Water-soil-food resources coupling optimization allocation and greenhouse gas environmental impact assessment system

Abstract

This invention relates to the field of resource and environmental science and technology, specifically to a system for coupled optimization of water-soil-food resources and assessment of greenhouse gas environmental impact. The system includes a data acquisition module, a multi-factor dynamic coupling model module, a dynamic adaptive configuration algorithm module, and a closed-loop feedback control module. This invention significantly reduces the carbon emission intensity of food production by improving resource allocation efficiency. It can track changes in precipitation, soil moisture, fertility, and market supply and demand in real time. Using a four-dimensional coupled model of water, soil, food, and greenhouse gases, it accurately calculates the output efficiency and carbon emission cost of each unit of water and fertilizer in different plots. When water and soil inputs per unit of food yield are too high or greenhouse gas emissions surge in a certain area, the system automatically adjusts the resource allocation scheme, prioritizing irrigation water and fertilizer to plots with lower emissions and higher yields.

Description

Technical Field

[0001] This invention relates to the field of resource and environmental science and technology, specifically to a system for coupled optimal allocation of water, soil and food resources and assessment of the environmental impact of greenhouse gases. Background Technology

[0002] Agricultural development cannot be separated from the coordinated support of water resources, land resources, and grain production. These three are the core foundation for ensuring food security and sustainable agricultural development. Currently, my country's agriculture is transforming towards green development, which requires not only ensuring stable grain output but also focusing on the efficient use of resources, while also addressing the environmental pressure brought about by greenhouse gas emissions.

[0003] In agricultural production practices, water resources, land resources, and food production are closely linked, and greenhouse gas emissions are closely related to the utilization of these resources. With the advancement of dual-carbon goals and the deepening of the concept of green agricultural development, traditional agricultural resource management models are no longer sufficient to meet actual needs. Therefore, it is necessary to combine the allocation of water, soil, and food resources with greenhouse gas environmental impact assessments for management. According to authorization announcement number CN119395231A, a method for assessing forage digestion and greenhouse gas emissions from ruminant livestock in grazing-type artificial mixed-sowing grasslands is disclosed. This invention features real-time monitoring and precise assessment: through real-time data monitoring and intelligent analysis, the system can accurately assess the forage digestion efficiency of grazing-type mixed-sowing grasslands and the greenhouse gas emissions from ruminant livestock, improving the accuracy and timeliness of data. It is also simple to operate and cost-effective: automated data collection and analysis reduce the demand for human and physical resources, simplify operational processes, and lower management costs. The implementation of the system improves the efficiency of grassland management and helps farmers or managers make data-driven decisions quickly. Environmental Impact Assessment and Management Improvement: By assessing the environmental impact of different grazing strategies and grassland combinations, the system helps to optimize grazing management, reduce greenhouse gas emissions, and thus contribute to ecological environmental protection.

[0004] Currently, traditional models mostly focus on the optimal allocation of a single resource or a single greenhouse gas environmental assessment, failing to integrate multiple factors such as water, soil, food, and greenhouse gases. Consequently, they cannot achieve simultaneous quantification of resource allocation and environmental impact. Moreover, traditional resource allocation models lack flexibility and cannot be adjusted in real time according to changes in precipitation, soil moisture, food market demand, and emission reduction targets. Furthermore, resource allocation and environmental assessment are often disconnected, lacking an effective feedback and control mechanism, resulting in a failure to promote the coordinated advancement of resource utilization efficiency improvement and greenhouse gas emission reduction. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a water-soil-food resource coupling optimization allocation and greenhouse gas environmental impact assessment system. This system solves the problems of traditional models that only focus on single resource optimization or single greenhouse gas assessment, cannot integrate multiple factors, have rigid resource allocation models that are disconnected from environmental assessment, and lack feedback and control mechanisms, resulting in the inability to promote resource utilization efficiency and greenhouse gas emission reduction in a coordinated manner.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a water-soil-food resource coupled optimization configuration and greenhouse gas environmental impact assessment system, including a data acquisition module, a multi-element dynamic coupling model module, a dynamic adaptive configuration algorithm module, and a closed-loop feedback control module.

[0007] The data acquisition module is used to acquire in real time regional-scale precipitation time series data, soil moisture spatial distribution data, soil fertility level data, supply and demand structure data of major grain crops, grain market price fluctuation data, and greenhouse gas emission inventory data.

[0008] The multi-factor dynamic coupling model module has a built-in four-dimensional coupling framework of water-soil-grain-greenhouse gas. It uses the available water resources, soil fertility correction coefficient, grain production and consumption balance difference and greenhouse gas emission intensity as state variables to establish a set of simultaneous equations including water balance equation, soil nitrogen and carbon turnover equation, grain supply and demand difference equation and greenhouse gas emission factor.

[0009] The dynamic adaptive configuration algorithm module receives real-time data from the data acquisition module, takes the current state variables output by the multi-factor dynamic coupling model module as the initial condition, takes minimizing the water and soil resource input per unit of grain output as the sub-objective, and takes the greenhouse gas emission increment per unit area not exceeding a set threshold as the constraint to generate a water and soil resource spatial allocation scheme.

[0010] The closed-loop feedback control module compares the measured greenhouse gas emission data after the implementation of the allocation scheme with the estimated value of the multi-factor dynamic coupling model module, calculates the feedback deviation, and transmits the deviation signal simultaneously to the parameter correction unit of the multi-factor dynamic coupling model module and the constraint boundary update unit of the dynamic adaptive configuration algorithm module, forming a closed-loop control structure that is optimized every 24 hours.

[0011] Preferably, the simultaneous equations in the multi-factor dynamic coupling model module include a sub-model of soil organic carbon response to nitrogen fertilizer application, wherein the soil nitrous oxide emission flux is calculated using the following formula.

[0012]

[0013] in, This represents the daily average emission flux of nitrous oxide from the soil, expressed in units of... ;

[0014] The reference emission factor is in the same unit.

[0015] This represents the actual amount of nitrogen fertilizer applied. The recommended nitrogen application rate for the region is shown in both units. ;

[0016] α is the temperature sensitivity coefficient, with units of °C. -1 ;

[0017] For soil temperature, Both are reference temperatures, and the unit is °C.

[0018] Soil water-filling porosity, dimensionless;

[0019] β is the porosity index coefficient.

[0020] Preferably, the data acquisition module includes a satellite remote sensing inversion submodule, a ground-based Internet of Things (IoT) sensor network submodule, and a statistical data interface submodule. The satellite remote sensing inversion submodule is used to acquire normalized vegetation index, surface temperature, and soil moisture products, and invert to generate a spatial distribution map of crop water requirements. The ground-based IoT sensor network submodule is used to collect soil moisture content, soil temperature, atmospheric temperature and humidity, and carbon dioxide concentration at the point scale. The statistical data interface submodule is used to access grain yield, sown area, agricultural input consumption, and rural population data from local statistical yearbooks.

[0021] Preferably, the dynamic adaptive configuration algorithm module adopts a dynamic programming solver based on reinforcement learning, wherein the state space consists of water resource availability, soil moisture level, grain stock-to-consumption ratio and current cumulative greenhouse gas emissions, the action space is the water-soil resource allocation quota of each sub-region, and the reward function introduces a greenhouse gas emission penalty term.

[0022] In each rolling optimization, the solver uses the latest monitoring data from the data acquisition module to calculate the resource replacement rate under the constraint of unit grain carbon emission intensity using the following formula:

[0023]

[0024] in, Let be the resource replacement rate at time t when adjusting the unit grain output from subregion i to subregion j, in units of . ;

[0025] For adjusting the amount of water, The adjustment amount refers to the land area, and the units for both are m.3 and ha;

[0026] c w Let c1 be the marginal cost of water and c2 be the marginal cost of land.

[0027] This represents the change in grain production, expressed in kg.

[0028] λ is the carbon emission penalty coefficient, which is dimensionless;

[0029] The change in greenhouse gas emissions, in units .

[0030] Preferably, the closed-loop feedback control module includes an error detection unit, a parameter adaptive updater, and a constraint boundary softening unit. The error detection unit is used to calculate the relative deviation and cumulative deviation between the measured greenhouse gas emission flux and the model prediction. The parameter adaptive updater updates the soil nitrogen turnover rate constant, water evaporation coefficient, and crop growth coefficient in the multi-factor dynamic coupling model module according to the deviation signal using the Kalman filter algorithm. When the deviation exceeds a preset threshold for three consecutive cycles, the constraint boundary softening unit relaxes the upper and lower limits of the greenhouse gas emission constraint per unit area by a preset step size in the next optimization cycle.

[0031] Preferably, in the four-dimensional coupling framework of water-soil-food-greenhouse gas, the water molecule module adopts the reference crop evapotranspiration calculation method based on the Penman-Montess formula, the soil submodule adopts a layered soil hydrodynamic model coupled with the first-order kinetic equation of soil organic carbon decomposition, the food submodule adopts the food yield response function based on the harvest index theory, and the greenhouse gas submodule simultaneously includes source and sink terms for methane emissions, nitrous oxide emissions, and carbon dioxide fixation and release.

[0032] Preferably, after receiving the deviation signal from the closed-loop feedback control module, the multi-factor dynamic coupling model module performs online correction on the synergy coefficient between the water footprint and carbon footprint of grain production. The correction process uses the following recursive formula:

[0033]

[0034] in, and are the water-carbon synergistic coefficients at time t+1 and time t, respectively, and are dimensionless.

[0035] The learning rate is between 0.01 and 0.1.

[0036] This represents the total measured greenhouse gas emissions during the period. The values ​​are model estimates, and both are in units of 1 / 2. ;

[0037] The measured total water consumption within period t. The values ​​are model estimates, and both are in units of 1 / 2. .

[0038] Preferably, the water-soil resource spatial allocation scheme output by the dynamic adaptive configuration algorithm module uses a 1:10000 scale gridded map as the output carrier. Each grid cell contains recommended irrigation quotas, recommended fertilization amounts, recommended planting systems, and corresponding estimated greenhouse gas emissions per unit area, and generates a zoning operation instruction file that can be imported into an agricultural machinery navigation system.

[0039] Preferably, the greenhouse gas emission inventory data in the data acquisition module includes at least the direct emission factor of nitrous oxide from farmland, the daily methane emission flux from paddy fields, the carbon dioxide emission coefficient from fuel combustion in agricultural machinery, and the annual change rate of soil organic carbon; the time resolution of the emission inventory data is daily, and the spatial resolution is administrative villages or equal-area grids.

[0040] Preferably, the data interaction frequency between the dynamic adaptive configuration algorithm module and the multi-factor dynamic coupling model module is once per hour;

[0041] When an extreme precipitation or drought warning event is triggered, the interaction frequency is automatically encrypted to once every 15 minutes. At the same time, the dynamic adaptive configuration algorithm module temporarily blocks food market fluctuation signals below the set time limit and prioritizes the response to the coupling effect of water stress and sudden changes in greenhouse gas emissions.

[0042] The beneficial effects of this invention are reflected in:

[0043] 1. This invention significantly reduces the carbon emission intensity of grain production by improving resource allocation efficiency. It can track changes in precipitation, soil moisture, fertility, and market supply and demand in real time. Using a four-dimensional coupling model of water-soil-grain-greenhouse gas, it can accurately calculate the output efficiency and carbon emission cost of each unit of water and fertilizer in different plots. When the water and soil input per unit of grain yield in a certain area is too high or the greenhouse gas emissions surge, the system will automatically adjust the resource allocation plan, giving priority to directing irrigation water and fertilizer to plots with lower emissions and higher output.

[0044] 2. This invention continuously improves the accuracy and reliability of model prediction and control through closed-loop rolling correction. Every 24 hours, the measured greenhouse gas emission data is compared with the model prediction results, the deviation is calculated and fed back in a timely manner. If it is found that the actual value of nitrous oxide emission in a certain plot is consistently higher than the value calculated by the model, the system will automatically correct key parameters such as soil nitrogen turnover rate and evaporation coefficient using the Kalman filter algorithm to make the model more in line with the real situation.

[0045] 3. This invention significantly improves the system's ability to cope with extreme weather by using a high-frequency collaboration mechanism between modules and an emergency response mechanism. The dynamic adaptive configuration algorithm module and the multi-factor dynamic coupling model module exchange data once an hour. Once an extreme precipitation or drought warning is triggered, the exchange frequency will be automatically encrypted to once every fifteen minutes. At this time, the system will temporarily ignore short-term grain market fluctuation signals that are not significant and concentrate its efforts on dealing with the coupling effect of water stress and sudden changes in greenhouse gas emissions. Attached Figure Description

[0046] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0047] Figure 1 This is a schematic diagram of the data acquisition module structure and data flow of the present invention;

[0048] Figure 2 This is a schematic diagram illustrating the interaction process between the closed-loop feedback control module and the model and algorithm of the present invention.

[0049] Figure 3 This diagram illustrates the interaction frequency and event response of the dynamic adaptive configuration algorithm and model module of the present invention. Detailed Implementation

[0050] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0051] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0052] Example 1

[0053] like Figures 1-3 As shown, the water-soil-food resource coupling optimization configuration and greenhouse gas environmental impact assessment system in this embodiment is characterized by including a data acquisition module, a multi-element dynamic coupling model module, a dynamic adaptive configuration algorithm module, and a closed-loop feedback control module.

[0054] The data acquisition module is used to acquire multi-source heterogeneous data at the regional scale in real time, including: precipitation time-series data and soil moisture spatial distribution data retrieved by satellite remote sensing; soil fertility level data at the point scale collected by the ground Internet of Things sensor network; and supply and demand structure data of major grain crops, grain market price fluctuation data and greenhouse gas emission inventory data accessed through statistical data interface; the data acquisition module has a built-in spatiotemporal unified scale data fusion unit to resample the above multi-source data to a unified spatial grid of 1km×1km or administrative village boundary, and synchronize it to daily time resolution;

[0055] The multi-factor dynamic coupling model module incorporates a four-dimensional coupling framework of water, soil, food, and greenhouse gases. Its key feature is that this coupling framework is not a simple list of sub-equations, but rather achieves multi-factor dynamic coupling through simultaneous solution and interaction with state variables. Specifically, it uses the available water resources... Soil fertility correction coefficient The imbalance between grain production and consumption and greenhouse gas emission intensity For the state variables, establish and solve the following system of equations simultaneously in real time:

[0056]

[0057] in, The root zone soil water storage (mm); For precipitation ; Actual evaporation ; , These are surface runoff and deep infiltration, respectively.

[0058] Soil organic carbon pool ; This refers to the rate of organic matter input. The decomposition rate constant; These are the correction functions for the decomposition effect, respectively, based on temperature and moisture content.

[0059] For the region's total food demand and total supply; This represents the change in inventory.

[0060] These are the emission factors for nitrous oxide, methane, and carbon dioxide, respectively.

[0061] The above-mentioned state variables of water, soil carbon, and food supply and demand The solution is obtained through simultaneous iterations at each time step, where... Depends on Exported soil hydraulic parameters, Feedback is used to regulate the priority of irrigation water allocation, thereby achieving dynamic coupling of four-dimensional state variables of water, soil, food and greenhouse gases;

[0062] The dynamic adaptive configuration algorithm module receives real-time data from the data acquisition module, uses the current state variables output by the multi-factor dynamic coupling model module as initial conditions, minimizes water and soil resource input per unit of grain yield as a sub-objective, and simultaneously constrains the increase in greenhouse gas emissions per unit area to not exceed a set threshold as a constraint, to generate a spatial allocation scheme for water and soil resources; its key feature is that this module embeds a hybrid solver based on improved particle swarm optimization and rolling time-domain control, and its core algorithm steps are as follows:

[0063] Step A1 (Initialization): Assume there are N spatial sub-regions (grids or administrative villages), and the time optimization window length is... =24 hours, at time t, the soil moisture of each sub-region is obtained from the data acquisition module. fertility The predicted value of current greenhouse gas emission intensity is obtained from the multi-factor dynamic coupling model module. ;

[0064] Step A2 (Construction of Optimization Variables and Objective Function): Define the decision variable matrix. ,in, The amount of irrigation water allocated to sub-region i , Nitrogen application rate Construct a multi-objective optimization problem:

[0065]

[0066] in, Forecast of grain yield in sub-region i The grain yield response function in the multi-factor dynamic coupling model module is used for calculation.

[0067] These are weighting coefficients (dimensionless). Water-fertilizer equivalent conversion coefficient ;

[0068] Let i be the greenhouse gas emission intensity of subregion i under given irrigation and fertilization conditions. ;

[0069] The first term of the objective function is the water and soil resource input index per unit of grain output, and the second term is the penalty term for exceeding emission limits;

[0070] Step A3 (Constraints): The following constraints must be satisfied:

[0071]

[0072] in, Let the area of ​​the subregion be (ha). Root depth (mm) The slack variable is dynamically adjusted by the closed-loop feedback control module;

[0073] Step A4 (Solution Algorithm): The above nonlinear constrained optimization problem is solved using a particle swarm optimization algorithm with inertia weights. Each particle represents a group of particles. The scheme approximates the global optimum by iteratively updating particle velocities and positions, and finally outputs the optimal allocation scheme. ;

[0074] The closed-loop feedback control module will allocate the scheme After implementation, greenhouse gas emission data will be measured using ground-based IoT sensors. Estimated values ​​with multi-factor dynamic coupling model module The comparison and feedback deviation calculation are performed. The module is characterized by including a deviation decoupling and dual-channel correction unit, which simultaneously corrects model parameters and algorithm constraint boundaries according to the following steps, forming a closed-loop control structure that is continuously optimized every 24 hours:

[0075] Step B1 (Obstacle Calculation): Calculate the relative deviation of each sub-region i. and cumulative deviation :

[0076]

[0077]

[0078] Step B2 (Parameter Correction Channel): The deviation signal... The parameter correction unit, which transmits data to the multi-factor dynamic coupling model module, updates the soil nitrogen turnover rate constant online using a recursive least squares algorithm. and water evaporation coefficient :

[0079]

[0080] in, The vector of parameters to be corrected. This is the sensitivity matrix. Kalman gain;

[0081] When the cumulative deviation over three consecutive periods At that time, the emission constraint threshold for the next optimization cycle is adjusted according to the following adaptive law. :

[0082]

[0083] in, For softening step size coefficient , The function guarantees that the adjustment range is bounded. Let be the historical emission fluctuation standard deviation of sub-region i. This design enables the system to moderately relax emission constraints to avoid excessive suppression of production when the measured emissions are consistently higher than the estimated values, and to tighten constraints to strengthen emission reduction, thus forming an adaptive soft constraint mechanism.

[0084] Step B4 (Rolling Time Domain Advancement): After completing parameter correction and constraint boundary update, the system enters the next optimization cycle t+1, repeating steps A1-A4 and B1-B4 to achieve closed-loop control of rolling optimization every 24 hours.

[0085] The data acquisition module is used to acquire real-time regional-scale precipitation time-series data, soil moisture spatial distribution data, soil fertility level data, supply and demand structure data of major food crops, grain market price fluctuation data, and greenhouse gas emission inventory data, specifically including:

[0086] Water resources data:

[0087] Precipitation time series data, satellite inversion and ground rain gauge correction;

[0088] Surface water and groundwater resource data, including river runoff, reservoir storage, water diversion, irrigation area water intake, groundwater depth and extraction intensity;

[0089] Agricultural irrigation water data, including irrigation quotas, irrigation efficiency, and water transmission and distribution losses for different water sources (surface water, groundwater, and reclaimed water);

[0090] Evapotranspiration data, including actual evapotranspiration and potential evapotranspiration;

[0091] Land resource dimension data:

[0092] Soil property data, including soil texture, organic carbon content, pH value, bulk density, field water holding capacity, wilting coefficient and salinization degree;

[0093] Land use and cover data, including the spatial distribution and dynamic changes of cultivated land, construction land, forest land, grassland, wetlands, etc.

[0094] Data on arable land quality and utilization intensity, including the delineation of basic farmland, multiple cropping index, distribution of abandoned land, land consolidation projects, and information on the balance of arable land occupation and compensation;

[0095] Food system dimension data:

[0096] Data on the structure of grain crops, including sown area, planting system, crop rotation pattern and growth period division of different crop types (wheat, corn, rice, soybean, etc.);

[0097] Grain supply and demand and consumption data, including production, inventory, import and export volume, processing and conversion volume, feed grain, industrial grain, and staple food consumption structure;

[0098] Grain market price fluctuation data, including purchase price at the place of origin, wholesale market price, retail price, and price warning threshold;

[0099] Data on residents' dietary structure and nutritional needs, including per capita staple food consumption, and the composition of protein and calorie intake sources;

[0100] Greenhouse gas and carbon sink data:

[0101] Greenhouse gas emission inventory data includes direct / indirect emission factors of nitrous oxide from farmland, daily methane emission flux from paddy fields, carbon dioxide emission coefficients from fuel combustion in agricultural machinery, emission factors from livestock and poultry farming, and annual change rate of soil organic carbon.

[0102] Carbon satellite remote sensing data, including column concentration dry air mixing ratio, methane concentration, and inverted regional carbon flux changes;

[0103] Farmland carbon sequestration data includes crop biomass carbon accumulation, straw return carbon input, changes in soil organic carbon pool, and the carbon sequestration potential of conservation tillage;

[0104] High-resolution emission grid data, with a temporal resolution of no less than daily and a spatial resolution reaching the administrative village level or a 1km×1km grid.

[0105] The multi-factor dynamic coupling model module has a built-in four-dimensional coupling framework of water-soil-food-greenhouse gas. It uses water availability, soil fertility correction coefficient, food production-consumption balance difference and greenhouse gas emission intensity as state variables to establish a set of simultaneous equations including water balance equation, soil nitrogen and carbon turnover equation, food supply and demand difference equation and greenhouse gas emission factors.

[0106] The dynamic adaptive configuration algorithm module receives real-time data from the data acquisition module, takes the current state variables output by the multi-factor dynamic coupling model module as the initial condition, takes minimizing the water and soil resource input per unit of grain output as the sub-objective, and takes the greenhouse gas emission increment per unit area not exceeding the set threshold as the constraint to generate a water and soil resource spatial allocation scheme.

[0107] The closed-loop feedback control module compares the measured greenhouse gas emission data after the allocation scheme is implemented with the estimated value of the multi-factor dynamic coupling model module, calculates the feedback deviation, and transmits the deviation signal to the parameter correction unit of the multi-factor dynamic coupling model module and the constraint boundary update unit of the dynamic adaptive configuration algorithm module at the same time, forming a closed-loop control structure that is optimized every 24 hours.

[0108] The simultaneous equations in the multi-factor dynamic coupling model module include a sub-model of soil organic carbon response to nitrogen fertilizer application, where soil nitrous oxide emission flux is calculated using the following formula.

[0109]

[0110] in, This represents the daily average emission flux of nitrous oxide from the soil, in units of... ;

[0111] The reference emission factor is in the same unit.

[0112] This represents the actual amount of nitrogen fertilizer applied. The recommended nitrogen application rate for the region is shown in both units. ;

[0113] α is the temperature sensitivity coefficient, with units of °C. -1 ;

[0114] For soil temperature, Both are reference temperatures, and the unit is °C.

[0115] Soil water-filling porosity, dimensionless;

[0116] β is the porosity index coefficient.

[0117] The data acquisition module includes a satellite remote sensing inversion submodule, a ground-based IoT sensor network submodule, and a statistical data interface submodule. The satellite remote sensing inversion submodule is used to acquire normalized vegetation index, surface temperature, and soil moisture products, and invert to generate a spatial distribution map of crop water requirements. The ground-based IoT sensor network submodule is used to collect soil moisture content, soil temperature, atmospheric temperature and humidity, and carbon dioxide concentration at the point scale. The statistical data interface submodule is used to access grain yield, sown area, agricultural input consumption, and rural population data from local statistical yearbooks.

[0118] Among them, the dynamic adaptive configuration algorithm module adopts a dynamic programming solver based on reinforcement learning. The state space consists of water resource availability, soil moisture level, grain stock consumption ratio and current cumulative greenhouse gas emissions. The action space is the water-soil resource allocation quota for each sub-region. The reward function introduces a greenhouse gas emission penalty term.

[0119] In each rolling optimization, the solver uses the latest monitoring data from the data acquisition module to calculate the resource replacement rate under the constraint of unit grain carbon emission intensity using the following formula:

[0120]

[0121] in, Let be the resource replacement rate at time t when adjusting the unit grain output from subregion i to subregion j, in units of . ;

[0122] For adjusting the amount of water, The adjustment amount refers to the land area, and the units for both are m. 3 and ha;

[0123] c w Let c1 be the marginal cost of water and c2 be the marginal cost of land.

[0124] This represents the change in grain production, expressed in kg.

[0125] λ is the carbon emission penalty coefficient, which is dimensionless;

[0126] The change in greenhouse gas emissions, in units .

[0127] The closed-loop feedback control module includes an error detection unit, a parameter adaptive updater, and a constraint boundary softening unit. The error detection unit is used to calculate the relative and cumulative deviations between the measured greenhouse gas emission fluxes and the model predictions. The parameter adaptive updater uses the Kalman filter algorithm to update the soil nitrogen turnover rate constant, water evaporation coefficient, and crop growth coefficient in the multi-factor dynamic coupling model module based on the deviation signal. When the deviation exceeds the preset threshold for three consecutive cycles, the constraint boundary softening unit relaxes the upper and lower limits of the greenhouse gas emission constraints per unit area by a preset step size in the next optimization cycle.

[0128] In the four-dimensional coupling framework of water-soil-food-greenhouse gas, the water molecule module adopts the reference crop evapotranspiration calculation method based on the Penman-Montess formula, the soil submodule adopts a layered soil hydrodynamic model coupled with the first-order kinetic equation of soil organic carbon decomposition, the food submodule adopts the food yield response function based on the harvest index theory, and the greenhouse gas submodule includes source and sink terms for methane emissions, nitrous oxide emissions, and carbon dioxide fixation and release.

[0129] Among them, the multi-factor dynamic coupling model module, after receiving the deviation signal from the closed-loop feedback control module, performs online correction on the synergy coefficient between the water footprint and carbon footprint of grain production. The correction process uses the following recursive formula:

[0130]

[0131] in, and are the water-carbon synergistic coefficients at time t+1 and time t, respectively, and are dimensionless.

[0132] The learning rate is between 0.01 and 0.1.

[0133] This represents the total measured greenhouse gas emissions during the period. The values ​​are model estimates, and both are in units of 1 / 2. ;

[0134] The measured total water consumption within period t. The values ​​are model estimates, and both are in units of 1 / 2. .

[0135] The dynamic adaptive configuration algorithm module outputs a spatial allocation scheme for water and soil resources using a 1:10000 scale gridded map as its output carrier. Each grid cell contains recommended irrigation quotas, recommended fertilization amounts, recommended planting systems, and corresponding estimated greenhouse gas emissions per unit area. It also generates a zoning operation instruction file that can be imported into an agricultural machinery navigation system.

[0136] The greenhouse gas emission inventory data in the data acquisition module includes at least the direct emission factor of nitrous oxide from farmland, the daily methane emission flux from paddy fields, the carbon dioxide emission coefficient of fuel combustion in agricultural machinery, and the annual change rate of soil organic carbon; the time resolution of the emission inventory data is daily, and the spatial resolution is administrative village or equal area grid.

[0137] The data interaction frequency between the dynamic adaptive configuration algorithm module and the multi-factor dynamic coupling model module is once per hour.

[0138] When an extreme precipitation or drought warning event is triggered, the interaction frequency is automatically encrypted to once every 15 minutes. At the same time, the dynamic adaptive configuration algorithm module temporarily blocks food market fluctuation signals below the set time limit and prioritizes the response to the coupling effect of water stress and sudden changes in greenhouse gas emissions.

[0139] Example 2

[0140] Based on the adaptability under water resource fluctuation conditions in typical rain-fed agricultural areas, the specifics are as follows:

[0141] In areas with significant interannual precipitation fluctuations, the data acquisition module uses satellite remote sensing to invert daily precipitation time series and soil moisture spatial distribution. The multi-factor dynamic coupling model module uses the water balance equation and soil nitrogen and carbon turnover equation as its core to dynamically calculate the effective soil water storage and organic carbon decomposition rate. The dynamic adaptive configuration algorithm module aims to minimize water and soil input per unit of grain yield. Under the constraint that the increase in greenhouse gas emissions does not exceed a set threshold, it generates irrigation quotas and fertilization plans for each grid unit. The closed-loop feedback control module compares the measured nitrous oxide flux with the model prediction every 24 hours and updates the soil nitrogen turnover rate constant using Kalman filtering. When the measured emissions are high for three consecutive days, the upper limit of emissions per unit area in the constraint boundary softening unit is relaxed by step size to avoid excessive suppression of yield.

[0142] Example 3

[0143] Based on the water and carbon balance regulation in major grain-producing areas and the response to market fluctuations, the specifics are as follows:

[0144] In regions where the structure of grain supply and demand changes significantly, the statistical data interface submodule accesses data on grain output, sown area, and market price fluctuations. The multi-factor dynamic coupling model module combines the grain supply and demand difference equation with the greenhouse gas emission factor, outputting the current stock-to-consumption ratio and emission intensity. The dynamic adaptive configuration algorithm module introduces a reinforcement learning solver, with the state space including water availability, soil moisture level, stock-to-consumption ratio, and cumulative emissions. A carbon emission penalty term is set in the reward function to calculate the resource replacement rate under the constraint of carbon emissions per unit of grain. The closed-loop feedback control module compares the measured emissions with the estimated values ​​and recursively corrects the water-carbon synergy coefficient. When an extreme drought warning is triggered, the algorithm module temporarily blocks market fluctuation signals below 72 hours and prioritizes the response to the coupling effect of water stress and sudden changes in nitrous oxide emissions.

[0145] Example 4

[0146] The optimization of methane and nitrous oxide emission reduction in rice-growing areas is as follows:

[0147] In rice-dryland rotation areas, greenhouse gas emission inventory data simultaneously includes daily methane emission flux from paddy fields, nitrous oxide emission factors during the dry season, and carbon dioxide emission coefficients from agricultural machinery fuel combustion. The multi-factor dynamic coupling model module adopts a layered soil hydrodynamic model and couples it with the first-order kinetic equation of soil organic carbon decomposition. The dynamic adaptive configuration algorithm module outputs recommended irrigation quotas and planting systems at a 1:10000 scale grid. During the rice season, an intermittent irrigation scheme is adopted to reduce methane emissions, and during the dry season, nitrogen fertilizer application is optimized to control nitrous oxide. The closed-loop feedback control module interacts with data every 15 minutes. The error detection unit calculates the relative deviation between measured and estimated emissions, and the parameter adaptive updater performs online correction of the methane oxidation rate constant and denitrification rate constant, generating a zoned operation instruction file that can be imported into the agricultural machinery navigation system.

[0148] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A system for coupled optimal allocation of water, soil, and food resources and assessment of greenhouse gas environmental impact, characterized in that, It includes a data acquisition module, a multi-factor dynamic coupling model module, a dynamic adaptive configuration algorithm module, and a closed-loop feedback control module; The data acquisition module is used to acquire in real time regional-scale precipitation time series data, soil moisture spatial distribution data, soil fertility level data, supply and demand structure data of major grain crops, grain market price fluctuation data, and greenhouse gas emission inventory data. The multi-factor dynamic coupling model module has a built-in four-dimensional coupling framework of water-soil-grain-greenhouse gas. It uses the available water resources, soil fertility correction coefficient, grain production and consumption balance difference and greenhouse gas emission intensity as state variables to establish a set of simultaneous equations including water balance equation, soil nitrogen and carbon turnover equation, grain supply and demand difference equation and greenhouse gas emission factor. The dynamic adaptive configuration algorithm module receives real-time data from the data acquisition module, takes the current state variables output by the multi-factor dynamic coupling model module as the initial condition, takes minimizing the water and soil resource input per unit of grain output as the sub-objective, and takes the greenhouse gas emission increment per unit area not exceeding a set threshold as the constraint to generate a water and soil resource spatial allocation scheme. The closed-loop feedback control module compares the measured greenhouse gas emission data after the implementation of the allocation scheme with the estimated value of the multi-factor dynamic coupling model module, calculates the feedback deviation, and transmits the deviation signal simultaneously to the parameter correction unit of the multi-factor dynamic coupling model module and the constraint boundary update unit of the dynamic adaptive configuration algorithm module, forming a closed-loop control structure that is optimized every 24 hours.

2. The water-soil-food resource coupling optimization allocation and greenhouse gas environmental impact assessment system according to claim 1, characterized in that: The simultaneous equations in the multi-factor dynamic coupling model module include a sub-model of soil organic carbon response to nitrogen fertilizer application rate, wherein soil nitrous oxide emission flux is calculated using the following formula. in, This represents the daily average emission flux of nitrous oxide from the soil, in units of... ; The reference emission factor is in the same unit. This represents the actual amount of nitrogen fertilizer applied. The recommended nitrogen application rate for the region is shown in both units. ; α is the temperature sensitivity coefficient, with units of °C. -1 ; For soil temperature, Both are reference temperatures, and the unit is °C. Soil water-filling porosity, dimensionless; β is the porosity index coefficient.

3. The water-soil-food resource coupling optimization allocation and greenhouse gas environmental impact assessment system according to claim 1, characterized in that: The data acquisition module includes a satellite remote sensing inversion submodule, a ground-based Internet of Things (IoT) sensor network submodule, and a statistical data interface submodule. The satellite remote sensing inversion submodule is used to acquire normalized vegetation index, surface temperature, and soil moisture products, and invert to generate a spatial distribution map of crop water requirements. The ground-based IoT sensor network submodule is used to collect soil moisture content, soil temperature, atmospheric temperature and humidity, and carbon dioxide concentration at the point scale. The statistical data interface submodule is used to access grain yield, sown area, agricultural input consumption, and rural population data from local statistical yearbooks.

4. The water-soil-food resource coupling optimization allocation and greenhouse gas environmental impact assessment system according to claim 1, characterized in that: The dynamic adaptive configuration algorithm module adopts a dynamic programming solver based on reinforcement learning. The state space consists of water resource availability, soil moisture level, grain stock-to-consumption ratio and current cumulative greenhouse gas emissions. The action space is the water-soil resource allocation quota for each sub-region. The reward function introduces a greenhouse gas emission penalty term. In each rolling optimization, the solver uses the latest monitoring data from the data acquisition module to calculate the resource replacement rate under the constraint of unit grain carbon emission intensity using the following formula: in, Let be the resource replacement rate at time t when adjusting the unit grain output from subregion i to subregion j, in units of . ; For adjusting the amount of water, The adjustment amount refers to the land area, and the units for both are m. 3 and ha; c w Let c1 be the marginal cost of water and c2 be the marginal cost of land. This represents the change in grain production, expressed in kg. λ is the carbon emission penalty coefficient, which is dimensionless; The change in greenhouse gas emissions, in units .

5. The water-soil-food resource coupling optimization allocation and greenhouse gas environmental impact assessment system according to claim 1 is characterized in that: The closed-loop feedback control module includes an error detection unit, a parameter adaptive updater, and a constraint boundary softening unit. The error detection unit is used to calculate the relative deviation and cumulative deviation between the measured greenhouse gas emission flux and the model prediction. The parameter adaptive updater uses the Kalman filter algorithm to update the soil nitrogen turnover rate constant, water evaporation coefficient, and crop growth coefficient in the multi-factor dynamic coupling model module according to the deviation signal. When the deviation exceeds the preset threshold for three consecutive cycles, the constraint boundary softening unit relaxes the upper and lower limits of the greenhouse gas emission constraint per unit area by a preset step size in the next optimization cycle.

6. The water-soil-food resource coupling optimization allocation and greenhouse gas environmental impact assessment system according to claim 1, characterized in that: In the proposed four-dimensional coupling framework of water-soil-food-greenhouse gas, the water molecule module adopts the reference crop evapotranspiration calculation method based on the Penman-Montes formula, the soil submodule adopts a layered soil hydrodynamic model coupled with the first-order kinetic equation of soil organic carbon decomposition, the food submodule adopts the food yield response function based on the harvest index theory, and the greenhouse gas submodule simultaneously includes source and sink terms for methane emissions, nitrous oxide emissions, and carbon dioxide fixation and release.

7. The water-soil-food resource coupling optimization allocation and greenhouse gas environmental impact assessment system according to claim 1, characterized in that: After receiving the deviation signal from the closed-loop feedback control module, the multi-factor dynamic coupling model module performs online correction on the synergy coefficient between the water footprint and carbon footprint of grain production. The correction process uses the following recursive formula: in, and are the water-carbon synergistic coefficients at time t+1 and time t, respectively, and are dimensionless. The learning rate is between 0.01 and 0.

1. This represents the total measured greenhouse gas emissions during the period. The values ​​are model estimates, and both are in units of 1 / 2. ; The measured total water consumption within period t. The values ​​are model estimates, and both are in units of 1 / 2. .

8. The water-soil-food resource coupling optimization allocation and greenhouse gas environmental impact assessment system according to claim 1, characterized in that: The water-soil resource spatial allocation scheme output by the dynamic adaptive configuration algorithm module uses a 1:10000 scale gridded map as the output carrier. Each grid cell contains recommended irrigation quotas, recommended fertilization amounts, recommended planting systems, and corresponding estimated greenhouse gas emissions per unit area, and generates a zoning operation instruction file that can be imported into an agricultural machinery navigation system.

9. The water-soil-food resource coupling optimization allocation and greenhouse gas environmental impact assessment system according to claim 1, characterized in that: The greenhouse gas emission inventory data in the data acquisition module includes at least the direct emission factor of nitrous oxide from farmland, the daily methane emission flux from paddy fields, the carbon dioxide emission coefficient from fuel combustion in agricultural machinery, and the annual change rate of soil organic carbon; the time resolution of the emission inventory data is daily, and the spatial resolution is administrative villages or equal-area grids.

10. The water-soil-food resource coupling optimization allocation and greenhouse gas environmental impact assessment system according to claim 1, characterized in that: The data interaction frequency between the dynamic adaptive configuration algorithm module and the multi-factor dynamic coupling model module is once per hour. When an extreme precipitation or drought warning event is triggered, the interaction frequency is automatically encrypted to once every 15 minutes. At the same time, the dynamic adaptive configuration algorithm module temporarily blocks food market fluctuation signals below the set time limit and prioritizes the response to the coupling effect of water stress and sudden changes in greenhouse gas emissions.

Images