Land Remediation Potential Index Assessment Method Based on Implementation Health and Existing Pressure

The land remediation potential energy index assessment method solves the problem of existing land remediation assessments that emphasize results over process, enabling scientific and efficient land remediation project assessments and improving the accuracy and consistency of the assessments.

CN123117902APending Publication Date: 2026-11-17GUANGXI ZHUANG AUTONOMOUS REGION NATURAL RESOURCES ECOLOGICAL RESTORATION CENT
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611300349.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-26
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing land remediation assessment methods prioritize results over process, resulting in redundant indicators, a large workload for data collection, and significant subjective bias in assessment results. This makes it difficult to scientifically and efficiently assess the health of land remediation projects and the existing pressure on land remediation projects.

Method used

A land remediation potential energy index assessment method based on execution health and stock pressure is adopted. Through outlier detection, multi-dimensional feature variable calculation, coefficient of variation weighting and minimum-maximum normalization processing, a standardized dataset is generated, and the reclamation potential energy index is calculated and classified for diagnosis.

Benefits of technology

It improved the ability to obtain process information in the assessment, reduced data collection costs, eliminated subjective bias, improved the accuracy and consistency of the assessment, and achieved concise and comparable assessment results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123117902A_ABST
    Figure CN123117902A_ABST
Patent Text Reader

Abstract

This invention discloses a land remediation potential energy index assessment method based on the health of land use and existing pressure, relating to the fields of land resource management and data analysis technology. The method includes: performing outlier detection and missing value imputation on the basic data of each assessment unit within the assessment area to generate a standardized dataset; calculating multi-dimensional feature variables for each assessment unit; calculating the coefficient of variation corresponding to the multi-dimensional feature variables and determining the final weight of each dimension in conjunction with physical constraints; using a minimum-maximum normalization algorithm to dynamically normalize the multi-dimensional feature variables, obtaining the normalized baseline value of each assessment unit in each dimension; calculating the reclamation potential energy index of each assessment unit, and classifying each assessment unit into several levels based on the reclamation potential energy index to output a diagnostic report for each assessment unit. This invention achieves a highly simplified indicator system and a significant reduction in data collection costs while ensuring assessment accuracy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of land resource management and data analysis technology, and more specifically, to a method for assessing the land remediation potential index based on the health of implementation and the pressure of existing stock. Background Technology

[0002] Temporary land use for construction projects refers to land temporarily used during the construction process, including construction access roads, material storage yards, beam fabrication yards, and borrow pits. After these lands are used, they must be reclaimed according to a land remediation plan to restore them to a usable state. Furthermore, land damage and degradation caused by mining, pollution incidents, and natural disasters also require systematic land remediation projects to restore their ecological or productive functions. In recent years, with the rapid advancement of infrastructure construction and the continuous improvement of ecological protection requirements, the number of various land remediation projects, such as temporary land reclamation, mine ecological restoration, and contaminated site remediation, has continued to increase, making the regulatory tasks increasingly heavy. How to scientifically and efficiently assess the progress of each project's remediation and promptly identify lagging projects has become an urgent problem to be solved in the fields of natural resource management and ecological environment supervision. In particular, how to quantitatively assess the land remediation potential from the two dimensions of implementation health and existing pressure is key to improving regulatory effectiveness.

[0003] Currently, the assessment of the effectiveness of land reclamation and restoration suffers from the following technical deficiencies: (1) The assessment method focuses on the final result and lacks tracking of the reclamation process: Existing reclamation assessment methods mostly focus on evaluating the final state after reclamation is completed. For example, they compare changes in land use types before and after reclamation through remote sensing imagery, or test soil quality through on-site sampling. This "results-oriented, process-neglecting" assessment approach makes it difficult to obtain process information reflecting the health of project implementation. This makes it difficult for managers to promptly identify projects with lagging reclamation progress and to make basic judgments on the changing trends of regional reclamation stock pressure.

[0004] (2) Too many evaluation indicators, resulting in a large workload for data collection: Existing land remediation assessment methods generally employ complex, multi-level, and multi-dimensional indicator systems. For example, in the suitability evaluation of temporary land reclamation for transportation projects, seven evaluation factors are required, including topographic slope, effective soil layer thickness, irrigation and drainage conditions, surface materials, land use compatibility, location conditions, and degree of damage. In the decision-making process for temporary construction land reclamation in coal mines, more than 20 evaluation factors are involved. The evaluation of the effectiveness of rural construction land demolition and reclamation requires 14 indicators covering three aspects: economic, social, and ecological benefits. The more indicators there are, the more data needs to be collected, increasing the workload for grassroots managers. More importantly, many indicators are interconnected; if a project's reclamation progresses rapidly, multiple indicators will improve simultaneously. Many indicators essentially merely describe the same state repeatedly without providing new information. This redundancy not only increases management costs but also fails to make the assessment more accurate and hinders the formation of concise, comparable quantitative indices for implementation health and existing pressure.

[0005] (3) The weighting of indicators depends on expert scoring, which can easily introduce subjective bias: Weighting indicators is a crucial step in land reclamation assessment. Different indicators have varying importance and require corresponding weights. Currently, the most commonly used method is the Analytic Hierarchy Process (AHP), which requires experts to compare and score indicators pairwise, then calculate their weights. The problem with this approach is that different experts have different knowledge structures and experience backgrounds, leading to potentially large discrepancies in scores and inconsistent assessment results. Although some studies have attempted to combine AHP with entropy weighting, the AHP step still relies heavily on expert judgment, making it difficult to guarantee the consistency and objectivity of the assessment results regarding the health of the land reclamation and the pressure on existing assets.

[0006] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0007] To address the problems in related technologies, this invention proposes a land remediation potential energy index assessment method based on execution health and stock pressure, in order to overcome the aforementioned technical problems existing in the current related technologies.

[0008] Therefore, the specific technical solution adopted by the present invention is as follows: Based on the land remediation potential index assessment method that considers both the health of the land and the existing pressure, this method includes: S1. Perform outlier detection and missing value imputation on the basic data of each assessment unit within the area to be assessed, and generate a standardized dataset; the basic data includes time series data of reclamation rate, existing area data and project completion data; S2. Based on the standardized dataset, calculate the multi-dimensional feature variables of each evaluation unit; among which, the multi-dimensional feature variables include target output rate, execution health, stock pressure coefficient and correction momentum; S3. Calculate the coefficient of variation for the multi-dimensional feature variables, assign initial weights based on the coefficient of variation, and determine the final weights for each dimension in combination with physical constraints. S4. Using the min-max normalization algorithm, the multi-dimensional feature variables are dynamically normalized to obtain the normalized benchmark values ​​of each evaluation unit in each dimension. S5. Calculate the reclamation potential energy index of each assessment unit based on the final weight and normalized benchmark value, and divide each assessment unit into several levels according to the reclamation potential energy index to output a diagnostic report for each assessment unit.

[0009] Furthermore, the expression for the target output rate is: ; The expression for calculating health is: ; The expression for the stock pressure coefficient is: ; The expression for the corrected momentum is: ; In the formula, Target output rate; The area to be reclaimed in stages; Initial stock area; To implement health status; The average monthly reclamation rate within the time window; The standard deviation of the monthly reclamation rate; This refers to the existing pressure coefficient. Ending stock area; To correct momentum; This represents the difference between the final reclamation rate and the early reclamation rate.

[0010] Furthermore, S3 includes: S31. Based on the multi-dimensional characteristic variables of each evaluation unit, calculate the mean and standard deviation of each dimension characteristic variable in all evaluation units, and generate the coefficient of variation of each dimension characteristic variable based on the calculation results to determine the corresponding initial weight. S32. Based on the physical constraints, with the goal of minimizing the Euclidean distance between the final weights and the initial weights, the final weights of the feature variables in each dimension are obtained.

[0011] Furthermore, the expression for the mean is: ; The expression for standard deviation is: ; In the formula, The mean; The total number of evaluation units; Index for the evaluation unit; For the first The evaluation unit in the first The unnormalized values ​​of each dimension; The standard deviation is denoted as .

[0012] Furthermore, the expression for the coefficient of variation is: ; In the formula, The coefficient of variation; Standard deviation; This is the mean.

[0013] Furthermore, the physical constraints include: The final weight of the target output rate is greater than or equal to the first preset threshold; the final weight of the execution health is less than or equal to the second preset threshold; and the sum of the final weights of the target output rate, execution health, stock pressure coefficient and correction momentum is equal to the target threshold.

[0014] Furthermore, S4 includes: S41. Use the target output rate, execution health and correction momentum in the multi-dimensional characteristic variables as positive indicators, and the stock pressure coefficient in the multi-dimensional characteristic variables as negative indicators. S42. Based on the characteristic variables of each dimension corresponding to each evaluation unit, determine the maximum and minimum values ​​of each positive and negative indicator in all evaluation units. S43. Based on the maximum and minimum values, the positive and negative indicators of each evaluation unit are dynamically normalized to obtain the normalized benchmark values ​​of each evaluation unit in each dimension.

[0015] Furthermore, the expression for the dynamic normalization of positive indicators is: ; In the formula, For the first The evaluation unit in the first Normalized baseline values ​​in each dimension; For the first The evaluation unit in the first Original values ​​of feature variables in each dimension; For the first The minimum value of each dimension across all evaluation units; For the first The maximum value of each dimension across all evaluation units.

[0016] Furthermore, the expression for the dynamic normalization of negative indicators is: ; In the formula, For the first Normalized baseline value of each assessment unit in the dimension of stock pressure coefficient; For the first Original values ​​of characteristic variables for each assessment unit in the dimension of stock pressure coefficient; The minimum value of the stock pressure coefficient among all assessment units; This represents the maximum value of the stock pressure coefficient across all assessment units.

[0017] Furthermore, the expression for the reclamation potential energy index is: ; ; In the formula, The land reclamation potential energy index; For dimension indexing; For the first The final weights of each dimension; For the first The evaluation unit in the first Normalized baseline values ​​in each dimension; The final weight for the target output rate; For the first The normalized baseline value of each evaluation unit in the target output dimension; To determine the final weight for health; For the first The normalized baseline value of each assessment unit on the health dimension; The final weight of the stock pressure coefficient; For the first Normalized baseline value of each assessment unit in the dimension of stock pressure coefficient; To adjust the final weight of momentum; For the first The normalized baseline value of each evaluation unit in the modified momentum dimension.

[0018] The beneficial effects of this invention are as follows: 1. This invention sets the execution health level to the form of mean divided by one plus standard deviation, so that the execution health level of the zero-output evaluation unit is automatically reset to zero. This effectively eliminates the paradox of stable zero output being misjudged as stable and healthy in traditional stability evaluation, and improves the evaluation method's ability to identify invalid output.

[0019] 2. This invention sets the corrected momentum as the mean multiplied by an acceleration, and uses the mean as the natural weighting coefficient of the acceleration. This prevents a slight rebound in the low base region from being misjudged as a trend reversal, thus overcoming the oversensitivity of traditional momentum indicators to small sample fluctuations and improving the accuracy and robustness of trend judgment.

[0020] 3. By imposing a physical constraint that the final weight of the target output rate is greater than or equal to the first preset threshold, this invention ensures that the evaluation results are dominated by the objective fact of the completed reclamation area, so that trend indicators and stability indicators only play a regulatory role, thereby enhancing the physical interpretability and credibility of the evaluation results.

[0021] 4. This invention automatically determines the weights of each dimension by using the coefficient of variation method, so that the weight allocation is driven entirely by the distribution characteristics of the data itself, without the need for expert scoring, thus eliminating the evaluation bias caused by subjective weighting and achieving objectivity and repeatability of the evaluation process.

[0022] 5. This invention uses Occam's razor variable screening mechanism to eliminate candidate variables that are highly redundant with the target output rate through correlation analysis, and retains only dimensions with marginal contributions greater than or equal to the preset values ​​through information gain verification. This results in the selection of four non-redundant core variables: target output rate, execution health, stock pressure coefficient, and correction momentum. This achieves a simplification of the indicator system and a significant reduction in data collection costs while ensuring the accuracy of the evaluation. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is one of the flowcharts of the land remediation potential energy index assessment method based on execution health and stock pressure according to an embodiment of the present invention; Figure 2 This is the second flowchart of the land remediation potential energy index assessment method based on execution health and stock pressure according to an embodiment of the present invention. Figure 3 This is a trend chart of reclamation rates in cities A, B, C, D, and E from January to May 2026, based on the land remediation potential energy index assessment method according to an embodiment of the present invention, which is based on the health of the implementation and the pressure of existing stock. Figure 4This is a trend chart of reclamation rates in cities F, G, H, I, and J from January to May 2026, based on the land remediation potential energy index assessment method of execution health and stock pressure according to an embodiment of the present invention. Figure 5 This is a trend chart of reclamation rate in cities K, L, M, and N and the entire province from January to May 2026, based on the land remediation potential index assessment method based on execution health and stock pressure according to an embodiment of the present invention. Detailed Implementation

[0025] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention.

[0026] According to embodiments of the present invention, a method for assessing the land remediation potential index based on the health of the land and the pressure of existing land is provided.

[0027] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figures 1-2 As shown, according to an embodiment of the present invention, a land remediation potential energy index assessment method based on execution health and stock pressure includes: S1. Perform outlier detection and missing value imputation on the basic data of each assessment unit within the area to be assessed, and generate a standardized dataset; the basic data includes time series data of reclamation rate, existing area data and project completion data; It should be added that the following data should be collected for each assessment unit in the area to be assessed within a continuous time window: historical reclamation rate time series data, initial stock area and final stock area, phased completed reclamation area and the number of completed reclamation projects and the number of projects in the database.

[0028] S2. Based on the standardized dataset, calculate the multi-dimensional feature variables of each evaluation unit; among which, the multi-dimensional feature variables include target output rate, execution health, stock pressure coefficient and correction momentum; In this optional embodiment, the expression for the target output rate is: ; The expression for calculating health is: ; The expression for the stock pressure coefficient is: ; The expression for the corrected momentum is: ; In the formula, Target output rate; The area to be reclaimed in stages; Initial stock area; To implement health status; The average monthly reclamation rate within the time window; The standard deviation of the monthly reclamation rate; This refers to the existing pressure coefficient. Ending stock area; To correct momentum; This represents the difference between the final reclamation rate and the early reclamation rate.

[0029] It's important to note that in traditional statistics, a standard deviation of 0 is generally considered the most stable state. However, in land remediation assessments, if a city's monthly reclamation rate is consistently 0, resulting in a standard deviation of 0, traditional methods mistakenly classify it as "most stable." This contradicts the fundamental physical meaning of reclamation assessment: sustained zero output is precisely the most severe state of failure. When the reclamation rate of any unit within the assessment area is 0, its... (Execution health) reaches zero, that is: if ,but Perform health assessment This invention effectively distinguishes between stable zero output and stable high output by using the formula "mean divided by (1 + standard deviation)". The invention introduces the mean... As an activity penalty factor, construct .when hour, Automatically reset to zero; when and Time (e.g., stable monthly reclamation). To achieve the highest level of health, the standard deviation is a measure of dispersion independent of the mean, and the two are usually reported separately. This invention couples the mean and standard deviation into a single indicator and gives it the physical meaning of "activity × stability"; this is not a simple transformation of statistical formulas, but a redefinition of the applicable boundaries of statistical tools.

[0030] in, In the formula, The reclamation rate at the last time point within the time window. The reclamation rate is the average reclamation rate at the midpoint of a time window, with a window length starting from one month, and the difference between two time points is taken. In existing technologies, traditional trend analysis of growth rates or accelerations suffers from a severe base effect: a city's increase from 1% to 2% (+100%) is considered a strong trend, while a city's increase from 20% to 22% (+10%) is underestimated. In reclamation assessments, slight rebounds in low-base areas should not be given high weight because the absolute progress is extremely limited. This invention constructs... , output level As acceleration The natural weighting coefficients. When Even at that time , The increase was still being Suppress; when hour, The contribution of [the economy / entity] is amplified. In existing technologies, the growth rate is a relative indicator independent of the absolute level. This invention couples the absolute level and relative change into a single indicator, allowing trend judgment to return to the physical fact of "output level". This is not a simple multiplication operation, but a modification of the trend analysis paradigm, that is, acknowledging a fundamental principle: progress at a low base does not constitute a trend reversal.

[0031] Furthermore, the traditional approach believes that more indicators equate to greater comprehensiveness, but multiple indicators have the following drawbacks: high data collection costs, multicollinearity among indicators, obscuring core driving factors, and excessive complexity leading to implementation difficulties. This invention, through correlation analysis and principal component contribution verification, selects four non-redundant core variables from a large pool of candidate variables (…). Target output rate Perform health status, Existing pressure coefficient (Momentum Correction). This invention demonstrates through systematic variable screening that "fewer indicators = more accurate assessment," proposes the concept of a minimum effective variable set in the field of reclamation assessment, and provides a specific screening method, namely, a variable screening mechanism based on Occam's Razor, which selects the model with the fewest variables while achieving the same predictive accuracy. The steps of the Occam's Razor variable screening mechanism are as follows: (1) From a business logic perspective, the original variables that may be related to the effectiveness of reclamation include core output rate (i.e., target output rate), stock digestion capacity, historical repayment capacity, average monthly reclamation rate, execution stability, corrected momentum, stock scale, and project conversion efficiency. Among these, execution stability is a statistical indicator that measures the degree of process fluctuation, reflecting the dispersion of the reclamation rate over time. It does not concern itself with the level of output, but only with whether the output is stable. Execution health is a comprehensive evaluation indicator that couples output activity with volatility, reflecting the healthy state of the reclamation process, which is both productive and stable. The candidate variable pool includes execution stability, which reflects the degree of fluctuation in the reclamation rate. After Occam's razor screening, execution stability was further improved into execution health. — By multiplying by the average monthly reclamation rate This automatically resets zero-output units, thus resolving the paradox of traditional stability indicators misjudging stable zero output as stable and healthy. The filtered four-variable simplified indicator system includes execution health. Instead of the original stability metrics.

[0032] (2) Calculate the correlation coefficient between each original variable and the reclamation rate, and the stock absorption capacity. Historical solvency Average monthly reclamation rate Implement health status Corrected momentum Existing scale Project conversion efficiency Highly correlated with core variables ( Parameters that are low in correlation or uncorrelated with the core variables are removed. The parameters of the core variables are retained. Variables highly correlated with the core variables are removed because their information overlaps significantly, and adding them does not provide new insights for the evaluation model. Variables with low correlation to the core variables are retained because they provide independent new information. For example, the monthly average reclamation rate is simply a time-averaged form of the reclamation rate; both share the numerator information of "completed area" and are essentially measuring the output level of the same parameter. Retaining both would create information redundancy. Therefore, the retained variables include reclamation rate, execution health, corrective momentum, stock size, and project conversion efficiency.

[0033] (3) Perform information gain verification: Add variables one by one and calculate The change in quantity. The coefficient of determination (COD) is a statistical indicator that measures how well a model fits the data; it represents the proportion of the total variation in the target variable that can be explained by the model. Only those with a percentage of ≥2% are retained; as shown in Table 1.

[0034] Table 1. Example of Information Gain Verification (4) Occam's Razor Optimal Model Selection: Define the objective function: ,in For the number of variables, The penalty coefficient is 0.02; as shown in Table 2.

[0035] Table 2. Example of optimal model selection using Occam's razor (5) The final set of variables is obtained, as shown in Table 3.

[0036] Table 3 Example of the final variable set S3. Calculate the coefficient of variation for the multi-dimensional feature variables, assign initial weights based on the coefficient of variation, and determine the final weights for each dimension in combination with physical constraints. In this optional embodiment, S3 includes: S31. Based on the multi-dimensional characteristic variables of each evaluation unit, calculate the mean and standard deviation of each dimension characteristic variable in all evaluation units, and generate the coefficient of variation of each dimension characteristic variable based on the calculation results to determine the corresponding initial weight. In this optional embodiment, the expression for the mean is: ; The expression for standard deviation is: ; In the formula, The mean; The total number of evaluation units; Index for the evaluation unit; For the first The evaluation unit in the first The unnormalized values ​​of each dimension; The standard deviation is denoted as .

[0037] In this optional embodiment, the expression for the coefficient of variation is: ; In the formula, The coefficient of variation; Standard deviation; This is the mean.

[0038] S32. Based on the physical constraints, with the goal of minimizing the Euclidean distance between the final weights and the initial weights, the final weights of the feature variables in each dimension are obtained.

[0039] In this optional embodiment, the physical constraints include: The final weight of the target output rate is greater than or equal to the first preset threshold; the final weight of the execution health is less than or equal to the second preset threshold; and the sum of the final weights of the target output rate, execution health, stock pressure coefficient and correction momentum is equal to the target threshold.

[0040] It should be added that the initial weights for each dimension are calculated. : Apply physical constraints to obtain the final weights. : (Core output rate priority principle). (Stability penalty cap) .

[0041] Existing technologies often employ a combination of the Analytic Hierarchy Process (AHP) and entropy weighting for weight assignment, relying on expert scoring and thus subject to subjective bias. While the purely objective entropy weighting method eliminates the need for experts, it may result in weight allocations that contradict common sense in physics. This invention constructs a two-layer weighting mechanism: the first layer involves objective calculation using the coefficient of variation method. Calculating data-driven weights for each dimension requires no expert intervention. The second layer of physical constraints applies rigid conditions. (Core output rate priority principle) (Stability penalty cap) In existing technologies, the coefficient of variation method is typically used only for the dispersion analysis of a single indicator. This invention extends it to multi-dimensional weight calculation and combines physical constraints with objective weight calculation into a two-layer mechanism. This is equivalent to superimposing the rigid anchoring of domain prior knowledge on data-driven weight calculation, ensuring objectivity while preventing the physical paradoxes that may arise from purely data-driven approaches.

[0042] S4. Using the min-max normalization algorithm, the multi-dimensional feature variables are dynamically normalized to obtain the normalized benchmark values ​​of each evaluation unit in each dimension. In this optional embodiment, S4 includes: S41. Use the target output rate, execution health and correction momentum in the multi-dimensional characteristic variables as positive indicators, and the stock pressure coefficient in the multi-dimensional characteristic variables as negative indicators. S42. Based on the characteristic variables of each dimension corresponding to each evaluation unit, determine the maximum and minimum values ​​of each positive and negative indicator in all evaluation units. S43. Based on the maximum and minimum values, the positive and negative indicators of each evaluation unit are dynamically normalized to obtain the normalized benchmark values ​​of each evaluation unit in each dimension.

[0043] In this optional embodiment, the expression for the dynamic normalization of the positive index is: ; In the formula, For the first The evaluation unit in the first Normalized baseline values ​​in each dimension; For the first The evaluation unit in the first Original values ​​of feature variables in each dimension; For the first The minimum value of each dimension across all evaluation units; For the first The maximum value of each dimension across all evaluation units.

[0044] In this optional embodiment, the expression for the dynamic normalization of negative indicators is: ; In the formula, For the first Normalized baseline value of each assessment unit in the dimension of stock pressure coefficient; For the first Original values ​​of characteristic variables for each assessment unit in the dimension of stock pressure coefficient; The minimum value of the stock pressure coefficient among all assessment units; This represents the maximum value of the stock pressure coefficient across all assessment units.

[0045] In this optional embodiment, the expression for the reclamation potential energy index is: ; ; In the formula, The land reclamation potential energy index; For dimensional indexing, Corresponding to the target output rate, Corresponding to the health status, Corresponding stock pressure coefficient Corresponding corrected momentum; For the first The final weights of each dimension; For the first The evaluation unit in the first Normalized baseline values ​​in each dimension; The final weight for the target output rate; For the first The normalized baseline value of each evaluation unit in the target output dimension; To determine the final weight for health; For the first The normalized baseline value of each assessment unit on the health dimension; The final weight of the stock pressure coefficient; For the first Normalized baseline value of each assessment unit in the dimension of stock pressure coefficient; To adjust the final weight of momentum; For the first The normalized baseline value of each evaluation unit in the modified momentum dimension.

[0046] It should be added that the normalized benchmark value is dynamically calculated based on the current data of all evaluation units, without the need for an external reference benchmark, thus achieving adaptive normalization. The Min-Max normalization algorithm is used to map the data of each dimension to the [0,1] interval: positive indicators ( , , ): Negative indicators ( ): .

[0047] Existing techniques mostly employ traditional Min-Max normalization, which uses fixed global extrema (such as the maximum / minimum value of all historical data). When the data distribution shifts, fixed extrema can lead to systematic biases in the normalized data. For example, if a city's land reclamation rate increases from 2% to 8%, in a normalization process that includes 40% of historical extrema, both the normalization rate and the fixed extrema will be close to 0, failing to reflect the progress trend.

[0048] The normalized benchmark value in this invention is dynamically calculated based solely on the data of all evaluation units in the current period, without the need for an external reference benchmark. The extreme values ​​are derived from the current period's data. The normalization benchmark automatically adapts as the data distribution changes. In traditional normalization, a fixed benchmark is considered a necessary condition to ensure the comparability of data from different periods. In this invention, the current data distribution is the core factor determining the relative ranking in the evaluation scenario; historical extreme values, on the contrary, freeze the evaluation scale, obscuring subtle real progress. This invention, through the concept of dynamic normalization, allows the sensitivity of the evaluation scale to adaptively adjust with the data distribution.

[0049] S5. Calculate the reclamation potential energy index of each assessment unit based on the final weight and normalized benchmark value, and divide each assessment unit into several levels according to the reclamation potential energy index to output a diagnostic report for each assessment unit.

[0050] It should be added that the graded diagnosis and output are based on... The evaluation units are divided into four levels: Benchmark level: Good level: General level: Alert Level: Output of each evaluation unit Values, rankings, grading diagnoses, and corresponding policy recommendations.

[0051] Example 1: The following are land remediation data for fourteen cities in a certain province, as shown in Tables 4 and 5.

[0052] Table 4. Example of Land Remediation Data (One of the Tables) Among them, the phase reclamation rate 1 = phase completed reclamation area / overdue unreclaimed area in the first quarter of 2026; the phase reclamation rate 2 = phase completed reclamation area / overdue unreclaimed area in the third quarter of 2025; the completion progress = (phase completed area - newly added area) / overdue unreclaimed area in the first quarter of 2026.

[0053] Table 5. Example Table of Land Remediation Data (Part Two) like Figures 3 to 5As shown, the calculations for each city are based on the above data. , , , As shown in Table 6.

[0054] Table 6. Example Table of Multidimensional Variables Min-Max normalization was performed, as shown in Table 7.

[0055] Table 7. Examples of Normalization Calculate the reclamation potential index for each city : ;Right now: As shown in Table 8.

[0056] Table 8 Example Table of Reclamation Potential Energy Index Graded diagnosis and output, based on The assessment of the reclamation potential level of each city is as follows: Benchmark level: Good level: General level: Alert Level: Output the reclamation potential of each city Values, rankings, level diagnoses, and corresponding policy recommendations are shown in Table 9.

[0057] Table 9. Example Table of Reclamation Potential Energy Index According to another embodiment of the present invention, a land remediation potential energy index assessment system based on performance health and stock pressure is also provided, the system comprising: The data acquisition module is used to perform outlier detection and missing value imputation on the basic data of each assessment unit in the area to be assessed, and generate a standardized dataset; wherein, the basic data includes reclamation rate time series data, existing area data and project completion data; The variable calculation module is used to calculate multi-dimensional feature variables for each evaluation unit based on a standardized dataset; wherein, the multi-dimensional feature variables include target output rate, execution health, stock pressure coefficient, and correction momentum; The weight optimization module is used to calculate the coefficient of variation corresponding to multi-dimensional feature variables, assign initial weights based on the coefficient of variation, and determine the final weights of each dimension in combination with physical constraints. The normalization module is used to dynamically normalize multi-dimensional feature variables using the min-max normalization algorithm to obtain the normalized baseline value of each evaluation unit in each dimension. The comprehensive assessment and grading diagnosis module is used to calculate the reclamation potential energy index of each assessment unit based on the final weight and normalized benchmark value, and to divide each assessment unit into several levels according to the reclamation potential energy index, so as to output a diagnosis report for each assessment unit.

[0058] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A land remediation potential index assessment method based on implementation health and existing pressure, characterized in that, The method includes: S1. Perform outlier detection and missing value imputation on the basic data of each assessment unit within the area to be assessed, and generate a standardized dataset; wherein, the basic data includes time series data of reclamation rate, existing area data and project completion data; S2. Based on the standardized dataset, calculate the multi-dimensional feature variables of each evaluation unit; wherein, the multi-dimensional feature variables include target output rate, execution health, stock pressure coefficient, and correction momentum; S3. Calculate the coefficient of variation for the multi-dimensional feature variables, assign initial weights based on the coefficient of variation, and determine the final weights for each dimension in combination with physical constraints. S4. Using the min-max normalization algorithm, the multi-dimensional feature variables are dynamically normalized to obtain the normalized benchmark values ​​of each evaluation unit in each dimension. S5. Calculate the reclamation potential energy index of each assessment unit based on the final weight and normalized benchmark value, and divide each assessment unit into several levels according to the reclamation potential energy index to output a diagnostic report for each assessment unit.

2. The land remediation potential index assessment method based on execution health and existing pressure as described in claim 1, characterized in that, The expression for the target output rate is: ; The expression for the execution health status is: ; The expression for the stock pressure coefficient is: ; The expression for the corrected momentum is: ; In the formula, Target output rate; The area to be reclaimed in stages; Initial stock area; To implement health status; The average monthly reclamation rate within the time window; The standard deviation of the monthly reclamation rate; This refers to the existing pressure coefficient. Ending stock area; To correct momentum; This represents the difference between the final reclamation rate and the early reclamation rate.

3. The land remediation potential index assessment method based on execution health and existing pressure as described in claim 1, characterized in that, S3 includes: S31. Based on the multi-dimensional characteristic variables of each evaluation unit, calculate the mean and standard deviation of each dimension characteristic variable in all evaluation units, and generate the coefficient of variation of each dimension characteristic variable based on the calculation results to determine the corresponding initial weight. S32. Based on the physical constraints, with the goal of minimizing the Euclidean distance between the final weights and the initial weights, the final weights of the feature variables in each dimension are obtained.

4. The land remediation potential index assessment method based on execution health and stock pressure according to claim 3, characterized in that, The expression for the mean is: ; The expression for the standard deviation is: ; In the formula, The mean; The total number of evaluation units; Index for the evaluation unit; For the first The evaluation unit in the first The unnormalized values ​​of each dimension; The standard deviation is denoted as .

5. The land remediation potential index assessment method based on execution health and stock pressure according to claim 3, characterized in that, The expression for the coefficient of variation is: ; In the formula, The coefficient of variation; Standard deviation; This is the mean.

6. The land remediation potential index assessment method based on execution health and stock pressure according to claim 3, characterized in that, The physical constraints include: The final weight of the target output rate is greater than or equal to the first preset threshold; the final weight of the execution health is less than or equal to the second preset threshold; and the sum of the final weights of the target output rate, execution health, stock pressure coefficient and correction momentum is equal to the target threshold.

7. The land remediation potential index assessment method based on execution health and existing pressure according to claim 1, characterized in that, S4 includes: S41. Use the target output rate, execution health and correction momentum in the multi-dimensional characteristic variables as positive indicators, and the stock pressure coefficient in the multi-dimensional characteristic variables as negative indicators. S42. Based on the characteristic variables of each dimension corresponding to each evaluation unit, determine the maximum and minimum values ​​of each positive and negative indicator in all evaluation units. S43. Based on the maximum and minimum values, the positive and negative indicators of each evaluation unit are dynamically normalized to obtain the normalized benchmark values ​​of each evaluation unit in each dimension.

8. The land remediation potential index assessment method based on execution health and stock pressure according to claim 7, characterized in that, The expression for the dynamic normalization of the positive index is: ; In the formula, For the first The evaluation unit in the first Normalized baseline values ​​in each dimension; For the first The evaluation unit in the first Original values ​​of feature variables in each dimension; For the first The minimum value of each dimension across all evaluation units; For the first The maximum value of each dimension across all evaluation units.

9. The land remediation potential index assessment method based on execution health and stock pressure according to claim 7, characterized in that, The expression for the dynamic normalization of the negative index is: ; In the formula, For the first Normalized baseline value of each assessment unit in the dimension of stock pressure coefficient; For the first Original values ​​of characteristic variables for each assessment unit in the dimension of stock pressure coefficient; The minimum value of the stock pressure coefficient among all assessment units; This represents the maximum value of the stock pressure coefficient across all assessment units.

10. The land remediation potential index assessment method based on execution health and stock pressure according to claim 1, characterized in that, The expression for the reclamation potential energy index is: ; ; In the formula, The land reclamation potential energy index; For dimension indexing; For the first The final weights of each dimension; For the first The evaluation unit in the first Normalized baseline values ​​in each dimension; The final weight for the target output rate; For the first The normalized baseline value of each evaluation unit in the target output dimension; To determine the final weight for health; For the first The normalized baseline value of each assessment unit on the health dimension; The final weight of the stock pressure coefficient; For the first Normalized baseline value of each assessment unit in the dimension of stock pressure coefficient; To adjust the final weight of momentum; For the first The normalized baseline value of each evaluation unit in the modified momentum dimension.