Method and system for optimizing urban green land ecosystem based on cultural benefit prediction

Abstract

The invention discloses an urban green land ecosystem optimization method and system based on cultural benefit prediction, and the method comprises the steps: carrying out the optimization of an urban green land ecosystem based on an environment physical feature-environment perception feature-recreation behavior feature-cultural benefit cascade model; environmental physical characteristics having significant influence on culture benefits are screened through a backward stepwise regression method; on the basis of the screened environmental physical characteristics, through path analysis of a multi-intermediary model, screening intermediary variables and paths with an intermediary effect in the influence of the environmental physical characteristics on culture benefits; integrating the screened environmental physical characteristics and the screened intermediary variables and paths into a structural equation model, performing fitting degree analysis on the structural equation model, and correcting highly related variables according to correction indexes; the corrected structural equation model reveals a cultural benefit influence mechanism of the urban green land ecosystem; building a Bayesian network model having the same structure as the urban green land ecosystem culture benefit influence mechanism to predict the level probability value of the culture benefit; and based on the collected environmental physical feature level of the current situation of the to-be-designed site, predicting a level probability value of cultural benefits by using a Bayesian network model, and then selecting an optimal scheme. The urban green land ecological system is optimized according to quantitative culture benefits, and a more sufficient and comprehensive basis is provided for evidence-based design and scientific decision-making of the urban green land.

Description

Technical Field

[0001] The present invention relates to the technical field of urban green space planning, and in particular to an optimization method, system, terminal equipment and computer-readable storage medium for an urban green space ecosystem based on cultural benefit prediction. Background Art

[0002] In the process of promoting construction in an all-round way, the country is committed to improving the ecological and environmental governance capabilities of large and medium-sized cities to ensure that the service functions and benefit levels of natural ecosystems match the population and economic scale. At present, the direct and tangible benefits brought by urban green spaces through regulating services and supply services have been widely and deeply explored, and have become an important reference for land use decision-making and planning design. In contrast, the cultural benefits of ecosystems, as the contribution of ecosystem cultural services to intangible human well-being, are related to personal happiness. Due to their subjective and intangible characteristics, they are difficult to define and even more difficult to measure at the technical level, and the frequency of being mentioned and included in decision-making is extremely limited. As the ecosystem benefit most closely related to human daily life, cultural benefits are considered to be one of the potential driving forces for promoting natural resource management, land use decisions and ecosystem protection. It is crucial to comprehensively improve the contribution of urban green spaces to the cultural benefits of residents. Therefore, in the face of multiple challenges such as the gradual degradation and reduction of urban nature, the increasing urban population, and the growing demand for intangible well-being of residents, how to give full play to the ecosystem cultural benefits of urban green spaces and enhance people's happiness through scientific urban green space design has become an important challenge faced by urban planners.

[0003] Current research on cultural benefit prediction to assist urban green space design has the following limitations: first, traditional tools mostly rely on monetary value prediction methods, which often leads to the underestimation of cultural benefits based on the public's subjective perception; second, there is a mismatch between the scale of theoretical research and the scale of actual design. At the same time, the mechanism of how urban green space affects cultural benefits is overly simplified, which leads to a lack of directly available auxiliary tools and theoretical support when designing green space at a specific site scale. Summary of the invention

[0004] In order to address the deficiencies of the above-mentioned prior art, the present invention provides an optimization method, system, terminal device and computer-readable storage medium for an urban green space ecosystem based on cultural benefit prediction, which can quantify the cultural benefits of the ecosystem that have long been neglected in the design and decision-making process of urban green spaces, and optimize the urban green space ecosystem based on the quantified cultural benefits, thereby providing a more sufficient and comprehensive basis for the evidence-based design and scientific decision-making of urban green spaces.

[0005] The first object of the present invention is to provide an optimization method for an urban green space ecosystem based on cultural benefit prediction.

[0006] The second object of the present invention is to provide an optimization system for urban green space ecosystem based on cultural benefit prediction.

[0007] The third object of the present invention is to provide a terminal device.

[0008] A fourth object of the present invention is to provide a computer-readable storage medium.

[0009] The first object of the present invention can be achieved by adopting the following technical solutions:

[0010] An optimization method for an urban green space ecosystem based on cultural benefit prediction, the method comprising:

[0011] Based on the cascade model of "environmental physical characteristics-environmental perception characteristics-recreational behavior characteristics-cultural benefits", the backward stepwise regression method was used to screen the environmental physical characteristics that have a significant impact on cultural benefits; among them, multiple sub-dimensions of cultural benefits and total benefits were used as dependent variables, environmental physical characteristics were used as independent variables, and environmental perception characteristics and recreational behavior characteristics were used as mediating variables;

[0012] Based on the selected environmental physical characteristics, the mediating variables and paths with mediating effects in the impact of environmental physical characteristics on cultural benefits are screened through path analysis of multiple mediation models.

[0013] The screened environmental physical characteristics, as well as the screened mediating variables and paths, were integrated into the structural equation model, and the goodness of fit of the structural equation model was analyzed. The variables whose correction index was greater than the set threshold were corrected. The corrected structural equation model revealed the influencing mechanism of the cultural benefits of the urban green space ecosystem.

[0014] Build a Bayesian network model with the same structure as the cultural benefit impact mechanism of urban green space ecosystem to predict the level probability value of cultural benefits;

[0015] Collect the environmental physical characteristic levels of the current site to be designed; the environmental physical characteristic levels include environmental physical characteristics and corresponding quantitative indicators;

[0016] If it is a single-factor optimization scheme, the environmental physical characteristics of the site to be designed and the different quantitative indicators of the target characteristics to be transformed are input as variables into the Bayesian network model to predict the level probability values ​​of the cultural benefits corresponding to the different quantitative indicators of the target characteristics to be transformed; the scheme with the largest level probability value is taken as the optimal scheme;

[0017] If it is a multi-factor optimization scheme, all possible combinations of environmental physical characteristic levels are enumerated, and the environmental physical characteristic levels of the current site to be designed and the environmental physical characteristic levels of each combination are input as variables into the Bayesian network model to predict the horizontal probability value of the cultural benefits corresponding to each combination of environmental physical characteristic levels; the scheme with the largest horizontal probability value is taken as the optimal scheme.

[0018] The second object of the present invention can be achieved by adopting the following technical solutions:

[0019] An optimization system for an urban green space ecosystem based on cultural benefit prediction, the system comprising:

[0020] The first screening module is used to screen environmental physical characteristics that have a significant impact on cultural benefits through the backward stepwise regression method based on the cascade model of "environmental physical characteristics-environmental perception characteristics-recreational behavior characteristics-cultural benefits". Among them, multiple sub-dimensions of cultural benefits and total benefits are used as dependent variables, environmental physical characteristics are used as independent variables, and environmental perception characteristics and recreational behavior characteristics are used as mediating variables.

[0021] The second screening module is used to screen the mediating variables and paths with mediating effects in the impact of environmental physical characteristics on cultural benefits through path analysis of multiple mediation models based on the screened environmental physical characteristics;

[0022] The integration and analysis module is used to integrate the selected environmental physical characteristics, as well as the selected mediating variables and paths into the structural equation model, and to perform a goodness of fit analysis on the structural equation model, and to modify the variables whose modification index is greater than the set threshold. The modified structural equation model reveals the influencing mechanism of the cultural benefits of the urban green space ecosystem.

[0023] A module is built to build a Bayesian network model with the same structure as the cultural benefit impact mechanism of the urban green space ecosystem to predict the horizontal probability value of cultural benefits;

[0024] A collection module, used to collect the environmental physical characteristic levels of the current status of the site to be designed; the environmental physical characteristic levels include environmental physical characteristics and corresponding quantitative indicators;

[0025] The first optimization module is used for, if it is a single-factor optimization scheme, inputting the environmental physical characteristic level of the current site to be designed and the different quantitative indicators of the target characteristics to be transformed as variables into the Bayesian network model, predicting the level probability value of the cultural benefit corresponding to the different quantitative indicators of the target characteristics to be transformed; the scheme with the largest level probability value is taken as the optimal scheme;

[0026] The second optimization module is used to input the environmental physical characteristic levels of the current site to be designed as variables into the Bayesian network model if it is a multi-factor optimization plan, generate all possible combinations of environmental physical characteristic levels, and predict the horizontal probability value of the cultural benefits corresponding to each combination of environmental physical characteristic levels; the plan with the largest horizontal probability value is taken as the optimal plan.

[0027] The third object of the present invention can be achieved by adopting the following technical solutions:

[0028] A terminal device includes a processor and a memory for storing a program executable by the processor. When the processor executes the program stored in the memory, the above-mentioned method for optimizing the urban green space ecosystem based on cultural benefit prediction is implemented.

[0029] The fourth object of the present invention can be achieved by adopting the following technical solutions:

[0030] A computer-readable storage medium stores a program, which, when executed by a processor, implements the above-mentioned method for optimizing the urban green space ecosystem based on cultural benefit prediction.

[0031] The present invention has the following beneficial effects compared with the prior art:

[0032] The present invention reveals the influencing mechanism of the cultural benefits of urban green space ecosystems through a modified structural equation model, and builds a Bayesian network model with the same structure as the influencing mechanism of the cultural benefits of urban green space ecosystems. The Bayesian network model is used to predict the horizontal probability value of cultural benefits; the urban green space ecosystem is optimized according to the quantified cultural benefits, providing a more sufficient and comprehensive basis for the evidence-based design and scientific decision-making of urban green space. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0034] Figure 1 This is a flow chart of an optimization method for an urban green space ecosystem based on cultural benefit prediction according to Example 1 of the present invention.

[0035] Figure 2 This is a schematic diagram of the minimum data collection unit division, handheld navigation auxiliary positioning, and iPad-side recording unit attributes during field survey in Example 1 of the present invention.

[0036] Figure 3This is a schematic diagram of the recreational unit after the division is completed in Example 1 of the present invention.

[0037] Figure 4 This is a diagram for analyzing the impact path of the benefits of “identity construction” in Example 1 of the present invention.

[0038] Figure 5 This is a diagram for analyzing the impact path of the benefits of “physical and mental experience” according to Example 1 of the present invention.

[0039] Figure 6 This is a diagram for analyzing the impact path of the benefits of “capacity enhancement” in Example 1 of the present invention.

[0040] Figure 7 This is a benefit impact path analysis diagram of “total benefit of ecosystem cultural services” of Example 1 of the present invention.

[0041] Figure 8 Schematic diagram of the environmental physical characteristics measurement model of Example 1 of the present invention.

[0042] Fig. 9 This is a schematic diagram of the cultural benefit measurement model of Example 1 of the present invention.

[0043] Fig.10 This is a standardized parameter estimation path diagram of the cultural benefit impact mechanism model of Example 1 of the present invention.

[0044] Fig.11 This is a schematic diagram of predicting the cultural benefits of an urban green space ecosystem based on a Bayesian network according to Example 1 of the present invention.

[0045] Fig.12 This is an interactive and visual schematic diagram of the prediction of the cultural benefits of the urban green space ecosystem according to Example 1 of the present invention.

[0046] Fig.13 Schematic diagram of the physical characteristics of various environments in Example 1 of the present invention.

[0047] Fig.14 The present characteristics of the recreational unit and the corresponding ecosystem cultural benefits of Example 1 of the present invention.

[0048] Fig.15 These are multiple solutions for the transformation of “recreational facilities” according to Example 1 of the present invention.

[0049] Fig.16 These are all the feature level combination schemes and the corresponding high benefit probabilities of Example 1 of the present invention.

[0050] Fig.17 The initial status quo of the solution of Example 1 of the present invention is limited.

[0051] Fig.18All feature level combination schemes based on “full space search” and corresponding high benefit probabilities of embodiment 1 of the present invention.

[0052] Fig.19 This is a structural block diagram of an optimization system for an urban green space ecosystem based on cultural benefit prediction according to Example 2 of the present invention.

[0053] Fig. 20 This is a structural block diagram of the terminal device of embodiment 3 of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in 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. It should be understood that the specific embodiments described are only used to explain the present application and are not used to limit the present application.

[0055] Embodiment 1:

[0056] like Figure 1 As shown, this embodiment provides an optimization method for an urban green space ecosystem based on cultural benefit prediction, comprising the following steps:

[0057] S101. Construct a revised structural equation model to reveal the influencing mechanism of cultural benefits of urban green space ecosystem.

[0058] This embodiment uses the cascade concept described in the ecosystem service cascade model as the underlying logic, decomposes the model into three parts: "natural ecosystem", "social ecosystem" and "interaction process of the two systems", and establishes the basic impact path and transmission direction of the ecosystem cultural benefits. And the research framework of "environment-practice-benefit" proposed by NEA is used as the core framework of the theoretical model. On the one hand, the ecosystem cultural services are decomposed into specific physical environments and practical activities, and the physical characteristics of the environment and the activities occurring within them are introduced. On the other hand, drawing on this theoretical model, the ecosystem cultural benefits are quantified and the cultural benefits are decomposed into three dimensions: "identity construction", "physical and mental experience" and "capacity improvement". On this basis, the theoretical model is also based on the matching idea of ​​supply and demand of the "ecosystem cultural service supply and demand confluence model", and "people" are included in the model as an independent link that affects the use of services and the generation of benefits, including social demographic characteristics and recreational behavior in the environment, which are used in parallel with the natural ecosystem to analyze the impact mechanism of benefit generation, and the benefit evaluation results are matched with the corresponding spatial characteristics of service provision in the analysis. In addition, the theoretical model is supplemented by the “ecosystem socio-cultural value model” to further refine the research model and incorporate human perception of the surrounding physical environment into the influencing mechanism.

[0059] Under the guidance of the above model, the impact mechanism of the cultural benefits of NEA is expanded into a cascade model of "environmental physical characteristics - environmental perception characteristics - recreational behavior characteristics - cultural benefits" as a theoretical model for exploring the impact mechanism of the cultural benefits of ecosystems in this embodiment.

[0060] (1) Collect data.

[0061] The team members collected 1,451 self-reported data from respondents in Guangzhou Haizhu Wetland Park through PPGIS (URL: http: / / 47.107.118.212 / mapsurvey / ), covering five aspects: ecosystem cultural benefits, environmental perception characteristics, recreational behavior characteristics, socio-demographic characteristics, and activity types. The specific process of collecting environmental physical characteristics data through field surveys is as follows:

[0062] refer to Figure 2 , 3First, various physical obstacles such as roads, rivers, hedges, etc. in the park are used as the basis for the division of the minimum data collection unit. Thus, 510 units are divided on GIS as the minimum data collection unit, called "recreation unit", to collect the characteristic level of environmental physical characteristics (independent variables). Secondly, each recreational unit is numbered with a unique value. Then, the divided recreational units are exported in KML format with geospatial information attributes and loaded into the Aowei interactive map (handheld navigation software) to facilitate the investigators to check the corresponding recreational unit according to the real-time GPS positioning. Finally, during the field survey, iPad is used to select the corresponding characteristic level in the pre-prepared environmental physical characteristics electronic evaluation form (Table 1) according to the environmental characteristics of each recreational unit, and the corresponding environmental physical characteristic level of the recreational unit is recorded.

[0063] Table 1 Environmental physical characteristics attribute evaluation table

[0064]

[0065] (2) Screen environmental physical features that have a significant impact on cultural benefits.

[0066] In the stage of independent variable screening, this embodiment uses 11 environmental physical characteristics that have important contributions to the level of cultural benefits as initial independent variables, including: "Land cover type: forest land", "Connectivity with surrounding roads", "Recreation facilities", "Terrain changes", "Activity space size", "Sight openness", "Distance from the main entrance", "Visible waterscape type", "Noise level", "Convenience facilities" and "Vegetation management level". The total benefit of cultural services and the benefits of the three sub-dimensions are used as dependent variables, and multivariate linear stepwise regression is performed on the environmental physical characteristics and the four dependent variables. Through the backward stepwise regression method, the factors that have no significant effect on the dependent variable (P>0.05) and the factors with high autocorrelation (VIF>10) are gradually removed, and the environmental physical characteristics that have a significant impact on the dependent variable are retained. Finally, the union of the above results is taken, and the environmental physical characteristics in the union are used as independent variables to be included in the mediation model analysis in the next stage.

[0067] The reason why this embodiment considers four dependent variables instead of the total benefit of ecosystem cultural services is that the total benefit of ecosystem cultural services is a latent variable constructed by the benefits of three sub-dimensions, and the benefits of the three sub-dimensions are latent variables composed of various specific evaluation items. Since the process of constructing latent variables is the process of factor extraction and dimensionality reduction of high-dimensional data, each factor will lose some variable information in this process, and the relationship between the total benefit and the independent variable cannot fully represent the relationship between the benefits of each sub-dimension and the independent variable. If only the relationship between the independent variable and the total benefit is considered in the initial screening stage, some important independent variables may be excluded.

[0068] (2-1) Stepwise regression analysis based on the total benefits of ecosystem cultural services.

[0069] In the stepwise regression with “total benefit” as the dependent variable, four characteristic variables, “land cover type: forest land”, “vegetation management level”, “noise level” and “view openness”, were retained after backward stepwise screening. The detailed results are shown in Table 2.

[0070] Table 2 Summary of model data

[0071]

[0072] (2-2) Stepwise regression analysis based on cultural benefit sub-dimensions.

[0073] In the stepwise regression with “identity construction” as the dependent variable, four characteristic variables, “land cover type: woodland”, “visible waterscape type”, “noise level” and “view openness”, were retained after backward stepwise screening. The detailed results are shown in Table 3 Model Data Summary.

[0074] Table 3 Summary of model data

[0075]

[0076] In the stepwise regression with “physical and mental experience” as the dependent variable, two characteristic variables, “noise level” and “terrain change”, were retained after backward stepwise screening. The detailed results are shown in Table 4 Model data summary.

[0077] Table 4 Summary of model data

[0078]

[0079] In the stepwise regression with “capacity improvement” as the dependent variable, four characteristic variables, “activity space size”, “land cover type: woodland”, “visible waterscape type”, and “connectivity to surrounding roads”, were retained after backward stepwise screening. The detailed results are shown in Table 5 Model data summary.

[0080] Table 5 Summary of model data

[0081]

[0082] After stepwise backward regression screening, the union of the above retained features was taken, and finally 9 environmental physical characteristic variables were retained, including: “land cover type: woodland”, “connectivity to surrounding roads”, “recreational facilities”, “terrain changes”, “activity space size”, “view openness”, “visible waterscape type”, “noise level”, and “vegetation management level”, which were included as independent variables in the next stage of the path analysis of the mediation model.

[0083] In this embodiment, the total benefit of ecosystem cultural services, and the benefits of the three sub-dimensions of "identity construction", "physical and mental experience", and "ability improvement" are used as the dependent variables respectively, and multiple linear stepwise regressions of environmental physical characteristics and the four dependent variables are performed respectively. Through the method of backward stepwise regression, factors that have insignificant effects on the dependent variable (P>0.05) and those with high autocorrelation (VIF>10) are gradually removed, and environmental physical characteristics that have significant effects on the dependent variable are retained. Finally, the union of the above results is taken, and the environmental physical characteristics in the union are used as the independent variables for inclusion in the mediation model analysis in the next stage.

[0084] It can be understood that in the backward stepwise regression in this embodiment, all independent variables are selected into the model, and then partial F-tests are performed on each independent variable. The smallest F value is denoted as F L , which is compared with the pre-specified significance level F0. If F L <F0, the variable is removed, and the remaining variables are refitted into a regression model. The above steps are repeated until all independent variables in the model cannot be removed.

[0085] (3) Screening of significant mediating variables and influence paths based on the multiple mediation model.

[0086] In this embodiment, the Lavaan software package in Rstudio 4.0.0 is used to construct a multiple mediation model to test the mediating effects of "environmental perception characteristics (mediating variable 1: M1)" and "recreational behavior characteristics (mediating variable 2: M2)" in the influence of "environmental physical characteristics (independent variable, X)" on "ecosystem cultural benefit level (dependent variable, Y)", and to screen out significant mediating variables and influence paths.

[0087] Specifically, the multiple mediation model is:

[0088] Y = cX + e1 (1)

[0089] M = aX + e2 (2)

[0090] Y = c'X + bM + e3 (3)

[0091] Among them, the coefficient c of equation (1) is the total effect of the independent variable X on the dependent variable Y; the coefficient a of equation (2) is the effect of the independent variable X on the mediating variable M; the coefficient b of equation (3) is the effect of the mediating variable M on the dependent variable Y after controlling the influence of the independent variable X; the coefficient c′ is the direct effect of the independent variable X on the dependent variable Y after controlling the influence of the mediating variable M; e1 to e3 are regression residuals.

[0092] Specifically, the test process of the mediating effect is as follows:

[0093] The first step is to test the coefficient c of equation (1). If it is significant, the argument is based on the mediation effect. Otherwise, the argument is based on the masking effect. However, no matter whether it is significant or not, subsequent tests are carried out.

[0094] In the second step, the coefficient a of equation (2) and the coefficient b of equation (3) are tested in turn. If both are significant, the indirect effect is significant, and we proceed to the fourth step. If at least one is not significant, we proceed to the third step.

[0095] The third step is to use the Bootstrap method to directly test H0: ab = 0. If it is significant, the indirect effect is significant, and the fourth step is carried out; otherwise, the indirect effect is not significant, and the analysis is stopped.

[0096] The fourth step is to test the coefficient c' of equation (3). If it is not significant, that is, the direct effect is not significant, indicating that there is only a mediating effect. If it is significant, that is, the direct effect is significant, proceed to the fifth step.

[0097] The fifth step is to compare the signs of ab and c'. If they have the same sign, it is a partial mediation effect, and report the ratio of the mediation effect to the total effect, ab / c. If they have different signs, it is a masking effect, and report the absolute value of the ratio of the indirect effect to the direct effect, |ab / c'|.

[0098] The full mediation effect is as follows:

[0099] When environmental perception characteristics / recreational behavior characteristics enter the relationship analysis between "environmental physical characteristics" and "cultural benefits", the relationship between "cultural benefits" and "environmental physical characteristics" (direct effect) is no longer significant, and the corresponding perception characteristics / recreational behavior characteristics play a complete mediating role, which means that the impact of "environmental physical characteristics" on "cultural benefits" completely depends on the corresponding environmental perception characteristics or recreational behavior characteristics to affect cultural benefits.

[0100] Among them, some mediating effects are as follows:

[0101] If the relationship between "cultural benefits" and "environmental physical characteristics" is still significant, but the direct effect is weaker than the total effect, then the corresponding perception characteristics / recreational behavior characteristics play a partial mediating role, which means that the impact of "environmental physical characteristics" on "cultural benefits" partially depends on the corresponding environmental perception characteristics or recreational behavior characteristics to affect cultural benefits, and the rest is due to the direct impact of "environmental physical characteristics" on "cultural benefits".

[0102] The masking effect is specifically:

[0103] If the relationship between "cultural benefits" and "environmental physical characteristics" is still significant, but the direct effect is greater than the total effect (or the indirect effect is judged to be of opposite signs to the direct effect), then the corresponding perceptual characteristics / recreational behavior characteristics will have a masking effect, which will reduce the original impact of environmental physical characteristics on cultural benefits.

[0104] (3-1) Environmental perception characteristics and recreational behavior characteristics mediate the relationship between environmental physical characteristics and “identity construction”.

[0105] According to the test results of the mediation effect (Table 6), among the four environmental physical characteristics that have a significant impact on "identity construction", the impact of "noise level" and "visible waterscape type" on the benefits of "identity construction" will be transmitted through the mediating variables, which can be referred to Figure 4 .

[0106] Table 6 Results of the mediation effect test

[0107]

[0108] Note: ***, **, * represent the significance levels of 1%, 5%, and 10%, respectively; the indirect effect interval does not include 0, which means it is significant.

[0109] (3-2) Environmental perception characteristics and recreational methods mediate the relationship between environmental physical characteristics and “physical and mental experience”.

[0110] According to the test results of the mediation effect (Table 7), among the two environmental physical characteristics that have a significant impact on "physical and mental experience", "noise level" and "terrain change" will both transmit the influence of the independent variable on the dependent variable through the mediation of "environmental perception characteristics", which can be referred to Figure 5 .

[0111] Table 7 Results of the mediation effect test

[0112]

[0113]

[0114] Note: ***, **, * represent the significance levels of 1%, 5%, and 10%, respectively; the indirect effect interval does not include 0, which means it is significant.

[0115] (3-3) The mediating role of environmental perception characteristics and recreational methods between environmental physical characteristics and “capacity improvement”.

[0116] According to the test results of the mediation effect (Table 8), among the four environmental physical characteristics that have a significant impact on the benefits of "capacity improvement", except for the characteristic of "connectivity with surrounding roads", the other three environmental physical characteristics "visible waterscape type", "activity space size" and "land cover type: forest land" will all transmit the influence of the independent variable on the dependent variable through the mediation of "environmental perception characteristics" and "recreation frequency", which can be referred to Figure 6 .

[0117] Table 8 Results of the mediation effect test

[0118]

[0119]

[0120] Note: ***, **, * represent the significance levels of 1%, 5%, and 10%, respectively; the indirect effect interval does not include 0, which means it is significant.

[0121] (3-4) The mediating role of environmental perception characteristics and recreational methods between environmental physical characteristics and the total benefits of cultural services.

[0122] According to the test results of the mediation effect (Table 9), among the four environmental physical characteristics that have a significant impact on "total benefit", except for the characteristic of "view openness", the other three environmental physical characteristics "noise level", "vegetation management level" and "land cover type: forest land" will all transmit the influence of the independent variable on the dependent variable through the mediation of "environmental perception characteristics" and "recreation frequency", which can be referred to Figure 7 .

[0123] Table 9 Results of the mediation effect test

[0124]

[0125] Note: ***, **, * represent the significance levels of 1%, 5%, and 10%, respectively; the indirect effect interval does not include 0, which means it is significant.

[0126] (4) Construct a structural equation model.

[0127] (4-1) Construction of measurement model of independent variables.

[0128] This example packages the nine environmental physical characteristics by factor analysis, constructs latent variables of environmental physical characteristics of different dimensions, and explores the impact mechanism of environmental physical characteristics of different dimensions on cultural benefits (Table 10). The measurement model of environmental physical characteristics is as follows: Figure 8 shown.

[0129] (4-2) Construction of measurement model of dependent variables.

[0130] The measurement model of the dependent variable can be constructed as a second-order latent variable model consisting of three dimensions: "identity construction", "physical and mental experience" and "ability improvement", such as Fig. 9 shown.

[0131] (4-3) Construction of mediating variable measurement model.

[0132] Since the dimensions represented by the screened mediating variables are different, in order to ensure the reliability of the measurement model, this embodiment no longer packages the various mediating variable factors into items, but instead introduces each mediating variable independently into the mediation model in the form of explicit variables, thereby clearly revealing the role of each mediating variable in the cultural benefit influencing mechanism.

[0133] (5) Conduct model fit analysis to obtain the revised structural equation model.

[0134] This example uses the weighted least squares estimation method with mean and variance correction to estimate the model parameters, corrects the variables with correction index>5, removes the mediating variables "diversity" and "sense of tranquility", and obtains the final model parameter estimation results (Table 10). All indicators meet the fitting standards, which means that the constructed structural equation model is ideal and the influence mechanism of cultural benefits revealed is credible. The corrected structural equation model is shown in the figure below: Fig.10 shown.

[0135] Table 10 Model fit results

[0136]

[0137] The specific fitting criteria are:

[0138] 1) Goodness of fit index (GFI) ≥ 0.90;

[0139] 2) comparative fit index (CFI) ≥ 0.90;

[0140] 3) Root mean square error of approximation (RMSEA) ≤ 0.08;

[0141] 4) Use standardized root mean square residual (SRMR) ≤ 0.1.

[0142] (6) Conduct impact path analysis of the modified structural equation model.

[0143] Table 11 reports all the model path coefficients in the structural equation model. The size of the standardized path coefficient shows the relationship between the variables and the degree of influence. Whether the path coefficient between the variables is valid is judged by the P value. As long as P < 0.05, the path coefficient can be judged to be significant.

[0144] The P values ​​of the direct impact paths of "natural characteristics of the site", "sensory characteristics of the site" and "basic conditions and facilities of the site" on "cultural benefits" were all greater than 0.05, which did not reach the significance level, indicating that in this model, "natural characteristics of the site", "sensory characteristics of the site" and "basic conditions and facilities of the site" had no direct effect on cultural benefits.

[0145] The standardized coefficients of the environmental perception characteristics of "sociality", "consistency", "openness", "sheltered feeling", and "recreation frequency" of tourists on "cultural benefits" are 0.1993, 0.1443, 0.2719, 0.2472, and 0.1300, respectively, and the P values ​​are all less than 0.05, reaching a significant level, indicating that the "sociality", "consistency", "openness", "sheltered feeling" level and "recreation frequency" of the venue perceived by tourists have a direct effect on cultural benefits. Among them, the standardized path coefficient of "openness" in environmental perception characteristics is the largest, indicating that "openness" has the greatest impact on cultural benefits and is also the dimension that has the greatest impact on environmental perception characteristics among all the physical environment characteristics of the venue. "Recreation frequency" has the smallest impact on cultural benefits.

[0146] Secondly, the results of the influence paths of "site natural characteristics", "site sensory characteristics" and "site basic conditions and facilities" on the perception characteristics of the environment by recreationists are as follows: First, "site sensory characteristics" have an impact on the levels of "sociality", "consistency" and "openness" perceived by recreationists, with P values ​​less than 0.05, and the corresponding standardized coefficients are -0.0704, 0.1211, and 0.2056. Among them, "site natural characteristics" have the greatest effect on the perception characteristics of "openness". In addition, the influence path of "site natural characteristics" on the "sense of shelter" characteristic has a P value greater than 0.05, indicating that it has not reached a significant level. Therefore, "site natural characteristics" have no direct effect on the sense of shelter. Secondly, the P values ​​of the influence paths of "site natural characteristics" on the perception characteristics are all greater than 0.05. Therefore, "site natural characteristics" have no direct effect on the environmental perception characteristics. In addition, among the influence paths of the dimension of "site basic conditions and facilities" on the environmental perception characteristics, only the influence path on "sense of shelter" has a P value less than 0.05, which is significant, and the corresponding standardized path is -0.1417. The P values ​​of the remaining paths were all greater than 0.05 and did not reach the significance level.

[0147] In addition, the impact paths and coefficients of environmental physical characteristics and environmental perception characteristics on recreational frequency are as follows: First, among the three dimensions of environmental physical characteristics, the P values ​​of the impact paths of "site natural characteristics" and "site basic conditions and facilities" on recreational frequency are both less than 0.05, which shows that the above two dimensions have a direct effect on the recreational frequency of recreationists, and the corresponding effect sizes (absolute values ​​of standardized coefficients) are 0.1435 and 0.2149, that is, the impact of "site basic conditions and facilities" on "recreation frequency" is greater than the impact of "site natural characteristics" on "recreation frequency", and it is also the largest effect size among the many factors that can affect "recreation frequency". The P value of the impact path of "site sensory characteristics" on recreational frequency is greater than 0.05, which did not reach the significance level, indicating that "site sensory characteristics" do not have a direct impact on recreational frequency.

[0148] Secondly, among the environmental perception characteristics, only the "consistency" and "openness" characteristics have an impact path P value of less than 0.05 on the "recreation frequency", reaching a significant level, indicating that the "consistency" and "openness" characteristics of the venue perceived by the recreationists have a direct effect on the recreation frequency of the venue visited by the recreationists, with corresponding coefficients of 0.0773 and -0.0735, which means that as the level of "consistency" characteristics perceived by the recreationists in the green space increases, it has a positive impact on the recreation frequency of the recreationists. The "openness" characteristic is the opposite.

[0149] Table 11 Results of path analysis between variables

[0150]

[0151]

[0152] According to the results of the structural equation model impact path analysis, the impact mechanism of the cultural benefits of urban green space ecosystems is as follows:

[0153] "Natural characteristics of the site" affect cultural benefits through "recreation frequency"; "sensory characteristics of the site" have a positive impact on cultural benefits by affecting the recreationists' perception of the "sociality" characteristics of the space; "sensory characteristics of the site" have a positive impact on cultural benefits by affecting the recreationists' perception of the "consistency" characteristics of the space; "sensory characteristics of the site" have a positive impact on cultural benefits by affecting the recreationists' perception of the "openness" characteristics of the space; "sensory characteristics of the site" have a positive impact on cultural benefits by affecting the recreationists' perception of the "consistency" characteristics of the space, thereby having a positive impact on the recreationists' recreation frequency, and thus having a positive impact on cultural benefits; "sensory characteristics of the site" have an impact on the recreationists' recreation frequency by affecting the recreationists' perception of the "openness" characteristics of the space, and thus having an impact on cultural benefits; the dimension of "basic conditions and facilities of the site" has a negative impact on cultural benefits by negatively affecting the recreationists' perception of the "sense of shelter" of the space; the dimension of "basic conditions and facilities of the site" has a positive impact on cultural benefits by positively affecting the recreationists' "recreation frequency".

[0154] S102. Build a Bayesian network model based on the influencing mechanism of cultural benefits of urban green space ecosystem.

[0155] Using the Bnlear package in R Studio, referring to the influencing mechanism of the cultural benefits of the urban green space ecosystem revealed by the structural equation model, a Bayesian network model with the same structure as the influencing mechanism was built to predict the relationship between variables. In the Bayesian network, the circle represents the variable, the arrow represents the relationship between the variables, and the direction of the arrow means the direction of the influence transmission ( Fig.11 ).

[0156] In the Bayesian network, circles are used to represent variables, environmental physical characteristics are used as independent variables, environmental perception characteristics are used as mediating variables 1, recreational behavior characteristics are used as mediating variables 2, and ecosystem service benefits are used as dependent variables. Arrows are used to represent the causal relationship between variables. The starting end of the arrow is the "cause", that is, the "parent node", and the direction it points to is the "result", that is, the "child node". The direction of the arrow represents the direction of the transmission of the influence. The conditional probability table is used to describe the degree of dependence of the child node on the parent node. The probability of the parent node is determined by the marginal probability distribution, and its distribution is generally not affected by other factors in the network; the conditional probability of the child node is determined under the joint action of all the parent nodes it depends on; when the attribute of a parent node changes, the conditional probability of a series of corresponding child nodes will change.

[0157] Since the Bayesian network is a classification prediction tool, the model requires that the input variable levels are all categorical data. Therefore, different types of variables need to be preprocessed. First, for the independent variable environmental physical characteristics, its essence is an ordered categorical variable, which can be regarded as a continuous variable and included in the structural equation, or it can be maintained as a categorical variable and included in the analysis of the Bayesian network. Therefore, the original classification level of the independent variable will be followed in the Bayesian network. Secondly, for the environmental perception characteristics and recreation frequency as mediating variables, their essence is also an ordered categorical variable. Therefore, like the independent variables, their original classification level is also used as their corresponding characteristic level in the Bayesian network. Finally, for the dependent variable cultural benefits, its essence is a continuous variable. Therefore, in the Bayesian network, the total benefit of ecosystem cultural services and its corresponding sub-benefits are divided into three gradients: low level, medium level and high level.

[0158] This embodiment adopts a 10-fold cross-validation method to verify the accuracy of the built Bayesian network model by calculating the "posterior classification error" of the model. The advantage of the posterior classification error is that it can verify the accuracy of any group of nodes. Therefore, in this study, the accuracy of the four nodes of "total benefit", "ecosystem cultural service sub-benefit: identity construction", "ecosystem cultural service sub-benefit: physical and mental experience", and "ecosystem cultural service sub-benefit: capacity improvement" were verified to ensure that the constructed Bayesian network is effective in different dimensions.

[0159] The specific verification process is as follows: First, the data set is evenly divided into 10 parts, one of which is taken as the test set and the others as the training set. A training is started to obtain the corresponding error rate. After that, one of them is selected as the test set and the others as the training set in turn, and the model is trained in turn to obtain the error rate. Finally, based on the error rate of 10 iterations, the average error rate is calculated as the final model validity verification result (Table 12). The verification results show that the accuracy of the sub-dimension benefits of ecosystem cultural services is higher than the evaluation accuracy of the total benefit. Among them, the accuracy of the total benefit measurement model is 71%, the evaluation accuracy of the "identity construction" dimension is 83%, the evaluation accuracy of the "physical and mental experience" dimension is 75%, and the evaluation accuracy of the "capacity improvement" dimension is the highest, with a corresponding level of 84%. According to relevant research experience, this level of accuracy has reached the accuracy of machine learning in related fields such as landscape preference (60%-85%), so the prediction results of the model have reliable accuracy.

[0160] Table 12 Bayesian network model validity verification results

[0161]

[0162] In this embodiment, the typical combination of environmental physical characteristic levels corresponding to high-level cultural benefits and the typical combination of environmental physical characteristic levels corresponding to low-level cultural benefits generated by machine learning are input into the Bayesian network model to test whether the predicted cultural benefit results are consistent with the results of machine learning.

[0163] The results show that when the characteristic levels of the recreational unit are [extremely low visual openness (VO1), still water surface visible in the unit (WV2), noise level >65dB (NL4), small undulating hills (TP1), land cover type is woodland (LU1), vegetation has no signs of maintenance (VM1), no recreational facilities (AM1), no staying space (ASS1), no unit entrance connected to main and secondary roads (including those with obstacles that are inaccessible) (CSR1)], the probability that people perceive the ecosystem cultural benefits of the unit as a low benefit level is as high as 63.0%, which is dominant. Therefore, it can be considered that the prediction results of the Bayesian network echo the low benefit level corresponding to this combination reported by machine learning. When the characteristics of the recreational unit are [extremely high visual openness (VO4), visible moving water surface in the unit (WV3), noise level <45dB (quiet) (NL1), micro-topography (TP2), non-forest land (LU1), well-pruned vegetation, no weeds or dead trees (VM4), landmark structures / attractions (A㎡), area >460㎡, capacity >100 people (ASS6), unit entrance connected to secondary garden path (CSR3)], the probability that people perceive the ecosystem cultural benefits as high-benefit level is as high as 41.3%, which is dominant. Therefore, it can be considered that the prediction results of the Bayesian network are consistent with the high-benefit level of this combination reported by machine learning.

[0164] The cross-validation results of the combination of environmental physical characteristics corresponding to different benefit levels show that the Bayesian network constructed to predict the ecosystem cultural benefits of urban green space through environmental physical characteristics is applicable.

[0165] This embodiment uses the python high-dimensional visualization tool - plotly to build an interactive and visualization module. The color band represents the probability of high ecosystem cultural benefits. The larger the value, the higher the probability. The vertical axis represents the different levels of environmental physical characteristics and their subordinates. The pink vertical axis represents the selected feature level. The lines between the axes represent the combination of feature levels. The color of the line corresponds to the color band, indicating the probability level of selecting this feature level combination to bring about a high level of ecosystem cultural benefits. For more information, please refer to Fig.12 .

[0166] S103. Use the Bayesian network model to predict the level probability value of cultural benefits and optimize the urban green space ecosystem.

[0167] In order to transform the graphical design scheme into an object suitable for quantitative evaluation by Bayesian network, it is necessary to take the recreational unit that constitutes the scheme as the smallest unit, disassemble the environmental physical characteristics of the recreational unit according to the given level division standard, and determine the dominant environmental object characteristic level of each recreational unit (refer to Fig.13 ). In addition, since the noise level depends on the surrounding environment of the recreational unit and the vegetation management level depends on the daily maintenance of the urban green space, it is not included in the variable variables of the design scheme in the demonstration process of the scheme selection. The noise level can be determined according to the surrounding environment of the site as a given condition before the design scheme is generated. In the demonstration process of this embodiment, the median value of the noise level, 45-55dB, is taken as the given value, and the vegetation management level is "vegetation is carefully trimmed, no weeds and dead trees" as a fixed value, and the best scheme is found by changing other feature levels.

[0168] The quantitative indicators corresponding to the nine environmental physical characteristics selected in this embodiment are:

[0169] The quantitative indicators for “Land cover type: forest land” are: forest land and non-forest land;

[0170] The quantitative indicators of visible waterscape types are: visible combinations of multiple water bodies, visible dynamic water surface, visible static water surface, and no water body can be seen;

[0171] The quantitative indicators of connectivity with surrounding roads are: connected to the primary garden road, connected to the secondary garden road, connected to the branch road, and not connected;

[0172] The quantitative indicators of sight openness are: extremely high sight openness, high sight openness, low sight openness, and extremely low sight openness;

[0173] The quantitative indicators of recreational facilities are: children's play facilities, fitness equipment, jogging tracks, landmark structures / attractions and none;

[0174] The quantitative indicators of terrain changes are: flat, micro-topography and small undulating mountains;

[0175] The quantitative indicators of the size of the activity space are: area > 460 m2, accommodating > 100 people, area 230-460 m2, accommodating 50-100 people, area 92-230 m2, accommodating 20-50 people, area 46-92 m2, accommodating 10-20 people, area < 46 m2, accommodating < 10 people, area < 2.4 m2;

[0176] The quantitative indicators of noise level are: <45dB (quiet), 45dB-55dB (normal conversation can be heard), 55dB-65dB (a bit noisy), >65dB (noise);

[0177] The quantitative indicators of "vegetation management level" are: vegetation with no signs of maintenance, mainly wild grass and dead trees, vegetation with little pruning, wild grass and dead trees, vegetation with neat pruning, small amount of wild grass, vegetation with careful pruning, and no wild grass and dead trees.

[0178] (1) Comparison and selection of single-factor optimization and transformation plans.

[0179] Assume that the recreational unit to be optimized is a "woodland with a static water surface visible, connected to a branch road, a noise level of 45-55dB (i.e., the level at which normal conversations can be heard), carefully trimmed vegetation without weeds and dead trees, undulating terrain, a hard activity space of 46-92㎡, and no recreational facilities." The goal of the transformation and optimization is to select the type of recreational facilities for the recreational unit, that is, to analyze what type of recreational facilities are suitable for placement in the site, so that recreationists can obtain higher ecosystem cultural benefits.

[0180] The following is the process of comparing the optimal transformation plan for a single factor based on the Bayesian network:

[0181] The existing environmental physical characteristics of the recreational unit to be transformed are transformed according to the level classification given by the model to generate the initial cultural benefits of the site as a comparison benchmark for the scheme. Next, the different levels of the target characteristics to be transformed are input into the Bayesian network as variables to obtain the cultural benefit levels of the corresponding schemes. Then, according to the level of cultural benefits, the high-value scheme is selected as the optimal scheme. If there are multiple groups of high-value schemes, all of them are considered as alternatives, and by associating other evaluation criteria, such as cost, other ecosystem benefits, etc., the optimal scheme that meets multiple objectives is selected from the high-level schemes.

[0182] like Fig.14 The figure shows the current environmental characteristics of the site and its corresponding cultural benefits. The low, medium and high probabilities of the cultural benefits corresponding to the current site are 0.317, 0.317 and 0.366 respectively. Gray-green indicates the established and unchangeable characteristic level of the site.

[0183] According to the level classification of recreational facilities: "structures / attractions", "jogging trails", "fitness equipment" and "children's play facilities", excluding the existing "no recreational facilities" on the site, the following four schemes can be generated by superimposing the above recreational facilities on the level of the site's current environmental characteristics. You can refer to Fig.15 .

[0184] Based on the multiple transformation plans of “recreational facility types”, the corresponding cultural benefits are as follows: Fig.16As shown in Table 1, when the recreational facilities are "landmark structures / attractions" (Scheme 1), the high probability line corresponding to the cultural benefits of the recreational unit is the darkest, that is, when the recreational unit is equipped with landmark structures or other special attractions, the probability that the site can produce high-level cultural benefits is the highest, at 52.91%, which is much higher than the current status of the site and other renovation plans, and can be put into practice as the optimal renovation plan. The specific benefit prediction results of each renovation plan are shown in Table 13.

[0185] Table 13 Cultural benefit prediction results of all characteristic level combination schemes

[0186]

[0187] (2) Comparison and selection of multi-factor optimization and transformation plans.

[0188] Different from the renovation design, when facing a brand new site design, in addition to the site's limited conditions, the levels of other environmental physical characteristics are all dependent on the given when the scheme is designed. Therefore, in this case, what needs to be faced is a small number of feature level limitations, and the remaining multiple feature levels are in an unknown state. At this time, the "full space search" function of the Bayesian network can be used to search for all possible combinations of feature levels, and then select the best one. "Full space search" means including multiple environmental physical feature levels and enumerating all possible combinations of environmental physical feature levels.

[0189] Assuming that the recreation unit to be designed is a "non-forested land with a static water surface visible, connected to a secondary garden path, with a noise level of 45-55dB, carefully trimmed vegetation without weeds and dead trees, and children's play facilities", how to design the remaining characteristic levels, including: terrain changes, recreational facilities, size of activity space, and openness of sight, so that recreationists can obtain higher cultural benefits?

[0190] The following is the process of selecting the optimal design solution for multiple factors based on the Bayesian network:

[0191] The site status is input as a constraint into the Bayesian network (refer to Fig.17 ). Then, all the target design features are used as the permutation and combination elements to generate new solutions. Through the "full space search" function of the Bayesian network, all possible combinations of feature levels are enumerated, and the cultural benefit visualization results of the corresponding combinations are obtained at the same time (refer to Fig.18 ). Then, the schemes with high cultural benefit levels are selected as alternatives. Due to the combination of multiple features, there is often more than one set of high-value schemes. Finally, other evaluation criteria are associated, such as cost, other ecosystem benefits, etc., to select the best scheme that meets multiple objectives from the high-level schemes.

[0192] Under the above-mentioned conditions, all characteristic levels can be combined into 72 schemes. Considering that the attribute of no activity space conflicts with the requirement of placing children's amusement facilities, the level of "no activity space, or the short side of the activity space is less than 2.4m" is eliminated, and finally 60 feasible schemes are obtained. The cultural benefit prediction results corresponding to the specific schemes are shown in Table 14. The high-level prediction probability of cultural benefits corresponding to Schemes 1 to 10 is 41.71%, the highest among all schemes. At the same time, the low-level probability is 31.40%, and the medium-level probability is 26.9%. Therefore, Schemes 1 to 10 can be considered as ideal alternatives that meet the site's limited conditions. The 10 schemes can be selected and further combined with other evaluation indicators to select the optimal scheme that meets multiple objectives.

[0193] Table 14 Cultural benefit prediction results of all characteristic level combination schemes

[0194]

[0195]

[0196] Those skilled in the art will appreciate that all or part of the steps in the method for implementing the above embodiments may be completed by instructing related hardware through a program, and the corresponding program may be stored in a computer-readable storage medium.

[0197] It should be noted that although the method operations of the above embodiments are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. On the contrary, the steps depicted can be performed in a different order. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step, and / or one step can be decomposed into multiple steps.

[0198] Embodiment 2:

[0199] like Fig.19 As shown, this embodiment provides an optimization system for an urban green space ecosystem based on cultural benefit prediction, the system comprising a first screening module 1901, a second screening module 1902, an integration and analysis module 1903, a construction module 1904, a collection module 1905, a first optimization module 1906 and a second optimization module 1907, wherein:

[0200] The first screening module 1901 is used to screen the environmental physical characteristics that have a significant impact on cultural benefits through a backward stepwise regression method based on the "environmental physical characteristics-environmental perception characteristics-recreational behavior characteristics-cultural benefits" cascade model; wherein the multiple sub-dimensions of cultural benefits and the total benefit are respectively used as dependent variables, the environmental physical characteristics are used as independent variables, and the environmental perception characteristics and recreational behavior characteristics are used as mediating variables;

[0201] The second screening module 1902 is used to screen the mediating variables and paths with mediating effects in the impact of environmental physical characteristics on cultural benefits through path analysis of multiple mediating models based on the screened environmental physical characteristics;

[0202] Integration and analysis module 1903 is used to integrate the selected environmental physical characteristics, as well as the selected mediating variables and paths into the structural equation model, and to perform a goodness of fit analysis on the structural equation model, and to modify the variables whose modification index is greater than the set threshold; the modified structural equation model reveals the influencing mechanism of the cultural benefits of the urban green space ecosystem;

[0203] Building module 1904, for building a Bayesian network model with the same structure as the cultural benefit impact mechanism of the urban green space ecosystem, so as to predict the level probability value of the cultural benefit;

[0204] The collection module 1905 is used to collect the environmental physical characteristic level of the current situation of the site to be designed; the environmental physical characteristic level includes environmental physical characteristics and corresponding quantitative indicators;

[0205] The first optimization module 1906 is used to input the environmental physical characteristic level of the site to be designed and the different quantitative indicators of the target characteristics to be transformed as variables into the Bayesian network model if it is a single-factor optimization scheme, and predict the level probability value of the cultural benefit corresponding to the different quantitative indicators of the target characteristics to be transformed; the scheme with the largest level probability value is taken as the optimal scheme;

[0206] The second optimization module 1907 is used to input the environmental physical characteristic level of the current site to be designed as a variable into the Bayesian network model if it is a multi-factor optimization plan, generate all possible combinations of environmental physical characteristic levels, and predict the horizontal probability value of the cultural benefit corresponding to each combination of environmental physical characteristic levels; the plan with the largest horizontal probability value is taken as the optimal plan.

[0207] The specific implementation of each module in this embodiment can refer to the above-mentioned embodiment 1, which will not be described one by one here; it should be noted that the system provided in this embodiment is only illustrated by the division of the above-mentioned functional modules. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure can be divided into different functional modules to complete all or part of the functions described above.

[0208] Embodiment 3:

[0209] This embodiment provides a terminal device, which can be a computer. Fig. 20 As shown, it comprises a processor 2002, a memory, an input device 2003, a display 2004 and a network interface 2005 connected via a system bus 2001, wherein the processor is used to provide computing and control capabilities, the memory comprises a non-volatile storage medium 2006 and an internal memory 2007, wherein the non-volatile storage medium 2006 stores an operating system, a computer program and a database, and the internal memory 2007 provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. When the processor 2002 executes the computer program stored in the memory, the optimization method for the urban green space ecosystem based on cultural benefit prediction of the above-mentioned embodiment 1 is implemented.

[0210] Embodiment 4:

[0211] This embodiment provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the method for optimizing the urban green space ecosystem based on cultural benefit prediction of the above-mentioned embodiment 1 is implemented.

[0212] It should be noted that the computer-readable storage medium of the present embodiment may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: 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.

[0213] The above is only a preferred embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solution and inventive concept of the present invention within the scope disclosed by the present invention, which shall fall within the protection scope of the present invention.

Claims

1. A method for optimizing the urban green space ecosystem based on cultural benefit prediction, characterized in that: The method comprises: Based on the cascade model of "environmental physical characteristics-environmental perception characteristics-recreational behavior characteristics-cultural benefits", the backward stepwise regression method was used to screen the environmental physical characteristics that had a significant impact on cultural benefits; among them, multiple sub-dimensions of cultural benefits and total benefits were used as dependent variables, environmental physical characteristics were used as independent variables, and environmental perception characteristics and recreational behavior characteristics were used as mediating variables. Based on the selected environmental physical characteristics, the mediating variables and paths with mediating effects in the impact of environmental physical characteristics on cultural benefits are screened through path analysis of multiple mediation models. The screened environmental physical characteristics, as well as the screened mediating variables and paths, were integrated into the structural equation model, and the goodness of fit of the structural equation model was analyzed. The variables whose correction index was greater than the set threshold were corrected. The corrected structural equation model revealed the influencing mechanism of the cultural benefits of the urban green space ecosystem. Build a Bayesian network model with the same structure as the cultural benefit impact mechanism of urban green space ecosystem to predict the level probability value of cultural benefits; Collect the environmental physical characteristic levels of the current site to be designed; the environmental physical characteristic levels include environmental physical characteristics and corresponding quantitative indicators; If it is a single-factor optimization scheme, the environmental physical characteristics of the site to be designed and the different quantitative indicators of the target characteristics to be transformed are input as variables into the Bayesian network model to predict the level probability values ​​of the cultural benefits corresponding to the different quantitative indicators of the target characteristics to be transformed; the scheme with the largest level probability value is taken as the optimal scheme; If it is a multi-factor optimization scheme, all possible combinations of environmental physical characteristic levels are enumerated, and the environmental physical characteristic levels of the current site to be designed and the environmental physical characteristic levels of each combination are input as variables into the Bayesian network model to predict the horizontal probability value of the cultural benefits corresponding to each combination of environmental physical characteristic levels; the scheme with the largest horizontal probability value is taken as the optimal scheme.

2. The optimization method according to claim 1, characterized in that: The environmental physical characteristics that have a significant impact on cultural benefits are screened by the backward stepwise regression method, including: Perform multiple linear stepwise regression for independent variables and dependent variables respectively; Through backward stepwise regression, environmental physical characteristics with an effect on the dependent variable of P <= 0.05 and autocorrelation factor VIF <= 10 were retained; The union of the retained independent variables is taken as the screened environmental physical characteristics.

3. The optimization method according to claim 1, characterized in that: The multiple mediation model is: Y=cX+e1 (1) M=aX+e2 (2) Y=c'X+bM+e3 (3) Among them, the coefficient c of equation (1) is the total effect of the independent variable X on the dependent variable Y; the coefficient a of equation (2) is the effect of the independent variable X on the mediating variable M; the coefficient b of equation (3) is the effect of the mediating variable M on the dependent variable Y after controlling the influence of the independent variable X; the coefficient c′ is the direct effect of the independent variable X on the dependent variable Y after controlling the influence of the mediating variable M; e1~e3 are regression residuals.

4. The optimization method according to claim 1, characterized in that: The integration of the screened environmental physical characteristics, as well as the screened mediating variables and paths into the structural equation model includes: Construct the measurement model of independent variables: By packaging the environmental physical characteristics, construct latent variables of environmental physical characteristics of different dimensions to explore the impact mechanism of environmental physical characteristics of different dimensions on cultural benefits, so that the results revealed by the measurement model of independent variables are universal and instructive; Construct the measurement model of the dependent variable: a second-order latent variable model consisting of multiple sub-dimensions; Construct a measurement model of mediating variables: Introduce each mediating variable independently into the measurement model of the mediating variables in the form of explicit variables to clearly reveal the role of each mediating variable in the cultural benefit influencing mechanism.

5. The optimization method according to claim 1, characterized in that: The structural equation model was analyzed for goodness of fit using weighted least squares estimation with mean and variance correction.

6. The optimization method according to claim 1, characterized in that: The Bayesian network model having the same structure as the cultural benefit impact mechanism of the urban green space ecosystem includes: The variables and influence paths in the modified structural equation model are used as nodes and causal relationships in the Bayesian network, respectively; wherein the variables include the screened environmental physical characteristics, the screened mediating variables and the dependent variables; the causal relationship is represented by an arrow, the starting end of the arrow is the "parent node", the direction it points to is the "child node", and the direction indicated by the arrow represents the direction of the influence transmission; the conditional probability table is used to describe the degree of dependence of the child node on the parent node, the probability of the parent node is determined by the marginal probability distribution, and its distribution is generally not affected by other factors in the network; the conditional probability of the child node is determined under the joint action of all the parent nodes it depends on; when the attribute of a parent node changes, the conditional probabilities of a series of corresponding child nodes will change.

7. The optimization method according to any one of claims 1 to 6, characterized in that: The physical characteristics of the environment include the sensory characteristics of the site, the natural characteristics of the site, and the basic conditions and facilities of the site. Among them, the sensory characteristics of the site include the openness of sight, the type of visible waterscape and the noise level. The natural characteristics of the site include terrain changes and land cover types: woodland. The basic conditions and facilities of the site include the level of vegetation management, recreational facilities, size of activity space, connectivity with surrounding roads, distance from the main entrance and convenient facilities.

8. The optimization method according to any one of claims 1 to 6, characterized in that: The multiple sub-dimensions of cultural benefits include identity construction, physical and mental experience, and ability improvement.

9. An optimization system for urban green space ecosystem based on cultural benefit prediction, characterized in that: The system comprises: The first screening module is used to screen environmental physical characteristics that have a significant impact on cultural benefits through the backward stepwise regression method based on the "environmental physical characteristics-environmental perception characteristics-recreational behavior characteristics-cultural benefits" cascade model; among them, multiple sub-dimensions of cultural benefits and total benefits are used as dependent variables, environmental physical characteristics are used as independent variables, and environmental perception characteristics and recreational behavior characteristics are used as mediating variables; The second screening module is used to screen the mediating variables and paths with mediating effects in the impact of environmental physical characteristics on cultural benefits through path analysis of multiple mediation models based on the screened environmental physical characteristics; The integration and analysis module is used to integrate the selected environmental physical characteristics, as well as the selected mediating variables and paths into the structural equation model, and to perform a goodness of fit analysis on the structural equation model, and to modify the variables whose modification index is greater than the set threshold. The modified structural equation model reveals the influencing mechanism of the cultural benefits of the urban green space ecosystem. A module is built to build a Bayesian network model with the same structure as the cultural benefit impact mechanism of the urban green space ecosystem to predict the horizontal probability value of cultural benefits; A collection module, used to collect the environmental physical characteristic levels of the current status of the site to be designed; the environmental physical characteristic levels include environmental physical characteristics and corresponding quantitative indicators; The first optimization module is used for, if it is a single-factor optimization scheme, inputting the environmental physical characteristic level of the current site to be designed and the different quantitative indicators of the target characteristics to be transformed as variables into the Bayesian network model, predicting the level probability value of the cultural benefit corresponding to the different quantitative indicators of the target characteristics to be transformed; the scheme with the largest level probability value is taken as the optimal scheme; The second optimization module is used to input the environmental physical characteristic levels of the current site to be designed as variables into the Bayesian network model if it is a multi-factor optimization plan, generate all possible combinations of environmental physical characteristic levels, and predict the horizontal probability value of the cultural benefits corresponding to each combination of environmental physical characteristic levels; the plan with the largest horizontal probability value is taken as the optimal plan.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the optimization method described in any one of claims 1 to 8 is implemented.

Images