A soil improvement and soil quality optimization comprehensive management method based on big data analysis

Abstract

The application discloses a kind of based on big data analysis's soil improvement and soil quality optimization comprehensive management method, comprising the following steps: S1, collection and pre-processing multi-source soil data, construct soil data input matrix;S2, initialize fuzzy C means clustering model parameters, construct model basic structure;S3, introduce the optimization algorithm of hermit crab, optimize clustering model parameters;S4, execute clustering analysis, generate fuzzy soil type distribution atlas;S5, construct improvement strategy scheme space, integrate crop type, historical record and resource information;S6, formulate multi-objective scoring standard, select optimal improvement strategy;S7, deploy optimal improvement strategy to farmland operation system, and real-time acquisition soil feedback data;S8, based on feedback data update model input and parameters, re-execute optimization and realize closed loop.The application fuses big data and intelligent optimization technology, realizes the accurate matching and efficient configuration of soil improvement strategy.

Description

Technical Field

[0001] This invention relates to the field of big data technology, and in particular to a comprehensive management method for soil improvement and soil quality optimization based on big data analysis. Background Technology

[0002] As agricultural production develops towards higher efficiency, intensification, and precision, farmland soil, as the core foundation of the arable land ecosystem, has a decisive impact on crop yield, quality, and the sustainability of the ecological environment. However, long-term irrational farming practices, excessive fertilization, and a single management model have led to soil compaction, organic matter degradation, pH imbalance, and weakened water retention capacity in many parts of my country, seriously affecting agricultural production efficiency and resource utilization efficiency. Therefore, promoting soil improvement and soil quality optimization has become a key aspect of modern agricultural management.

[0003] Current soil improvement technologies primarily rely on expert systems based on soil testing data and field experience, or on rule bases formed from historical improvement records for crop planting management. While these approaches can achieve targeted management to some extent, they have significant limitations: First, traditional methods typically rely on single-point or small-sample data collection, making it difficult to reflect the spatial variability and dynamic trends of soil types in large-scale, diverse farmland areas, leading to uncertainty in strategy application. Second, soil properties exhibit highly nonlinear coupling characteristics; for example, there are complex relationships between properties such as structure, fertility level, water retention capacity, and pH state. Existing methods lack systematic modeling tools to comprehensively analyze the synergistic mechanisms between these variables. Furthermore, traditional management methods often struggle to integrate multiple constraints such as differences in target crop needs, regional resource availability, and economic implementation costs, resulting in soil improvement strategies failing to achieve true site-specific and economically adaptable implementation.

[0004] In recent years, some studies have introduced data mining and machine learning methods to analyze and process soil information. For example, support vector machines and random forests have been attempted for soil classification and crop yield prediction. However, these methods generally focus on classification modeling, lacking a comprehensive management framework for improvement strategies based on multi-source soil attributes. Furthermore, most of these models emphasize static analysis results, lacking a closed-loop management path and failing to form a complete "perception-analysis-decision-feedback" cycle. They are also ill-suited to complex situations such as ambiguous transitions between different plots and unclear soil type boundaries. In addition, most models neglect resource constraints and economic cost assessments in actual agricultural operations, resulting in poor implementation and adaptability of the recommended strategies in practice, failing to meet production management needs.

[0005] In terms of strategy selection, existing systems mostly use rule-based, linear scoring, or expert scoring methods to evaluate the merits of strategies, making it difficult to simultaneously consider the multi-objective balance between soil improvement effects, crop response potential, and input costs. Furthermore, they lack the ability to model uncertainties and fuzziness, making it difficult to handle the conditions of plots with multiple overlapping attributes and blurred boundaries in real soil spaces. At the same time, most studies only address a one-way allocation method of "selecting one strategy for a certain soil category" at the strategy level, failing to design integrated strategies for mixed plots with multiple membership degrees, severely restricting the level of intelligent management of heterogeneous large-scale farmland.

[0006] Therefore, how to provide a comprehensive management method for soil improvement and soil quality optimization based on big data analysis is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0007] One objective of this invention is to propose a comprehensive management method for soil improvement and soil quality optimization based on big data analysis. This invention integrates multi-source heterogeneous data processing technology, fuzzy C-means clustering modeling, an improved hermit crab optimization algorithm, and a multi-objective scoring strategy to systematically construct an intelligent identification and refined decision-making mechanism for soil characteristics of different plots. The invention details the entire process from soil data collection, clustering modeling, strategy selection to personalized fusion and feedback optimization. This method possesses advantages such as strong adaptability, high strategy matching accuracy, controllable improvement effects, and strong intelligent feedback closed-loop capability. It can significantly improve the scientific, economic, and sustainable nature of farmland soil management and has broad prospects for agricultural application.

[0008] According to an embodiment of the present invention, a comprehensive management method for soil improvement and soil quality optimization based on big data analysis includes the following steps:

[0009] S1. Collect multi-source soil data within the farmland area, preprocess the multi-source soil data, and construct a soil data input matrix;

[0010] S2. Initialize the key parameters of the fuzzy C-means clustering model and construct the basic structure of the fuzzy C-means clustering model;

[0011] S3. Introduce the hermit crab optimization algorithm to collaboratively optimize the key parameters of the fuzzy C-means clustering model and obtain the optimized fuzzy C-means clustering model.

[0012] S4. Input the soil data input matrix into the optimized fuzzy C-means clustering model, perform clustering operations, output the membership degree of each cluster category for each plot, and generate a fuzzy soil type distribution map.

[0013] S5. Based on the fuzzy soil type distribution map, construct a space of improvement strategy schemes for various soil types, and generate a set of candidate improvement strategies by combining the target crop type, historical improvement records and resource availability.

[0014] S6. Based on the membership distribution of each plot in the fuzzy soil type distribution map, formulate a multi-objective scoring standard and select the optimal improvement strategy corresponding to each soil type from the candidate improvement strategy set.

[0015] S7. Deploy the optimal improvement strategy to the farmland operation system and collect soil condition feedback data in real time during the execution process, including the time-series changes of target soil parameters and crop growth response.

[0016] S8. Based on soil condition feedback data, update the input data and parameter distribution of the fuzzy C-means clustering model, re-execute the hermit crab optimization algorithm to reconstruct the model, and complete the closed-loop optimization.

[0017] Optionally, the multi-source soil data specifically includes the soil's physicochemical properties, historical application records, and spatial location data, which are used to construct a soil data input matrix.

[0018] Optionally, the preprocessing of the multi-source soil data specifically includes missing value imputation, outlier removal, standardization, and spatial information alignment to improve the quality and consistency of the multi-source soil data.

[0019] Optionally, S2 specifically includes:

[0020] S21. Set the number of cluster categories to guide the fuzzy C-means clustering model to divide soil data into multiple soil type categories. The number of cluster categories determines the number of cluster centers.

[0021] S22. Set a fuzzy factor to adjust the degree of membership of each sample data to different cluster centers during the clustering process. The fuzzy factor value is greater than 1 to control the degree of fuzziness of the clustering results.

[0022] S23. Based on the constructed soil data input matrix, extract the feature vector of each sample and use it as the input data of the fuzzy C-means clustering model to establish the input layer of the fuzzy C-means clustering model.

[0023] S24. Based on the set number of cluster categories, initialize the initial positions of multiple cluster centers in the soil feature space. Each cluster center corresponds to a representative feature configuration of a soil type, which serves as the basis for the clustering objective of the fuzzy C-means clustering model.

[0024] S25. Initialize the sample membership matrix. The membership matrix is ​​used to store the degree of membership of each sample to all cluster centers. The membership value is a real number, and the sum of all membership degrees of each sample is 1.

[0025] S26. Based on the set number of cluster categories, fuzzy factor, initial cluster center positions, and sample membership matrix, construct the basic structure of the fuzzy C-means clustering model, including the model's input layer, cluster center parameter set, and membership calculation mechanism.

[0026] Optionally, S3 specifically includes:

[0027] S31. Initialize the population of the hermit crab optimization algorithm, and divide the population into a main population and multiple sub-populations. The sub-populations include a boundary exploration population, a local refinement population, and an elite memory population, which are used to perform global search, local search, and optimal solution preservation, respectively.

[0028] S32. Randomly assign a set of key parameters of a fuzzy C-means clustering model to each hermit crab individual in the main population to form a parameter set:

[0029] [C i ,m i V i ,μ i ];

[0030] Among them, C i m is the number of clusters. i V is the fuzzy factor. i For the set of cluster centers, μ i This is the membership matrix of the samples;

[0031] S33. Construct a corresponding fuzzy C-means clustering model based on each parameter group, and calculate the clustering error objective function value J. i :

[0032]

[0033] Where, x n For the nth soil sample, v j For the j-th cluster center, μ nj C represents the membership value of the nth sample to the jth cluster center. i This represents the number of cluster categories set for the i-th individual, and N represents the total number of soil samples;

[0034] S34. Calculate the overall fitness function value F of all candidate individuals in the main population. i This includes initially generated individuals, individuals migrating from subgroups, and new individuals generated by shell fusion:

[0035]

[0036] Among them, DBI s S is the Davies-Bouldin index.i The contour coefficient is α, and β are weighting factors.

[0037] S35. Deploy several reconnaissance crab individuals in the boundary region of the solution space. Each reconnaissance crab individual samples multiple candidate points in the boundary region. Evaluate the sampled points based on the fitness function, and select the candidate point with the best performance as the boundary optimum x. s And feed back the boundary optimum to the main group;

[0038] S36. Execute the search strategies of each subgroup: the boundary exploration group updates the parameters with large perturbations, the local refinement group updates the neighborhood of the current best solution with small perturbations, and the elite memory group saves the best solutions of previous generations and participates in the rotation and recombination.

[0039] S37. In the main population, select any two individuals A and B, and calculate the complementarity synergy Comp(A,B):

[0040] Comp(A,B)=γ1·‖x A -x B ‖+γ2·|F A -F B |+γ3·D local ;

[0041] Where, x A x B Let F be a parameter vector. A Let F be the comprehensive fitness function value of individual A. B Let be the comprehensive fitness function value of individual B, γ1 be the weighting coefficient of the solution vector difference term, γ2 be the weighting coefficient of the fitness difference term, γ3 be the weighting coefficient of the local density term, and D be the weighting coefficient of the local density term. local For the local solution density, when Comp(A,B)>θ, shell fusion is performed to generate a new individual, where θ represents the trigger threshold for shell fusion:

[0042] x new =λ o ·x A +(1-λ)·x B ;

[0043] Where, x new λ represents the solution vector of the new individual generated after shell fusion. o The fusion coefficient;

[0044] S38. For all candidate individuals, including new individuals generated after shell fusion, combine the boundary optimum x fed back by the scout crab. s Perform guided location update:

[0045]

[0046] in, This represents the new position of the i-th candidate individual after the (t+1)-th iteration. Let δi be the position of the i-th candidate individual in the current main population at the t-th iteration, and δ1, δ2, and δ3 be the update coefficients. The sample membership matrix representing an individual. The changing trend For the nonlinear interaction function between the individual and the population center, This represents the average of the parameter vectors of all individuals in the current main group;

[0047] S39. Recalculate the fitness function values ​​of all individuals after the position update, perform individual replacement, retention or migration operations on the main population and each subpopulation according to the fitness changes, and update the global optimal solution;

[0048] S310. When the number of iterations reaches the set maximum number of iteration rounds T. max Finally, the parameter combination with the highest fitness is output [C]. * ,m * V * ,μ * ], and construct an optimized fuzzy C-means clustering model.

[0049] Optionally, S4 specifically includes:

[0050] S41. Input the constructed soil data into the matrix X = {x1, x2, ..., x...} N The input x is fed into the optimized fuzzy C-means clustering model. i Let i represent the feature vector of the i-th soil sample, where i = 1, 2, ..., N, and N represents the total number of soil samples.

[0051] S42. Utilize the parameter combination with the highest fitness obtained [C] * ,m * V * ,μ * For each soil sample x i Calculate v for each cluster center j membership degree μ ij :

[0052]

[0053] Where, μ ij C represents the membership value of the i-th sample belonging to the j-th class. * m represents the final number of clusters. * v is the fuzzy factor. j ∈V * Let represent the j-th cluster center vector, and ||| represent the Euclidean distance;

[0054] S43, The construction dimension is N×C * The sample membership matrix μ = {μ ij}, where each row represents the degree of membership of a soil sample in each cluster category;

[0055] S44. Based on the membership matrix μ, generate a fuzzy soil type distribution map, which is used to express the fuzzy attribution relationship distribution of each plot in the farmland on different soil types.

[0056] Optionally, S5 specifically includes:

[0057] S51. Based on the fuzzy soil type distribution map, classify the soil samples according to their respective cluster categories to form multiple soil type groups;

[0058] S52. For each type of soil, construct a preliminary improvement strategy scheme space, which includes a combination of soil improvement measures, covering physical structure conditioning, nutrient regulation, organic matter replenishment, acid-base neutralization and microbial activation.

[0059] S53. Introduce target crop type information and analyze the specific soil requirements of the crop as constraints on the preliminary improvement strategy scheme space.

[0060] S54. Call historical improvement records to conduct correlation analysis on the operation combinations, implementation effects and crop responses of similar soil types in past improvement practices, for reference and optimization of the current strategy construction process;

[0061] S55. Based on the available resources in the current period, including available materials, technical equipment, operational capacity and cost budget, identify the feasibility of implementing various improvement measures in actual farmland, and use this information to limit infeasible or costly strategy options.

[0062] S56. Based on the target crop type, historical improvement records, and resource availability constraints, the preliminary strategy scheme space is filtered, screened, and reconstructed to form a set of candidate improvement strategies suitable for the current agricultural scenario.

[0063] Optionally, the specific soil requirements parameters of the crop include structural adaptability, fertility level, and water retention capacity, which are used to guide the adaptation and matching of improvement strategies with the growth conditions of the target crop.

[0064] Optionally, S6 specifically includes:

[0065] S61. Extract the membership distribution information of each plot in the fuzzy soil type distribution map, and obtain the membership weight of the plot under different soil type categories.

[0066] S62. For each set of candidate improvement strategies corresponding to various soil types, develop a multi-objective scoring standard for comprehensively evaluating the merits of the strategies.

[0067] S63. For the candidate improvement strategies for various types of soil, evaluate them item by item according to the established multi-objective scoring criteria to form corresponding strategy scoring results;

[0068] S64. In the strategy set for each soil type, select the strategy with the best strategy score as the optimal improvement strategy for the soil type category.

[0069] S65. Combining the membership weights of each plot under different soil type categories, the selected optimal improvement strategies for each category are weighted and integrated to generate a plot-level integrated personalized improvement strategy.

[0070] Optionally, the multi-objective scoring criteria specifically include soil improvement effectiveness, target crop response performance, and implementation cost, comprehensively measuring the adaptability and superiority of candidate improvement strategies.

[0071] The beneficial effects of this invention are:

[0072] This invention integrates big data analytics, fuzzy clustering modeling, and intelligent optimization algorithms to achieve comprehensive management and intelligent decision-making for farmland soil improvement and soil quality optimization. It significantly overcomes the shortcomings of existing technologies, such as coarse soil classification, singular strategy matching, lack of personalization, and weak feedback mechanisms. By collecting a large amount of multi-source soil data from farmland areas and performing standardized preprocessing, a high-quality soil data input matrix is ​​constructed. This effectively overcomes the information loss problem caused by traditional technologies relying on small samples and point data, providing a comprehensive and detailed data foundation for subsequent analysis. The introduction of a fuzzy C-means clustering model combined with the hermit crab optimization algorithm for collaborative optimization of clustering parameters enhances the spatial continuity and boundary fuzziness recognition capabilities of soil type classification, significantly improving the accuracy and stability of soil classification. This is particularly suitable for real-world scenarios with complex plot transitions and overlapping soil properties.

[0073] This invention further integrates key elements such as crop growth needs, historical improvement experience, and resource availability to construct a multi-factor driven strategy set. It comprehensively evaluates various strategies by setting a multi-objective scoring standard that includes soil improvement effectiveness, target crop response performance, and implementation costs. This mechanism effectively solves the shortcomings of traditional methods that select the best strategy based on a single dimension or rule-based scoring, achieving a balance between improvement effects, crop benefits, and economic feasibility. Furthermore, based on the membership distribution of each plot in the fuzzy clustering results, this invention innovatively weights and fuses the optimal strategies to generate personalized soil improvement plans for specific plots, fully embodying the modern agricultural management concept of "adapting to local conditions, implementing differentiated policies, and precise regulation."

[0074] Through the above improvements, this invention constructs a complete closed-loop management process encompassing data collection, intelligent identification, strategy construction, optimization screening, and personalized integration. It possesses significant advantages such as full information utilization, high intelligence, accurate strategy matching, and strong feasibility of execution. This method can be widely applied to scenarios such as farmland soil improvement, arable land quality enhancement, and optimal allocation of agricultural production resources, demonstrating significant practical value and promotion potential. It can effectively support the achievement of high-quality agricultural development and green production goals. Attached Figure Description

[0075] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0076] Figure 1 This is a flowchart of a comprehensive management method for soil improvement and soil quality optimization based on big data analysis proposed in this invention;

[0077] Figure 2 This is a schematic diagram of the processing flow for constructing a soil type distribution map based on a fuzzy C-means clustering model, which is part of the integrated management method for soil improvement and soil quality optimization based on big data analysis proposed in this invention.

[0078] Figure 3 This invention presents a collaborative optimization structure diagram of a soil improvement and soil quality optimization integrated management method based on big data analysis, which combines the hermit crab optimization algorithm to optimize fuzzy clustering parameters. Detailed Implementation

[0079] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0080] refer to Figure 1 , Figure 2 and Figure 3A comprehensive management method for soil improvement and soil quality optimization based on big data analysis includes the following steps:

[0081] S1. Collect multi-source soil data within the farmland area, preprocess the multi-source soil data, and construct a soil data input matrix;

[0082] S2. Initialize the key parameters of the fuzzy C-means clustering model and construct the basic structure of the fuzzy C-means clustering model;

[0083] S3. Introduce the hermit crab optimization algorithm to collaboratively optimize the key parameters of the fuzzy C-means clustering model and obtain the optimized fuzzy C-means clustering model.

[0084] S4. Input the soil data input matrix into the optimized fuzzy C-means clustering model, perform clustering operations, output the membership degree of each cluster category for each plot, and generate a fuzzy soil type distribution map.

[0085] S5. Based on the fuzzy soil type distribution map, construct a space of improvement strategy schemes for various soil types, and generate a set of candidate improvement strategies by combining the target crop type, historical improvement records and resource availability.

[0086] S6. Based on the membership distribution of each plot in the fuzzy soil type distribution map, formulate a multi-objective scoring standard and select the optimal improvement strategy corresponding to each soil type from the candidate improvement strategy set.

[0087] S7. Deploy the optimal improvement strategy to the farmland operation system and collect soil condition feedback data in real time during the execution process, including the time-series changes of target soil parameters and crop growth response.

[0088] S8. Based on soil condition feedback data, update the input data and parameter distribution of the fuzzy C-means clustering model, re-execute the hermit crab optimization algorithm to reconstruct the model, and complete the closed-loop optimization.

[0089] This invention provides a comprehensive management method for soil improvement and soil quality optimization based on big data analysis. By constructing an intelligent analysis system with a fuzzy C-means clustering model at its core and incorporating a hermit crab optimization algorithm for parameter optimization, it achieves in-depth mining and refined modeling of multi-source soil data, accurately identifying the distribution characteristics of different soil types in farmland. By generating fuzzy soil type distribution maps, it reflects the membership information of soil types in each plot, providing a basis for subsequent strategy matching. At the strategy formulation level, the method integrates target crop needs, historical improvement records, and resource availability information, establishing a multi-objective scoring mechanism to comprehensively evaluate the feasibility and superiority of candidate strategies, achieving personalized recommendations tailored to local conditions. The system possesses closed-loop optimization capabilities, dynamically adjusting model parameters and improvement schemes based on soil condition feedback, continuously improving management accuracy and responsiveness. Compared to traditional methods, this invention has significant advantages such as high modeling accuracy, scientific strategy matching, strong execution flexibility, and a complete decision-making closed loop, significantly improving soil improvement efficiency and crop yield, and promoting the intelligent and sustainable development of agriculture.

[0090] In this embodiment, the multi-source soil data specifically includes the soil's physical and chemical properties, historical application records, and spatial location data, which are used to construct a soil data input matrix.

[0091] In this embodiment, the preprocessing of the multi-source soil data specifically includes missing value filling, outlier removal, standardization processing, and spatial information alignment, which are used to improve the quality and consistency of the multi-source soil data.

[0092] In this embodiment, S2 specifically includes:

[0093] S21. Set the number of cluster categories to guide the fuzzy C-means clustering model to divide soil data into multiple soil type categories. The number of cluster categories determines the number of cluster centers.

[0094] S22. Set a fuzzy factor to adjust the degree of membership of each sample data to different cluster centers during the clustering process. The fuzzy factor value is greater than 1 to control the degree of fuzziness of the clustering results.

[0095] S23. Based on the constructed soil data input matrix, extract the feature vector of each sample and use it as the input data of the fuzzy C-means clustering model to establish the input layer of the fuzzy C-means clustering model.

[0096] S24. Based on the set number of cluster categories, initialize the initial positions of multiple cluster centers in the soil feature space. Each cluster center corresponds to a representative feature configuration of a soil type, which serves as the basis for the clustering objective of the fuzzy C-means clustering model.

[0097] S25. Initialize the sample membership matrix. The membership matrix is ​​used to store the degree of membership of each sample to all cluster centers. The membership value is a real number, and the sum of all membership degrees of each sample is 1.

[0098] S26. Based on the set number of cluster categories, fuzzy factor, initial cluster center positions, and sample membership matrix, construct the basic structure of the fuzzy C-means clustering model, including the model's input layer, cluster center parameter set, and membership calculation mechanism.

[0099] The initialization and structure construction steps of the fuzzy C-means clustering model in this invention effectively improve the scientific rigor and clustering accuracy of the clustering model in soil type identification by systematically setting key parameters such as the number of clusters, fuzziness factor, sample input features, initial cluster centers, and membership matrix. A reasonable setting of the number of clusters ensures the model has sufficient expressive power to cover diverse soil type distributions; the introduction of the fuzziness factor enhances the model's flexible expression of boundary samples, avoiding misjudgment problems caused by traditional hard classification; and by initializing the sample membership matrix and cluster center positions, a reasonable starting point for clustering is constructed, improving the model's stability and convergence speed during iterative solution. This model initialization process ensures stronger robustness and adaptability of the subsequent clustering process in complex, high-dimensional soil feature spaces, accurately identifying the fuzzy transition characteristics of soil types in farmland. It lays a solid data foundation and model structure guarantee for subsequent intelligent decision-making, strategy matching, and dynamic optimization, and has broad application value.

[0100] In this embodiment, S3 specifically includes:

[0101] S31. Initialize the population of the hermit crab optimization algorithm, and divide the population into a main population and multiple sub-populations. The sub-populations include a boundary exploration population, a local refinement population, and an elite memory population, which are used to perform global search, local search, and optimal solution preservation, respectively.

[0102] S32. Randomly assign a set of key parameters of a fuzzy C-means clustering model to each hermit crab individual in the main population to form a parameter set:

[0103] [C i ,m i V i ,μ i ];

[0104] Among them, C i m is the number of clusters. i V is the fuzzy factor. i For the set of cluster centers, μ i This is the membership matrix of the samples;

[0105] S33. Construct a corresponding fuzzy C-means clustering model based on each parameter group, and calculate the clustering error objective function value J. i :

[0106]

[0107] Where, x n For the nth soil sample, v j For the j-th cluster center, μ nj C represents the membership value of the nth sample to the jth cluster center. i This represents the number of cluster categories set for the i-th individual, and N represents the total number of soil samples;

[0108] S34. Calculate the overall fitness function value F of all candidate individuals in the main population. i This includes initially generated individuals, individuals migrating from subgroups, and new individuals generated by shell fusion:

[0109]

[0110] Among them, DBI s S is the Davies-Bouldin index. i The contour coefficient is α, and β are weighting factors.

[0111] S35. Deploy several reconnaissance crab individuals in the boundary region of the solution space. Each reconnaissance crab individual samples multiple candidate points in the boundary region. Evaluate the sampled points based on the fitness function, and select the candidate point with the best performance as the boundary optimum x. s And feed back the boundary optimum to the main group;

[0112] S36. Execute the search strategies of each subgroup: the boundary exploration group updates the parameters with large perturbations, the local refinement group updates the neighborhood of the current best solution with small perturbations, and the elite memory group saves the best solutions of previous generations and participates in the rotation and recombination.

[0113] S37. In the main population, select any two individuals A and B, and calculate the complementarity synergy Comp(A,B):

[0114] Comp(A,B)=γ1·‖x A -x B ‖+γ2·|F A -F B |+γ3·D local ;

[0115] Where, x A x B Let F be a parameter vector. A Let F be the comprehensive fitness function value of individual A. BLet be the comprehensive fitness function value of individual B, γ1 be the weighting coefficient of the solution vector difference term, γ2 be the weighting coefficient of the fitness difference term, γ3 be the weighting coefficient of the local density term, and D be the weighting coefficient of the local density term. local For the local solution density, when Comp(A,B)>θ, shell fusion is performed to generate a new individual, where θ represents the trigger threshold for shell fusion:

[0116] x new =λ o ·x A +(1-λ)·x B ;

[0117] Where, x new λ represents the solution vector of the new individual generated after shell fusion. o The fusion coefficient;

[0118] S38. For all candidate individuals, including new individuals generated after shell fusion, combine the boundary optimum x fed back by the scout crab. s Perform guided location update:

[0119]

[0120] in, This represents the new position of the i-th candidate individual after the (t+1)-th iteration. Let δi be the position of the i-th candidate individual in the current main population at the t-th iteration, and δ1, δ2, and δ3 be the update coefficients. The sample membership matrix representing an individual. The changing trend For the nonlinear interaction function between the individual and the population center, This represents the average of the parameter vectors of all individuals in the current main group;

[0121] S39. Recalculate the fitness function values ​​of all individuals after the position update, perform individual replacement, retention or migration operations on the main population and each subpopulation according to the fitness changes, and update the global optimal solution;

[0122] S310. When the number of iterations reaches the set maximum number of iteration rounds T. max Finally, the parameter combination with the highest fitness is output [C]. * ,m * V * ,μ * ], and construct an optimized fuzzy C-means clustering model.

[0123] This invention significantly improves the adaptability and clustering accuracy of a fuzzy C-means clustering model by introducing an improved hermit crab optimization algorithm to collaboratively optimize key parameters. The algorithm innovatively constructs a co-evolutionary structure of the main population and multiple sub-populations, with sub-populations undertaking boundary exploration, local refinement, and elite retention tasks respectively, thus achieving a dynamic balance between global search and local convergence capabilities. By defining a comprehensive fitness function and integrating clustering quality indices such as the Davies-Bouldin index and silhouette coefficient, the clustering effect of individual solutions is comprehensively evaluated. Furthermore, a scout crab mechanism is introduced, deploying feedback points at the solution space boundary to effectively prevent the algorithm from getting trapped in local optima. A complementary synergy judgment mechanism is proposed, combining solution vector differences, fitness differences, and local density to dynamically trigger shell fusion operations, enhancing population diversity and improving population evolution quality. In the individual position update strategy, boundary optimum guidance, individual historical trends, adaptive membership perturbation, and nonlinear interaction mechanisms are integrated to achieve fine-grained control and rapid optimization of the parameter space. Finally, through iterative control strategies, the globally optimal parameter set is stably output. The optimized clustering model has stronger stability and generalization ability, and can more accurately characterize the fuzzy type distribution of farmland soil, providing a high-quality modeling foundation for strategy matching and personalized improvement.

[0124] In this embodiment, S4 specifically includes:

[0125] S41. Input the constructed soil data into the matrix X = {x1, x2, ..., x...} N The input x is fed into the optimized fuzzy C-means clustering model. i Let i represent the feature vector of the i-th soil sample, where i = 1, 2, ..., N, and N represents the total number of soil samples.

[0126] S42. Utilize the parameter combination with the highest fitness obtained [C] * ,m * V * ,μ * For each soil sample x i Calculate v for each cluster center j membership degree μ ij :

[0127]

[0128] Where, μ ij C represents the membership value of the i-th sample belonging to the j-th class. * m represents the final number of clusters. * v is the fuzzy factor. j ∈V * Let represent the j-th cluster center vector, and ||| represent the Euclidean distance;

[0129] S43, The construction dimension is N×C * The sample membership matrix μ = {μ ij}, where each row represents the degree of membership of a soil sample in each cluster category;

[0130] S44. Based on the membership matrix μ, generate a fuzzy soil type distribution map, which is used to express the fuzzy attribution relationship distribution of each plot in the farmland on different soil types.

[0131] This invention applies an optimized fuzzy C-means clustering model to a high-dimensional soil data input matrix, accurately calculating the membership distribution of each plot sample under various soil types, significantly improving the expressiveness and practical adaptability of soil type classification. Compared to traditional hard clustering or single-category classification methods, this invention introduces the concept of fuzzy membership during the clustering process, which not only reflects the transitional and mixed characteristics between soil types but also effectively avoids the problem of inaccurate classification of boundary samples. By constructing a membership matrix, the system comprehensively expresses the fuzzy membership degree of each sample under different cluster centers, enhancing the modeling ability of soil spatial continuity characteristics. Furthermore, the generated fuzzy soil type distribution map not only possesses high resolution and stability but can also serve as a basis for subsequent improvement strategy allocation, crop planting structure optimization, and field management decisions. This map presents the multi-dimensional attribute fusion characteristics of soil classification in a data-driven manner, realizing dynamic expression from point data to the plot level, effectively improving the accuracy and intelligence of agricultural management, and has significant practical value and scalability.

[0132] In this embodiment, S5 specifically includes:

[0133] S51. Based on the fuzzy soil type distribution map, classify the soil samples according to their respective cluster categories to form multiple soil type groups;

[0134] S52. For each type of soil, construct a preliminary improvement strategy scheme space, which includes a combination of soil improvement measures, covering physical structure conditioning, nutrient regulation, organic matter replenishment, acid-base neutralization and microbial activation.

[0135] S53. Introduce target crop type information and analyze the specific soil requirements of the crop as constraints on the preliminary improvement strategy scheme space.

[0136] S54. Call historical improvement records to conduct correlation analysis on the operation combinations, implementation effects and crop responses of similar soil types in past improvement practices, for reference and optimization of the current strategy construction process;

[0137] S55. Based on the available resources in the current period, including available materials, technical equipment, operational capacity and cost budget, identify the feasibility of implementing various improvement measures in actual farmland, and use this information to limit infeasible or costly strategy options.

[0138] S56. Based on the target crop type, historical improvement records, and resource availability constraints, the preliminary strategy scheme space is filtered, screened, and reconstructed to form a set of candidate improvement strategies suitable for the current agricultural scenario.

[0139] This invention constructs a multi-soil improvement strategy space based on a fuzzy soil type distribution map, systematically integrating target crop needs, historical improvement experience, and resource availability, significantly improving the scientific rigor, relevance, and feasibility of soil improvement strategy construction. By grouping soil samples under different cluster categories, the underlying differences between soil types are clarified, constructing a multi-dimensional, combinable strategy space that effectively supports site-specific and categorized regulation. Introducing crop-specific requirement parameters ensures that strategy construction not only focuses on the soil's physicochemical characteristics but also considers biological coupling relationships such as crop root structure and nutrient absorption mechanisms, improving the strategy's suitability for crop growth. Simultaneously, by combining historical improvement records, a data-driven approach is used to uncover successful experiences and lessons learned, achieving experience accumulation and knowledge transfer in strategy design and reducing trial-and-error costs. Furthermore, various resource factors such as materials, equipment, labor, and budget are incorporated into a feasibility filtering logic, effectively avoiding impractical theoretical solutions and enhancing the strategy's practicality and executability. The final set of candidate improvement strategies combines agricultural science with engineering constraints, forming a well-structured, multi-faceted, and highly adaptable strategy foundation that supports subsequent intelligent evaluation and precise deployment, demonstrating significant application value and potential for implementation.

[0140] In this embodiment, the specific soil requirement parameters of the crop include structural adaptability, fertility level, and water retention capacity, which are used to guide the adaptation and matching of improvement strategies with the growth conditions of the target crop.

[0141] In this embodiment, S6 specifically includes:

[0142] S61. Extract the membership distribution information of each plot in the fuzzy soil type distribution map, and obtain the membership weight of the plot under different soil type categories.

[0143] S62. For each set of candidate improvement strategies corresponding to various soil types, develop a multi-objective scoring standard for comprehensively evaluating the merits of the strategies.

[0144] S63. For the candidate improvement strategies for various types of soil, evaluate them item by item according to the established multi-objective scoring criteria to form corresponding strategy scoring results;

[0145] S64. In the strategy set for each soil type, select the strategy with the best strategy score as the optimal improvement strategy for the soil type category.

[0146] S65. Combining the membership weights of each plot under different soil type categories, the selected optimal improvement strategies for each category are weighted and integrated to generate a plot-level integrated personalized improvement strategy.

[0147] This invention, by introducing a multi-objective scoring mechanism and a fuzzy membership weighting strategy, achieves a precise decision-making process from soil type identification to personalized improvement strategy generation, significantly improving the scientific rigor and plot-level adaptability of soil improvement. By extracting the membership weights of each plot in the fuzzy soil type distribution map, it comprehensively reflects the mixed characteristics of different plots for multiple soil attributes, breaking through the traditional rigid "one plot, one type" classification method and effectively addressing the issues of continuity and fuzzy boundaries in soil type distribution. In the candidate improvement strategy evaluation stage, a multi-objective scoring standard is constructed with soil improvement effect, crop response potential, and implementation cost as its core, ensuring that the selected strategies possess ecological adaptability, economic feasibility, and production benefits. Based on this, the optimal strategies for different soil types are weighted and fused, combined with the membership degree of each plot, to achieve plot-specific strategy personalization reconstruction, ensuring that the improvement plan obtained for each plot highly matches its soil condition characteristics and crop growth needs. This method fully embodies the concept of "precision agriculture," ensuring both the scientific rigor of strategy allocation and improving the execution effect of actual deployment, providing strong decision support for intelligent agricultural management.

[0148] In this embodiment, the multi-objective scoring criteria specifically include soil improvement effectiveness, target crop response performance, and implementation cost, comprehensively measuring the adaptability and superiority of candidate improvement strategies.

[0149] Example 1:

[0150] To verify the feasibility of this invention in practice, it was applied to a farmland experimental area. Due to long-term improper application of chemical fertilizers and over-cultivation, the soil generally suffered from problems such as acid-base imbalance, low organic matter content, and poor water retention capacity, resulting in a harsh crop growth environment and average crop yields far below expectations. To improve this situation, this embodiment adopts the integrated management method for soil improvement and soil quality optimization based on big data analysis proposed in this invention. This method first collects a large amount of soil data from the experimental field, using multiple deployed sensors to monitor soil pH, organic matter content, moisture content, electrical conductivity, and other key physicochemical indicators in real time. Simultaneously, crop planting information and historical application records are collected, and a unified input matrix is ​​obtained after data preprocessing. Based on this data, a soil type distribution map is constructed using the fuzzy C-means clustering method, dividing the farmland soil into multiple fuzzy categories, thus reflecting the gradual and overlapping characteristics between plots. Subsequently, this invention uses an improved hermit crab optimization algorithm to collaboratively optimize the key parameters of the clustering model, making the clustering results more refined and accurate, and enabling accurate estimation of the membership degree of each soil type in each plot.

[0151] Based on this, a set of candidate soil improvement strategies was constructed according to soil type distribution maps. Taking into account specific crop requirements for the soil, such as structural adaptability, fertility level, and water retention capacity, and integrating factors such as historical improvement experience, resource availability, and actual input costs, a multi-objective scoring standard was developed to comprehensively evaluate the candidate strategies. The optimal strategy was then selected, and a weighted fusion of the membership degrees of different plots under different soil types was used to generate personalized soil improvement plans. Experimental results show that this integrated management method can significantly improve soil physicochemical properties while promoting healthy crop growth and yield increase.

[0152] Table 1 Comparison of key soil indicators and crop yields before and after farmland improvement.

[0153]

[0154] As shown in Table 1, the method of this invention has achieved significant results in improving soil quality and crop yield. Firstly, regarding soil pH, the pH value increased from 5.2 to 6.8, an improvement of 30.8%, effectively alleviating the problem of acidic soil and bringing the soil environment closer to neutral, making it more suitable for the growth of most crops. Secondly, the organic matter content increased from 1.9% to 3.2%, an increase of 68.4%, indicating that the application of organic amendments such as humus and organic fertilizers greatly enhanced soil fertility, providing crops with a more stable and continuous nutrient supply.

[0155] In terms of water retention capacity, soil moisture content increased from 12% before improvement to 18%, an increase of 50.0%, indicating that optimizing tillage structure and applying biochar or mulch effectively enhanced the soil's water retention capacity and reduced the occurrence of soil drought stress. Electrical conductivity decreased from 300 μS / cm to 180 μS / cm, an improvement of 40.0%. This reduction signifies a decrease in soil salinity, a more stable electrolyte environment, and further improvements in soil aeration and root growth conditions.

[0156] The most direct benefit is reflected in crop yield. Before implementation, the crop yield per unit area was 2800 kg / ha, and after applying the method of this invention, the yield increased to 4200 kg / ha, an increase of 50.0%. This data not only reflects the positive feedback of improved soil physicochemical properties but also fully verifies the scientific nature and adaptability of the improvement strategy. This invention, through a fuzzy clustering model constructed using big data analysis and an intelligent strategy scoring and optimization mechanism, successfully achieves site-specific and differentiated allocation of improvement resources, making farmland management more refined and efficient.

[0157] In summary, the data presented in this table fully demonstrates the comprehensive advantages of this invention in improving basic soil conditions, enhancing the farming environment, and promoting crop yield increases. It provides agricultural production with a highly practical, intelligent, and economical method for soil improvement and optimization management. This method also has broad application value and practical significance in a wider range of and more complex environments.

[0158] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A comprehensive management method for soil improvement and soil quality optimization based on big data analysis, characterized in that, The method comprises the following steps: S1, collecting multi-source soil data in a farmland area, and preprocessing the multi-source soil data to construct a soil data input matrix; S2, initializing key parameters of a fuzzy C-means clustering model, and constructing a basic structure of the fuzzy C-means clustering model; S3, introducing a hermit crab optimization algorithm to cooperatively optimize the key parameters of the fuzzy C-means clustering model, and obtaining an optimized fuzzy C-means clustering model; S4, inputting the soil data input matrix into the optimized fuzzy C-means clustering model, performing clustering operation, outputting the membership degree of each plot corresponding to each clustering category, and generating a fuzzy soil type distribution map; S5, based on the fuzzy soil type distribution map, constructing an improvement strategy scheme space for each type of soil, combining a target crop type, historical improvement records and resource availability, and generating a candidate improvement strategy set; S6, according to the membership degree distribution of each plot in the fuzzy soil type distribution map, formulating a multi-objective scoring standard, and selecting an optimal improvement strategy corresponding to each soil type in the candidate improvement strategy set; S7, deploying the optimal improvement strategy to a farmland operation system, and collecting soil state feedback data in real time during the execution process, including time series change values of target soil parameters and crop growth response; S8, based on the soil state feedback data, updating the input data and parameter distribution of the fuzzy C-means clustering model, re-executing the hermit crab optimization algorithm to reconstruct the model, and completing closed-loop optimization.

2. The method according to claim 1, wherein the method is characterized by, The multi-source soil data specifically includes soil physical and chemical properties, historical improvement records and spatial location data, and is used to construct the soil data input matrix.

3. The method according to claim 1, wherein the method is characterized by, The preprocessing of the multi-source soil data specifically includes missing value filling, outlier removal, standardization processing and spatial information alignment, and is used to improve the quality and consistency of the multi-source soil data.

4. The method according to claim 1, wherein the method is characterized by, The S2 specifically includes: S21, setting the number of clustering categories, which is used to guide the fuzzy C-means clustering model to divide the soil data into multiple soil type categories, and the number of clustering categories determines the number of clustering centers; S22, setting a fuzzy factor, which is used to adjust the membership degree distribution of each sample data to different clustering centers during the clustering process, and the fuzzy factor is greater than 1, which controls the fuzzy degree of the clustering result; S23, based on the constructed soil data input matrix, extracting the feature vector of each sample as the input data of the fuzzy C-means clustering model, and establishing the input layer of the fuzzy C-means clustering model; S24, according to the set number of clustering categories, initializing the initial positions of multiple clustering centers in the soil feature space, each clustering center corresponding to a representative feature configuration of a soil type, as the basis of the clustering target of the fuzzy C-means clustering model; S25, initializing a sample membership degree matrix, which is used to store the membership degree of each sample to all clustering centers, and the membership degree value is a real number, and the sum of all membership degrees of each sample is 1; S26, according to the set number of clustering categories, the fuzzy factor, the initial clustering center position and the sample membership degree matrix, constructing the basic structure of the fuzzy C-means clustering model, including the input layer of the model, the clustering center parameter set and the membership degree calculation mechanism.

5. The method for soil improvement and soil quality optimization based on big data analysis according to claim 1, characterized in that, The S3 specifically includes: S31, initialize the population of the hermit crab optimization algorithm, divide the population into a main group and multiple subgroups, the subgroups include boundary exploration group, local refinement group and elite memory group, for performing global search, local search and optimal solution reservation respectively; S32, randomly assign a set of key parameters of fuzzy C-means clustering model to each hermit crab individual in the main group to form a parameter set: ； wherein, is the number of cluster classes, is the fuzzifier, is the set of cluster centers, is the sample membership matrix; S33, constructing a corresponding fuzzy C-means clustering model based on each parameter group, and calculating a clustering error objective function value : ； in, For the first One soil sample, For the first Cluster centers, This represents the membership value of the nth sample to the jth cluster center. This represents the number of cluster categories set for the i-th individual. This indicates the total number of soil samples; S34, calculate the comprehensive fitness function value of all candidate individuals in the main group including the initial generated individuals, the sub-group migrated individuals and the new individuals generated by the shell fusion: ； wherein is the Davies-Bouldin index, is the silhouette coefficient, , is the weighting factor; S35、in the solution space boundary region, deploy several scout crab individuals, each of which samples multiple candidate points in the boundary region, evaluates the sampled points based on a fitness function, and selects the best-performing candidate point as the boundary optimal point and feeds the boundary optimal point back to the main population; S36, execute the search strategy of each subgroup: the boundary exploration group updates the parameters with large disturbance, the local refinement group updates the current optimal solution neighborhood with small disturbance, and the elite memory group saves the historical optimal solution and participates in rotation and recombination; S37. In the main population, select any two individuals With , calculate the complementary synergy : ； wherein, , is a parameter vector, is a fitness function value of the individual A, is a fitness function value of the individual , is a weighting coefficient of the solution vector difference term, is a weighting coefficient of the fitness difference term, is a weighting coefficient of the local density term, is a local solution density, and when a shell fusion is performed to generate a new individual, wherein, denotes a trigger threshold for the shell fusion; ； wherein, denotes the unit vector of the new individual generated after the fusion of the shells, is the fusion coefficient; S38, for all candidate individuals, including new individuals generated after shell fusion, combine the feedback from the boundary optimal points of the scout crabs perform a guided position update: ； wherein, denotes the new position of the i-th candidate individual after the t+1 iteration, is the position of the i-th candidate individual in the current main population at the t iteration, , and is the update coefficient, denotes the change trend of the sample membership matrix corresponding to the individual , is the non-linear interaction function between the individual and the population center, is the average of all individual parameter vectors of the current main population; S39, recalculate the fitness function value of all individuals after position update, replace, retain or migrate the individuals in the main group and each subgroup according to the fitness change, and update the global optimal solution; S310、When the number of iterations reaches the set maximum number of iterations Finally, the parameter combination with the highest fitness is output And the optimized fuzzy C-means clustering model is constructed.

6. The method according to claim 1, wherein the method is characterized by, The S4 specifically comprises: S41, inputting the constructed soil data into a matrix into the fuzzy C-means clustering model obtained by optimization, indicates the feature vector of the th soil sample, , indicates the total number of soil samples; S42, using the obtained parameter combination with the highest fitness For each soil sample Calculate the membership to each cluster center : ； wherein, represents a membership value of the th sample belonging to the th cluster, is the final number of clustering categories, is a fuzzy factor, represents the th clustering center vector, represents the Euclidean distance; S43, constructing a sample membership matrix of dimension where each row represents the degree of membership of a soil sample under each cluster class; where each row represents the degree of membership of a soil sample under each cluster class; S44、based on the membership matrix generate a fuzzy soil type distribution map for expressing the fuzzy attribution relationship distribution of each plot in the farmland on different soil types.

7. The method according to claim 1, wherein the method is characterized by, The S5 specifically comprises: S51, according to the fuzzy soil type distribution map, classify the soil samples according to the clustering categories to which they belong to form multiple soil type groups; S52, for each soil type, construct a preliminary improvement strategy scheme space, which includes soil improvement measure combination, covering physical structure conditioning, nutrient regulation, organic matter supplement, acid-base neutralization and microbial activation; S53, introduce the target crop type information, analyze the specific demand parameters of crops for soil as constraint conditions for the preliminary improvement strategy scheme space; S54, call the historical improvement record data, correlate the operation combination, implementation effect and crop response of similar soil types in past improvement practices for reference and optimization of the current strategy construction process; S55, combined with the available resources in the current period, including available materials, technical equipment, operation capacity and cost budget, identify the implementation feasibility of various improvement measures in actual farmland, which is used to limit the strategy options that are not feasible or too expensive; S56, filter, screen and reconstruct the preliminary strategy scheme space to form a candidate improvement strategy set suitable for the current agricultural scene by comprehensively considering the target crop type, historical improvement record and resource availability constraints.

8. The method according to claim 7, wherein the method is characterized by, The specific demand parameters of crops for soil specifically include structure adaptability, fertility level, and water retention capacity, which are used to guide the adaptation and matching of improvement strategies and target crop growth conditions.

9. The method according to claim 1, wherein the method is characterized by, The S6 specifically comprises: S61, extract the membership degree distribution information corresponding to each plot in the fuzzy soil type distribution map to obtain the membership degree weight of the plot under different soil type categories; S62, for the candidate improvement strategy set corresponding to each soil type, develop a multi-objective scoring standard for comprehensive evaluation of strategy advantages and disadvantages; S63, evaluate each soil candidate improvement strategy set according to the developed multi-objective scoring standard to form the corresponding strategy scoring result; S64, in the strategy set of each soil type, select the strategy with the optimal strategy scoring result as the optimal improvement strategy of the soil type category; S65, combine the membership weight of each plot under different soil type categories, and weight and integrate the selected optimal improvement strategies of each category to generate a fusion personalized improvement strategy at the plot level.

10. The method according to claim 9, wherein the method is characterized by, The multi-target scoring standard specifically includes soil improvement effectiveness, target crop response performance, and implementation cost, and comprehensively measures the adaptability and superiority of the candidate improvement strategy.

Images