CN-122022135-A - Comprehensive evaluation method suitable for tropical coral sand soil conditioning technology

CN122022135ACN 122022135 ACN122022135 ACN 122022135ACN-122022135-A

Abstract

The invention discloses a comprehensive evaluation method suitable for a tropical coral sand soil conditioning technology, which is characterized in that an entropy weight-TOPSIS-gray correlation integrated decision algorithm is adopted to embed tropical coral island hard constraint conditions by constructing a soil-vegetation-environment three-coupling quantization index system, and a dynamic weight rewarding mechanism is combined to comprehensively evaluate a coral sand conditioning technical scheme of any source, so that the comprehensive evaluation method has the advantages of full objective weighting, quantitative ordering, robustness verification and directional scene constraint, and improves the overall efficiency and sustainability of coral sand soil remediation technology evaluation while promoting local material drawing and environment-friendly guiding of coral sand soil conditioning.

Inventors

ZHOU WEIWEN
CHENG HAO
SUN CUICI
WANG YOUSHAO

Assignees

中国科学院南海海洋研究所

Dates

Publication Date: 20260512
Application Date: 20260115

Claims (7)

1. The comprehensive evaluation method suitable for the tropical coral sand soil conditioning technology is characterized by comprising the following steps of: Collecting evaluation data of coral sand conditioning technical schemes from any source, forming a multi-scheme evaluation data set, and constructing an original data matrix based on the multi-scheme evaluation data set and a preset evaluation system, wherein the preset evaluation system comprises a plurality of evaluation indexes and corresponding dimension attributes; All evaluation data associated with schemes which do not meet preset hard constraint conditions are removed from the original data matrix to form a target matrix; traversing all schemes in the target matrix, and dynamically adjusting corresponding index weights in the evaluation data according to a preset rewarding scheme if corresponding evaluation data in the current scheme accords with rewarding conditions; Carrying out standardization processing on the target matrix to obtain a standardized matrix; calculating the initial weight of each evaluation index in the standardized matrix by adopting an entropy weight method; Fusing the index weight obtained by dynamic adjustment with the initial weight to generate a final weight; based on the final weight, calculating the comprehensive closeness of each scheme by adopting a TOPSIS method, and grading the corresponding scheme according to the comprehensive closeness; And carrying out robustness verification on the sequencing result of the comprehensive closeness by adopting a gray correlation analysis method, and outputting a comprehensive evaluation result.
2. The comprehensive evaluation method suitable for the tropical coral sand soil conditioning technology according to claim 1, wherein for missing values existing in the raw data matrix, a mean value of evaluation indexes corresponding to the missing values is adopted for filling.
3. The comprehensive evaluation method suitable for the tropical coral sand soil conditioning technology according to claim 1, wherein the preset hard constraint conditions are selected from the evaluation indexes, and include four hard constraint items including a conditioner addition proportion, a heavy metal increment, a pH reduction amplitude and a unit area cost, a trigger condition is set for each hard constraint item, all schemes in the original data matrix are traversed, and all evaluation data associated with the current scheme are removed from the original data matrix if the evaluation index of any one of the current schemes meets the trigger conditions.
4. The comprehensive evaluation method for the tropical coral sand soil conditioning technology according to claim 1, wherein the reward condition includes selecting a local material rate from the evaluation indexes as a unique reward item, and setting a value range of the local material rate to be more than 60%.
5. The comprehensive evaluation method suitable for the tropical coral sand soil conditioning technology according to claim 4, wherein for the scheme with the local material yield greater than 60%, the index weight of the local material yield in the scheme is temporarily increased to 2 times of the original value to be used as the rewarding scheme, and all weights in the scheme are normalized after dynamic adjustment, so that the total weight of the same scheme is always 1.
6. The method for comprehensively evaluating the conditioning technology of the tropical coral sand soil according to claim 1, wherein the normalizing treatment comprises converting all indexes in the target matrix into dimensionless values, and the numerical ranges are unified to be [0,1].
7. The method for comprehensively evaluating the tropical coral sand soil conditioning technology according to claim 1, wherein grading the corresponding schemes according to the comprehensive closeness comprises defining the scheme as excellent when the comprehensive closeness is more than or equal to 0.75, defining the scheme as good when the comprehensive closeness is more than 0.75 and less than or equal to 0.6, and defining the scheme as medium when the comprehensive closeness is less than 0.6.

Description

Comprehensive evaluation method suitable for tropical coral sand soil conditioning technology Technical Field The invention relates to the technical field of soil conditioning scheme evaluation, in particular to a comprehensive evaluation method suitable for a tropical coral sand soil conditioning technology. Background Tropical coral sand soil mainly consists of weathered coral sand and is widely distributed in isolated islands and reef areas of tropical sea areas in China. The soil is composed of sand grains serving as main particles, lacks sticky particles, has the core characteristics of high alkalinity (pH 8.5-9.5), high salinity and low fertility (organic matters <5g/kg and CEC <10 cmol/kg), and has poor water and fertilizer retaining capacity, so that natural vegetation growth is severely restricted, and construction and recovery of island ecosystems are hindered. Therefore, the conditioning and improvement of coral sand soil improves the ecological application value of the coral sand soil, and becomes a key subject in the development of tropical islands (especially artificial islands). At present, the evaluation scheme of the coral sand soil conditioning technology is mainly based on traditional qualitative evaluation, and has a plurality of defects that the actual application requirement is difficult to meet, firstly, an evaluation system is single and is limited to soil physicochemical indexes such as pH, EC, CEC, physiological response and economic safety of vegetation are not coupled, so that an evaluation result is one-sided and the overall ecological benefit of the conditioning technology cannot be comprehensively reflected, secondly, decision subjectivity is strong, evaluation weight is set randomly, objective quantitative ordering of the conditioning scheme is difficult to realize, thirdly, a scientific constraint and excitation mechanism is lacked, a poor scheme which damages the soil structure or has too high cost is easily introduced, reasonable excitation is not given to a local material drawing scheme according to the 'zero output' strategic requirement, fourthly, the output form is extensive, most of qualitative description is lacked, a visual chart and an automatic report support are lacked, and the decision efficiency is low. In summary, the existing evaluation method cannot realize objective and accurate evaluation of the conditioning effect of coral sand soil, and a comprehensive evaluation method with comprehensiveness, objectivity and high operability is needed. Disclosure of Invention Aiming at the problems, the invention provides a comprehensive evaluation method suitable for a tropical coral sand soil conditioning technology, which mainly solves the problems of the background technology. In order to solve the technical problems, the technical scheme of the invention is as follows: a comprehensive evaluation method suitable for a tropical coral sand soil conditioning technology comprises the following steps: Collecting evaluation data of coral sand conditioning technical schemes from any source, forming a multi-scheme evaluation data set, and constructing an original data matrix based on the multi-scheme evaluation data set and a preset evaluation system, wherein the preset evaluation system comprises a plurality of evaluation indexes and corresponding dimension attributes; All evaluation data associated with schemes which do not meet preset hard constraint conditions are removed from the original data matrix to form a target matrix; traversing all schemes in the target matrix, and dynamically adjusting corresponding index weights in the evaluation data according to a preset rewarding scheme if corresponding evaluation data in the current scheme accords with rewarding conditions; Carrying out standardization processing on the target matrix to obtain a standardized matrix; calculating the initial weight of each evaluation index in the standardized matrix by adopting an entropy weight method; Fusing the index weight obtained by dynamic adjustment with the initial weight to generate a final weight; based on the final weight, calculating the comprehensive closeness of each scheme by adopting a TOPSIS method, and grading the corresponding scheme according to the comprehensive closeness; And carrying out robustness verification on the sequencing result of the comprehensive closeness by adopting a gray correlation analysis method, and outputting a comprehensive evaluation result. In some embodiments, for missing values existing in the original data matrix, the average value of the evaluation index corresponding to the missing value is adopted for filling. In some embodiments, the preset hard constraint conditions are selected from the evaluation indexes, including four hard constraint terms including an addition proportion of a conditioning agent, a heavy metal increment, a pH reduction range and a unit area cost, a trigger condition is set for each hard constraint term, all schemes i