Multi-source data fusion irrigation regulation and control system and method for smart agriculture

Abstract

The invention belongs to the technical field of irrigation regulation and control, and discloses a multi-source data fusion irrigation regulation and control system and method for smart agriculture. Comprising the following steps: collecting historical irrigation records and crop yield data of each farmland plot, and constructing a benefit response model library of each farmland plot; underground water level monitoring data and soil salinity accumulation data are collected in real time, and an ecological health degree attenuation curve of each farmland plot is generated; obtaining water resource management data and ecological red line constraint parameters, determining an irrigation amount feasible region of each farmland plot, and determining a sensitive plot in combination with an ecological health degree attenuation curve; according to the irrigation amount feasible region and the ecological health degree attenuation curve of each sensitive land parcel, an optimal benefit-ecological balance point is intelligently identified, and an eco-friendly irrigation regulation and control scheme is generated; according to the invention, intelligent irrigation water volume scheduling is realized on the premise of considering agricultural production benefits and regional ecological sustainability, and progressive degradation of an agricultural ecological system is effectively avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of irrigation regulation, more particularly, the present application relates to a multi-source data fusion irrigation regulation system and method for smart agriculture. BACKGROUND

[0002] With the wide penetration of Internet of Things, big data and artificial intelligence technology in the field of agriculture, smart agriculture has become the core driving force to promote the transformation of agricultural modernization, and precise irrigation regulation as one of the key application scenarios of smart agriculture has important significance for improving the efficiency of agricultural water use, ensuring stable crop yield and promoting sustainable development of agriculture. The existing irrigation regulation system usually adopts a control strategy based on soil moisture monitoring or crop water demand model, that is, by deploying soil humidity sensors in farmland plots to collect real-time soil water content data, combining the water demand coefficient of crops at different growth stages and the current meteorological evapotranspiration parameters, the irrigation demand of each farmland plot in a specific period is calculated, and then the corresponding irrigation execution scheme is formulated according to the demand and issued to the field irrigation equipment to realize the on-demand precise supply of water resources.

[0003] However, the above existing technology has the following shortcomings: on the one hand, the existing method mainly focuses on the short-term water demand of crops and the immediate yield improvement, and fails to establish a benefit response model between irrigation investment and yield output, making it difficult to optimize irrigation decisions from the long-term economic benefit perspective; on the other hand, the existing method ignores the negative impact of excessive irrigation on the decline of groundwater level and the accumulation of soil salt, lacks a continuous monitoring and constraint mechanism for the ecological health status of farmland, and is prone to cause progressive degradation of the agricultural ecosystem, making it difficult to achieve a coordinated balance between irrigation benefits and ecological protection; therefore, there is an urgent need for a sustainable irrigation regulation system that can comprehensively consider the economic benefits of irrigation and the ecological health constraints, and support the benefit-ecological balance optimization.

[0004] In view of this, the present application proposes a multi-source data fusion irrigation regulation system and method for smart agriculture to solve the above problems. SUMMARY

[0005] In order to overcome the above-mentioned defects of the prior art and achieve the above-mentioned purposes, the present application provides the following technical scheme: a multi-source data fusion irrigation regulation method for smart agriculture, comprising: collecting historical irrigation records and corresponding crop yield data of each farmland plot, modeling the response relationship between different irrigation water amounts and crop yield data in the historical irrigation records, and constructing a benefit response model library for each farmland plot; real-time collection of underground water level monitoring data and soil salt accumulation data around the farmland, evaluation of the evolution risk degree of each farmland plot from the current ecological state to the critical state of ecological degradation, and generation of an ecological health degree decay curve for each farmland plot; Obtain water resource management data and ecological red line constraint parameters, calculate the theoretical optimal irrigation amount of each farmland plot under the constraint of the benefit response model library, determine the irrigation amount feasible region of each farmland plot based on the theoretical optimal irrigation amount, and determine the sensitive plot in combination with the ecological health degree attenuation curve; According to the irrigation amount feasible region and the ecological health degree attenuation curve of each sensitive plot, analyze the long-term benefit net present value and the ecological recovery cycle estimation value of each sensitive plot under different irrigation intensity schemes, and intelligently identify the optimal benefit-ecological balance point; Based on the optimal benefit-ecological balance point of each sensitive plot, generate an ecological-friendly irrigation regulation scheme, and intelligently schedule the irrigation water of each sensitive plot according to the ecological-friendly irrigation regulation scheme.

[0006] Further, the method for constructing the benefit response model library of each farmland plot comprises: According to historical irrigation records and crop yield data, construct irrigation-yield paired data for each farmland plot; the irrigation-yield paired data includes unit area irrigation amount, unit area yield, and unit area benefit; for the unit area irrigation amount and unit area yield in each irrigation-yield paired data, perform a data quality screening operation to obtain irrigation-yield screening data; For the irrigation-yield screening data of each farmland plot, respectively perform yield response relationship modeling and benefit response relationship modeling to obtain the yield response function and the benefit response function of each farmland plot; according to each yield response function, calculate the corresponding yield optimal irrigation amount and theoretical maximum yield; according to each benefit response function, calculate the corresponding benefit optimal irrigation amount and theoretical maximum benefit; take the first derivative of the benefit response function to obtain the marginal benefit function; and according to the marginal benefit function, calculate the benefit critical irrigation amount; Integrate the yield response function, the benefit response function, the yield optimal irrigation amount, the benefit optimal irrigation amount, the theoretical maximum yield, the theoretical maximum benefit, and the benefit critical irrigation amount of each farmland plot to form the benefit response model of each farmland plot; and aggregate the benefit response models of all farmland plots to construct a benefit response model library.

[0007] Further, the method for evaluating the evolution risk degree of each farmland plot from the current ecological state to the ecological degradation critical state comprises: The groundwater level monitoring data includes monitoring point basic information and multiple water level monitoring records, and the soil salt accumulation data includes sampling point basic information and multiple salt monitoring records; obtain the associated plot code in each water level monitoring record and salt monitoring record, associate all water level monitoring records and salt monitoring records with each farmland plot according to the associated plot code; integrate the water level monitoring records and the salt monitoring records of the same farmland plot to form an ecological monitoring original data set of each farmland plot; According to the ecological monitoring original data set, the groundwater level health index and the soil salt health index of each farmland plot are calculated; in a preset historical analysis period, all water level monitoring records and salt content monitoring records of each farmland plot are obtained, and the water level change rate and the salt accumulation trend of each farmland plot are obtained from the water level monitoring records and the salt content monitoring records respectively; based on the water level change rate and the salt accumulation trend, the water level decline rate and the salt accumulation rate of each farmland plot are calculated respectively; Based on the preset ecological weight, the groundwater level health index and the soil salt health index of the same farmland plot are weighted and summed to obtain the comprehensive ecological health index of each farmland plot; based on the preset degradation weight, the water level decline rate and the salt accumulation rate of the same farmland plot are weighted and summed to obtain the ecological degradation rate of each farmland plot; according to the comprehensive ecological health index and the ecological degradation rate of each farmland plot, the evolution risk degree of each farmland plot is calculated.

[0008] Further, the method for generating the ecological health degree attenuation curve of each farmland plot comprises: According to the ecological degradation rate and the evolution risk degree, the ecological attenuation coefficient of each farmland plot is calculated; taking time as the independent variable and ecological health degree as the dependent variable, and combining the ecological attenuation coefficient and the comprehensive ecological health index of each farmland plot, the change law of the ecological health degree with time is described by using an exponential attenuation function to obtain the ecological health degree attenuation curve of each farmland plot.

[0009] Further, the method for calculating the theoretical optimal irrigation amount of the farmland plot comprises: According to the yield optimal irrigation amount and the benefit optimal irrigation amount, the comprehensive optimal irrigation amount is calculated; according to the theoretical maximum yield, the yield guarantee constraint irrigation amount is calculated; according to the water resource management data, the water resource constraint irrigation amount is calculated; according to the ecological red line constraint parameter, the ecological constraint irrigation amount is calculated; The water resource constraint irrigation amount and the ecological constraint irrigation amount are compared, and the smaller value of the two is taken as the hard constraint upper limit; the yield guarantee constraint irrigation amount and the hard constraint upper limit are compared; if the yield guarantee constraint irrigation amount is greater than the hard constraint upper limit, the hard constraint upper limit is taken as the theoretical optimal irrigation amount, and the corresponding farmland plot is marked as a yield limited plot; if the yield guarantee constraint irrigation amount is less than or equal to the hard constraint upper limit, the comprehensive optimal irrigation amount and the hard constraint upper limit are compared; If the comprehensive optimal irrigation amount is greater than the hard constraint upper limit, the hard constraint upper limit is taken as the theoretical optimal irrigation amount; if the comprehensive optimal irrigation amount is less than or equal to the hard constraint upper limit, the comprehensive optimal irrigation amount and the yield guarantee constraint irrigation amount are compared, and the larger value of the two is taken as the theoretical optimal irrigation amount.

[0010] Further, the method for determining the irrigation amount feasible region of the farmland plot comprises: According to the benefit response model, the benefit constraint irrigation amount is determined; the yield guarantee constraint irrigation amount and the benefit constraint irrigation amount are compared, and the larger value of the two is taken as the lower limit of the irrigation amount feasible region; and the theoretical optimal irrigation amount is taken as the upper limit of the irrigation amount feasible region; If the lower limit of the irrigation amount feasible region is greater than the upper limit of the irrigation amount feasible region, the irrigation amount feasible region corresponding to the farmland plot is marked as an empty region; if the lower limit of the irrigation amount feasible region is less than or equal to the upper limit of the irrigation amount feasible region, the irrigation amount feasible region corresponding to the farmland plot is expressed in the form of a closed interval; The method for determining the sensitive plot comprises: A planning time span is preset, the planning time span is substituted into the ecological health degree attenuation curve of each farmland plot respectively, the ecological health degree of each farmland plot is calculated, and is marked as a predicted health degree; according to the irrigation amount feasible region of each farmland plot, the feasible region width of each farmland plot is calculated; according to the comprehensive optimal irrigation amount and the theoretical optimal irrigation amount, the benefit-ecological conflict degree of each farmland plot is calculated; If the feasible region width is less than a preset feasible region width threshold value, the corresponding farmland plot is marked as a sensitive plot; if the benefit-ecological conflict degree is greater than a preset conflict degree threshold value, the corresponding farmland plot is marked as a sensitive plot; if the predicted health degree is less than a preset ecological health degree early warning threshold value, the corresponding farmland plot is marked as a sensitive plot; if the farmland plot is marked as a yield limited plot, the corresponding farmland plot is marked as a sensitive plot.

[0011] Further, the method for analyzing the long-term benefit net present value and the ecological recovery cycle estimation value of the sensitive plot under different irrigation intensity schemes comprises: According to the irrigation amount feasible region of the sensitive plot, a plurality of irrigation intensity schemes are generated; a benefit evaluation period and an annual discount rate are preset; the irrigation intensity scheme is substituted into the benefit response function, and the annual expected benefit is calculated; the water supply guarantee rate is obtained from the water resource management data; according to the annual expected benefit and the water supply guarantee rate, the annual actual benefit is calculated; for each year within the benefit evaluation period, the corresponding discount coefficient is calculated; according to the annual actual benefit and the discount coefficient corresponding to each year, the discount benefit of each year is calculated; according to the discount benefit of all years within the benefit evaluation period, the long-term benefit net present value is calculated; If the irrigation amount feasible region is null space, the ecological adjustment coefficient is calculated according to the irrigation intensity scheme; the minimum value of the underground water level is obtained from the ecological red line constraint parameter; the lower threshold of the ecological health degree is calculated according to the minimum value of the underground water level; the ecological adjustment coefficient, the comprehensive ecological health index and the lower threshold of the ecological health degree are substituted into the ecological health degree attenuation curve to obtain the ecological degradation period; the ecological restoration target health degree and the restoration completion ratio are preset to calculate the restoration target level; if the comprehensive ecological health index is less than the ecological restoration target health degree, the ecological restoration function is constructed; the function value of the ecological restoration function is equal to the restoration target level to obtain the ecological restoration period estimate value; if the current comprehensive ecological health index is greater than or equal to the ecological restoration target health degree, the ecological restoration period estimate value is set to zero.

[0012] Further, the method of intelligently identifying the optimal benefit-ecological balance point comprises: The ecological sustainability judgment condition is preset; for each sensitive plot, the irrigation intensity scheme with the largest value is selected from all irrigation intensity schemes satisfying the ecological sustainability judgment condition as the sustainable irrigation intensity threshold; if there is no irrigation intensity scheme satisfying the ecological sustainability judgment condition, the corresponding sensitive plot is marked as an ecological priority plot, and it is determined whether the irrigation amount feasible region is null space; if the irrigation amount feasible region is null space, the theoretical optimal irrigation amount is taken as the sustainable irrigation intensity threshold; if the irrigation amount feasible region is not null space, the lower limit of the irrigation amount feasible region is taken as the sustainable irrigation intensity threshold; The long-term benefit net present value and the ecological restoration period estimate value of each sensitive plot are normalized to obtain the standard benefit net present value and the standard restoration period estimate value; the standard benefit net present value and the standard restoration period estimate value of the same irrigation intensity scheme corresponding to the same sensitive plot are weighted and subtracted to obtain the comprehensive evaluation score of each sensitive plot corresponding to each irrigation intensity scheme; the irrigation intensity scheme with the highest comprehensive evaluation score is taken as the optimal benefit-ecological balance point of the corresponding sensitive plot; the regional water consumption total amount is checked for all sensitive plots to determine whether the optimal benefit-ecological balance point needs to be updated, and the optimal benefit-ecological balance point that needs to be updated is updated according to the sustainable irrigation intensity threshold.

[0013] Further, the method of generating an eco-friendly irrigation control scheme comprises: obtaining the corresponding geographic coordinates of the land plots and the unique codes of the land plots from historical irrigation records of each sensitive land plot; constructing a sensitive land plot neighborhood correlation graph according to the geographic coordinates of each sensitive land plot; calculating the neighborhood ecological coupling factor of each sensitive land plot according to the sensitive land plot neighborhood correlation graph and the ecological degradation rate of each sensitive land plot; obtaining the crop growth stage of each sensitive land plot, and calculating the growth stage sensitivity coefficient of each sensitive land plot; obtaining meteorological forecast data, and calculating the meteorological response adjustment coefficient of each sensitive land plot; According to the optimal benefit-ecological balance point, the neighborhood ecological coupling factor, the growth stage sensitivity coefficient and the meteorological response adjustment coefficient of each sensitive land plot, the comprehensive adjustment irrigation amount of each sensitive land plot is calculated; according to the comprehensive ecological health index and the ecological degradation rate of each sensitive land plot, the ecological urgency index of each sensitive land plot is calculated; according to the ecological urgency index and the comprehensive adjustment irrigation amount of each sensitive land plot, the irrigation control priority of each sensitive land plot is determined; according to the irrigation control priority and the comprehensive adjustment irrigation amount of each sensitive land plot, cross-plot ecological compensation adjustment is performed to obtain the comprehensive compensation irrigation amount of each sensitive land plot; The comprehensive compensation irrigation amount, the irrigation control priority and the unique code of each sensitive land plot are integrated to form the irrigation control parameters of each sensitive land plot; the irrigation control parameters of all sensitive land plots are summarized to generate an ecological-friendly irrigation control scheme.

[0014] The multi-source data fusion irrigation control system for smart agriculture implements the multi-source data fusion irrigation control method for smart agriculture, which comprises: The benefit modeling module is used to collect the historical irrigation records and the corresponding crop yield data of each farmland plot, and to regress model the response relationship between different irrigation water amounts and crop yield data in the historical irrigation records to construct a benefit response model library of each farmland plot. The ecological monitoring module is used to collect the underground water level monitoring data and the soil salt accumulation data around the farmland in real time, to evaluate the evolution risk degree of each farmland plot from the current ecological state to the ecological degradation critical state, and to generate an ecological health degree attenuation curve of each farmland plot. The constraint quantification module is used to obtain water resource management data and ecological red line constraint parameters, to calculate the theoretical optimal irrigation amount of each farmland plot under the constraint of the benefit response model library, to determine the irrigation amount feasible region of each farmland plot based on the theoretical optimal irrigation amount, and to determine the sensitive land plot in combination with the ecological health degree attenuation curve. The balance optimization module is used to analyze the long-term benefit net present value and the ecological recovery cycle estimate of each sensitive land plot under different irrigation intensity schemes according to the irrigation amount feasible region and the ecological health degree attenuation curve of each sensitive land plot, and to intelligently identify the optimal benefit-ecological balance point. A sustainable regulation module is configured to generate an eco-friendly irrigation regulation scheme based on the optimal benefit-ecological balance point of each sensitive plot and intelligently schedule the irrigation water of each sensitive plot according to the eco-friendly irrigation regulation scheme.

[0015] The technical effects and advantages of the multi-source data fusion irrigation regulation system and method for smart agriculture of the present application are as follows: By collecting the historical irrigation records and crop yield data of each farmland plot, a response relationship model between irrigation water and yield output is established, achieving the goal of optimizing irrigation decision from the long-term economic benefit perspective, which can effectively overcome the shortcomings of the prior art that only focuses on short-term water demand satisfaction while ignoring the irrigation input-output benefit; by collecting real-time underground water level monitoring data and soil salt accumulation data, an ecological health degree decay curve is constructed, establishing a continuous monitoring and quantitative evaluation mechanism for the ecological health status of farmland, effectively warning the ecological degradation risks such as underground water level decline and soil salt accumulation caused by excessive irrigation; by comprehensively considering the water resource quota constraint and ecological red line constraint parameters, the irrigation amount feasible region and theoretical optimal irrigation amount of each farmland plot are determined, realizing the optimization of irrigation decision under multiple resource and ecological constraints; by analyzing the long-term benefit net present value and ecological recovery cycle estimate value under different irrigation intensity schemes, the optimal benefit-ecological balance point is intelligently identified, realizing the collaborative balance of irrigation economic benefit and ecological protection goal; by constructing the sensitive plot neighborhood correlation graph, calculating the neighborhood ecological coupling factor, and combining the crop growth stage sensitivity and meteorological response adjustment mechanism, fine irrigation regulation considering spatial ecological correlation and time-varying environmental factors is realized; by performing cross-plot ecological compensation adjustment, the irrigation quota is dynamically redistributed according to the ecological urgency priority, an eco-friendly irrigation regulation scheme is generated, and finally intelligent irrigation water scheduling is realized under the premise of considering agricultural production benefit and regional ecological sustainability, thereby effectively avoiding the progressive degradation of the agricultural ecosystem and providing technical support for the sustainable development of smart agriculture. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 FIG. 1 is a schematic diagram of the multi-source data fusion irrigation regulation system for smart agriculture of the present application; Figure 2 FIG. 2 is a flowchart of the multi-source data fusion irrigation regulation method for smart agriculture of the present application. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application. Embodiment One:

[0018] Referring to Figure 1 The smart agriculture-oriented multi-source data fusion irrigation control system described in the embodiment comprises a benefit modeling module, an ecological monitoring module, a constraint quantification module, a balance optimization module and a sustainable control module. The modules are connected through wired and / or wireless means to realize data transmission between the modules.

[0019] The benefit modeling module is configured to collect historical irrigation records and corresponding crop yield data of each farmland plot, perform regression modeling on the response relationship between different irrigation water quantities and crop yield data in the historical irrigation records, and construct a benefit response model library of each farmland plot.

[0020] The method for collecting historical irrigation records and corresponding crop yield data of each farmland plot comprises: The historical irrigation records of each farmland plot are obtained from an irrigation control system (i.e. an automatic control platform for controlling and recording the operation state of farmland irrigation equipment and irrigation execution data). The farmland plot refers to an agricultural production area that is divided according to geographical boundaries and planting management units and has independent irrigation control capability. The historical irrigation records are used to record all irrigation activity information of the farmland plot in each historical growing season, specifically including plot basic information and multiple irrigation event records. The plot basic information includes plot unique code, plot area, plot geographical coordinates, soil type and main crop type. The plot unique code is a pre-defined unique identification code of the farmland plot. The plot geographical coordinates include the longitude and latitude values of the plot center point. The soil type includes, but is not limited to, sandy soil, loam, clay, sandy loam, etc. The main crop type includes food crops (such as wheat, corn, soybeans, etc.), economic crops (such as cotton, tea, tobacco, etc.) and vegetable crops (such as tomatoes, spinach, cabbage, etc.). Each irrigation event record contains a growing season identifier, an irrigation execution time, an irrigation water quantity, an irrigation duration and an irrigation method. The growing season identifier is used to identify the complete growing cycle of the crop from sowing or transplanting to harvesting. The irrigation method includes, but is not limited to, drip irrigation, sprinkler irrigation, flood irrigation, micro-sprinkler irrigation, etc. The corresponding crop yield data of each farmland plot is obtained from an agricultural production management system (i.e. an information management platform for recording and managing agricultural production processes and harvesting information). The crop yield data is used to record the crop harvesting information of the farmland plot in each historical growing season, specifically including multiple yield records. Each yield record contains a plot unique code, a growing season identifier, a harvesting time, a crop variety, a total yield and a product grade distribution. The product grade distribution includes the proportion of superior products, the proportion of good products and the proportion of ordinary products. The market purchase price corresponding to each crop variety at each harvesting time is obtained from a farm product price information system (i.e., a data service platform for collecting and releasing market price information of farm products); wherein the market purchase price comprises a purchase price of superior products, a purchase price of good products, and a purchase price of ordinary products; the market purchase price is associated with the corresponding yield record to form crop yield data of each farmland plot containing price information.

[0021] The method for constructing the benefit response model library of each farmland plot comprises: According to the historical irrigation record and the crop yield data, irrigation-yield pairing data of each farmland plot is constructed; specifically, according to the unique plot code and the growth season identifier, the irrigation event record in the historical irrigation record is matched and associated with the yield record in the crop yield data; the irrigation water volume of all irrigation event records corresponding to the same growth season identifier is summed up to obtain the cumulative irrigation water volume of each growth season identifier; the ratio of the cumulative irrigation water volume to the corresponding plot area is calculated to obtain the unit area irrigation volume of each growth season identifier; the ratio of the total harvest yield to the plot area is calculated to obtain the unit area yield of each growth season identifier; the product of the proportion of superior products and the purchase price of superior products is calculated to obtain the unit price contribution of superior products; the product of the proportion of good products and the purchase price of good products is calculated to obtain the unit price contribution of good products; the product of the proportion of ordinary products and the purchase price of ordinary products is calculated to obtain the unit price contribution of ordinary products; the unit price contribution of superior products, the unit price contribution of good products, and the unit price contribution of ordinary products of the same yield record are sequentially added to obtain the comprehensive unit price of each yield record; the product of the total harvest yield of the same yield record and the comprehensive unit price is calculated to obtain the yield value of each yield record; the ratio between the yield value of each yield record and the plot area of the corresponding farmland plot is calculated to obtain the unit area benefit of each growth season identifier; the unit area irrigation volume, the unit area yield, and the unit area benefit of each growth season identifier are integrated to form the irrigation-yield pairing data of the farmland plot. For the unit area irrigation volume and the unit area yield in each irrigation-yield pairing data, a data quality screening operation is performed to obtain irrigation-yield screening data; specifically, a yield abnormality threshold and an irrigation abnormality threshold are preset, each abnormality threshold is preset by a person skilled in the art according to the statistical characteristics of historical data; the mean and the standard deviation of all unit area yields in the irrigation-yield pairing data corresponding to the farmland plot are calculated to obtain the yield mean and the yield standard deviation; the absolute value of the difference between the unit area yield of each growth season identifier and the corresponding yield mean is calculated, and then divided by the corresponding yield standard deviation to obtain the yield deviation; if the yield deviation is greater than the yield abnormality threshold, the irrigation-yield pairing data of the corresponding growth season identifier is excluded; wherein the method for data quality screening of the unit area irrigation volume is the same as the method for data quality screening of the unit area yield. The irrigation-yield screening data of each farmland plot is modeled for a yield response relationship with irrigation amount per unit area as the independent variable and yield per unit area as the dependent variable to obtain a yield response function of each farmland plot; specifically, a quadratic polynomial function is selected as a basic regression model according to the typical characteristics of a crop water production function in agronomy theory; a least squares method is used to regress and fit the irrigation-yield screening data to solve model parameters of the corresponding basic regression model; the mean of all yield per unit areas in the irrigation-yield screening data is calculated to obtain a total mean yield; the square of the difference between the actual yield per unit area of each growth season identifier and the total mean yield is calculated, and the sum of all difference squares is calculated to obtain a total sum of squares; wherein the actual yield per unit area is the actual yield per unit area of each growth season identifier recorded in the irrigation-yield screening data; the irrigation amount per unit area of each growth season identifier is substituted into the basic regression model to calculate the corresponding yield per unit area prediction value; the square of the difference between the actual yield per unit area of each growth season identifier and the yield per unit area prediction value is calculated, and the sum of all difference squares is calculated to obtain a residual sum of squares; the ratio of the residual sum of squares to the total sum of squares is calculated, and one is subtracted from the calculation result to obtain a determination coefficient; wherein the determination coefficient is used to reflect the degree of explanation of the actual data by the basic regression model, and the closer the determination coefficient is to 1, the better the fitting effect of the basic regression model; a preset goodness-of-fit threshold is compared with the determination coefficient; if the determination coefficient is greater than or equal to the goodness-of-fit threshold, it is determined that the basic regression model is fitted and qualified, and the basic regression model is taken as the yield response function of the corresponding farmland plot; if the determination coefficient is less than the goodness-of-fit threshold, a higher order polynomial (such as a cubic polynomial, a quartic polynomial, etc.) or a nonlinear regression model (such as a saturation function, an exponential function, etc.) is used to re-fit until the determination coefficient meets the requirements; wherein the goodness-of-fit threshold is pre-set by a person skilled in the art according to the model accuracy requirements; the least squares method is a known technology in the art, and the specific solving process is not described in detail here; According to the yield response functions, the corresponding optimal irrigation amounts for yield are calculated; specifically, the first derivative of the yield response function is taken and set to zero, and the unit area irrigation amount that makes the yield per unit area reach a maximum value is obtained by analytical or numerical solving method, and is taken as the optimal irrigation amount for yield; each optimal irrigation amount for yield is substituted into the corresponding yield response function to calculate the corresponding yield per unit area, which is taken as the theoretical maximum yield; it should be noted that the numerical solving method is a known technology in the art, and the specific solving process is not described in detail here; The benefit response function of each farmland plot is obtained by modeling the benefit response relationship with unit area irrigation amount as the independent variable and unit area benefit as the dependent variable according to the irrigation-yield screening data of each farmland plot; the benefit optimal irrigation amount of each farmland plot is calculated according to the benefit response function; the unit area benefit corresponding to each benefit optimal irrigation amount is calculated by substituting each benefit optimal irrigation amount into the corresponding benefit response function, and is taken as the theoretical maximum benefit; wherein the method for modeling the benefit response relationship is the same as the method for modeling the yield response relationship; the method for calculating the benefit optimal irrigation amount is the same as the method for calculating the yield optimal irrigation amount; the marginal benefit function is obtained by taking the first derivative of the benefit response function; wherein the marginal benefit function is used to describe the benefit increment brought by the increase of unit irrigation water amount; the benefit critical irrigation amount is obtained by calculating the unit area irrigation amount corresponding to the marginal benefit function equal to the preset marginal benefit lower limit value, and is used to identify the critical point at which the irrigation input-output ratio starts to be lower than the economically reasonable level; wherein the marginal benefit lower limit value is preset by the person skilled in the art according to the irrigation cost and the value of water resources; The benefit response model of each farmland plot is formed by integrating the yield response function, the benefit response function, the yield optimal irrigation amount, the benefit optimal irrigation amount, the theoretical maximum yield, the theoretical maximum benefit and the benefit critical irrigation amount of each farmland plot; the benefit response model library is constructed by summarizing the benefit response models of all farmland plots.

[0022] The ecological monitoring module is used to collect the underground water level monitoring data and the soil salt accumulation data around the farmland in real time, evaluate the evolution risk degree of each farmland plot from the current ecological state to the ecological degradation critical state, and generate the ecological health degree attenuation curve of each farmland plot.

[0023] The method for collecting the underground water level monitoring data and the soil salt accumulation data around the farmland in real time comprises the following steps: The underground water level monitoring data around each farmland plot is obtained from an underground water level monitoring system (i.e. a hydrological monitoring platform used for monitoring and recording the dynamic change of the underground water level in the farmland area in real time); wherein the underground water level monitoring data is used to record the time sequence change information of the underground water level in the surrounding area of the farmland plot, and specifically includes monitoring point basic information and multiple water level monitoring records; the monitoring point basic information includes monitoring point unique code, monitoring point geographic coordinates, monitoring point depth and associated plot code; the monitoring point unique code is the unique identification code of the pre-defined underground water level monitoring point; the monitoring point geographic coordinates include the longitude value and the latitude value of the position where the monitoring point is located; the monitoring point depth is the vertical distance from the monitoring point to the ground surface; the associated plot code is used to identify the farmland plot served by the corresponding monitoring point, and corresponds to the plot unique code; each water level monitoring record includes real-time water level depth and water level change rate; the real-time water level depth is the vertical distance from the underground water surface to the ground surface; the water level change rate is the rise-fall amplitude of the underground water level per unit time; Soil salinity accumulation data for each farmland plot is obtained from the soil salinity monitoring system (i.e., a soil environmental monitoring platform used to monitor and record the dynamic changes in soil salinity in farmland in real time). This accumulated data records the temporal changes in salinity content within the soil layer of each farmland plot, specifically including basic information about sampling points and multiple salinity monitoring records. Basic sampling point information includes a unique sampling point code, geographical coordinates, sampling depth level, and associated plot code. Sampling depth levels include topsoil, middle soil, and deep soil. Each salinity monitoring record includes salinity concentration and salinity accumulation trend. Salinity concentration is the mass concentration of soluble salts in the soil. Salinity accumulation trends include upward, stable, and downward trends.

[0024] Methods for assessing the risk of each farmland plot evolving from its current ecological state to a critical state of ecological degradation include: Based on the associated plot codes of each monitoring point and sampling point, each monitoring point and sampling point is associated and matched with the corresponding farmland plot; for each farmland plot, the water level monitoring records of all associated monitoring points and the salinity monitoring records of all sampling points are obtained; the water level monitoring records and salinity monitoring records of the same farmland plot are integrated to form the original ecological monitoring dataset for each farmland plot; Based on the original ecological monitoring dataset, groundwater health indicators for each farmland plot were calculated. Specifically, a groundwater safety threshold and a groundwater warning threshold were preset, both pre-set by those skilled in the art based on regional hydrogeological conditions. The groundwater safety threshold is the lower limit of groundwater level required to maintain the normal operation of the farmland ecosystem; the groundwater warning threshold is the critical groundwater level that may cause ecological degradation. The average water level depth was calculated by averaging all real-time water level depths corresponding to the farmland plots. The difference between the average water level depth and the groundwater level safety threshold is used to obtain the water level deviation. The difference between the groundwater level warning threshold and the groundwater level safety threshold is used to obtain the water level warning interval. The ratio of the water level deviation to the water level warning interval is used to obtain the water level deviation degree. If the water level deviation degree is less than zero, it is set to zero; if the water level deviation degree is greater than one, it is set to one. The difference between one and the water level deviation degree is used to obtain the groundwater level health index of the farmland plot. The groundwater level health index is used to reflect the health of the groundwater level of the farmland plot relative to the safety threshold. All water level monitoring records for each farmland plot and all monitoring points within a preset historical analysis period are obtained. The historical analysis period is preset by a person skilled in the art based on the needs of ecological assessment. The average water level change rate of each farmland plot is calculated by averaging all water level change rates within the same historical analysis period. If the average water level change rate is negative, it indicates that the groundwater level is declining, and the absolute value of the average water level change rate is taken as the water level decline rate. If the average water level change rate is positive or zero, the water level decline rate is set to zero.

[0025] Based on the original ecological monitoring dataset, soil salinity health indicators for each farmland plot were calculated. Specifically, a soil salinity safety threshold and a soil salinity warning threshold were preset, both of which were pre-set by those skilled in the art based on crop salt tolerance characteristics and soil type. The soil salinity safety threshold is the upper limit of salt concentration that does not affect normal crop growth; the soil salinity warning threshold is the critical value of salt concentration that may lead to crop yield reduction or soil degradation. Based on preset salinity weights, a weighted average of all salinity concentrations corresponding to each farmland plot was calculated to obtain the comprehensive salinity concentration. The salinity weight of the topsoil is greater than that of the middle soil, and the salinity weight of the middle soil is greater than that of the topsoil. The salinity weights of deep soil layers are calculated. Each salinity weight is pre-set by a person skilled in the art based on the characteristics of crop root distribution. The difference between the comprehensive salinity concentration and the soil salinity safety threshold is calculated to obtain the amount of salinity exceeding the standard. The difference between the soil salinity warning threshold and the soil salinity safety threshold is calculated to obtain the salinity warning interval. The ratio of the amount of salinity exceeding the standard to the salinity warning interval is calculated to obtain the degree of salinity exceeding the standard. If the degree of salinity exceeding the standard is less than zero, it is set to zero. If the degree of salinity exceeding the standard is greater than one, it is set to one. The difference between one and the degree of salinity exceeding the standard is calculated to obtain the soil salinity health index of the farmland plot. The soil salinity health index is used to reflect the health level of soil salinity in the farmland plot relative to the safety threshold. Obtain all salt monitoring records for all sampling points corresponding to each farmland plot within the historical analysis period; for each farmland plot, count the number of salt monitoring records with an upward trend in salt accumulation within the historical analysis period to obtain the number of increases; count the total number of all salt monitoring records within the historical analysis period to obtain the number of records; calculate the ratio of the number of increases to the number of records to obtain the percentage of salt increase trend for each farmland plot, and use this as the salt accumulation rate.

[0026] Ecological weights were assigned to groundwater level health indicators and soil salinity health indicators, respectively. Based on these ecological weights, the groundwater level health indicators and soil salinity health indicators of the same farmland plot were weighted and summed to obtain the comprehensive ecological health index of each farmland plot. Each ecological weight was pre-set by a person skilled in the art based on the regional ecological characteristics and the main types of ecological risks. Corresponding degradation weights were set for the rate of water level decline and the rate of salinity accumulation, and the rate of water level decline and the rate of salinity accumulation for the same farmland plot were weighted and summed based on the degradation weights to obtain the ecological degradation rate of each farmland plot; wherein, each degradation weight was pre-set by a person skilled in the art based on the regional ecological degradation driving factors.

[0027] Based on the comprehensive ecological health index and the ecological degradation rate, the degree of evolution risk for each farmland plot is calculated. Specifically, a baseline value for the ecological degradation rate is preset, which is pre-set by those skilled in the art based on the regional ecological carrying capacity. The ratio of the ecological degradation rate to the baseline value is calculated to obtain the degradation rate ratio. The difference between the degradation rate ratio and the comprehensive ecological health index is calculated to obtain the ecological vulnerability. Corresponding risk weights are set for the degradation rate ratio and the ecological vulnerability, and the degradation rate ratio and the ecological vulnerability are weighted and summed based on the risk weights to obtain the original risk value. The original risk value is normalized to obtain the degree of evolution risk of the farmland plot. The degree of evolution risk reflects the degree of risk of the farmland plot evolving from its current ecological state to a critical state of ecological degradation. The higher the degree of evolution risk, the greater the ecological degradation risk faced by the farmland plot. Each risk weight is pre-set by those skilled in the art based on the actual situation.

[0028] Methods for generating ecological health degradation curves for each farmland plot include: Based on the rate of ecological degradation and the degree of evolution risk, the ecological degradation coefficient of each farmland plot is calculated. Specifically, the ratio of the rate of ecological degradation to the comprehensive ecological health index is calculated to obtain the relative degradation rate; the sum of the relative degradation rate and the degree of evolution risk is calculated to obtain the risk amplification factor; the risk amplification factor reflects the amplification effect of the degree of evolution risk on the degradation rate. The higher the degree of evolution risk, the larger the risk amplification factor, and the faster the degradation rate; the product of the relative degradation rate and the risk amplification factor is calculated to obtain the basic degradation coefficient; a preset correction set is established, which includes correction coefficients corresponding to different soil types, and is pre-set by those skilled in the art based on the regional ecological characteristics; the corresponding correction coefficients are obtained from the correction set according to the soil type of the farmland plot; the product of the correction coefficient and the basic degradation coefficient is calculated to obtain the ecological degradation coefficient. Using time as the independent variable and ecological health as the dependent variable, and combining the ecological decay coefficient and comprehensive ecological health index of each farmland plot, an exponential decay function is used to describe the change in ecological health over time, resulting in the ecological health decay curve for each farmland plot. The ecological health decay curve describes the expected trajectory of ecological health changes in the farmland plot over a future time period. The expression for the exponential decay function is: In the formula, For a moment Ecological health To achieve a comprehensive ecological health index, This represents the ecological degradation coefficient.

[0029] The constraint quantification module is used to acquire water resource management data and ecological red line constraint parameters, calculate the theoretical optimal irrigation amount for each farmland plot under the constraints of the benefit response model library, determine the feasible irrigation area for each farmland plot based on the theoretical optimal irrigation amount, and identify sensitive plots by combining the ecological health decay curve.

[0030] Methods for obtaining water resource management data and ecological red line constraint parameters include: Water resource management data is obtained from the water resource management system (i.e., a water affairs management platform used for unified allocation and management of regional water resource distribution and use). This data records the regional water supply capacity and allocation constraints, specifically including water source supply information and water use quota information. Water source supply information specifically refers to the water supply guarantee rate, which indicates the reliability of water source supply. Water use quota information includes the regional total water use control index, the agricultural water use quota ratio, and the water use quota for each farmland plot. The regional total water use control index is the upper limit of the total water use for all types of water within the region (i.e., the area where all farmland plots are located). The agricultural water use quota ratio is the proportion of agricultural irrigation water used to control the total water use index. The water use quota for each farmland plot is the upper limit of irrigation water allocated to a single farmland plot based on its area and main crop type. Ecological red line constraint parameters are obtained from the ecological environment management system (i.e., the environmental management platform used to monitor and enforce regional ecological protection policies and red line constraints). These parameters define the ecological protection red lines that agricultural irrigation activities must adhere to, specifically including groundwater extraction constraint parameters and soil salinization control parameters. Groundwater extraction constraint parameters include the minimum groundwater level and the annual exploitable groundwater volume. The minimum groundwater level is the lower limit of the groundwater level required to maintain the stability of the regional ecosystem. The annual exploitable groundwater volume is the total amount of groundwater available for irrigation within a year without causing over-extraction. Soil salinization control parameters include the upper limit of soil salinity concentration and the target for salinity accumulation rate. The upper limit of soil salinity concentration is the salt concentration threshold to prevent irreversible soil salinization. The target for salinity accumulation rate is the maximum allowable annual increase in soil salinity caused by irrigation activities.

[0031] Methods for calculating the theoretical optimal irrigation amount for each farmland plot include: Obtain the benefit response model for each farmland plot from the benefit response model library; for each farmland plot, extract the yield-optimal irrigation amount, benefit-optimal irrigation amount, theoretical maximum yield, theoretical maximum benefit, benefit-critical irrigation amount, and yield response function from the corresponding benefit response model; Based on the optimal irrigation amount for yield and the optimal irrigation amount for economic benefit, the comprehensive optimal irrigation amount for each farmland plot is calculated. Specifically, based on preset comprehensive weights, the optimal irrigation amount for yield and the optimal irrigation amount for economic benefit are weighted and summed to obtain the comprehensive optimal irrigation amount. The comprehensive optimal irrigation amount is used to balance the difference between the yield target and the economic benefit target. Each comprehensive weight is preset by those skilled in the art based on the regional agricultural production target orientation. Based on the theoretical maximum yield, the yield guarantee constraint irrigation amount for each farmland plot is calculated. Specifically, a yield guarantee ratio is preset, which is pre-set by those skilled in the art based on the requirements of food security and agricultural production stability. The product of the theoretical maximum yield and the yield guarantee ratio is calculated to obtain the yield guarantee threshold. The yield guarantee threshold is substituted into the yield response function, and the irrigation amount per unit area corresponding to the function value of the yield response function being equal to the yield guarantee threshold is calculated through numerical solution methods, and this is used as the yield guarantee constraint irrigation amount. Here, the yield guarantee constraint irrigation amount is used to represent the minimum irrigation water required to maintain a basic yield level. It should be noted that if there are multiple irrigation amounts per unit area corresponding to the function value of the yield response function being equal to the yield guarantee threshold, the irrigation amount per unit area with the smallest value is taken as the yield guarantee constraint irrigation amount. Based on water resource management data, calculate the water resource constraint irrigation amount for each farmland plot; specifically, obtain the water use quota for each farmland plot and calculate the ratio of the water use quota to the corresponding plot area to obtain the water resource constraint irrigation amount for the farmland plot; whereby the water resource constraint irrigation amount is used to represent the upper limit of irrigation amount to meet the water resource constraint conditions. Based on the ecological red line constraint parameters, the ecological constraint irrigation volume for each farmland plot is calculated. Specifically, the ratio of the annual exploitable groundwater volume to the total area of ​​the regional plots (i.e., the sum of the areas of all corresponding farmland plots) is calculated to obtain the usable groundwater volume per unit area. The difference between the upper limit of soil salinity concentration and the comprehensive salinity concentration is calculated to obtain the salinity capacity space, which represents the amount of salt accumulation that the soil can still accommodate before reaching the upper limit of salinity concentration. A set of salinity accumulation response coefficients is preset, which includes salinity accumulation response coefficients corresponding to different soil types. Each salinity accumulation response coefficient is determined by a person skilled in the art based on soil salinity. The transport characteristics are pre-set; based on the soil type of the farmland plot, the corresponding salt accumulation response coefficient is obtained from the salt accumulation response coefficient set; the salt accumulation rate control target is compared with the salt capacity space, and the smaller value of the two is taken as the allowable salt accumulation amount; the ratio of the allowable salt accumulation amount to the salt accumulation response coefficient is calculated to obtain the maximum allowable irrigation amount; the available groundwater per unit area is compared with the maximum allowable irrigation amount, and the smaller value of the two is taken as the ecological constraint irrigation amount; whereby the ecological constraint irrigation amount is used to represent the upper limit of irrigation amount under the condition of simultaneously satisfying groundwater extraction constraints and soil salinization control constraints; By comprehensively considering the optimal irrigation amount, yield-guaranteed irrigation amount, water resource-guaranteed irrigation amount, and ecological-guaranteed irrigation amount, the theoretical optimal irrigation amount for each farmland plot is determined. Specifically, the water resource-guaranteed irrigation amount is compared with the ecological-guaranteed irrigation amount, and the smaller of the two is taken as the upper limit of the hard constraint. The yield-guaranteed irrigation amount is compared with the upper limit of the hard constraint. If the yield-guaranteed irrigation amount is greater than the upper limit of the hard constraint, it indicates that there is a conflict between yield guarantee and resource and ecological constraints in the farmland plot. In this case, the upper limit of the hard constraint is taken as the theoretical optimal irrigation amount, and the corresponding farmland plot is marked as a yield-limited area. If the irrigation amount constrained by the production guarantee is less than or equal to the upper limit of the hard constraint, then the comprehensive optimal irrigation amount is compared with the upper limit of the hard constraint; if the comprehensive optimal irrigation amount is greater than the upper limit of the hard constraint, then the upper limit of the hard constraint is taken as the theoretical optimal irrigation amount; if the comprehensive optimal irrigation amount is less than or equal to the upper limit of the hard constraint, then the comprehensive optimal irrigation amount is compared with the irrigation amount constrained by the production guarantee, and the larger of the two is taken as the theoretical optimal irrigation amount; whereby the theoretical optimal irrigation amount is used to represent the optimal irrigation water amount that, under the premise of satisfying water resource constraints and ecological red line constraints, takes into account both production guarantee and benefit objectives as much as possible.

[0032] Methods for determining the feasible irrigation zone for each farmland plot include: Based on the benefit response model, the benefit-constrained irrigation amount for each farmland plot is determined. Specifically, the benefit response function and the theoretical maximum benefit are obtained from the benefit response model; a benefit guarantee ratio is preset, which is pre-set by those skilled in the art according to the economic requirements of agricultural production; the product of the theoretical maximum benefit and the benefit guarantee ratio is calculated to obtain the benefit guarantee threshold; through numerical solution methods, the irrigation amount per unit area corresponding to the function value of the benefit response function being equal to the benefit guarantee threshold is calculated and used as the benefit-constrained irrigation amount; wherein, the benefit-constrained irrigation amount is used to represent the minimum irrigation water required to maintain basic economic benefits; it should be noted that if there are multiple irrigation amounts per unit area corresponding to the function value of the benefit response function being equal to the benefit guarantee threshold, the irrigation amount per unit area with the smallest value is taken as the benefit-constrained irrigation amount; Based on the irrigation amounts constrained by yield guarantee, efficiency guarantee, and theoretical optimal irrigation, the feasible irrigation region for each farmland plot is determined. Specifically, the irrigation amounts constrained by yield guarantee and efficiency guarantee are compared, and the larger value is taken as the lower limit of the feasible irrigation region. The theoretical optimal irrigation amount is taken as the upper limit of the feasible irrigation region. If the lower limit of the feasible irrigation region is greater than the upper limit, it is determined that the feasible irrigation region for the corresponding farmland plot does not exist, and the feasible irrigation region is marked as an empty region. If the lower limit of the feasible irrigation region is less than or equal to the upper limit, the feasible irrigation region for the corresponding farmland plot is represented as a closed interval. The feasible irrigation region is used to represent the range of irrigation values ​​for a farmland plot under the multiple constraints of yield guarantee, efficiency guarantee, and resource ecology.

[0033] Methods for identifying sensitive land parcels include: The planning time span is preset and is set in advance by those skilled in the art based on the regional ecological protection plan. The planning time span is substituted into the ecological health decay curve of each farmland plot to calculate the ecological health of each farmland plot and mark it as the predicted health. The predicted health is used to reflect the expected level of ecological health of the corresponding farmland plot at the end of the planning time span. Based on the feasible irrigation area of ​​each farmland plot, the feasible area width of each farmland plot is calculated; specifically, the difference between the upper limit and the lower limit of the feasible irrigation area is calculated to obtain the feasible area width; if the feasible irrigation area is empty, the feasible area width is set to zero. The benefit-ecological conflict degree of each farmland plot is calculated. Specifically, the difference between the comprehensive optimal irrigation amount and the theoretical optimal irrigation amount is calculated to obtain the irrigation amount constraint. If the irrigation amount constraint is less than or equal to zero, it indicates that the comprehensive optimal irrigation amount is not restricted by resource and ecological constraints, and the benefit-ecological conflict degree is set to zero. If the irrigation amount constraint is greater than zero, it indicates that the comprehensive optimal irrigation amount is restricted by resource and ecological constraints, and the ratio of the irrigation amount constraint to the comprehensive optimal irrigation amount is calculated to obtain the benefit-ecological conflict degree. The benefit-ecological conflict degree is used to reflect the degree of contradiction between the economic benefit objectives and the ecological protection objectives of the farmland plot. Based on the feasible domain width, the degree of benefit-ecological conflict, predicted health, and yield-limited plot marking, it is determined whether each farmland plot is a sensitive plot. Specifically, if the feasible domain width is less than a preset feasible domain width threshold, the corresponding farmland plot is marked as a sensitive plot; if the degree of benefit-ecological conflict is greater than a preset conflict degree threshold, the corresponding farmland plot is marked as a sensitive plot; if the predicted health is less than a preset ecological health warning threshold, the corresponding farmland plot is marked as a sensitive plot; if the farmland plot is marked as a yield-limited plot, the corresponding farmland plot is marked as a sensitive plot; if none of the above conditions are met, the corresponding farmland plot is determined not to be a sensitive plot. Among them, sensitive plots refer to farmland plots where irrigation volume regulation space is limited, there is a significant conflict between benefit and ecological goals, ecological health faces the risk of decline, or yield guarantee is constrained and limited, and fine-grained balance optimization is required.

[0034] The balance optimization module is used to analyze the long-term net present value of benefits and the estimated ecological restoration cycle of each sensitive plot under different irrigation intensity schemes, based on the feasible irrigation volume domain and ecological health decay curve of each sensitive plot, and intelligently identify the optimal benefit-ecological balance point.

[0035] The methods for analyzing the long-term net present value of benefits and the estimated ecological restoration cycle of various sensitive plots under different irrigation intensity schemes include: Based on the feasible irrigation volume domain of each sensitive plot, multiple irrigation intensity schemes are generated. Specifically, a scheme generation step size is preset, which is pre-set by those skilled in the art according to the irrigation control accuracy requirements. For each sensitive plot, if the feasible irrigation volume domain is empty, the theoretically optimal irrigation volume is used as the only irrigation intensity scheme. If the feasible irrigation volume domain is not empty, the lower limit and upper limit of the feasible irrigation volume domain are obtained. Starting from the lower limit of the feasible irrigation volume domain, multiple irrigation intensity values ​​are generated sequentially at intervals of the scheme generation step size until the upper limit of the feasible irrigation volume domain is reached. Each irrigation intensity value is used as an irrigation intensity scheme to obtain the irrigation intensity scheme for each sensitive plot. For each irrigation intensity scheme for each sensitive plot, the corresponding long-term benefit net present value is calculated. Specifically, a benefit assessment period and annual discount rate are preset, both of which are pre-set by those skilled in the art based on agricultural production planning and economic evaluation standards. The benefit assessment period is the time span for assessing long-term benefits; the annual discount rate is used to convert the benefits of future years into present value at the current point in time. The irrigation intensity value corresponding to the irrigation intensity scheme is substituted into the benefit response function to calculate the corresponding benefit per unit area, which is then used as the expected annual benefit. The water supply guarantee rate is obtained from water resource management data. The product of the expected annual benefit and the water supply guarantee rate is calculated to obtain the risk-adjusted expected annual benefit, which is then marked as the actual annual benefit. For each year within the benefit assessment period, the corresponding discount factor is calculated. The expression for the discount factor is: In the formula, For the first The discount factor for the year The annual discount rate is used to calculate the discounted benefit for each year by multiplying the actual annual benefit by the corresponding discount factor for each year. The discounted benefits for all years within the benefit assessment period are summed to obtain the net present value of long-term benefits. The net present value of long-term benefits is used to reflect the current value of the cumulative economic benefits that can be obtained by adopting the corresponding irrigation intensity scheme within the benefit assessment period.

[0036] For each irrigation intensity scheme of each sensitive plot, the corresponding estimated ecological restoration cycle is calculated. Specifically, if the feasible irrigation area is an empty area, the ecological decay coefficient is used as the ecological adjustment coefficient; if the feasible irrigation area is not an empty area, the ecological adjustment coefficient is calculated based on the irrigation intensity scheme. The expression for the ecological adjustment coefficient is: In the formula, This is the ecological adjustment coefficient. This is the ecological degradation coefficient. The irrigation impact coefficient, This represents the irrigation intensity value. This represents the lower limit of the feasible irrigation volume range. The upper limit of the feasible irrigation volume is defined as follows: The irrigation impact coefficient reflects the degree of influence of irrigation volume changes on the rate of ecological degradation and is pre-set by those skilled in the art based on the regional ecological sensitivity; the minimum groundwater level is obtained from the ecological redline constraint parameters; the difference between the minimum groundwater level and the groundwater level safety threshold is calculated and then divided by the water level warning interval to obtain the critical deviation of the water level; the difference between this and the critical deviation of the water level is calculated to obtain the lower limit of ecological health; the lower limit of ecological health represents the minimum ecological health level required to maintain the stability of the regional ecosystem. Substituting the ecological adjustment coefficient and the comprehensive ecological health index into the ecological health decay curve, and using numerical methods, the time corresponding to when the function value of the ecological health decay curve equals the lower limit threshold of ecological health is calculated and used as the ecological degradation cycle. A target ecological health level and ecological restoration rate are preset, both of which are pre-set by those skilled in the art based on regional ecological protection requirements and the natural recovery capacity of the ecosystem. If the comprehensive ecological health index is less than the target ecological health level, an ecological restoration function is constructed. The expression of the ecological restoration function is: In the formula, For the recovery process at all times Ecological health To achieve the goal of ecological restoration and health, The ecological restoration rate is defined as follows: Those skilled in the art pre-set the restoration completion ratio based on actual conditions, calculate the product of the ecological restoration target health level and the restoration completion ratio, and obtain the restoration target level. The function value of the ecological restoration function is set equal to the restoration target level, and the time required for the ecological health level to recover to the restoration target level is calculated using numerical methods, and this time is used as the ecological restoration cycle estimate. If the current comprehensive ecological health index is greater than or equal to the ecological restoration target health level, the ecological restoration cycle estimate is set to zero.

[0037] Methods for intelligently identifying the optimal benefit-ecological balance point include: The ecological sustainability assessment criteria are preset, including a lower limit for the ecological degradation cycle and a lower limit for benefit guarantee. The lower limit for the ecological degradation cycle is the minimum ecological degradation cycle required to maintain ecological sustainability. The lower limit for benefit guarantee is the minimum long-term net present value of benefits required to maintain the sustainability of agricultural production. Both the lower limit for the ecological degradation cycle and the lower limit for benefit guarantee are preset by those skilled in the art based on the regional sustainable development goals. From the irrigation intensity schemes for each sensitive plot, irrigation intensity schemes that meet the ecological sustainability criteria are selected. Specifically, if the ecological degradation cycle is greater than or equal to the lower limit of the ecological degradation cycle, and the net present value of long-term benefits is greater than or equal to the lower limit of benefit guarantee, then the corresponding irrigation intensity scheme is determined to meet the ecological sustainability criteria. For each sensitive plot, the maximum value of the irrigation intensity value is obtained from all irrigation intensity schemes that meet the ecological sustainability criteria and used as the sustainable irrigation intensity threshold. If no irrigation intensity scheme meets the ecological sustainability criteria, the corresponding sensitive plot is marked as an ecological priority plot, and it is determined whether the corresponding feasible irrigation volume region is empty. If the feasible irrigation volume region is empty, the theoretically optimal irrigation volume is used as the sustainable irrigation intensity threshold for the corresponding sensitive plot. If the feasible irrigation volume region is not empty, the lower limit of the feasible irrigation volume region is used as the sustainable irrigation intensity threshold for the corresponding sensitive plot. The net present value of each long-term benefit and the estimated value of the ecological restoration cycle are normalized to obtain the standard net present value of benefits and the estimated value of the standard restoration cycle. Corresponding scoring weights are set for the standard net present value of benefits and the estimated value of the standard restoration cycle. Each scoring weight is pre-set by a person skilled in the art based on the priority of regional benefits and ecological protection. Based on the scoring weights, the standard net present value of benefits and the estimated standard recovery period are calculated by weighted difference for the same irrigation intensity scheme for the same sensitive plot, resulting in a comprehensive evaluation score for each irrigation intensity scheme for each sensitive plot. The comprehensive evaluation scores of all irrigation intensity schemes for the same sensitive plot are compared, and the irrigation intensity value corresponding to the irrigation intensity scheme with the highest comprehensive evaluation score is taken as the optimal benefit-ecological balance point for the corresponding sensitive plot. The optimal benefit-ecological balance point is used to represent the best irrigation intensity that balances economic benefits and ecological protection. From water resource management data, obtain the regional total water consumption control index and the agricultural water quota ratio; calculate the product of the regional total water consumption control index and the agricultural water quota ratio to obtain the total available agricultural water resources; for all sensitive plots, perform regional total water consumption verification to determine whether the optimal benefit-ecological balance point needs to be updated; specifically, for farmland plots that are not sensitive, calculate the product of the corresponding plot area and the theoretical optimal irrigation amount to obtain the non-sensitive expected water consumption; calculate the product of the plot area and the optimal benefit-ecological balance point for sensitive plots to obtain the sensitive expected water consumption; sum all sensitive and non-sensitive expected water consumption separately to obtain the total expected water consumption for sensitive plots and the total expected water consumption for non-sensitive plots; calculate the difference between the total available agricultural water resources and the total expected water consumption for non-sensitive plots to obtain the total available water for sensitive plots; and then apply the expected water consumption of sensitive plots to the calculation of the total available agricultural water resources. The total water volume is compared with the total available water volume of sensitive plots. If the expected total water consumption of a sensitive plot is greater than the total available water volume, the ratio of the total available water volume to the expected total water consumption is calculated to obtain a water consumption compression coefficient. The optimal benefit-ecological balance point is multiplied by the water consumption compression coefficient to obtain the compressed benefit-ecological balance point. The compressed benefit-ecological balance point is compared with the corresponding sustainable irrigation intensity threshold. If the compressed benefit-ecological balance point is greater than the sustainable irrigation intensity threshold, the optimal benefit-ecological balance point is updated according to the sustainable irrigation intensity threshold. If the compressed benefit-ecological balance point is less than or equal to the sustainable irrigation intensity threshold, the optimal benefit-ecological balance point is updated according to the compressed benefit-ecological balance point. If the expected total water consumption of a sensitive plot is less than or equal to the total available water volume of the sensitive plot, the optimal benefit-ecological balance point is not updated.

[0038] The sustainable regulation module is used to generate an eco-friendly irrigation regulation scheme based on the optimal benefit-ecological balance point of each sensitive plot, and to intelligently schedule the irrigation water volume of each sensitive plot according to the eco-friendly irrigation regulation scheme.

[0039] Methods for generating eco-friendly irrigation regulation schemes include: Based on the geographic coordinates of each sensitive plot, a neighborhood association map of sensitive plots is constructed. Specifically, for each sensitive plot, its corresponding geographic coordinates are obtained; a neighborhood association radius is preset, which is determined by those skilled in the art based on the regional groundwater hydrological connectivity and soil salinity migration characteristics; the spatial distance between any two sensitive plots is calculated based on their geographic coordinates; if the spatial distance is less than or equal to the neighborhood association radius, the two sensitive plots are determined to have a neighborhood association relationship; if the spatial distance is greater than the neighborhood association radius, the two sensitive plots are determined not to have a neighborhood association relationship; sensitive plots with neighborhood association relationships are connected to form a neighborhood association map of sensitive plots; the neighborhood association map of sensitive plots is used to describe the spatial proximity relationships and potential ecological impact transmission paths between sensitive plots. Based on the neighborhood association map of sensitive land parcels and the ecological degradation rate of each sensitive land parcel, the neighborhood ecological coupling factor of each sensitive land parcel is calculated. Specifically, for each sensitive land parcel, all sensitive land parcels with neighborhood association relationships with it are obtained from the neighborhood association map and marked as neighborhood parcels. If no neighborhood parcels exist, the neighborhood ecological coupling factor of the corresponding sensitive land parcel is set to one. If the set of neighborhood parcels is not empty, the ecological degradation rate of each neighborhood parcel corresponding to the corresponding sensitive land parcel is obtained and the average value is calculated to obtain the result. The average degradation rate of the neighborhood is calculated; the ratio of the ecological degradation rate of the corresponding sensitive plot to the average degradation rate of the neighborhood is obtained; a neighborhood coupling strength coefficient is preset, which is pre-set by those skilled in the art based on the connectivity of the regional ecosystem; the product of the degradation rate ratio of the corresponding sensitive plot and the neighborhood coupling strength coefficient is calculated and then added to one to obtain the neighborhood ecological coupling factor; wherein, the neighborhood ecological coupling factor is used to reflect the degree of constraint and influence of the ecological state of the neighborhood plot on the current irrigation decision of the sensitive plot; The crop growth stages of each sensitive plot are obtained from the agricultural production management system, and the growth stage sensitivity coefficients of each sensitive plot are calculated. Specifically, a water demand sensitivity coefficient set is preset, which includes water demand sensitivity coefficients corresponding to different main crop types at different crop growth stages. Each water demand sensitivity coefficient is preset by a person skilled in the art based on the physiological water demand patterns of crops. The crop growth stages include, but are not limited to, seedling stage, growth stage, flowering stage, fruiting stage, and maturity stage. Based on the main crop type and crop growth stage of the sensitive plot, the corresponding water demand sensitivity coefficients are obtained from the water demand sensitivity coefficient set and used as growth stage sensitivity coefficients. The growth stage sensitivity coefficients are used to reflect the sensitivity of the crop growth stage to changes in irrigation water volume. Meteorological forecast data within a preset weather forecast time window is obtained from the meteorological information service system (i.e., a meteorological data service platform used to collect and publish regional meteorological monitoring data and forecast information), and the meteorological response adjustment coefficient for each sensitive plot is calculated. Specifically, the preset weather forecast time window is set by those skilled in the art based on the irrigation control cycle; the weather forecast data includes the cumulative precipitation forecast value and the daily average evaporation forecast value; the effective precipitation replenishment is obtained by multiplying the cumulative precipitation forecast value by the preset precipitation infiltration coefficient; the precipitation infiltration coefficient is set by those skilled in the art based on the regional soil permeability characteristics; the cumulative evaporation loss is obtained by multiplying the daily average evaporation forecast value by the number of days corresponding to the weather forecast time window; the effective precipitation replenishment is then calculated by multiplying the cumulative evaporation loss by the product of the daily average evaporation forecast value and the number of days corresponding to the weather forecast time window; and the effective precipitation replenishment is then calculated by multiplying the cumulative evaporation loss by the product of the daily average evaporation forecast value and the number of days corresponding to the weather forecast time window. The difference between the water loss and the expected water balance is obtained; a preset water balance benchmark value is established by those skilled in the art based on the regional water balance characteristics; the ratio of the expected water balance to the benchmark value is calculated to obtain the water balance deviation; the difference between the expected water balance and the water balance deviation is calculated to obtain the meteorological response adjustment coefficient; if the meteorological response adjustment coefficient is less than the preset lower limit, it is set as the lower limit; if it is greater than the preset upper limit, it is set as the upper limit; both the lower and upper limits are preset by those skilled in the art based on the requirements for irrigation regulation stability; the meteorological response adjustment coefficient is used to reflect the pre-emptive adjustment effect of future meteorological conditions on current irrigation demand; Based on the optimal benefit-ecological balance point, neighborhood ecological coupling factor, growth stage sensitivity coefficient, and meteorological response regulation coefficient of each sensitive plot, the comprehensive adjusted irrigation amount for each sensitive plot is calculated. Specifically, the ratio of the optimal benefit-ecological balance point to the neighborhood ecological coupling factor is calculated to obtain the neighborhood coupling adjustment amount; the product of the neighborhood coupling adjustment amount and the growth stage sensitivity coefficient is calculated to obtain the growth stage adjustment amount; the product of the growth stage adjustment amount and the meteorological response regulation coefficient is calculated to obtain the comprehensive adjusted irrigation amount; the comprehensive adjusted irrigation amount is compared with the sustainable irrigation intensity threshold of the corresponding sensitive plot; if the comprehensive adjusted irrigation amount is greater than the sustainable irrigation intensity threshold, the sustainable irrigation intensity threshold is used as the comprehensive adjusted irrigation amount; if the comprehensive adjusted irrigation amount is less than or equal to the sustainable irrigation intensity threshold, the comprehensive adjusted irrigation amount remains unchanged; where the comprehensive adjusted irrigation amount is used to represent the irrigation water volume after comprehensively considering the neighborhood ecological coupling effect, crop growth stage requirements, and meteorological condition response. Based on the comprehensive ecological health index and ecological degradation rate of each sensitive plot, an ecological urgency index is calculated for each sensitive plot. Specifically, the difference between the comprehensive ecological health index and the ecological health index is calculated to obtain the ecological health gap; the ratio of the ecological degradation rate to a preset degradation rate reference value is calculated to obtain the relative degradation intensity; wherein, the degradation rate reference value is preset by those skilled in the art based on historical data of regional ecological degradation; corresponding urgency weights are set for the ecological health gap and the relative degradation intensity, and the ecological health gap and the relative degradation intensity are weighted and summed based on the urgency weights to obtain the ecological urgency index of the sensitive plot; wherein, each urgency weight is preset by those skilled in the art based on the ecological protection priority strategy; the ecological urgency index is used to reflect the urgency of the ecological status of the sensitive plot. The higher the ecological urgency index, the more the corresponding sensitive plot needs priority for ecological protection irrigation regulation. Based on the ecological urgency index and the comprehensive adjustment of irrigation volume for each sensitive plot, the irrigation regulation priority of each sensitive plot is determined. Specifically, all sensitive plots are sorted from high to low according to their corresponding ecological urgency index to obtain an ecological priority ranking sequence. Irrigation regulation priorities are assigned to each sensitive plot in ascending order of the ecological priority ranking sequence. Among them, the sensitive plot with the highest ecological urgency index has the highest irrigation regulation priority. Based on the irrigation control priority and comprehensive adjustment irrigation volume of each sensitive plot, cross-plot ecological compensation regulation is implemented to obtain the comprehensive compensation irrigation volume for each sensitive plot. Specifically, the difference between the optimal benefit-ecological balance point and the corresponding comprehensive adjustment irrigation volume for each sensitive plot is calculated to obtain the irrigation adjustment surplus. If the irrigation adjustment surplus is greater than zero, it indicates that the corresponding sensitive plot has an irrigation quota surplus, and the corresponding sensitive plot is marked as a quota surplus plot, and the comprehensive adjustment irrigation volume of each quota surplus plot is used as the corresponding comprehensive compensation irrigation volume. If the irrigation adjustment surplus is equal to zero, the comprehensive adjustment irrigation volume is used as the comprehensive compensation irrigation volume for the corresponding sensitive plot. If the irrigation adjustment surplus is less than zero, it indicates that the corresponding sensitive plot needs additional irrigation quota support, and the corresponding sensitive plot is marked as a quota shortage plot. The irrigation adjustment surplus of all quota-surplus plots is summed to obtain the total regional adjustable quota. Quota compensation is then applied to each quota-scarce plot in descending order of irrigation control priority. The specific process of quota compensation is as follows: the smaller value between the absolute value of the irrigation adjustment surplus corresponding to the quota-scarce plot and the total regional adjustable quota is selected as the actual compensation quota; the actual compensation quota is added to the comprehensive adjustment irrigation volume of the corresponding quota-scarce plot to obtain the comprehensive compensation irrigation volume; the total regional adjustable quota is subtracted from the actual compensation quota to update the total regional adjustable quota; if the total regional adjustable quota is less than or equal to zero, quota compensation is stopped, and the comprehensive adjustment irrigation volume of the remaining quota-scarce plots that have not received quota compensation is used as the corresponding comprehensive compensation irrigation volume. The comprehensive compensation irrigation volume, irrigation regulation priority, and unique code of each sensitive plot are integrated to form irrigation regulation parameters for each sensitive plot. The irrigation regulation parameters of all sensitive plots are summarized to generate an eco-friendly irrigation regulation scheme. The eco-friendly irrigation regulation scheme is used to guide the allocation and scheduling of irrigation water for each sensitive plot while taking into account both ecological protection and benefit objectives. According to the irrigation regulation priority of each sensitive plot, the irrigation water volume of each sensitive plot is intelligently scheduled in sequence according to the corresponding comprehensive compensation irrigation volume.

[0040] This embodiment establishes a response relationship model between irrigation water volume and yield output by collecting historical irrigation records and crop yield data for each farmland plot. This achieves the goal of optimizing irrigation decisions from a long-term economic perspective, effectively overcoming the shortcomings of existing technologies that only focus on short-term water demand satisfaction while neglecting the input-output benefits of irrigation. By collecting real-time groundwater level monitoring data and soil salinity accumulation data, an ecological health decay curve is constructed, establishing a continuous monitoring and quantitative assessment mechanism for the ecological health status of farmland. This effectively warns of ecological degradation risks such as groundwater level decline and soil salinity accumulation that may result from over-irrigation. By integrating water resource quota constraints and ecological red line constraints, the feasible irrigation volume and theoretically optimal irrigation volume for each farmland plot are determined, achieving irrigation decision optimization under multiple resource and ecological constraints. By analyzing the long-term net present value of benefits and the estimated ecological restoration cycle under different irrigation intensity schemes, the optimal benefit-ecological balance point is intelligently identified, achieving a synergistic balance between irrigation economic benefits and ecological protection goals. By constructing a neighborhood association map of sensitive plots, calculating neighborhood ecological coupling factors, and combining crop growth stage sensitivity and meteorological response regulation mechanisms, refined irrigation control considering spatial ecological correlations and time-varying environmental factors is achieved. Through cross-plot ecological compensation regulation and dynamic redistribution of irrigation quotas according to ecological urgency priorities, an eco-friendly irrigation control scheme is generated. Ultimately, intelligent irrigation water scheduling is achieved while balancing agricultural production benefits and regional ecological sustainability, effectively preventing the gradual degradation of agricultural ecosystems and providing technical support for the sustainable development of smart agriculture. Example 2:

[0041] Please see Figure 2 As shown, the parts not described in detail in this embodiment are described in Embodiment 1. A multi-source data fusion irrigation control method for smart agriculture is provided, the method including: Collect historical irrigation records and corresponding crop yield data for each farmland plot, perform regression modeling on the response relationship between different irrigation water volumes and crop yield data in the historical irrigation records, and construct a benefit response model library for each farmland plot. Real-time data collection of groundwater level monitoring and soil salinity accumulation around farmland is used to assess the risk of each farmland plot evolving from its current ecological state to a critical state of ecological degradation, and to generate ecological health decay curves for each farmland plot. Obtain water resource management data and ecological red line constraint parameters, calculate the theoretical optimal irrigation amount for each farmland plot under the constraint of the benefit response model library, determine the feasible irrigation area for each farmland plot based on the theoretical optimal irrigation amount, and identify sensitive plots by combining the ecological health decay curve; Based on the feasible irrigation volume domain and ecological health decay curve of each sensitive plot, the long-term net present value of benefits and the estimated ecological restoration cycle of each sensitive plot under different irrigation intensity schemes are analyzed, and the optimal benefit-ecological balance point is intelligently identified. Based on the optimal benefit-ecological balance point of each sensitive plot, an eco-friendly irrigation regulation scheme is generated, and the irrigation water volume of each sensitive plot is intelligently scheduled according to the eco-friendly irrigation regulation scheme. Example 3:

[0042] This application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable code, which, when executed by the one or more processors, can perform the multi-source data fusion irrigation control method for smart agriculture as described above.

[0043] The method or system according to the embodiments of this application can also be implemented using the architecture of the electronic device shown in this application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output, a hard disk, etc. The storage device in the electronic device, such as a ROM or hard disk, may store the multi-source data fusion irrigation control method for smart agriculture provided in this application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in this application is merely exemplary; when implementing different devices, one or more components of the electronic device shown in this application may be omitted according to actual needs. Example 4:

[0044] One embodiment of this application discloses a computer-readable storage medium. The computer-readable storage medium stores computer-readable instructions. When the computer-readable instructions are executed by a processor, the multi-source data fusion irrigation control method for smart agriculture according to the embodiment of this application, as described with reference to the above figures, can be executed. The storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0045] Furthermore, according to embodiments of this application, the processes described in the above-referenced flowcharts can be implemented as computer software programs. For example, this application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be executed by a processor to perform instructions corresponding to the method steps provided in this application, such as a multi-source data fusion irrigation control method for smart agriculture. When this computer program is executed by a central processing unit (CPU), it performs the functions defined in the method of this application.

[0046] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0047] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0048] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for irrigation regulation based on multi-source data fusion for smart agriculture, characterized in that, The method for constructing the benefit response model library of each farmland plot comprises the following steps: According to the historical irrigation records and crop yield data, irrigation-yield paired data of each farmland plot is constructed, which includes unit area irrigation amount, unit area yield and unit area benefit; for the unit area irrigation amount and unit area yield in each irrigation-yield paired data, data quality screening operation is performed to obtain irrigation-yield screening data; For the irrigation-yield screening data of each farmland plot, yield response relationship modeling and benefit response relationship modeling are respectively performed to obtain the yield response function and the benefit response function of each farmland plot; according to the yield response functions, the corresponding optimal irrigation amount for yield and the theoretical maximum yield are calculated; according to the benefit response functions, the corresponding optimal irrigation amount for benefit and the theoretical maximum benefit are calculated; the first derivative of the benefit response function is obtained to obtain the marginal benefit function; according to the marginal benefit function, the critical irrigation amount for benefit is calculated; The yield response function, the benefit response function, the optimal irrigation amount for yield, the optimal irrigation amount for benefit, the theoretical maximum yield, the theoretical maximum benefit and the critical irrigation amount for benefit of each farmland plot are integrated to form the benefit response model of each farmland plot; all the benefit response models of the farmland plots are summarized to construct the benefit response model library. The method for evaluating the evolution risk degree of each farmland plot from the current ecological state to the critical state of ecological degradation comprises the following steps: The groundwater level monitoring data comprises monitoring point basic information and multiple water level monitoring records, and the soil salt accumulation data comprises sampling point basic information and multiple salt monitoring records; the associated plot code in each water level monitoring record and salt monitoring record is obtained, and all the water level monitoring records and salt monitoring records are associated with each farmland plot according to the associated plot code; the water level monitoring records and the salt monitoring records of the same farmland plot are integrated to form an ecological monitoring original data set of each farmland plot; 2.The smart agriculture oriented multi-source data fusion irrigation regulation method according to claim 1, wherein, The method for constructing the benefit response model library of each farmland plot comprises the following steps: According to the historical irrigation records and crop yield data, irrigation-yield paired data of each farmland plot is constructed, which includes unit area irrigation amount, unit area yield and unit area benefit; for the unit area irrigation amount and unit area yield in each irrigation-yield paired data, data quality screening operation is performed to obtain irrigation-yield screening data; For the irrigation-yield screening data of each farmland plot, yield response relationship modeling and benefit response relationship modeling are respectively performed to obtain the yield response function and the benefit response function of each farmland plot; according to the yield response functions, the corresponding optimal irrigation amount for yield and the theoretical maximum yield are calculated; according to the benefit response functions, the corresponding optimal irrigation amount for benefit and the theoretical maximum benefit are calculated; the first derivative of the benefit response function is obtained to obtain the marginal benefit function; according to the marginal benefit function, the critical irrigation amount for benefit is calculated; The yield response function, the benefit response function, the optimal irrigation amount for yield, the optimal irrigation amount for benefit, the theoretical maximum yield, the theoretical maximum benefit and the critical irrigation amount for benefit of each farmland plot are integrated to form the benefit response model of each farmland plot; all the benefit response models of the farmland plots are summarized to construct the benefit response model library. 3.The smart agriculture oriented multi-source data fusion irrigation regulation method according to claim 2, characterized in that, The method for evaluating the evolution risk degree of each farmland plot from the current ecological state to the critical state of ecological degradation comprises the following steps: The groundwater level monitoring data comprises monitoring point basic information and multiple water level monitoring records, and the soil salt accumulation data comprises sampling point basic information and multiple salt monitoring records; the associated plot code in each water level monitoring record and salt monitoring record is obtained, and all the water level monitoring records and salt monitoring records are associated with each farmland plot according to the associated plot code; the water level monitoring records and the salt monitoring records of the same farmland plot are integrated to form an ecological monitoring original data set of each farmland plot; According to the ecological monitoring original data set, the groundwater level health index and the soil salt health index of each farmland plot are calculated; in a preset historical analysis period, all water level monitoring records and salt content monitoring records of each farmland plot are obtained, and the water level change rate and the salt accumulation trend of each farmland plot are obtained from the water level monitoring records and the salt content monitoring records respectively; based on the water level change rate and the salt accumulation trend, the water level decline rate and the salt accumulation rate of each farmland plot are calculated respectively; Based on the preset ecological weight, the groundwater level health index and the soil salt health index of the same farmland plot are weighted and summed to obtain the comprehensive ecological health index of each farmland plot; based on the preset degradation weight, the water level decline rate and the salt accumulation rate of the same farmland plot are weighted and summed to obtain the ecological degradation rate of each farmland plot; according to the comprehensive ecological health index and the ecological degradation rate of each farmland plot, the evolution risk degree of each farmland plot is calculated.

4. The smart agriculture oriented multi-source data fusion irrigation regulation method according to claim 3, characterized in that, The method for generating the ecological health degree attenuation curve of each farmland plot comprises: According to the ecological degradation rate and the evolution risk degree, the ecological attenuation coefficient of each farmland plot is calculated; taking time as the independent variable and ecological health degree as the dependent variable, and combining the ecological attenuation coefficient and the comprehensive ecological health index of each farmland plot, the change law of the ecological health degree with time is described by using an exponential attenuation function to obtain the ecological health degree attenuation curve of each farmland plot.

5. The smart agriculture oriented multi-source data fusion irrigation regulation method according to claim 4, characterized in that, The method for calculating the theoretical optimal irrigation amount of the farmland plot comprises: According to the yield optimal irrigation amount and the benefit optimal irrigation amount, the comprehensive optimal irrigation amount is calculated; according to the theoretical maximum yield, the yield guarantee constraint irrigation amount is calculated; according to the water resource management data, the water resource constraint irrigation amount is calculated; according to the ecological red line constraint parameter, the ecological constraint irrigation amount is calculated; The water resource constraint irrigation amount and the ecological constraint irrigation amount are compared, and the smaller value of the two is taken as the hard constraint upper limit; the yield guarantee constraint irrigation amount and the hard constraint upper limit are compared; if the yield guarantee constraint irrigation amount is greater than the hard constraint upper limit, the hard constraint upper limit is taken as the theoretical optimal irrigation amount, and the corresponding farmland plot is marked as a yield limited plot; if the yield guarantee constraint irrigation amount is less than or equal to the hard constraint upper limit, the comprehensive optimal irrigation amount and the hard constraint upper limit are compared; If the comprehensive optimal irrigation amount is greater than the hard constraint upper limit, the hard constraint upper limit is taken as the theoretical optimal irrigation amount; if the comprehensive optimal irrigation amount is less than or equal to the hard constraint upper limit, the comprehensive optimal irrigation amount and the yield guarantee constraint irrigation amount are compared, and the larger value of the two is taken as the theoretical optimal irrigation amount. 6.The smart agriculture oriented multi-source data fusion irrigation regulation method according to claim 5, characterized in that, The method for determining the irrigation amount feasible region of the farmland plot comprises: According to the benefit response model, the benefit constraint irrigation amount is determined; the yield guarantee constraint irrigation amount and the benefit constraint irrigation amount are compared, and the larger value of the two is taken as the irrigation amount feasible region lower limit; the theoretical optimal irrigation amount is taken as the irrigation amount feasible region upper limit; If the lower limit of the irrigation amount feasible region is greater than the upper limit of the irrigation amount feasible region, the irrigation amount feasible region corresponding to the farmland plot is marked as an empty region; if the lower limit of the irrigation amount feasible region is less than or equal to the upper limit of the irrigation amount feasible region, the irrigation amount feasible region corresponding to the farmland plot is expressed in the form of a closed interval; The method for determining the sensitive plot comprises: presetting a planning time span, substituting the planning time span into the ecological health degree attenuation curve of each farmland plot respectively, calculating the ecological health degree of each farmland plot, and marking it as a predicted health degree; calculating the feasible region width of each farmland plot according to the irrigation amount feasible region of each farmland plot; calculating the benefit-ecological conflict degree of each farmland plot according to the comprehensive optimal irrigation amount and the theoretical optimal irrigation amount; if the feasible region width is less than a preset feasible region width threshold, the corresponding farmland plot is marked as a sensitive plot; if the benefit-ecological conflict degree is greater than a preset conflict degree threshold, the corresponding farmland plot is marked as a sensitive plot; if the predicted health degree is less than a preset ecological health degree early warning threshold, the corresponding farmland plot is marked as a sensitive plot; if the farmland plot is marked as a yield limited plot, the corresponding farmland plot is marked as a sensitive plot.

7. The smart agriculture oriented multi-source data fusion irrigation regulation method according to claim 6, characterized in that, The method for analyzing the long-term benefit net present value and the ecological recovery cycle estimate value of the sensitive plot under different irrigation intensity schemes comprises: generating a plurality of irrigation intensity schemes according to the irrigation amount feasible region of the sensitive plot; presetting a benefit evaluation period and an annual discount rate; substituting the irrigation intensity scheme into the benefit response function to calculate the annual expected benefit; obtaining the water supply guarantee rate from the water resource management data; calculating the annual actual benefit according to the annual expected benefit and the water supply guarantee rate; calculating the discount coefficient corresponding to each year within the benefit evaluation period; calculating the discounted benefit of each year according to the annual actual benefit and the discount coefficient corresponding to each year; calculating the long-term benefit net present value according to the discounted benefits of all years within the benefit evaluation period; if the irrigation amount feasible region is an empty region, the ecological attenuation coefficient is used as the ecological adjustment coefficient; if the irrigation amount feasible region is not an empty region, the ecological adjustment coefficient is calculated according to the irrigation intensity scheme; obtaining the minimum value of underground water level from the ecological red line constraint parameter; calculating the lower limit threshold of ecological health degree according to the minimum value of underground water level; substituting the ecological adjustment coefficient, the comprehensive ecological health index and the lower limit threshold of ecological health degree into the ecological health degree attenuation curve to calculate the ecological degradation cycle; presetting the ecological recovery target health degree and the recovery completion ratio to calculate the recovery target level; if the comprehensive ecological health index is less than the ecological recovery target health degree, the ecological recovery function is constructed; setting the function value of the ecological recovery function to be equal to the recovery target level to calculate the ecological recovery cycle estimate value; if the current comprehensive ecological health index is greater than or equal to the ecological recovery target health degree, the ecological recovery cycle estimate value is set to zero. 8.The smart agriculture oriented multi-source data fusion irrigation regulation method according to claim 7, characterized in that, The method for intelligently identifying the optimal benefit-ecological balance point comprises: The preset ecological sustainability determination condition is obtained; for each sensitive land plot, the maximum irrigation intensity scheme satisfying the ecological sustainability determination condition is selected as a sustainable irrigation intensity threshold from all irrigation intensity schemes satisfying the ecological sustainability determination condition; if there is no irrigation intensity scheme satisfying the ecological sustainability determination condition, the corresponding sensitive land plot is marked as an ecological priority land plot, and it is determined whether an irrigation amount feasible region is empty; if the irrigation amount feasible region is empty, a theoretical optimal irrigation amount is taken as the sustainable irrigation intensity threshold; if the irrigation amount feasible region is not empty, a lower limit of the irrigation amount feasible region is taken as the sustainable irrigation intensity threshold; The long-term net present value and the ecological restoration cycle estimation value of each sensitive land plot are normalized respectively to obtain a standard net present value and a standard restoration cycle estimation value; the standard net present value and the standard restoration cycle estimation value of the same irrigation intensity scheme corresponding to the same sensitive land plot are weighted and subtracted based on a scoring weight to obtain a comprehensive evaluation score of each sensitive land plot corresponding to each irrigation intensity scheme; the irrigation intensity scheme with the highest comprehensive evaluation score is taken as an optimal benefit-ecological balance point of the corresponding sensitive land plot; the total regional water consumption is checked for all sensitive land plots, it is determined whether the optimal benefit-ecological balance point needs to be updated, and the optimal benefit-ecological balance point needing to be updated is updated according to the sustainable irrigation intensity threshold. 9.The smart agriculture oriented multi-source data fusion irrigation regulation method according to claim 8, characterized in that, The method for generating the ecological-friendly irrigation control scheme comprises the following steps: The geographical coordinates and unique codes of the sensitive land plots are obtained from historical irrigation records of the sensitive land plots; a sensitive land plot neighborhood correlation graph is constructed according to the geographical coordinates of the sensitive land plots; a neighborhood ecological coupling factor of each sensitive land plot is calculated according to the sensitive land plot neighborhood correlation graph and the ecological degradation rate of each sensitive land plot; a crop growth stage of each sensitive land plot is obtained, and a growth stage sensitivity coefficient of each sensitive land plot is calculated; meteorological forecast data are obtained, and a meteorological response adjustment coefficient of each sensitive land plot is calculated; The optimal benefit-ecological balance point, the neighborhood ecological coupling factor, the growth stage sensitivity coefficient and the meteorological response adjustment coefficient of each sensitive land plot are used to calculate a comprehensive adjusted irrigation amount of each sensitive land plot; an ecological urgency index of each sensitive land plot is calculated according to the comprehensive ecological health index and the ecological degradation rate of each sensitive land plot; the irrigation control priority of each sensitive land plot is determined according to the ecological urgency index and the comprehensive adjusted irrigation amount of each sensitive land plot; cross-land plot ecological compensation adjustment is performed according to the irrigation control priority and the comprehensive adjusted irrigation amount of each sensitive land plot to obtain a comprehensive compensation irrigation amount of each sensitive land plot; The comprehensive compensation irrigation amount, the irrigation control priority and the unique code of each sensitive land plot are integrated to form an irrigation control parameter of each sensitive land plot; the irrigation control parameters of all sensitive land plots are summarized to generate an ecological-friendly irrigation control scheme.

10. The multi-source data fusion irrigation regulation system for smart agriculture, implementing the multi-source data fusion irrigation regulation method for smart agriculture according to any one of claims 1-9, characterized in that, The method comprises the following steps: An efficiency modeling module is used to collect historical irrigation records and corresponding crop yield data of each farmland plot, regress the response relationship between different irrigation water amounts and crop yield data in the historical irrigation records, and construct an efficiency response model library of each farmland plot; An ecological monitoring module is configured to collect groundwater level monitoring data and soil salt accumulation data in real time around the farmland, evaluate the evolution risk degree of each farmland plot from the current ecological state to the critical state of ecological degradation, and generate an ecological health degree attenuation curve of each farmland plot; A constraint quantification module is configured to acquire water resource management data and ecological red line constraint parameters, calculate a theoretical optimal irrigation amount of each farmland plot under the constraint of the benefit response model library, determine an irrigation amount feasible region of each farmland plot based on the theoretical optimal irrigation amount, and determine a sensitive plot in combination with the ecological health degree attenuation curve; A balance optimization module is configured to analyze long-term benefit net present value and ecological recovery cycle estimation of each sensitive plot under different irrigation intensity schemes according to the irrigation amount feasible region and the ecological health degree attenuation curve of each sensitive plot, and intelligently identify an optimal benefit-ecological balance point; A sustainable regulation module is configured to generate an ecological friendly irrigation regulation scheme based on the optimal benefit-ecological balance point of each sensitive plot, and intelligently schedule irrigation water of each sensitive plot according to the ecological friendly irrigation regulation scheme.

Images