A comprehensive evaluation method for the quality of plowshare substructure construction based on multi-index fusion
By constructing a plow pan quality state matrix and a fuzzy measurement matrix, the nonlinear correlation and conflict relationships of soil indicators are identified. Combined with the Schogey integral calculation, the problem of evaluation result deviation and weak indicator location in the existing technology is solved, and the accurate evaluation of plow pan construction quality and process adjustment are realized.
Patent Information
- Application Number
- CN202611358969.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-03
- Publication Date
- 2026-11-17
AI Technical Summary
In the evaluation of the quality of plow pan construction, the existing technology, which uses fixed-weight linear weighting, cannot characterize the non-linear coupling relationship between soil indicators, leading to biased evaluation results. It is also difficult to identify weak indicators and cannot provide targeted guidance for process adjustments. The evaluation results deviate significantly, especially when tillage conditions change.
By constructing a plow pan quality state matrix, the degree of soil degradation of test indicators is calculated, strong nonlinear correlations and conflict relationships are identified, fuzzy measure matrix and Shapley value are used to calculate the importance of indicators, and the dynamic reconstruction of the comprehensive evaluation index is achieved by combining the Schogey integral, weak indicators are identified and process adjustment suggestions are provided.
It enables precise evaluation of the quality of the plow pan structure, identifies weak indicators and provides targeted adjustment suggestions, improves the accuracy and stability of the evaluation, reduces evaluation bias and blind control, and ensures the long-term improvement of the quality of the plow pan.
Smart Images

Figure CN123117918A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of agricultural soil monitoring and intelligent evaluation technology, and more specifically, to a comprehensive evaluation method for the quality of plow pan construction based on multi-index fusion. Background Technology
[0002] In the field of agricultural soil management, methods for evaluating the quality of plow pan construction based on field sampling and physicochemical analysis have been widely applied. Typically, soil penetration resistance, saturated hydraulic conductivity, bulk density, and porosity are collected at different soil depths through manual sampling or sensors, and individual indicators are judged to meet preset thresholds. In the specific evaluation process, existing technical solutions focus on the independent analysis of each detection indicator, calculating a comprehensive score through linear weighting with fixed weights to classify the quality of the plow pan construction. These methods generally employ a simple summation of scores from multiple indicators, and their technical implementation involves the collaborative application of soil physical property testing and agricultural engineering evaluation technologies. However, because each indicator mechanistically represents the degree of compaction, water permeability, and structural looseness, their importance fluctuates dynamically with changes in tillage conditions, and static settings with fixed weights cannot cover the complexity of actual scenarios.
[0003] However, existing technologies have significant limitations in indicator fusion and evaluation decision-making. First, there is an inherent contradiction between fixed-weight linear weighting and the complex interactions between plow pan indicators. Since indicators such as soil mechanical support, water regulation, and structural integrity influence each other and may exhibit non-linear coupling, simple weighting cannot characterize this relationship, leading to deviations between evaluation results and the actual functional state of the plow pan. Second, traditional methods often only output a total score, making it difficult to pinpoint the weak indicators causing quality decline, thus failing to provide targeted guidance for adjusting plow pan construction process parameters. Furthermore, in field scenarios with significant differences in tillage depth and soil texture, fixed weighting struggles to adapt to changing environments, further exacerbating the deviation in evaluation results. This lack of evaluation information weakens the decision-making support value of the results for improving plow pan quality, affecting not only the accuracy of a single evaluation but also potentially leading to errors in subsequent tillage decisions. Therefore, it is difficult to meet the actual needs of refined farmland management for diagnostic depth and control precision. Summary of the Invention
[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide a comprehensive evaluation method for the construction quality of the plowshare layer based on multi-index fusion to solve the problems mentioned in the background art.
[0005] To achieve the above objectives, the present invention provides the following technical solution: A comprehensive evaluation method for the quality of plowshare substructure construction based on multi-index fusion includes the following steps: S1. Receive soil test index data collected on-site according to the test section location at the site of the plow bottom construction quality acceptance evaluation. The soil test index includes mechanical support index, moisture control index and structural integrity index at different soil depths. Perform positive and standardization processing on each soil test index, calculate the degree of deterioration of each soil test index relative to the preset corresponding functional benchmark value, generate a degree of deterioration sequence, arrange the degree of deterioration sequence according to the test section location, and construct the plow bottom quality state matrix. S2. Read the plow bottom quality state matrix, calculate the maximum information coefficient and Spearman rank correlation coefficient between any two soil detection indicators, identify indicator pairs with strong nonlinear correlation based on the maximum information coefficient, select representative indicators from the indicator pairs according to the preset retention indicator selection rules and remove redundant indicators to obtain the retention indicator set, determine the positive synergy or negative conflict relationship between indicators based on the sign of the Spearman rank correlation coefficient between each indicator in the retention indicator set, and construct a fuzzy measure matrix to characterize the interaction tendency between indicators based on the retention indicator set and the conflict relationship. S3. Based on the fuzzy measure matrix, calculate the Shapley value of each indicator in the retained indicator set as the indicator importance. Input the degree to which the average deterioration of each indicator exceeds the total average deterioration into the variable weight equilibrium function to perform short-board strengthening correction on the indicator importance. Reconstruct the fuzzy measure matrix based on the corrected indicator importance to obtain the corrected fuzzy measure matrix. S4. For each detection section, based on the corrected fuzzy measure matrix, the degradation degree of each indicator is converted into a positive quality score. The positive quality scores of each indicator of the section are aggregated into the plowshare of the detection section to construct a comprehensive evaluation index. S5. Based on the preset quality level range, map the comprehensive evaluation index of each test section to the quality level of the plowshare substructure of the test section. Based on the marginal contribution rate of each indicator to the comprehensive evaluation index of the test section during the Shape integral calculation, identify the weak indicators of the test section. Based on the preset correspondence between the weak indicators and the plowshare substructure construction process parameters, output a diagnostic evaluation report containing the comprehensive evaluation index, quality level, weak indicators and corresponding process adjustment suggestions for each test section.
[0006] In a preferred embodiment, soil testing index data collected at the site of the plow pan quality acceptance evaluation according to the test section location is received. The soil testing indexes include mechanical support indexes, moisture regulation indexes, and structural integrity indexes at different soil depths. Each soil testing index is positively oriented and standardized, and the degree of deterioration of each soil testing index relative to a preset corresponding functional benchmark value is calculated to generate a deterioration degree sequence. The deterioration degree sequence is arranged according to the test section location to construct a plow pan quality state matrix, specifically including: The soil testing index data is collected on-site at the test section locations for the quality acceptance evaluation of the plow layer construction. The soil testing index data is obtained by determining multiple test section locations at preset intervals along the extension direction of the plow layer, dividing multiple soil layer depths at preset depth intervals at each test section location, and collecting soil samples at each soil layer depth. The soil testing indexes include mechanical support indexes, moisture control indexes, and structural integrity indexes. The mechanical support index is soil penetration resistance, the moisture control index is soil saturated hydraulic conductivity and soil moisture content, and the structural integrity index is soil bulk density and soil porosity. A baseline value table is established in advance according to each soil testing index. Under each soil testing index, functional baseline value, preferred type and preset optimal value are set according to different soil depths. The preferred type is one of the following: the larger the value, the better the quality; the smaller the value, the better the quality; or the closer the value is to the optimal value, the better the quality. The preferred type for soil penetration resistance and soil bulk density is that the smaller the value, the better the quality. The preferred type for soil saturated hydraulic conductivity and soil moisture content is that the closer the value is to the optimal value, the better the quality. The preferred type for soil porosity is that the larger the value, the better the quality. For each soil testing index, a positive transformation process is performed according to the corresponding preferred type. For soil testing indices with larger values and better quality, the soil testing index value is retained as the positive transformation value. For soil testing indices with smaller values and better quality, the inverse transformation method is used to take the reciprocal of the soil testing index value as the positive transformation value. For soil testing indices with values closer to the optimal value and better quality, the optimal distance compensation method is used to calculate the distance between the soil testing index value and the preset optimal value, and the distance is converted into a positive transformation value, so that the smaller the distance, the larger the positive transformation value. At the same time, for each functional benchmark value in the benchmark value table, the same positive transformation process is performed according to the preferred type of the index and the preset optimal value to obtain the positive benchmark value. The range standardization method is adopted. For each soil test index, based on the maximum and minimum values of the positive values of the soil test index in all test sections, each positively normalized soil test index is linearly mapped to a unified numerical range to obtain a standardized test value. At the same time, the positively normalized benchmark value of the soil test index is linearly mapped to a unified numerical range according to the same maximum and minimum values to obtain a standardized benchmark value, so that the standardized test value and the standardized benchmark value have a consistent measurement scale. Based on the soil layer depth to which each soil testing index belongs, the standardized benchmark value corresponding to the soil testing index is obtained. When the standardized test value is less than the standardized benchmark value, the difference between the standardized benchmark value and the standardized test value is taken as the degree of deterioration. When the standardized test value is greater than or equal to the standardized benchmark value, the degree of deterioration is recorded as zero. The degradation degree of each soil test index corresponding to each soil layer depth at each test section location is arranged in a preset order to generate a degradation degree sequence corresponding to the test section location. All degradation degree sequences corresponding to the test section locations are arranged according to the test section location, with the test section location as the row and the combination of soil layer depth and soil test index as the column, to construct the plow layer quality state matrix.
[0007] In a preferred embodiment, for the plow pan quality state matrix, the maximum information coefficient and Spearman rank correlation coefficient between any two soil detection indicators are calculated. Based on the maximum information coefficient, indicator pairs with strong nonlinear correlations are identified. Representative indicators are selected from the indicator pairs according to a preset retention indicator selection rule, and redundant indicators are removed to obtain a retention indicator set. The positive or negative synergistic relationship between indicators is determined based on the sign of the Spearman rank correlation coefficient between the indicators in the retention indicator set. A fuzzy measure matrix characterizing the interaction tendency between indicators is constructed based on the retention indicator set and the conflict relationship, specifically including: Read the plow pan quality state matrix and assemble the deterioration degree of each soil test index at each test section position into a soil test index deterioration degree sequence according to the test section position order. For any two soil testing indicators, the deterioration degree sequences corresponding to the two soil testing indicators are mapped to a two-dimensional plane. The deterioration degree value range of the two soil testing indicators is divided into multiple grids according to a preset number of grids. The number of data points falling into each grid is counted. The mutual information value is calculated based on the number of data points. After normalizing the mutual information value to a preset interval, the maximum information coefficient between the two soil testing indicators is obtained. At the same time, the deterioration degree of the two soil testing indicators at each testing section is sorted according to the numerical value. Each deterioration degree is assigned a sorting number. The Spearman rank correlation coefficient is calculated based on the difference between the sorting numbers of the two soil testing indicators at the same testing section. Soil detection index pairs with a maximum information coefficient greater than a preset strong correlation threshold are identified as index pairs with strong nonlinear correlation. According to the preset retention index selection rules, representative indicators are selected from the index pairs with strong nonlinear correlation, redundant indicators that are not selected are removed, and soil detection indicators that do not appear in the index pairs with strong nonlinear correlation are retained to obtain the retention index set. For any two soil testing indicators in the retained indicator set, when the Spearman rank correlation coefficient between the two soil testing indicators is positive, it is determined that the two soil testing indicators have a positive synergistic relationship; when the Spearman rank correlation coefficient is negative, it is determined that the two soil testing indicators have an inverse conflict relationship. A fuzzy measure matrix is constructed to characterize the interaction tendency among indicators based on the positive synergy and negative conflict relationships among the retained indicator set and soil testing indicators. The elements in the fuzzy measure matrix represent the comprehensive interaction tendency of the corresponding soil testing indicator set. The fuzzy measure of a single soil testing indicator is determined by the sum of the absolute values of the Spearman rank correlation coefficients between the soil testing indicator and the remaining soil testing indicators in the retained indicator set. The fuzzy measure of a soil testing indicator set consisting of at least two soil testing indicators is determined by the positive synergy and negative conflict relationships between all pairs of soil testing indicators within the soil testing indicator set.
[0008] In a preferred embodiment, the Shapley value of each indicator in the retained indicator set is calculated as the indicator importance based on the fuzzy measure matrix, specifically including: Read the degree of deterioration of each soil detection index in the retained index set and the plow bottom quality state matrix at each detection section. The fuzzy measure matrix contains the fuzzy measure values of all index subsets in the retained index set. Based on the fuzzy measure matrix, for each soil detection index in the retained index set, all index subsets in the retained index set that do not contain soil detection indexes are traversed. For each index subset traversed, the fuzzy measure value of the index subset is read from the fuzzy measure matrix, and then the fuzzy measure value of the new index set formed after adding the soil detection index to the index subset is read. The difference between the two fuzzy measure values is taken as the marginal contribution value corresponding to the index subset. Calculate the factorial of the number of indicators contained in the indicator subset, the factorial of the number of indicators that do not include soil testing indicators and are not in the indicator subset, and the factorial of the total number of indicators in the retained indicator set. Divide the product of the factorial of the number of indicators contained in the indicator subset and the factorial of the number of indicators that do not include soil testing indicators and are not in the indicator subset by the factorial of the total number of indicators in the retained indicator set to obtain the cumulative weight corresponding to the marginal contribution value. Sum the products of the marginal contribution value of the soil testing indicator under all indicator subsets and the corresponding cumulative weight to obtain the Shapley value of the soil testing indicator. The Shapley value is used as the importance of the soil testing indicator.
[0009] In a preferred embodiment, the degree to which the average degradation of each indicator exceeds the overall average degradation is input into the variable-weight equilibrium function to perform a weak link reinforcement correction on the indicator importance. The fuzzy measure matrix is then reconstructed based on the corrected indicator importance to obtain the corrected fuzzy measure matrix, specifically including: Calculate the average degree of deterioration of all soil testing indicators in the retained indicator set at all testing sections, as the total average degree of deterioration. Then calculate the average degree of deterioration of each soil testing indicator at all testing sections, as the indicator average degree of deterioration of the soil testing indicator. When the indicator average degree of deterioration of the soil testing indicator is greater than the total average degree of deterioration, the difference between the indicator average degree of deterioration and the total average degree of deterioration is taken as the degree of deviation of the soil testing indicator. When the indicator average degree of deterioration of the soil testing indicator is less than or equal to the total average degree of deterioration, the degree of deviation of the soil testing indicator is set to zero. When the sum of the deviations of all soil test indicators is zero, no correction is made to the importance. The corrected importance of each soil test indicator is kept as the importance before correction, and the importance is used as the final corrected importance. When the sum of the deviations of all soil testing indicators is greater than zero, the deviation of each soil testing indicator is input into the variable weight equilibrium function. The variable weight equilibrium function divides the deviation of each soil testing indicator by the sum of all deviations to obtain the deviation ratio of the soil testing indicator. The deviation ratio is used as the exponent to perform a power operation on the natural constant to obtain the correction coefficient of the soil testing indicator. The importance of each soil testing indicator is multiplied by the correction coefficient to obtain the corrected importance of the soil testing indicator. All corrected importance values are normalized so that the sum of the corrected importance values of each soil testing indicator is equal to the sum of the importance values of each soil testing indicator before correction. The normalized importance value is used as the final corrected importance value. The fuzzy measure value of each individual soil test index in the fuzzy measure matrix is corrected to the final corrected importance of the corresponding soil test index. For a set of soil test indices consisting of at least two soil test indices, the ratio of the sum of the final corrected importance of each soil test index in the set to the sum of the importance before correction is calculated. The ratio is multiplied by the fuzzy measure value of the set of soil test indices in the fuzzy measure matrix to obtain the corrected fuzzy measure value of the set of soil test indices. At the same time, the positive synergistic relationship or the negative conflict relationship between soil test indices in the set of soil test indices remains unchanged. The reconstructed fuzzy measure matrix is used as the corrected fuzzy measure matrix.
[0010] In a preferred embodiment, for each detection section, based on the modified fuzzy measure matrix, the degradation degree of each indicator is converted into a positive quality score. The Schogai integral is then used to aggregate the positive quality scores of each indicator of the section into a plowshare layer to construct a comprehensive evaluation index, specifically including: Read the modified fuzzy metric matrix and the plow bottom quality state matrix to retain the degree of deterioration of each soil detection index in the index set at each detection section location. For each detection section location, extract the degree of deterioration of each soil detection index in the index set at the detection section location from the plow bottom quality state matrix. According to the preset positive quality score conversion rules, the degree of deterioration of each soil test index at the test section is converted into a positive quality score. The positive quality score conversion rules are as follows: when the degree of deterioration is zero, the positive quality score takes the maximum value within the preset range; when the degree of deterioration reaches the preset upper limit of deterioration, the positive quality score takes the minimum value within the preset range; and when the degree of deterioration is between zero and the preset upper limit of deterioration, the positive quality score decreases linearly as the degree of deterioration increases. Using the Schottky integral, the positive quality scores of all soil test indicators in the retained index set at the test section location are arranged in descending order of value to obtain the positive quality score sequence, and the soil test indicator corresponding to each positive quality score is recorded. For any two adjacent positive quality scores in the positive quality score sequence, the difference between the positive quality score that comes first and the positive quality score that comes later is taken as the first difference value corresponding to the adjacent position. The set of soil test indicators corresponding to the first difference value consists of the soil test indicators corresponding to the positive quality score that comes first and all soil test indicators that come before the soil test indicators. The difference between the smallest positive quality score and zero is taken as the last difference. The set of soil test indicators corresponding to the last difference consists of the soil test indicator corresponding to the smallest positive quality score and all soil test indicators listed before the soil test indicator. Read the fuzzy measure values of the soil detection index set corresponding to the first difference and the soil detection index set corresponding to the last difference from the corrected fuzzy measure matrix respectively. Multiply each first difference by the fuzzy measure value of the corresponding soil detection index set to obtain the aggregated contribution value corresponding to the first difference. Multiply the last difference by the fuzzy measure value of the corresponding soil detection index set to obtain the aggregated contribution value corresponding to the last difference. Add the aggregated contribution values corresponding to all first differences and the aggregated contribution values corresponding to the last difference to obtain the comprehensive evaluation index of the plow layer construction at the detection section location. Repeat the above steps for each detection section location to obtain the comprehensive evaluation index of the plow layer construction at all detection section locations.
[0011] In a preferred embodiment, the comprehensive evaluation index of each detection section is mapped to the plowshare subsurface construction quality level of the detection section according to a preset quality level range, specifically including: Read the comprehensive evaluation index of the plow layer at all detection section locations, the corrected fuzzy measure matrix, and the positive quality score of each soil detection index in the retained index set at each detection section location; According to the preset quality grade range, the comprehensive evaluation index of the plow layer construction at each test section location is compared with the boundary values of each quality grade in the preset quality grade range to determine the quality grade of the plow layer construction at the test section location.
[0012] In a preferred embodiment, based on the marginal contribution rate of each indicator to the comprehensive evaluation index of the test section during the Schoker integral calculation process, weak indicators of the test section are identified. Based on the preset correspondence between weak indicators and plow layer construction process parameters, a diagnostic evaluation report is output, including the comprehensive evaluation index, quality level, weak indicators, and corresponding process adjustment suggestions for each test section. Specifically, this includes: For each detection section location, the positive quality scores of all soil detection indicators in the retained indicator set are arranged in descending order of value to obtain a positive quality score sequence. The soil detection indicator corresponding to each sorting position in the positive quality score sequence is recorded. When there are soil detection indicators with the same positive quality score, the sorting position is determined according to the preset indicator sorting order. According to the calculation process of the Schogey integral, for each detection section position, for each sorting position in the positive quality score sequence, when the sorting position is the first sorting position, the set of soil detection indicators corresponding to the sorting position is recorded as the set composed of the soil detection indicators corresponding to the sorting position, and the fuzzy measure value before the sorting position is recorded as zero. When the sorting position is not the first sorting position, the set of soil detection indicators corresponding to the sorting position is recorded as the set composed of all soil detection indicators from the first sorting position to the sorting position, and the fuzzy measure value of the set of soil detection indicators corresponding to the sorting position before the sorting position is read from the corrected fuzzy measure matrix. Read the fuzzy measure value of the soil test index set corresponding to the sorting position from the corrected fuzzy measure matrix, subtract the fuzzy measure value before the sorting position from the fuzzy measure value of the soil test index set corresponding to the sorting position to obtain the fuzzy measure increment corresponding to the sorting position, and multiply the positive quality score of the soil test index corresponding to the sorting position by the fuzzy measure increment corresponding to the sorting position to obtain the marginal contribution value of the soil test index corresponding to the sorting position. Divide the marginal contribution value corresponding to the ranking position by the comprehensive evaluation index of the plow layer at the detection section position to obtain the marginal contribution rate of the soil detection index corresponding to the ranking position. Soil testing indicators whose marginal contribution rate at each testing section location does not meet the preset weak indicator judgment standard are identified as weak indicators at the testing section location. Based on the preset weak indicator and the plow layer construction process parameters, corresponding plow layer construction process adjustment suggestions are matched for each weak indicator. The output is a diagnostic evaluation report containing the comprehensive evaluation index of the plow layer construction at each test section location, the quality grade of the plow layer construction, the weak indicators, and the corresponding suggestions for adjusting the plow layer construction process.
[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. By dynamically coupling multi-index degradation characteristics and fuzzy measurement interaction analysis, this method achieves accurate capture and collaborative evaluation of the micro and macroscopic states of plow pan formation quality. Based on the positive transformation, standardization, and degradation degree calculation of soil detection indicators at multiple soil depths at the detection section location, a plow pan quality state matrix is constructed. This effectively identifies strong nonlinear correlations and positive synergies or negative conflicts among mechanical support indicators, moisture regulation indicators, and structural integrity indicators. Combined with the joint analysis of the maximum information coefficient and Spearman rank correlation coefficient, the complex interactions of indicators such as soil penetration resistance, soil saturated hydraulic conductivity, and soil porosity are physically integrated into the fuzzy measurement matrix. Compared to existing technologies, this method can perceive the dynamic changes in the interaction tendencies between various indicators in real time during plow pan formation. Through Shapley value importance calculation and variable weight equilibrium function correction, it significantly improves the reflectivity of weakest indicators in comprehensive evaluation, suppressing evaluation bias caused by neglecting the coupling relationships between indicators. Furthermore, based on the corrected fuzzy measurement matrix and Schoker integral aggregation, robust quantification of the comprehensive evaluation index of plow pan formation is achieved.
[0014] 2. By identifying weak indicators and dynamically adapting process parameters based on the marginal contribution rate of the Schogey integral, this method solves the problem of traditional evaluation methods that only output comprehensive scores but cannot locate key obstacle factors. It constructs a marginal contribution rate based on the correspondence between the positive quality score sequence of each soil testing indicator and the corrected fuzzy measure value. This rate can match the degree of indicator degradation with the quality shortcomings of the plow pan construction in real time, achieving the coordinated output of comprehensive evaluation index, quality level, weak indicators, and process adjustment suggestions. Through the pre-set criteria for weak indicator judgment and the coordinated constraints of the correspondence between weak indicators and plow pan construction process parameters, it ensures that the evaluation results always match the actual plow pan construction quality status and indicator shortcomings, avoiding blind control measures due to unclear indicator degradation mechanisms. While maintaining the accuracy of plow pan evaluation, it significantly reduces the risk of plow pan construction quality problems caused by missing evaluation information, ensuring the long-term and stable improvement of plow pan construction quality under agricultural farming conditions. Attached Figure Description
[0015] 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.
[0016] Figure 1 This is a flowchart of a comprehensive evaluation method for the construction quality of plowshare layers based on multi-index fusion, according to the present invention. Detailed Implementation
[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0018] Example 1: Figure 1 A flowchart of a comprehensive evaluation method for the quality of plowshare substructure construction based on multi-index fusion is provided, which includes the following steps: S1. Receive soil test index data collected on-site according to the test section location at the site of the plow bottom construction quality acceptance evaluation. The soil test index includes mechanical support index, moisture control index and structural integrity index at different soil depths. Perform positive and standardization processing on each soil test index, calculate the degree of deterioration of each soil test index relative to the preset corresponding functional benchmark value, generate a degree of deterioration sequence, arrange the degree of deterioration sequence according to the test section location, and construct the plow bottom quality state matrix. S2. Read the plow bottom quality state matrix, calculate the maximum information coefficient and Spearman rank correlation coefficient between any two soil detection indicators, identify indicator pairs with strong nonlinear correlation based on the maximum information coefficient, select representative indicators from the indicator pairs according to the preset retention indicator selection rules and remove redundant indicators to obtain the retention indicator set, determine the positive synergy or negative conflict relationship between indicators based on the sign of the Spearman rank correlation coefficient between each indicator in the retention indicator set, and construct a fuzzy measure matrix to characterize the interaction tendency between indicators based on the retention indicator set and the conflict relationship. S3. Based on the fuzzy measure matrix, calculate the Shapley value of each indicator in the retained indicator set as the indicator importance. Input the degree to which the average deterioration of each indicator exceeds the total average deterioration into the variable weight equilibrium function to perform short-board strengthening correction on the indicator importance. Reconstruct the fuzzy measure matrix based on the corrected indicator importance to obtain the corrected fuzzy measure matrix. S4. For each detection section, based on the corrected fuzzy measure matrix, the degradation degree of each indicator is converted into a positive quality score. The positive quality scores of each indicator of the section are aggregated into the plowshare of the detection section to construct a comprehensive evaluation index. S5. Based on the preset quality level range, map the comprehensive evaluation index of each test section to the quality level of the plowshare substructure of the test section. Based on the marginal contribution rate of each indicator to the comprehensive evaluation index of the test section during the Shape integral calculation, identify the weak indicators of the test section. Based on the preset correspondence between the weak indicators and the plowshare substructure construction process parameters, output a diagnostic evaluation report containing the comprehensive evaluation index, quality level, weak indicators and corresponding process adjustment suggestions for each test section.
[0019] The system receives soil testing data collected at the testing section locations during the on-site quality acceptance evaluation of the plow pan substructure. These soil testing data include mechanical support indicators, moisture regulation indicators, and structural integrity indicators at different soil depths. Each soil testing indicator is normalized and standardized, and the degree of degradation of each indicator relative to a preset corresponding functional benchmark value is calculated, generating a degradation degree sequence. This sequence is then arranged according to the testing section location to construct a plow pan substructure quality state matrix. The specific implementation is as follows: The soil testing data collected on-site at the test section locations are received for the quality acceptance evaluation of the plow layer construction. Five test section locations are determined at a preset interval of 10 meters along the extension direction of the plow layer. At each test section location, the soil layer depth is divided at a preset depth interval of 5 centimeters. The soil layer depth is divided into four levels from 20 centimeters to 40 centimeters below the surface, namely 20 centimeters to 25 centimeters, 25 centimeters to 30 centimeters, 30 centimeters to 35 centimeters, and 35 centimeters to 40 centimeters.
[0020] Undisturbed and disturbed soil samples were collected at each soil depth, and soil testing indicators were measured. These indicators included mechanical support, moisture control, and structural integrity. The mechanical support indicator was soil penetration resistance, measured using a portable penetration meter. The moisture control indicators were soil saturated hydraulic conductivity and soil moisture content, measured using the ring cutter method and the oven-drying method, respectively. The structural integrity indicators were soil bulk density and soil porosity, with soil porosity calculated from the bulk density measured by the ring cutter method.
[0021] A benchmark value table is established in advance according to each soil testing index. Under each soil testing index, functional benchmark value, preferred type and preset optimal value are set according to different soil depths. The preferred type of soil penetration resistance and soil bulk density is that the smaller the value, the better the quality. The preferred type of soil saturated hydraulic conductivity and soil moisture content is that the closer the value is to the optimal value, the better the quality. The preferred type of soil porosity is that the larger the value, the better the quality.
[0022] For indicators where the smaller or larger numerical value is considered better, the preset optimal value is not included in the forward pass calculation and is only recorded in the baseline value table. For indicators where the closer the numerical value is to the optimal value, the better, the preset optimal value is included in the forward pass calculation. Specifically, the functional baseline values for soil penetration resistance at the four soil depths are set to 1800 kPa, 2000 kPa, 2200 kPa, and 2400 kPa, respectively; the functional baseline values for soil bulk density at the four soil depths are set to 1.40 g / cm³, 1.45 g / cm³, 1.50 g / cm³, and 1.55 g / cm³, respectively; and the soil saturation... The functional benchmark values for hydraulic conductivity at four soil depths were set to 0.50 cm / h, 0.45 cm / h, 0.40 cm / h, and 0.35 cm / h, respectively, with the preset optimal values set to 0.80 cm / h, 0.75 cm / h, 0.70 cm / h, and 0.65 cm / h, respectively. The functional benchmark values for soil moisture content at four soil depths were set to 25%, 24%, 23%, and 22%, respectively, with the preset optimal values set to 30%, 29%, 28%, and 27%, respectively. The functional benchmark values for soil porosity at four soil depths were set to 45%, 44%, 43%, and 42%, respectively.
[0023] For each soil testing index, a positive transformation is performed according to the corresponding eigenvalue type. For soil porosity, where a larger value indicates better quality, the index value is directly retained as the positive transformation value. For soil penetration resistance and soil bulk density, where a smaller value indicates better quality, the reciprocal transformation method is used, taking the reciprocal of the index value as the positive transformation value. For soil saturated hydraulic conductivity and soil moisture content, where a value closer to the optimal value indicates better quality, the optimal distance compensation method is used. First, the absolute difference between the index value and the preset optimal value is calculated as the distance. Then, the maximum distance of the index in all testing sections is found. The difference between the maximum distance and the maximum distance is then divided by the maximum distance. When the maximum distance is greater than 0, the difference between the maximum distance and the maximum distance is divided by the maximum distance to obtain the positive transformation value. The smaller the distance, the larger the positive transformation value. When the distance is 0, the positive transformation value is 1. When the maximum distance is 0, the positive transformation value of the index at this testing section is directly set to 1.
[0024] Simultaneously, for each functional benchmark value in the benchmark value table, the same positive transformation process is performed according to the preferred type and preset optimal value of the soil testing index to obtain a positive benchmark value. For soil testing indices using the optimal distance compensation method, the positive transformation calculation of the functional benchmark value uses the same maximum distance as the index's test value. When the maximum distance is 0, the positive benchmark value is also set to 1. Then, using the range standardization method, for each soil testing index, the maximum and minimum values of the positive transformation test values at all test section locations are found. Each positive transformation test value is linearly mapped to a unified numerical range of 0 to 1 to obtain a standardized test value, ensuring that the standardized test value is within the range of 0 to 1. The standardized benchmark value is obtained by linearly mapping the positive benchmark value to the same numerical range according to the same maximum and minimum values. When the calculated result of the positive benchmark value exceeds the range of 0 to 1, the calculated result is retained without truncation. Specifically, the standardized value of a positive detection value is equal to the positive detection value minus the minimum positive detection value of the index, and then divided by the difference between the maximum and minimum values. When the maximum and minimum values of the positive detection values of the index are equal at all detection sections, the standardized detection value of the index is 1, and the standardized value of the positive benchmark value is also 1. Thus, the standardized detection value and the standardized benchmark value have a consistent measurement scale.
[0025] Based on the soil layer depth to which each soil testing index belongs, the corresponding standardized benchmark value is obtained from the benchmark value table. When the standardized test value is less than the standardized benchmark value, the difference between the standardized benchmark value and the standardized test value is taken as the degree of deterioration of the index at that soil layer depth. When the standardized test value is greater than or equal to the standardized benchmark value, the degree of deterioration is recorded as 0.
[0026] Then, the deterioration degree of each soil test index corresponding to each soil layer depth at each test section location is arranged in a preset order to generate a deterioration degree sequence corresponding to that test section location. The preset order is to first arrange the soil layer depth from smallest to largest, and then arrange them in the order of soil penetration resistance, soil saturated hydraulic conductivity, soil moisture content, soil bulk density, and soil porosity within each soil layer depth. All the deterioration degree sequences corresponding to the test section locations are arranged in sequence according to the test section locations, with the test section location as the row and the combination of soil layer depth and soil test index as the column, to construct the plow pan quality state matrix.
[0027] For the plow pan quality state matrix, calculate the maximum information coefficient and Spearman rank correlation coefficient between any two soil detection indicators. Identify indicator pairs with strong nonlinear correlations based on the maximum information coefficient. Select representative indicators from these pairs according to a pre-defined retention indicator selection rule and remove redundant indicators to obtain a retention indicator set. Determine the positive or negative synergistic relationship between indicators based on the sign of the Spearman rank correlation coefficient between each indicator in the retention indicator set. Construct a fuzzy measure matrix characterizing the interaction tendency between indicators based on the retention indicator set and the conflict relationship. The specific implementation is as follows: The read plow pan quality state matrix has 5 rows, corresponding to 5 detection cross-section locations. The columns consist of 4 soil depths and 5 soil testing indicators, totaling 20 columns. For each soil testing indicator, the degree of degradation at the corresponding soil depth is extracted sequentially according to the detection cross-section location from smallest to largest, and within each cross-section location, according to the soil depth from shallowest to deepest, forming a degradation degree sequence for that indicator. Since there are 5 detection cross-section locations and 4 soil depths, each soil testing indicator's degradation degree sequence contains 20 values.
[0028] Then, for any two soil testing indicators, the deterioration sequence of the two indicators is mapped to a two-dimensional plane. Specifically, the minimum to maximum value range of the first soil testing indicator among 20 values is divided into 5 equal segments, and the minimum to maximum value range of the second soil testing indicator among 20 values is divided into 5 equal segments, forming a grid of 5 by 5, totaling 25 grids.
[0029] To calculate the mutual information value, count the number of data points falling into each grid for each of the 20 data points. For each grid, count the number of data points falling into that grid and record this number as the number of falling points. The joint probability of that grid is the number of falling points divided by 20. The total number of data points in the interval where the first soil test indicator is located is recorded as the number of points in the first interval. The marginal probability of the first soil test indicator is the number of points in the first interval divided by 20. The total number of data points in the interval where the second soil test indicator is located is recorded as the number of points in the second interval. The marginal probability of the second indicator is the number of points in the second interval divided by 20.
[0030] If the number of points falling into a grid is greater than 0, a product is accumulated. This product is the number of points falling into a grid divided by 20, multiplied by a logarithmic value with a base of 2. The argument of this logarithm is calculated as follows: first, calculate the number of points falling into a grid divided by 20 to obtain the first quotient; then, calculate the product of the number of points in the first interval divided by 20 and the number of points in the second interval divided by 20; then, divide the first quotient by this product to obtain the argument of the logarithm. After accumulating all grids, the mutual information value is obtained. The mutual information value is divided by a logarithmic value with a base of 2 and an argument of 5 to obtain the maximum information coefficient between the two indicators.
[0031] Simultaneously, for the two soil testing indicators, their respective 20 degradation values are sorted in ascending order, with the minimum value assigned the sorting number 1 and the maximum value assigned the sorting number 20. If there are identical values, the average sorting number is taken to obtain the sorting number corresponding to each data point. For each pair of data points, the difference between the sorting numbers of the two soil testing indicators is calculated. The squares of all 20 sorting number differences are added together to obtain a sum of squares. A fraction is subtracted from 1. The numerator of this fraction is 6 multiplied by the sum of squares, and the denominator of this fraction is 20 multiplied by the product of a difference, which is 20 squared minus 1. The result is the Spearman rank correlation coefficient between the two indicators. The above process is repeated for all soil testing indicators in pairs to obtain the maximum information coefficient and Spearman rank correlation coefficient of any two soil testing indicators.
[0032] A preset strong correlation threshold of 0.6 is set. Indicator pairs with a maximum information coefficient greater than 0.6 are identified as having a strong nonlinear correlation. In this embodiment, based on measured data, the maximum information coefficient between soil penetration resistance and soil bulk density is 0.71, and the maximum information coefficient between soil saturated hydraulic conductivity and soil moisture content is 0.65. The maximum information coefficients of any other two soil testing indicators do not exceed 0.6. Therefore, the strongly nonlinearly correlated indicator pairs are soil penetration resistance and soil bulk density, and soil saturated hydraulic conductivity and soil moisture content. According to the preset retention indicator selection rules, for each strongly nonlinearly correlated indicator pair, the average deterioration degree of the two soil testing indicators is calculated across all 20 values. The pair with an average deterioration degree greater than the average deterioration degree of the other soil testing indicator is considered a strong nonlinearly correlated indicator pair. The soil testing indicators with the highest values are retained as representative indicators. Soil testing indicators with average deterioration values lower than the average deterioration values of another soil testing indicator are removed as redundant indicators. If the average deterioration values of two soil testing indicators are equal, the soil testing indicator with the higher preset order is retained. After calculation, the average deterioration value of soil penetration resistance is greater than the average deterioration value of soil bulk density, so soil penetration resistance is retained and soil bulk density is removed. The average deterioration value of soil saturated hydraulic conductivity is greater than the average deterioration value of soil moisture content, so soil saturated hydraulic conductivity is retained and soil moisture content is removed. Soil porosity, which does not appear in any strongly nonlinear correlation indicator pair, is retained. The final retained indicator set consists of three soil testing indicators: soil penetration resistance, soil saturated hydraulic conductivity, and soil porosity.
[0033] For any two soil testing indicators in the retained indicator set, a positive Spearman rank correlation coefficient indicates a positive synergistic relationship, while a negative value indicates an inverse conflict relationship. In this embodiment, the Spearman rank correlation coefficient between soil penetration resistance and soil saturated hydraulic conductivity is -0.35, which is negative, therefore they are inversely conflicting. The Spearman rank correlation coefficient between soil penetration resistance and soil porosity is -0.42, which is also negative, therefore they are inversely conflicting. The Spearman rank correlation coefficient between soil saturated hydraulic conductivity and soil porosity is 0.28, which is positive, therefore they are positively synergistic.
[0034] Then, a fuzzy measure matrix is constructed to characterize the interaction tendency between indicators. Specifically, for each soil detection indicator in the retained indicator set, its individual base number is calculated. The individual base number is equal to 1 plus the sum of the absolute values of the Spearman rank correlation coefficients between the indicator and the other two indicators. The individual base numbers for soil penetration resistance are 1.77, soil saturated hydraulic conductivity is 1.63, and soil porosity is 1.70.
[0035] For any non-empty subset of the retained index set, the unnormalized fuzzy measure is first calculated. Specifically, this is the sum of the individual cardinality of all soil detection indicators within the subset, plus a preset interaction strength coefficient of 0.2 multiplied by the sum of the Spearman rank correlation coefficients between all pairs of indicators within the subset. For example, the unnormalized fuzzy measure of the subset consisting of soil penetration resistance and soil saturated hydraulic conductivity is 3.33, and the unnormalized fuzzy measure of the entire set consisting of soil penetration resistance, soil saturated hydraulic conductivity, and soil porosity is 5.002. Divide the unnormalized fuzzy measure of each non-empty subset by the unnormalized fuzzy measure of the entire set to obtain the fuzzy measure of that subset. Set the fuzzy measure of the empty set to 0 and the fuzzy measure of the entire set to 1. Calculate the fuzzy measures of all non-empty subsets of the retained index set according to the above method. Arrange all non-empty subsets in a preset order according to the number of indicators contained in the subset from fewest to most, and for the same number of indicators, arrange them according to the soil testing indicators. Record the fuzzy measure values of each subset in this order to obtain an ordered sequence of fuzzy measures, which is the fuzzy measure matrix.
[0036] Based on the fuzzy measure matrix, the Shapley value of each indicator in the retained indicator set is calculated as the indicator importance. The specific implementation is as follows: The fuzzy measure matrix, the retained index set, and the plow pan quality state matrix are used to read the degree of deterioration of each soil test index at each test section location within the retained index set. The retained index set consists of three soil test indices: soil penetration resistance, soil saturated hydraulic conductivity, and soil porosity. In the fuzzy measure matrix, the fuzzy measure value of the empty set is 0. In the subset consisting of a single soil test index, the fuzzy measure value of soil penetration resistance is 0.3539, the fuzzy measure value of soil saturated hydraulic conductivity is 0.3259, and the fuzzy measure value of soil porosity is 0.3399. In the subset consisting of two soil test indices, the fuzzy measure values of soil penetration resistance and soil saturated hydraulic conductivity are 0.6657, the fuzzy measure values of soil penetration resistance and soil porosity are 0.6769, and the fuzzy measure value of soil saturated hydraulic conductivity and soil porosity are 0.6769. The fuzzy measure value of the entire set consisting of all three soil test indices is 1.
[0037] For each soil testing indicator in the retained indicator set, iterate through all indicator subsets that do not contain that soil testing indicator. For each traversed indicator subset, read the fuzzy measure value of that indicator subset from the fuzzy measure matrix, and then read the fuzzy measure value of the new indicator set formed after adding the soil testing indicator to that indicator subset. The difference between these two fuzzy measure values is taken as the marginal contribution value corresponding to that indicator subset. The cumulative weight is calculated by multiplying the factorial of the number of indicators contained in the indicator subset by the factorial of the number of indicators that do not contain the soil testing indicator and are not in the indicator subset, and then dividing by the factorial of the total number of indicators in the retained indicator set. Since the retained indicator set has a total of 3 soil testing indicators, when the indicator subset contains 0 indicators, its factorial... The factorial of 0 equals 1. The number of indicators that do not contain the soil testing index and are not in the subset of indicators is 2. The factorial of 2 equals 2. The factorial of the total number of indicators, 3, equals 6. Therefore, the weight is 2 divided by 6, which equals one-third. When the subset of indicators contains 1 indicator, the factorial of the number of indicators in the subset is 1. The number of indicators that do not contain the soil testing index and are not in the subset of indicators is 1. The factorial of 1 equals 1. Dividing by the total factorial of 6, we get a weight of one-sixth. When the subset of indicators contains 2 indicators, the factorial of the number of indicators in the subset is 2. The number of indicators that do not contain the soil testing index and are not in the subset of indicators is 0. The factorial of 0 equals 1. Dividing by the total factorial of 6, we get a weight of one-third.
[0038] Taking soil penetration resistance as an example, there are four subsets of indicators that do not include soil penetration resistance: the empty set, soil saturated hydraulic conductivity, soil porosity, and soil saturated hydraulic conductivity and soil porosity. The marginal contribution value corresponding to the empty set is the fuzzy measure value of soil penetration resistance (0.3539) minus the fuzzy measure value of the empty set (0), which equals 0.3539. The cumulative weight is one-third. The marginal contribution value corresponding to soil saturated hydraulic conductivity is the fuzzy measure value of soil penetration resistance and soil saturated hydraulic conductivity (0.6657) minus the fuzzy measure value of soil saturated hydraulic conductivity (0.32). 59, equal to 0.3398, with a cumulative weight of one-sixth. The marginal contribution value corresponding to soil porosity is 0.3370, which is the fuzzy measure value of soil penetration resistance and soil porosity minus the fuzzy measure value of soil porosity of 0.6769. The cumulative weight is one-sixth. The marginal contribution value corresponding to the subset of indicators composed of soil saturated hydraulic conductivity and soil porosity is 0.3231, which is the fuzzy measure value of the whole set of 1 minus the fuzzy measure value of soil saturated hydraulic conductivity and soil porosity of 0.6769. The cumulative weight is one-third.
[0039] The products of the four marginal contribution values and their corresponding cumulative weights are added together: 0.3539 multiplied by one-third, 0.3398 multiplied by one-sixth, 0.3370 multiplied by one-sixth, and 0.3231 multiplied by one-third. This yields a Shapley value of 0.3385 for soil penetration resistance, which is the importance of soil penetration resistance. Similarly, the Shapley value for soil saturated hydraulic conductivity is calculated to be 0.3245, meaning the importance of soil saturated hydraulic conductivity is 0.3245. The Shapley value for soil porosity is 0.3371, meaning the importance of soil porosity is 0.3371. The sum of the three importance values is 1.
[0040] The degree to which the average degradation of each indicator exceeds the overall average degradation is input into the variable-weight equilibrium function to correct the weakness in the indicator importance. The fuzzy measure matrix is then reconstructed based on the corrected indicator importance, resulting in the corrected fuzzy measure matrix. The specific implementation is as follows: The average and overall average degradation of each soil testing index were calculated. The degradation of each soil testing index within the retained index set was read from the plow pan quality state matrix at each testing section location. The average degradation of soil penetration resistance across all 5 testing sections and 4 soil depths (a total of 20 values) was 0.45; the average degradation of soil saturated hydraulic conductivity was 0.38; and the average degradation of soil porosity was 0.32. The overall average degradation of all soil testing indices in the retained index set across all testing sections was calculated as 0.45 + 0.38 + 0.32, divided by 3, resulting in 0.3833.
[0041] Since the average degradation degree of soil penetration resistance (0.45) is greater than the total average degradation degree (0.3833), the deviation of soil penetration resistance is 0.45 minus 0.3833, which equals 0.0667. The average degradation degree of soil saturated hydraulic conductivity (0.38) is less than the total average degradation degree (0.3833), therefore the deviation of soil saturated hydraulic conductivity is 0. The average degradation degree of soil porosity (0.32) is less than the total average degradation degree (0.3833), therefore the deviation of soil porosity is 0. The sum of all deviations is 0.0667, which is greater than 0. Therefore, the importance is corrected according to the variable weight equilibrium function. Specifically... The deviation percentage of each soil test index is obtained by dividing the deviation of each index by the sum of all deviations. The deviation percentage of soil penetration resistance is 0.0667 divided by 0.0667 equals 1, the deviation percentage of soil saturated hydraulic conductivity is 0, and the deviation percentage of soil porosity is 0. The natural constant is raised to the power of the deviation percentage. The value of the natural constant is approximately 2.7183. When the exponent is 1, the result of the power operation is 2.7183, and when the exponent is 0, the result of the power operation is 1. Thus, the correction coefficients for soil penetration resistance, soil saturated hydraulic conductivity, and soil porosity are 2.7183, 1, and 1, respectively.
[0042] The importance of each soil test index is multiplied by the corresponding correction factor to obtain the corrected importance. The corrected importance of soil penetration resistance is 0.3385 multiplied by 2.7183, which equals 0.9200. The corrected importance of soil saturated hydraulic conductivity is 0.3245 multiplied by 1, which equals 0.3245. The corrected importance of soil porosity is 0.3371 multiplied by 1, which equals 0.3371. The sum of the three corrected importance values is 1.5816. Before correction, the sum of the importance of each soil test index was 1. Therefore, all corrected importance values were normalized so that the sum of the corrected importance values of each soil test index was equal to 1. The normalization coefficient was 1 divided by 1.5816, which equaled 0.6323. Each corrected importance value was multiplied by the normalization coefficient to obtain the final corrected importance value. Specifically, soil penetration resistance was 0.9200 multiplied by 0.6323, which equaled 0.5817; soil saturated hydraulic conductivity was 0.3245 multiplied by 0.6323, which equaled 0.2052; and soil porosity was 0.3371 multiplied by 0.6323, which equaled 0.2131. The sum of the three final corrected importance values was 1.
[0043] The fuzzy measure matrix is reconstructed based on the final corrected importance values to obtain the corrected fuzzy measure matrix. The fuzzy measure values of each individual soil test index in the fuzzy measure matrix are then corrected to the final corrected importance values of the corresponding soil test index. That is, the fuzzy measure value of soil penetration resistance is corrected to 0.5817, the fuzzy measure value of soil saturated hydraulic conductivity is corrected to 0.2052, the fuzzy measure value of soil porosity is corrected to 0.2131, and the fuzzy measure value of the empty set remains 0.
[0044] For an index set consisting of two soil testing indicators, calculate the ratio of the sum of the final importance of each soil testing indicator after correction to the sum of the importance before correction. Then multiply the ratio by the fuzzy measure value of the index set in the fuzzy measure matrix to obtain the corrected fuzzy measure value. In the index set consisting of soil penetration resistance and soil saturated hydraulic conductivity, the sum of importance before correction is 0.3385 plus 0.3245 equals 0.6630, and the sum of importance after correction is 0.5817 plus 0.2052 equals 0.7869. The ratio is 0.7869 divided by 0.6630 equals 1.1869. The original fuzzy measure value 0.6657 multiplied by 1.1869 equals 0.7902.
[0045] In the set of indicators consisting of soil penetration resistance and soil porosity, the sum of importance before correction is 0.3385 plus 0.3371 equals 0.6756, and the sum of importance after correction is 0.5817 plus 0.2131 equals 0.7948. The ratio is 0.7948 divided by 0.6756 equals 1.1764. The original fuzzy measure value of 0.6769 multiplied by 1.1764 equals 0.7963.
[0046] In the set of indicators consisting of soil saturated hydraulic conductivity and soil porosity, the sum of importance before correction is 0.3245 plus 0.3371 equals 0.6616, and the sum of importance after correction is 0.2052 plus 0.2131 equals 0.4183. The ratio is 0.4183 divided by 0.6616 equals 0.6322. The original fuzzy measure value of 0.6769 multiplied by 0.6322 equals 0.4279.
[0047] The total set consisting of all three soil testing indicators has a sum of importance of 1 before correction and a sum of importance of 1 after correction, with a ratio of 1. Therefore, the fuzzy measure value of the total set remains 1. During the reconstruction process, the original positive synergistic relationship and negative conflict relationship between soil testing indicators within each indicator set remain unchanged, thus obtaining the corrected fuzzy measure matrix.
[0048] For each detection section, based on the corrected fuzzy measure matrix, the degradation degree of each indicator is converted into a positive quality score. The Schogai integral is then used to aggregate the positive quality scores of each indicator in the section into a plowshare layer for constructing a comprehensive evaluation index. The specific implementation is as follows: The deterioration degree of each soil test index at each test section location is read from the corrected fuzzy measure matrix and the plow pan quality state matrix. The retained index set consists of three soil test indices: soil penetration resistance, soil saturated hydraulic conductivity, and soil porosity. In the corrected fuzzy measure matrix, the fuzzy measure value of the empty set is 0. In the subset consisting of a single soil test index, the fuzzy measure value of soil penetration resistance is 0.5817, the fuzzy measure value of soil saturated hydraulic conductivity is 0.2052, and the fuzzy measure value of soil porosity is 0.2131. In the subset consisting of two soil test indices, the fuzzy measure values of soil penetration resistance and soil saturated hydraulic conductivity are 0.7902, the fuzzy measure values of soil penetration resistance and soil porosity are 0.7963, and the fuzzy measure value of soil saturated hydraulic conductivity and soil porosity is 0.4279. The fuzzy measure value of the entire set consisting of all three soil test indices is 1.
[0049] For each detection section, the degradation degree of each soil indicator at each soil depth is extracted from the plow pan quality state matrix within the retained indicator set at that detection section location. Since each detection section location corresponds to four soil depths, the arithmetic mean of the degradation degree of the soil indicator at each of the four soil depths is taken as the degradation degree of the soil indicator at that detection section location. The calculated degradation degree of soil penetration resistance at the first detection section location is 0.30, soil saturated hydraulic conductivity is 0.20, and soil porosity is 0.15. The degradation degree of soil penetration resistance at the second detection section location is... The degradation degree of soil saturated hydraulic conductivity is 0.35, and the degradation degree of soil porosity is 0.30 at the third detection section. The degradation degree of soil penetration resistance at the fourth detection section is 0.55, the degradation degree of soil saturated hydraulic conductivity is 0.50, and the degradation degree of soil porosity is 0.40 at the fifth detection section.
[0050] According to the preset positive quality score conversion rules, the degree of deterioration of each soil test index at each test section is converted into a positive quality score. The preset range is 0 to 1, and the preset upper limit of deterioration is 1. That is, when the degree of deterioration is 0, the positive quality score takes the maximum value of 1 in the preset range. When the degree of deterioration reaches the preset upper limit of 1, the positive quality score takes the minimum value of 0 in the preset range. When the degree of deterioration is between 0 and 1, the positive quality score decreases linearly as the degree of deterioration increases. Therefore, the positive quality score is equal to 1 minus the degree of deterioration.
[0051] According to the conversion rule, the positive mass scores for soil penetration resistance, saturated hydraulic conductivity, and soil porosity at the first detection section are 0.70, 0.80, and 0.85, respectively; at the second detection section, the positive mass scores are 0.60, 0.65, and 0.70, respectively; and at the third detection section, the positive mass scores are 0.55, 0.60, 0.65, and 0.70, respectively. The positive mass scores for soil saturated hydraulic conductivity and soil porosity were 0.60 and 0.65, respectively. At the fourth detection section, the positive mass scores for soil penetration resistance were 0.45, 0.50, and 0.60, respectively. At the fifth detection section, the positive mass scores for soil penetration resistance and soil saturated hydraulic conductivity were 0.45, 0.55, and 0.60, respectively.
[0052] The comprehensive evaluation index of the plow layer at each test section location is calculated using the Schogey integral. For each test section location, the positive quality scores of all soil test indicators in the retained index set are arranged in descending order of value, and the soil test indicator corresponding to each positive quality score is recorded.
[0053] Since the positive quality scores of soil porosity are the highest at each detection section, the positive quality scores of soil saturated hydraulic conductivity are all in the middle, and the positive quality scores of soil penetration resistance are all the lowest, the positive quality score sequence is soil porosity, soil saturated hydraulic conductivity, and soil penetration resistance.
[0054] Taking the first detection section as an example, the positive quality score of soil porosity (0.85) and the positive quality score of soil saturated hydraulic conductivity (0.80) are adjacent. The soil detection index corresponding to the preceding positive quality score of 0.85 is soil porosity. There are no soil detection indices preceding soil porosity. Therefore, the first difference corresponding to this adjacent position is 0.85 minus 0.80, which equals 0.05. The set of soil detection indices corresponding to this first difference is a set containing only soil porosity. The fuzzy measure value of this set is read from the corrected fuzzy measure matrix as 0.2131. Multiplying the first difference of 0.05 by the fuzzy measure value of 0.2131 yields the corresponding aggregated contribution value of 0.010655. The positive quality score of 0.80 for soil saturated hydraulic conductivity and the positive quality score of 0.70 for soil penetration resistance are adjacent. The soil test index corresponding to the preceding positive quality score of 0.80 is soil saturated hydraulic conductivity, and the soil test index preceding soil saturated hydraulic conductivity is soil porosity. Therefore, the first difference corresponding to this adjacent position is 0.80 minus 0.70, which equals 0.10. The set of soil test indices corresponding to the first difference is the set composed of soil porosity and soil saturated hydraulic conductivity. The fuzzy measure value of this set is read from the corrected fuzzy measure matrix as 0.4279. Multiplying the first difference of 0.10 by the fuzzy measure value of 0.4279 yields the corresponding aggregate contribution value of 0.04279.
[0055] The soil test index corresponding to the smallest positive quality score of 0.70 is soil penetration resistance. The difference between the smallest positive quality score of 0.70 and 0 is taken as the last difference, i.e., the last difference is 0.70. The set of soil test indices corresponding to the last difference is soil penetration resistance and all soil test indices arranged before soil penetration resistance, i.e., soil porosity, soil saturated hydraulic conductivity and soil penetration resistance. The fuzzy measure value of this set is read from the corrected fuzzy measure matrix as 1. The last difference of 0.70 is multiplied by the fuzzy measure value of 1 to obtain the aggregate contribution value corresponding to the last difference as 0.70.
[0056] The aggregated contribution values corresponding to all the first differences are added together with the aggregated contribution values corresponding to the last differences, i.e., 0.010655 plus 0.04279 plus 0.70, to obtain the comprehensive evaluation index of the plow layer construction at the first detection section location as 0.7534, which is rounded to four decimal places.
[0057] Following the same Schoenberger integral calculation process as the first detection section, the comprehensive evaluation index of the plow pan structure at the remaining detection sections was calculated sequentially. The first difference between soil porosity and soil saturated hydraulic conductivity at the second detection section was 0.05, corresponding to a fuzzy measure value of 0.2131 for aggregated soil porosity and a aggregated contribution value of 0.010655. The first difference between soil saturated hydraulic conductivity and soil penetration resistance was 0.05, corresponding to a fuzzy measure value of 0.4279 for aggregated soil porosity and soil saturated hydraulic conductivity and a aggregated contribution value of 0.021395. The final difference was 0.60, corresponding to a fuzzy measure value of 1 for the entire set and an aggregated contribution value of 0.60. After summing, the comprehensive evaluation index of the plow pan structure at the second detection section was 0.6321, rounded to four decimal places.
[0058] The first difference between soil porosity and soil saturated hydraulic conductivity at the third detection section is 0.05, with a cumulative contribution of 0.010655. The first difference between soil saturated hydraulic conductivity and soil penetration resistance is 0.05, with a cumulative contribution of 0.021395. The last difference is 0.55, with a cumulative contribution of 0.55. After summing, the comprehensive evaluation index of the plow layer structure at the third detection section is 0.5821, rounded to four decimal places.
[0059] The first difference between soil porosity and soil saturated hydraulic conductivity at the fourth detection section is 0.10, with a cumulative contribution of 0.02131. The first difference between soil saturated hydraulic conductivity and soil penetration resistance is 0.05, with a cumulative contribution of 0.021395. The last difference is 0.45, with a cumulative contribution of 0.45. After summing, the comprehensive evaluation index of the plow layer structure at the fourth detection section is 0.4927, rounded to four decimal places.
[0060] The first difference between soil porosity and soil saturated hydraulic conductivity at the fifth detection section is 0.05, with a cumulative contribution of 0.010655. The first difference between soil saturated hydraulic conductivity and soil penetration resistance is 0.10, with a cumulative contribution of 0.04279. The last difference is 0.45, with a cumulative contribution of 0.45. After summing, the comprehensive evaluation index of the plow layer structure at the fifth detection section is 0.5034, rounded to four decimal places.
[0061] Thus far, the comprehensive evaluation indices of the plow layer structure at all five detection sections are 0.7534, 0.6321, 0.5821, 0.4927, and 0.5034, respectively.
[0062] Based on the preset quality level range, the comprehensive evaluation index of each inspection section is mapped to the quality level of the plowshare layer construction of the inspection section. The specific implementation is as follows: The comprehensive evaluation index of the plow layer structure at all five detection sections was read, and the values were 0.7534, 0.6321, 0.5821, 0.4927, and 0.5034, respectively. The corrected fuzzy measure matrix and the positive quality scores of each soil test index within the retained index set at each detection section were read. In the corrected fuzzy measure matrix, the fuzzy measure value of the empty set was 0, the fuzzy measure value of soil penetration resistance was 0.5817, the fuzzy measure value of soil saturated hydraulic conductivity was 0.2052, the fuzzy measure value of soil porosity was 0.2131, the fuzzy measure value of the set consisting of soil penetration resistance and soil saturated hydraulic conductivity was 0.7902, the fuzzy measure value of the set consisting of soil penetration resistance and soil porosity was 0.7963, the fuzzy measure value of the set consisting of soil saturated hydraulic conductivity and soil porosity was 0.4279, and the fuzzy measure value of the entire set consisting of all three soil test indexes was 1.
[0063] The preset quality grade range is as follows: a comprehensive evaluation index greater than or equal to 0.8 is excellent; greater than or equal to 0.6 and less than 0.8 is good; greater than or equal to 0.4 and less than 0.6 is medium; and less than 0.4 is poor. The comprehensive evaluation index of the plowshare substructure at each test section is compared with the above-mentioned limit values item by item. Specifically, the comprehensive evaluation index of the first test section is 0.7534, which is greater than 0.6 and less than 0.8, indicating a good quality grade of the plowshare substructure; the comprehensive evaluation index of the second test section is 0.6321, which is greater than 0.6 and less than 0.8, indicating a good quality grade of the plowshare substructure; the comprehensive evaluation index of the third test section is 0.5821, which is greater than 0.4 and less than 0.6, indicating a medium quality grade of the plowshare substructure; the comprehensive evaluation index of the fourth test section is 0.4927, which is greater than 0.4 and less than 0.6, indicating a medium quality grade of the plowshare substructure; and the comprehensive evaluation index of the fifth test section is 0.5034, which is greater than 0.4 and less than 0.6, indicating a medium quality grade of the plowshare substructure.
[0064] Based on the marginal contribution rate of each indicator to the comprehensive evaluation index of the test section during the Schogley integral calculation process, the weak indicators of the test section are identified. Based on the preset correspondence between the weak indicators and the plow layer construction process parameters, a diagnostic evaluation report is output, which includes the comprehensive evaluation index, quality level, weak indicators, and corresponding process adjustment suggestions for each test section. The specific implementation is as follows: For each detection section, the positive quality scores of all soil testing indicators within the retained indicator set are arranged in descending order of value to obtain a positive quality score sequence. The soil testing indicator corresponding to each sorted position is recorded. The positive quality score sequence for each detection section position is as follows: sorted position 1 is soil porosity, sorted position 2 is soil saturated hydraulic conductivity, and sorted position 3 is soil penetration resistance. Specifically, the values at the first detection section position are: soil porosity 0.85, soil saturated hydraulic conductivity 0.80, and soil penetration resistance... The soil porosity at the second test section is 0.70, the saturated hydraulic conductivity at the third test section is 0.65, and the penetration resistance at the fourth test section is 0.45. The soil porosity at the fifth test section is 0.60, the saturated hydraulic conductivity at the fifth test section is 0.55, and the penetration resistance at the sixth test section is 0.45.
[0065] When soil testing indicators have the same positive quality score, the sorting position is determined according to the preset indicator arrangement order, namely soil penetration resistance, soil saturated hydraulic conductivity, and soil porosity. In this embodiment, no situation with the same positive quality score occurred. According to the Schogey integral calculation process, for each test section position, the set of soil testing indicators corresponding to sorting position 1 is a set composed of soil porosity. Its previous fuzzy measure value is recorded as zero. The fuzzy measure value of this set is read from the corrected fuzzy measure matrix as 0.2131. Therefore, the fuzzy measure increment corresponding to sorting position 1 is 0.2131.
[0066] The set of soil test indicators corresponding to the ranking position 2 is a set consisting of soil porosity and soil saturated hydraulic conductivity. The fuzzy measure value of the set of soil test indicators corresponding to the previous ranking position 1 is 0.2131. The fuzzy measure value of this set is read from the corrected fuzzy measure matrix as 0.4279. Therefore, the fuzzy measure increment corresponding to the ranking position 2 is 0.4279 minus 0.2131, which equals 0.2148.
[0067] The set of soil test indicators corresponding to the sorting position 3 is the complete set consisting of soil porosity, soil saturated hydraulic conductivity and soil penetration resistance. The fuzzy measure value of the set of soil test indicators corresponding to the previous sorting position 2 is 0.4279. The fuzzy measure value of this set is read as 1 from the corrected fuzzy measure matrix. Therefore, the fuzzy measure increment corresponding to the sorting position 3 is 1 minus 0.4279, which equals 0.5721.
[0068] Multiply the positive quality score of the soil testing index corresponding to each ranking position by the fuzzy measure increment corresponding to that ranking position to obtain the marginal contribution value of the soil testing index corresponding to that ranking position. Then divide the marginal contribution value by the comprehensive evaluation index of the plow pan at that testing section position to obtain the marginal contribution rate. At the first testing section position, the positive quality score of soil porosity 0.85 multiplied by 0.2131 gives a marginal contribution value of 0.181135, which is divided by 0.7534 to obtain a marginal contribution rate of 0.2404. The positive quality score of soil saturated hydraulic conductivity 0.80 multiplied by 0.2148 gives a marginal contribution value of 0.17184, which is divided by 0.7534 to obtain a marginal contribution rate of 0.2281. The positive quality score of soil penetration resistance 0.70 multiplied by 0.5721 gives a marginal contribution value of 0.40047, which is divided by 0.7534 to obtain a marginal contribution rate of 0.5315.
[0069] At the second detection section, the positive quality score of soil porosity (0.70) multiplied by 0.2131 yields 0.14917, which, when divided by 0.6321, gives a marginal contribution rate of 0.2360. The positive quality score of soil saturated hydraulic conductivity (0.65) multiplied by 0.2148 yields 0.13962, which, when divided by 0.6321, gives a marginal contribution rate of 0.2209. The positive quality score of soil penetration resistance (0.60) multiplied by 0.5721 yields 0.34326, which, when divided by 0.6321, gives a marginal contribution rate of 0.5431.
[0070] At the third detection section, the positive quality score of soil porosity (0.65) multiplied by 0.2131 yields 0.138515, which is then divided by 0.5821 to obtain a marginal contribution rate of 0.2380; the positive quality score of soil saturated hydraulic conductivity (0.60) multiplied by 0.2148 yields 0.12888, which is then divided by 0.5821 to obtain a marginal contribution rate of 0.2214; and the positive quality score of soil penetration resistance (0.55) multiplied by 0.5721 yields 0.314655, which is then divided by 0.5821 to obtain a marginal contribution rate of 0.5406.
[0071] At the fourth detection section, the positive quality score of soil porosity (0.60) multiplied by 0.2131 yields 0.12786, which is then divided by 0.4927 to obtain a marginal contribution rate of 0.2595; the positive quality score of soil saturated hydraulic conductivity (0.50) multiplied by 0.2148 yields 0.1074, which is then divided by 0.4927 to obtain a marginal contribution rate of 0.2180; and the positive quality score of soil penetration resistance (0.45) multiplied by 0.5721 yields 0.257445, which is then divided by 0.4927 to obtain a marginal contribution rate of 0.5225.
[0072] At the fifth detection section, the positive quality score of soil porosity (0.60) multiplied by 0.2131 yields 0.12786, which is then divided by 0.5034 to obtain a marginal contribution rate of 0.2540; the positive quality score of soil saturated hydraulic conductivity (0.55) multiplied by 0.2148 yields 0.11814, which is then divided by 0.5034 to obtain a marginal contribution rate of 0.2347; and the positive quality score of soil penetration resistance (0.45) multiplied by 0.5721 yields 0.257445, which is then divided by 0.5034 to obtain a marginal contribution rate of 0.5114.
[0073] The preset criterion for determining weak indicators is a marginal contribution rate of less than 0.25. The marginal contribution rates of each soil test indicator at each test section are compared with this standard. At the first test section, the marginal contribution rate of soil porosity (0.2404) is less than 0.25, the marginal contribution rate of soil saturated hydraulic conductivity (0.2281) is less than 0.25, and the marginal contribution rate of soil penetration resistance (0.5315) is greater than 0.25. Therefore, soil porosity and soil saturated hydraulic conductivity are determined to be weak indicators.
[0074] At the second detection section, the marginal contribution rate of soil porosity was 0.2360, which was less than 0.25; the marginal contribution rate of soil saturated hydraulic conductivity was 0.2209, which was less than 0.25; and the marginal contribution rate of soil penetration resistance was 0.5431, which was greater than 0.25. Therefore, soil porosity and soil saturated hydraulic conductivity were determined to be weak indicators.
[0075] At the third detection section, the marginal contribution rate of soil porosity was 0.2380, which was less than 0.25; the marginal contribution rate of soil saturated hydraulic conductivity was 0.2214, which was less than 0.25; and the marginal contribution rate of soil penetration resistance was 0.5406, which was greater than 0.25. Therefore, soil porosity and soil saturated hydraulic conductivity were determined to be weak indicators.
[0076] At the fourth detection section, the marginal contribution rate of soil porosity is 0.2595, which is greater than 0.25; the marginal contribution rate of soil saturated hydraulic conductivity is 0.2180, which is less than 0.25; and the marginal contribution rate of soil penetration resistance is 0.5225, which is greater than 0.25. Therefore, soil saturated hydraulic conductivity is determined to be a weak indicator.
[0077] At the fifth detection section, the marginal contribution rate of soil porosity is 0.2540, which is greater than 0.25; the marginal contribution rate of soil saturated hydraulic conductivity is 0.2347, which is less than 0.25; and the marginal contribution rate of soil penetration resistance is 0.5114, which is greater than 0.25. Therefore, soil saturated hydraulic conductivity is determined to be a weak indicator.
[0078] The pre-defined relationship between the weak index and the plow pan construction process parameters is as follows: When the weak index is soil penetration resistance, the corresponding adjustment suggestion for the plow pan construction process is to adopt deep tillage to reduce soil compaction and decrease soil penetration resistance. When the weak index is soil saturated hydraulic conductivity, the corresponding adjustment suggestion for the plow pan construction process is to adopt a combination of deep tillage and organic fertilization to improve soil pore connectivity and increase soil saturated hydraulic conductivity. When the weak index is soil porosity, the corresponding adjustment suggestion for the plow pan construction process is to increase the application of organic materials and reduce mechanical compaction to maintain or increase soil porosity.
[0079] Based on this correspondence, the recommended adjustment for the plowshare construction process at the first detection section is to improve soil pore connectivity and increase soil saturated hydraulic conductivity by combining deep tillage with organic fertilization. Simultaneously, increase the application of organic materials and reduce mechanical compaction to maintain or improve soil porosity. The recommended adjustment for the plowshare construction process at the second and third detection sections is the same as at the first and third detection sections. The recommended adjustment for the plowshare construction process at the fourth and fifth detection sections is the same as at the fourth detection section.
[0080] The output is a diagnostic evaluation report containing the comprehensive evaluation index of the plow pan structure at each detection section, the quality grade of the plow pan structure, the weak indicators, and the corresponding suggestions for adjusting the plow pan structure process. The report shows that the comprehensive evaluation index of the plow pan structure at the first detection section is 0.7534, the quality grade of the plow pan structure is good, the weak indicators are soil porosity and soil saturated hydraulic conductivity, and the corresponding suggestions for adjusting the plow pan structure process are to increase the application of organic materials and reduce mechanical compaction to maintain or increase soil porosity, and at the same time, to improve soil pore connectivity and increase soil saturated hydraulic conductivity by combining deep loosening and organic fertilization.
[0081] The comprehensive evaluation index of the plow layer construction at the second test section was 0.6321, indicating a good quality level. The weak indicators were soil porosity and soil saturated hydraulic conductivity. The corresponding recommendations for adjusting the plow layer construction process were to increase the application of organic materials and reduce mechanical compaction to maintain or improve soil porosity. At the same time, a combination of deep loosening and organic fertilization should be adopted to improve soil pore connectivity and increase soil saturated hydraulic conductivity.
[0082] The comprehensive evaluation index of the plow layer construction at the third testing section was 0.5821, indicating a medium quality level. The weak indicators were soil porosity and soil saturated hydraulic conductivity. The corresponding recommendations for adjusting the plow layer construction process were to increase the application of organic materials and reduce mechanical compaction to maintain or improve soil porosity. At the same time, a combination of deep loosening and organic fertilization should be adopted to improve soil pore connectivity and increase soil saturated hydraulic conductivity.
[0083] The comprehensive evaluation index of the plow layer construction at the fourth detection section was 0.4927, the quality level of the plow layer construction was medium, the weak index was soil saturated hydraulic conductivity, and the corresponding adjustment suggestion for the plow layer construction process was to improve soil pore connectivity and increase soil saturated hydraulic conductivity by combining deep loosening and organic fertilization.
[0084] The comprehensive evaluation index of the plow layer construction at the fifth detection section was 0.5034, the quality level of the plow layer construction was medium, the weak index was soil saturated hydraulic conductivity, and the corresponding adjustment suggestion for the plow layer construction process was to improve soil pore connectivity and increase soil saturated hydraulic conductivity by combining deep loosening and organic fertilization.
[0085] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.
[0086] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, and a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0087] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.
[0088] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
[0089] In conclusion, 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 comprehensive evaluation method for the quality of plowshare subsurface construction based on multi-index fusion, characterized in that, Includes the following steps: S1. Receive soil test index data collected on-site according to the test section location at the site of the plow bottom construction quality acceptance evaluation. The soil test index includes mechanical support index, moisture control index and structural integrity index at different soil depths. Perform positive and standardization processing on each soil test index, calculate the degree of deterioration of each soil test index relative to the preset corresponding functional benchmark value, generate a degree of deterioration sequence, arrange the degree of deterioration sequence according to the test section location, and construct the plow bottom quality state matrix. S2. Read the plow bottom quality state matrix, calculate the maximum information coefficient and Spearman rank correlation coefficient between any two soil detection indicators, identify indicator pairs with strong nonlinear correlation based on the maximum information coefficient, select representative indicators from the indicator pairs according to the preset retention indicator selection rules and remove redundant indicators to obtain the retention indicator set, determine the positive synergy or negative conflict relationship between indicators based on the sign of the Spearman rank correlation coefficient between each indicator in the retention indicator set, and construct a fuzzy measure matrix to characterize the interaction tendency between indicators based on the retention indicator set and the conflict relationship. S3. Based on the fuzzy measure matrix, calculate the Shapley value of each indicator in the retained indicator set as the indicator importance. Input the degree to which the average deterioration of each indicator exceeds the total average deterioration into the variable weight equilibrium function to perform short-board strengthening correction on the indicator importance. Reconstruct the fuzzy measure matrix based on the corrected indicator importance to obtain the corrected fuzzy measure matrix. S4. For each detection section, based on the corrected fuzzy measure matrix, the degradation degree of each indicator is converted into a positive quality score. The positive quality scores of each indicator of the section are aggregated into the plowshare of the detection section to construct a comprehensive evaluation index. S5. Based on the preset quality level range, map the comprehensive evaluation index of each test section to the quality level of the plowshare substructure of the test section. Based on the marginal contribution rate of each indicator to the comprehensive evaluation index of the test section during the Shape integral calculation, identify the weak indicators of the test section. Based on the preset correspondence between the weak indicators and the plowshare substructure construction process parameters, output a diagnostic evaluation report containing the comprehensive evaluation index, quality level, weak indicators and corresponding process adjustment suggestions for each test section.
2. The method for comprehensive evaluation of plowshare substructure construction quality based on multi-index fusion as described in claim 1, characterized in that: The system receives soil testing data collected at the testing section locations during the on-site quality acceptance evaluation of the plow pan substructure. These data include mechanical support indicators, moisture regulation indicators, and structural integrity indicators at different soil depths. Each soil testing indicator is normalized and standardized, and the degree of degradation of each indicator relative to a preset functional benchmark value is calculated, generating a degradation degree sequence. This sequence is then arranged according to the testing section location to construct a plow pan substructure quality state matrix, specifically including: The soil testing index data is collected on-site at the test section locations for the quality acceptance evaluation of the plow layer construction. The soil testing index data is obtained by determining multiple test section locations at preset intervals along the extension direction of the plow layer, dividing multiple soil layer depths at preset depth intervals at each test section location, and collecting soil samples at each soil layer depth. The soil testing indexes include mechanical support indexes, moisture control indexes, and structural integrity indexes. The mechanical support index is soil penetration resistance, the moisture control index is soil saturated hydraulic conductivity and soil moisture content, and the structural integrity index is soil bulk density and soil porosity. A baseline value table is established in advance according to each soil testing index. Under each soil testing index, functional baseline value, preferred type and preset optimal value are set according to different soil depths. The preferred type is one of the following: the larger the value, the better the quality; the smaller the value, the better the quality; or the closer the value is to the optimal value, the better the quality. The preferred type for soil penetration resistance and soil bulk density is that the smaller the value, the better the quality. The preferred type for soil saturated hydraulic conductivity and soil moisture content is that the closer the value is to the optimal value, the better the quality. The preferred type for soil porosity is that the larger the value, the better the quality. For each soil testing index, a positive transformation process is performed according to the corresponding preferred type. For soil testing indices with larger values and better quality, the soil testing index value is retained as the positive transformation value. For soil testing indices with smaller values and better quality, the inverse transformation method is used to take the reciprocal of the soil testing index value as the positive transformation value. For soil testing indices with values closer to the optimal value and better quality, the optimal distance compensation method is used to calculate the distance between the soil testing index value and the preset optimal value, and the distance is converted into a positive transformation value, so that the smaller the distance, the larger the positive transformation value. At the same time, for each functional benchmark value in the benchmark value table, the same positive transformation process is performed according to the preferred type of the index and the preset optimal value to obtain the positive benchmark value. The range standardization method is adopted. For each soil test index, based on the maximum and minimum values of the positive values of the soil test index in all test sections, each positively normalized soil test index is linearly mapped to a unified numerical range to obtain a standardized test value. At the same time, the positively normalized benchmark value of the soil test index is linearly mapped to a unified numerical range according to the same maximum and minimum values to obtain a standardized benchmark value, so that the standardized test value and the standardized benchmark value have a consistent measurement scale. Based on the soil layer depth to which each soil testing index belongs, the standardized benchmark value corresponding to the soil testing index is obtained. When the standardized test value is less than the standardized benchmark value, the difference between the standardized benchmark value and the standardized test value is taken as the degree of deterioration. When the standardized test value is greater than or equal to the standardized benchmark value, the degree of deterioration is recorded as zero. The degradation degree of each soil test index corresponding to each soil layer depth at each test section location is arranged in a preset order to generate a degradation degree sequence corresponding to the test section location. All degradation degree sequences corresponding to the test section locations are arranged according to the test section location, with the test section location as the row and the combination of soil layer depth and soil test index as the column, to construct the plow layer quality state matrix.
3. The method for comprehensive evaluation of plowshare substructure construction quality based on multi-index fusion as described in claim 1, characterized in that: For the plow pan quality state matrix, the maximum information coefficient and Spearman rank correlation coefficient between any two soil detection indicators are calculated. Indicator pairs with strong nonlinear correlations are identified based on the maximum information coefficient. Representative indicators are selected from these pairs according to a pre-defined retention indicator selection rule, and redundant indicators are removed to obtain a retention indicator set. The positive or negative synergistic relationship between indicators is determined based on the sign of the Spearman rank correlation coefficient between each indicator in the retention indicator set. A fuzzy measure matrix characterizing the interaction tendency between indicators is constructed based on the retention indicator set and the conflict relationship, specifically including: Read the plow pan quality state matrix and assemble the deterioration degree of each soil test index at each test section position into a soil test index deterioration degree sequence according to the test section position order. For any two soil testing indicators, the deterioration degree sequences corresponding to the two soil testing indicators are mapped to a two-dimensional plane. The deterioration degree value range of the two soil testing indicators is divided into multiple grids according to a preset number of grids. The number of data points falling into each grid is counted. The mutual information value is calculated based on the number of data points. After normalizing the mutual information value to a preset interval, the maximum information coefficient between the two soil testing indicators is obtained. At the same time, the deterioration degree of the two soil testing indicators at each testing section is sorted according to the numerical value. Each deterioration degree is assigned a sorting number. The Spearman rank correlation coefficient is calculated based on the difference between the sorting numbers of the two soil testing indicators at the same testing section. Soil detection index pairs with a maximum information coefficient greater than a preset strong correlation threshold are identified as index pairs with strong nonlinear correlation. According to the preset retention index selection rules, representative indicators are selected from the index pairs with strong nonlinear correlation, redundant indicators that are not selected are removed, and soil detection indicators that do not appear in the index pairs with strong nonlinear correlation are retained to obtain the retention index set. For any two soil testing indicators in the retained indicator set, when the Spearman rank correlation coefficient between the two soil testing indicators is positive, it is determined that the two soil testing indicators have a positive synergistic relationship; when the Spearman rank correlation coefficient is negative, it is determined that the two soil testing indicators have an inverse conflict relationship. A fuzzy measure matrix is constructed to characterize the interaction tendency among indicators based on the positive synergy and negative conflict relationships among the retained indicator set and soil testing indicators. The elements in the fuzzy measure matrix represent the comprehensive interaction tendency of the corresponding soil testing indicator set. The fuzzy measure of a single soil testing indicator is determined by the sum of the absolute values of the Spearman rank correlation coefficients between the soil testing indicator and the remaining soil testing indicators in the retained indicator set. The fuzzy measure of a soil testing indicator set consisting of at least two soil testing indicators is determined by the positive synergy and negative conflict relationships between all pairs of soil testing indicators within the soil testing indicator set.
4. The method for comprehensive evaluation of plowshare substructure construction quality based on multi-index fusion as described in claim 1, characterized in that: Based on the fuzzy measure matrix, the Shapley value of each indicator in the retained indicator set is calculated as the indicator importance, specifically including: Read the degree of deterioration of each soil detection index in the retained index set and the plow bottom quality state matrix at each detection section. The fuzzy measure matrix contains the fuzzy measure values of all index subsets in the retained index set. Based on the fuzzy measure matrix, for each soil detection index in the retained index set, all index subsets in the retained index set that do not contain soil detection indexes are traversed. For each index subset traversed, the fuzzy measure value of the index subset is read from the fuzzy measure matrix, and then the fuzzy measure value of the new index set formed after adding the soil detection index to the index subset is read. The difference between the two fuzzy measure values is taken as the marginal contribution value corresponding to the index subset. Calculate the factorial of the number of indicators contained in the indicator subset, the factorial of the number of indicators that do not include soil testing indicators and are not in the indicator subset, and the factorial of the total number of indicators in the retained indicator set. Divide the product of the factorial of the number of indicators contained in the indicator subset and the factorial of the number of indicators that do not include soil testing indicators and are not in the indicator subset by the factorial of the total number of indicators in the retained indicator set to obtain the cumulative weight corresponding to the marginal contribution value. Sum the products of the marginal contribution value of the soil testing indicator under all indicator subsets and the corresponding cumulative weight to obtain the Shapley value of the soil testing indicator. The Shapley value is used as the importance of the soil testing indicator.
5. The method for comprehensive evaluation of plowshare substructure construction quality based on multi-index fusion as described in claim 4, characterized in that: The degree to which the average degradation of each indicator exceeds the overall average degradation is input into the variable-weight equilibrium function to correct the weakness in the indicator importance. The fuzzy measure matrix is then reconstructed based on the corrected indicator importance, resulting in the corrected fuzzy measure matrix, which specifically includes: Calculate the average degree of deterioration of all soil testing indicators in the retained indicator set at all testing sections, as the total average degree of deterioration. Then calculate the average degree of deterioration of each soil testing indicator at all testing sections, as the indicator average degree of deterioration of the soil testing indicator. When the indicator average degree of deterioration of the soil testing indicator is greater than the total average degree of deterioration, the difference between the indicator average degree of deterioration and the total average degree of deterioration is taken as the degree of deviation of the soil testing indicator. When the indicator average degree of deterioration of the soil testing indicator is less than or equal to the total average degree of deterioration, the degree of deviation of the soil testing indicator is set to zero. When the sum of the deviations of all soil test indicators is zero, no correction is made to the importance. The corrected importance of each soil test indicator is kept as the importance before correction, and the importance is used as the final corrected importance. When the sum of the deviations of all soil testing indicators is greater than zero, the deviation of each soil testing indicator is input into the variable weight equilibrium function. The variable weight equilibrium function divides the deviation of each soil testing indicator by the sum of all deviations to obtain the deviation ratio of the soil testing indicator. The deviation ratio is used as the exponent to perform a power operation on the natural constant to obtain the correction coefficient of the soil testing indicator. The importance of each soil testing indicator is multiplied by the correction coefficient to obtain the corrected importance of the soil testing indicator. All corrected importance values are normalized so that the sum of the corrected importance values of each soil testing indicator is equal to the sum of the importance values of each soil testing indicator before correction. The normalized importance value is used as the final corrected importance value. The fuzzy measure value of each individual soil test index in the fuzzy measure matrix is corrected to the final corrected importance of the corresponding soil test index. For a set of soil test indices consisting of at least two soil test indices, the ratio of the sum of the final corrected importance of each soil test index in the set to the sum of the importance before correction is calculated. The ratio is multiplied by the fuzzy measure value of the set of soil test indices in the fuzzy measure matrix to obtain the corrected fuzzy measure value of the set of soil test indices. At the same time, the positive synergistic relationship or the negative conflict relationship between soil test indices in the set of soil test indices remains unchanged. The reconstructed fuzzy measure matrix is used as the corrected fuzzy measure matrix.
6. The method for comprehensive evaluation of plowshare substructure construction quality based on multi-index fusion as described in claim 1, characterized in that: For each detection section, based on the corrected fuzzy measure matrix, the degradation degree of each indicator is converted into a positive quality score. The Schogai integral is then used to aggregate the positive quality scores of each indicator in the section into a plowshare layer, constructing a comprehensive evaluation index, which specifically includes: Read the modified fuzzy metric matrix and the plow bottom quality state matrix to retain the degree of deterioration of each soil detection index in the index set at each detection section location. For each detection section location, extract the degree of deterioration of each soil detection index in the index set at the detection section location from the plow bottom quality state matrix. According to the preset positive quality score conversion rules, the degree of deterioration of each soil test index at the test section is converted into a positive quality score. The positive quality score conversion rules are as follows: when the degree of deterioration is zero, the positive quality score takes the maximum value within the preset range; when the degree of deterioration reaches the preset upper limit of deterioration, the positive quality score takes the minimum value within the preset range; and when the degree of deterioration is between zero and the preset upper limit of deterioration, the positive quality score decreases linearly as the degree of deterioration increases. Using the Schottky integral, the positive quality scores of all soil test indicators in the retained index set at the test section location are arranged in descending order of value to obtain the positive quality score sequence, and the soil test indicator corresponding to each positive quality score is recorded. For any two adjacent positive quality scores in the positive quality score sequence, the difference between the positive quality score that comes first and the positive quality score that comes later is taken as the first difference value corresponding to the adjacent position. The set of soil test indicators corresponding to the first difference value consists of the soil test indicators corresponding to the positive quality score that comes first and all soil test indicators that come before the soil test indicators. The difference between the smallest positive quality score and zero is taken as the last difference. The set of soil test indicators corresponding to the last difference consists of the soil test indicator corresponding to the smallest positive quality score and all soil test indicators listed before the soil test indicator. Read the fuzzy measure values of the soil detection index set corresponding to the first difference and the soil detection index set corresponding to the last difference from the corrected fuzzy measure matrix respectively. Multiply each first difference by the fuzzy measure value of the corresponding soil detection index set to obtain the aggregated contribution value corresponding to the first difference. Multiply the last difference by the fuzzy measure value of the corresponding soil detection index set to obtain the aggregated contribution value corresponding to the last difference. Add the aggregated contribution values corresponding to all first differences and the aggregated contribution values corresponding to the last difference to obtain the comprehensive evaluation index of the plow layer construction at the detection section location. Repeat the above steps for each detection section location to obtain the comprehensive evaluation index of the plow layer construction at all detection section locations.
7. The method for comprehensive evaluation of plowshare substructure construction quality based on multi-index fusion as described in claim 1, characterized in that: Based on the preset quality level range, the comprehensive evaluation index of each inspection section is mapped to the quality level of the plowshare layer construction of the inspection section, specifically including: Read the comprehensive evaluation index of the plow layer at all detection section locations, the corrected fuzzy measure matrix, and the positive quality score of each soil detection index in the retained index set at each detection section location; According to the preset quality grade range, the comprehensive evaluation index of the plow layer construction at each test section location is compared with the boundary values of each quality grade in the preset quality grade range to determine the quality grade of the plow layer construction at the test section location.
8. The method for comprehensive evaluation of plowshare substructure construction quality based on multi-index fusion as described in claim 7, characterized in that: Based on the marginal contribution rate of each indicator to the comprehensive evaluation index of the test section during the Schogley integral calculation process, the weak indicators of the test section are identified. Based on the preset correspondence between the weak indicators and the plow layer construction process parameters, a diagnostic evaluation report is output, which includes the comprehensive evaluation index, quality level, weak indicators, and corresponding process adjustment suggestions for each test section. Specifically, it includes: For each detection section location, the positive quality scores of all soil detection indicators in the retained indicator set are arranged in descending order of value to obtain a positive quality score sequence. The soil detection indicator corresponding to each sorting position in the positive quality score sequence is recorded. When there are soil detection indicators with the same positive quality score, the sorting position is determined according to the preset indicator sorting order. According to the calculation process of the Schogey integral, for each detection section position, for each sorting position in the positive quality score sequence, when the sorting position is the first sorting position, the set of soil detection indicators corresponding to the sorting position is recorded as the set composed of the soil detection indicators corresponding to the sorting position, and the fuzzy measure value before the sorting position is recorded as zero. When the sorting position is not the first sorting position, the set of soil detection indicators corresponding to the sorting position is recorded as the set composed of all soil detection indicators from the first sorting position to the sorting position, and the fuzzy measure value of the set of soil detection indicators corresponding to the sorting position before the sorting position is read from the corrected fuzzy measure matrix. Read the fuzzy measure value of the soil test index set corresponding to the sorting position from the corrected fuzzy measure matrix, subtract the fuzzy measure value before the sorting position from the fuzzy measure value of the soil test index set corresponding to the sorting position to obtain the fuzzy measure increment corresponding to the sorting position, and multiply the positive quality score of the soil test index corresponding to the sorting position by the fuzzy measure increment corresponding to the sorting position to obtain the marginal contribution value of the soil test index corresponding to the sorting position. Divide the marginal contribution value corresponding to the ranking position by the comprehensive evaluation index of the plow layer at the detection section position to obtain the marginal contribution rate of the soil detection index corresponding to the ranking position. Soil testing indicators whose marginal contribution rate at each testing section location does not meet the preset weak indicator judgment standard are identified as weak indicators at the testing section location. Based on the preset weak indicator and the plow layer construction process parameters, corresponding plow layer construction process adjustment suggestions are matched for each weak indicator. The output is a diagnostic evaluation report containing the comprehensive evaluation index of the plow layer construction at each test section location, the quality grade of the plow layer construction, the weak indicators, and the corresponding suggestions for adjusting the plow layer construction process.
