A method and system for analyzing and providing early warning of environmental carrying capacity

Abstract

This invention provides a method and system for analyzing and providing early warning of environmental carrying capacity, applied in the field of data processing technology. This application performs data forwarding and standardization processing on cleaned environmental carrying capacity information to generate target indicator features; processes the target indicator features to generate cloud model parameter information; performs weight allocation processing on the cloud model parameter information to generate weight information corresponding to each indicator; processes the target indicator features based on the weight information corresponding to each indicator to generate the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree; and processes the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree based on the target environmental carrying capacity early warning model to generate environmental carrying capacity early warning information for the area to be evaluated.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for analyzing and providing early warning of environmental carrying capacity. Background Technology

[0002] Traditional methods for environmental carrying capacity analysis and early warning have been widely used in past practices, providing some support for regional development planning. However, environmental systems are complex, and data is characterized by uncertainty, inaccuracy, and incompleteness. Traditional methods have limitations in processing this data, making it difficult to effectively quantify the impact of data uncertainty on assessment results. For example, in water quality monitoring, data is subject to errors due to the accuracy of monitoring equipment and spatiotemporal limitations. However, traditional indicator-based methods directly use raw data for calculations, which may lead to assessment results that deviate from reality.

[0003] Environmental factors change constantly over time, and traditional assessment methods mostly focus on static evaluation, failing to reflect the dynamic evolution of these factors in real time. When assessing regional atmospheric carrying capacity, it is difficult to reflect the dynamic impact of factors such as industrial development and energy structure adjustments on air pollutant emissions and environmental quality, thus failing to provide timely and accurate dynamic information for environmental management.

[0004] Furthermore, environmental carrying capacity involves multiple interconnected systems, and traditional methods often analyze these systems in isolation, neglecting the complex interactions between them. In the assessment of environmental carrying capacity in tourist attractions, traditional methods evaluate the ecological environment and tourism reception capacity separately, but fail to fully consider the impact of tourism activities on the ecological environment and the feedback of ecological changes on sustainable tourism development. This results in assessment results that cannot comprehensively and accurately reflect the true environmental carrying capacity of the tourist attraction.

[0005] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore includes information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this application is to provide a method and system for analyzing and providing early warning of environmental carrying capacity, which at least to some extent overcomes the problems existing in the prior art. This method involves cleaning environmental carrying capacity information, classifying it into normally distributed and non-normally distributed data, processing each type separately to filter out outliers and imput missing values, resulting in cleaned data. The cleaned data is then forward-oriented and standardized, generating different types of indicators based on their properties and converting them into a unified and comparable form, thereby obtaining the target indicator characteristics and laying the foundation for subsequent analysis. Cloud model parameters are calculated based on the target indicator characteristics. Expected features are obtained by filtering feature values, and entropy features are calculated by combining these with the target indicator characteristics. Finally, the three are integrated to obtain hyper-entropy features. Simultaneously, the cloud model parameters are processed to generate data standard deviation, correlation coefficient matrix, etc., to determine the comprehensive weight information of each indicator and normalize it, obtaining objective weight information. Furthermore, a cloud evaluation scale and comparison matrix are constructed to calculate relative weight values, and finally, subjective and objective weights are integrated to determine the weight of each indicator. Using the weights of each indicator to process the target indicator characteristics, the average membership degree, fluctuation range value, and noise level value of environmental carrying capacity are obtained. Inputting these values ​​into the target environmental carrying capacity early warning model and comparing them with preset thresholds determines the early warning level of the environmental carrying capacity of the area to be assessed, and generates corresponding early warning information. This helps improve the scientific nature of environmental carrying capacity assessment and provides a more reliable basis for environmental planning, management and decision-making, thus promoting regional sustainable development.

[0007] Other features and advantages of this application will become apparent from the following detailed description, or may be learned in part by practice of the invention.

[0008] According to one aspect of this application, a method for analyzing and providing early warning of environmental carrying capacity is provided, comprising: acquiring environmental carrying capacity information of an area to be assessed and a target environmental carrying capacity early warning model; performing data cleaning processing on the environmental carrying capacity information of the area to be assessed to generate cleaned environmental carrying capacity information; performing data forwarding and data standardization processing on the cleaned environmental carrying capacity information to generate target indicator features; processing the target indicator features to generate cloud model parameter information; performing weight allocation processing on the cloud model parameter information to generate weight information corresponding to each indicator; processing the target indicator features based on the weight information corresponding to each indicator to generate the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree; and processing the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree based on the target environmental carrying capacity early warning model to generate environmental carrying capacity early warning information for the area to be assessed.

[0009] Another aspect of this application discloses an environmental carrying capacity analysis and early warning device, characterized in that it comprises: an acquisition module for acquiring environmental carrying capacity information of the area to be assessed and a target environmental carrying capacity early warning model; and a processing module for performing data cleaning on the environmental carrying capacity information of the area to be assessed to generate cleaned environmental carrying capacity information; performing data forwarding and data standardization on the cleaned environmental carrying capacity information to generate target indicator features; processing the target indicator features to generate cloud model parameter information; performing weight allocation processing on the cloud model parameter information to generate weight information corresponding to each indicator; processing the target indicator features based on the weight information corresponding to each indicator to generate the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree; and processing the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree based on the target environmental carrying capacity early warning model to generate environmental carrying capacity early warning information for the area to be assessed.

[0010] According to another aspect of this application, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a second processor, implements the above-described method for analyzing and warning of environmental carrying capacity.

[0011] This application provides a method and system for analyzing and providing early warning of environmental carrying capacity. The server first acquires environmental carrying capacity information and a target early warning model for the area to be assessed. Then, the environmental carrying capacity information is cleaned, categorized into normally distributed and non-normally distributed data, and processed separately to filter out outliers and imput missing values, resulting in cleaned data. Subsequently, the cleaned data undergoes positive transformation and standardization, generating different types of indicators based on their properties and converting them into a unified and comparable form, thereby obtaining the characteristics of the target indicators and laying the foundation for subsequent analysis.

[0012] Based on the target indicator features, cloud model parameter information is calculated. Expected features are obtained by filtering feature values, and entropy features are calculated by combining these with the target indicator features. Finally, the three are integrated to obtain hyper-entropy features. Simultaneously, the cloud model parameter information is processed to generate data standard deviation, correlation coefficient matrix, etc., to determine the comprehensive weight information of each indicator and normalize it, obtaining objective weight information. Furthermore, a cloud evaluation scale and comparison matrix are constructed to calculate relative weight values. Finally, the subjective and objective weights are integrated to determine the weight of each indicator. Using the weights of each indicator, the target indicator features are processed to obtain the average membership degree, fluctuation range value, and noise level value of environmental carrying capacity. These values ​​are input into the target environmental carrying capacity early warning model and compared with preset thresholds to determine the early warning level of the environmental carrying capacity of the area to be evaluated, generating corresponding early warning information to assist in relevant decision-making, thereby achieving an accurate grasp and reasonable response to the environmental carrying capacity status.

[0013] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0014] Figure 1 This document illustrates a flowchart of an environmental carrying capacity analysis and early warning method provided in an embodiment of this application.

[0015] Figure 2 A schematic diagram of the structure of an environmental carrying capacity analysis and early warning device provided in an embodiment of this application is shown. Detailed Implementation

[0016] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0017] The following is combined Figure 1 This describes a method for analyzing and providing early warning of environmental carrying capacity according to exemplary embodiments of this application. For example... Figure 1 As shown, this method is applied to a server and includes:

[0018] S101, Obtain environmental carrying capacity information and target environmental carrying capacity early warning model of the area to be evaluated.

[0019] In one implementation, for a specific coastal city, environmental carrying capacity information comes from a wide range of sources. Regarding natural resources, water resource data, such as annual available freshwater volume, can be obtained from the local water resources department. Data from the past five years shows that annual available freshwater volume fluctuates within a [specific range], influenced by rainfall and water demand. Land resource data, including arable land area and construction land area, can be obtained from land use survey data from the land resources department. In recent years, arable land area has slightly decreased due to urbanization. Mineral resource data covers reserves and extraction volume, with detailed records kept by the local mineral resources management department. Regarding environmental quality, air pollution indicators, such as the annual average concentration of PM2.5, are provided by real-time monitoring from environmental monitoring stations, and have varied within a [specific value range] over the past few years. Water quality indicators include pollutant content in rivers and oceans; monitoring data from environmental protection departments shows that some marine water quality is affected by land-based pollution. Soil pollution indicators are obtained through soil sampling analysis; some areas have certain levels of heavy metal contamination. Regarding ecological resilience, biodiversity data can be obtained through biodiversity surveys, vegetation cover is determined through a combination of remote sensing monitoring and on-site measurements, and ecological resilience is assessed based on monitoring data from relevant ecological restoration projects. In terms of socioeconomic aspects, population density data, available from statistical departments, shows a year-on-year upward trend. Industrial output data comes from the local statistics bureau, while environmental protection investment data can be obtained from government fiscal expenditure and corporate environmental protection investment statistics. These data reflect the multifaceted state of the coastal city's environmental system, providing rich information for environmental carrying capacity analysis.

[0020] The target environmental carrying capacity early warning model integrates the Analytic Hierarchy Process (AHP), cloud model theory, and the CRITIC expert fusion weighting method. AHP is used to construct a multi-level indicator system for systematic decomposition and analysis of environmental carrying capacity; cloud model theory is used to handle uncertainties and fuzziness in the data; and the CRITIC expert fusion weighting method combines objective data statistics with expert subjective knowledge to determine indicator weights, making the model results more scientific and reliable. The model decomposes environmental carrying capacity into a target layer, a criterion layer, and an indicator layer. The target layer is "regional environmental carrying capacity," which is the core objective of the entire evaluation. The criterion layer is divided into key subsystems such as natural resources, environmental quality, ecological resilience, and socio-economic factors based on the complexity of the environmental system. The indicator layer consists of specific quantifiable indicators for each criterion layer, such as annual available water volume and arable land area under the natural resources criterion layer; and annual average PM2.5 concentration and COD emissions under the environmental quality criterion layer.

[0021] The target environmental carrying capacity early warning model first acquires environmental carrying capacity information for the area to be assessed and performs data cleaning, including outlier handling and missing value imputation. Then, the data undergoes forward normalization and standardization, transforming different types of indicators into a unified and comparable form. Next, cloud model parameters are calculated to determine the weights of each indicator. Based on these parameters and weights, hierarchical aggregation calculations are performed, from the indicator layer to the criterion layer and then to the target layer, to obtain the average membership degree, the fluctuation range of the membership degree, and the noise level of the membership degree. According to the preset thresholds corresponding to different levels of carrying capacity, and in conjunction with the target environmental carrying capacity early warning model, the early warning level of the environmental carrying capacity of the area to be assessed is determined, thereby generating corresponding early warning information.

[0022] S102, perform data cleaning processing on the environmental carrying capacity information of the area to be assessed, and generate cleaned environmental carrying capacity information.

[0023] In one implementation, the environmental carrying capacity information of the area to be assessed is classified and processed to generate normally distributed data and non-normally distributed data. The collected environmental carrying capacity information includes numerous indicators, such as annual available water volume, annual average PM2.5 concentration, and population density. The distribution type of these data is determined, revealing that the annual available water volume data approximately follows a normal distribution, while the population density data exhibits a non-normal distribution due to the influence of various factors such as urban development planning and policies. This step forms the basis for subsequent processing methods for different data distributions, as normally distributed and non-normally distributed data have different characteristics and require different methods to handle outliers.

[0024] The normally distributed data is processed to generate the mean and standard deviation for each indicator, and the first outlier is generated based on the mean and standard deviation of each indicator. For the annual available water volume, which is approximately normally distributed, its mean and standard deviation are calculated. Assuming that statistical calculations show the city's annual available water volume over the past five years has a mean μ = 5 billion cubic meters and a standard deviation σ = 500 million cubic meters, according to the 3σ rule, the outlier determination condition is |x-μ|>3σ. Therefore, the first outlier range is data points less than 50-3×5 = 35 billion cubic meters or greater than 50+3×5 = 65 billion cubic meters. This is based on the characteristics of the normal distribution, identifying outliers within a certain probability range, excluding data that deviates from the normal range, and avoiding misleading subsequent analysis.

[0025] Box plots are used to process non-normally distributed data and generate second outliers. For population density, a non-normally distributed data point, box plots are employed. The quartiles are calculated; assuming Q1 = 500 people / km² and Q3 = 800 people / km², then IQR = Q3 - Q1 = 300 people / km². The second outlier is defined as data points less than Q1 - 1.5IQR = 800 - 1.5 × 300 = 50 people / km² or greater than Q1 + 1.5IQR = 800 + 1.5 × 300 = 1250 people / km². Box plots are suitable for non-normally distributed data and can effectively identify outliers. Based on the first and second outliers, outlier filtering is performed on the environmental carrying capacity information of the area to be evaluated, generating noise-filtered environmental carrying capacity information. Data points with annual available water volume less than 3.5 billion cubic meters and greater than 6.5 billion cubic meters, as well as population densities less than 50 people / km² and greater than 1250 people / km², were considered outliers. After filtering, the environmental carrying capacity information after noise filtering was obtained, removing the interference of these outliers.

[0026] Missing value imputation is performed on the noise-filtered environmental carrying capacity information to generate cleaned environmental carrying capacity information. If missing data exists in the noise-filtered environmental carrying capacity information, imputation is required. For example, if missing values ​​exist in the industrial output data for some years, and the missing data rate is less than 20%, with high correlation between variables, the KNN method is used for imputation. Assuming that the Euclidean distance between the sample containing the missing value and other samples is calculated, the five nearest neighbor samples are selected, and the average industrial output value of these five neighbor samples is 50 billion yuan; then, the missing value is filled with 50 billion yuan. If KNN imputation is not possible, such as missing data for certain soil pollution indicators, and low correlation between variables, the MICE algorithm is used for multiple imputation. Missing values ​​are predicted using an iterative regression model, generating multiple complete datasets. The results are then aggregated to fill in the missing values, thus obtaining cleaned environmental carrying capacity information, laying the foundation for subsequent accurate analysis of environmental carrying capacity.

[0027] S103 performs data forwarding and standardization processing on the cleaned environmental carrying capacity information to generate target indicator features.

[0028] In one implementation, the environmental carrying capacity information of the area to be assessed is processed to generate target-layer features, criterion-layer features, and indicator-layer features. The target-layer features represent "regional environmental carrying capacity," which is the core of the comprehensive assessment. Criterion-layer features are determined based on the complexity of the environmental system. The natural resource criterion layer (B1) covers features related to water resources, land resources, and mineral resources; the environmental quality criterion layer (B2) includes air, water, and soil pollution indicators; the ecological resilience criterion layer (B3) includes features related to biodiversity, vegetation coverage, and ecological resilience; and the socio-economic criterion layer (B4) involves features related to population density, industrial output, and environmental protection input. Indicator-layer features are a refinement of the criterion layers. For example, under the natural resource criterion layer (B1), the annual available water volume (C1, unit: 100 million m³) is... 3 ) and cultivated land area (C2, unit: km²) 2 ) is a specific indicator; under the environmental quality standard layer (B2), the annual average concentration of PM2.5 (C3, μg / m³) is... 3 The specific indicators are (C4) emissions and COD emissions (tons / year). These characteristics constitute a preliminary classification and quantification of environmental carrying capacity information, laying the foundation for subsequent analysis.

[0029] The characteristics of the target layer, criterion layer, and indicator layer are processed to generate large-scale indicators, small-scale indicators, intermediate-scale indicators, and interval-scale indicators. In the environmental carrying capacity analysis of this coastal city, the indicators are classified according to their nature and the direction of their impact on environmental carrying capacity. Indicators such as annual available water volume and vegetation coverage, where larger values ​​are more beneficial to improving environmental carrying capacity, are classified as large-scale indicators; while indicators such as annual average PM2.5 concentration and COD emissions, where smaller values ​​are better, are classified as small-scale indicators. For some indicators with optimal values, such as certain soil nutrient contents, where soil fertility is optimal within a certain range, these are intermediate-scale indicators. If the dissolved oxygen content of a certain sea area is specified to be most suitable for the marine ecosystem within a specific range [5,7] mg / L, this is an interval-scale indicator. This classification helps to process the indicators in a targeted manner afterward, making different types of indicators comparable.

[0030] Based on the first calculation formula, the smaller indicators are positively processed to generate the larger indicators. The first calculation formula is: x i,new =max{x1,x2,…,x n}-x i ; where x i,new For a large-scale indicator after positive transformation, x iFor the original small index values, max{x1,x2,…,x n} represents the maximum value of all original small-scale indicators. Taking the annual average PM2.5 concentration as an example, assuming the annual average PM2.5 concentration data for this coastal city and its surrounding area over the past five years is x1 = 35 μg / m³,... 3 x2 = 40 μg / m 3 x3 = 30 μg / m 3 x4 = 45 μg / m 3 x5 = 38 μg / m 3 The maximum value is max{x1,x2,x3,x4,x5} = 45 μg / m 3 For x i =40μg / m 3 This data, according to the first calculation formula x i,new =max{x1,x2,…,x n}-x i The positiveized value is 4540 = 5. By positiveizing, the small indicators are transformed into a form where the larger the better, which facilitates subsequent unified calculation and comparison.

[0031] The intermediate indicator is positively processed based on the second calculation formula to generate a larger positively processed indicator. The second calculation formula is as follows: Where, x' i,new For a large-scale indicator after positive transformation, M is the maximum absolute deviation, and x' i x' is the original intermediate index value. best This represents the target value for an intermediate-type indicator. Assume the optimal value x' for a certain nutrient content in the city's soil. best =50mg / kg, the soil nutrient content in a certain area is x' i = 45mg / kg. First calculate M = max{|x' i -x' best The maximum absolute value of the difference from the optimal value is found among multiple sample data. Assuming the maximum absolute value of the difference between other sample data and 50 mg / kg is 10 mg / kg, i.e., M = 10, the second calculation formula is used... The value after the positive direction is This makes intermediate indicators conform to the calculation logic that the larger the better after positive conversion.

[0032] The interval-type indicator is positively oriented based on the third calculation formula to generate a larger positively oriented indicator. The third calculation formula is as follows: M = max{a - minx' i ' ,max{x' i '}-b};where, x'i, ‘ new This is a large-scale indicator after positive transformation, where M is the maximum absolute deviation, a and b are the lower and upper limits of the interval indicator, respectively, and x' i ' This represents the original interval-type index value. For the dissolved oxygen content (interval-type index) in a certain sea area of ​​this coastal city, the optimal interval is [5, 7] mg / L. Assume a monitoring value of x' i ' =4mg / L, first calculate M = max{5-minx' i ' ,max{x' i ' If the minimum dissolved oxygen content in this sea area during historical monitoring is 3 mg / L and the maximum is 8 mg / L, then M = max{5-3, 8-7} = 2. According to the third calculation formula, This formula converts interval-type indicators into a unified positive indicator.

[0033] Based on the fourth calculation formula, several large indicators are standardized to generate target indicator features. The fourth calculation formula is as follows: Among them, X norm The target indicator features are defined as follows. After positive transformation, a series of large indicators are obtained. Taking annual available water volume and the positively transformed annual average PM2.5 concentration as examples, let's assume the original data for annual available water volume is x. 11 = 4 billion cubic meters, x 12 = 5 billion cubic meters, x 13 =3.5 billion cubic meters, its minimum value min{x 11 ,x 12 ,x 13 = 3.5 billion cubic meters, maximum value max{x 11 ,x 12 ,x 13} = 5 billion cubic meters. According to the fourth calculation formula, for x... 11 =4 billion cubic meters, the standardized value is Calculate all positiveized large indicators, unify indicators of different dimensions and magnitudes into the [0,1] interval, eliminate dimensional differences, generate target indicator features, and provide a standardized data foundation for subsequent cloud model parameter calculation and environmental carrying capacity evaluation.

[0034] S104 processes the target indicator features to generate cloud model parameter information.

[0035] In one implementation, feature filtering is performed on the target indicator features to generate desired features. These desired features characterize feature values ​​in the target indicator features that exceed a preset threshold. After data standardization and positive conversion, target indicator feature data related to the environmental carrying capacity of the coastal city are obtained, including multiple indicator data such as annual available water volume and positively converted annual average PM2.5 concentration. A preset threshold of 0.6 is set (this threshold can be determined according to actual research needs and data characteristics) to filter the target indicator features. Taking the annual available water volume indicator as an example, its standardized data includes [0.4, 0.7, 0.5, 0.8, 0.65], among which feature values ​​greater than 0.6 are 0.7, 0.8, and 0.65. These values ​​constitute the desired features of the annual available water volume indicator. The desired features reflect the relatively superior portion of the indicator in the data distribution, representing data features that positively contribute to and stand out in environmental carrying capacity assessment.

[0036] The target indicator features and expected features are processed to generate entropy features, which characterize the degree of dispersion of the target indicator features around the expected features. Taking the annual available water volume indicator as an example, the mean Ex of the expected feature (assumed to be calculated as 0.72) is calculated according to the entropy feature formula. Where En is the entropy feature, used to represent the uncertainty of the target indicator feature; avg(X'-Ex) is the average deviation between the target indicator feature X' and the expected feature Ex. For the data point 0.4, (X'-Ex) = (0.4-0.72) = -0.32. A similar calculation is performed for all data points, and the average deviation avg(X'-Ex) is taken. Assuming that after calculation, avg(X'-Ex) = 0.15, then the entropy feature... The larger the entropy value, the greater the dispersion of the target indicator feature around the expected feature, that is, the higher the uncertainty of the data. In this example, the entropy value of 0.13 for the annual available water volume indicator reflects the dispersion of its data around the expected feature. If the entropy value is small, it means that the indicator data is relatively concentrated near the expected feature.

[0037] The target indicator features, expected features, and entropy features are processed to generate hyperentropy features. The formula for calculating hyperentropy features is as follows: Where He is the hyperentropy feature, used to represent the uncertainty of the entropy feature; N is the number of samples, X' iLet be the feature value of the i-th sample. Taking the annual available water volume index as an example, the sample size N = 5 (i.e., the 5 data points [0.4, 0.7, 0.5, 0.8, 0.65] mentioned earlier). According to the formula for calculating the hyperentropy feature, for the data point 0.4, |0.4 + 0.72| + 0.13 = 0.19. The square of each data point is calculated and summed, assuming the result is 0.05. The larger the hyperentropy value, the greater the uncertainty of the entropy feature itself, and the higher the noise level of the data. In this scheme, the hyperentropy value of 0.11 for the annual available water volume index represents the fluctuation of its entropy value. If the hyperentropy value is larger, it means that the stability of the index data is poor, and there may be many interfering factors affecting the assessment of its uncertainty.

[0038] Cloud model parameter information is generated based on expected features, entropy features, and hyperentropy features. In the cloud model, expected features represent typical values ​​or central points of qualitative concepts in the quantitative domain. Mathematically, they are the mean of cloud droplet distribution, equivalent to the x value (ideal center) when membership μ(x) = 1. In the annual available water volume index of this coastal city, the calculated mean of expected features, Ex = 0.72. This value indicates that, from a data characteristic perspective, 0.72 can be considered a representative "central" value for the city's annual available water volume in environmental carrying capacity assessment. If the ideal state of environmental carrying capacity-related indicators is considered a standard, 0.72 represents a reference value for the annual available water volume in the current dataset that is relatively close to the ideal state. For example, when comprehensively assessing the city's environmental carrying capacity, if 1 represents the most ideal annual available water volume state (indicating that it fully meets all aspects of urban development needs and there is no water resource pressure), and 0 represents the worst state, then 0.72 means that the annual available water volume is at a relatively good level, but there is still some room for improvement.

[0039] Entropy has a dual function: qualitatively, it reflects the ambiguity of a concept, i.e., its coverage; quantitatively, it characterizes the standard deviation of a random distribution, determining the dispersion of cloud droplet distribution, i.e., the fluctuation range of membership. In the annual available water volume index, the entropy feature En = 0.13. From the perspective of coverage, it reflects the ambiguity of the annual available water volume concept; the larger the value, the wider the coverage of the annual available water volume, meaning the larger the range of variation of this indicator under different conditions. In terms of dispersion, an entropy value of 0.13 indicates that the annual available water volume data has a moderate degree of dispersion around the expected feature of 0.72. If the entropy value is small, such as close to 0, it means that the data is relatively concentrated near the expected feature, and the change in annual available water volume is relatively stable; if the entropy value is large, it indicates that the data has a large degree of dispersion, and the annual available water volume varies significantly across different samples. In actual environmental carrying capacity assessment, this reflects the fluctuation of annual available water volume under the influence of various factors. An entropy value of 0.13 indicates that although the city's annual available water volume fluctuates, the fluctuation range is within a certain acceptable range.

[0040] The hyperentropy feature is used to characterize the uncertainty of the entropy feature itself, reflecting the degree of cloud droplet aggregation. Intuitively, the larger the hyperentropy, the thicker the cloud layer and the more blurred the edges; when He = 0, the cloud degenerates into a precise Gaussian distribution. The hyperentropy feature He = 0.11 for the city's annual available water volume index, indicating that the uncertainty of the entropy feature is at a certain level. If the hyperentropy value is high, such as greater than 0.2, it means that the entropy value fluctuates greatly, the data noise level is high, and there may be many uncertain factors affecting the assessment of annual available water volume, such as measurement errors and climate anomalies in special years. The current hyperentropy value of 0.11 indicates that although there are some interfering factors in assessing the uncertainty of annual available water volume, the overall entropy estimation is relatively reliable, and the model's stability is acceptable.

[0041] Based on the above-mentioned expected characteristic mean Ex = 0.72, entropy characteristic En = 0.13, and hyperentropy characteristic He = 0.11, these parameters collectively constitute the cloud model parameter information for the annual available water volume index of this coastal city. These parameters are interrelated and can more comprehensively reflect the characteristics of annual available water volume in environmental carrying capacity assessment, providing a scientific basis for subsequent analysis and decision-making. For example, when formulating urban water resource planning, these parameters can be used to understand the stability, range of variation, and degree of uncertainty of annual available water volume, thereby rationally planning water resource development, utilization, and protection strategies.

[0042] S105 performs weight allocation processing on the cloud model parameter information to generate the weight information corresponding to each indicator.

[0043] In one implementation, the cloud model parameter information is processed to generate the standard deviation of each indicator. The standard deviation characterizes the dispersion between indicators, reflecting the degree of dispersion among them. Suppose we have an environmental carrying capacity evaluation indicator system for a coastal city, including the following indicators: annual available water volume (C1), arable land area (C2), annual average PM2.5 concentration (C3), and COD emissions (C4) (this is only a simple example). The standard deviation of each indicator needs to be calculated to characterize the degree of dispersion between them. Assuming that after data cleaning and standardization, the values ​​of each indicator are: annual available water (C1): [0.4, 0.6, 0.5, 0.7, 0.8]; cultivated land area (C2): [0.3, 0.5, 0.4, 0.6, 0.7]; annual average PM2.5 concentration (C3): [0.8, 0.7, 0.6, 0.5, 0.4]; COD emissions (C4): [0.2, 0.3, 0.4, 0.5, 0.6].

[0044] The formula for calculating the standard deviation is: in, Let σ be the mean of the j-th indicator. jLet be the standard deviation of the j-th indicator. The formula for standard deviation is:

[0045] The cloud model parameter information is processed to generate linear correlation coefficients between various indicators. Based on these linear correlation coefficients, a correlation coefficient matrix and indicator conflict information are generated. The correlation coefficient matrix characterizes the correlation between the indicators. The linear correlation coefficients between these indicators are calculated using the cloud model parameter information to characterize their correlation. The formula for calculating the linear correlation coefficient is: Thus, the correlation coefficient matrix is ​​obtained. Specifically, r12 = 0.8 indicates a strong positive correlation between annual available water volume and arable land area, suggesting that these two indicators have similar trends. r13 = -0.6 indicates a negative correlation between annual available water volume and annual average PM2.5 concentration, suggesting that these two indicators have opposite trends. r14 = 0.2 indicates a weak positive correlation between annual available water volume and COD emissions, suggesting that these two indicators have some correlation but it is not strong. The same logic applies to other parameters.

[0046] The correlation coefficient matrix is ​​used to characterize the correlation between indicators. For further analysis, we can characterize the conflict between indicators by taking the negative of the correlation coefficient. The greater the conflict, the higher the redundancy between the indicators, and the relatively lower their importance. This can be achieved through the formula... The conflict information of the indicators is calculated, where δ j This reflects the conflict level of the j-th indicator relative to other indicators; the larger the value, the more severe the conflict. m represents the total number of indicator data points. Assume we calculate the following conflict level information: r1: 0.2 (low conflict), r2: 0.2 (low conflict), r3: 0.4 (moderate conflict), r4: 0.3 (moderate conflict). We process the data standard deviation and conflict level information for each indicator to generate a comprehensive weight information value for each indicator. The comprehensive weight information value is calculated by combining the data standard deviation and conflict level information. The specific formula is as follows: C j =σ j δ j , where C j It is the comprehensive weight information of the j-th indicator, σ j It is the standard deviation of the indicator, δ j This is the conflict score of the indicator. The greater the comprehensive weight information, the higher the importance of the indicator in the evaluation.

[0047] The comprehensive weight information of each indicator is normalized to generate objective weight information for each indicator. To ensure that the sum of the weights of all indicators is 1, these comprehensive weight information quantities need to be normalized to obtain the objective weight information of each indicator. The specific formula is as follows: Where, ω j is the objective weight of the j-th indicator, and m is the total number of indicators.

[0048] The cloud model parameter information is processed to generate a cloud evaluation scale. Specifically, the calculation formula for the cloud evaluation scale is as follows:

[0049]

[0050] The cloud model parameters include the following four indicators: annual available water (C1), arable land area (C2), annual average PM2.5 concentration (C3), and COD emissions (C4). The cloud model uses expectation (Ex), entropy (En), and hyperentropy (He) to reflect the importance of these indicators. Assuming the cloud model parameters for each indicator have been calculated, a cloud evaluation scale is generated as follows:

[0051] index Ex En He C1 0.65 0.1 0.05 C2 0.7 0.08 0.03 C3 0.3 0.2 0.1 C4 0.5 0.15 0.08

[0052] A comparison matrix of evaluation indicators for cloud models and cloud evaluation scales is constructed. The golden ratio method is used to allocate parameters of the cloud model, generating a cloud parameter table with a 9-level scale for cloud importance, as follows:

[0053]

[0054]

[0055] After providing the cloud parameters for the importance scale, construct a comparison matrix of evaluation indicators between the cloud model and the cloud evaluation scale:

[0056]

[0057] Where n is the number of indicators to be evaluated, the expected value of the diagonal elements is Ex = 1, and En and He are both 0. The two indicators are compared pairwise using the formula... Obtain the specific values ​​of each indicator and generate an evaluation indicator comparison matrix.

[0058] The following is an example; the evaluation index comparison matrix is ​​as follows:

[0059]

[0060] The evaluation index comparison matrix of the cloud model and cloud evaluation scale is processed to generate the relative weight values ​​of each index. The square root method is used to calculate the relative weight value of each index, as follows: the relative weight value W' of index i. i (Ex' i ,En' i He' i The relative weight value of index j

[0061] Where, λ max The largest eigenvalue of the matrix is ​​used to calculate the relative weights of the indicators. Ex i ' is the expected value of the normalized index i, H e 'En is the entropy of the normalized index i. i 'Ex is the hyperentropy of the normalized index i. i Let En be the expected value of index i. i H is the entropy of index i, used to characterize the uncertainty of the index. e Let Ex be the hyperentropy of index i, used to represent the uncertainty of entropy. i,j Let En be the entropy of indicator i in the evaluation scale, representing the uncertainty or fuzziness of indicator i. The greater the entropy, the higher the uncertainty of the indicator. i,j is the expected value of indicator i in the evaluation scale, representing the typical value or central value of indicator i, which reflects the main characteristics of the indicator.

[0062] The following relative weight values ​​were obtained through calculation: annual available water (C1): 0.2, arable land area (C2): 0.35, annual average PM2.5 concentration (C3): 0.20, and COD emissions (C4): 0.25. These relative weight values ​​reflect the importance distribution of each indicator in the environmental carrying capacity assessment.

[0063] The objective weight information and relative weight values ​​of each indicator are normalized to generate the corresponding weight information for each indicator. The objective weight of each indicator is calculated using the CRITIC method, and then combined with the relative weight of expert scores using a subjective and objective preference coefficient α.

[0064] Assume the objective weights calculated by the CRITIC method are:

[0065] index Objective weight C1 0.25 C2 0.30 C3 0.15 C4 0.30

[0066] Assume the relative weights of the expert ratings are:

[0067] index relative weight C1 0.20 C2 0.35 C3 0.20 C4 0.25

[0068] The fusion was performed using a subjective and objective preference coefficient α = 0.5, and the result was obtained through w = αw. s +(1-α)w o Calculate the overall weight, where w is the overall weight of the indicator; w s The weights calculated using the expert scoring method for the cloud model refer to the relative weights; w o The weights calculated by the THECRITIC method are objective weights; α is the subjective and objective preference coefficient, which is set to 0.5 in this invention.

[0069] ω C 1=0.5×0.25+0.5×0.20=0.225;ω C 2 = 0.5 × 0.30 + 0.5 × 0.35 = 0.325;

[0070] ω C 3=0.5×0.15+0.5×0.20=0.175;ω C 4 = 0.5 × 0.30 + 0.5 × 0.25 = 0.275.

[0071] index Overall weight C1 0.23 C2 0.31 C3 0.17 C4 0.28

[0072] S106, based on the weight information corresponding to each indicator, the target indicator features are processed to generate the average membership degree of environmental carrying capacity, the fluctuation range value of membership degree, and the noise level value of membership degree.

[0073] In one implementation, the criterion-level indicators in the target indicator features are processed based on the weight information corresponding to the criterion-level indicators to generate membership information for the criterion-level. Assume we have an environmental carrying capacity evaluation index system for a coastal city, including four criterion levels: natural resources (B1), environmental quality (B2), ecological resilience (B3), and socio-economic factors (B4). Each criterion level has multiple specific indicators, for example: Natural Resources (B1): Annual available water volume (C1), arable land area (C2).

[0074] Environmental quality (B2): Annual average PM2.5 concentration (C3), COD emissions (C4)

[0075] Ecological resilience (B3): Biodiversity (C5), Vegetation cover (C6)

[0076] Socioeconomic factors (B4): Population density (C7), Industrial output (C8)

[0077] Assume we have already calculated the weight information for each indicator using the cloud model and the CRITIC-expert fusion weighting method. For example, the weight of annual available water (C1) is 0.2, the weight of arable land area (C2) is 0.1, the weight of annual average PM2.5 concentration (C3) is 0.15, the weight of COD emissions (C4) is 0.1, the weight of biodiversity (C5) is 0.1, the weight of vegetation cover (C6) is 0.1, the weight of population density (C7) is 0.15, and the weight of industrial output (C8) is 0.2.

[0078] For each criterion layer, we calculate the membership information of the criterion layer based on the weight information of its subordinate indicators. For example, for natural resources (B1):

[0079] Based on the calculation formula of the membership degree value of the criterion layer, it can be seen that... For criterion layer B k membership degree, w j For indicator layer C j The weight, μ Cj For indicator layer C j The membership degree is denoted by n, where n is the number of index layers and m is the number of criterion layers. in, and These are the membership values ​​for annual available water volume and cultivated land area, respectively.

[0080] The membership information of the criterion layer is processed based on the weight information corresponding to the target layer indicators to generate the membership information of the target layer. Assume that weights have already been assigned to each criterion layer, for example:

[0081] Natural Resources (B1): 0.2; Environmental Quality (B2): 0.2;

[0082] Ecological resilience (B3): 0.3; Socioeconomic resilience (B4): 0.3;

[0083] The formula for calculating the membership value of the target layer is: Where, μ A The membership degree of the target layer. The weights for the criteria layer.

[0084] Based on these weights, we can calculate the membership information of the target layer (regional environmental carrying capacity):

[0085]

[0086] The membership information of the target layer is processed to generate the average membership degree of the environmental carrying capacity. Assume we obtain multiple membership values ​​of the target layer through Monte Carlo simulation. Where N = 1000. We calculate the average membership degree: Where Ex represents the average membership degree of environmental carrying capacity, and N represents the number of samples for the membership degree information of the target layer. Let be the target layer membership degree of the i-th sample. Assume the average value of the simulation results is 0.65.

[0087] The membership information of the target layer and the average membership of the environmental carrying capacity are processed to generate the fluctuation range value of the membership. First, the mean absolute deviation between the membership and the average membership is calculated. in, Let Ex be the average absolute deviation of the membership degree from the mean, and let Ex be the average membership degree of the environmental carrying capacity. The calculated mean deviation is 0.12. Then, the fluctuation range of the membership degree is calculated. Where En is the fluctuation range of the membership degree.

[0088] The fluctuation range of membership values ​​is processed to generate noise level values ​​for membership. First, the variance of membership values ​​deviating from the mean is calculated. Assuming the calculated variance is 0.03, calculate the noise level value of the membership degree.

[0089] S107. Based on the target environmental carrying capacity early warning model, the average membership degree, the fluctuation range of the membership degree, and the noise level of the membership degree are processed to generate environmental carrying capacity early warning information for the area to be evaluated.

[0090] In one implementation, threshold values ​​for average membership degree, fluctuation range of membership degree, and noise level of membership degree are constructed for different levels of carrying capacity. In environmental carrying capacity assessment, carrying capacity is divided into several levels, such as "low," "medium," "high," and "extremely high." Each level corresponds to different threshold values ​​for average membership degree, fluctuation range of membership degree, and noise level. Assume that based on historical data and expert opinions, we have determined the following threshold values:

[0091] Low load capacity: Average membership threshold: ≤0.4; Fluctuation range threshold: ≥0.2; Noise level threshold: ≥0.1.

[0092] Medium bearing capacity: Average membership threshold: 0.4 < ≤ 0.6; Fluctuation range threshold: 0.1 < ≤ 0.2; Noise level threshold: 0.05 < ≤ 0.1.

[0093] High load capacity: Average membership threshold: 0.6 < ≤ 0.8; Fluctuation range threshold: < 0.1; Noise level threshold: < 0.05.

[0094] Extremely high load-bearing capacity: Average membership threshold: >0.8; Fluctuation range threshold: <0.05; Noise level threshold: <0.02.

[0095] Based on the target environmental carrying capacity early warning model, the average membership threshold, the membership fluctuation range threshold, and the membership noise level threshold corresponding to different carrying capacity levels, the average membership, membership fluctuation range, and membership noise level values ​​of the environmental carrying capacity are processed to generate the early warning level of the environmental carrying capacity of the area to be assessed. Assuming we have environmental carrying capacity data for the area to be assessed, the following values ​​are obtained after calculation:

[0096] Average membership: 0.65; fluctuation range: 0.15; noise level: 0.087.

[0097] Based on the above thresholds, we can determine the bearing capacity level of this area:

[0098] The average membership degree of 0.65 falls within the "high load-bearing capacity" range.

[0099] The fluctuation range of 0.15 exceeds the fluctuation range threshold of "high load capacity".

[0100] The noise level of 0.087 exceeds the threshold for "extremely high load capacity".

[0101] Based on the above information, we can conclude that the carrying capacity of this area is at the "high carrying capacity" level, but it fluctuates greatly and the noise level is high.

[0102] The environmental carrying capacity warning level of the area to be assessed is processed to generate environmental carrying capacity warning information for the area. Based on the calculation results and threshold judgment, we can generate the following warning information: Carrying capacity level: High carrying capacity. Warning information: Although the current carrying capacity is at a high level, the fluctuation range is large (0.12), and the noise level is also high (0.08). It is recommended to strengthen monitoring and data analysis to ensure the stability and reliability of the data. At the same time, due to the large fluctuation, it is recommended to exercise caution in planning and decision-making. In this way, we can combine the average membership degree, fluctuation range, and noise level value to generate detailed environmental carrying capacity warning information, helping decision-makers to better understand and respond to changes in environmental carrying capacity.

[0103] The server first acquires environmental carrying capacity information and target early warning models for the area to be evaluated. Then, the environmental carrying capacity information undergoes data cleaning, classifying it into normally distributed and non-normally distributed data. These are processed separately to filter out outliers and impute missing values, resulting in cleaned data. Next, the cleaned data undergoes positive transformation and standardization, generating different types of indicators based on their properties and converting them into a unified, comparable form. This yields the target indicator characteristics, laying the foundation for subsequent analysis.

[0104] Based on the target indicator features, cloud model parameter information is calculated. Expected features are obtained by filtering feature values, and entropy features are calculated by combining these with the target indicator features. Finally, the three are integrated to obtain hyper-entropy features. Simultaneously, the cloud model parameter information is processed to generate data standard deviation, correlation coefficient matrix, etc., to determine the comprehensive weight information of each indicator and normalize it, obtaining objective weight information. Furthermore, a cloud evaluation scale and comparison matrix are constructed to calculate relative weight values. Finally, the subjective and objective weights are integrated to determine the weight of each indicator. Using the weights of each indicator, the target indicator features are processed to obtain the average membership degree, fluctuation range value, and noise level value of environmental carrying capacity. These values ​​are input into the target environmental carrying capacity early warning model and compared with preset thresholds to determine the early warning level of the environmental carrying capacity of the area to be evaluated, generating corresponding early warning information to assist in relevant decision-making, thereby achieving an accurate grasp and reasonable response to the environmental carrying capacity status.

[0105] In one implementation, such as Figure 2 As shown, this application also provides an environmental carrying capacity analysis and early warning device, comprising:

[0106] Module 201 is used to acquire environmental carrying capacity information of the area to be evaluated and the target environmental carrying capacity early warning model.

[0107] The processing module 202 is used to perform data cleaning on the environmental carrying capacity information of the area to be evaluated, generating cleaned environmental carrying capacity information; perform data forwarding and standardization on the cleaned environmental carrying capacity information to generate target indicator features; process the target indicator features to generate cloud model parameter information; perform weight allocation on the cloud model parameter information to generate weight information corresponding to each indicator; process the target indicator features based on the weight information corresponding to each indicator to generate the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree; and process the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree based on the target environmental carrying capacity early warning model to generate environmental carrying capacity early warning information for the area to be evaluated.

[0108] The various embodiments in this application are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of the analysis and early warning method for assessing environmental carrying capacity, the electronic device, the electronic device, and the readable storage medium are basically similar to the environmental carrying capacity analysis and early warning method embodiments described above, so the descriptions are relatively simple. Relevant parts can be referred to in the descriptions of the environmental carrying capacity analysis and early warning method embodiments described above.

Claims

1. A method for analyzing and providing early warning of environmental carrying capacity, characterized in that, include: The process involves acquiring environmental carrying capacity information for the area to be assessed, as well as a target environmental carrying capacity early warning model. This environmental carrying capacity information includes: water resource data, land resource data, and mineral resource data. Water resource data includes annual available freshwater volume; land resource data includes arable land area and building land area; and mineral resource data includes mineral reserves and extraction volume. Other data include air pollution indicators, water quality indicators, soil pollution indicators, biodiversity data, vegetation coverage, and ecological resilience data; population density data, industrial output data, and environmental protection input data. The environmental carrying capacity information of the area to be assessed is cleaned to generate cleaned environmental carrying capacity information. After data cleaning, environmental carrying capacity information is processed for data forwarding and standardization to generate target indicator features; The target indicator features are processed to generate cloud model parameter information; The cloud model parameter information is weighted to generate weight information for each indicator. This includes processing the cloud model parameter information to generate the standard deviation of each indicator, which characterizes the difference between indicators and reflects their dispersion; processing the cloud model parameter information to generate linear correlation coefficients between indicators, and generating a correlation coefficient matrix and indicator conflict information based on these coefficients, where the correlation coefficient matrix characterizes the correlation between indicators; processing the standard deviation and conflict information of each indicator to generate a comprehensive weight information for each indicator; normalizing the comprehensive weight information for each indicator to generate objective weight information for each indicator; processing the cloud model parameter information to generate a cloud evaluation scale; constructing a comparison matrix of evaluation indicators between the cloud model and the cloud evaluation scale; processing each indicator in the comparison matrix to generate a relative weight value for each indicator; and normalizing the objective weight information and relative weight values ​​of each indicator to generate the corresponding weight information for each indicator. Based on the weight information corresponding to each indicator, the characteristics of the target indicator are processed to generate the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree of the environmental carrying capacity. Based on the target environmental carrying capacity early warning model, the average membership degree, the fluctuation range of membership degree, and the noise level of membership degree are processed to generate environmental carrying capacity early warning information for the area to be evaluated.

2. The method as described in claim 1, characterized in that, The environmental carrying capacity information of the area to be assessed is cleaned to generate cleaned environmental carrying capacity information, including: The environmental carrying capacity information of the area to be assessed is classified and processed to generate normally distributed data and non-normally distributed data; The normally distributed data is processed to generate the mean and standard deviation of each indicator, and the first outlier is generated based on the mean and standard deviation of each indicator. The box plot method is used to process non-normally distributed data and generate a second outlier. Based on the first and second outliers, outlier screening is performed on the environmental carrying capacity information of the area to be evaluated to generate noise-filtered environmental carrying capacity information. Missing values ​​are imputed in the noise-filtered environmental carrying capacity information to generate cleaned environmental carrying capacity information.

3. The method as described in claim 1, characterized in that, The cleaned environmental carrying capacity information is then processed for data forwarding and standardization to generate target indicator features, including: The environmental carrying capacity information of the area to be assessed is processed to generate target layer features, criterion layer features, and indicator layer features. The target layer features, criterion layer features, and indicator layer features are processed to generate large indicators, small indicators, intermediate indicators, and interval indicators. The small indicators are positiveized based on the first calculation formula to generate the large indicators after positiveization. The intermediate indicators are positiveized based on the second calculation formula to generate the positiveized large indicators. The interval-type indicators are positively processed based on the third calculation formula to generate a large positively processed indicator. Based on the fourth calculation formula, several large indicators are standardized to generate target indicator features. The first calculation formula is: ;in, For large indicators after positive transformation, These are the original, small-scale indicator values. This is the maximum value of all original small indicators; The second calculation formula is: ; ;in, For large indicators after positive transformation, M is the maximum absolute deviation. These are the original intermediate index values. The target value for intermediate indicators; The third calculation formula is: ; ； in, This is a large-scale indicator after positive transformation, where M is the maximum absolute deviation, and a and b are the lower and upper limits of the interval indicator, respectively. These are the original interval-type index values; The fourth calculation formula is: ; in, The target indicator features.

4. The method as described in claim 3, characterized in that, The target indicator features are processed to generate cloud model parameter information, including: The target indicator features are subjected to feature filtering to generate desired features, where the desired features are used to characterize the feature values ​​of the target indicator features that are greater than a preset threshold. The target indicator features and expected features are processed to generate entropy features, where entropy features are used to characterize the degree of dispersion of the target indicator features around the expected features; The target indicator features, expected features, and entropy features are processed to generate hyperentropy features, which are used to characterize the uncertainty of the entropy features. Generate cloud model parameter information based on expected features, entropy features, and hyperentropy features; The formula for calculating entropy features is as follows: ; in, Entropy features are used to represent the uncertainty of target indicator features; For target indicator characteristics With expectations The average deviation; The formula for calculating hyperentropy features is as follows: ; in, This represents the hyperentropy feature, used to indicate the uncertainty of the entropy feature; N is the number of samples. Let be the feature value of the i-th sample.

5. The method as described in claim 1, characterized in that, Based on the weight information corresponding to each indicator, the characteristics of the target indicator are processed to generate the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree for environmental carrying capacity, including: Based on the weight information corresponding to the criteria layer indicators, the criteria layer indicators in the target indicator features are processed to generate the membership information of the criteria layer. The membership information of the criterion layer is processed based on the weight information corresponding to the target layer indicators to generate the membership information of the target layer. The membership information of the target layer is processed to generate the average membership of environmental carrying capacity; The membership information of the target layer and the average membership of the environmental carrying capacity are processed to generate the fluctuation range value of the membership. The fluctuation range of membership values ​​is processed to generate noise level values ​​for membership.

6. The method as described in claim 1, characterized in that, Based on the target environmental carrying capacity early warning model, the average membership degree, the fluctuation range of the membership degree, and the noise level of the membership degree are processed to generate environmental carrying capacity early warning information for the area to be assessed, including: Construct the average membership threshold, the membership fluctuation range threshold, and the membership noise level threshold corresponding to different levels of bearing capacity; Based on the target environmental carrying capacity early warning model, the average membership threshold, the membership fluctuation range threshold, and the membership noise level threshold corresponding to different levels of carrying capacity, the average membership, the membership fluctuation range, and the membership noise level are processed to generate the early warning level of the environmental carrying capacity of the area to be evaluated. The warning level of the environmental carrying capacity of the area to be assessed is processed to generate environmental carrying capacity warning information for the area to be assessed.

7. An environmental carrying capacity analysis and early warning device, used to implement the method of claim 1, characterized in that, The device includes: The acquisition module is used to acquire environmental carrying capacity information of the area to be assessed and the target environmental carrying capacity early warning model; The processing module is used to perform data cleaning on the environmental carrying capacity information of the area to be assessed, generating cleaned environmental carrying capacity information; perform data forwarding and standardization on the cleaned environmental carrying capacity information to generate target indicator features; process the target indicator features to generate cloud model parameter information; perform weight allocation on the cloud model parameter information to generate weight information corresponding to each indicator; process the target indicator features based on the weight information corresponding to each indicator to generate the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree; and process the average membership degree, the fluctuation range value of the membership degree, and the noise level value of the membership degree based on the target environmental carrying capacity early warning model to generate environmental carrying capacity early warning information for the area to be assessed.

8. An electronic device, characterized in that, include: First processor; and memory for storing executable instructions of the first processor; The first processor is configured to execute the environmental carrying capacity analysis and early warning method according to any one of claims 1 to 6 by executing the executable instructions.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the second processor, it implements the environmental carrying capacity analysis and early warning method according to any one of claims 1 to 6.

Images