Method and system for measuring and calculating coupling degree of green space structure and built environment

Abstract

The invention discloses a method and system for measuring and calculating the coupling degree of a green space structure and a built environment, and relates to the technical field of urban planning, landscape gardens and artificial intelligence. Through establishment of a coupling structure, based on a random forest regression model, a significant relation between urban observation data and a green space structure-built environment correlation index is established; urban green space performance evaluation is converted into a series of coupling values, the problems that an urban green space performance evaluation system is huge, indexes are redundant and quantification is difficult are solved, and the benefit evaluation accuracy and the measurement and calculation efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the fields of urban planning, landscape architecture and artificial intelligence technology, and in particular to a method and system for calculating the coupling degree between a green space structure and a built environment. Background Art

[0002] In the context of urban shrinkage, urban green space has become an important means to maintain urban vitality and improve the quality of life. However, the current addition of urban green space in the built environment is mostly based on green space performance evaluation based on expert opinions. There are defects such as difficulty in unifying indicators, excessive subjectivity and randomness, and difficulty in quantifying performance, which makes the specific practice of urban green space structural adjustment inefficient and lacks scientificity. Summary of the invention

[0003] In order to solve the deficiencies mentioned in the above background technology, the purpose of the present invention is to provide a method and system for measuring the coupling degree between green space structure and built environment.

[0004] In a first aspect, the purpose of the present invention can be achieved by the following technical solution: a method for measuring the coupling degree between green space structure and built environment, the method comprising the following steps:

[0005] Receive data related to sample cities and divide the sample cities into multiple analysis units; wherein the data related to sample cities include urban residential buildings, commercial buildings, industrial buildings, office buildings, urban green space locations and sizes, road network structure data, POI distribution data, and population distribution data;

[0006] Obtaining comprehensive characterization index data and correlation index change rate data of each analysis unit, performing data processing on the comprehensive characterization index data and correlation index change rate data of each analysis unit, inputting the processed comprehensive characterization index data and correlation index change rate data of each analysis unit into a pre-established random forest regression model, and outputting the analysis results;

[0007] The importance of the change rate data of the related indicators is sorted, and the significant related indicators of the target performance in each analysis unit are screened based on the importance sorting results. The significant related indicators of the target performance in each analysis unit are combined with the analysis results to obtain the weights of the significant related indicators;

[0008] Obtain the significant correlation indicators of the target performance in each analysis unit in the target study area, and perform standardization to obtain the standardized values ​​of the significant correlation indicators. Based on the standardized values ​​of the significant correlation indicators and the weights of the significant correlation indicators, the coupling degree is calculated to obtain the target coupling degree.

[0009] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the process of acquiring the sample city-related data includes:

[0010] In the geographic information system, the location and size of urban green spaces are obtained based on the interpretation of remote sensing data of sample cities. On the map open platform, the API is called to crawl urban residential buildings, commercial buildings, industrial buildings, office buildings, POI data, and road network structure data, and combined with the land use status map of the sample city, manual corrections are made in the geographic information system; census data is obtained, or population distribution data with an accuracy that matches the analysis unit is obtained based on the population data website.

[0011] In combination with the first aspect, in some implementations of the first aspect, the method further includes: calculating the comprehensive characterization index data as follows:

[0012] For each target performance P, calculate the change rate of each characterization index E of all analysis units in two periods: define the oth characterization index of the i-th target performance, and the early characterization index value is E io _before, the later characterization index value is E io _after, then the rate of change of the characteristic index ΔE io for:

[0013]

[0014] Calculate the comprehensive representation index of each target performance of all analysis units: the comprehensive representation index is the sum of the weights of the change rates of each representation index, and define the i-th target performance P i The comprehensive characterization index is E iC , then:

[0015]

[0016] Among them, o is the number of indicators representing the performance of the i-th target.

[0017] In combination with the first aspect, in some implementations of the first aspect, the method further includes: a calculation process of the correlation index change rate data is as follows:

[0018] For each correlation index I, calculate the rate of change: Define the qth correlation index I of the jth correlation item Cj jq , whose early characterization index is I jq _before, the later characterization index is I jq _after, the change rate of the associated indicator is:

[0019]

[0020] In the formula, ΔI jqis the rate of change of the associated indicator.

[0021] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the pre-established random forest regression model is as follows:

[0022] Feature extraction is performed based on the random forest regression model. The random forest consists of multiple decision trees. For the training data set D = {(x 1 ,y 1 ),(x 2 ,y 2 ),...,(x n ,y n )}, where x i is the sample size containing multiple features, that is, the independent variable, y i is the corresponding target continuous value; when constructing each decision tree, the same number of samples as the original training set are randomly selected from the original training data with replacement as the sub-training set. At the same time, at each node, m features are randomly selected from all features, and the optimal features and splitting points are selected for node splitting based on the principle of minimizing the mean square error. This process is repeated until the preset tree depth or node sample number threshold is reached, and the construction of the decision tree is completed, thereby constructing a random forest model containing T decision trees;

[0023] The feature extraction method based on feature contribution is used to obtain significant association indicators: the Gini index is used as a measure of feature importance. For each decision tree T k (k=1,2,...,T), Gini impurity of node t Among them, K is the number of categories after discretizing the target value, p t,k is the proportion of samples belonging to the kth class in node t, and feature j is in tree T k Importance score in:

[0024]

[0025] Among them, T jk For feature j in tree T k The set of nodes used in t left and t right are the left and right child nodes after node t is split;

[0026] The importance scores of feature j in all decision trees are accumulated, that is, Get the final importance score of feature j, sort all features by importance, and select the first n features as the key indicators for extraction.

[0027] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: variable selection of the pre-established random forest regression model: using the comprehensive representation index of each target performance as a dependent variable and all related indicators as independent variables, respectively, and inputting them into the random forest regression model for regression analysis;

[0028] The related indicators include one or more of per capita artificial green space area, per capita natural green space area, per capita vacant land area, average shape index, mean artificial green space proximity distance, mean natural green space proximity distance, mean vacant land proximity distance, artificial green space service population coverage rate and landscape aggregation index; the characterization indicators include one or more of block activity, block housing price average, block heat island value and block carbon emission;

[0029] The parameters of the pre-built random forest regression model were adjusted based on the mean square error (MSE), mean absolute error (MAE), and coefficient of determination of the model.

[0030] In combination with the first aspect, in some implementations of the first aspect, the method further includes: the significant correlation index of the target performance in each analysis unit is obtained by screening based on the importance ranking result, and the corresponding correlation index with an importance greater than a preset threshold is selected as the significant correlation index of the target performance, and the vth significant correlation index of the i-th target performance is defined as SI iv , assuming that the target performance has n significant correlation indicators, then v = 1, 2, 3, ..., n; according to the feature importance percentage presented by the random forest regression model analysis results, the SI of each significant correlation indicator is obtained iv The initial weight IW iv ;

[0031] Based on the selected significant correlation indicators and their initial weights, the weights are modified and the weight of the vth significant correlation indicator of the i-th target performance is defined as W iv , then:

[0032]

[0033] Where n is the number of significant correlation indicators of target performance i.

[0034] In combination with the first aspect, in some implementations of the first aspect, the method further includes: a process of calculating the coupling degree based on the standardized value of the significant correlation index and the weight of the significant correlation index:

[0035] Obtain the value of the significant correlation index SI of each target performance P of each analysis unit in the target study area;

[0036] The significant correlation index SI is standardized, and the vth significant correlation index SI of the i-th target performance is iv , standardized value SI iv _S is:

[0037]

[0038] Among them, SI iv max is the maximum value of the significant association index SI of all analysis units in the target study area. iv min is the minimum value of the significant association index SI of all analysis units in the target study area. iv mean is the average value of the significant association index SI of all analysis units in the target study area;

[0039] Based on the standardized value of the significant correlation index and the weight of the significant correlation index, the coupling degree CD of each coupling target is calculated. Assuming there are m coupling targets in total, the coupling degree CD of the i-th coupling target is i ,exist:

[0040] CD i =SI i1 _S×W i1 +SI i2 _S×W i2 +SI i3 _S×W i3 +...+SI in _S×W in .

[0041] In a second aspect, in order to achieve the above-mentioned purpose, the present invention discloses a system for calculating the coupling degree between green space structure and built environment, comprising:

[0042] A data division module is used to receive sample city related data and divide the sample city into multiple analysis units; wherein the sample city related data includes urban residential buildings, commercial buildings, industrial buildings, office buildings, urban green space locations and sizes, road network structure data, POI distribution data and population distribution data;

[0043] A data analysis module, used to obtain the comprehensive characterization index data and the correlation index change rate data of each analysis unit, perform data processing on the comprehensive characterization index data and the correlation index change rate data of each analysis unit, input the processed comprehensive characterization index data and the correlation index change rate data of each analysis unit into a pre-established random forest regression model, and output the analysis results;

[0044] The weight acquisition module is used to sort the importance of the correlation indicator change rate data, screen and obtain the significant correlation indicators of the target performance in each analysis unit based on the importance sorting results, and combine the significant correlation indicators of the target performance in each analysis unit with the analysis results to obtain the weights of the significant correlation indicators;

[0045] The coupling measurement module is used to obtain the significant correlation indicators of the target performance in each analysis unit in the target study area, and perform standardization to obtain the standardized values ​​of the significant correlation indicators. The coupling degree is calculated based on the standardized values ​​of the significant correlation indicators and the weights of the significant correlation indicators to obtain the target coupling degree.

[0046] In another aspect of the present invention, in order to achieve the above-mentioned purpose, a terminal device is disclosed, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the memory stores a computer program capable of running on the processor, and when the processor loads and executes the computer program, a method for measuring the coupling degree between a green space structure and a built environment as described above is adopted.

[0047] Beneficial effects of the present invention:

[0048] The present invention establishes a coupling structure and constructs a significant relationship between urban observation data and green space structure-built environment correlation indicators based on a random forest regression model, transforming the urban green space performance evaluation into a series of coupling values, thereby optimizing the problems of a large urban green space performance evaluation system, complicated indicators, and difficult quantification, and improving the accuracy of benefit evaluation and the efficiency of measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0050] Figure 1 It is a schematic flow chart of the method of the present invention;

[0051] Figure 2 It is a schematic diagram of the technical process of the present invention;

[0052] Figure 3 It is a schematic diagram of the system structure of the present invention;

[0053] Figure 4 It is a schematic diagram of the per capita artificial green space change rate and per capita natural green space change rate of the sample cities of the present invention;

[0054] Figure 5It is a comprehensive characterization index diagram of the living performance of sample cities in 2020 and 2024 of the present invention;

[0055] Figure 6 It is a schematic diagram of the importance of sample data features of the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Embodiment 1:

[0058] like Figure 1 As shown, a method for calculating the coupling degree between green space structure and built environment includes the following steps:

[0059] Receive data related to sample cities and divide the sample cities into multiple analysis units; wherein the data related to sample cities include urban residential buildings, commercial buildings, industrial buildings, office buildings, urban green space locations and sizes, road network structure data, POI distribution data, and population distribution data;

[0060] Specifically, in principle, two time periods with obvious differences in urban green space structure should be selected. The process of obtaining relevant data of the sample cities includes:

[0061] In the geographic information system, the location and size of urban green spaces are obtained based on the interpretation of remote sensing data of sample cities. The green space information extraction based on remote sensing data interpretation can identify all urban green spaces, not just the green spaces classified into land use classification.

[0062] On the map open platform, call the API to crawl urban residential buildings, commercial buildings, industrial buildings, office buildings, POI data, and road network structure data, and combine the land use status map of the sample city to make manual corrections in the geographic information system. This step can directly obtain the complete vector data of the sample data, which is more efficient and commonly used;

[0063] Obtain census data from official channels, or obtain population distribution data with accuracy that matches the analysis unit from population data websites such as WorldPop;

[0064] Specifically, the present invention is further described below by way of embodiments:

[0065] Based on the service functions and research needs of urban green space, the coupling targets of the green space structure to be measured and the built space are determined. The coupling targets refer to the goals that are interrelated, mutually influential, and coordinated among multiple systems, elements, or processes in order to achieve specific overall functions, effects, or states. The coupling targets of green space and built environment refer to the efforts to coordinate and integrate green space (such as parks, green spaces, woodlands, water bodies, etc.) and built environment (such as buildings, roads, infrastructure, etc.) in the process of urban planning, design, and development, so as to achieve comprehensive benefits and sustainable development in many aspects. The coupling targets can be directly reflected by urban observation data;

[0066] Based on the coupling target, a coupling structure is formulated, in which the coupling structure is composed of target performance and its related projects. This structure can realize the exploration of the coordination and integration mechanism between green space and built environment under the guidance of multiple coupling targets, condense and extract the green space related structural factors that have a significant impact on the built environment, so as to accurately guide the construction of urban green space and respond to the demand of urban development for green space service functions. Compared with the traditional formulation of related performance indicators that rely on expert knowledge, this conclusion improves the scientificity and objectivity of the process and avoids the defects of excessive subjectivity and one-sidedness of the research:

[0067] Define performance goals

[0068] In this method, coupling targets refer to the ideal targets achieved through the mutual coordination, influence and dependence between different elements and subsystems of urban green space structure and built environment elements in urban development. The final performance of each coupling target is represented by the performance value of the corresponding urban green space structure. Referring to the classification and definition of performance in the Millennium Ecosystem Assessment, target performance under the guidance of multiple coupling targets can be formulated according to research needs:

[0069] P1, P2, P3, ..., Pm, where m is the target performance number;

[0070] Furthermore, for each target performance P, multiple representation indicators E are defined. The representation indicators are specific observed values, representing the changes in a certain aspect of the city directly caused by the changes in the urban green space structure. The final effect of the target performance is calculated by the change rate of its specific value. Let the i-th target performance P i There are o characterization indicators in total, so the characterization indicators of the target performance are defined as:

[0071] E i1 ,E i2 ,E i3 ,...,E io , where o is the number of indicators representing the performance of the i-th target;

[0072] In this example, the proposed coupling target is the coupling of the three lives (production, life, and ecology) of the urban green space structure. Accordingly, the target performance in the coupling structure is defined as: P1 life performance, P2 production performance, and P3 ecological performance;

[0073] For P1 life performance, its representative index E 11 =Neighborhood activity;

[0074] For P2 production performance, its characterization index E 21 = Average housing price in the neighborhood;

[0075] For P3 ecological performance, its characterization index E 31 = Block heat island value.

[0076] Defining related projects

[0077] According to the relationship between multiple factors involved in the urban green space structure and the built environment and the research needs, the associated projects are constructed. The associated projects are various green space attributes extracted from the green space structure, such as one or more of the attributes such as shape, scale, area, space, and location. The associated project is defined as C. Assuming there are n associated projects in total, the associated projects have:

[0078] C1, C2, C3, ..., Cn, where n is the number of associated items;

[0079] Furthermore, for each associated item C, a series of associated indicators I are set; in this method, the associated indicators are the associated data obtained by secondary calculation between the green space structural elements (such as area, shape, location, etc.) and other elements in the built environment (such as residential areas, roads, POIs, population, etc.) rather than isolated data; assuming that the j-th associated item Cj has a total of p associated indicators, then the total number of associated indicators is:

[0080] I j1 , I j2 , I j3 , ..., I jp , where p is the number of associated indicators of the j-th associated project Cj;

[0081] In this example, the associated projects are:

[0082] C1 scale-related projects, C2 form-related projects, C3 space-related projects;

[0083] For C1 scale-related projects, the scale-related indicators are:

[0084] I 11 = artificial green space per capita, I 12 = natural green area per capita, I 13 = per capita vacant land area;

[0085] For C2 form-related projects, the form-related indicators are:

[0086] I 21 =Proportion of artificial green space, I 22 =Ratio of natural green space, I 23 = percentage of vacant land, I 24 = average shape index;

[0087] For C3 spatial correlation projects, the spatial correlation indicators are:

[0088] I 31 = the mean value of the proximity distance of artificial green space, I 32 = the mean distance to natural green space,

[0089] I 33 = average distance to vacant land, I 34 = Population coverage rate of artificial green space services.

[0090] Obtaining comprehensive characterization index data and correlation index change rate data of each analysis unit, performing data processing on the comprehensive characterization index data and correlation index change rate data of each analysis unit, inputting the processed comprehensive characterization index data and correlation index change rate data of each analysis unit into a pre-established random forest regression model, and outputting the analysis results;

[0091] The calculation of the comprehensive characterization index data is as follows:

[0092] For each target performance P, calculate the change rate of each characterization index E of all analysis units in two periods: define the oth characterization index of the i-th target performance, and the early characterization index value is E io _before, the later characterization index value is E io _after, then the rate of change of the characteristic index ΔE io for:

[0093]

[0094] Calculate the comprehensive representation index of each target performance of all analysis units: the comprehensive representation index is the sum of the weights of the change rates of each representation index, and define the i-th target performance P i The comprehensive characterization index is E iC , then:

[0095]

[0096] Among them, o is the number of indicators representing the performance of the i-th target.

[0097] The calculation process of the correlation index change rate data is as follows:

[0098] For each correlation index I, calculate the rate of change: Define the qth correlation index I of the jth correlation item Cj jq , whose early characterization index is I jq _before, the later characterization index is I jq _after, the change rate of the associated indicator is:

[0099]

[0100] In the formula, ΔI jq is the rate of change of the associated indicator.

[0101] In this example, taking life performance as an example, the related indicator I 11 , I 12 The rate of change of Figure 4 As shown in the figure, the comprehensive indicators of living performance in the two stages of the sample cities are as follows: Figure 5 shown.

[0102] The pre-built random forest regression model is as follows:

[0103] Feature extraction is performed based on the random forest regression model. The random forest consists of multiple decision trees. For the training data set D = {(x 1 ,y 1 ),(x 2 ,y 2 ),...,(x n ,y n )}, where x i is the sample size containing multiple features, that is, the independent variable, y i is the corresponding target continuous value, i.e., the dependent variable; when constructing each decision tree, the same number of samples as the original training set are randomly selected from the original training data with replacement as the sub-training set. At the same time, at each node, m features are randomly selected from all features, and the optimal features and splitting points are selected for node splitting based on the principle of minimizing the mean square error. This process is repeated until the preset tree depth or node sample number threshold is reached, and the construction of a decision tree is completed. In this way, a random forest model containing T decision trees is constructed;

[0104] The feature extraction method based on feature contribution is used to obtain significant association indicators: This method uses the Gini index as a measure of feature importance. For each decision tree T k (k=1,2,...,T), Gini impurity of node t Among them, K is the number of categories after discretizing the target value, p t,k is the proportion of samples belonging to the kth class in node t, and feature j is in tree T k Importance score in:

[0105]

[0106] Among them, T jk For feature j in tree T k The set of nodes used in t left and t right are the left and right child nodes after node t is split;

[0107] The importance scores of feature j in all decision trees are accumulated, that is, Get the final importance score of feature j, sort all features according to importance, and select the first n features (determined according to actual needs) as the extracted significant association indicators.

[0108] Significant correlation indicators can be extracted by principal component regression and least squares regression. The random forest model is selected because it has the following advantages:

[0109] 1. Stability: Random forest integrates multiple decision trees and has good anti-overfitting ability. Least squares method and principal component regression are prone to overfitting when the amount of data is small or there are too many features, resulting in a decrease in the generalization ability of the model on the test set;

[0110] 2. Scalability: When processing large-scale data, principal component regression and least squares regression have high computational complexity, long training time, and relatively complex implementation of parallel computing. Their efficiency and scalability are not as good as the random forest regression algorithm.

[0111] 3. Adapt to high-dimensional data: Random forest uses a random feature selection method in the process of building a decision tree to automatically screen out important features and reduce the problem of collinearity between features; principal component regression needs to perform principal component analysis to reduce the dimension before processing the data, which may cause loss of original data; the least squares regression model is unstable in the case of high-dimensional data and multicollinearity.

[0112] Variable selection of the pre-established random forest regression model: in the imported sample data, the comprehensive representation index of each target performance is used as the dependent variable, and all related indicators are used as independent variables, which are input into the random forest regression model for regression analysis;

[0113] The related indicators include one or more of per capita artificial green space area, per capita natural green space area, per capita vacant land area, average shape index, mean artificial green space proximity distance, mean natural green space proximity distance, mean vacant land proximity distance, artificial green space service population coverage rate and landscape aggregation index; the characterization indicators include one or more of block activity, block housing price average, block heat island value and block carbon emission;

[0114] Specifically, in this embodiment, the comprehensive life representation index E is selected 1C = The change rate of the block activity is put into the variable Y analysis module as the dependent variable, and the change rate of each related indicator ΔI is selected 11 = Change rate of per capita artificial green space area, ΔI 12 = Change rate of per capita natural green space area, ΔI 13 = Change rate of vacant land area per capita, ΔI 21 = Change rate of artificial green space proportion, ΔI 22 = Change rate of natural green space proportion, ΔI 23 = Change rate of vacant land proportion, ΔI 24 = Average shape index change rate, ΔI 31 = Change rate of mean value of artificial green space proximity distance, ΔI 32 = Change rate of the mean distance to natural green space, ΔI 33 = Change rate of the mean distance to vacant land, ΔI 34 = Change rate of population coverage of artificial green space services, put into the independent variable X module;

[0115] Specifically, according to the optimization results of the correlation indicators, the sample data is divided into a training set and a test set at a ratio of 7:3 or 8:2, or the algorithm code is written based on Python, or a random forest regression model is established based on data statistical analysis software such as SPSSPRO, and the comprehensive representation index E of each type of target performance is used. C As the dependent variable, the change rate ΔI of all related indicators is used as the independent variable for model training;

[0116] In this example, taking life performance P1 as an example, SPSSPRO software is used to perform random forest regression analysis. The specific operations are as follows:

[0117] Data cleaning and data import: Check whether the data is complete, delete missing values, outliers, etc.; include all comprehensive characterization indicators E C The data file of the correlation index ΔI was imported into SPSSPRO in the form of an EXCEL table;

[0118] Correlation analysis of related indicators: Click "Correlation Analysis-Spearman Correlation Coefficient", drag all independent variables into "Analysis Item Y (Quantitative)", click "Start Analysis", and find and delete redundant variables by observing the size and significance of the correlation coefficient. In this example, there are no related indicators with multicollinearity.

[0119] Select the training ratio as 0.7 (i.e. 70% of the input samples are used as training sets and 30% of the samples are used as test sets), whether to shuffle the data or not, and whether to cross-validate; click Start Analysis to output the results of this analysis;

[0120] The parameters of the pre-established random forest regression model were adjusted based on the mean square error (MSE), mean absolute error (MAE) and coefficient of determination (R) of the model. 2 Make adjustments.

[0121] Among them, the calculation of mean square error MSE is:

[0122]

[0123] n is the total number of samples, y i is the true value of the i-th sample, ^y i is the predicted value of the i-th sample;

[0124] Calculation of absolute error MAE:

[0125]

[0126] n is the total number of samples, y i is the true value of the i-th sample, ^y i is the predicted value of the i-th sample;

[0127] Coefficient of determination R 2 The calculation process:

[0128]

[0129] n is the total number of samples, y i is the true value of the i-th sample, ^y i is the predicted value of the i-th sample, is the mean of the true values;

[0130] The details are shown in Table 1:

[0131] Table 1 Effect related indicators

[0132]

[0133] The importance of the change rate data of the related indicators is sorted, and the significant related indicators of the target performance in each analysis unit are screened based on the importance sorting results. The significant related indicators of the target performance in each analysis unit are combined with the analysis results to obtain the weights of the significant related indicators;

[0134] In the analysis results, observe the importance ranking of the change rate of the associated indicators, select the corresponding associated indicators with an importance greater than 10% as the significant associated indicators of the target performance, and define the vth significant associated indicator of the ith target performance as SI iv, assuming that the target performance has n significant correlation indicators, then v = 1, 2, 3, ..., n; further, according to the feature importance percentage presented by the random forest regression model analysis results, the SI of each significant correlation indicator is obtained iv The initial weight IW iv ;

[0135] Based on the screened significant correlation indicators and their initial weights, their weights are modified, and the weight of the vth significant correlation indicator of the i-th target performance is defined as W iv , then:

[0136]

[0137] Where n is the number of significant correlation indicators of target performance i;

[0138] In this example, taking life performance P1 as an example, according to output result 2: the bar chart result of feature importance value, it can be seen that the change rate of the associated indicators whose importance of life performance P1 is greater than 10% is:

[0139] ΔI 11 = Change rate of per capita artificial green space area, ΔI 24 = Average shape index change rate, ΔI 31 = Change rate of mean value of proximity distance of artificial green space;

[0140] Extract the corresponding correlation index, and the significant correlation index of life performance P1 is:

[0141] SI 11 = per capita artificial green area, SI 12 = Mean shape index, SI 13 = the mean value of the proximity distance of artificial green space;

[0142] According to the analysis results, the initial weight of each significant correlation indicator is:

[0143] IW 11 =61.10%, IW 12 =14.10%, IW 13 =17.40%;

[0144] After correction, the weight of each significant correlation indicator is:

[0145] W 11 =66.00%, W 12 =15.23%, W 13 =18.77%.

[0146] The specific importance of the associated indicator features and their initial weights are as follows: Figure 6 shown.

[0147] Obtain the significant correlation indicators of target performance in each analysis unit in the target study area, and perform standardization to obtain the standardized values ​​of the significant correlation indicators. Based on the standardized values ​​of the significant correlation indicators and the weights of the significant correlation indicators, the coupling degree is calculated to obtain the life coupling degree.

[0148] Extraction of significant correlation indicators of cities in the target area

[0149] Obtain the value of the significant correlation index SI of each target performance P of each research unit in the target research area;

[0150] Each significant correlation index is standardized, and the vth significant correlation index SI of the i-th target performance is iv , its standardized value SI iv _S is:

[0151]

[0152] Among them, SI iv max is the maximum value of the significant association index SI of all research units in the target research area. iv min is the minimum value of the significant association index SI of all research units in the target research area. iv Mean is the average value of the significant association index SI of all research units in the target study area.

[0153] Coupling degree calculation

[0154] Based on the standardized value SI_S of the significant correlation index of each target performance and the weight of the significant correlation index obtained by S33, the coupling degree CD of each coupling target is calculated. If there are m coupling targets in total, the coupling degree CD of the i-th coupling target is i ,exist:

[0155] CD i =SI i1 _S×W i1 +SI i2 _S×W i2 +SI i3 _S×W i3 +...+SI in _S×W in

[0156] In this example, taking life performance P1 as an example, the standardized value of the significant correlation indicator of a target research unit in the sample area is:

[0157] SI 11 _S=0.8,SI 12 _S=0.2, SI13 _S = 0.3;

[0158] Then the degree of life coupling between the green space structure and the built environment of the research unit is:

[0159] CD 1 =0.8×66.00%+0.2×15.23%+0.3×18.77%=0.61.

[0160] Embodiment 2: In the second aspect, as Figure 3 As shown, in order to achieve the above-mentioned purpose, the present invention discloses a system for calculating the coupling degree between green space structure and built environment, comprising:

[0161] The data division module 11 is used to receive sample city related data and divide the sample city into multiple analysis units; wherein the sample city related data includes urban residential buildings, commercial buildings, industrial buildings, office buildings, urban green space locations and sizes, road network structure data, POI distribution data and population distribution data;

[0162] The data analysis module 12 is used to obtain the comprehensive characterization index data and the correlation index change rate data of each analysis unit, perform data processing on the comprehensive characterization index data and the correlation index change rate data of each analysis unit, input the processed comprehensive characterization index data and the correlation index change rate data of each analysis unit into a pre-established random forest regression model, and output the analysis results;

[0163] The weight acquisition module 13 is used to sort the importance of the correlation indicator change rate data, screen and obtain the significant correlation indicators of the target performance in each analysis unit based on the importance sorting results, and combine the significant correlation indicators of the target performance in each analysis unit with the analysis results to obtain the weights of the significant correlation indicators;

[0164] The coupling measurement module 14 is used to obtain the significant correlation indicators of the target performance in each analysis unit in the target research area, and perform standardization processing to obtain the standardized values ​​of the significant correlation indicators. The coupling degree is calculated based on the standardized values ​​of the significant correlation indicators and the weights of the significant correlation indicators to obtain the life coupling degree.

[0165] Based on the same inventive concept, the present invention also provides a computer device, which includes: one or more processors, and a memory for storing one or more computer programs; the program includes program instructions, and the processor is used to execute the program instructions stored in the memory. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is used to implement one or more instructions, specifically for loading and executing one or more instructions in a computer storage medium to implement the above method.

[0166] It needs to be further explained that, based on the same inventive concept, the present invention also provides a computer storage medium, on which a computer program is stored, and the computer program is executed by a processor to execute the above method. The storage medium can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electrical, magnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0167] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0168] The above shows and describes the basic principles, main features and advantages of the present disclosure. Those skilled in the art should understand that the present disclosure is not limited by the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present disclosure. Without departing from the spirit and scope of the present disclosure, the present disclosure may have various changes and improvements, and these changes and improvements fall within the scope of the present disclosure to be protected.

Claims

1. A method for calculating the coupling degree between green space structure and built environment, characterized in that: The method comprises the following steps: Receive data related to sample cities and divide the sample cities into multiple analysis units; wherein the data related to sample cities include urban residential buildings, commercial buildings, industrial buildings, office buildings, urban green space locations and sizes, road network structure data, POI distribution data, and population distribution data; Obtaining comprehensive characterization index data and correlation index change rate data of each analysis unit, performing data processing on the comprehensive characterization index data and correlation index change rate data of each analysis unit, inputting the processed comprehensive characterization index data and correlation index change rate data of each analysis unit into a pre-established random forest regression model, and outputting the analysis results; The importance of the change rate data of the related indicators is sorted, and the significant related indicators of the target performance in each analysis unit are screened based on the importance sorting results. The significant related indicators of the target performance in each analysis unit are combined with the analysis results to obtain the weights of the significant related indicators; Obtain the significant correlation indicators of the target performance in each analysis unit in the target study area, and perform standardization to obtain the standardized values ​​of the significant correlation indicators. Based on the standardized values ​​of the significant correlation indicators and the weights of the significant correlation indicators, the coupling degree is calculated to obtain the target coupling degree.

2. A method for calculating the coupling degree between green space structure and built environment according to claim 1, characterized in that: The process of obtaining the sample city-related data includes: In the geographic information system, the location and size of urban green spaces are obtained based on the interpretation of remote sensing data of sample cities. On the map open platform, the API is called to crawl urban residential buildings, commercial buildings, industrial buildings, office buildings, POI data, and road network structure data, and combined with the land use status map of the sample city, manual corrections are made in the geographic information system; census data is obtained, or population distribution data with an accuracy that matches the analysis unit is obtained based on the population data website.

3. The method for calculating the coupling degree between green space structure and built environment according to claim 1, characterized in that: The calculation of the comprehensive characterization index data is as follows: For each target performance P, calculate the change rate of each characterization index E of all analysis units in two periods: define the oth characterization index of the i-th target performance, and the early characterization index value is E io _before, the later characterization index value is E io _after, then the rate of change of the characteristic index ΔE io for: Calculate the comprehensive representation index of each target performance of all analysis units: the comprehensive representation index is the sum of the weights of the change rates of each representation index, and define the i-th target performance P i The comprehensive characterization index is E iC , then: Among them, o is the number of indicators representing the performance of the i-th target.

4. The method for calculating the coupling degree between green space structure and built environment according to claim 1, characterized in that: The calculation process of the correlation index change rate data is as follows: For each correlation index I, calculate the rate of change: Define the qth correlation index I of the jth correlation item Cj jq , whose early characterization index is I jq _before, the later characterization index is I jq _after, the change rate of the associated indicator is: In the formula, ΔI jq is the rate of change of the associated indicator.

5. The method for calculating the coupling degree between green space structure and built environment according to claim 1, characterized in that: The pre-built random forest regression model is as follows: Feature extraction is performed based on the random forest regression model. The random forest consists of multiple decision trees. For the training data set D = {(x1, y1), (x2, y2), ..., (x n ,y n )}, where x i is the sample size containing multiple features, that is, the independent variable, y i is the corresponding target continuous value; when constructing each decision tree, the same number of samples as the original training set are randomly selected from the original training data with replacement as the sub-training set. At the same time, at each node, m features are randomly selected from all features, and the optimal features and splitting points are selected for node splitting based on the principle of minimizing the mean square error. This process is repeated until the preset tree depth or node sample number threshold is reached, and the construction of the decision tree is completed, thereby constructing a random forest model containing T decision trees; The feature extraction method based on feature contribution is used to obtain significant association indicators: the Gini index is used as a measure of feature importance. For each decision tree T k (k=1,2,...,T), Gini impurity of node t Among them, K is the number of categories after discretizing the target value, p t,k is the proportion of samples belonging to the kth class in node t, and feature j is in tree T k Importance score in: Among them, T jk For feature j in tree T k The set of nodes used in t left and t right are the left and right child nodes after node t is split; The importance scores of feature j in all decision trees are accumulated, that is, Get the final importance score of feature j, sort all features by importance, and select the first n features as the key indicators for extraction.

6. A method for calculating the coupling degree between green space structure and built environment according to claim 5, characterized in that: Variable selection of the pre-established random forest regression model: in the imported sample data, the comprehensive representation index of each target performance is used as the dependent variable, and all related indicators are used as independent variables, which are input into the random forest regression model for regression analysis; The related indicators include one or more of per capita artificial green space area, per capita natural green space area, per capita vacant land area, average shape index, mean artificial green space proximity distance, mean natural green space proximity distance, mean vacant land proximity distance, artificial green space service population coverage rate and landscape aggregation index; the characterization indicators include one or more of block activity, block housing price average, block heat island value and block carbon emission; The parameters of the pre-built random forest regression model were adjusted based on the mean square error (MSE), mean absolute error (MAE), and coefficient of determination of the model.

7. The method for calculating the coupling degree between green space structure and built environment according to claim 1, characterized in that: The significant correlation index of the target performance in each analysis unit is obtained by screening based on the importance ranking result. The corresponding correlation index with importance greater than the preset threshold is selected as the significant correlation index of the target performance, and the vth significant correlation index of the i-th target performance is defined as SI iv , assuming that the target performance has n significant correlation indicators, then v = 1, 2, 3, ..., n; according to the feature importance percentage presented by the random forest regression model analysis results, the SI of each significant correlation indicator is obtained iv The initial weight IW iv ; Based on the selected significant correlation indicators and their initial weights, the weights are modified and the weight of the vth significant correlation indicator of the i-th target performance is defined as W iv , then: Where n is the number of significant correlation indicators of target performance i.

8. The method for calculating the coupling degree between green space structure and built environment according to claim 1, characterized in that: The process of calculating the coupling degree based on the standardized value of the significant correlation index and the weight of the significant correlation index is as follows: Obtain the value of the significant correlation index SI of each target performance P of each analysis unit in the target study area; The significant correlation index SI is standardized, and the vth significant correlation index SI of the i-th target performance is iv , standardized value SI iv _S is: Among them, SI iv max is the maximum value of the significant association index SI of all analysis units in the target study area. iv min is the minimum value of the significant association index SI of all analysis units in the target study area. iv mean is the average value of the significant association index SI of all analysis units in the target study area; Based on the standardized value of the significant correlation index and the weight of the significant correlation index, the coupling degree CD of each coupling target is calculated. Assuming there are m coupling targets in total, the coupling degree CD of the i-th coupling target is i ,exist: CD i =SI i1 _S×W i1 +SI i2 _S×W i2 +SI i3 _S×W i3 +...+SI in _S×W in .

9. A system for calculating the coupling degree between green space structure and built environment, which adopts a method for calculating the coupling degree between green space structure and built environment as claimed in any one of claims 1 to 8, characterized in that: include: A data division module is used to receive sample city related data and divide the sample city into multiple analysis units; wherein the sample city related data includes urban residential buildings, commercial buildings, industrial buildings, office buildings, urban green space locations and sizes, road network structure data, POI distribution data and population distribution data; A data analysis module, used to obtain the comprehensive characterization index data and the correlation index change rate data of each analysis unit, perform data processing on the comprehensive characterization index data and the correlation index change rate data of each analysis unit, input the processed comprehensive characterization index data and the correlation index change rate data of each analysis unit into a pre-established random forest regression model, and output the analysis results; The weight acquisition module is used to sort the importance of the correlation indicator change rate data, screen and obtain the significant correlation indicators of the target performance in each analysis unit based on the importance sorting results, and combine the significant correlation indicators of the target performance in each analysis unit with the analysis results to obtain the weights of the significant correlation indicators; The coupling measurement module is used to obtain the significant correlation indicators of the target performance in each analysis unit in the target study area, and perform standardization to obtain the standardized values ​​of the significant correlation indicators. The coupling degree is calculated based on the standardized values ​​of the significant correlation indicators and the weights of the significant correlation indicators to obtain the target coupling degree.

10. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that: The memory stores a computer program that can be run on the processor. When the processor loads and executes the computer program, a method for calculating the coupling degree between a green space structure and a built environment according to any one of claims 1 to 8 is adopted.

Images