Alfalfa high-yield planting method

Abstract

The invention relates to the technical field of alfalfa planting, and discloses a high-yield alfalfa planting method, which realizes high-temporal-spatial-resolution dynamic perception of key environment parameters by constructing a growth environment database fusing historical and real-time data, and overcomes the hysteresis of traditional macroscopic static data. And furthermore, an expert knowledge base formed by extracting an optimization parameter threshold value based on a high-yield data sample provides a scientific basis for management decision making. The crop growth state and environment data are collected in real time and are accurately compared with the expert knowledge base, so that the moisture, accumulated temperature and nutrient demand gaps of alfalfa in different growth stages can be identified in time, and differential irrigation and topdressing instructions are generated. Finally, the intelligent execution equipment is driven by the central control system to realize accurate delivery of water and fertilizer, and a strategy is continuously optimized according to execution feedback, so that high matching of environmental resources and physiological needs of alfalfa is ensured, and the utilization efficiency of moisture and nutrients is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of alfalfa cultivation technology, and in particular to a high-yield alfalfa cultivation method. Background Technology

[0002] Current high-yield alfalfa cultivation and management methods primarily rely on regional, macro-scale environmental data as the basis for decision-making. The core technology lies in collecting historical meteorological data (such as cumulative precipitation, effective accumulated temperature, and diurnal temperature range) and soil profile data (such as soil pH at different depths, organic matter content, and nitrogen, phosphorus, and potassium content) for the target planting area. These parameters are key environmental factors directly affecting alfalfa root development, overwintering ability, and protein accumulation. The data typically originates from spatially limited meteorological stations and periodic soil sampling tests, and is manually processed to form static environmental background parameters. In practical applications, managers use these macro-level parameters to formulate unified field management strategies for the entire planting area, such as setting fixed irrigation systems, basal fertilizer formulas, and sowing densities, in order to meet the general conditions required for high alfalfa yields.

[0003] The aforementioned and existing related technologies often suffer from the following drawbacks: Traditional methods relying on macroscopic static data cannot accurately reflect the dynamic spatiotemporal changes of meteorological factors (such as precipitation and effective accumulated temperature) and soil properties (such as organic matter content and pH) that directly affect alfalfa growth at the field scale. Due to the lack of high spatiotemporal resolution for real-time monitoring and data fusion, management decisions struggle to match the real-time water, accumulated temperature, and nutrient requirements of alfalfa at different growth stages. Water and fertilizer supply either lags behind the critical water requirement period or the period of maximum nutrient efficiency for alfalfa, or exceeds actual needs, not only wasting resources but also affecting alfalfa yield and quality due to soil environmental imbalance, ultimately hindering the full realization of its high-yield potential. Summary of the Invention

[0004] The technical problem to be solved by this invention is that the existing technology cannot accurately reflect the actual and dynamically changing weather and soil conditions in the field, resulting in a mismatch between water and fertilizer management decisions and the real-time needs of crops, thus restricting yield improvement. To address this, we propose a high-yield alfalfa planting method.

[0005] To achieve the above objectives, this application adopts the following technical solution: a high-yield alfalfa cultivation method, comprising the following steps: S1. Collect historical and real-time environmental data of the target planting area to construct an alfalfa growth environment database; environmental data includes climate data and soil moisture data; S2. Based on the growth environment database, extract the threshold range of optimized environmental parameters that conform to the high-yield alfalfa growth model to form an expert knowledge base; S3. Through a sensor network deployed in the field, real-time data on crop growth status and environmental monitoring of the target alfalfa field are collected. S4. Compare crop growth status data and environmental monitoring data with the threshold range of optimized environmental parameters in the expert knowledge base, and generate precision agronomic operation decision instructions based on the comparison results. S5. The central control system sends agronomic operation decision instructions to the intelligent execution equipment in the field, driving it to complete automated operations.

[0006] Furthermore, an alfalfa growth environment database is constructed, including: collecting historical meteorological data and soil profile data for at least one complete growing season in the target planting area; historical meteorological data includes precipitation, sunshine hours, effective accumulated temperature, and diurnal temperature range; soil profile data includes soil pH, organic matter content, nitrogen, phosphorus, and potassium content, soil bulk density, and field water holding capacity at different depths; integrating historical meteorological data and soil profile data through a geographic information system and unifying them under the same spatial coordinate framework to form a basic database of the growth environment with spatiotemporal attributes; and setting a data update mechanism to continuously collect real-time meteorological monitoring data and soil sensor data through IoT nodes to dynamically update and expand the basic database of the growth environment.

[0007] Furthermore, historical meteorological data and soil profile data are integrated through a geographic information system and unified under the same spatial coordinate framework to form a basic database of growth environment with spatiotemporal attributes, and the following operations are performed; By overlaying and fusing historical meteorological data and soil profile data after gridding processing using a Geographic Information System (GIS), a multivariate environmental feature vector is generated for each geographic grid cell. A comprehensive assessment model for alfalfa growth suitability is then constructed, and the comprehensive suitability index is calculated for all geographic grid cells. The comprehensive suitability index of each geographic grid unit is calculated. As a newly added attribute, it is stored in association with the existing spatial location information in the database; based on The spatial distribution of values ​​is used to divide the target planting area into zones and assign differentiated variety selection and initial management strategies to areas with different suitability levels. These strategies include those based on... The value allows for quantitative adjustments to sowing density and basal fertilizer formulation; The mathematical expression for the comprehensive assessment model of alfalfa growth suitability is: , in, The comprehensive suitability index represents the i-th geographic grid unit; This represents the cumulative precipitation in that grid cell during the critical growth period; This represents the optimal water requirement threshold for this growth stage under the high-yield alfalfa model. This represents the effective accumulated temperature of the grid cell during a specific reproductive stage; This represents the optimal accumulated temperature threshold required for alfalfa to complete this growth stage. This represents the soil organic matter content of that grid cell; The optimal soil organic matter content threshold required for high alfalfa yield; The soil pH value represents the value of the grid cell. The optimal soil pH value for alfalfa growth; These are the weighting coefficients for climate factors, soil fertility factors, and soil environmental factors, respectively, and they satisfy the following conditions: .

[0008] Furthermore, a data update mechanism is established to continuously collect real-time meteorological monitoring data and soil sensor data through IoT nodes, dynamically updating and expanding the basic database of the growth environment, and performing the following operations: Based on a comprehensive assessment model of alfalfa growth suitability, a dynamic data value assessment function is constructed to intelligently filter and integrate real-time monitoring data, and to evaluate the degree of value change. Set a dynamic threshold When calculated in real time Value exceeds When the real-time data set is determined to be high-value data, the following operations are automatically triggered: the high-value data set and its corresponding spatial location information are synchronously updated to the basic database of the growth environment, and the current comprehensive suitability index of the grid cell is recalculated using this new data point. If the updated If the value decreases significantly, an environmental stress warning is generated, and the central control system is triggered to reassess the current irrigation or fertilization decision instructions, so as to achieve data-driven, highly timely dynamic regulation. The mathematical expression of the function is: , in, This represents the degree of value change of a real-time environmental data unit collected at time t; These represent the real-time precipitation, real-time accumulated temperature, and real-time soil organic matter content monitoring values ​​collected at time t, respectively. These represent the corresponding monitoring values ​​stored in the database at the previous data collection time point t-1; These are the weighting coefficients for changes in precipitation, accumulated temperature, and organic matter content, respectively.

[0009] Furthermore, an expert knowledge base is formed, including: selecting environmental data sample sets corresponding to the high-yield alfalfa record years from the growth environment database; performing multivariate statistical analysis on the high-yield environmental data sample sets to calculate the mean and standard deviation of each environmental parameter under high-yield conditions; based on the mean and standard deviation, setting an optimal target interval and an acceptable warning interval for each environmental parameter, with the optimal target interval used to guide precise management and the acceptable warning interval used to trigger intervention measures; and associating the interval thresholds of each environmental parameter with the corresponding alfalfa growth period to construct the expert knowledge base.

[0010] Furthermore, from the growth environment database, a sample set of environmental data corresponding to the years with high alfalfa yield records was selected, and the following operations were performed: A comprehensive scoring model for high-yield years is constructed. Based on the degree of matching between various environmental parameters and optimal thresholds, the contribution of environmental conditions to high alfalfa yield in each historical year is quantitatively evaluated, and the model is used to calculate the yield of alfalfa for all historical years. Value, and according to The values ​​are sorted from high to low; all environmental data corresponding to the years in the top K% of the sorted years are selected to form a sample set of environmental data for alfalfa high-yield record years; based on this sample set, the kernel density estimation method is used to calculate the probability density distribution of each environmental parameter under high-yield conditions, and then the threshold boundary between the optimal target interval and the acceptable warning interval is determined to ensure that the expert knowledge base is derived from the statistical nature of high-yield environmental data. The mathematical expression for the comprehensive scoring model for high-yield years is: , in, The overall high-yield score represents the j-th historical year; These represent the annual average values ​​of each environmental parameter in year j. The effective range of pH values ​​allowed; These are the index weights of climate factors, soil fertility factors, and soil environmental factors, respectively, and they satisfy the following conditions: .

[0011] Furthermore, multivariate statistical analysis was performed on the high-yield environmental data sample set to calculate the mean and standard deviation of each environmental parameter under high-yield conditions, and the following operations were performed: A dynamic weighted statistical model based on factor sensitivity is constructed. The model calculates the impact of various environmental parameters on high yield scores. The contribution sensitivity is dynamically adjusted in statistical calculations to more accurately reflect the core distribution characteristics of each parameter under high-yield conditions, and its weighted average value is used. with weighted standard deviation The calculation formula is as follows: , , in, This represents the value of a certain environmental parameter to be statistically analyzed in the sample of the j-th high-yield year; This represents the dynamic weight assigned to the j-th sample, and its value is the high-yield score for that sample. The exponential weight of this environmental parameter in the high-yield scoring model ( The product of ), i.e. (k takes the index of the corresponding parameter in 1, 2, or 3); The total number of samples in the high-yield environment data sample set; This is the weighted average of the calculated environmental parameter; This is the weighted standard deviation of the calculated environmental parameter; The above dynamic weighted statistical model was applied sequentially to the four key environmental parameters—precipitation, effective accumulated temperature, organic matter content, and pH value—to calculate their respective values. and Based on the calculation results, the optimal target interval is set as [ -0.5× , +0.5× The acceptable warning range is set as []. -1.5× , +1.5× ].

[0012] Furthermore, the threshold ranges of various environmental parameters are linked to the corresponding alfalfa growth stages to construct an expert knowledge base, and the following operations are performed: A dynamic threshold adjustment model for the reproductive period is constructed. Based on the differences in physiological needs and environmental factor sensitivities of alfalfa at different reproductive stages, the model finely modulates the basic threshold range. The formula for the dynamic threshold adjustment model for the reproductive period is as follows: , in, This represents the upper or lower limit of the adjusted environmental parameter threshold under a specific reproductive phase. It is a multiple of the standard deviation; This is the demand sensitivity coefficient of this environmental parameter under a specific reproductive phase; This refers to the specific reproductive phase. This represents the reproductive phase. This represents the historical variation range of the environmental parameter within the reproductive phase; based on the model, the optimal target range and acceptable warning range of each environmental parameter after adjustment are calculated for each major reproductive phase; the adjusted range thresholds for all reproductive phases and the parameters required for calculation are structured, stored, and mapped according to reproductive phase and environmental parameter type to form an expert knowledge base.

[0013] Furthermore, the system generates precise agronomic operation decision instructions, including: comparing the real-time monitored soil volumetric moisture content with the optimal target range of soil moisture content for the current growth stage in the expert knowledge base; if it is lower than the optimal target range, an irrigation decision instruction is generated, which includes an estimated amount of water required for irrigation; comparing the real-time monitored soil electrical conductivity data with the warning threshold in the expert knowledge base; if the nitrogen content is low, a topdressing decision instruction is generated, which includes a suggested fertilizer formula and dosage; the decision instructions are transmitted to the central control system via a wireless communication network.

[0014] Furthermore, the central control system drives the field intelligent execution equipment to complete automated operations, including: the central control system receives irrigation decision instructions, analyzes the estimated amount of water to be irrigated, calculates the intelligent valve groups to be opened and their duration, and controls the integrated water and fertilizer system to perform precision irrigation operations; it receives topdressing decision instructions, controls the automatic fertilizer dispenser to prepare liquid fertilizer according to the fertilizer formula and dosage, and performs fertilization operations simultaneously through the integrated water and fertilizer system during irrigation; after the operation is completed, the sensor network collects environmental data again and feeds the executed data back to the system to evaluate the decision effect and optimize subsequent instructions.

[0015] The technical effects and advantages of this invention are as follows: This invention constructs a growth environment database integrating historical and real-time data, achieving high spatiotemporal resolution dynamic perception of key environmental parameters such as precipitation, effective accumulated temperature, soil organic matter content, and pH value, overcoming the lag of traditional macroscopic static data. Furthermore, an expert knowledge base, formed by extracting and optimizing parameter thresholds based on high-yield data samples, provides a scientific basis for management decisions. By collecting crop growth status and environmental data in real time and accurately comparing them with the expert knowledge base, the invention can promptly identify the water, accumulated temperature, and nutrient requirements gaps of alfalfa at different growth stages (such as branching and budding stages) and generate differentiated irrigation and fertilization instructions. Finally, a central control system drives intelligent execution equipment to achieve precise water and fertilizer delivery, and continuously optimizes strategies based on execution feedback, forming an adaptive control closed loop. Ultimately, this ensures a high degree of matching between environmental resources and alfalfa physiological needs, significantly improving water and nutrient utilization efficiency and providing a solid foundation for high alfalfa yields. Attached Figure Description

[0016] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts: Figure 1 This is the overall flowchart of the present invention. Detailed Implementation

[0017] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0018] Reference Figure 1 As shown, the present invention provides a technical solution: a high-yield intelligent alfalfa planting and management method, comprising the following steps: Step 1: Collect historical and real-time environmental data of the target planting area to construct an alfalfa growth environment database; environmental data includes climate data and soil moisture data; Step 2: Based on the growth environment database, extract the threshold range of optimized environmental parameters that conform to the high-yield alfalfa growth model, and form an expert knowledge base; Step 3: Collect real-time crop growth data and environmental monitoring data of the target alfalfa field through a sensor network deployed in the field; Step 4: Compare crop growth status data and environmental monitoring data with the optimized environmental parameter threshold range in the expert knowledge base, and generate precision agronomic operation decision instructions based on the comparison results. Step 5: The central control system sends agronomic operation decision instructions to the intelligent execution equipment in the field, drives it to complete automated operations, and provides feedback and records the execution results to achieve closed-loop control of planting management.

[0019] Specifically, by constructing a growth environment database integrating historical and real-time data, high spatiotemporal resolution dynamic perception of key environmental parameters such as precipitation, effective accumulated temperature, soil organic matter content, and pH value was achieved, overcoming the lag of traditional macroscopic static data. Furthermore, an expert knowledge base, formed by extracting and optimizing parameter thresholds based on high-yield data samples, provides a scientific basis for management decisions. By collecting crop growth status and environmental data in real time and accurately comparing them with the expert knowledge base, the system can promptly identify the water, accumulated temperature, and nutrient requirements of alfalfa at different growth stages (such as branching and budding stages) and generate differentiated irrigation and topdressing instructions. Finally, a central control system drives intelligent execution equipment to achieve precise water and fertilizer delivery, and continuously optimizes strategies based on execution feedback, forming an adaptive control closed loop. This ensures a high degree of matching between environmental resources and alfalfa physiological needs, significantly improving water and nutrient utilization efficiency and providing a solid foundation for high alfalfa yields.

[0020] In this preferred embodiment, constructing the alfalfa growth environment database specifically includes: collecting historical meteorological data and soil profile data for at least one complete growing season in the target planting area; historical meteorological data includes precipitation, sunshine hours, effective accumulated temperature, and diurnal temperature range; soil profile data includes soil pH, organic matter content, nitrogen, phosphorus, and potassium content, soil bulk density, and field water holding capacity at different depths; integrating historical meteorological data and soil profile data through a geographic information system and unifying them under the same spatial coordinate framework to form a basic database of the growth environment with spatiotemporal attributes; and setting a data update mechanism to continuously collect real-time meteorological monitoring data and soil sensor data through IoT nodes to dynamically update and expand the basic database of the growth environment.

[0021] Specifically, by collecting multi-dimensional historical data and integrating it into a grid using a geographic information system, environmental feature vectors with precise spatial location attributes were generated, laying a data foundation for subsequent precise zoning management. More importantly, by continuously collecting real-time data through IoT nodes and setting a dynamic update mechanism, continuous monitoring and database updates of key environmental factors (such as precipitation, effective accumulated temperature, soil organic matter content, and pH value) were achieved. This enables timely capture of environmental changes at the field scale, providing highly timely data support for precise water and fertilizer management at different growth stages of alfalfa (such as branching stage and budding stage), thereby significantly improving water and nutrient utilization efficiency and ultimately providing a reliable data-driven guarantee for high alfalfa yield.

[0022] Furthermore, in this preferred embodiment, historical meteorological data and soil profile data are fused together using a geographic information system and unified under the same spatial coordinate framework to form a basic database of growth environment with spatiotemporal attributes, and the following operations are performed; By overlaying and fusing gridded historical meteorological data and soil profile data using a geographic information system, a multivariate environmental feature vector is generated for each geographic grid unit. This generates a comprehensive assessment model for alfalfa growth suitability. The mathematical expression for this comprehensive alfalfa growth suitability assessment model is as follows: , in, The comprehensive suitability index, representing the i-th geographical grid unit, is a dimensionless scalar. The higher the value, the more favorable the natural endowment of the grid unit is for high alfalfa yield. This represents the cumulative precipitation (in mm) of the grid cell during the critical growth period. The optimal water requirement threshold (in mm) represents the growth period under the high-yield alfalfa model. The effective accumulated temperature (in °C·d) represents the grid cell during a specific growth stage. The optimal accumulated temperature threshold (in °C·d) required for alfalfa to complete this growth stage. Represents the soil organic matter content of this grid cell (unit: %). The optimal soil organic matter content threshold (in %) represents the required high alfalfa yield. The soil pH value represents the value of the grid cell. The optimal soil pH value for alfalfa growth; These are the weighting coefficients for climate factors, soil fertility factors, and soil environmental factors, respectively, and they satisfy the following conditions: The specific value was determined by performing multiple linear regression fitting on a high-yield environment data sample set. Traverse all geographic grid cells and calculate their comprehensive suitability index. The comprehensive suitability index of each geographic grid unit is calculated. As a newly added attribute, it is stored in association with the existing spatial location information in the database; based on The spatial distribution of values ​​is used to divide the target planting area into zones and assign differentiated variety selection and initial management strategies to areas with different suitability levels. These strategies include those based on... The value allows for quantitative adjustments to sowing density and basal fertilizer formulation.

[0023] Specifically, the comprehensive suitability index of each geographic grid unit is quantified by weighted fusion of key meteorological factors and soil properties. This enables a refined assessment of the environmental endowment within the planting area. Based on After dividing the spatial distribution of alfalfa values ​​into zones, differentiated initial management strategies, such as sowing density and basal fertilizer formulation, can be developed for areas with different suitability levels: higher sowing densities and optimized fertilization are used in high-suitability areas to fully realize high-yield potential, while adjustments to variety selection and reduced inputs are made in low-suitability areas to avoid resource waste. This precision management strategy based on spatial variability significantly improves the efficiency of water and fertilizer resource utilization, ensuring that alfalfa grows under optimal environmental conditions throughout its entire growth cycle, thus providing a solid foundation for ultimately increasing hay yield per unit area.

[0024] Furthermore, in this preferred embodiment, real-time meteorological monitoring data and soil sensor data are continuously collected through IoT nodes to dynamically update and expand the basic database of the growth environment, and the following operations are performed: Based on the comprehensive assessment model of alfalfa growth suitability, a dynamic data value assessment function is constructed for intelligent screening and integration of real-time monitoring data. The mathematical expression of the function is: , in, It represents the degree of value change of a real-time environmental data unit collected at time t. It is a dimensionless scalar. The larger the value, the more significant the change in the data at that time point compared to the previous time point t-1, and the higher the value for updating the database and triggering decisions. These represent the real-time precipitation, real-time accumulated temperature, and real-time soil organic matter content monitored at time t, respectively. Their physical units and thresholds are the same as those of the data collected at time t. , , completely consistent; These represent the corresponding monitoring values ​​stored in the database at the previous data collection time point t-1; These are the weighting coefficients for changes in precipitation, accumulated temperature, and organic matter content, used to measure the relative importance of changes in different environmental factors. Their values ​​are determined through sensitivity analysis of the impact of each factor on the final yield of alfalfa in historical data. Value change Set a dynamic threshold When calculated in real time Value exceeds When the real-time data set is determined to be high-value data, the following operations are automatically triggered: the high-value data set and its corresponding spatial location information are synchronously updated to the basic database of the growth environment, and the current comprehensive suitability index of the grid cell is recalculated using this new data point. If the updated If the value decreases significantly, an environmental stress warning is generated, triggering the central control system to reassess the current irrigation or fertilization decision instructions, thus achieving data-driven, highly timely dynamic regulation.

[0025] Specifically, by constructing a data value assessment function It can intelligently identify significant changes in key environmental factors such as precipitation, effective accumulated temperature, and organic matter content, and will detect when the changes in monitoring data exceed a threshold. When this happens, the database will be automatically updated and the grid suitability index will be recalculated. This ensures that the database always reflects the latest environmental conditions in the field, and is particularly capable of promptly detecting sudden environmental stresses, such as a sudden drop in rainfall, abnormal accumulated temperature, or loss of organic matter. When the value drops significantly, an early warning is immediately generated, triggering the central control system to reassess irrigation and fertilization decisions, achieving a minute-level response from data collection to decision adjustment. This highly timely dynamic control capability ensures that alfalfa receives the most suitable water and fertilizer supply at different growth stages, significantly improving resource utilization efficiency and stress resistance, and providing strong technical support for high and stable alfalfa yields.

[0026] In this preferred embodiment, forming an expert knowledge base specifically includes: From the growth environment database, environmental data samples corresponding to the high-yield records of alfalfa were selected; multivariate statistical analysis was performed on the high-yield environmental data samples to calculate the mean and standard deviation of each environmental parameter under high-yield conditions; based on the mean and standard deviation, an optimal target interval and an acceptable warning interval were set for each environmental parameter. The optimal target interval is used to guide precise management, and the acceptable warning interval is used to trigger intervention measures; the interval thresholds of each environmental parameter are associated with the corresponding alfalfa growth period to construct an expert knowledge base.

[0027] Specifically, by conducting multivariate statistical analysis on environmental data samples from high-yield years, the optimal target range and acceptable warning range for each environmental parameter under high-yield conditions can be scientifically determined. These threshold ranges are more statistically reliable and regionally adaptable than traditional empirical values. More importantly, by dynamically linking the threshold values ​​of each parameter with different growth stages of alfalfa, the expert knowledge base can provide differentiated management standards for each key growth stage, ensuring the most appropriate water and fertilizer regulation at different stages such as seedling, branching, and budding. Data-driven precision management strategies significantly improve water and nutrient utilization efficiency, creating optimal growth conditions for high alfalfa yields and ultimately achieving a significant increase in yield per unit area.

[0028] Furthermore, in this preferred embodiment, an environmental data sample set corresponding to the high-yield alfalfa record year is selected from the growth environment database, and the following operations are performed: A comprehensive scoring model for high-yield years is constructed. Based on the degree of matching between various environmental parameters and optimal thresholds, the contribution of environmental conditions in each historical year to alfalfa high yield is quantitatively evaluated. The mathematical expression of the comprehensive scoring model for high-yield years is as follows: , in, The comprehensive high-yield score representing the j-th historical year is a dimensionless scalar. The closer its value is to 1, the closer the environmental conditions of that year are to the optimal state for high alfalfa yield. These represent the annual average values ​​of various environmental parameters (precipitation, effective accumulated temperature, organic matter content, and pH value) in year j, with their physical units and thresholds as described above. , , , The definitions are completely consistent; The optimal threshold values ​​for each parameter defined above; This represents the effective range of pH values ​​that can be tolerated; it is a constant used to normalize pH deviations. The index weights for climate factors, soil fertility factors, and soil environmental factors are determined by performing a multiple nonlinear regression fitting on historical yield data and environmental data, and satisfy the following conditions: ; Calculate its for all historical years Value, and according to The values ​​are sorted from high to low; all environmental data corresponding to the years in the top K% of the sorted years are selected to form a sample set of environmental data for alfalfa high-yield record years; based on this sample set, the kernel density estimation method is used to calculate the probability density distribution of each environmental parameter under high-yield conditions, and then the threshold boundaries of the optimal target interval and the acceptable warning interval are determined to ensure that the expert knowledge base is derived from the statistical nature of high-yield environmental data.

[0029] Specifically, using mathematical models This study quantitatively assesses the degree of matching between environmental conditions and the optimal state for high alfalfa yield in each historical year. By introducing exponential weights for climate factors, soil fertility factors, and soil environmental factors, it scientifically reflects the relative contributions of different environmental parameters to yield formation, ensuring the objectivity and regional adaptability of the scoring results. A high-yield sample set is constructed by selecting data from the top K% of high-scoring years, and the probability density distribution of each parameter is determined using kernel density estimation. The resulting optimal target intervals and acceptable warning intervals (such as threshold ranges for precipitation, effective accumulated temperature, organic matter content, and pH) are no longer static empirical values ​​but rather derived from the statistical nature of high-yield environmental data, possessing stronger scientific validity and representativeness. This enables the constructed expert knowledge base to accurately characterize the optimal combination of environmental parameters required for high alfalfa yield, providing a reliable data foundation for subsequent precise management decisions. Ultimately, by guiding optimal environmental regulation at different growth stages, it significantly improves alfalfa yield.

[0030] Furthermore, in this preferred embodiment, multivariate statistical analysis is performed on the high-yield environmental data sample set to calculate the mean and standard deviation of each environmental parameter under high-yield conditions, and the following operations are performed: A dynamic weighted statistical model based on factor sensitivity is constructed, which calculates the impact of various environmental parameters on high-yield scores. The contribution sensitivity is dynamically adjusted in statistical calculations to more accurately reflect the core distribution characteristics of each parameter under high-yield conditions, and its weighted average value is used. with weighted standard deviation The calculation formula is as follows: , , in, This represents the value of a certain environmental parameter to be statistically analyzed in the sample of the j-th high-yield year (e.g., ); This represents the dynamic weight assigned to the j-th sample, and its value is the high-yield score for that sample. The exponential weight of this environmental parameter in the high-yield scoring model ( The product of ) Right now (k takes the index of the corresponding parameter in 1, 2, 3). This design makes the sample that contributes more to the final high yield and is more important in the parameter itself have a higher weight in the statistics. The total number of samples in the high-yield environment data sample set; This is the weighted average of the calculated environmental parameter, which is more likely to be determined by data from years that are closer to the ideal high-yield state; This is the weighted standard deviation of the calculated environmental parameter, which measures the dispersion around the weighted mean within the high-yield sample group and emphasizes the volatility of the high-weighted sample. The above dynamic weighted statistical model was applied sequentially to the four key environmental parameters—precipitation, effective accumulated temperature, organic matter content, and pH value—to calculate their respective values. and Based on the calculation results, the optimal target interval is set as [ -0.5× , +0.5× The acceptable warning range is set as []. -1.5× , +1.5× This interval definition method based on weighted statistical results can more accurately characterize the optimal environmental parameter range required for high-yield alfalfa compared to traditional static thresholds or simple average statistics.

[0031] Specifically, by scoring the high yield of samples from each high-yield year... The exponential weights of the corresponding environmental parameters Combined to generate dynamic weights This weighting mechanism ensures that samples closer to the ideal high-yield state, and whose environmental parameters have a more significant impact on yield, receive higher weights in the statistics. This weighting mechanism guarantees the weighted average value obtained in the calculation. and weighted standard deviation It can more accurately reflect the core environmental characteristics required for high-yield alfalfa production, rather than the biases that may arise from simple arithmetic averages. Based on and The set optimal target interval [ -0.5× , +0.5× ] and acceptable warning range[ -1.5× , +1.5× This method not only statistically captures the core distribution range of high-yield environmental parameters, but also amplifies the contribution of important samples through weighted amplification. This makes the final determined threshold ranges (such as the suitable ranges for precipitation, effective accumulated temperature, organic matter content, and pH) more scientific and representative than traditional methods. Furthermore, it provides accurate and reliable environmental parameter thresholds for the expert knowledge base, laying a solid foundation for precise control of the alfalfa growth environment. Ultimately, by meeting the optimal environmental conditions required for high yield, alfalfa production is significantly increased.

[0032] Furthermore, in this preferred embodiment, the threshold ranges of each environmental parameter are associated with the corresponding alfalfa growth period to construct an expert knowledge base, and the following operations are performed: A dynamic threshold adjustment model for the reproductive period is constructed. This model refines the basic threshold range based on the differences in alfalfa's physiological needs and sensitivity to environmental factors at different reproductive stages. The formula for the dynamic threshold adjustment model for the reproductive period is as follows: , in, This represents the upper or lower limit of the adjusted environmental parameter threshold under a specific growth phase, which includes the seedling stage, branching stage, budding stage, flowering stage, and pod-setting stage. and The weighted average and weighted standard deviation of the environmental parameters calculated above are used as the basic statistics; N is a multiple of the standard deviation used to define the width of the base interval; N=0.5 for the optimal target interval and N=1.5 for the acceptable warning interval. The demand sensitivity coefficient of an environmental parameter under a specific reproductive phase is a dimensionless parameter. Its value is determined by analyzing the deviation of the environmental parameter from its value under a specific reproductive phase in a high-yield environmental data sample set. At that time, the final yield (or high yield score) was evaluated. The degree of influence is determined by the amount of precipitation during the water-sensitive period (such as the budding stage). The OM content is relatively high, especially during the period when nutrients are most sensitive (such as the branching stage). The value is relatively large; This represents the actual monitored or predicted environmental parameter values ​​during that specific reproductive phase. This represents the historical average value of this environmental parameter during this growth phase, which can be extracted from a basic database of growth environments. This represents the historical range of variation of the environmental parameter within the reproductive phase, and the difference between the maximum and minimum values ​​is used to normalize the deviation. Based on the above model, the optimal target range and acceptable warning range for each environmental parameter after adjustment are calculated for each major reproductive period; the threshold values ​​for all adjusted ranges for all reproductive periods, and the parameters required to calculate them are also calculated. , , , , The system stores and maps information in a structured manner according to reproductive period and environmental parameter type to form an expert knowledge base. This dynamic binding method ensures that the thresholds in the knowledge base can respond to the specific needs of different reproductive periods, providing a more scientific basis for subsequent accurate decision-making.

[0033] Specifically, based on the demand sensitivity coefficients of various environmental parameters at different reproductive stages... This involves precisely modulating the basic threshold range. For example, appropriately raising the upper and lower limits of the precipitation threshold during water-sensitive periods (such as the budding stage), and adjusting the appropriate range of organic matter content during critical nutrient demand periods (such as the branching stage), so that the threshold settings better align with the physiological characteristics of alfalfa at each growth stage. This is achieved by introducing real-time monitoring values. Compared with historical average The model can also dynamically adjust the threshold boundaries in response to changes in actual environmental conditions, thanks to the deviation normalization term. This refined control ensures that the thresholds in the expert knowledge base are no longer static values, but intelligent standards that adaptively adjust with changes in the growth stage and environment. This embodiment provides a tailored environmental management basis for different growth stages, enabling water and fertilizer control measures to accurately match the dynamic needs of alfalfa, greatly improving resource utilization efficiency and environmental adaptability, and ultimately significantly increasing alfalfa yield by optimizing growth conditions at each growth stage.

[0034] In this preferred embodiment, the collection of crop growth status data and environmental monitoring data of the target alfalfa field specifically includes: Using a multispectral UAV remote sensing platform, the normalized difference vegetation index, leaf area index, and canopy temperature of alfalfa canopy are regularly acquired; a network of sensor nodes buried in the soil is used to monitor soil temperature, soil volumetric water content, and electrical conductivity at different soil depths in real time; and field weather stations are used to collect real-time data on air temperature and humidity, photosynthetically active radiation, and wind speed. Crop growth status data and environmental monitoring data are synchronized in time and spatially registered and stored in a cloud data platform.

[0035] Specifically, through the collaborative operation of drone remote sensing, soil sensor networks, and field weather stations, simultaneous monitoring and spatiotemporal registration of alfalfa canopy physiological indicators, soil profile environmental parameters, and near-surface meteorological elements were achieved. This allows for precise capture of alfalfa's growth status at different growth stages and its interactions with environmental factors, providing high-resolution data support for subsequent precise decision-making. Through centralized storage and processing on a cloud-based data platform, this multi-source data can be compared in real-time with an expert knowledge base, triggering timely control measures such as precision irrigation and topdressing to ensure optimal growth conditions for alfalfa throughout its entire growth cycle, ultimately providing reliable data assurance for achieving high alfalfa yields.

[0036] In this preferred embodiment, generating precision agronomic operation decision instructions specifically includes: The system compares the real-time monitored soil volumetric water content with the optimal target range for soil water content during the current growth stage in the expert knowledge base. If the water content is lower than the optimal target range, an irrigation decision instruction is generated, which includes an estimated amount of water required for irrigation. The system also compares the real-time monitored soil electrical conductivity data with the warning threshold in the expert knowledge base. If the nitrogen content is low, a topdressing decision instruction is generated, which includes a suggested fertilizer formula and dosage. The decision instruction is transmitted to the central control system via a wireless communication network.

[0037] Specifically, by comparing real-time monitored soil volumetric water content and electrical conductivity data with optimal thresholds dynamically adjusted according to the growth stage in an expert knowledge base, the system can accurately identify the water and nutrient gaps in alfalfa at its current growth stage. Once soil moisture content is found to be below the optimal target range or nitrogen content is indicated to be low, a decision instruction containing specific water volume estimates, fertilizer formulas, and dosages is immediately generated and transmitted to the central control system via a wireless network. This ensures that water and fertilizer supply accurately matches the dynamic needs of alfalfa at different growth stages, significantly improving the timeliness and accuracy of irrigation and fertilization, and providing a crucial guarantee for high alfalfa yields.

[0038] In this preferred embodiment, the central control system drives the intelligent field execution equipment to complete automated operations, specifically including: The central control system receives irrigation decision commands, analyzes the estimated amount of water needed for irrigation, calculates the smart valve groups to be opened and their duration, and controls the integrated water and fertilizer system to perform precision irrigation operations. It also receives topdressing decision commands, controls the automatic fertilizer dispenser to prepare liquid fertilizer according to the fertilizer formula and dosage, and performs the fertilization operation synchronously through the integrated water and fertilizer system during the irrigation process. After the operation is completed, the sensor network collects environmental data again and feeds the executed data back to the system to evaluate the decision effect and optimize subsequent commands.

[0039] Specifically, the central control system analyzes irrigation and fertilization decisions to precisely control intelligent valve groups and the integrated water and fertilizer system, achieving on-demand and precise delivery of water and fertilizer. After the operation is completed, environmental data is immediately collected via a sensor network, and this feedback data is compared and analyzed with the expected targets to evaluate the effectiveness of the decision execution. Finally, the system continuously learns and optimizes, developing adaptive control capabilities to ensure that alfalfa remains in the optimal water and fertilizer environment throughout its growth cycle, ultimately significantly increasing alfalfa yield by improving water and nutrient utilization efficiency.

[0040] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A high-yield alfalfa cultivation method, characterized in that, Includes the following steps: S1. Collect historical and real-time environmental data of the target planting area to construct an alfalfa growth environment database; the environmental data includes climate data and soil moisture data; S2. Based on the growth environment database, extract the threshold range of optimized environmental parameters that conform to the high-yield alfalfa growth model to form an expert knowledge base; S3. Through a sensor network deployed in the field, real-time data on crop growth status and environmental monitoring of the target alfalfa field are collected. S4. Compare the crop growth status data and the environmental monitoring data with the optimized environmental parameter threshold range in the expert knowledge base, and generate precision agronomic operation decision instructions based on the comparison results. S5. The central control system sends the agronomic operation decision instructions to the field intelligent execution equipment to drive it to complete automated operations.

2. The method for high-yield alfalfa cultivation according to claim 1, characterized in that: The construction of the alfalfa growth environment database in S1 includes: collecting historical meteorological data and soil profile data for at least one complete growing season in the target planting area; the historical meteorological data includes precipitation, sunshine hours, effective accumulated temperature, and diurnal temperature range; the soil profile data includes soil pH, organic matter content, nitrogen, phosphorus, and potassium content, soil bulk density, and field water holding capacity at different depths; integrating the historical meteorological data and the soil profile data through a geographic information system and unifying them under the same spatial coordinate framework to form a basic database of the growth environment with spatiotemporal attributes; and setting a data update mechanism to continuously collect real-time meteorological monitoring data and soil sensor data through IoT nodes to dynamically update and expand the basic database of the growth environment.

3. The method for high-yield alfalfa cultivation according to claim 2, characterized in that: The historical meteorological data and soil profile data are integrated through a geographic information system and unified under the same spatial coordinate framework to form a basic database of growth environment with spatiotemporal attributes. The following operations are performed: By overlaying and fusing the gridded historical meteorological data with the soil profile data using a geographic information system, a multivariate environmental feature vector for each geographic grid cell is generated. This leads to the construction of a comprehensive alfalfa growth suitability assessment model, which iterates through all geographic grid cells to calculate their comprehensive suitability index. The comprehensive suitability index of each geographic grid unit is calculated. As a newly added attribute, it is stored in association with the existing spatial location information in the database; based on Based on the spatial distribution of values, the target planting area is divided into zones and plots, and differentiated variety selection and initial management strategies are assigned to areas with different suitability levels. These strategies include those based on… The value allows for quantitative adjustments to sowing density and basal fertilizer formulation; The mathematical expression for the comprehensive assessment model of alfalfa growth suitability is: , in, The comprehensive suitability index represents the i-th geographic grid unit; This represents the cumulative precipitation in that grid cell during the critical growth period; This represents the optimal water requirement threshold for this growth stage under the high-yield alfalfa model. This represents the effective accumulated temperature of the grid cell during a specific reproductive stage; This represents the optimal accumulated temperature threshold required for alfalfa to complete this growth stage. This represents the soil organic matter content of that grid cell; The optimal soil organic matter content threshold required for high alfalfa yield; The soil pH value represents the value of the grid cell. The optimal soil pH value for alfalfa growth; These are the weighting coefficients for climate factors, soil fertility factors, and soil environmental factors, respectively, and they satisfy the following conditions: .

4. The method for high-yield alfalfa cultivation according to claim 3, characterized in that: The established data update mechanism continuously collects real-time meteorological monitoring data and soil sensor data through IoT nodes to dynamically update and expand the basic database of the growth environment, and performs the following operations: Based on the comprehensive evaluation model for alfalfa growth suitability, a dynamic data value evaluation function is constructed to intelligently filter and integrate real-time monitoring data, representing the degree of value change. Set a dynamic threshold ; When calculated in real time Value exceeds When the real-time data set is determined to be high-value data, the following operations are automatically triggered: the high-value data set and its corresponding spatial location information are synchronously updated to the growth environment basic database, and the current comprehensive suitability index of the grid cell is recalculated using this new data point. If the updated If the value decreases significantly, an environmental stress warning is generated, and the central control system is triggered to reassess the current irrigation or fertilization decision instructions, thereby achieving data-driven, highly timely dynamic regulation. The mathematical expression of the function is: , in, This represents the degree of value change of a real-time environmental data unit collected at time t; These represent the real-time precipitation, real-time accumulated temperature, and real-time soil organic matter content monitoring values ​​collected at time t, respectively. These represent the corresponding monitoring values ​​stored in the database at the previous data collection time point t-1; These are the weighting coefficients for changes in precipitation, accumulated temperature, and organic matter content, respectively.

5. A method for high-yield alfalfa cultivation according to claim 1 or 4, characterized in that: The formation of the expert knowledge base mentioned in S2 includes: From the aforementioned growth environment database, environmental data sample sets corresponding to the years of high alfalfa yield records were selected; Multivariate statistical analysis was performed on the high-yield environmental data sample set to calculate the mean and standard deviation of each environmental parameter under high-yield conditions. Based on the mean and standard deviation, an optimal target range and an acceptable warning range are set for each environmental parameter. The optimal target range is used to guide precise management, and the acceptable warning range is used to trigger intervention measures. The expert knowledge base is constructed by associating the range thresholds of each environmental parameter with the corresponding alfalfa growth period.

6. The method for high-yield alfalfa cultivation according to claim 5, characterized in that: From the aforementioned growth environment database, select the environmental data sample set corresponding to the years with high alfalfa yield records, and perform the following operations: A comprehensive scoring model for high-yield years is constructed. Based on the degree of matching between various environmental parameters and optimal thresholds, the contribution of environmental conditions to high alfalfa yield in each historical year is quantitatively evaluated, and the model is used to calculate the yield of alfalfa for all historical years. Value, and according to The values ​​are sorted from high to low; all environmental data corresponding to the years in the top K% of the sorted years are selected to form the environmental data sample set of the alfalfa high-yield record years; based on this sample set, the kernel density estimation method is used to calculate the probability density distribution of each environmental parameter under high-yield conditions, and then the threshold boundary between the optimal target interval and the acceptable warning interval is determined to ensure that the expert knowledge base is derived from the statistical nature of high-yield environmental data. The mathematical expression for the comprehensive scoring model for high-yield years is: , in, The overall high-yield score represents the j-th historical year; These represent the annual average values ​​of each environmental parameter in year j. The effective range of pH values ​​allowed; These are the index weights of climate factors, soil fertility factors, and soil environmental factors, respectively, and they satisfy the following conditions: .

7. The method for high-yield alfalfa cultivation according to claim 6, characterized in that: Multivariate statistical analysis was performed on the high-yield environmental data sample set to calculate the mean and standard deviation of each environmental parameter under high-yield conditions, and the following operations were performed: A dynamic weighted statistical model based on factor sensitivity is constructed. The model calculates the impact of various environmental parameters on high yield scores. The contribution sensitivity is dynamically adjusted in statistical calculations to more accurately reflect the core distribution characteristics of each parameter under high-yield conditions, and its weighted average value is used. with weighted standard deviation The calculation formula is as follows: , ,in, This represents the value of a certain environmental parameter to be statistically analyzed in the sample of the j-th high-yield year; This represents the dynamic weight assigned to the j-th sample, and its value is the high-yield score for that sample. The product of this environmental parameter and its exponential weight in the high-yield scoring model, i.e. ; The total number of samples in the high-yield environment data sample set; This is the weighted average of the calculated environmental parameter; This is the weighted standard deviation of the calculated environmental parameter; The above dynamic weighted statistical model was applied sequentially to the four key environmental parameters—precipitation, effective accumulated temperature, organic matter content, and pH value—to calculate their respective values. and Based on the calculation results, the optimal target interval is set as [ -0.5× , +0.5× The acceptable warning range is set as [ ]. -1.5× , +1.5× ].

8. The method for high-yield alfalfa cultivation according to claim 7, characterized in that: The process involves associating and binding the range thresholds of various environmental parameters with the corresponding alfalfa growth stages to construct the expert knowledge base, and then performing the following operations: A dynamic threshold adjustment model for the reproductive period is constructed. Based on the differences in physiological needs and environmental factor sensitivities of alfalfa at different reproductive stages, the model finely modulates the basic threshold range. The formula for the dynamic threshold adjustment model for the reproductive period is as follows: ， in, This represents the upper or lower limit of the adjusted environmental parameter threshold under a specific reproductive phase. It is a multiple of the standard deviation; This is the demand sensitivity coefficient of this environmental parameter under a specific reproductive phase; This refers to the specific reproductive phase. This represents the reproductive phase. This represents the historical variation range of the environmental parameter within the reproductive phase; based on the model, the optimal target range and acceptable warning range of each environmental parameter after adjustment are calculated for each major reproductive phase; the adjusted range thresholds for all reproductive phases and the parameters required for calculation are structured, stored, and mapped according to reproductive phase and environmental parameter type to form the expert knowledge base.

9. The method for high-yield alfalfa cultivation according to claim 1, characterized in that: The generation of precision agronomic operation decision instructions in S4 includes: The real-time monitored soil volumetric water content is compared with the optimal target range of soil water content under the current growth period in the expert knowledge base; If the value is below the optimal target range, an irrigation decision instruction is generated, which includes an estimated amount of water to be irrigated. The real-time monitored soil electrical conductivity data is compared with the early warning threshold in the expert knowledge base; If the indicated nitrogen content is low, a topdressing decision instruction is generated, which includes a suggested fertilizer formula and dosage. The decision-making instructions are transmitted to the central control system via a wireless communication network.

10. A method for high-yield alfalfa cultivation according to claim 1, characterized in that: The central control system in S5 drives the intelligent field execution equipment to complete automated operations, including: The central control system receives the irrigation decision command, parses the estimated amount of water to be irrigated, calculates the intelligent valve group to be opened and its duration, and controls the integrated water and fertilizer system to perform precision irrigation operations. Upon receiving the topdressing decision instruction, the system controls the automatic fertilizer mixing machine to prepare liquid fertilizer according to the fertilizer formula and dosage, and simultaneously performs the fertilization operation through the integrated water and fertilizer system during the irrigation process; After the task is completed, the sensor network collects environmental data again and feeds the executed data back to the system to evaluate the decision-making effect and optimize subsequent instructions.

Images