Soil monitoring method for fruit honey black tea planting

Abstract

The invention discloses a soil monitoring method for fruit honey black tea planting, and aims to improve the yield and quality of fruit honey black tea by accurately monitoring and dynamically managing and optimizing soil conditions. The method comprises the following steps: S1, arranging soil sampling points and generating a preliminary data set; s2, monitoring soil physical parameters in real time, and generating and cleaning a physical parameter data set; s3, integrating the data to form a comprehensive soil physical state data set; s4, applying multiple excitation response spectral imaging to generate a spectral data set; s5, integrating the spectral data set and the physical state data set into a soil comprehensive data set; s6, analyzing by using a self-adaptive space-time ripple model and a non-exchangeable algebraic topology method to generate a soil health assessment data set; s7, generating fertilization and irrigation suggestions according to the evaluation data, and transmitting the suggestions to the field management equipment; s8, continuously updating the evaluation data to adapt to environmental changes; and S9, recording and storing all data for future planting decisions.

Description

Technical Field

[0001] The invention relates to the technical field of soil monitoring, and in particular to a soil monitoring method for planting fruit honey black tea. Background Art

[0002] In the process of growing fruit honey black tea, soil management directly affects the quality and yield of tea. Existing soil monitoring methods mainly rely on single physical or chemical indicators, such as soil moisture, pH value, nutrient content, etc. In practical applications, these methods often have problems such as delayed data collection, single monitoring indicators, and incomplete analysis results. It is difficult to fully and timely reflect the health status of the soil, thus limiting the accuracy and scientificity of agricultural management.

[0003] First, traditional soil monitoring methods are mainly based on manual sampling and laboratory analysis. Although this method can obtain certain soil information, the sampling and analysis process is cumbersome, resulting in significant data lag, making it difficult to meet the requirements of real-time monitoring of soil conditions for crop growth. Especially in complex planting environments, changes in soil conditions have significant temporal and spatial heterogeneity, and traditional methods cannot track these changes in real time.

[0004] Secondly, with the development of sensor technology, modern agriculture has gradually introduced sensor-based soil monitoring systems. These systems can collect physical parameters such as soil temperature, humidity, and pH value in real time, providing data support for agricultural management. However, these sensors can usually only monitor a single physical indicator, making it difficult to form a comprehensive assessment of soil health, especially when it comes to monitoring soil chemical composition and biological activity, which is often beyond the capabilities of existing sensor systems.

[0005] Third, existing soil monitoring methods lack the ability to deeply analyze and comprehensively process data, which limits the application value of monitoring results. Soil health depends not only on physical parameters, but also on complex chemical and biological processes. A single physical monitoring indicator is difficult to reflect these complex soil dynamics. Although some methods have tried to solve this problem through multi-source data fusion and big data analysis technology, the application of these technologies in agricultural production is not mature enough and lacks optimization for specific crops and environments.

[0006] In summary, the existing soil monitoring technology has many deficiencies in practical applications, mainly manifested in the problems of poor data real-time performance, single monitoring indicators, and limited analysis capabilities. Therefore, how to provide a soil monitoring method for fruit honey black tea planting is an urgent problem to be solved by those skilled in the art. Summary of the invention

[0007] One purpose of the present invention is to propose a soil monitoring method for fruit honey black tea planting. By comprehensively utilizing multiple excitation response spectral imaging, adaptive spatiotemporal ripple model and non-commutative algebraic topology method, the real-time monitoring and dynamic analysis of soil health status are described in detail. It has the advantages of high accuracy, strong real-time performance and wide adaptability, effectively improves the yield and quality of fruit honey black tea, and optimizes soil conditions through intelligent management means.

[0008] A soil monitoring method for planting nectar black tea according to an embodiment of the present invention comprises the following steps:

[0009] S1. Set up multiple soil sampling points in the fruit honey black tea planting area, determine the location and depth of the sampling points according to the growth requirements and geographical characteristics of the tea trees, collect representative soil samples, and generate a preliminary soil data set;

[0010] S2. Sensors at each sampling point monitor the basic physical parameters of the soil in real time, and generate physical parameter data sets after preliminary processing of the raw data. The processing includes data cleaning and outlier removal;

[0011] S3, physical parameter data sets are transmitted from each sampling point to the data set aggregation process, and the data from each sampling point is integrated into a comprehensive soil physical state data set, which provides an overview of the overall health status of the soil;

[0012] S4, performing multiple excitation response spectral imaging on the soil samples at the sampling points to capture the spectral response patterns of different components in the soil and generate a spectral data set;

[0013] S5, integrating the spectral dataset with the comprehensive soil physical state dataset to form a complete soil comprehensive dataset;

[0014] S6. Use the adaptive spatiotemporal ripple model and non-commutative algebraic topology method to conduct in-depth analysis of the soil comprehensive data set, identify the complex dynamic changes in the soil, especially the spatiotemporal changes of key nutrients closely related to the growth of fruit honey black tea, and generate a soil health assessment data set;

[0015] S7. Generate a fertilization and irrigation recommendation dataset based on the soil health assessment dataset, and transmit the fertilization and irrigation recommendation dataset to the field management equipment. The field management equipment performs corresponding fertilization and irrigation operations according to the recommendations to optimize soil conditions and meet the growth requirements of fruit honey black tea.

[0016] S8. The soil health assessment dataset continuously receives and integrates new physical parameters and spectral data, and is dynamically updated to adapt to changing soil conditions and environments;

[0017] S9. Throughout the entire fruit honey black tea planting cycle, all data sets are recorded and saved for future planting decisions and for continuous optimization of soil management methods for fruit honey black tea planting.

[0018] Optionally, S4 includes the following steps:

[0019] S41, placing the collected soil samples in a multiple excitation response spectral imaging device, setting multiple excitation wavelengths (λ 1 ,λ 2 , ..., λ n ), these wavelengths cover multiple frequency bands from visible light to near-infrared light; the soil sample is continuously excited by each excitation wavelength, and the spectral response of the sample at each excitation wavelength is recorded to generate an initial spectral response data set in Represents the spectral response intensity of the sample at position (x, y);

[0020] S42, initial spectral response data set Perform adaptive nonlinear enhancement processing:

[0021]

[0022] Among them, α is the enhancement coefficient, and β is the adjustment parameter used to control the nonlinear degree of enhancement. is the enhanced spectral dataset;

[0023] S43. Enhanced spectral dataset Applying the two-dimensional generalized hyperbolic cosine transform, Transform from the spatial domain to the frequency domain to capture the spectral characteristics implicit in the soil sample:

[0024]

[0025] Among them, cosh(ax+by) is the hyperbolic cosine function, a and b are generalized transformation parameters, is the spectral data in the frequency domain;

[0026] S44. Frequency domain spectral data sets at different excitation wavelengths Perform weighted superposition in the complex domain to generate a complex domain spectral dataset G total (u, v):

[0027]

[0028] Where j represents the imaginary unit, w i is the excitation wavelength λ i The weight coefficient of

[0029] S45. For the complex domain spectral dataset G total (u, v) is transformed by Fourier Laplace inverse transform and returned to the time domain to extract the composite spectral features in space to generate the final time domain spectral feature dataset F final (x, y);

[0030] S46, the final time domain spectral feature data set F final (x, y) to extract features and identify key spectral components related to soil health and F final The distribution of (x, y) in space generates a standardized spectral feature dataset S final (x, y).

[0031] Optionally, S5 includes the following steps:

[0032] S51, receiving the generated spectral data set S final (x, y) and a comprehensive soil physical state data set P(x, y), where P(x, y) represents the soil physical state parameters at the spatial position (x, y);

[0033] S52, for the spectral data set S final (x, y) and soil physical state data set P(x, y) are standardized in data format;

[0034] S53, the standardized spectral data set Standardized soil physical state dataset Perform multi-dimensional feature fusion to generate a comprehensive soil dataset C(x, y);

[0035] S54. Apply spatial analysis algorithms to analyze the soil comprehensive data set C(x, y) and extract key soil characteristics related to the growth of fruit honey black tea.

[0036] Optionally, the S6 includes the following steps:

[0037] S61. Input the soil comprehensive data set C(x, y) into the adaptive spatiotemporal ripple model. The adaptive spatiotemporal ripple model constructs a multi-scale dynamic disturbance equation based on the spatiotemporal evolution of soil characteristics, and adjusts the disturbance propagation speed c(x, y, t) and damping coefficient in real time to cope with the complex changes of soil characteristics under different environmental conditions:

[0038]

[0039] Among them, C(x, y, t) represents the comprehensive characteristics of soil at spatial coordinates (x, y) and time t, reflecting the comprehensive state of soil; It is the second-order derivative of the comprehensive soil properties changing with time, indicating the acceleration of soil properties; is the adaptive disturbance propagation velocity, where E(x, y, t) is the elastic modulus of the soil, ρ(x, y, t) is the density of the soil, both of which vary with time and space; α(x, y, t) is the adaptive amplitude coefficient, which controls the fluctuation amplitude of the disturbance propagation velocity and adjusts dynamically with space and time; sin(ωt) is a sine function, which represents the periodic change over time, and ω is the oscillation frequency; is the spatial Laplace operator of the comprehensive soil characteristics, which represents the diffusion of soil characteristics in space; is the spatiotemporal damping coefficient, which represents the energy loss when the disturbance propagates in the soil, η(x, y, t) is the soil viscosity coefficient that varies with time and space; A(x, y, t)·Cos(k(x, y, t)x-ωt) is the external disturbance term, where A(x, y, t) is the external force amplitude and k(x, y, t) is the wave number, both of which vary with space and time;

[0040] S62. Discretize the dynamic perturbation equation and solve it by adaptive multi-scale finite difference method to generate adaptive dynamic soil property data set:

[0041]

[0042]

[0043] in, represents the comprehensive soil characteristic value at position (i, j) at the n+1th moment after discretization; is the comprehensive soil characteristic value at position (i, j) at the nth moment after discretization; is the comprehensive soil characteristic value at the position (i, j) at the nth moment after discretization; Δt is the time step, which represents the time interval after discretization; is the disturbance propagation speed at position (i, j) at the nth moment; is the adaptive amplitude coefficient at position (i, j); is the damping coefficient at position (i, j) at the nth moment; is the external disturbance term;

[0044] S63. Discrete spatiotemporal data sets Perform spatial Fourier transform to convert soil characteristics from spatial domain to frequency domain and obtain frequency domain soil characteristic data set F(u, v):

[0045]

[0046] Among them, F(u, v) is the frequency domain soil property data set; represents Fourier transform; (u, v) is the frequency domain coordinate, representing the spatial frequency;

[0047] S64. Filter F(u, v) in the frequency domain, use a bandpass filter to eliminate high-frequency noise and low-frequency drift, and retain the effective soil characteristics in the mid-frequency band. The filtered data set is represented by F filtered (u, v):

[0048]

[0049] Among them, F filtered (u, v) is the frequency domain soil property data set after filtering; and is a rectangular function, defining the transfer function of the bandpass filter, u 0 and v 0 is the filter center frequency, Δu and Δv are the frequency bandwidth;

[0050] S65, the filtered frequency domain data F filtered (u, v) is inversely Fourier transformed back to the spatial domain to obtain the purified soil characteristic dataset C in the spatiotemporal domain filtered (x, y);

[0051] S66, the purified soil characteristic dataset C filtered (x, y) is input into the non-commutative algebraic topological analysis module to analyze the spatiotemporal variation characteristics of key nutrients in the soil. The non-commutative algebraic topological method is based on the non-commutative algebraic structure of soil characteristics to extract topological invariants:

[0052]

[0053] Among them, μ k is the kth topological invariant, which represents the changing characteristics of soil properties in the non-commutative algebraic topological structure; A is the connection form, which describes the structure of soil properties in the topological space and is obtained by solving the Yang-Mills equation dA+A∧A=0; M is the manifold in the space-time domain, which describes the overall structure of soil property changes; is the boundary of the manifold, which is used to calculate the topological invariant, through Calculate, where is the ith sub-boundary;

[0054] S67. Based on the results of non-commutative algebraic topological analysis, the spatiotemporal variation characteristics of key nutrients in the soil that are closely related to the growth of fruit honey black tea are identified, and the final soil health assessment data set H(x, y) is generated for the subsequent generation of fertilization and irrigation management recommendations.

[0055] Optionally, the S7 includes the following steps:

[0056] S71, extracting soil characteristic parameters closely related to the growth of fruit honey black tea from the soil health assessment data set H(x, y);

[0057] S72. Based on the extracted soil characteristic parameters, an optimization algorithm is used to generate multiple fertilization and irrigation plans. The optimization algorithm comprehensively considers the relationship between the soil characteristic parameters and the growth requirements of the fruit honey black tea, and evaluates the effectiveness and resource utilization of each plan;

[0058] S73, perform multi-objective optimization on the generated fertilization and irrigation schemes, taking into account factors including nutrient absorption efficiency, water resource utilization rate and potential environmental impact, and select the scheme with the best comprehensive benefits;

[0059] S74, selecting the fertilization and irrigation scheme with the best comprehensive benefits, and generating a final fertilization and irrigation recommendation data set;

[0060] S75, transmitting the final recommended data set to the field management equipment through wireless communication or other means, the field management equipment refers to an automated mechanical or electronic equipment system used to perform specific operations of fertilization, irrigation, weeding, and pesticide application in agricultural production, and the equipment automatically adjusts the operating parameters of the fertilization and irrigation equipment according to the instructions in the recommended data set, specifically including the amount of fertilizer, the amount of irrigation, and the time node of execution;

[0061] S76. After the field management equipment performs fertilization and irrigation operations, it monitors the soil response in real time and feeds back to the soil monitoring system.

[0062] Optionally, the S8 includes the following steps:

[0063] S8 1. The soil monitoring system continuously collects and updates physical parameter data sets and spectral data sets. Physical parameters include soil moisture, temperature, pH value, conductivity, and key nutrient concentrations, and uses a multi-source data fusion algorithm to unify data from different sources;

[0064] S82. After each data update, the physical parameter data set and the spectral data set are passed to the soil health assessment module, and the spatiotemporal correlation model is used to dynamically couple the new data with the existing data. The spatiotemporal correlation kernel function ψ(x′, y′, t′):

[0065]

[0066] Among them, σ s is the spatial scale parameter, which controls the effect of spatial distance on the association strength; σ t is the time scale parameter, which controls the influence of time distance on the correlation strength; x, y are the spatial coordinates of the current moment; t is the time of the current moment; x′, y′ are the spatial coordinates of the historical data; t′ is the time of the historical data;

[0067] S83. Use the spatiotemporal correlation kernel function ψ(x′, y′, t′) to calculate the coupling degree between the newly collected data and the historical data to obtain the updated soil health assessment dataset H n ew(x,y,t):

[0068] H new (x, y, t) = ∫ Ω ψ(x′,y′,t′)H(x′,y′,t′)dΩ+∈(x,y,t);

[0069] Among them, H new (x, y, t) is the updated soil health assessment dataset; H(x′, y′, t′) is the historical soil health assessment dataset; ∈(x, y, t) is the data noise term; Ω is the spatial integration domain;

[0070] S84. Updated soil health assessment dataset H n ew(x, y, t) is used for deep learning prediction, and a recurrent neural network model is used to predict the evolution trend of soil health status in the future time period;

[0071] S85. Dynamically adjust the fertilization and irrigation recommendation data set according to the predicted soil health status data, and transmit the adjusted recommendation data set to the field management equipment;

[0072] S86. The dynamic update process of the soil health assessment dataset is performed automatically at specific time intervals and encrypted updates are performed when important growth stages or environmental conditions change significantly.

[0073] The beneficial effects of the present invention are:

[0074] (1) The present invention proposes a soil monitoring method for the cultivation of fruit honey black tea. By deploying soil sampling points in the planting area and combining a variety of sensor technologies, the physical parameters of the soil are monitored in real time, and a soil physical state data set is generated and processed. The noise and outliers in the data collection process are eliminated, the accuracy and reliability of soil monitoring are improved, and the soil conditions in the process of cultivating fruit honey black tea are accurately grasped, especially under complex geographical conditions.

[0075] (2) The present invention uses multiple excitation response spectral imaging technology to capture the spectral response patterns of different components in the soil by performing multi-wavelength spectral excitation on soil samples, and integrates them with the physical state data set to generate a complete soil comprehensive data set. This process effectively reveals the key nutrients and organic matter distribution characteristics in the soil, provides a more scientific soil health assessment for the growth of fruit honey black tea, and significantly improves the quality and yield of tea.

[0076] (3) The present invention introduces an adaptive spatiotemporal ripple model and a non-commutative algebraic topology method. By constructing and solving dynamic perturbation equations, the comprehensive soil data set is analyzed in real time to identify complex dynamic changes in the soil, especially the spatiotemporal changes of key nutrients closely related to the growth of fruit honey black tea. The application of this model ensures an in-depth analysis of soil health, making soil management more precise and intelligent, and can effectively respond to the dynamic changes of soil conditions during the planting process, thereby improving the scientificity and operability of agricultural management.

[0077] (4) After receiving and integrating new physical parameters and spectral data, the soil health assessment dataset of the present invention can be dynamically updated to adapt to the ever-changing soil conditions and environment. Through continuous dynamic monitoring and analysis, the present invention realizes precise fertilization and irrigation management during the cultivation of fruit honey black tea, which not only optimizes resource utilization, but also effectively reduces environmental impact, and significantly improves the overall management efficiency and sustainable development capacity of the tea garden. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0079] Figure 1 This is a flow chart of a soil monitoring method for fruit honey black tea planting proposed by the present invention;

[0080] Figure 2 Schematic diagram of the adaptive space-time ripple model and non-commutative algebraic topological method analysis proposed in the present invention. DETAILED DESCRIPTION

[0081] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0082] refer to Figure 1-2 , a soil monitoring method for fruit honey black tea planting, comprising the following steps:

[0083] S1. Set up multiple soil sampling points in the fruit honey black tea planting area, determine the location and depth of the sampling points according to the growth requirements and geographical characteristics of the tea trees, collect representative soil samples, and generate a preliminary soil data set;

[0084] S2. Sensors at each sampling point monitor the basic physical parameters of the soil in real time, and generate physical parameter data sets after preliminary processing of the raw data. The processing includes data cleaning and outlier removal;

[0085] S3, physical parameter data sets are transmitted from each sampling point to the data set aggregation process, and the data from each sampling point is integrated into a comprehensive soil physical state data set, which provides an overview of the overall health status of the soil;

[0086] S4, performing multiple excitation response spectral imaging on the soil samples at the sampling points to capture the spectral response patterns of different components in the soil and generate a spectral data set;

[0087] S5, integrating the spectral dataset with the comprehensive soil physical state dataset to form a complete soil comprehensive dataset;

[0088] S6. Use the adaptive spatiotemporal ripple model and non-commutative algebraic topology method to conduct in-depth analysis of the soil comprehensive data set, identify the complex dynamic changes in the soil, especially the spatiotemporal changes of key nutrients closely related to the growth of fruit honey black tea, and generate a soil health assessment data set;

[0089] S7. Generate a fertilization and irrigation recommendation dataset based on the soil health assessment dataset, and transmit the fertilization and irrigation recommendation dataset to the field management equipment. The field management equipment performs corresponding fertilization and irrigation operations according to the recommendations to optimize soil conditions and meet the growth requirements of fruit honey black tea.

[0090] S8. The soil health assessment dataset continuously receives and integrates new physical parameters and spectral data, and is dynamically updated to adapt to changing soil conditions and environments;

[0091] S9. Throughout the entire fruit honey black tea planting cycle, all data sets are recorded and saved for future planting decisions and for continuous optimization of soil management methods for fruit honey black tea planting.

[0092] In this implementation, S4 includes the following steps:

[0093] S41, placing the collected soil samples in a multiple excitation response spectral imaging device, setting multiple excitation wavelengths (λ 1 ,λ 2 , ..., λ n ), these wavelengths cover multiple frequency bands from visible light to near-infrared light; the soil sample is continuously excited by each excitation wavelength, and the spectral response of the sample at each excitation wavelength is recorded to generate an initial spectral response data set in Represents the spectral response intensity of the sample at position (x, y);

[0094] S42, initial spectral response data set Perform adaptive nonlinear enhancement processing:

[0095]

[0096] Among them, α is the enhancement coefficient, and β is the adjustment parameter used to control the nonlinear degree of enhancement. is the enhanced spectral dataset;

[0097] S43. Enhanced spectral dataset Applying the two-dimensional generalized hyperbolic cosine transform, Transform from the spatial domain to the frequency domain to capture the spectral characteristics implicit in the soil sample:

[0098]

[0099] Among them, cosh(ax+by) is the hyperbolic cosine function, a and b are generalized transformation parameters, is the spectral data in the frequency domain;

[0100] S44. Frequency domain spectral data sets at different excitation wavelengths Perform weighted superposition in the complex domain to generate a complex domain spectral dataset G total (u, v):

[0101]

[0102] Where j represents the imaginary unit, w i is the excitation wavelength λ i The weight coefficient of

[0103] S45. For the complex domain spectral dataset G total (u, v) is transformed by Fourier Laplace inverse transform and returned to the time domain to extract the composite spectral features in space to generate the final time domain spectral feature dataset F final (x, y);

[0104] S46, the final time domain spectral feature data set F final (x, y) to extract features and identify key spectral components related to soil health and F final The distribution of (x, y) in space generates a standardized spectral feature dataset S final (x, y).

[0105] In this implementation, S5 includes the following steps:

[0106] S51, receiving the generated spectral data set S final (x, y) and a comprehensive soil physical state data set P(x, y), where P(x, y) represents the soil physical state parameters at the spatial position (x, y);

[0107] S52, for the spectral data set S final(x, y) and soil physical state data set P(x, y) are standardized in data format;

[0108] S53, the standardized spectral data set Standardized soil physical state dataset Perform multi-dimensional feature fusion to generate a comprehensive soil dataset C(x, y);

[0109] S54. Apply spatial analysis algorithms to analyze the soil comprehensive data set C(x, y) and extract key soil characteristics related to the growth of fruit honey black tea.

[0110] In this implementation, S6 includes the following steps:

[0111] S61. Input the soil comprehensive data set C(x, y) into the adaptive spatiotemporal ripple model. The adaptive spatiotemporal ripple model constructs a multi-scale dynamic disturbance equation based on the spatiotemporal evolution of soil characteristics, and adjusts the disturbance propagation speed c(x, y, t) and damping coefficient in real time to cope with the complex changes of soil characteristics under different environmental conditions:

[0112]

[0113] Among them, C(x, y, t) represents the comprehensive characteristics of soil at spatial coordinates (x, y) and time t, reflecting the comprehensive state of soil; It is the second-order derivative of the comprehensive soil properties changing with time, indicating the acceleration of soil properties; is the adaptive disturbance propagation velocity, where E(x, y, t) is the elastic modulus of the soil, ρ(x, y, t) is the density of the soil, both of which vary with time and space; α(x, y, t) is the adaptive amplitude coefficient, which controls the fluctuation amplitude of the disturbance propagation velocity and adjusts dynamically with space and time; sin(ωt) is a sine function, which represents the periodic change over time, and ω is the oscillation frequency; is the spatial Laplace operator of the comprehensive soil characteristics, which represents the diffusion of soil characteristics in space; is the spatiotemporal damping coefficient, which represents the energy loss when the disturbance propagates in the soil, η(x, y, t) is the soil viscosity coefficient that varies with time and space; A(x, y, t)·Cos(k(x, y, t)x-ωt) is the external disturbance term, where A(x, y, t) is the external force amplitude and k(x, y, t) is the wave number, both of which vary with space and time;

[0114] S62. Discretize the dynamic perturbation equation and solve it by adaptive multi-scale finite difference method to generate adaptive dynamic soil property data set:

[0115]

[0116] in, represents the comprehensive soil characteristic value at position (i, j) at the n+1th moment after discretization; is the comprehensive soil characteristic value at position (i, j) at the nth moment after discretization; is the comprehensive soil characteristic value at the position (i, j) at the nth moment after discretization; Δt is the time step, which represents the time interval after discretization; is the disturbance propagation speed at position (i, j) at the nth moment; is the adaptive amplitude coefficient at position (i, j); is the damping coefficient at position (i, j) at the nth moment; is the external disturbance term;

[0117] S63. Discrete spatiotemporal data sets Perform spatial Fourier transform to convert soil characteristics from spatial domain to frequency domain and obtain frequency domain soil characteristic data set F(u, v):

[0118]

[0119] Among them, F(u, v) is the frequency domain soil property data set; represents Fourier transform; (u, v) is the frequency domain coordinate, representing the spatial frequency;

[0120] S64. Filter F(u, v) in the frequency domain, use a bandpass filter to eliminate high-frequency noise and low-frequency drift, and retain the effective soil characteristics in the mid-frequency band. The filtered data set is represented by F filtered (u, v):

[0121]

[0122] Among them, F filtered (u, v) is the frequency domain soil property data set after filtering; and is a rectangular function, defining the transfer function of the bandpass filter, u 0 and v 0 is the filter center frequency, Δu and Δv are the frequency bandwidth;

[0123] S65, the filtered frequency domain data F filtered (u, v) is inversely Fourier transformed back to the spatial domain to obtain the purified soil characteristic dataset C in the spatiotemporal domain filtered (x, y);

[0124] S66, the purified soil characteristic dataset C filtered(x, y) is input into the non-commutative algebraic topological analysis module to analyze the spatiotemporal variation characteristics of key nutrients in the soil. The non-commutative algebraic topological method is based on the non-commutative algebraic structure of soil characteristics to extract topological invariants:

[0125]

[0126] Among them, μ k is the kth topological invariant, which represents the changing characteristics of soil properties in the non-commutative algebraic topological structure; A is the connection form, which describes the structure of soil properties in the topological space and is obtained by solving the Yang-Mills equation dA+A∧A=0; M is the manifold in the space-time domain, which describes the overall structure of soil property changes; is the boundary of the manifold, which is used to calculate the topological invariant, through Calculate, where is the ith sub-boundary;

[0127] S67. Based on the results of non-commutative algebraic topological analysis, the spatiotemporal variation characteristics of key nutrients in the soil that are closely related to the growth of fruit honey black tea are identified, and the final soil health assessment data set H(x, y) is generated for the subsequent generation of fertilization and irrigation management recommendations.

[0128] In this implementation, S7 includes the following steps:

[0129] S71, extracting soil characteristic parameters closely related to the growth of fruit honey black tea from the soil health assessment data set H(x, y);

[0130] S72. Based on the extracted soil characteristic parameters, an optimization algorithm is used to generate multiple fertilization and irrigation plans. The optimization algorithm comprehensively considers the relationship between the soil characteristic parameters and the growth requirements of the fruit honey black tea, and evaluates the effectiveness and resource utilization of each plan;

[0131] S73, perform multi-objective optimization on the generated fertilization and irrigation schemes, taking into account factors including nutrient absorption efficiency, water resource utilization rate and potential environmental impact, and select the scheme with the best comprehensive benefits;

[0132] S74, selecting the fertilization and irrigation scheme with the best comprehensive benefits, and generating a final fertilization and irrigation recommendation data set;

[0133] S75, transmitting the final recommended data set to the field management equipment through wireless communication or other means, the field management equipment refers to an automated mechanical or electronic equipment system used to perform specific operations of fertilization, irrigation, weeding, and pesticide application in agricultural production, and the equipment automatically adjusts the operating parameters of the fertilization and irrigation equipment according to the instructions in the recommended data set, specifically including the amount of fertilizer, the amount of irrigation, and the time node of execution;

[0134] S76. After the field management equipment performs fertilization and irrigation operations, it monitors the soil response in real time and feeds back to the soil monitoring system.

[0135] In this implementation, S8 includes the following steps:

[0136] S81. The soil monitoring system continuously collects and updates physical parameter data sets and spectral data sets. Physical parameters include soil moisture, temperature, pH value, conductivity, and key nutrient concentrations, and uses a multi-source data fusion algorithm to unify data from different sources.

[0137] S82. After each data update, the physical parameter data set and the spectral data set are passed to the soil health assessment module, and the spatiotemporal correlation model is used to dynamically couple the new data with the existing data. The spatiotemporal correlation kernel function ψ(x′, y′, t′):

[0138]

[0139] Among them, σ s is the spatial scale parameter, which controls the effect of spatial distance on the association strength; σ t is the time scale parameter, which controls the influence of time distance on the correlation strength; x, y are the spatial coordinates of the current moment; t is the time of the current moment; x′, y′ are the spatial coordinates of the historical data; t′ is the time of the historical data;

[0140] S83. Use the spatiotemporal correlation kernel function ψ(x′, y′, t′) to calculate the coupling degree between the newly collected data and the historical data to obtain the updated soil health assessment dataset H n ew(x, y, t):

[0141] H n ew(x,y ， t)=∫Ωψ(x′,y′,t′)H(x′,y′,t′)dΩ+∈(x,y ， t);

[0142] Among them, H new (x, y, t) is the updated soil health assessment dataset; H(x′, y′, t′) is the historical soil health assessment dataset; ∈(x, y, t) is the data noise term; Ω is the spatial integration domain;

[0143] S84. Updated soil health assessment dataset H new (x, y, t) is used for deep learning prediction, and a recurrent neural network model is used to predict the evolution trend of soil health status in the future time period;

[0144] S85. Dynamically adjust the fertilization and irrigation recommendation data set according to the predicted soil health status data, and transmit the adjusted recommendation data set to the field management equipment;

[0145] S86. The dynamic update process of the soil health assessment dataset is performed automatically at specific time intervals and encrypted updates are performed when important growth stages or environmental conditions change significantly.

[0146] Embodiment 1:

[0147] This example was conducted at a fruit honey black tea planting base in a certain county. The area is famous for its natural selenium-rich soil and has unique geographical and climatic conditions, making it an ideal place for growing high-quality black tea. Fruit honey black tea is very popular in the market for its unique taste and high selenium content, but the quality and yield of tea are easily affected by soil conditions. In order to improve the yield and quality of fruit honey black tea, this example adopts an innovative soil monitoring method, which optimizes the soil management strategy of the tea garden by real-time monitoring of the physical and chemical properties of the soil and combining it with an intelligent data analysis model.

[0148] At a fruit honey black tea planting base in a certain county, multiple soil sampling points were first set up in the planting area. The locations and depths of these sampling points were scientifically determined based on the growth requirements of tea trees and local geographical features. The collected soil samples generated a preliminary soil data set, which included soil moisture, temperature, pH, conductivity, selenium content, and the content of major nutrients (such as nitrogen, phosphorus, and potassium).

[0149] High-precision sensors are installed at each sampling point to monitor the physical and chemical parameters of the soil in real time, and the raw data is cleaned and outliers are removed. The processed data generates physical parameter data sets, which are then integrated into comprehensive soil physical status data sets. These data sets provide an overall overview of soil health, allowing managers to understand the soil conditions of the entire tea garden in real time.

[0150] Based on the data collection, the soil samples were further analyzed using multiple excitation response spectral imaging technology. This technology captures the spectral response patterns of various components in the soil through spectral excitation at different wavelengths, generating a detailed spectral data set. These spectral data are integrated with the comprehensive soil physical state data set to form a more comprehensive soil comprehensive data set.

[0151] In order to deeply analyze the soil health status, this example introduces an adaptive spatiotemporal ripple model and a non-commutative algebraic topology method. These methods dynamically analyze the soil comprehensive data set and identify the complex dynamic changes in the soil, especially the spatiotemporal changes of key nutrients closely related to the growth of fruit honey black tea. Based on these analysis results, a soil health assessment data set is generated.

[0152] Based on the soil health assessment dataset, the system generates fertilization and irrigation recommendation datasets and transmits these datasets to field management equipment. The management equipment performs precise fertilization and irrigation operations based on the recommendations to optimize soil conditions and meet the growth needs of fruit honey black tea. In order to adapt to the dynamic changes in soil conditions, the soil health assessment dataset continuously receives and integrates new physical parameters and spectral data, and updates them dynamically.

[0153] Data Proof

[0154] During a complete fruit honey black tea planting cycle from April 2023 to April 2024, the fruit honey black tea planting base using the method of the present invention conducted a comprehensive monitoring of various soil indicators, and the results are as follows:

[0155] Table 1 Test data

[0156] Test items Traditional method (average) The method of the present invention (average value) Improvement Soil moisture (%) 35 38.5 10％ Soil temperature (℃) 18 19.2 6.70％ pH 6.2 6.5 4.80％ Electrical conductivity (ms / cm) 0.9 1.1 22.20％ Selenium content (mg / kg) 0.12 0.18 50％ Nitrogen content (mg / kg) 35 40 14.30％ Phosphorus content (mg / kg) 15 17.5 16.70％ Potassium content (mg / kg) 25 28 12％

[0157] With the support of these basic data, the growth of fruit honey black tea has been significantly improved. In the same period, the yield and tea quality of the planting base using the method of the present invention have also been significantly improved compared with the traditional method, as follows:

[0158] Table 2 Data table

[0159]

[0160] Data show that the yield of the method of the present invention in each quarter has increased compared with the traditional method, with an average increase of about 12%-15%. At the same time, the quality of tea has also been improved from the first and second grades in the traditional method to the special grade and first grade. This shows that the method of the present invention has a significant effect in increasing tea yield and improving tea quality.

[0161] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A soil monitoring method for planting fruit honey black tea, characterized in that: The steps include: S1. Set up multiple soil sampling points in the fruit honey black tea planting area, determine the location and depth of the sampling points according to the growth requirements and geographical characteristics of the tea trees, collect representative soil samples, and generate a preliminary soil data set; S2. Sensors at each sampling point monitor the basic physical parameters of the soil in real time, and generate physical parameter data sets after preliminary processing of the raw data. The processing includes data cleaning and outlier removal; S3, physical parameter data sets are transmitted from each sampling point to the data set aggregation process, and the data from each sampling point is integrated into a comprehensive soil physical state data set, which provides an overview of the overall health status of the soil; S4, performing multiple excitation response spectral imaging on the soil samples at the sampling points to capture the spectral response patterns of different components in the soil and generate a spectral data set; S5, integrating the spectral dataset with the comprehensive soil physical state dataset to form a complete soil comprehensive dataset; S6. Use the adaptive spatiotemporal ripple model and non-commutative algebraic topology method to conduct in-depth analysis of the soil comprehensive data set, identify the complex dynamic changes in the soil, especially the spatiotemporal changes of key nutrients closely related to the growth of fruit honey black tea, and generate a soil health assessment data set; S7. Generate a fertilization and irrigation recommendation dataset based on the soil health assessment dataset, and transmit the fertilization and irrigation recommendation dataset to the field management equipment. The field management equipment performs corresponding fertilization and irrigation operations according to the recommendations to optimize soil conditions and meet the growth requirements of fruit honey black tea. S8. The soil health assessment dataset continuously receives and integrates new physical parameters and spectral data, and is dynamically updated to adapt to changing soil conditions and environments; S9. Throughout the entire fruit honey black tea planting cycle, all data sets are recorded and saved for future planting decisions and for continuous optimization of soil management methods for fruit honey black tea planting.

2. A soil monitoring method for planting nectar black tea according to claim 1, characterized in that: The S4 comprises the following steps: S41, placing the collected soil sample in a multiple excitation response spectral imaging device, setting multiple excitation wavelengths (λ1, λ2, ..., λ n ), these wavelengths cover multiple frequency bands from visible light to near-infrared light; the soil sample is continuously excited by each excitation wavelength, and the spectral response of the sample at each excitation wavelength is recorded to generate an initial spectral response data set in Represents the spectral response intensity of the sample at position (x, y); S42, initial spectral response data set Perform adaptive nonlinear enhancement processing: Among them, α is the enhancement coefficient, and β is the adjustment parameter used to control the nonlinear degree of enhancement. is the enhanced spectral dataset; S43. Enhanced spectral dataset Applying the two-dimensional generalized hyperbolic cosine transform, Transform from the spatial domain to the frequency domain to capture the spectral characteristics implicit in the soil sample: Among them, cosh(ax+by) is the hyperbolic cosine function, a and b are generalized transformation parameters, is the spectral data in the frequency domain; S44. Frequency domain spectral data sets at different excitation wavelengths Perform weighted superposition in the complex domain to generate a complex domain spectral dataset G total (u,v): Where j represents the imaginary unit, w i is the excitation wavelength λ i The weight coefficient of S45. For the complex domain spectral dataset G total (u, v) is transformed by Fourier Laplace inverse transform back to the time domain to extract the composite spectral features in space and generate the final time domain spectral feature dataset F final (x,y); S46, the final time domain spectral feature data set F final (x, y) to extract features and identify key spectral components related to soil health and F final The distribution of (x, y) in space generates a standardized spectral feature dataset S final (x,y).

3. The soil monitoring method for planting nectar black tea according to claim 1, characterized in that: The S5 comprises the following steps: S51, receiving the generated spectral data set S final (x, y) and a comprehensive soil physical state data set P(x, y), where P(x, y) represents the soil physical state parameters at the spatial position (x, y); S52, for the spectral data set S final (x, y) and soil physical state data set P(x, y) are standardized in data format; S53, the standardized spectral data set Standardized soil physical state dataset Perform multi-dimensional feature fusion to generate a comprehensive soil dataset C(x,y); S54. Apply spatial analysis algorithms to analyze the soil comprehensive dataset C(x,y) and extract key soil characteristics related to the growth of fruit honey black tea.

4. The soil monitoring method for planting nectar black tea according to claim 1, characterized in that: The S6 comprises the following steps: S61. Input the soil comprehensive data set C(x,y) into the adaptive spatiotemporal ripple model. The adaptive spatiotemporal ripple model constructs a multi-scale dynamic disturbance equation based on the spatiotemporal evolution of soil characteristics, and adjusts the disturbance propagation speed c(x,y,t) and damping coefficient in real time to cope with the complex changes of soil characteristics under different environmental conditions: Among them, C(x,y,t) represents the comprehensive characteristics of soil at spatial coordinates (x,y) and time t, reflecting the comprehensive state of soil; It is the second-order derivative of the comprehensive soil properties changing with time, indicating the acceleration of soil properties; is the adaptive disturbance propagation speed, where E(x, y, t) is the elastic modulus of the soil, ρ(x, y, t) is the density of the soil, both of which vary with time and space; α(x, y, t) is the adaptive amplitude coefficient, which controls the fluctuation amplitude of the disturbance propagation speed and adjusts dynamically with space and time; sin(ωt) is a sine function, which represents the periodic change over time, and ω is the oscillation frequency; is the spatial Laplace operator of the comprehensive soil characteristics, which represents the diffusion of soil characteristics in space; is the spatiotemporal damping coefficient, which represents the energy loss when the disturbance propagates in the soil, η(x,y,t) is the soil viscosity coefficient that varies with time and space; A(x,y,t)·cos(k(x,y,t)x-ωt) is the external disturbance term, where A(x,y,t) is the external force amplitude and k(x,y,t) is the wave number, both of which vary with space and time; S62. Discretize the dynamic perturbation equation and solve it by adaptive multi-scale finite difference method to generate adaptive dynamic soil property data set: in, It represents the comprehensive soil characteristic value at the position (i, j) at the n+1th moment after discretization; is the comprehensive soil characteristic value at position (i, j) at the nth moment after discretization; is the comprehensive soil characteristic value at the position (i, j) at the nth moment after discretization; Δt is the time step, which represents the time interval after discretization; is the disturbance propagation speed at position (i, j) at the nth moment; is the adaptive amplitude coefficient at position (i, j); is the damping coefficient at position (i, j) at the nth moment; is the external disturbance term; S63. Discrete spatiotemporal data sets Perform spatial Fourier transform to convert soil characteristics from the spatial domain to the frequency domain and obtain the frequency domain soil characteristic data set F(u,v): Among them, F(u,v) is the frequency domain soil property data set; represents Fourier transform; (u,v) is the frequency domain coordinate, representing the spatial frequency; S64. Filter F(u,v) in the frequency domain, use a bandpass filter to eliminate high-frequency noise and low-frequency drift, and retain the effective soil characteristics in the mid-frequency band. The filtered data set is represented by F filtered (u,v): Among them, F filtered (u, v) is the frequency domain soil property data set after filtering; and is a rectangular function, defining the transfer function of the bandpass filter, u0 and v0 are the center frequencies of the filter, Δu and Δv are the frequency bandwidths; S65, the filtered frequency domain data F filtered (u, v) is inversely Fourier transformed back to the spatial domain to obtain the purified soil characteristic dataset C in the spatiotemporal domain filtered (x,y); S66, the purified soil characteristic dataset C filtered (x,y) is input into the non-commutative algebraic topological analysis module to analyze the spatiotemporal variation characteristics of key nutrients in the soil. The non-commutative algebraic topological method is based on the non-commutative algebraic structure of soil characteristics to extract topological invariants: Among them, μ k is the kth topological invariant, which represents the changing characteristics of soil properties in the non-commutative algebraic topological structure; A is the connection form, which describes the structure of soil properties in the topological space and is obtained by solving the Yang-Mills equation dA+A∧A=0; M is the manifold in the space-time domain, which describes the overall structure of soil property changes; is the boundary of the manifold, which is used to calculate the topological invariant, through Calculate, where is the ith sub-boundary; S67. Based on the results of non-commutative algebraic topological analysis, the spatiotemporal variation characteristics of key nutrients in the soil that are closely related to the growth of fruit honey black tea are identified, and the final soil health assessment dataset H(x,y) is generated for the subsequent generation of fertilization and irrigation management recommendations.

5. The soil monitoring method for planting nectar black tea according to claim 1, characterized in that: The S7 comprises the following steps: S71, extracting soil characteristic parameters closely related to the growth of fruit honey black tea from the soil health assessment data set H(x,y); S72. Based on the extracted soil characteristic parameters, an optimization algorithm is used to generate multiple fertilization and irrigation plans. The optimization algorithm comprehensively considers the relationship between the soil characteristic parameters and the growth requirements of the fruit honey black tea, and evaluates the effectiveness and resource utilization of each plan; S73, perform multi-objective optimization on the generated fertilization and irrigation schemes, taking into account factors including nutrient absorption efficiency, water resource utilization rate and potential environmental impact, and select the scheme with the best comprehensive benefits; S74, selecting the fertilization and irrigation scheme with the best comprehensive benefits, and generating a final fertilization and irrigation recommendation data set; S75, transmitting the final recommended data set to the field management equipment through wireless communication or other means, the field management equipment refers to an automated mechanical or electronic equipment system used to perform specific operations of fertilization, irrigation, weeding, and pesticide application in agricultural production, and the equipment automatically adjusts the operating parameters of the fertilization and irrigation equipment according to the instructions in the recommended data set, specifically including the amount of fertilizer, the amount of irrigation, and the time node of execution; S76. After the field management equipment performs fertilization and irrigation operations, it monitors the soil response in real time and feeds back to the soil monitoring system.

6. The soil monitoring method for planting nectar black tea according to claim 1, characterized in that: The S8 comprises the following steps: S81. The soil monitoring system continuously collects and updates physical parameter data sets and spectral data sets. Physical parameters include soil moisture, temperature, pH value, conductivity, and key nutrient concentrations, and uses a multi-source data fusion algorithm to unify data from different sources. S82. After each data update, the physical parameter data set and the spectral data set are passed to the soil health assessment module, and the spatiotemporal correlation model is used to dynamically couple the new data with the existing data. The spatiotemporal correlation kernel function ψ(x′, y′, t′): Among them, σ s is the spatial scale parameter, which controls the effect of spatial distance on the association strength; σ t is the time scale parameter, which controls the influence of time distance on the correlation strength; x, y are the spatial coordinates of the current moment; t is the time of the current moment; x′, y′ are the spatial coordinates of the historical data; t′ is the time of the historical data; S83. Use the spatiotemporal correlation kernel function ψ(x′, y′, t′) to calculate the coupling degree between the newly collected data and the historical data to obtain the updated soil health assessment dataset H new (x,y,t): H new (x,y,t)=∫ Ω ψ(x′,y′,t′)H(x′,y′,t′)dΩ+∈(x,y,t); Among them, H new (x, y, t) is the updated soil health assessment dataset; H(x′, y′, t′) is the historical soil health assessment dataset; ∈(x, y, t) is the data noise term; Ω is the spatial integration domain; S84. Updated soil health assessment dataset H new (x, y, t) is used for deep learning prediction, and a recurrent neural network model is used to predict the evolution trend of soil health status in the future time period; S85. Dynamically adjust the fertilization and irrigation recommendation data set according to the predicted soil health status data, and transmit the adjusted recommendation data set to the field management equipment; S86. The dynamic update process of the soil health assessment dataset is performed automatically at specific time intervals and encrypted updates are performed when important growth stages or environmental conditions change significantly.

Images