A method and apparatus for spatial decision making for skyscraper site selection

Abstract

The application discloses a skyscraper site selection space decision method and device. The candidate area is gridded, the distance from each grid to a landmark and the Gi*Z-score value of an interest point are assigned to the grid, and a skyscraper construction candidate area A is obtained; the candidate area A is input into two skyscraper site selection evaluation prediction models, and two groups of important indexes and target areas are output. The important indexes are intersected to obtain key indexes; the target areas are unionized to obtain a skyscraper construction prediction area; the key indexes are combined to obtain index combinations with different functional preferences; for each index combination, corresponding grid areas with Gi*Z-score values greater than or equal to a preset value are screened out from the prediction area, the grid areas are intersected to obtain a high space welfare skyscraper construction site selection area under the index combination, and then skyscraper construction site selection areas under different index combinations are obtained, so that the technical problems of strong subjectivity of index selection and lack of multi-scenario simulation analysis are solved.

Description

Technical Field

[0001] This invention relates to the field of skyscraper site selection technology, and in particular to a spatial decision-making method and apparatus for skyscraper site selection. Background Technology

[0002] Skyscrapers are an indispensable part of urban development, especially in China's urbanization process. With an increasing rural population migrating to Chinese cities, skyscrapers will undoubtedly remain a key solution to alleviate land use tensions. However, existing methods and technologies for skyscraper site selection and layout often face the following problems:

[0003] First, the selection of indicators is highly subjective. When selecting sites for skyscrapers, the selection of influencing factors often relies on experience and subjective judgment. This method lacks objective quantitative standards and is prone to uncertainty and randomness in the site selection results.

[0004] Second, there is a lack of multi-scenario simulation analysis: current site selection methods usually only consider a single scenario and lack comprehensive consideration of different functional requirements and scenarios, resulting in site selection results that cannot fully reflect the actual needs and future development trends of the city. Summary of the Invention

[0005] This invention provides a spatial decision-making method and apparatus for skyscraper site selection, which solves the technical problems of strong subjectivity in indicator selection and lack of multi-scenario simulation analysis in the prior art.

[0006] This invention provides a spatial decision-making method for skyscraper site selection, comprising:

[0007] The candidate area is gridded, and the distance from each grid to the preset landmark is calculated;

[0008] Calculate the Gi*Z-score values ​​of the preset M points of interest data;

[0009] The distance and the Gi*Z-score value are assigned to each grid after gridding to obtain a set A of candidate areas for skyscraper construction containing M indicators;

[0010] The set A of candidate areas for skyscraper construction containing M indicators is input into the first skyscraper site selection evaluation and prediction model and the second skyscraper site selection evaluation and prediction model, respectively. The first skyscraper site selection evaluation and prediction model outputs P important indicators and a first target area, and the second skyscraper site selection evaluation and prediction model outputs H important indicators and a second target area.

[0011] By taking the intersection of the P important indicators and the H important indicators, N key indicators are obtained;

[0012] The union of the first target region and the second target region is used to obtain the predicted area for skyscraper construction.

[0013] By combining the N key indicators, we can obtain indicator combinations with different functional preferences;

[0014] For each combination of indicators for the aforementioned functional preferences, the corresponding grid areas whose Gi*Z-score values ​​for each key indicator are greater than or equal to preset values ​​are selected from the skyscraper construction prediction area. The intersection of the corresponding grid areas is then used to obtain the high-space welfare skyscraper construction site selection area under the combination of indicators for that functional preference.

[0015] Specifically, the gridding of the candidate regions includes:

[0016] Based on the preset service radius of the skyscraper, the candidate area is gridded using a grid spatial representation method.

[0017] Specifically, before assigning the distance and the Gi*Z-score value to each of the meshed cells, the method further includes:

[0018] The distance and the Gi*Z-score value are standardized, and the standardized distance and Gi*Z-score value are assigned to each grid cell after meshing.

[0019] Specifically, the step of inputting the set A of candidate areas for skyscraper construction, which includes M indicators, into the first skyscraper site selection assessment and prediction model and the second skyscraper site selection assessment and prediction model respectively includes:

[0020] Calculate the information contribution parameter of each indicator in the set A of candidate areas for skyscraper construction containing M indicators;

[0021] Remove the indicators whose information contribution parameters are less than a set threshold from the set A of candidate areas for skyscraper construction containing M indicators, and obtain a new set B of candidate areas for skyscraper construction.

[0022] The new set of candidate areas for skyscraper construction, B, is input into the first skyscraper site selection assessment and prediction model and the second skyscraper site selection assessment and prediction model, respectively.

[0023] Specifically, it also includes:

[0024] The Gi*Z-score values ​​of each key indicator in the high-space welfare skyscraper construction site selection area under the different combinations of functional preference indicators are projected into a two-dimensional graph, and the magnitude of the Gi*Z-score value is represented by a bipolar color mapping. The high-space welfare skyscraper construction site selection area under the different combinations of functional preference indicators is also identified by different colors and icons.

[0025] The present invention also provides a spatial decision-making device for skyscraper site selection, comprising:

[0026] The gridding module is used to grid the candidate area and calculate the distance from each grid to the preset landmark;

[0027] The calculation module is used to calculate the Gi*Z-score values ​​of the preset M points of interest data;

[0028] The grid assignment module is used to assign the distance and the Gi*Z-score value to each grid after gridding, so as to obtain a set A of candidate areas for skyscraper construction containing M indicators;

[0029] The prediction module is used to input the set A of candidate areas for skyscraper construction containing M indicators into the first skyscraper site selection evaluation and prediction model and the second skyscraper site selection evaluation and prediction model respectively. The first skyscraper site selection evaluation and prediction model outputs P important indicators and a first target area, and the second skyscraper site selection evaluation and prediction model outputs H important indicators and a second target area.

[0030] The important indicator intersection module is used to find the intersection of the P important indicators and the H important indicators to obtain N key indicators;

[0031] The target region union module is used to take the union of the first target region and the second target region to obtain the skyscraper construction prediction region;

[0032] The key indicator combination module is used to combine the N key indicators to obtain indicator combinations with different functional preferences.

[0033] The site selection decision module is used to select, for each combination of indicators of the aforementioned functional preferences, the corresponding grid areas whose Gi*Z-score values ​​of each key indicator are greater than or equal to preset values ​​from the skyscraper construction prediction area, and take the intersection of the corresponding grid areas to obtain the high space welfare skyscraper construction site selection area under the indicator combination of the functional preferences.

[0034] Specifically, the meshing module includes:

[0035] The gridding execution unit is used to grid the candidate area according to the preset service radius of the skyscraper using a grid spatial representation method;

[0036] The distance measurement unit is used to measure the distance from each grid to a preset landmark.

[0037] Specifically, the grid assignment module is used to standardize the distance and the Gi*Z-score value, and assign the standardized distance and Gi*Z-score value to each grid after gridding, so as to obtain a set A of candidate areas for skyscraper construction containing M indicators.

[0038] Specifically, the prediction module includes:

[0039] The information contribution parameter calculation unit is used to calculate the information contribution parameters of each indicator in the set A of candidate areas for skyscraper construction containing M indicators;

[0040] The indicator filtering unit is used to delete indicators whose information contribution parameters are less than a set threshold from the skyscraper construction candidate area set A containing M indicators, and obtain a new skyscraper construction candidate area set B.

[0041] The prediction execution unit is used to input the new set of candidate areas for skyscraper construction B into the first skyscraper site selection evaluation and prediction model and the second skyscraper site selection evaluation and prediction model, respectively. The first skyscraper site selection evaluation and prediction model outputs P important indicators and a first target area, and the second skyscraper site selection evaluation and prediction model outputs H important indicators and a second target area.

[0042] Specifically, it also includes:

[0043] The visualization module is used to project the Gi*Z-score values ​​of each key indicator in the high-space-welfare skyscraper construction site selection area under different combinations of functional preference indicators into a two-dimensional graph, and to represent the magnitude of the Gi*Z-score values ​​through bipolar color mapping. It also identifies the high-space-welfare skyscraper construction site selection area under different combinations of functional preference indicators through different colors and icons.

[0044] One or more technical solutions provided in this invention have at least the following technical effects or advantages:

[0045] 1. Grid the candidate areas, assigning the distance to a preset landmark and the Gi*Z-score values ​​of M preset points of interest to each grid cell, resulting in a set A of candidate areas for skyscraper construction containing M indicators. Input this set A into the first and second skyscraper site selection assessment and prediction models, respectively, to obtain two sets of important indicators and target areas. Take the intersection of these two sets of important indicators to obtain N key indicators. Take the union of the two sets of target areas to obtain the predicted area for skyscraper construction. By combining N key indicators, different functional preference indicator combinations are obtained. For each functional preference indicator combination, grid areas corresponding to the Gi*Z-score values ​​of each key indicator greater than or equal to preset values ​​are selected from the skyscraper construction prediction area. The intersection of these grid areas yields the high-space-welfare skyscraper construction site selection area under that functional preference indicator combination. This process, through multi-scenario simulation analysis combined with different functional requirements, forms multiple decision-making scenarios and provides diversified skyscraper site selection solutions. Therefore, this invention effectively solves the technical problems of strong subjectivity in indicator selection and lack of multi-scenario simulation analysis in existing technologies.

[0046] 2. Based on the information contribution of M indicators to skyscraper site selection decisions, key indicators were screened out through quantitative analysis methods, which further improved the scientificity and reliability of skyscraper site selection decisions.

[0047] 3. The results of skyscraper site selection are visualized using geospatial and statistical data, which can intuitively present the distribution of skyscraper site selection areas under different scenarios, helping decision-makers to make optimal decisions quickly in complex data.

[0048] Through the above-mentioned technical means, the present invention can significantly improve the scientificity, accuracy and applicability of skyscraper site selection, and provide strong support for urban planning and construction. Attached Figure Description

[0049] Figure 1 A flowchart of a spatial decision-making method for skyscraper site selection provided in an embodiment of the present invention;

[0050] Figure 2 A block diagram of a spatial decision-making device for skyscraper site selection provided in an embodiment of the present invention;

[0051] Figure 3 A flowchart illustrating the workflow of a spatial decision support system for skyscraper site selection provided in this embodiment of the invention;

[0052] Figure 4An analysis result diagram of the "Rank" component in the spatial decision support system for skyscraper site selection provided in an embodiment of the present invention;

[0053] Figure 5 This is a schematic diagram illustrating the selection of indicator combinations in an example of the present invention;

[0054] Figure 6 These are visualizations of skyscraper construction under different scenarios in this invention. Detailed Implementation

[0055] The present invention provides a spatial decision-making method and apparatus for skyscraper site selection, which solves the technical problems of strong subjectivity in indicator selection and lack of multi-scenario simulation analysis in the prior art.

[0056] The technical solutions in the embodiments of the present invention are intended to solve the above-mentioned technical problems, and the overall approach is as follows:

[0057] 1. Construction of candidate areas for skyscraper construction: The construction of candidate areas for skyscraper construction mainly refers to: identifying urban areas where skyscrapers need to be built, firstly using a grid spatial representation method to grid the urban areas, then collecting POI data, the number of skyscrapers, and urban landscape data of the city, and calculating the clustering and dispersion of POIs in each grid and their distance to the urban landscape, in order to generate a candidate area set for skyscraper construction.

[0058] 2. Influencing Factor Judgment and Site Selection Prediction; Influencing factor judgment and site selection prediction mainly refers to: learning the relationship between different indicators and skyscraper construction site selection based on decision tree and random forest models, and using the trained decision tree and random forest models to evaluate and predict candidate areas. Specifically, the intersection method is used to obtain N key indicators that have a significant impact on skyscraper site selection; the union method is used to obtain all predicted areas suitable for skyscraper construction.

[0059] 3. Multi-scenario simulation and prediction area visualization: Scenario simulation and prediction area visualization mainly refers to: forming multiple decision scenarios in a cross-combination manner based on the Z-score and POI category of N key indicators that have a significant impact on skyscraper site selection, and displaying skyscraper layout schemes under multiple scenarios.

[0060] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0061] like Figure 1 As shown, the spatial decision-making method for skyscraper site selection provided in this embodiment of the invention includes:

[0062] Step S110: Grid the candidate area and calculate the distance from each grid to the preset landmark;

[0063] Specifically, the candidate regions are gridded, including:

[0064] Based on the preset service radius of skyscrapers, the candidate areas are gridded using a grid spatial representation method, and grids that cannot be built on, such as lakes and rivers, are deleted.

[0065] Calculate the distance from each grid cell to the preset landmark, including:

[0066] The distances from each grid point to urban landscapes such as mountains, rivers, lakes, green spaces, forests, and the city center are calculated sequentially using the following formula:

[0067] d = (x2 - x1) 2 +(y2-y1) 2 ; where (x1, y1) and (x2, y2) are the coordinates of two points.

[0068] Step S120: Calculate the Gi*Z-score values ​​of the preset M points of interest data;

[0069] This step is explained in detail, calculating the Gi*Z-score values ​​of the preset M points of interest data, including:

[0070] The Gi*Z-score values ​​of M points of interest (POIs) of various types, including medical services, scenic spots, office spaces, shopping areas, transportation facilities, financial services, science and education, commercial facilities, living services, sports and leisure, and accommodation services, are calculated in sequence.

[0071] Specifically, the formula for calculating the Gi*Z-score is as follows: in, It is the average of the attribute values. S is the standard deviation of the attribute values. x j It is the attribute value at position j, w ij is the spatial weight between position i and position j. n is the total number of points in the dataset.

[0072] Step S130: Assign the distance from each grid to the preset landmark and the Gi*Z-score value to each grid after gridding to obtain a set A of candidate areas for skyscraper construction containing M indicators;

[0073] To eliminate the influence of dimensions between data, before assigning the distance from each grid to the preset landmark and the Gi*Z-score value to each grid after gridding, the following steps are also included:

[0074] The distance and Gi*Z-score value from each grid to the preset landmark are standardized, and the standardized distance and Gi*Z-score value are assigned to each grid after gridding.

[0075] Specifically, the standardized processing formula is as follows:

[0076]

[0077] Where Z is the standardized Z-score, X is the data point value that needs to be standardized, μ is the mean of the dataset, and σ is the standard deviation of the dataset.

[0078] Step S140: Input the set A of candidate areas for skyscraper construction containing M indicators into the first skyscraper site selection evaluation and prediction model and the second skyscraper site selection evaluation and prediction model respectively. The first skyscraper site selection evaluation and prediction model outputs P important indicators and the first target area, and the second skyscraper site selection evaluation and prediction model outputs H important indicators and the second target area.

[0079] Specifically, the set A of candidate areas for skyscraper construction, containing M indicators, is input into the first skyscraper site selection assessment and prediction model and the second skyscraper site selection assessment and prediction model, respectively, including:

[0080] Calculate the information contribution parameter of each indicator in the set A of candidate areas for skyscraper construction containing M indicators; in this embodiment, the information contribution parameter is at least one of information gain, Gini coefficient, chi-square test, ANOVA F value, and correlation coefficient.

[0081] From the set of candidate areas for skyscraper construction A containing M indicators, remove indicators whose information contribution parameters are less than a set threshold to obtain a new set of candidate areas for skyscraper construction B. Specifically, calculate the information gain, Gini coefficient, chi-square test, ANOVA F value, and correlation coefficient of the M indicators in sequence. By comparing the average value of the five parameters, determine the information contribution of each indicator to the decision-making. Remove the S indicators with smaller average values ​​to form a new set of candidate areas for skyscraper construction B containing MS indicators.

[0082] The new set of candidate areas for skyscraper construction, B, is input into the first skyscraper site selection assessment and prediction model and the second skyscraper site selection assessment and prediction model, respectively.

[0083] Specifically, the selection methods for the first and second skyscraper site selection assessment and prediction models are as follows:

[0084] Determine whether all grids in the candidate area set A for skyscraper construction contain Z-score values, and handle missing values ​​by filling grids that do not contain Z-score values ​​with specific values;

[0085] Remove the indicators whose information contribution parameters are less than a set threshold from the set of candidate areas for skyscraper construction after processing for missing values, and obtain a new set of candidate areas for skyscraper construction, B.

[0086] Stratified sampling was used to extract 80% of the data in the candidate area set B for skyscraper construction as the training set. Each candidate area was labeled to record whether there was a skyscraper in each candidate area. Candidate areas with skyscrapers were labeled as positive (1), and candidate areas without skyscrapers were labeled as negative (0).

[0087] The labeled training set is input into three machine learning models—random forest, decision tree, and support vector machine—for cross-training. Three evaluation parameters—accuracy, recall, and F1 score—are obtained for each model. The fit of the three models is evaluated by comparing the average of the three parameters, and the two machine learning models with the larger average parameter values ​​are retained.

[0088] Step S150: Take the intersection of P important indicators and H important indicators to obtain N key indicators;

[0089] Step S160: Take the union of the first target area and the second target area to obtain the predicted area for skyscraper construction;

[0090] Step S170: Combine N key indicators to obtain indicator combinations with different functional preferences;

[0091] Step S180: For each combination of indicators for functional preferences, select the corresponding grid areas from the skyscraper construction prediction area where the Gi*Z-score value of each key indicator is greater than or equal to the preset value, and take the intersection of the corresponding grid areas to obtain the high-space welfare skyscraper construction site selection area under the indicator combination of functional preferences.

[0092] Specifically, the N key indicators that have a significant impact on skyscraper site selection are classified according to POI categories to form indicator combinations a, b, and c with different functional preferences.

[0093] Based on the Z-score value of each indicator in indicator combination a, the resulting predicted areas are classified. Specifically, the Z-score values ​​of each indicator in indicator combination a are sorted, and the predicted areas with the highest Z-score values ​​for each indicator are selected. The intersection of these predicted areas is named the "High-Speed ​​Welfare Construction Site Selection Area". Then, the predicted areas with the highest Z-score values ​​for each indicator are selected, and the intersection of these predicted areas is named the "Second-High-Speed ​​Welfare Construction Site Selection Area". The determination of the high-speed and second-high-speed welfare construction site selection areas for indicator combinations b and c is completed sequentially, resulting in a total of 6 different scenarios for skyscraper site selection predicted areas.

[0094] The predicted areas for skyscraper site selection under six different scenarios are geographically visualized, and the results of skyscraper site selection are displayed through geospatial and statistical visualization, presenting the distribution schemes of skyscraper site selection areas under different scenarios in an intuitive way.

[0095] Specifically, it also includes:

[0096] The Gi*Z-score values ​​of key indicators in the high-space-welfare skyscraper construction site selection area under different combinations of indicator preferences are projected into a two-dimensional graph, and the magnitude of the Gi*Z-score values ​​is represented by bipolar color mapping. Different colors and icons are also used to identify the high-space-welfare skyscraper construction site selection area under different combinations of indicator preferences.

[0097] In this embodiment, the numerical selection range of high and second-high welfare spaces can be flexibly adjusted by the decision-maker according to the actual decision-making scenario. This allows for the automatic re-selection of "high welfare space construction site areas" and "second-high welfare space construction site areas" based on new parameters, and the layout display can be updated in real time to ensure that users can receive rapid visual feedback for each interactive operation.

[0098] like Figure 2 As shown, the spatial decision-making device for skyscraper site selection provided in this embodiment of the invention includes:

[0099] The gridding module 100 is used to grid the candidate area and calculate the distance from each grid to the preset landmark.

[0100] Specifically, the meshing module 100 includes:

[0101] The gridded execution unit is used to grid the candidate area according to the preset service radius of the skyscraper, using a grid spatial representation method, and delete grids that cannot be built on, such as lakes and rivers.

[0102] The distance measurement unit is used to measure the distance from each grid to a preset landmark.

[0103] In this embodiment, the distance measurement unit is specifically used to sequentially measure the distance from each grid to urban landscapes such as mountains, rivers, lakes, green spaces, forests, and the city center. The calculation formula is: d = (x2 - x1) 2 +(y2-y1) 2 ; where (x1, y1) and (x2, y2) are the coordinates of two points.

[0104] The calculation module 200 is used to calculate the Gi*Z-score values ​​of the preset M points of interest data;

[0105] In this embodiment, the calculation module 200 is specifically used to sequentially calculate the Gi*Z-score values ​​of M points of interest (POIs) of various types, including medical services, scenic spots, office spaces, shopping malls, transportation facilities, financial services, science and education, commercial facilities, living services, sports and leisure, and accommodation services. The formula for calculating the Gi*Z-score value is as follows: in, It is the average of the attribute values. S is the standard deviation of the attribute values. x j It is the attribute value at position j, w ij is the spatial weight between position i and position j. n is the total number of points in the dataset.

[0106] The grid assignment module 300 is used to assign the distance from each grid to the preset landmark and the Gi*Z-score value to each grid after gridding, so as to obtain a set A of candidate areas for skyscraper construction containing M indicators;

[0107] Specifically, the grid assignment module 300 is used to standardize the distance and Gi*Z-score value of each grid to the preset landmark, and assign the standardized distance and Gi*Z-score value to each grid after gridding, so as to obtain a set A of candidate areas for skyscraper construction containing M indicators.

[0108] The prediction module 400 is used to input the set A of candidate areas for skyscraper construction containing M indicators into the first skyscraper site selection evaluation and prediction model and the second skyscraper site selection evaluation and prediction model respectively. The first skyscraper site selection evaluation and prediction model outputs P important indicators and the first target area, and the second skyscraper site selection evaluation and prediction model outputs H important indicators and the second target area.

[0109] Specifically, the prediction module 400 includes:

[0110] The information contribution parameter calculation unit is used to calculate the information contribution parameters of each indicator in the set A of candidate areas for skyscraper construction containing M indicators.

[0111] The indicator screening unit is used to delete indicators whose information contribution parameters are less than a set threshold from the candidate area set A for skyscraper construction containing M indicators, and obtain a new candidate area set B for skyscraper construction. In this embodiment, the information contribution parameter is at least one of information gain, Gini coefficient, chi-square test, ANOVA F value, and correlation coefficient.

[0112] In this embodiment, the indicator screening unit is specifically used to sequentially calculate the information gain, Gini coefficient, chi-square test, ANOVA F value, and correlation coefficient of M indicators. By comparing the average value of the five parameters, the information contribution of each indicator to the decision is determined, and the S indicators with smaller average values ​​are eliminated, thereby forming a new set of candidate areas for skyscraper construction B containing MS indicators.

[0113] The prediction execution unit is used to input the new set of candidate areas for skyscraper construction B into the first skyscraper site selection evaluation and prediction model and the second skyscraper site selection evaluation and prediction model respectively. The first skyscraper site selection evaluation and prediction model outputs P important indicators and the first target area, and the second skyscraper site selection evaluation and prediction model outputs H important indicators and the second target area.

[0114] The important indicator intersection module 500 is used to find the intersection of P important indicators and H important indicators to obtain N key indicators;

[0115] The target area union module 600 is used to take the union of the first target area and the second target area to obtain the skyscraper construction prediction area;

[0116] The key indicator combination module 700 is used to combine N key indicators to obtain indicator combinations with different functional preferences.

[0117] The site selection decision module 800 is used to select the corresponding grid areas from the skyscraper construction prediction area where the Gi*Z-score value of each key indicator is greater than or equal to the preset value for each combination of indicators of functional preference. The intersection of the corresponding grid areas is then used to obtain the high space welfare skyscraper construction site selection area under the indicator combination of that functional preference.

[0118] To provide geospatial and statistical visualization of skyscraper site selection results and to intuitively present the regional distribution schemes of skyscraper site selection under different scenarios, the following also includes:

[0119] The visualization module projects the Gi*Z-score values ​​of key indicators in the high-space-welfare skyscraper construction site selection area under different combinations of indicator preferences into a two-dimensional graph. It uses bipolar color mapping to represent the magnitude of the Gi*Z-score values ​​and uses different colors and icons to identify the high-space-welfare skyscraper construction site selection area under different combinations of indicator preferences.

[0120] Based on the skyscraper site selection method and apparatus provided in this invention, a spatial decision support system for skyscraper site selection is also designed. This system adopts a B / S architecture, with the front-end built using HTML5, CSS, and JavaScript technologies, the back-end using the Python and Django frameworks, the database using PostgreSQL, and geographic information processing using ArcGIS Server. Figure 3 As shown, from the perspective of data acquisition, the basic data mainly consists of POI big data obtained through web crawling technology via the API (Application Programming Interface https: / / api.map.baidu.com) provided by Baidu Maps, as well as vector data of urban landscapes and city centers, and skyscraper data provided by CTBUH (Global Tall Building Database www.skyscrapercenter.com).

[0121] From the basic data processing flow, firstly, based on the service radius of skyscrapers, a grid with a spatial resolution of 500 meters is selected as the unit for spatial analysis. Grids that are unsuitable for construction, such as lakes and rivers, are deleted, and the remaining grids are the candidate areas for skyscraper construction. Then, using GIS hotspot analysis tools, Gi*Z-score data of POIs are obtained. Using distance calculation tools, the distance data from each candidate area to the nearest urban landscape and city center is obtained, and Z-Score standardization is performed on them to eliminate the influence of dimensions between data. The standardized Z-Score values ​​are then assigned to the corresponding grids. After data processing, this embodiment has 2759 candidate areas and 18 spatial indicators, excluding descriptive data.

[0122] Next, the multi-source spatial data with spatial attributes generated by GIS tools is input into the Orange platform for missing value filling and indicator filtering. Missing value handling primarily employs a specific value filling method, where the specific value is 0. Specifically, for grids with missing spatial features, the `Imput` component is used to fill the missing values ​​with 0. Simultaneously, 2759 candidate areas containing 18 spatial indicators are input into the "Rank" component. Five parameters—information gain, Gini coefficient, chi-square test, ANOVA F-value, and correlation coefficient—are calculated for each of the 18 indicators. Based on the average value of these parameters, the five indicators with the highest average values ​​are selected. The four indicators with the lowest average values ​​are removed, retaining the top 14 indicators. This filters the candidate areas for skyscraper construction with 18 indicator characteristics into candidate areas for skyscraper construction with 14 indicator characteristics. Figure 4 As shown.

[0123] A stratified sampling method was used to extract 80% of the data as the training set to train the machine learning model. First, each candidate region was labeled to record whether a skyscraper exists in each candidate region. Candidate regions with skyscrapers were labeled as positive (1), and candidate regions without skyscrapers were labeled as negative (0). Three machine learning models, Random Forest (RF), Decision Tree (Tree), and Support Vector Machine (SVM), were selected to cross-train the training dataset. The accuracy, recall, and F1 score of the three models were obtained respectively. The fit of the three models was evaluated by comparing the average of the three parameters. It was confirmed that Decision Tree and Random Forest were the two models suitable for this spatial decision-making, namely the first skyscraper location evaluation and prediction model and the second skyscraper location evaluation and prediction model.

[0124] Next, decision tree and random forest machine learning models were used for prediction. The decision tree identified eight main influencing factors affecting site selection decisions, predicting 148 positively classified buildable areas. The random forest also identified eight main influencing factors, predicting 124 positively classified buildable areas. By taking the intersection of these factors, eight influencing factors affecting skyscraper site selection decisions were determined. By taking the union of these factors, suitable buildable areas for skyscraper construction were determined, resulting in 170 candidate areas being predicted as suitable for skyscraper construction.

[0125] like Figure 5As shown, based on the POI categories of eight indicators, four indicators—distance to the city center, office space, shopping, and dining services—are combined into a "business-first" functional group; four indicators—distance to the city center, dining services, medical services, and science and education—are combined into a "life-first" functional group; and four indicators—distance to urban landscape, leisure and entertainment, science and education, and medical services—are combined into a "public service-first" functional group. Simultaneously, the Z-Score values ​​within the grid are sorted from highest to lowest. The predicted areas with the top 30% Z-Score values ​​are designated as high-spatial-welfare functional areas, and the predicted areas with the top 70% Z-Score values ​​are designated as second-highest-spatial-welfare functional areas. Through cross-combination, the three functional groups are paired with the two types of predicted areas to form six multi-category, hierarchical skyscraper spatial location schemes.

[0126] The geospatial and statistical data visualization of skyscraper site selection decisions under six different scenarios is presented, intuitively showing the regional distribution schemes of skyscraper site selection under different scenarios. For example... Figure 6 As shown, SI represents the prediction result under the "Business Priority - High Space Welfare Zone" decision scenario, SII represents the prediction result under the "Business Priority - Second High Space Welfare Zone" decision scenario, SIII represents the prediction result under the "Living Priority - High Space Welfare Zone" decision scenario, SIV represents the prediction result under the "Living Priority - Second High Space Welfare Zone" decision scenario, SV represents the prediction result under the "Public Service Priority - High Space Welfare Zone" decision scenario, and SVI represents the prediction result under the "Public Service Priority - High Space Welfare Zone" decision scenario.

[0127] In summary, this invention addresses the shortcomings of existing technologies by proposing a skyscraper site selection method that achieves the following technical effects: First, it provides an objective indicator screening method: This invention uses quantitative methods such as information gain, Gini coefficient, chi-square test, ANOVA F-value, and correlation coefficient to objectively evaluate and rank various indicators, selecting those that contribute significantly to site selection decisions, reducing the influence of subjective judgment, and improving the scientific rigor and reliability of the site selection results. Second, it combines multiple machine learning models: This invention employs three machine learning models—random forest, decision tree, and support vector machine—and selects the optimal model for site selection prediction through cross-training and evaluation, improving the accuracy and stability of the prediction. The model performance is comprehensively evaluated using multiple evaluation indicators such as accuracy, recall, and F1 score, ensuring the reliability of the site selection prediction. Third, it incorporates multi-scenario simulation and comprehensive analysis: This invention uses multi-scenario simulation analysis to form multiple decision scenarios through cross-combinations of different functional preferences (such as business priority, living priority, and public service priority), providing diversified skyscraper site selection schemes to meet the needs of multifaceted urban development. Furthermore, different decision-making scenarios are input into the geographic visualization module, intuitively presenting the distribution schemes of skyscraper site selection areas under different scenarios, thereby improving the transparency and interpretability of decision-making. Through the above technical means, the embodiments of the present invention can significantly improve the scientificity, accuracy, and applicability of skyscraper site selection, providing strong support for urban planning and construction.

[0128] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0129] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0130] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0131] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0132] Any aspects of this invention not described in detail in the embodiments are well-known techniques to those skilled in the art. Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this invention and not to limit it. Although this invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of this invention without departing from the spirit and scope of this invention, and all such modifications and substitutions should be covered within the scope of the claims of this invention.

Claims

1. A spatial decision-making method for skyscraper site selection, characterized in that, include: The candidate area is gridded, and the distance from each grid to the preset landmark is calculated; Calculate the Gi*Z-score values ​​of the preset M points of interest data; The distance and the Gi*Z-score value are assigned to each grid after gridding to obtain a set A of candidate areas for skyscraper construction containing M indicators; The set A of candidate areas for skyscraper construction containing M indicators is input into the first skyscraper site selection evaluation and prediction model and the second skyscraper site selection evaluation and prediction model, respectively. The first skyscraper site selection evaluation and prediction model outputs P important indicators and a first target area, and the second skyscraper site selection evaluation and prediction model outputs H important indicators and a second target area. By taking the intersection of the P important indicators and the H important indicators, N key indicators are obtained; The union of the first target region and the second target region is used to obtain the predicted area for skyscraper construction. By combining the N key indicators, we can obtain indicator combinations with different functional preferences; For each combination of indicators for the aforementioned functional preferences, the corresponding grid areas whose Gi*Z-score values ​​for each key indicator are greater than or equal to preset values ​​are selected from the skyscraper construction prediction area. The intersection of the corresponding grid areas is then used to obtain the high-space welfare skyscraper construction site selection area under the combination of indicators for that functional preference.

2. The spatial decision-making method for skyscraper site selection as described in claim 1, characterized in that, The process of gridding the candidate regions includes: Based on the preset service radius of the skyscraper, the candidate area is gridded using a grid spatial representation method.

3. The spatial decision-making method for skyscraper site selection as described in claim 1, characterized in that, Before assigning the distance and the Gi*Z-score value to each of the meshed cells, the method further includes: The distance and the Gi*Z-score value are standardized, and the standardized distance and Gi*Z-score value are assigned to each grid cell after meshing.

4. The spatial decision-making method for skyscraper site selection as described in claim 1, characterized in that, The step of inputting the set A of candidate areas for skyscraper construction, which includes M indicators, into the first skyscraper site selection assessment and prediction model and the second skyscraper site selection assessment and prediction model, respectively, includes: Calculate the information contribution parameter of each indicator in the set A of candidate areas for skyscraper construction containing M indicators; Remove the indicators whose information contribution parameters are less than a set threshold from the set A of candidate areas for skyscraper construction containing M indicators, and obtain a new set B of candidate areas for skyscraper construction. The new set of candidate areas for skyscraper construction, B, is input into the first skyscraper site selection assessment and prediction model and the second skyscraper site selection assessment and prediction model, respectively.

5. The spatial decision-making method for skyscraper site selection as described in any one of claims 1-4, characterized in that, Also includes: The Gi*Z-score values ​​of each key indicator in the high-space welfare skyscraper construction site selection area under the different combinations of functional preference indicators are projected into a two-dimensional graph, and the magnitude of the Gi*Z-score value is represented by a bipolar color mapping. The high-space welfare skyscraper construction site selection area under the different combinations of functional preference indicators is also identified by different colors and icons.

6. A spatial decision-making device for skyscraper site selection, characterized in that, include: The gridding module is used to grid the candidate area and calculate the distance from each grid to the preset landmark; The calculation module is used to calculate the Gi*Z-score values ​​of the preset M points of interest data; The grid assignment module is used to assign the distance and the Gi*Z-score value to each grid after gridding, so as to obtain a set A of candidate areas for skyscraper construction containing M indicators; The prediction module is used to input the set A of candidate areas for skyscraper construction containing M indicators into the first skyscraper site selection evaluation and prediction model and the second skyscraper site selection evaluation and prediction model respectively. The first skyscraper site selection evaluation and prediction model outputs P important indicators and a first target area, and the second skyscraper site selection evaluation and prediction model outputs H important indicators and a second target area. The important indicator intersection module is used to find the intersection of the P important indicators and the H important indicators to obtain N key indicators; The target region union module is used to take the union of the first target region and the second target region to obtain the skyscraper construction prediction region; The key indicator combination module is used to combine the N key indicators to obtain indicator combinations with different functional preferences. The site selection decision module is used to select, for each combination of indicators of the aforementioned functional preferences, the corresponding grid areas whose Gi*Z-score values ​​of each key indicator are greater than or equal to preset values ​​from the skyscraper construction prediction area, and take the intersection of the corresponding grid areas to obtain the high space welfare skyscraper construction site selection area under the indicator combination of the functional preferences.

7. The spatial decision-making device for skyscraper site selection as described in claim 6, characterized in that, The meshing module includes: The gridding execution unit is used to grid the candidate area according to the preset service radius of the skyscraper using a grid spatial representation method; The distance measurement unit is used to measure the distance from each grid to a preset landmark.

8. The spatial decision-making device for skyscraper site selection as described in claim 6, characterized in that, The grid assignment module is specifically used to standardize the distance and the Gi*Z-score value, and assign the standardized distance and Gi*Z-score value to each grid after gridding, so as to obtain a set A of candidate areas for skyscraper construction containing M indicators.

9. The spatial decision-making device for skyscraper site selection as described in claim 6, characterized in that, The prediction module includes: The information contribution parameter calculation unit is used to calculate the information contribution parameters of each indicator in the set A of candidate areas for skyscraper construction containing M indicators; The indicator filtering unit is used to delete indicators whose information contribution parameters are less than a set threshold from the skyscraper construction candidate area set A containing M indicators, and obtain a new skyscraper construction candidate area set B. The prediction execution unit is used to input the new set of candidate areas for skyscraper construction B into the first skyscraper site selection evaluation and prediction model and the second skyscraper site selection evaluation and prediction model, respectively. The first skyscraper site selection evaluation and prediction model outputs P important indicators and a first target area, and the second skyscraper site selection evaluation and prediction model outputs H important indicators and a second target area.

10. The spatial decision-making device for skyscraper site selection as described in any one of claims 6-9, characterized in that, Also includes: The visualization module is used to project the Gi*Z-score values ​​of each key indicator in the high-space-welfare skyscraper construction site selection area under different combinations of functional preference indicators into a two-dimensional graph, and to represent the magnitude of the Gi*Z-score values ​​through bipolar color mapping. It also identifies the high-space-welfare skyscraper construction site selection area under different combinations of functional preference indicators through different colors and icons.

Images