Intelligent soil structure improvement method and system

Abstract

The application discloses a kind of soil structure intelligent improvement method and system, specifically related to soil structure improvement technical field;It is by collecting the original soil sample data of target cultivated area and constructing basic data set, after standardization and normalization processing, reconstructs soil structure state atlas using multidimensional feature mapping model;And through structure pattern recognition, the key structure deterioration index that influences tillage performance is extracted;Based on key structure deterioration index and crop root zone environmental demand parameter, establish structure improvement adaptation model, determine the adaptation weight relation matrix;According to the adaptation weight relation matrix, soil structure improvement measures are matched, and soil improvement strategy is formed;Soil improvement strategy is applied to target cultivated area, and the structure improvement adaptation model is updated, to realize the continuous improvement of soil structure in target cultivated area.It can accurately evaluate soil structure state and realize differentiated, intelligent soil improvement strategy formulation and implementation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of soil structure improvement, and more particularly to a soil structure intelligent improvement method and system. BACKGROUND

[0002] At present, soil structure degradation phenomenon widely exists in the process of agricultural production, which seriously affects the growth environment of crop root system. The existing soil improvement method depends on experience, lacks fine identification and quantitative analysis of soil structure state, and cannot realize the targeted identification and dynamic response of different soil problems in different farmland areas, so that the improvement measures are difficult to accurately adapt, which restricts the improvement of the level of agricultural precision and intelligent management.

[0003] In order to solve the above problems, a technical scheme is provided. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a soil structure intelligent improvement method and system to solve the problems raised in the background art.

[0005] In order to achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0006] A soil structure intelligent improvement method comprises the following steps:

[0007] S1: obtaining original soil sample data of a target farmland area, and constructing a basic data set;

[0008] S2: performing standardization and normalization processing on the basic data set, and reconstructing a soil structure state atlas through a multi-dimensional feature mapping model to obtain atlas data representing the integrity of the current soil structure;

[0009] S3: performing structure pattern recognition on the atlas data to identify key structure degradation indicators affecting the tillage performance;

[0010] S4: constructing a structure improvement adaptation model, fusing and analyzing the key structure degradation indicators and crop root zone environment demand parameters to generate an adaptation weight relationship matrix between soil structure improvement measures and soil response;

[0011] S5: matching the soil structure improvement measures according to the adaptation weight relationship matrix, generating a soil improvement strategy, and adjusting the implementation order and implementation strength according to a multi-objective optimization algorithm;

[0012] S6: applying the soil improvement strategy to the target farmland area, recording real-time soil sample data, updating the structure improvement adaptation model through an improvement feedback mechanism, and realizing the improvement of the soil structure of the target farmland area.

[0013] In a preferred embodiment, S1, specifically:

[0014] The soil sampling points are arranged according to a predetermined grid in the target farmland area, and the surface soil samples and deep soil samples are collected by soil profile method;

[0015] The physical property parameters, chemical property parameters and biological property parameters of the surface soil samples and deep soil samples are obtained respectively;

[0016] The physical property parameters, chemical property parameters and biological property parameters of the surface soil samples and deep soil samples are associated and matched according to the spatial coordinates of the soil sampling points and the sampling time sequence, and a basic data set is constructed.

[0017] In a preferred embodiment, S2, specifically:

[0018] The physical property parameters, chemical property parameters and biological property parameters of the surface soil samples and deep soil samples in the basic data set are standardized and normalized;

[0019] The basic data set after standardization and normalization is reconstructed by using a multi-dimensional feature mapping model, a soil structure state atlas represented by physical property parameters, chemical property parameters and biological property parameters is established, and atlas data reflecting the current soil structure integrity state of the target farmland area are generated.

[0020] In a preferred embodiment, S3, specifically:

[0021] The physical property parameters, chemical property parameters and biological property parameters of each soil sampling point in the soil structure state atlas data are subjected to multi-dimensional spatial clustering and principal component analysis;

[0022] The soil sampling points are divided into several structure similar groups by multi-dimensional spatial clustering, and the contribution rate of each principal component to the total variance is calculated by combining principal component analysis;

[0023] Based on the principal component contribution rate from high to low, the main structure variables affecting the tillage performance are screened;

[0024] According to the distribution difference degree of the main structure variables in the grid, the key structure deterioration index is extracted.

[0025] In a preferred embodiment, S4, specifically:

[0026] A structure improvement adaptation model is constructed based on the key structure deterioration index and the root zone environmental demand parameters of the crops to be planted in the target farmland area;

[0027] The structural improvement adaptation model is used for fusion analysis of the key structure deterioration indexes and the crop root zone environment demand parameters, to determine the response degree between the different types of soil structure improvement measures and the key structure deterioration indexes.

[0028] The adaptation weight is calculated according to the response degree between the different types of soil structure improvement measures and the key structure deterioration indexes, to form the adaptation weight relationship matrix between the soil structure improvement measures and the soil response.

[0029] In a preferred embodiment, S5, specifically:

[0030] Based on the adaptation weight relationship matrix between the soil structure improvement measures and the soil response, the different types of soil structure improvement measures are matched according to the pre-set adaptation weight threshold;

[0031] Based on the matching result, an initial soil structure improvement strategy set is formed;

[0032] The implementation sequence and the implementation intensity in the initial soil structure improvement strategy set are adjusted through a multi-objective optimization algorithm, to obtain the soil structure improvement strategy implemented at the spatial coordinates of the soil sampling points in the target arable land region.

[0033] In a preferred embodiment, S6, specifically:

[0034] The soil structure improvement strategy is implemented at the spatial coordinates of the soil sampling points in the target arable land region;

[0035] The soil sample data at the soil sampling points in the target arable land region are recorded;

[0036] The soil sample data and the basic data set are compared according to the soil sampling point spatial coordinates and the sampling time, to evaluate the improvement effect of the soil structure improvement strategy;

[0037] Based on the improvement effect of the soil structure improvement strategy, the structural improvement adaptation model is updated, to realize the continuous improvement of the soil structure in the target arable land region.

[0038] On the other hand, the present application provides a soil structure intelligent improvement system, comprising:

[0039] The data acquisition module acquires the original soil sample data of the target arable land region, to construct a basic data set;

[0040] The data processing module performs standardization and normalization processing on the basic data set, and reconstructs a soil structure state atlas through a multi-dimensional feature mapping model, to obtain atlas data representing the current soil structure integrity;

[0041] Structure identification module: structure pattern recognition is performed on the atlas data to identify key structure deterioration indexes affecting the tillage performance;

[0042] Adaptation modeling module: a structure improvement adaptation model is constructed to fuse and analyze the key structure deterioration indexes and crop root zone environment demand parameters, and an adaptation weight relationship matrix between the soil structure improvement measures and soil responses is generated;

[0043] Strategy generation module: the soil structure improvement measures are matched according to the adaptation weight relationship matrix, a soil improvement strategy is generated, and the implementation order and intensity are adjusted according to a multi-objective optimization algorithm;

[0044] Improvement execution module: the soil improvement strategy is applied to the target cultivated area, real-time soil sample data are recorded, the structure improvement adaptation model is updated through an improvement feedback mechanism, and the soil structure of the target cultivated area is improved.

[0045] The technical effects and advantages of the soil structure intelligent improvement method and system of the application are as follows:

[0046] By obtaining the original soil sample data of the target cultivated area, the real soil structure state can be reflected; by standardization and normalization processing and reconstruction of the structure state atlas based on a multi-dimensional feature mapping model, the fusion expression of the multi-dimensional properties of the soil is realized, and the recognition ability of the structure integrity change is improved; the key structure deterioration indexes affecting the tillage performance are mined through the structure pattern recognition, so that the soil structure improvement is targeted; the structure improvement adaptation model is constructed by combining the crop root zone environment demand parameters, and the responsiveness evaluation from the perspective of the root system suitability is realized; the soil structure improvement measures are matched according to the adaptation weight relationship matrix, and the order and intensity of the improvement strategy are adjusted by using the multi-objective optimization, so that the improvement effect and the implementation cost can be considered; the structure improvement adaptation model is continuously updated through the feedback mechanism, the dynamic perception and intelligent control of the soil structure are realized, and the precision and continuity of the improvement measures are improved. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 It is a schematic diagram of the soil structure intelligent improvement method of the application;

[0048] Figure 2 It is a structure schematic diagram of the soil structure intelligent improvement system of the application. DETAILED DESCRIPTION

[0049] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0050] Embodiment 1

[0051] Figure 1 A soil structure intelligent improvement method is given, which comprises the following steps:

[0052] S1: acquiring original soil sample data of a target farmland region, and constructing a basic data set;

[0053] S2: performing standardization and normalization processing on the basic data set, and reconstructing a soil structure state atlas through a multi-dimensional feature mapping model to obtain atlas data representing current soil structure integrity;

[0054] S3: performing structure mode recognition on the atlas data to identify key structure deterioration indexes affecting farmland performance;

[0055] S4: constructing a structure improvement adaptation model, fusing and analyzing the key structure deterioration indexes and crop root zone environment demand parameters to generate an adaptation weight relationship matrix between soil structure improvement measures and soil response;

[0056] S5: matching soil structure improvement measures according to the adaptation weight relationship matrix, generating soil improvement strategies, and adjusting implementation order and implementation strength according to a multi-objective optimization algorithm;

[0057] S6: applying the soil improvement strategies to the target farmland region, recording real-time soil sample data, updating the structure improvement adaptation model through an improvement feedback mechanism, and realizing improvement of the soil structure of the target farmland region.

[0058] Acquiring original soil sample data of a target farmland region, and constructing a basic data set, comprises:

[0059] Soil sampling points are arranged in the target farmland region according to a pre-determined grid division, and soil profile method is used to collect surface soil samples and deep soil samples respectively;

[0060] In the determined target farmland area range, first, according to the overall area size of the farmland area, the spatial heterogeneity characteristics of the soil properties and the expected analysis accuracy requirements, the target farmland area is divided into grids. The method of grid division is: in the boundary range of the target farmland area, the east-west direction and the north-south direction are divided into multiple grid units with equal area and regular shape at the same spatial distance, and the center point coordinates of each grid unit are taken as the spatial coordinate positions of the corresponding soil sampling points.

[0061] At the spatial coordinate positions corresponding to each grid unit, the soil samples are collected on site by using the soil profile method: a vertical profile is dug downward from the center of the spatial coordinate position of each soil sampling point, and two soil samples at different depth levels are collected, which are defined as surface soil samples and deep soil samples. Among them, the sampling depth range of the surface soil sample is from the ground surface to the shallow root growth zone below, and the sampling depth range of the deep soil sample is in the plough layer depth area below the sampling depth range of the surface soil sample.

[0062] The physical property parameters, chemical property parameters and biological property parameters of the surface soil samples and the deep soil samples are obtained respectively;

[0063] The physical property parameters, chemical property parameters and biological property parameters of the surface soil samples and the deep soil samples are obtained respectively;

[0064] The physical property parameters are obtained: the soil samples are subjected to indoor experimental analysis, and the soil density, soil porosity, soil particle composition ratio, soil moisture content and other parameters are measured by weighing method, specific gravity bottle method, cutting ring method and other experimental methods, and the type of instrument used for measurement and experimental steps are recorded.

[0065] The chemical property parameters are obtained: the soil samples are subjected to soil chemical property analysis, and the organic matter content, nitrogen content, phosphorus content, potassium content, acid-base degree and conductivity and other chemical properties of the soil samples are measured by laboratory analysis method, and the type of chemical reagent used, experimental method and model specification of analysis instrument are recorded.

[0066] The biological property parameters are obtained: the soil samples are subjected to microbial community activity, total amount of microorganisms and soil enzyme activity analysis, and the soil microbial quantity, biomass index and microbial activity level data are obtained by using fluorescence microscope counting method, dilution plate method and biological enzyme activity detection method and other experimental methods, and the test steps, experimental methods and instrument equipment types are recorded.

[0067] The physical property parameters, chemical property parameters and biological property parameters of the surface soil samples and the deep soil samples are associated and matched according to the soil sampling point spatial coordinates and sampling time sequence to construct a basic data set;

[0068] After the measurement of the physical property parameters, the chemical property parameters and the biological property parameters, the surface soil samples and the deep soil samples corresponding to each soil sampling point are uniformly coded and marked with the spatial coordinates and the sampling time of the soil sampling points as data tags to form a basic data set with spatial position identification and sampling time record. The basic data set contains the spatial coordinates, the sampling time, the soil physical property parameters, the soil chemical property parameters and the soil biological property parameters of each soil sampling point as the data source for the standardized processing, the normalized processing and the reconstruction of the soil structure state atlas.

[0069] The standardized and normalized processing is performed on the basic data set, and the soil structure state atlas is reconstructed through a multi-dimensional feature mapping model to obtain atlas data representing the current soil structure integrity, including:

[0070] The physical property parameters, the chemical property parameters and the biological property parameters of the surface soil samples and the deep soil samples in the basic data set are subjected to standardized and normalized processing.

[0071] The standardized processing is respectively implemented on each type of parameter in the basic data set. The standardized processing method is as follows: the arithmetic mean value of the same type of parameter values in all soil sampling points is calculated first, then the difference between the parameter value of each soil sampling point and the arithmetic mean value is calculated, and then the standard deviation of the same type of parameter values in all soil sampling points is calculated, and finally the difference between the parameter value of each soil sampling point and the arithmetic mean value is divided by the standard deviation to realize the standardized data conversion of each specific parameter type.

[0072] The above mean value standardization method is applied to the standardized processing of all physical property parameters, chemical property parameters and biological property parameters to ensure the comparability and consistency between the parameter data.

[0073] The normalized processing is performed on the standardized physical property parameters, the standardized chemical property parameters and the standardized biological property parameters. The normalization method is the minimum and maximum normalization method, which is specifically as follows: the minimum value and the maximum value of the standardized data of each parameter type in all soil sampling points are taken out, the value of each soil sampling point is subtracted from the taken minimum value, and then the obtained difference is divided by the difference between the maximum value and the minimum value to realize the normalized conversion of the standardized data of each parameter type between zero and one. The minimum and maximum normalization method is applied to all physical property parameters, chemical property parameters and biological property parameters in the basic data set, so that parameters of different scales and units can be unified into the same data interval range, which is convenient for effective construction of the soil structure state atlas.

[0074] After the standardization and normalization processing is completed, the basic data set with unified scale and in the same numerical range is obtained.

[0075] The multi-dimensional feature mapping model is used to reconstruct the basic data set after standardization and normalization processing, and a soil structure state atlas represented by physical property parameters, chemical property parameters and biological property parameters is established, and atlas data reflecting the current soil structure integrity state of the target farmland region is generated.

[0076] The selection method of the multi-dimensional feature mapping model is a nonlinear mapping analysis method, which can effectively capture the complex nonlinear relationship between soil sampling points. The steps of implementing the nonlinear mapping analysis method include:

[0077] The parameter configuration of the nonlinear mapping analysis method is determined, including the setting of the target mapping dimension. The target mapping dimension is determined by comprehensively considering the complexity of soil property parameters in the target farmland region and the interpretability of data analysis results. Usually, the target mapping dimension takes a lower dimension value that can cover enough variance information of the original data;

[0078] The basic data set after standardization and normalization processing is input into the nonlinear mapping analysis method, the data space distance between all soil sampling points is calculated, the data space distance matrix is constructed, and the local neighborhood correlation of each sampling point is calculated through the data space distance matrix, so as to determine the data structure similarity relationship between soil sampling points;

[0079] Based on the determined data structure similarity relationship between soil sampling points, an iterative optimization method is used to continuously adjust the position coordinates of each sampling point in the target low-dimensional space, and optimize the distance relationship between the positions of sampling points in the target low-dimensional space.

[0080] After multiple iterative optimization processes, the optimized coordinate positions of each soil sampling point in the target low-dimensional space are obtained, and these optimized coordinate positions are recorded, and the low-dimensional feature mapping process is completed.

[0081] After the low-dimensional feature mapping is completed, the soil structure state atlas represented by the optimized coordinate positions of each soil sampling point is obtained. Specifically, the optimized coordinate positions of the soil sampling points in the target low-dimensional space are combined with the spatial coordinate positions and sampling time information of the soil sampling points in the basic data set, and are marked in the soil structure state atlas, so as to ensure that the atlas data corresponds to the actual spatial position.

[0082] In the reconstructed soil structure state atlas, the optimized coordinate positions of the soil sampling points collectively express the comprehensive structure characteristics constituted by the physical property parameters, chemical property parameters and biological property parameters of the surface soil samples and deep soil samples of each sampling point. The atlas data reflect the overall spatial feature distribution of the soil structure in the target farmland region, facilitating structure mode recognition of the atlas data and accurate extraction of key structure deterioration indexes.

[0083] The structure mode recognition is performed on the atlas data to identify key structure deterioration indexes affecting the tillage performance, including:

[0084] The physical property parameters, chemical property parameters and biological property parameters of each soil sampling point in the soil structure state atlas data are subjected to multi-dimensional spatial clustering and principal component analysis;

[0085] The soil sampling points are divided into several structure similar groups through multi-dimensional spatial clustering, and the contribution rate of each principal component to the total variance is calculated in combination with the principal component analysis;

[0086] Based on the principal component contribution rate from high to low, the main structure variables affecting the tillage performance are screened;

[0087] According to the distribution difference degree of the main structure variables in the grid, the key structure deterioration indexes are extracted.

[0088] In order to accurately analyze the similarity relationship between each soil sampling point in the soil structure state atlas data, a multi-dimensional spatial clustering algorithm is used to perform structure mode recognition on the soil structure state atlas data. The selected multi-dimensional spatial clustering algorithm is a clustering analysis algorithm based on a density clustering method, which can effectively identify any shape of soil structure similar groups existing in the space.

[0089] The implementation process of the multi-dimensional spatial clustering algorithm is as follows:

[0090] The parameter configuration of the clustering analysis algorithm is determined. The parameter configuration of the clustering analysis algorithm includes a density threshold parameter and a neighborhood distance parameter. The density threshold parameter is obtained through preliminary analysis of the spatial distribution and parameter difference degree of the soil sampling points, and the neighborhood distance parameter is determined according to the spatial distance relationship between the soil sampling points in the soil structure state atlas data, so as to ensure that the soil sampling points can accurately express similar structure characteristics when being clustered;

[0091] Using the determined density threshold parameter and neighborhood distance parameter, the number of other sampling points in the neighborhood of each soil sampling point in the soil structure state atlas data is calculated, so as to determine whether each soil sampling point belongs to a dense region. The judgment rule of the dense region is that when the number of sampling points in the neighborhood range of the soil sampling point reaches or exceeds the density threshold, it is determined that the soil sampling point belongs to the dense region;

[0092] Based on the above rules, the soil structure state atlas data is searched multiple times, and soil sampling points in the same dense area and adjacent or overlapping neighborhood are summarized into the same structure similar group, thereby dividing the target cultivated area into several groups with similar structure characteristics.

[0093] After completing the multi-dimensional space clustering analysis, all soil sampling points in the soil structure state atlas data are attributed to a specific structure similar group, forming the classification results of structure pattern recognition.

[0094] In order to further analyze the comprehensive difference characteristics of soil property parameters in each structure similar group in the clustering results, principal component analysis method is used for dimension reduction analysis and principal component contribution rate calculation.

[0095] The principal component analysis implementation process is as follows:

[0096] An input data matrix for principal component analysis is constructed, which is composed of physical property parameters, chemical property parameters and biological property parameters of all soil sampling points after standardization and normalization, each row represents a soil sampling point, and each column represents a soil property parameter;

[0097] Based on the input data matrix, a covariance matrix is calculated, and each element in the covariance matrix is the covariance value of two parameters at all soil sampling points;

[0098] The eigenvectors and eigenvalues of the covariance matrix are calculated using linear algebra method, and the contribution degree of each principal component to the overall variance of soil property parameters is determined by the size of the eigenvalue, i.e. the principal component contribution rate, and the principal component with larger eigenvalue has higher contribution rate to the overall variance;

[0099] All calculated principal components are sorted from high to low according to the size of the corresponding eigenvalue to obtain the sorted principal component sequence;

[0100] Based on the principle that the cumulative principal component contribution rate reaches the preset variance explanation level, several principal components that have the most obvious influence on soil structure and tillage performance are selected from the sorted principal component sequence, which are defined as main structure variables, and the main structure variables can effectively reflect the main trend of the overall change of the atlas data.

[0101] According to the spatial distribution difference characteristics of the selected main structure variables, key structure deterioration indexes are extracted, which are as follows:

[0102] The value of each main structure variable is mapped to the grid cell defined in step S1, the value of the main structure variable corresponding to the soil sampling point in each grid cell is counted, and the absolute difference between the value of the main structure variable in each grid cell and the average value of the target cultivated area is calculated;

[0103] Calculate the variance of each main structure variable in all grid cells, and reflect the unevenness of the spatial distribution of the main structure variable in the target farmland area through the size of the variance;

[0104] A preset variance threshold is selected, and the main structure variable with a variance value greater than or equal to the variance threshold in the spatial grid cell is selected as the key structure deterioration index, which includes but is not limited to: deep soil compaction layer thickness, soil porosity spatial variation degree, soil aggregate spatial distribution unevenness index, etc. The arithmetic mean of the variance values of all main structure variables is three times the variance threshold.

[0105] Construct a structure improvement adaptation model to fuse and analyze the key structure deterioration index and the crop root zone environmental demand parameter, and generate an adaptation weight relationship matrix between the soil structure improvement measures and the soil response, including:

[0106] Based on the key structure deterioration index and the root zone environmental demand parameter of the crop to be planted in the target farmland area, a structure improvement adaptation model is constructed;

[0107] According to the agricultural scientific research literature or the field cultivation experimental data of the target crop, the optimal soil compaction degree range required for the root growth of the target crop, the soil porosity suitable range, and the soil aggregate distribution uniformity requirement standard are determined. The root zone environmental demand parameter is defined as the input variable of the structure improvement adaptation model to ensure that the improvement measures output by the structure improvement adaptation model can accurately meet the growth demand of the target crop.

[0108] Through the structure improvement adaptation model, the key structure deterioration index and the crop root zone environmental demand parameter are fused and analyzed to determine the response degree between different categories of soil structure improvement measures and the key structure deterioration index;

[0109] According to the response degree between different categories of soil structure improvement measures and each key structure deterioration index, the adaptation weight is calculated to form an adaptation weight relationship matrix between the structure improvement measures and the soil response.

[0110] Specifically, a fuzzy reasoning method is used to construct the structure improvement adaptation model, and the construction process of the structure improvement adaptation model includes:

[0111] A fuzzy reasoning rule base is established. The fuzzy reasoning rule base is developed through agricultural soil expert knowledge and field test data, and gives multiple rules. Each rule describes the response degree of a certain soil structure improvement measure to the crop root zone environmental demand under the condition of a specific key structure deterioration index in the form of language expression, for example, when the soil compaction layer thickness is greater than a certain specific thickness, the response degree of deep loosening measures is high. Similar rules such as

[0112] A membership function is defined for each parameter in the set of root zone environmental demand parameters and each key structure degradation indicator, and the membership function is selected from a triangular function, a trapezoidal function, or a Gaussian function. The parameters of the membership function are determined by optimizing the actual field test data to ensure that the membership function accurately represents the membership degree within the numerical range of each key structure degradation indicator and root zone environmental demand parameter;

[0113] The numerical values of the key structure degradation indicators and the parameters in the set of root zone environmental demand parameters are substituted into the respective membership functions to calculate the membership values of each parameter, which are used to describe the quantitative membership relationship of each parameter in the soil structure adaptation model.

[0114] According to each fuzzy inference rule in the fuzzy inference rule base, the membership values obtained by calculation are used to perform a fuzzy inference process. The inference method uses the maximum-minimum composition method. First, the minimum value of the membership values of each parameter involved in each rule is calculated to determine the membership level at which the rule is established. Then, the maximum value of the membership levels at which multiple rules are established under the same measure is calculated as the comprehensive response degree of each soil structure improvement measure to each key structure degradation indicator.

[0115] The comprehensive response degree of each soil structure improvement measure to each key structure degradation indicator is obtained through the above fuzzy inference method, and a response degree set is generated.

[0116] Based on the response degree set, the adaptation weight between each soil structure improvement measure and each key structure degradation indicator is calculated. The calculation method of the adaptation weight is as follows: first, the comprehensive response degree of each soil structure improvement measure is weighted and summarized, and the summary value of the comprehensive response degree is divided by the sum of the summary values of the comprehensive response degrees of all soil structure improvement measures to determine the adaptation weight value between each soil structure improvement measure and each key structure degradation indicator.

[0117] According to all the calculated adaptation weight values, an adaptation weight relationship matrix composed of soil structure improvement measures, key structure degradation indicators, and adaptation weights is constructed. The rows of the adaptation weight relationship matrix represent all soil structure improvement measures, and the columns represent all key structure degradation indicators. Each specific numerical value in the adaptation weight relationship matrix is the adaptation weight value between the corresponding soil structure improvement measure and the corresponding key structure degradation indicator.

[0118] According to the adaptation weight relationship matrix, the soil structure improvement measures are matched to generate a soil improvement strategy, and the implementation order and intensity are adjusted according to a multi-objective optimization algorithm, including:

[0119] Based on the matching weight relationship matrix between the structure improvement measures and the soil response, the soil structure improvement measures of different categories are matched according to the pre-set matching weight threshold value;

[0120] Based on the matching result, an initial soil structure improvement strategy set is formed;

[0121] The implementation sequence and intensity of the initial soil structure improvement strategy set are adjusted through a multi-objective optimization algorithm to obtain the soil structure improvement strategy implemented at the spatial coordinates of each soil sampling point in the target cultivated area.

[0122] Specifically, according to the matching weight relationship matrix between the soil structure improvement measures and the key structure deterioration indicators, the matching weight threshold value is set. The determination method of the matching weight threshold value is: calculating the arithmetic mean value of all matching weight values in the matching weight relationship matrix, and taking the arithmetic mean value as the matching weight threshold value. The arithmetic mean value can effectively reflect the overall matching level of all improvement measures, which is helpful to accurately distinguish the structure improvement measures with high and low matching degrees, and avoid errors caused by subjective setting of the threshold value.

[0123] Based on the matching weight threshold value, the matching weight values of each soil structure improvement measure on all key structure deterioration indicators in the matching weight relationship matrix are compared one by one. Specifically, the cases where the matching weight values of each structure improvement measure on all key structure deterioration indicators are higher than or equal to the matching weight threshold value are recorded and counted. The matching method is: if the number of key structure deterioration indicators whose matching weight values of a certain structure improvement measure exceed or equal to the matching weight threshold value reaches or exceeds half of the total number of key structure deterioration indicators corresponding to the structure improvement measure, the structure improvement measure is determined as a preliminary matching success.

[0124] For example: assuming that the deep soil mechanical deep loosening measure has three key structure deterioration indicators, namely deep soil compaction layer thickness, soil porosity spatial variability, and soil aggregate spatial distribution unevenness index. If the matching weights corresponding to the deep soil mechanical deep loosening measure are more than the threshold value, less than the threshold value, and more than the threshold value, respectively, the deep soil mechanical deep loosening measure successfully passes the matching and is determined as an initial soil structure improvement strategy candidate measure.

[0125] After the matching weight matching, all the matched structure improvement measures are summarized to form an initial soil structure improvement strategy set. The initial soil structure improvement strategy set includes all the soil structure improvement measures screened by the matching weight threshold value and the corresponding matching index conditions of each measure.

[0126] The multi-objective optimization algorithm type is a non-dominated sorting genetic algorithm, which can simultaneously consider multiple optimization objectives and achieve a reasonable balance between overall implementation efficiency and resource investment.

[0127] The multi-objective optimization algorithm implementation process is:

[0128] The specific optimization objectives and constraints of the multi-objective optimization algorithm are set. The optimization objectives include maximizing the soil improvement effect after the implementation of soil structure improvement measures, minimizing the improvement implementation cost, and minimizing the impact of improvement operations on farmland operations. The constraints include the limitation of actual resource investment capacity in the target cultivated area, the limitation of farmland machinery and equipment implementation capacity, and the improvement measure implementation time window, etc.

[0129] According to the optimization objectives and constraints, the specific parameters of the multi-objective optimization algorithm are determined, such as the initial population size, the maximum number of iterations, the crossover probability, and the mutation probability. The specific parameters are determined through pre-experiment verification to ensure the convergence and optimization efficiency of the algorithm.

[0130] The non-dominated sorting genetic algorithm is used to generate the initial population, and different combinations of implementation order and implementation strength parameters are formed by randomly combining soil structure improvement measures in the initial soil structure improvement strategy set.

[0131] The fitness function is used to evaluate the optimization objective satisfaction degree of each individual in the initial population. The fitness function calculation method is to quantitatively calculate multiple optimization objectives, and then determine the individual fitness value through weighted comprehensive evaluation.

[0132] The initial population is sorted according to the individual fitness value, and the individual with the optimal fitness value is selected as the parent for genetic operation.

[0133] The crossover and mutation operations are implemented. The crossover operation method is to exchange the implementation order and implementation strength parameters of the parent individuals, and the mutation operation is to randomly change the implementation strength or implementation order of individual soil structure improvement measures.

[0134] The fitness of the new generated offspring individuals is calculated again and sorted, and the individual with the optimal fitness is retained as the parent of the next generation.

[0135] The optimization process is stopped until the number of iterations reaches the maximum number of iterations. At this time, the combination of the implementation order and the implementation intensity parameter with the highest fitness value in the final optimization population is selected as the final implementation of the soil structure improvement strategy in the target cultivated area. The soil structure improvement strategy includes the implementation order, implementation intensity parameter, implementation frequency and implementation duration of each soil structure improvement measure at the spatial coordinates of each soil sampling point in the target cultivated area. The implemented measures include but are not limited to deep soil mechanical subsoiling measures, soil organic matter application measures, soil physical amendment application measures, etc.

[0136] The soil improvement strategy is applied to the target cultivated area, real-time soil sample data is recorded, and the structure improvement adaptation model is updated through the improvement feedback mechanism to realize the improvement of the soil structure in the target cultivated area, including:

[0137] The soil structure improvement strategy is implemented at each soil sampling point spatial coordinate in the target cultivated area;

[0138] Based on the soil sampling point spatial coordinates of the target cultivated area, the soil structure improvement operation is implemented according to the soil structure improvement measure type, implementation order, implementation intensity parameter and implementation frequency specified in the soil structure improvement strategy. The implementation equipment includes but is not limited to soil subsoiling mechanical equipment, organic matter application equipment and physical amendment application equipment, etc.

[0139] The soil sample data at the soil sampling points in the target cultivated area are sampled and recorded;

[0140] During the implementation of the soil structure improvement strategy and at a specific sampling time after the completion of the soil structure improvement strategy, the soil sample data of each soil sampling point in the target cultivated area are collected and recorded on site. The sampling method is soil profile method to ensure the comparability of the sampling process and the consistency of the data. The sampling depth is also divided into surface soil samples and deep soil samples. The collected soil samples are used for laboratory detection to obtain real-time soil sample data after the implementation of the soil structure improvement strategy, including the physical property parameters, chemical property parameters and biological property parameters of the surface soil and deep soil.

[0141] The soil sample data and the basic data set are compared according to the soil sampling point spatial coordinates and the sampling time to evaluate the improvement effect of the soil structure improvement strategy;

[0142] The improvement effect of the soil structure improvement strategy after implementation is evaluated by the following methods:

[0143] The evaluation standard is set, i.e. the improvement target value range of each soil parameter is defined;

[0144] Compare and analyze the parameter values ​​in the real-time soil sample data measured after the implementation of the soil structure improvement strategy with the improvement target value range;

[0145] When the parameter value of the real-time soil sample data enters the improvement target value range, it is determined to be effective improvement; when the parameter value of the real-time soil sample data does not enter the improvement target value range, it is determined to be ineffective improvement;

[0146] The proportion of each parameter of all soil sampling points that has reached an effective improvement state was counted. The improvement rate of each parameter was obtained by dividing the number of soil sampling points with effective improvement of each parameter by the total number of soil sampling points.

[0147] By comprehensively calculating the improvement compliance rates of all soil parameters and adopting the weighted average method, the improvement compliance rate of each soil parameter is multiplied by the weight of the importance of the soil parameter to the overall soil structure improvement, and then the sum is taken to obtain a comprehensive evaluation value of the overall soil structure improvement effect. The comprehensive evaluation value is used as an indicator to evaluate the improvement effect after the implementation of the soil structure improvement strategy.

[0148] Update the structural improvement adaptation model based on the improvement effect of the soil structure improvement strategy to achieve continuous improvement of the soil structure in the target cultivated land area;

[0149] Based on the calculated comprehensive evaluation values, the structural improvement adaptation model is updated and optimized to achieve continuous improvement of soil structure improvement.

[0150] The implementation process of structural improvement adaptation model update optimization is as follows:

[0151] The updating method is: the comprehensive evaluation values ​​are used to calibrate the membership function parameters in the structural improvement adaptation model and the rule weight parameters of the fuzzy inference rule base; the rule weight parameter refers to the membership level of each fuzzy inference rule in the fuzzy inference rule base calculated by inputting the membership values ​​of each parameter.

[0152] Based on the updated membership function parameters and rule weight parameters, the adaptive weight relationship matrix between soil structure improvement measures and key structural deterioration indicators is recalculated so that the adaptive weight relationship matrix reflects the latest actual soil structure and the soil response level after the implementation of the soil structure improvement strategy;

[0153] The updated structural improvement adaptation model and the adaptation weight relationship matrix are used in step S5 of the next cycle implementation to further optimize and improve the soil structure improvement strategy.

[0154] Example 2:

[0155] The embodiment 2 of the present application is different from the embodiment 1 in that the embodiment is to introduce a soil structure intelligent improvement system.

[0156] Figure 2 A structural schematic diagram of the soil structure intelligent improvement system is given, the soil structure intelligent improvement system comprises:

[0157] The data acquisition module acquires original soil sample data of a target farmland region, and constructs a basic data set.

[0158] The data processing module performs standardization and normalization processing on the basic data set, and reconstructs a soil structure state atlas through a multi-dimensional feature mapping model to obtain atlas data representing current soil structure integrity.

[0159] The structure recognition module performs structure pattern recognition on the atlas data to identify key structure deterioration indexes affecting the tillage performance.

[0160] The adaptive modeling module constructs a structure improvement adaptive model, fuses and analyzes the key structure deterioration indexes and crop root zone environment demand parameters, and generates an adaptive weight relationship matrix between soil structure improvement measures and soil responses.

[0161] The strategy generation module matches the soil structure improvement measures according to the adaptive weight relationship matrix, generates a soil improvement strategy, and adjusts the implementation order and implementation strength according to a multi-objective optimization algorithm.

[0162] The improvement execution module applies the soil improvement strategy to the target farmland region, records real-time soil sample data, updates the structure improvement adaptive model through an improvement feedback mechanism, and realizes the improvement of the soil structure of the target farmland region.

[0163] The above formulas are all dimensionless numerical calculations, the formulas are obtained by software simulation of a large amount of data to obtain a formula of the nearest real situation, and the preset parameters and threshold values in the formula are set by a person skilled in the art according to the actual situation.

[0164] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are wholly or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center through a wired (for example, infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available medium collections. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state disk.

[0165] Those of ordinary skill in the art can realize that the modules and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether the 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 the present application.

[0166] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the above-described system, device, and module can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0167] In several embodiments provided in the present application, it should be understood that the disclosed system, device, and method can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed ones can be indirect coupling or communication connection through some interfaces, devices, or modules, which can be electrical, mechanical, or other forms.

[0168] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0169] In addition, the functional modules in each embodiment of the present application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0170] The functions, if realized in the form of software function modules and sold or used as independent products, can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of software products, and the computer software products are stored in a storage medium, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and various program code storage media.

[0171] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0172] Finally: the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application, should be included in the protection scope of the present application.

Claims

1. A soil structure intelligent improvement method, characterized in that: The steps include: S1: Obtain the original soil sample data of the target cultivated land area and construct the basic data set; S2: Standardize and normalize the basic data set, and reconstruct the soil structure state map through a multidimensional feature mapping model to obtain map data representing the current soil structure integrity; The selection method of the multidimensional feature mapping model is the nonlinear mapping analysis method; S3: Perform structural pattern recognition on the atlas data to identify key structural deterioration indicators that affect tillage performance; perform multidimensional spatial clustering and principal component analysis on the physical, chemical, and biological property parameters of each soil sampling point in the soil structural state atlas data; S4: Construct a structural improvement adaptation model, integrate key structural deterioration indicators with crop root zone environmental demand parameters, and generate an adaptation weight relationship matrix between soil structural improvement measures and soil responses; use fuzzy reasoning methods to construct the structural improvement adaptation model; S5: Match soil structure improvement measures according to the adaptation weight relationship matrix, generate soil improvement strategies, and adjust the implementation sequence and intensity according to the multi-objective optimization algorithm; S6: Apply soil improvement strategies to target cultivated land areas, record real-time soil sample data, and update the structural improvement adaptation model through an improved feedback mechanism to achieve improvements in the soil structure of the target cultivated land areas.

2. A soil structure intelligent improvement method according to claim 1, characterized in that: S1, specifically: Soil sampling points were arranged in the target cultivated land area according to a predetermined grid division, and surface soil samples and deep soil samples were collected using the soil profile method; Obtain physical property parameters, chemical property parameters and biological property parameters of surface soil samples and deep soil samples respectively; The physical, chemical and biological parameters of surface soil samples and deep soil samples were correlated and matched according to the spatial coordinates of soil sampling points and the order of sampling time to construct a basic data set.

3. A soil structure intelligent improvement method according to claim 2, characterized in that: S2, specifically: Standardize and normalize the physical, chemical, and biological properties of surface soil samples and deep soil samples in the basic data set; A multidimensional feature mapping model is used to reconstruct the basic data set after standardization and normalization, and a soil structure state map characterized by physical property parameters, chemical property parameters and biological property parameters is established to generate map data that can reflect the current soil structure integrity status of the target cultivated land area.

4. A soil structure intelligent improvement method according to claim 3, characterized in that: S3, specifically: The soil sampling points were divided into several groups with similar structures by multidimensional spatial clustering, and the contribution rate of each principal component to the total variance was calculated by principal component analysis. Based on the ranking of principal component contribution rates from high to low, the main structural variables affecting tillage performance were screened; Key structural deterioration indicators are extracted based on the degree of distribution difference of the main structural variables within the grid.

5. A soil structure intelligent improvement method according to claim 4, characterized in that: S4, specifically: A structural improvement adaptation model is constructed based on key structural deterioration indicators and the root zone environmental requirement parameters of crops to be planted in the target cultivated land area; Through the structural improvement adaptation model, key structural deterioration indicators and crop root zone environmental demand parameters were integrated and analyzed to determine the response degree between different types of soil structure improvement measures and key structural deterioration indicators; The adaptation weights were calculated based on the response degree between different categories of soil structure improvement measures and key structural deterioration indicators to form an adaptation weight relationship matrix between structural improvement measures and soil responses.

6. A soil structure intelligent improvement method according to claim 5, characterized in that: S5, specifically: Based on the adaptive weight relationship matrix between structural improvement measures and soil response, different categories of soil structural improvement measures are matched according to the pre-set adaptive weight threshold; Based on the matching results, an initial set of soil structure improvement strategies is formed; The implementation order and intensity of the initial soil structure improvement strategy set were adjusted through a multi-objective optimization algorithm to obtain the soil structure improvement strategy implemented at the spatial coordinates of each soil sampling point in the target cultivated land area.

7. A soil structure intelligent improvement method according to claim 6, characterized in that: S6, specifically: Implement soil structure improvement strategies at the spatial coordinates of each soil sampling point within the target cultivated land area; Sampling and recording soil sample data at soil sampling points within the target cultivated land area; Compare soil sample data with the basic data set according to the spatial coordinates of soil sampling points and sampling time to evaluate the improvement effect of soil structure improvement strategies; Based on the improvement effect of the soil structure improvement strategy, the structure improvement adaptation model is updated to achieve continuous improvement of the soil structure in the target cultivated land area.

8. A soil structure intelligent improvement system, used to implement a soil structure intelligent improvement method according to any one of claims 1 to 7, characterized in that: include: Data acquisition module: obtains original soil sample data of the target cultivated land area and constructs a basic data set; Data processing module: performs standardization and normalization on the basic data set, and reconstructs the soil structure state map through a multidimensional feature mapping model to obtain map data representing the current soil structure integrity; The selection method of the multidimensional feature mapping model is the nonlinear mapping analysis method; Structure recognition module: This module performs structural pattern recognition on the atlas data to identify key structural deterioration indicators that affect tillage performance; it also performs multidimensional spatial clustering and principal component analysis on the physical, chemical, and biological property parameters of each soil sampling point in the soil structure state atlas data; Adaptation modeling module: Constructs a structural improvement adaptation model, integrates key structural deterioration indicators with crop root zone environmental demand parameters, and generates an adaptation weight relationship matrix between soil structure improvement measures and soil responses; uses fuzzy reasoning methods to construct the structural improvement adaptation model; Strategy generation module: matches soil structure improvement measures according to the adaptive weight relationship matrix, generates soil improvement strategies, and adjusts the implementation sequence and intensity according to the multi-objective optimization algorithm; Improvement execution module: Apply soil improvement strategies to the target cultivated land area, record real-time soil sample data, and update the structural improvement adaptation model through the improvement feedback mechanism to achieve improvement of the soil structure in the target cultivated land area.

Images