is constructed;

[0025] S302, input the grid information of the site in S1 into the conditional adversarial generation network model in S301, and generate a site multi-scheme spatial structure diagram;

[0026] S303, core points and axis grid information in the site multi-scheme spatial structure diagram are extracted through a Hog image feature extraction and recognition algorithm, and are converted into vector data;

[0027] S304, according to the buffer distance between the plot centroid and the spatial structure, the height potential value of the plot structure is assigned.

[0028] Further technical solutions: the specific steps of S4 are:

[0029] The height potential value of the plot structure is output in VR format and displayed on a holographic digital sand table platform for scheme comparison and selection, and a spatial structure scheme is obtained.

[0030] Further technical solutions: the specific steps of S5 are:

[0031] S501, according to the selected spatial structure scheme in S4, and according to the height potential value of the plot structure, the plot structure is divided into key plots and general plots;

[0032] S502, the height potential value of the key plot plot structure in S501 and the height potential value of the plot function in S102 are normalized and weighted to obtain the final height potential value of the key plot, and the weights of the two are each 50%, and the calculation formula is:

[0033]

[0034] Wherein, Pf(i) is the height potential value, P(i) is the final height potential value of the key plot, and Ps(i) is the height potential value of the key plot plot structure.

[0035] S503, the height distribution frequency histogram in S2 is automatically matched with the final height potential value of the key plot through an integral mapping algorithm, and the mapping relationship between the height of plot i and the final height potential value is:

[0036]

[0037] According to the above correspondence and the indefinite function f(x), the height Hi is calculated, wherein Pi is the final height potential value, Hi is the height of the plot, and the indefinite function f() is obtained according to the curve change of the height distribution frequency histogram;

[0038] S504, according to the plot area and function information of the general plot, the KNN algorithm is used to automatically match the plot height in the S2 urban design case library which is consistent in function, has a size ratio between 0.8 and 1.2, and is closest to 1; if the matching fails, the average maximum height of the adjacent plot is assigned to the plot, wherein the adjacent plot is adjacent to one side, and the average maximum height is calculated by plot area * plot maximum height / total plot area.

[0039] Further technical solutions: the specific steps of S6 are:

[0040] S601, merging the key plot and general plot data in S5 to form a plot height scheme, unit: meters;

[0041] S602, inputting the generated maximum height scheme into the holographic sand table platform, automatically positioning in the holographic sand table background through geographic spatial coordinate information, generating a plot height three-dimensional model and displaying it;

[0042] S603, the designer adjusts the plot height in the plot height three-dimensional model generated by S602 by the method in S402 to form a final height scheme.

[0043] Further technical solutions: the specific steps of S7 are:

[0044] S701, importing the upper planning index library to the holographic sand table platform, wherein the upper planning index library includes plot height limit index;

[0045] S702, using a computer workstation to calculate and determine whether the height of the plot meets the requirements of the upper planning index library, if yes, the result is exported to generate engineering drawings and displayed on the holographic sand table, if not, return to S6 for adjustment and output the result again.

[0046] Further technical solutions: the height potential value of the plot function in S102 is assigned by the plot function, wherein the assignment rule is:

[0047] The commercial business mixed land is valued as 9, the commercial land is valued as 8, the commercial-residential mixed land is valued as 8, the commercial land is valued as 7, the scientific research land is valued as 7, the administrative office land is valued as 6, the cultural facility land is valued as 5, the school education land is valued as 4, the sports land is valued as 5, the medical and health land is valued as 5, the second-class residential land is valued as 4, the industrial land is valued as 3, the public facility land is valued as 2, and the storage land is valued as 1.

[0048] The classification standard is classified according to the types in the urban land classification and planning construction land standard.

[0049] Further technical solutions: the value rule in S304 is:

[0050] The value is obtained by adding the distance between the mass center of the plot and the nearest core point element and the distance between the plot and the nearest axis element.

[0051] The distance between the mass center of the plot and the nearest core point element is less than 200m, 5 points, greater than or equal to 200m and less than 400m, 3 points, greater than or equal to 400m and less than 800m, 3 points, and greater than or equal to 800m, 1 point. The distance between the mass center of the plot and the nearest axis element is less than 200m, 4 points, greater than or equal to 200m and less than 600m, 2 points, and greater than or equal to 600m, 0 points.

[0052] Beneficial effects

[0053] The application provides a city plot height self-adaptive generation and demonstration platform based on an artificial intelligence algorithm, and has the following beneficial effects compared with the prior art:

[0054] 1. The application expands the application scene of artificial intelligence in city height form generation, and the application reflects the top-down designer thinking in traditional design through structural elements, and makes up for the intelligent generation demand of large and medium scale city height form in actual business;

[0055] 2. The application reduces the time cost and labor cost of the design scheme, and the application quickly generates multiple schemes of the height form of the plot and adjusts in real time through real-time man-machine interaction adjustment and display, finally exports the height index, and greatly improves the work efficiency of the related technical personnel. DETAILED DESCRIPTION

[0056] Figure 1 It is the overall method flow and device framework of the embodiment of the application;

[0057] Figure 2 It is the case height collection diagram of the embodiment of the application;

[0058] Figure 3 It is the space structure diagram generated by the embodiment of the application;

[0059] Figure 4 This is a structural potential value identification diagram of an embodiment of the present invention;

[0060] Figure 5 This is a schematic diagram illustrating the human-computer interaction principle of the structural scheme in an embodiment of the present invention.

[0061] Figure 6 This is a map showing the identification of key land parcels and general land parcels in an embodiment of the present invention;

[0062] Figure 7 This is a mapping diagram showing the relationship between the land parcel height and the final height potential value in an embodiment of the present invention;

[0063] Figure 8 A three-dimensional model diagram of the height morphology of an embodiment of the present invention is generated;

[0064] Figure 9 This is a schematic diagram illustrating the human-computer interaction principle of a height morphology model according to an embodiment of the present invention. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0066] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0067] like Figures 1-9 As shown in the figure, this invention discloses an artificial intelligence-based virtual reality real-time interactive platform for urban design, including the following steps:

[0068] S1. Collect and input geospatial information data of the site, calculate the characteristic index values ​​of each plot, and convert the geospatial information data into raster information;

[0069] S2. Collect geospatial information data and spatial structure raster information of existing urban design schemes, calculate and generate a height distribution frequency histogram, convert the geospatial information data into raster information, and import it into the urban design case library.

[0070] S3. A conditional adversarial generative network model is trained through the urban design scheme case library. The site grid information from S1 is imported to generate the spatial structure grid information of the site and convert it into vector data. The height potential value of the site structure is divided according to the buffer distance between the centroid of the plot and the spatial structure.

[0071] S4. Import the height potential value distribution map of the plot structure into the holographic sand table display device, and the planning designer selects the design structure scheme through gesture commands;

[0072] S5, according to the result of the high potential value of the plot structure in S4, the key plot and the general plot are divided, the high potential index of the key plot is calculated, and the highest height of the key plot is adaptively generated by integral mapping algorithm, the high potential index is obtained by adding the high potential value of the plot structure and the high potential value of the plot function, the plot area and the function information of the general plot of the site are imported into the S2 height generation case library, and the most similar plot height is matched through KNN algorithm;

[0073] S6, the above results are output to the holographic sand table device operation display to generate a three-dimensional height model, and the planner selects and adjusts the height of the plot of the site through gesture instructions to obtain a final scheme;

[0074] S7, import the upper planning index library, and judge whether the height of the plot of the site meets the index requirement, if yes, output the plot height index table, if not, return to S6 for adjustment and output again.

[0075] Specifically, the above S1 includes four steps of S101, S102, S103 and S104:

[0076] S101, using a computer workstation with a computing power of 2.5 petaFLOPS and a memory of 320 GB, the geographic spatial information data of a certain site of Chuzhou City with an area of 90 square kilometers is obtained through a public space information platform such as national geographic information public service platform, Google earth, etc., including plot data, road data, water system data, green land data, and manual input of plot function information. The above information is stored in DS1522+NAS storage and imported into geographic information processing software.

[0077] S102, calculate the geometric command of the geographic information processing software to calculate the characteristic index of each plot, including the area, perimeter, and high potential value of the plot function.

[0078] S103, the height potential value of the plot function is valued by the plot land function, wherein the commercial and business mixed land (land use code B1B2) is valued as 9; the business land (land use code B2) is valued as 8; the commercial and residential mixed land (land use code RB) is valued as 8; the commercial land (land use code B1) is valued as 7; the scientific research land (land use code A35) is valued as 7; the administrative office land (land use code A1) is valued as 6; the cultural facility land (land use code A2) is valued as 5; the school education land (land use code A3, not including A35 land here) is valued as 4; the sports land (land use code A4) is valued as 5; the medical and health land (land use code A5) is valued as 5; the second residential land (land use code R2) is valued as 4; including the industrial land (land use code M1) is valued as 3; the public facility land (land use codes U1, U2, U3, U9) is valued as 2; the storage land (land use code W1) is valued as 1, and the classification standard is classified according to the types in the “Urban Land Classification and Planning Construction Land Standard” (GB50137-2011).

[0079] S104, the site geographic space information data is converted into a grid information with a scale of 1:1000 and a resolution of 300dpi, wherein the plot function information and the color are converted according to the “Urban Planning Cartography Legend” and the “Urban Land Classification and Planning Construction Land Standard” (GB50137-2011).

[0080] Specifically, the above S2 includes four steps of S201, S202, S203 and S204:

[0081] S201, the geographic space information data of 5000 urban design cases is collected by the method of S1.1, including plot data, road data and water system data. The land use function and the space structure grid information are collected by the urban design text drawing. The scale of the grid information is 1:1000, and the resolution is 300dpi. The space structure information includes core point element grid information and axis element grid information;

[0082] S202, the building height information of the above urban design cases is collected by a surveying and mapping unmanned aerial vehicle with a resolution of 1920x1080 or above, as shown in Figure 2 The geographic information processing software is input, the highest building height of each plot is calculated as the plot height, the height distribution frequency histogram is generated by the data statistical command, the x-axis is the plot height (unit: meter), and the y-axis is the height frequency change;

[0083] S203, extract the core point and axis element in the spatial structure grid information of the above-mentioned urban design scheme by Hog image feature extraction recognition algorithm, and convert it into a picture with a scale of 1:1000 and a resolution of 300 dpi, wherein the core point is converted into grid information with RGB value of 255-0-0, and the axis is converted into grid information with RGB value of 255-165-0;

[0084] S204, import the geographic spatial information data and grid information of the above-mentioned 5000 urban design cases into the urban design case library and store them.

[0085] The above S3 includes three steps of S301, S302 and S303:

[0086] S301, machine learning training is performed on the mapping features of the land use function grid information and the spatial structure grid information in S2, and the mapping relationship of the two is generated after 4000 rounds of learning training, and a conditional adversarial generation network model is constructed;

[0087] S302, input the grid information of the site in step S1 into the conditional adversarial generation network model in step S301 to generate a site multi-scheme spatial structure diagram, as shown in Figure 3

[0088] S303, extract the core point and axis grid information in the site multi-scheme spatial structure diagram by Hog image feature extraction recognition algorithm, and convert it into vector data. According to the buffer distance between the plot centroid and the spatial structure, the height potential value of the plot structure is assigned, and the assignment is obtained by adding the distance between the plot centroid and the nearest core point element and the distance between the plot and the nearest axis element, wherein the distance between the plot centroid and the nearest core point element is less than 200m, 5 points, greater than or equal to 200m and less than 400m, 3 points, greater than or equal to 400m and less than 800m, 3 points, greater than or equal to 800m, 1 point; The distance between the plot centroid and the nearest axis element is less than 200m, 4 points, greater than or equal to 200m and less than 600m, 2 points, greater than or equal to 600m, 0 point, as shown in Figure 4

[0089] Specifically, the above S4 includes two steps of S401 and S402:

[0090] S401, output the height potential value of the plot structure in VR format and display it on the holographic digital sand table platform;

[0091] S402, the designer wears VR glasses and handle equipment to compare and select schemes, adjusts the distance and position of the positioning equipment and the digital large screen, and realizes real-time roaming of the scene through the head movement of the person, and realizes zooming, rotating, switching, selecting, lifting, compressing and other operations on the site multi-scheme spatial structure through the hand gestures, and selects the spatial structure scheme, as shown in Figure 5 ​​as shown;

[0092] Specifically, the above S5 includes four steps of S501, S502, S503 and S504:

[0093] S501, import the spatial structure scheme selected in S4, and divide it into key plots and general plots according to the height potential value of the plot structure, wherein the value higher than 6 is a key plot, and the rest is a general plot, such as Figure 6 as shown;

[0094] S502, the final height potential value of the key plot is obtained by weighting the height potential value of the plot structure of the key plot in S501 and the height potential value of the plot function in S102 after normalization, and the weight of the two is 50% respectively, and the calculation formula is:

[0095]

[0096] Wherein, Pf(i) is the height potential value of the plot function, P(i) is the final height potential value of the key plot, and Ps(i) is the height potential value of the plot structure of the key plot;

[0097] S503, the final height potential value of the key plot is automatically matched with the height distribution frequency histogram in S2 by integral mapping algorithm, and the mapping relationship between the height Hi of the plot i and the final height potential value P(i) is:

[0098]

[0099] The height Hi is calculated according to the above corresponding relationship and function formula f(x), wherein Pi is the final height potential value, Hi is the height of plot i, and the indefinite function f() is obtained according to the curve change of the height distribution frequency histogram;

[0100] S504, according to the plot area and function information of the general plot, the height of the plot with the same function, the area size ratio between 0.8-1.2 and the closest 1 in the urban design case library in step S2 is automatically matched by KNN algorithm; if the matching fails, the average maximum height of the adjacent plot is assigned to the plot, wherein one adjacent plot is regarded as adjacent plot, and the average maximum height is calculated by plot area * plot maximum height / total plot area.

[0101] Specifically, the above S6 includes three steps of S601, S602 and S603:

[0102] S601, merge the key plot and general plot data in S5 to form the maximum height scheme of the plot, which is in meters, such as Figure 7 as shown;

[0103] S602, input the generated maximum height scheme into the holographic sand table platform, automatically position in the holographic sand table background through geographic spatial coordinate information, generate a three-dimensional model of the plot height and display;

[0104] S603, as shown in the figure, the designer adjusts the plot height by the method in S402 to form the final height scheme. Figure 8

[0105] Specifically, the above S7 includes two steps of S701 and S702:

[0106] S701, import the upper planning index library to the holographic sand table platform, wherein the upper planning index library includes a plot height limit index;

[0107] S702, using a computer workstation with a computing power of 2.5 petaFLOPS and a memory of 320GB, calculate and judge whether the height of the site plot meets the requirements of the upper planning index library, if yes, export the result to generate engineering drawings with a scale of 1:1000, including plot information, and display on the holographic sand table, if not, return to S6 for adjustment and output the result again.

[0108] The application solves the problems of traditional manual scheme design process, such as complicated process, long working period, poor interactive design, high labor cost and the like, and further improves the efficiency of plot height generation; integrates the designer's thinking and computer logic, realizes the height form scheme generation method combining from top to bottom and from bottom to top, and responds to the plot height scheme generation of urban design in more scenes.

[0109] Although the embodiments of the application have been shown and described, it can be understood by those of ordinary skill in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the application, and the scope of the application is defined by the appended claims and their equivalents.​

Claims

1. A city plot height self-adaptive generation and demonstration platform based on an artificial intelligence algorithm, characterized in that, Comprise the following steps: S1, collect site geographic space information data and convert it into grid information, calculate the characteristic value of each plot; S2, collect case geographic space information data and convert it into grid information and import into the city design case library; S3, train the conditional adversarial generation network model through the city design scheme case library and import the site grid information in step S1, generate the site space structure grid information and convert it into vector data, and divide the plot structure height potential value according to the plot centroid and the space structure buffer distance; S4, import the plot structure height potential value distribution map into the holographic sand table display device, and select the design structure scheme; S5, divide the plot into key plots and general plots according to the plot structure height potential value in S4 and generate the height; S6, output the height result generated in S5 to the holographic sand table device for running and display, generate a three-dimensional height model, select the site plot in S5 and adjust the height of the site plot to obtain the final scheme; S7, import the upper planning index library, and judge whether the height of the site plot meets the index requirements, if yes, output the plot height index table, if not, return to S6 for adjustment and output again; The specific steps of S1 are: S101, use a computer workstation to obtain the geographic space information data of the site through a public space information platform, save the site space information and import it into geographic information processing software; S102, calculate the characteristic index of each plot through the geographic information processing software, the characteristic index includes the land area, the circumference, the height potential value of the plot function, and the height potential value of the plot function is valued by the plot land function; S103, convert the site geographic space information data into grid information; The specific steps of S5 are: S501, according to the selected space structure scheme in S4, and according to the plot structure height potential value in S4, divide the key plot and the general plot; S502, normalize and weight the height potential value of the key plot plot structure in S501 and the height potential value of the plot function in S102 to obtain the final height potential value of the key plot, the weight of the two is 50%, the calculation formula is: Wherein, Pf(i) is the height potential value of the plot function, P(i) is the final height potential value of the key plot, and Ps(i) is the height potential value of the key plot plot structure; S503, automatically match the height distribution frequency histogram in S2 and the final height potential value of the key plot through the integral mapping algorithm, the mapping relationship between the height Hi of plot i and the final height potential value P(i) is: According to the above correspondence and the function formula f(x), the height Hi is calculated, wherein P(i) is the final height potential value, Hi is the height of plot i, and the indefinite function f() is obtained according to the curve change of the height distribution frequency histogram; S504. Based on the area and functional information of a typical plot of land, the KNN algorithm is used to automatically match the height of the plot in the S2 urban design case library that has the same function, an area-to-size ratio between 0.8 and 1.2, and is closest to 1. If a match fails, the plot is assigned the average maximum height of its neighboring plots. Plots that are adjacent on one side are considered neighboring plots. The average maximum height is calculated as the area of ​​each plot. Maximum height of each plot / total area of ​​the plot.

2. The AI algorithm based urban plot height adaptive generation and demonstration platform according to claim 1, characterized in that, The specific steps of S2 are: S201, collect the geographic spatial information data of the urban design case by the method in step S101, collect the land use function and spatial structure grid information by the urban design text and drawing, and the spatial structure information includes core point element grid information and axis element grid information; S202, input the building height information of the urban design case collected by the surveying and mapping unmanned aerial vehicle into the geographic information processing software, calculate the highest building height of each plot as the plot height, generate a height distribution frequency histogram by a data statistics command, the x-axis is the plot height, and the y-axis is the height frequency change, wherein the plot height of the x-axis is meters; S203, extract the core points and axis elements in the spatial structure grid information of the urban design case by the Hog image feature extraction and recognition algorithm and convert them into pictures, wherein the core points and the axis lines are converted into grid information; S204, import the geographic spatial information data and the grid information of the urban design case into the urban design case library and store them.

3. The AI algorithm based urban plot height adaptive generation and demonstration platform according to claim 2, characterized in that, The specific steps of S3 are: S301, machine learning training is performed on the mapping features of the land use function grid information and the spatial structure grid information in S2, a mapping relationship between the two is generated, and a conditional adversarial generation network model is constructed; S302, input the grid information of the site in S1 into the conditional adversarial generation network model in S301 to generate a site multi-scheme spatial structure diagram; S303, extract the core points and axis grid information in the site multi-scheme spatial structure diagram by the Hog image feature extraction and recognition algorithm, and convert them into vector data; S304, according to the buffer distance between the plot centroid and the spatial structure, the height potential value of the plot structure is assigned.

4. The AI algorithm based urban plot height adaptive generation and demonstration platform according to claim 3, characterized in that, The specific steps of S4 are: The height potential value of the plot structure is output in VR format and displayed on the holographic digital sand table platform, and scheme comparison is performed to obtain a spatial structure scheme.

5. The AI algorithm based urban plot height adaptive generation and demonstration platform according to claim 4, characterized in that, The specific steps of S6 are: S601, merge the key plot and general plot data in S5 to form a highest height scheme of the plot, with the unit being meters; S602, input the generated highest height scheme into the holographic sand table platform, automatically position it in the holographic sand table background through geographic spatial coordinate information, generate a plot height three-dimensional model and display it; S603, adjust the plot height in the plot height three-dimensional model generated in S602 to form a final height scheme.

6. The AI algorithm based urban plot height adaptive generation and demonstration platform according to claim 5, characterized in that, The specific steps of S7 are: S701, import the upper planning index library to the holographic sand table platform, and the upper planning index library includes plot height limiting index; S702, use a computer workstation to calculate and determine whether the height of the site plot meets the requirements of the upper planning index library, if yes, output the result to generate engineering drawings and display them on the holographic sand table, if not, return to S6 for adjustment and then output the result again.

7. The AI algorithm based urban plot height adaptive generation and demonstration platform according to claim 1, characterized in that, The height potential value of the plot function in S102 is assigned by the plot land use function, and the assignment rule is: The commercial business mixed land is valued as 9, the commercial land is valued as 8, the commercial-residential mixed land is valued as 8, the commercial land is valued as 7, the scientific research land is valued as 7, the administrative office land is valued as 6, the cultural facility land is valued as 5, the school education land is valued as 4, the sports land is valued as 5, the medical and health land is valued as 5, the second-class residential land is valued as 4, the industrial land is valued as 3, the public facility land is valued as 2, and the storage land is valued as 1. The classification standard is classified according to the types in the urban land classification and the planning construction land standard.

8. The AI algorithm based urban plot height adaptive generation and demonstration platform according to claim 3, characterized in that, The valuation rule in the S304 is: The valuation is obtained by adding the distance between the mass center of the land and the nearest core point element and the distance between the land and the nearest axis element. The distance between the mass center of the land and the nearest core point element is less than 200m, 5 points are scored, greater than or equal to 200m and less than 400m, 3 points are scored, greater than or equal to 400m and less than 800m, 3 points are scored, and greater than or equal to 800m, 1 point is scored. The distance between the mass center of the land and the nearest axis element is less than 200m, 4 points are scored, greater than or equal to 200m and less than 600m, 2 points are scored, and greater than or equal to 600m, 0 points are scored.

Images