Time series-based resource utilization development degree comprehensive evaluation system

Abstract

The application discloses a time series-based resource utilization development degree comprehensive evaluation system, relates to the technical field of resource evaluation, and comprises the following modules: an index system construction module, which is used for constructing a three-level evaluation index system comprising a target layer, a criterion layer and an index layer; a coupling coordination degree analysis module, which is used for standardizing index data, calculating annual comprehensive evaluation indexes of each criterion layer, annual resource scores, annual comprehensive scores and coupling coordination degree indexes; a time series analysis module, which is used for decomposing time series of the annual comprehensive evaluation indexes of each criterion layer, separating short-term fluctuations and long-term development trends, and predicting the time series by means of a grey prediction model combined with a trend correction method; and a comprehensive evaluation module, which is used for combining data of historical years and prediction years, performing comprehensive evaluation from three dimensions of ecological protection effectiveness, resource utilization efficiency and resource coordination level, determining a development stage of resource utilization, and providing a scientific basis for forward-looking decision-making of regional resource management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of resource evaluation, in particular to a resource utilization development degree comprehensive evaluation system based on time series. BACKGROUND

[0002] Resource utilization development degree evaluation is an important research content in the field of resource science, and its core is to quantitatively evaluate the comprehensive performance of resource development, utilization and management through a systematic method. In the prior art, resource utilization evaluation mainly relies on traditional statistical index system and single resource dimension analysis. For example, through Malmquist index method, data envelopment analysis (DEA) and benefit analysis method, the resource input-output efficiency is calculated statically. Such methods are based on historical data, construct linear or nonlinear models, and combine resource quality, quantity and spatial distribution characteristics to form a phased judgment of the current situation of resources.

[0003] However, traditional methods such as Malmquist index method and DEA usually focus on the static efficiency calculation of independent resource types such as water resources or forest resources, and fail to systematically integrate the synergistic relationship of mountain, water, forest, grass, field and human resources, resulting in that the evaluation results are difficult to fully reflect the overall coordination of regional resource utilization. In addition, the static model is constructed relying on historical data, which lacks the ability to capture the dynamic evolution trend of resources, cannot distinguish short-term fluctuations from long-term trends, including resource degradation or recovery process, leading to that the evaluation conclusion lags behind the actual changes.

[0004] In addition, the traditional method does not introduce a time series prediction model such as gray prediction or trend decomposition, which cannot predict the future state of resource utilization, is difficult to support forward-looking decision-making, ignores the coupling and coordination mechanism between resource subsystems, such as the balance between ecological protection and economic development, and only evaluates through a single score (resource efficiency), without establishing a comprehensive framework covering ecological protection effectiveness, resource utilization efficiency and coordination level, resulting in rough development stage division and inability to accurately identify ecological imbalance period or benign cycle period. SUMMARY

[0005] (I) Technical problems solved

[0006] In view of the deficiencies of the prior art, the present application provides a resource utilization development degree comprehensive evaluation system based on time series, which realizes accurate capture of the dynamic trend of regional resource utilization and reliable prediction of the future state by constructing a multi-level evaluation index system, combining coupling coordination degree analysis and gray prediction model, and comprehensively evaluating the development degree of resource utilization from three dimensions of ecological protection, resource efficiency and coordination level through a comprehensive evaluation module. The development stage is accurately divided, and scientific basis is provided for forward-looking decision-making of regional resource management.

[0007] (II) Technical solutions

[0008] To achieve the above object, the application is implemented by the following technical solutions: a time series-based resource utilization development degree comprehensive evaluation system, comprising:

[0009] An index system construction module is configured to construct a three-level evaluation index system including a target layer, a criterion layer, and an index layer, the target layer being a regional common evaluation index system of mountains, water, forests, fields, and people, the criterion layer including six dimensions of mountains, water, forests, fields, and people, each criterion layer being provided with an index layer, and the evaluation direction and weight of each index being determined;

[0010] A coupling coordination degree analysis module is configured to standardize the index data, calculate the annual comprehensive evaluation index, annual resource score, annual comprehensive score, and coupling coordination degree index of each criterion layer;

[0011] A time series analysis module is configured to decompose the time series of the annual comprehensive evaluation index of each criterion layer, separate short-term fluctuations and long-term development trends, and predict the trends through a grey prediction model combined with a trend correction method;

[0012] A comprehensive evaluation module is configured to combine the data of historical years and predicted years to conduct comprehensive evaluation from three dimensions of ecological protection effectiveness, resource utilization efficiency, and resource coordination level, and determine the development stage of resource utilization.

[0013] Further, a three-level evaluation index system including a target layer, a criterion layer, and an index layer is constructed for the evaluation object of the mountains, water, forests, fields, and people life community;

[0014] The criterion layer includes six dimensions of mountains, water, forests, fields, and people, and each criterion layer is provided with an index layer;

[0015] The evaluation direction of each index is determined, and an index reflecting positive development of resources is set as a positive index, and an index having a negative impact on resource development is set as a negative index;

[0016] The analytic hierarchy process is used to determine the weight of each criterion layer, the entropy weight method is used to calculate the element weight of each index in the same criterion layer, and the system weight of each index is calculated through the formula: system weight = criterion layer weight × element weight.

[0017] Further, the indexes of the mountain criterion layer include bare land area;

[0018] The indexes of the water criterion layer include precipitation, runoff, evaporation, water area, and wetland area;

[0019] The indexes of the forest criterion layer include forestry output value and forest area;

[0020] The indexes of the grass criterion layer include grassland area, comprehensive coverage, animal husbandry output value, and livestock quantity;

[0021] The indicators of the farmland criterion layer include cultivated land area, agricultural output value, and grain yield;

[0022] The indicators of the human criterion layer include population, per capita GDP, area, and GDP.

[0023] Further, the original index data is standardized to map the data to the interval [0, 1];

[0024] Based on the standardized data and system weight, the annual comprehensive evaluation index of each criterion layer is calculated by a weighted summation formula;

[0025] The annual comprehensive score is obtained by summing the annual comprehensive evaluation indexes of all criterion layers;

[0026] The annual resource score is obtained by summing the annual comprehensive evaluation indexes of the mountain, water, forest, grass, and farmland criterion layers;

[0027] The coupling degree index is obtained by the annual comprehensive evaluation indexes of each criterion layer;

[0028] Based on the annual comprehensive evaluation indexes of each criterion layer and the criterion layer weight, the coordination degree index is calculated by a weighted summation formula;

[0029] The coupling coordination degree index is obtained by multiplying the coupling degree index and the coordination degree index and taking the square root.

[0030] Further, the time series of the annual comprehensive evaluation index of each criterion layer is decomposed by ensemble empirical mode decomposition to obtain intrinsic mode functions and trend items of different periods;

[0031] The trend items obtained by decomposition are subjected to linear regression analysis to test the significance of their change trend;

[0032] The time series of the annual comprehensive evaluation index of each criterion layer is predicted by a grey prediction model to obtain a grey prediction value;

[0033] The trend item obtained by ensemble empirical mode decomposition is extracted and extended to the prediction period by linear extrapolation;

[0034] The fitting error of the grey prediction value and the trend item is calculated, and the weight coefficient is optimized by the least squares method to obtain a corrected prediction value, which is a weighted combination of the grey prediction value and the extended trend item;

[0035] The reference region prediction value is introduced, the cosine similarity of the annual comprehensive evaluation index of each criterion layer with the reference region is calculated, the difference between the corrected prediction value and the reference region prediction value is calculated, and the corrected prediction value is obtained by multiplying the cosine similarity and adding the corrected prediction value.

[0036] Further, the time series of the annual comprehensive evaluation indexes of each criterion layer is decomposed by using ensemble empirical mode decomposition, including decomposing a single noisy data sequence by using empirical mode decomposition, and the specific steps are as follows:

[0037] Step 2: The maximum value points are connected by using a cubic spline interpolation method to form an upper envelope line, and the minimum value points are connected to form a lower envelope line;

[0038] Step 3: The mean values of the upper envelope line and the lower envelope line are calculated to obtain a mean envelope line, and a candidate intrinsic mode function component is obtained by subtracting the mean envelope line from the noisy data sequence;

[0039] Step 4: The candidate intrinsic mode function component is checked to see whether it meets the requirements of the intrinsic mode function:

[0040] In the entire sequence, the number of extreme points is equal to or differs from the number of zero-crossing points by at most 1;

[0041] At any time, the mean values of the upper envelope line and the lower envelope line are 0;

[0042] If not, the candidate intrinsic mode function component is replaced by the original noisy data sequence, and steps 2 and 3 are re-executed until a required intrinsic mode function component is obtained as a first intrinsic mode function component;

[0043] Step 5: A residual sequence is obtained by subtracting the first intrinsic mode function component from the noisy data sequence;

[0044] Step 6: Steps 2 to 5 are repeatedly executed on the residual sequence to extract subsequent intrinsic mode function components one by one until the residual sequence becomes a monotonically increasing or monotonically decreasing curve or a fixed value, and the decomposition is stopped. At this time, a plurality of IMF components and a last residual trend item are obtained.

[0045] Further, the annual comprehensive evaluation indexes, the annual resource scores, the annual comprehensive scores and the coupling coordination degree indexes of each criterion layer in the historical years and the predicted years are obtained respectively;

[0046] A matrix is constructed, with the historical years and the predicted years as row dimensions, and the ecological protection effectiveness, the resource utilization efficiency and the resource coordination level as column dimensions. The ecological protection effectiveness is composed of the weighted sum of the mountain, forest and grass criterion layer annual comprehensive evaluation indexes, the resource utilization efficiency is composed of the ratio of the annual resource score to the annual comprehensive score, and the coordination level is the coupling coordination degree index;

[0047] Linear normalization is used to map the matrix elements to the [0, 1] interval;

[0048] The CRITIC method is used to determine the column dimension weight;

[0049] The improved TOPSIS method is used to calculate the comprehensive score;

[0050] Based on the comprehensive score, the development stage of resource utilization is divided.

[0051] Further, the improved TOPSIS method is used to calculate the comprehensive score, including: determining the positive ideal solution and the negative ideal solution;

[0052] The weighted Euclidean distance of each year from the positive ideal solution and the negative ideal solution is calculated;

[0053] The comprehensive score is obtained by the weighted Euclidean distance of each year from the positive ideal solution and the negative ideal solution, and the comprehensive score is the weighted Euclidean distance of the year from the negative ideal solution divided by the sum of the weighted Euclidean distance of the year from the positive ideal solution and the weighted Euclidean distance of the year from the negative ideal solution;

[0054] Based on the comprehensive score, the development stage of resource utilization is divided, including: when the comprehensive score of any year is <M1, the development stage of the year is divided into an ecological imbalance period;

[0055] When the comprehensive score of any year is M1≤ the comprehensive score <M2, the development stage of the year is divided into a preliminary coordination period;

[0056] When the comprehensive score of any year is >M2, the development stage of the year is divided into a benign cycle period.

[0057] An electronic device includes a memory, a processor, and a computer program stored on the memory and running on the processor, and the processor implements the steps performed by the system of any of the above when executing the computer program.

[0058] A computer readable storage medium, the computer readable storage medium stores a computer program, and the computer program is executed to implement the steps performed by the system of any of the above.

[0059] (Three) beneficial effects

[0060] The present application provides a time series-based comprehensive evaluation system for the development degree of resource utilization, which has the following beneficial effects:

[0061] (1) By constructing a multi-level and multi-dimensional evaluation index system, covering six dimensions of mountain, water, forest, grass, field and people, the evaluation direction and weight of each index are scientifically determined, the overall coverage and system quantification of the regional resource utilization condition are realized, the scientific basis for subsequent analysis is provided, the comprehensiveness and accuracy of the evaluation results are ensured, the dynamic trend capturing and comprehensive evaluation are supported, and the foresight and decision-making scientificity of resource management are improved.

[0062] (2) Through the standardized processing of multi-source heterogeneous data, the annual comprehensive evaluation index, coupling degree and coordination index are calculated, and the synergistic level among the six resource dimensions of mountain, water, forest, grass, field and human is quantitatively revealed, which helps to accurately identify the coordination and short board of resource utilization and dynamically monitor the balanced state of the subsystem.

[0063] (3) Through ensemble empirical mode decomposition (EEMD) to separate the long-term trend and short-term fluctuation of resource data, and fuse gray prediction and trend correction technology to predict the future state, accurately capture the dynamic law of resource evolution, such as degradation or recovery trend, enhance the robustness and credibility of prediction, and optimize the prediction results through the reference regional calibration mechanism, provide forward-looking decision basis for resource management, and significantly improve the support ability of the system for long-term development strategy.

[0064] (4) Through the integration of historical and predicted data, a three-dimensional evaluation matrix is constructed, including ecological protection effect, resource utilization efficiency and coordination level, and the improved TOPSIS method is used to calculate the comprehensive score, taking ecological protection, efficiency and coordination level as the core, breaking through the limitation of single dimension evaluation, and accurately dividing the resource development stage based on the comprehensive score. BRIEF DESCRIPTION OF DRAWINGS

[0065] Fig. 1 It is a structure schematic diagram of the resource utilization development degree comprehensive evaluation system based on time series of the present application.

[0066] Fig. 2 It is a mountain-water-forest-field-human-grass life community comprehensive evaluation index schematic diagram of the present application. DETAILED DESCRIPTION

[0067] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0068] Please refer to Figs. 1-2The application provides a time series-based resource utilization development degree comprehensive evaluation system, a mountain-water-forest-river-grass-life community index evaluation system is preliminarily explored according to the natural resource investigation results of XX county, and coupling coordination degree analysis is performed, the project explores the connotation of the life community from the perspective of "mountain-water-forest-river-grass" plus "human", and constructs an evaluation analysis framework based on comprehensive evaluation, coordination degree and coupling degree, taking XX county as an example, an index system is constructed from six aspects of mountain, water, forest, field, grass and human with typical advantages and characteristics in the region, and by means of a comprehensive evaluation model and a coupling coordination degree model, the coupling coordination of the mountain-water-forest-river-grass-life community in the region and the time series change are explored, and the regularity and main influencing factors are revealed, so as to provide reference for regional natural resource control, including:

[0069] The index system construction module is used for constructing a three-level evaluation index system including a target layer, a criterion layer and an index layer, the target layer is a regional mountain-water-forest-river-grass common evaluation index system, the criterion layer includes six dimensions of mountain, water, forest, grass, field and human, each criterion layer is provided with an index layer, and the evaluation direction and weight of each index are determined;

[0070] Specifically, the mountain-water-forest-river-grass life community is taken as the evaluation object, a three-level evaluation index system including a target layer, a criterion layer and an index layer is constructed, wherein the target layer is a regional mountain-water-forest-river-grass common evaluation index system, the criterion layer includes six criterion layers of mountain, water, forest, grass, field and human, each criterion layer is provided with a corresponding index layer, and the index layer includes a plurality of indexes, wherein the indexes corresponding to the mountain criterion layer include bare land area, the indexes corresponding to the water criterion layer include precipitation, runoff, evaporation, water area and wetland area, the indexes corresponding to the forest criterion layer include forestry output value and forest area, the indexes corresponding to the grass criterion layer include grassland area, comprehensive coverage, animal husbandry output value and livestock quantity, the indexes corresponding to the field criterion layer include cultivated land area, agricultural output value and grain yield, and the indexes corresponding to the human criterion layer include population, per capita GDP, area and GDP, the evaluation direction of each index is determined, and for indexes reflecting positive development of resources (such as forest area and agricultural output value), positive indexes are set, and for indexes having negative influence on resource development (such as bare land area), negative indexes are set;

[0071] According to the constructed three-level evaluation index system, the index data of consecutive years are collected, the data sources include statistical yearbook, remote sensing image interpretation results and the like, for example, the data of cultivated land area and agricultural output value can be referred to XX County Statistical Yearbook 2023, the rainfall data adopts the annual precipitation records of XX county, and the forest area data is obtained through remote sensing image interpretation;

[0072] The analytic hierarchy process is used to determine the weight of each criterion layer, and the entropy weight method is used to calculate the internal weight of each index in the same criterion layer. The system weight of each index is the product of the criterion layer weight and the internal weight, and the formula is: system weight = criterion layer weight x internal weight. For example, in the "people" criterion layer of XX County, the system weight of urban area (p3) = 0.1 (criterion layer weight) x 0.683 (internal weight) = 0.0683, as shown in Table 1.

[0073] Table 1: Common evaluation index system of mountain, water, forest, grass, field and people in XX County:

[0074]

[0075] By constructing a multi-level and multi-dimensional evaluation index system covering six dimensions of mountain, water, forest, grass, field and people, scientifically determining the evaluation direction and weight of each index, realizing the comprehensive coverage and system quantification of regional resource utilization, providing a scientific basis for subsequent analysis, ensuring the comprehensiveness and accuracy of the evaluation results, supporting dynamic trend capture and comprehensive evaluation, and thus improving the foresight and decision-making scientificity of resource management.

[0076] The coupling coordination degree analysis module is used for standardizing the index data and calculating the annual comprehensive evaluation index, annual resource score, annual comprehensive score and coupling coordination degree index of each criterion layer;

[0077] Specifically, the collected index data of different dimensions is standardized by mapping the data to the interval [0, 1], and the formula is: wherein, is the original data of the i-th criterion layer and the k-th index, is the standardized data;

[0078] Based on the standardized index data and the corresponding system weight, the annual comprehensive evaluation index of each criterion layer is calculated by the weighted sum formula, and the formula is as follows: wherein, represents the annual comprehensive evaluation index of the i-th criterion layer in the t-th year, K represents the number of indexes of the i-th criterion layer, represents the system weight of the k-th index of the i-th criterion layer, represents the direction of the k-th index of the i-th criterion layer, is the standardized index value of the k-th index of the i-th criterion layer in the t-th year;

[0079] The annual comprehensive score is obtained by summing the annual comprehensive evaluation indexes of all criterion layers, and the annual resource score is obtained by summing the annual comprehensive evaluation indexes of mountain, water, forest, grass and field criterion layers. For example, in XX County, as shown in Table 2:

[0080] Table 2: Comprehensive evaluation index of the mountain-water-forest-field-human life community in XX County from 2017 to 2022:

[0081]

[0082] The coupling degree index is calculated through the annual comprehensive evaluation index of each criterion layer, and the formula is as follows:

[0083] wherein, represents the coupling degree in the t-th year, and n is the number of criterion layers;

[0084] Based on the annual comprehensive evaluation index of each criterion layer and the corresponding criterion layer weight, the coordination degree index is calculated by weighted summation formula, and the formula is as follows: wherein, represents the coordination degree index, represents the weight of the i-th criterion layer;

[0085] The coupling coordination degree index is calculated by combining the coupling degree index and the coordination degree index, and the formula is as follows: wherein, represents the coupling coordination degree index.

[0086] Through the standardization processing of multi-source heterogeneous data, the annual comprehensive evaluation index, the coupling degree and the coordination degree index are calculated, which quantitatively reveals the synergy level among the six resource dimensions of mountain, water, forest, grass, field and human, helps to accurately identify the coordination and short board of resource utilization, dynamically monitors the balanced state of the subsystem, provides data support for optimizing resource allocation and strengthening weak links, and thus improves the overall sustainability of the resource system.

[0087] The time series analysis module is used to decompose the time series of the annual comprehensive evaluation index of each criterion layer, separate the short-term fluctuations and long-term development trend, and predict through the gray prediction model combined with the trend correction method;

[0088] Specifically, the ensemble empirical mode decomposition (EEMD) is used to decompose the time series of the annual comprehensive evaluation index of each criterion layer, to obtain the intrinsic mode function (IMF) and the trend item of different periods, to separate the short-term fluctuations and the long-term development trend, and specifically includes:

[0089] The fluctuation amplitude (standard deviation) of white noise is set, which is usually 10% to 20% of the fluctuation amplitude of the original data, and the number of repeated decomposition (ensemble average number) is set, which is generally between 100 to 500 times;

[0090] Each time of decomposition, a set of random white noise data is generated, the original time series data is added to the set of white noise data to form a new noisy data sequence, and the process is repeated each time using different white noise data;

[0091] Decompose a single noisy data sequence using empirical mode decomposition (EMD), comprising:

[0092] Step 1: Traverse the noisy data sequence one by one, find all local maximum points (maxima) and minimum points (minima);

[0093] Step 2: Use cubic spline interpolation method to connect all maxima to form a curve that surrounds the upper part of the data sequence (upper envelope), and similarly, connect all minima to form a curve that surrounds the lower part of the data sequence (lower envelope);

[0094] Step 3: Calculate the mean of the upper envelope and the lower envelope to get a middle curve (mean envelope), subtract this mean envelope from the noisy data sequence to get a candidate intrinsic mode function (IMF) component;

[0095] Step 4: Check if the candidate intrinsic mode function component meets the requirements of intrinsic mode function:

[0096] The number of extreme points in the entire sequence is equal to or differs from the number of zero-crossing points by at most 1;

[0097] At any time, the mean of the upper envelope and the lower envelope is 0;

[0098] If not, replace the original noisy data sequence with the candidate intrinsic mode function component, re-execute steps 2 and 3 until a required IMF component is obtained as the first IMF component;

[0099] Step 5: Subtract the first IMF component from the noisy data sequence to get the remaining part (residual sequence), and take the residual sequence as the new data sequence to be decomposed;

[0100] Step 6: Repeat steps 2 to 5 for the residual sequence to extract the second, third, … IMF components in turn, until the residual sequence becomes a monotonically increasing or monotonically decreasing curve, or becomes a fixed value, stop decomposition, at this time, obtain multiple IMF components and a final residual trend item.

[0101] Take the average of the first IMF component obtained by repeated decomposition as the final first IMF component; similarly, take the average of the second IMF component obtained by repeated decomposition as the final second IMF component, and so on, take the average of the residual trend item obtained by repeated decomposition as the final trend item;

[0102] Linear regression analysis is performed on the trend item obtained by decomposition to test the significance of the change trend, and the significance of the slope of the trend item regression line is verified by t-test (p<0.05). If it is not significant, adjust the white noise standard deviation parameter (±0.05) of EEMD and re-decompose;

[0103] The grey prediction model is selected to generate a cumulative sequence by once cumulative generation (1-AGO) of the time series of the annual comprehensive evaluation index of each criterion layer, and a whitening differential equation is constructed: , wherein, represents the sequence generated by once cumulative generation, a represents the development coefficient, and b represents the grey action amount. The parameters a and b are solved by the least square method, and a prediction model is established based on the parameters: The cumulative sequence is predicted, and the grey prediction value of the original sequence is obtained by cumulative reduction;

[0104] The trend item obtained by EEMD decomposition is extracted, and the trend item is extended to the prediction period by linear extrapolation. According to the slope obtained by linear regression analysis of the trend item, the trend item data in the future is calculated in sequence based on the last year data of the existing trend item;

[0105] The fitting error of the grey prediction value and the trend item is calculated, and the least square method is used to optimize the weight coefficient so that the error between the weighted prediction value and the historical data is minimized. The corrected prediction value is a weighted combination of the grey prediction value and the extended trend item: , wherein, represents the corrected prediction value, represents the grey prediction value, represents the extended trend item, and w represents the weight coefficient, which is controlled between 0.3 and 0.7;

[0106] Select several regions similar to the evaluation region in climate zone, terrain type and resource endowment as reference samples, collect the annual comprehensive evaluation index data of each criterion layer of these regions in the same evaluation period, and use cosine similarity to quantify the similarity between the evaluation region and the reference region. By calculating the cosine similarity of the time series of the annual comprehensive evaluation index of each criterion layer of the two regions, the reference regions with cosine similarity ≥0.6 are retained. For the retained reference regions, the same grey prediction+trend correction method as the evaluation region is used to obtain their future prediction values, and then the cosine similarity is weighted and fused: , wherein, represents the prediction value of the annual comprehensive evaluation index of each criterion layer, represents the prediction value of the reference region, and s represents the cosine similarity. The annual comprehensive evaluation index of each criterion layer in the prediction year is obtained;

[0107] Collect historical error data generated when predicting each criterion layer index in the past, introduce a time decay factor (such as 0.95), and calculate the weight of prediction error in different years: wherein, represents the weight of prediction error in different years, and a represents the time decay factor, represents the current year, and the reliability index is calculated. The calculation of the reliability index is to subtract the ratio of the weighted average prediction error and the average of the historical data from 1;

[0108] When the reliability index is greater than or equal to G1, the prediction result is directly adopted;

[0109] When the reliability index is less than G1, the model is retrained (the amount of training data is increased or the EEMD decomposition parameters are adjusted) until the reliability meets the standard, i.e., the reliability index is greater than or equal to G1.

[0110] Through ensemble empirical mode decomposition (EEMD), the long-term trend and short-term fluctuation of resource data are separated, and the future state is predicted by combining gray prediction and trend correction technology, which accurately captures the dynamic law of resource evolution such as degradation or recovery trend, enhances the robustness and reliability of prediction, and optimizes the prediction result through the reference region calibration mechanism, providing a forward-looking decision basis for resource management, and significantly improving the support ability of the system for long-term development strategy.

[0111] The comprehensive evaluation module is used to combine the data of historical years and predicted years to comprehensively evaluate the ecological protection effectiveness, resource utilization efficiency and resource coordination level from three dimensions to determine the development stage of resource utilization.

[0112] Specifically, the annual comprehensive evaluation index, annual resource score, annual comprehensive score and coupling coordination degree index of each criterion layer in the historical years and predicted years are obtained, the historical years and predicted years are taken as row dimensions, and the ecological protection effectiveness, resource utilization efficiency and resource coordination level are taken as column dimensions. Among them, the ecological protection effectiveness is composed of the weighted sum of the mountain, forest and grass criterion layer annual comprehensive evaluation index, the resource utilization efficiency is composed of the ratio of the annual resource score and the annual comprehensive score, and the coordination level is the coupling coordination degree index.

[0113] Linear normalization is used to map the matrix elements to [0, 1], and CRITIC method is used to determine the column dimension weight: the standard deviation and correlation coefficient of each dimension are calculated, and the weight formula is: wherein, represents the weight of the jth column dimension, represents the standard deviation, represents the sum of the correlation coefficients of the jth column dimension and other dimensions;

[0114] The improved TOPSIS method is used to calculate the comprehensive score, the positive ideal solution (the maximum value of each column) and the negative ideal solution (the minimum value of each column) are determined, and the weighted Euclidean distance of each year and the positive ideal solution and the negative ideal solution is calculated: , , wherein, represents the weighted Euclidean distance of the tth year and the positive ideal solution, represents the weighted Euclidean distance of the tth year and the negative ideal solution, represents the positive ideal solution, represents the negative ideal solution, represents the value of the jth column dimension of the tth year; the comprehensive score is calculated by the weighted Euclidean distance of each year and the positive ideal solution and the negative ideal solution: ;

[0115] When the comprehensive score of any year is <M1, it indicates that the resource utilization efficiency is low, and the development stage of the year is divided into an ecological imbalance period;

[0116] When M1≤the comprehensive score of any year is <M2, it indicates that the dimension score is balanced and rising, but the level of human-earth coordination still has room for improvement, and the development stage of the year is divided into a preliminary coordination period;

[0117] When the comprehensive score of any year is >M2, it indicates that ecological protection and economic development form a positive feedback, and the development stage of the year is divided into a benign cycle period, M1 and M2 are threshold values for dividing the development stage according to the comprehensive score, and the specific numerical value is adjusted and set according to the actual situation of the evaluation region.

[0118] By integrating historical and predicted data, a three-dimensional evaluation matrix is constructed, including ecological protection effect, resource utilization efficiency and coordination level, and the improved TOPSIS method is used to calculate the comprehensive score, so as to break through the limitation of single dimension evaluation, accurately divide the resource development stage based on the comprehensive score, and take ecological protection, efficiency and coordination level as the core.

[0119] The application also provides an electronic device, including a memory, a processor and a computer program stored in the memory and running on the processor, and the processor implements the steps performed by the system according to any one of the above embodiments when executing the computer program.

[0120] The application also provides a computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program implements the steps performed by the system according to any one of the above embodiments when being executed.

[0121] In the application, the several formulas involved are dimensionless numerical calculation, and the formula is obtained by software simulation of a large amount of collected data to obtain a formula of the most real situation, and the coefficients in the formula are set by the person skilled in the art according to the actual situation.

[0122] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art can be aware that units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of both. Whether the functions are performed by hardware or software depends on the specific application and design constraints of the technical solutions.

[0123] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, and can be located in one place or distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiments according to actual needs.

[0124] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application.

Claims

1. A time series-based resource utilization development degree comprehensive evaluation system, characterized in that: Comprise: The index system construction module is used for taking the mountain-water-forest-river-human-grass life community as the evaluation object, constructing a three-level evaluation index system including a target layer, a criterion layer and an index layer, the target layer is a regional mountain-water-forest-river-human-grass evaluation index system, the criterion layer includes six dimensions of mountain, water, forest, grass, field and human, each criterion layer is provided with an index layer, and the evaluation direction of each index is determined, the index reflecting the positive development of resources is set as a positive index, and the index having a negative impact on the development of resources is set as a negative index; the analytic hierarchy process is used to determine the weight of each criterion layer, the entropy weight method is used to calculate the element internal weight of each index in the same criterion layer, and the system weight of each index is calculated through the formula: system weight = criterion layer weight * element internal weight; The coupling coordination degree analysis module is used for standardizing the index data, calculating the annual comprehensive evaluation index of each criterion layer based on the standardized data and the system weight through a weighted summation formula; the annual comprehensive score is obtained by summing all the annual comprehensive evaluation indexes of the criterion layers; the annual resource score is obtained by summing the annual comprehensive evaluation indexes of the mountain, water, forest, grass and field criterion layers; the coupling degree index is obtained by the annual comprehensive evaluation indexes of each criterion layer; the coordination degree index is calculated based on the annual comprehensive evaluation indexes of each criterion layer and the weight of the criterion layer through a weighted summation formula; the coupling coordination degree index is obtained by multiplying the coupling degree index and the coordination degree index and then taking the square root; The time series analysis module is used for decomposing the time series of the annual comprehensive evaluation indexes of each criterion layer, separating short-term fluctuations and long-term development trends, and predicting through a grey prediction model combined with a trend correction method; The comprehensive evaluation module is used for combining the data of the historical years and the predicted years to perform comprehensive evaluation from three dimensions of ecological protection effect, resource utilization efficiency and resource coordination level, and determining the development stage of resource utilization, comprising: The annual comprehensive evaluation indexes, the annual resource scores, the annual comprehensive scores and the coupling coordination degree indexes of each criterion layer in the historical years and the predicted years are obtained respectively; a matrix is constructed, the historical years and the predicted years are taken as the row dimensions, and the ecological protection effect, the resource utilization efficiency and the resource coordination level are taken as the column dimensions, wherein the ecological protection effect is composed of the weighted summation of the annual comprehensive evaluation indexes of the mountain, forest and grass criterion layers, the resource utilization efficiency is composed of the ratio of the annual resource score to the annual comprehensive score, and the coordination level is the coupling coordination degree index; the matrix elements are mapped to the interval [0, 1] by linear normalization; the column dimension weight is determined by the CRITIC method; the comprehensive score is calculated by the improved TOPSIS method; when the comprehensive score of any year is less than M1, the development stage of the year is divided into an ecological imbalance period; when M1 is less than the comprehensive score of any year and less than M2, the development stage of the year is divided into a preliminary coordination period; when the comprehensive score of any year is greater than M2, the development stage of the year is divided into a benign circulation period.

2. The resource utilization development degree comprehensive evaluation system based on time series according to claim 1, wherein: The indexes of the mountain criterion layer include bare land area; The indexes of the water criterion layer include precipitation, runoff, evaporation, water area and wetland area; The indexes of the forest criterion layer include forest output value and forest area; The indexes of the grass criterion layer include grassland area, comprehensive coverage, animal husbandry output value and livestock quantity; The indexes of the field criterion layer include cultivated land area, agricultural output value and grain output; The indexes of the human criterion layer include population, per capita GDP, area and GDP.

3. The time series-based resource utilization development degree comprehensive evaluation system according to claim 1, characterized in that: the time series of the annual comprehensive evaluation index of each criterion layer is decomposed by using ensemble empirical mode decomposition to obtain intrinsic mode functions and trend items of different periods; linear regression analysis is performed on the trend items obtained by decomposition to test the significance of the change trend; a grey prediction model is used to predict the time series of the annual comprehensive evaluation index of each criterion layer to obtain a grey prediction value; the trend item obtained by ensemble empirical mode decomposition is extracted and linearly extrapolated to the prediction period; the fitting error of the grey prediction value and the trend item is calculated, the weight coefficient is optimized by using the least square method, and a corrected prediction value is obtained, which is a weighted combination of the grey prediction value and the extrapolated trend item; a reference region prediction value is introduced, the cosine similarity of the annual comprehensive evaluation index of each criterion layer with the reference region is calculated, the difference between the corrected prediction value and the reference region prediction value is calculated, and the difference is multiplied by the cosine similarity and then added to the corrected prediction value to obtain the annual comprehensive evaluation index of each criterion layer in the prediction year.

4. The time series-based resource utilization development degree comprehensive evaluation system according to claim 3, characterized in that: The time series of the annual comprehensive evaluation index of each criterion layer is decomposed by using ensemble empirical mode decomposition, including decomposing a single noisy data sequence by using empirical mode decomposition, and the specific steps are as follows: Step 1: Traverse the noisy data sequence one by one to find all local maximum points and minimum points; Step 2: Use cubic spline interpolation to connect all maximum points to form an upper envelope line and connect all minimum points to form a lower envelope line; Step 3: Calculate the mean of the upper envelope line and the lower envelope line to obtain a mean envelope line, and subtract the mean envelope line from the noisy data sequence to obtain a candidate intrinsic mode function component; Step 4: Check whether the candidate intrinsic mode function component meets the requirements of the intrinsic mode function: In the entire sequence, the number of extreme points is equal to or differs from the number of zero-crossing points by at most 1. At any time, the mean of the upper envelope line and the lower envelope line is 0. If not, replace the candidate intrinsic mode function component with the original noisy data sequence, and re-execute steps 2 and 3 until a required intrinsic mode function component is obtained as the first intrinsic mode function component. Step 5: Subtract the first intrinsic mode function component from the noisy data sequence to obtain a residual sequence; Step 6: Repeat steps 2 to 5 for the residual sequence to extract subsequent intrinsic mode function components one by one until the residual sequence becomes a monotonically increasing or decreasing curve or a fixed value, and stop the decomposition. At this time, a plurality of IMF components and a last residual trend item are obtained.

5. The time series-based resource utilization development degree comprehensive evaluation system according to claim 1, characterized in that: the comprehensive score is calculated by improving the TOPSIS method, including determining the positive ideal solution and the negative ideal solution; calculating the weighted Euclidean distance of each year from the positive and negative ideal solutions; obtaining a comprehensive score by the weighted Euclidean distance of each year from the positive and negative ideal solutions, the comprehensive score being the weighted Euclidean distance of the year from the negative ideal solution divided by the sum of the weighted Euclidean distance of the year from the positive ideal solution and the weighted Euclidean distance of the year from the negative ideal solution. 6.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-5 when the computer program is executed by the processor. The processor, when executing the computer program, implements the steps performed by the system of any one of claims 1 to 5.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by the processor, implements the steps performed by the system of any one of claims 1 to 5.

Images