Collaborative remediation of farmland pollutants digital twin monitoring system

Abstract

The application discloses a farmland pollutant collaborative remediation digital twin monitoring system and relates to the technical field of digital twin monitoring.The system comprises a multi-source sensing quantification module, an acquisition accuracy regulation module, a multi-source data integration quantification module and an integration matching degree regulation module.The application quantifies the acquisition accuracy by collecting multi-source data accuracy parameters, judges whether to dynamically adjust the sampling point interval and the remediation period, quantifies the spatio-temporal consistency and the fusion efficiency by acquiring the integration matching degree parameters, judges whether to dynamically adjust the lag time interval and the spatial gradient threshold, optimizes the accuracy and efficiency of the farmland pollutant collaborative remediation digital twin monitoring, improves the accuracy of the farmland pollutant collaborative remediation digital twin monitoring, and solves the problem of low accuracy of the farmland pollutant collaborative remediation digital twin monitoring in the prior art.

Description

Technical Field

[0001] This invention relates to the field of digital twin monitoring technology, and in particular to a digital twin monitoring system for the collaborative remediation of farmland pollutants. Background Technology

[0002] In the collaborative remediation of farmland pollutants, the first stage is data acquisition. Utilizing IoT technology, various sensors, such as soil heavy metal sensors, pesticide residue sensors, and pH sensors, are densely deployed in farmland to collect soil pollution data in real time and with high precision. Simultaneously, satellite remote sensing and hyperspectral imaging equipment mounted on drones are used to acquire large-scale, high-resolution images of pollution distribution in farmland, providing a comprehensive data foundation for subsequent analysis. 5G communication technology is employed to rapidly and stably transmit the sensor-collected and remotely-sensed data to the data center, ensuring data timeliness and integrity and avoiding transmission delays or loss.

[0003] In the data modeling phase, based on the collected multi-source data, machine learning algorithms (such as random forests and support vector machines) and physical models (considering the physicochemical processes of pollutant migration and transformation) are used to construct a digital twin model of farmland pollutants. This model can simulate the migration and transformation processes of pollutants among soil, water bodies, and crops, reflecting the actual pollution status of farmland. Through the digital twin model, the simulation results are compared in real time with the actual data fed back by sensors to continuously monitor indicators such as the concentration and distribution range of farmland pollutants. If abnormal data is detected, such as the concentration of a certain pesticide residue exceeding a threshold, the system immediately activates an early warning mechanism to notify relevant personnel. When pollution problems are detected, the digital twin model is used to simulate and extrapolate different remediation schemes, such as simulating the remediation effects of bioremediation (introducing specific microorganisms to degrade pollutants), physical remediation (soil washing, etc.), and chemical remediation (adding chemical agents to solidify pollutants) under different parameter settings. Through comparative analysis, the optimal strategy is provided for actual remediation work, achieving efficient and coordinated remediation of farmland pollutants and ensuring the safety of the farmland ecological environment.

[0004] The above-mentioned technology has at least the following technical problems:

[0005] Existing farmland pollutant monitoring often relies on discrete sampling points, which makes it difficult to cover the spatial heterogeneity of pollutants in farmland. Furthermore, farmland pollutants (such as nitrogen and phosphorus) can migrate with surface runoff and interflow (such as the accumulation of nutrients in nearshore soil due to rainwater runoff). Existing monitoring technologies are mostly static sampling, which cannot track the dynamic migration process of pollutants in the soil-water-vegetation system in real time. In particular, there is a lack of continuous monitoring data on pollutant concentration changes under different hydrological conditions (such as rainfall and irrigation), resulting in insufficient accuracy of multi-source data acquisition.

[0006] The co-remediation of farmland pollutants involves multi-dimensional data, including soil physicochemical data, pollutant concentration data, hydrological data, and vegetation growth data. The collection time and spatial scale of different data vary. Soil sampling data is fixed-point data at a certain moment, while hydrological data is continuous monitoring data at the watershed scale. The two are difficult to match accurately in terms of time and space, and cannot be effectively integrated to reflect the spatiotemporal correlation of pollutant migration. This results in a high degree of difficulty in integrating multi-dimensional data. There is a lack of efficient real-time data transmission and feedback channels between the monitoring system and the digital twin model. Monitoring data must be manually sorted and format converted before it can be input into the model, resulting in a time lag in parameter updates. This leads to the problem of low accuracy in digital twin monitoring of farmland pollutant co-remediation. Summary of the Invention

[0007] This application provides a digital twin monitoring system for the co-remediation of farmland pollutants, which solves the problem of low accuracy in the existing digital twin monitoring of farmland pollutant co-remediation and improves the accuracy of farmland pollutant co-remediation digital twin monitoring.

[0008] On the one hand, a digital twin monitoring system for the collaborative remediation of farmland pollutants is provided, including: a multi-source sensing and quantification module, an acquisition accuracy control module, a multi-source data integration and quantification module, and an integration matching degree control module. The multi-source sensing and quantification module collects multi-source data accuracy parameters for the collaborative remediation of farmland pollutants in the target farmland area. Based on these accuracy parameters, the accuracy of acquiring multi-source data during pollutant monitoring in the target farmland area is quantified to obtain the multi-source data acquisition accuracy of the target farmland area. The acquisition accuracy control module determines whether to perform acquisition accuracy control based on the acquisition accuracy of the multi-source data in the target farmland area. If yes, the multi-source data integration stage is performed after acquisition accuracy control; otherwise, the multi-source data integration stage is performed directly to obtain the acquisition accuracy. The control measures include dynamic adjustment of sampling point spacing and dynamic adjustment of remediation cycle; the multi-source data integration and quantification module is used to obtain the integration matching degree parameter in the multi-source data integration process, and quantifies the consistency and fusion efficiency of multi-source data for co-remediation of pollutants in the target farmland area on a spatiotemporal scale based on the integration matching degree parameter to obtain the multi-source data integration matching degree of the target farmland area; the integration matching degree control module is used to determine whether to perform integration matching degree control based on the multi-source data integration matching degree of the target farmland area. If yes, digital twin monitoring of co-remediation of farmland pollutants is performed while performing integration matching degree control; otherwise, digital twin monitoring of co-remediation of farmland pollutants is performed directly. Integration matching degree control includes dynamic adjustment of lag time interval and dynamic adjustment of spatial gradient threshold.

[0009] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0010] 1. By collecting multi-source data accuracy parameters for the co-remediation of farmland pollutants in the target farmland area, the accuracy of multi-source data acquisition during pollutant monitoring in the target farmland area is quantified based on these parameters. The accuracy of multi-source data acquisition in the target farmland area is then used to determine whether to implement acquisition accuracy control. This allows for timely optimization of acquisition strategies or improvement of data quality when data accuracy is insufficient, ensuring sufficient reliability and representativeness of the data foundation upon which subsequent interflow migration simulation and pollutant risk assessment rely, avoiding simulation distortion or unreliable assessment results due to data accuracy deviations; and integrating multi-source data. The integration matching degree parameter in the process quantifies the consistency and fusion efficiency of multi-source data for the co-remediation of pollutants in the target farmland area across time and space, thus obtaining the integration matching degree of the multi-source data in the target farmland area. Based on the integration matching degree of the multi-source data in the target farmland area, it is determined whether to perform integration matching degree control. This allows for automatic alignment, fusion, or correction when there are significant differences between data, ensuring the overall quality and consistency of the integrated data, improving the compatibility of subsequent model inputs and the accuracy of analysis results, avoiding model bias or decision-making errors caused by data mismatch, and improving the accuracy of digital twin monitoring for the co-remediation of farmland pollutants.

[0011] 2. The system determines whether to implement acquisition accuracy control based on the multi-source data acquisition accuracy of the target farmland area. This avoids unnecessary adjustments that waste resources when data accuracy already meets standards, while also enabling timely detection and correction of insufficient accuracy. This prevents misjudgments of pollution status and mismatched remediation measures due to low-precision data. The system also determines whether to dynamically adjust the sampling point spacing based on the sampling frequency of farmland pollutants. This allows for denser sampling in areas with drastic pollution fluctuations to capture subtle changes, while appropriately widening the spacing in relatively stable areas to reduce monitoring costs. This optimizes sampling resource allocation, improves the spatiotemporal resolution and data representativeness of pollution monitoring, and provides more accurate data support for subsequent pollution source tracing, simulation prediction, and remediation decisions. Furthermore, the system determines whether to dynamically adjust the remediation cycle based on the abundance of soil microbial communities in the target farmland area. This allows for extending the remediation cycle during periods of slow microbial activity recovery to enhance remediation effects, while appropriately shortening the cycle during periods of rapid activity recovery to improve remediation efficiency. This optimizes remediation resource input, accelerates the restoration of soil ecological functions, and enhances the targeting and sustainability of remediation strategies, thereby improving the accuracy of digital twin monitoring for the collaborative remediation of farmland pollutants.

[0012] 3. The system determines whether to implement integration matching degree control based on the multi-source data integration matching degree of the target farmland area. This accurately identifies scenarios where data from different sources meet the requirements for farmland pollution monitoring, remediation simulation, and collaborative control decision-making in terms of spatiotemporal consistency, format compatibility, and logical correlation. This avoids the waste of human and time resources caused by unnecessary control when data integration has reached the collaborative standard (such as repeated data format conversion and spatiotemporal alignment operations), and can promptly identify and correct data integration contradictions. The system also determines whether to implement dynamic adjustment of lag time intervals based on pollutant migration time deviations, ensuring that remediation measures take effect before pollutants reach the target area. This effectively avoids remediation failure or secondary pollution risks due to time lags, improving the accuracy of the remediation process and environmental safety. Finally, the system determines whether to implement dynamic adjustment of spatial gradient thresholds based on the overlap of multi-source data acquisition spatial boundaries, ensuring improved monitoring sensitivity in sparse or unevenly covered areas and optimized allocation of computing resources in densely populated areas. This improves the accuracy of pollutant spatial distribution monitoring and the overall system operating efficiency, thereby enhancing the accuracy of digital twin monitoring for collaborative remediation of farmland pollutants. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic diagram of the structure of the digital twin monitoring system for the collaborative remediation of farmland pollutants provided in an embodiment of this application;

[0015] Figure 2 A flowchart illustrating the dynamic adjustment of the remediation cycle of the digital twin monitoring system for collaborative remediation of farmland pollutants provided in this application embodiment;

[0016] Figure 3 A flowchart illustrating the dynamic adjustment of the lag time interval in the digital twin monitoring system for collaborative remediation of farmland pollutants provided in this application embodiment. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0019] This application provides a digital twin monitoring system for the co-remediation of farmland pollutants, which solves the problem of low accuracy in existing digital twin monitoring systems for the co-remediation of farmland pollutants. The system obtains accuracy by quantifying multi-source data accuracy parameters to determine whether to dynamically adjust the sampling point spacing and remediation cycle; it also obtains integration matching parameters to quantify spatiotemporal consistency and fusion efficiency to determine whether to dynamically adjust the lag time interval and spatial gradient threshold, thereby optimizing the accuracy and efficiency of digital twin monitoring for the co-remediation of farmland pollutants and improving the overall accuracy of this monitoring system.

[0020] The technical solution in this application aims to address the problem of low accuracy in digital twin monitoring for the collaborative remediation of farmland pollutants. The overall approach is as follows:

[0021] By collecting multi-source data accuracy parameters of the co-remediation of farmland pollutants in the target farmland area, the accuracy of multi-source data acquisition during the monitoring process of pollutants in the target farmland area is quantified based on the multi-source data accuracy parameters. The acquisition accuracy of the multi-source data in the target farmland area is then determined based on the acquisition accuracy. If so, the multi-source data integration stage is performed after the acquisition accuracy adjustment; otherwise, the multi-source data integration stage is performed directly. The integration matching degree parameter in the multi-source data integration stage is obtained. Based on the integration matching degree parameter, the consistency and fusion efficiency of the multi-source data for the co-remediation of pollutants in the target farmland area on the spatiotemporal scale are quantified to obtain the integration matching degree of the multi-source data in the target farmland area. The integration matching degree of the multi-source data in the target farmland area is then determined based on the integration matching degree. If so, digital twin monitoring of the co-remediation of farmland pollutants is performed after the integration matching degree adjustment; otherwise, digital twin monitoring of the co-remediation of farmland pollutants is performed directly. This improves the accuracy of digital twin monitoring of the co-remediation of farmland pollutants.

[0022] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0023] like Figure 1The diagram shown is a structural schematic of the digital twin monitoring system for collaborative remediation of farmland pollutants provided in this application embodiment. The digital twin monitoring system for collaborative remediation of farmland pollutants provided in this application embodiment includes: a multi-source sensing and quantification module, an acquisition accuracy control module, a multi-source data integration and quantification module, and an integration matching degree control module.

[0024] As the first module of the digital twin monitoring system for collaborative remediation of farmland pollutants, the multi-source sensing and quantification module is used to collect multi-source data accuracy parameters for collaborative remediation of farmland pollutants in the target farmland area. Based on the multi-source data accuracy parameters, the accuracy of acquiring multi-source data during the monitoring of pollutants in the target farmland area is obtained to obtain the multi-source data acquisition accuracy of the target farmland area.

[0025] It should be noted that the specific steps to obtain the accuracy of multi-source data acquisition for the target farmland area are as follows:

[0026] The multi-source data accuracy parameters include the spatial density of farmland pollutant sampling points, the rate of change of farmland pollutant concentration, and the instantaneous pollutant flux. The spatial density of farmland pollutant sampling points refers to the number of sampling points arranged per unit area. Sampling points are set up within the target farmland area according to preset rules (such as grid method, random method, or gradient method), and the number of sampling points per unit area is counted. The rate of change of farmland pollutant concentration refers to the rate of change of farmland pollutant concentration per unit time. Pollutant concentrations are recorded in real time using in-situ monitoring equipment such as multi-parameter water quality monitors and soil sensors, and the rate of change of farmland pollutant concentration is calculated through data analysis. The instantaneous pollutant flux refers to the total amount of pollutants passing through a cross-section of the target farmland area per unit time. Pollutant concentrations and hydrological data such as flow velocity and flow rate are recorded in real time using automatic monitoring equipment at the monitoring section, such as multi-parameter water quality monitors, soil leachate collectors, and flow meters, and the instantaneous pollutant flux is calculated.

[0027] The spatial density influence value is obtained by combining the results of the relative deviation ratio between the spatial density of farmland pollutant sampling points and the spatial density set value through spatial density correction factor. The result of relative deviation ratio refers to the ratio of the absolute value of the difference between the spatial density of farmland pollutant sampling points and the spatial density set value to the spatial density set value. The combination processing refers to the multiplication operation.

[0028] The results of the analysis of the ratio of the rate of change of pollutant concentration in farmland to the setpoint of the rate of change were combined by the rate of change correction factor to obtain the rate of change influence value; where the ratio analysis refers to the division operation.

[0029] The flux impact value is obtained by combining the results of the analysis of the ratio of instantaneous pollutant flux to flux setpoint by flux correction factor.

[0030] The spatial density influence value, the rate of change influence value, and the flux influence value are superimposed to obtain the multi-source data acquisition accuracy of the target farmland area; where superposition processing refers to addition operation.

[0031] It should be understood that the spatial density correction factor, spatial density setpoint, rate of change correction factor, rate of change setpoint, flux correction factor, and flux setpoint are all obtained from the digital twin monitoring database. Specifically, there is a correlation between the spatial density of farmland pollutant sampling points, the rate of change of farmland pollutant concentration, and instantaneous pollutant flux, as follows: the spatial density of sampling points determines the monitoring network's ability to capture the spatial distribution of pollutants; the density within the reference range more accurately reflects the spatial heterogeneity of pollutants. The rate of change of pollutant concentration reflects the dynamic evolution of pollutants over time; the higher the rate, the more intense the pollutant migration or transformation process, and the higher the requirements for sampling frequency and spatial layout. Instantaneous pollutant flux combines concentration and migration rate, characterizing the total amount of pollutants passing through a certain cross-section per unit time; its change is directly affected by the rate of change of concentration and spatial distribution. There is a positive correlation between the spatial density of farmland pollutant sampling points, the rate of change of farmland pollutant concentration, and the instantaneous pollutant flux and the accuracy of multi-source data acquisition for the target farmland area, as follows: an excessively high sampling point density means that more sensors and acquisition equipment need to be deployed, leading to a significant increase in system construction and maintenance costs; a greater rate of concentration change indicates that pollutants migrate or diffuse faster, which may rapidly spread to the surrounding environment, expand the pollution range, and increase the difficulty of treatment; a greater flux indicates that the total amount of pollutants passing through the cross section per unit time is greater, the pollution load is heavier, and the potential harm to downstream water bodies, soil, or crops is greater.

[0032] As the second module of the digital twin monitoring system for collaborative remediation of farmland pollutants, the acquisition accuracy control module is used to determine whether to perform acquisition accuracy control based on the acquisition accuracy of multi-source data of the target farmland area. If so, the multi-source data integration stage is carried out after the acquisition accuracy control is performed; otherwise, the multi-source data integration stage is carried out directly. Acquisition accuracy control includes dynamic adjustment of sampling point spacing and dynamic adjustment of remediation cycle.

[0033] Furthermore, if the acquisition accuracy of multi-source data for the target farmland area is not greater than the acquisition accuracy reference value, then acquisition accuracy adjustment will not be performed; if the acquisition accuracy of multi-source data for the target farmland area is greater than the acquisition accuracy reference value, then it will be determined whether to perform dynamic adjustment of sampling point spacing based on the sampling frequency of farmland pollutants. If so, then it will be determined whether to perform dynamic adjustment of remediation cycle after the dynamic adjustment of sampling point spacing. If not, then it will be determined directly whether to perform dynamic adjustment of remediation cycle.

[0034] It should be understood that the sampling frequency and the spacing between sampling points are the core regulatory variables for monitoring the spatiotemporal density of farmland pollutants. The two achieve optimal allocation of monitoring resources through a dynamic balance between "temporal coverage" and "spatial coverage." This dynamic adjustment essentially establishes a balance between monitoring costs and data accuracy through complementary adaptation in the spatiotemporal dimensions, ultimately achieving high efficiency and accuracy in farmland pollutant monitoring.

[0035] As a further explanation, the specific steps for determining whether to perform dynamic adjustment of the sampling point spacing are as follows:

[0036] If the sampling frequency of farmland pollutants is greater than or equal to the upper limit of the sampling frequency reference, the sampling frequency correction amount is input into the sampling point spacing mapping table to obtain the spacing control factor. It is then determined whether the spacing control factor is greater than the set value of the spacing control factor. This can accurately identify the problem of time dimension data redundancy caused by excessive sampling frequency, and lay the foundation for balancing the spatiotemporal monitoring density, avoiding resource waste and ensuring monitoring accuracy by adjusting the sampling point spacing. The sampling frequency correction amount represents the difference between the sampling frequency of farmland pollutants and the upper limit of the sampling frequency reference.

[0037] If so, the arithmetic mean of the spacing control factor correction and the acquisition accuracy correction, rounded up, is input into the sampling point spacing mapping table to obtain the positive spacing adjustment coefficient. The positive spacing adjustment coefficient is then combined with the current farmland pollutant sampling point spacing to obtain the adjusted farmland pollutant sampling point spacing. This comprehensively and accurately considers the intensity of spacing control requirements and the deviation of multi-source data acquisition accuracy in scenarios with excessively high sampling frequencies. It scientifically increases the sampling point spacing to eliminate data redundancy in the time dimension, optimize the allocation of monitoring resources, and ensure that the multi-source data acquisition accuracy requirements are still met after adjustment. This achieves a synergy between improved monitoring efficiency and accuracy assurance. The spacing control factor correction represents the positive difference between the spacing control factor and the set value of the spacing control factor, while the acquisition accuracy correction represents the difference between the multi-source data acquisition accuracy of the target farmland area and the acquisition accuracy reference value.

[0038] If not, the arithmetic mean of the spacing control factor control value and the acquisition accuracy correction value is rounded down and input into the sampling point spacing mapping table to obtain the negative spacing adjustment coefficient. The negative spacing adjustment coefficient is combined with the current farmland pollutant sampling point spacing to obtain the adjusted farmland pollutant sampling point spacing. Based on the actual scenario where the sampling frequency exceeds the upper limit but the control demand is weak, as well as the deviation of multi-source data acquisition accuracy, the sampling point spacing is appropriately reduced. This avoids the spatial monitoring blind spot caused by excessively increasing the spacing and affecting the capture of pollutant migration spatial patterns. At the same time, it can balance the data redundancy in the time dimension and the spatial monitoring accuracy through fine spatial sampling density adjustment, ensuring the adaptability of multi-source data in the spatiotemporal dimension. It provides reliable spatial sampling data support for subsequent farmland pollutant monitoring data integration and accurate simulation of digital twin models. The spacing control factor control value represents the negative difference between the spacing control factor and the spacing control factor setting value.

[0039] It should be added that determining whether to perform dynamic adjustment of the sampling point spacing also includes:

[0040] If the sampling frequency of farmland pollutants is within the sampling frequency reference range, dynamic adjustment of the sampling point spacing will not be performed. This can accurately identify scenarios in farmland pollutant monitoring where the sampling frequency in the time dimension is already within a reasonable range and there is no need to balance the spatiotemporal monitoring density by adjusting the sampling point spacing in the spatial dimension. This avoids unnecessary spacing adjustment operations that interfere with the existing stable monitoring system (such as frequent adjustments leading to unstable sampling points and impaired data continuity), while maintaining the current compatibility between the sampling point spacing and the sampling frequency. This ensures the consistency and integrity of multi-source data in the spatiotemporal scale, providing stable basic monitoring data support for subsequent multi-source data integration, pollutant migration pattern analysis, and accurate simulation of digital twin models. At the same time, it reduces the waste of monitoring resources caused by ineffective regulation, achieving a balance between the efficiency and accuracy of farmland pollutant monitoring. The sampling frequency reference range represents the open interval formed by the lower limit and upper limit of the sampling frequency reference.

[0041] If the sampling frequency of farmland pollutants is less than or equal to the lower limit of the sampling frequency reference, the sampling frequency compensation amount is input into the sampling point spacing mapping table to obtain the spacing adjustment factor. It is then determined whether the spacing adjustment factor is greater than the set value of the spacing adjustment factor. This can accurately identify monitoring scenarios where the sampling frequency in the time dimension is insufficient (unable to capture the dynamic migration characteristics of pollutants). At the same time, based on the degree of negative deviation of the sampling frequency, the intensity of the adjustment demand for the spatial dimension sampling point spacing is quantified. This provides a scientific basis for whether to compensate for insufficient time sampling density by adjusting the sampling point spacing (such as reducing the spacing to increase spatial sampling density) and avoid pollutant monitoring omissions or decreased accuracy due to time data gaps. This ensures that the monitoring system can dynamically adapt to the actual problem of insufficient sampling frequency and avoids the waste of resources caused by blind adjustments. It achieves coordinated optimization and accuracy assurance of farmland pollutant monitoring in the spatiotemporal dimensions. The sampling frequency compensation amount represents the difference between the sampling frequency of farmland pollutants and the lower limit of the sampling frequency reference.

[0042] If so, the arithmetic mean of the spacing adjustment factor correction and the acquisition accuracy correction is rounded up and input into the sampling point spacing mapping table to obtain the negative spacing calibration coefficient. The current farmland pollutant sampling point spacing and the negative spacing calibration coefficient are combined to obtain the reduced farmland pollutant sampling point spacing. This can scientifically reduce the sampling point spacing based on the spatial densification requirements (based on the positive difference of the spacing adjustment factor) and the data accuracy guarantee requirements (based on the acquisition accuracy deviation) in scenarios where the precise quantification sampling frequency is lower than the reference lower limit. This allows for a more refined spatial sampling density to compensate for the monitoring gaps caused by insufficient temporal sampling frequency. It avoids missing key features of pollutant migration due to missing temporal data and ensures that the reduced spacing can support the accuracy of multi-source data by integrating accuracy requirements. This provides monitoring data with both spatial integrity and accuracy reliability for subsequent spatiotemporal correlation analysis of pollutants and accurate simulation of digital twin models. It achieves complementary optimization of the spatiotemporal dimensions of farmland pollutant monitoring and improves the decision support capability for remediation. The spacing adjustment factor correction represents the positive difference between the spacing adjustment factor and the set value of the spacing adjustment factor.

[0043] If not, the arithmetic mean of the spacing adjustment factor control and the accuracy correction is rounded down and input into the sampling point spacing mapping table to obtain the positive spacing calibration coefficient. The current farmland pollutant sampling point spacing is combined with the positive spacing calibration coefficient to obtain the adjusted farmland pollutant sampling point spacing. This can appropriately increase the sampling point spacing based on the actual scenario where the sampling frequency is lower than the reference lower limit but the spatial density requirement is weak (based on the negative difference of the spacing adjustment factor) and the accuracy deviation of multi-source data acquisition. This avoids the waste of sampling resources and the increase of data processing redundancy caused by excessively reducing the spacing. At the same time, it can balance the monitoring limitations caused by insufficient time sampling frequency while meeting the accuracy requirements of multi-source data acquisition through reasonable spatial density adjustment. This ensures the optimal resource allocation of the farmland pollutant monitoring system in the spatiotemporal dimension and provides basic data support that is both economical and reliable for subsequent multi-source data integration, pollutant migration law analysis and digital twin model simulation. The spacing adjustment factor control represents the negative difference between the spacing adjustment factor and the set value of the spacing adjustment factor.

[0044] In this embodiment, by dynamically determining whether the sampling frequency of farmland pollutants is within the reference range, exceeds the upper limit, or falls below the lower limit, and combining multi-dimensional parameters such as sampling frequency correction, compensation, spacing control factor, and acquisition accuracy correction, the sampling point spacing is scientifically adjusted upward or downward through arithmetic averaging and mapping table lookup. This accurately balances the sampling density in the time dimension and the spatial monitoring accuracy, effectively avoiding data redundancy or monitoring blind spots, optimizing resource allocation, and ensuring the consistency and integrity of multi-source data in the spatiotemporal scale. This provides efficient, reliable, and economical data support for pollutant migration pattern analysis, accurate simulation of digital twin models, and collaborative remediation decision-making, comprehensively improving the intelligence level and practicality of the farmland pollutant monitoring system.

[0045] It should be understood that the abundance of soil microbial communities in the target farmland area can serve as the core basis for dynamically adjusting the remediation cycle of pollutant reduction because soil microorganisms are the key functional carriers that drive the biodegradation and transformation of farmland pollutants. By dynamically adjusting the remediation cycle based on the abundance of soil microbial communities, the synergistic optimization of "remediation efficiency - ecological protection - resource cost" can be achieved, ensuring that pollutant reduction is both efficient and precise, and avoiding the waste of resources and remediation risks caused by blindly extending or shortening the cycle.

[0046] like Figure 2The diagram shows the dynamic adjustment flowchart of the remediation cycle of the digital twin monitoring system for collaborative remediation of farmland pollutants provided in this application embodiment. The specific logic is as follows: If the abundance of soil microbial communities in the target farmland area is within the abundance reference range, no dynamic adjustment of the remediation cycle is performed. If the abundance of soil microbial communities in the target farmland area is greater than or equal to the upper limit of the abundance reference, the abundance correction amount is input into the remediation cycle mapping table to obtain the cycle adjustment factor. It is then determined whether the cycle adjustment factor is greater than the set value of the cycle adjustment factor. If so, the rounded result of the harmonic average of the cycle adjustment factor control amount and the acquisition accuracy correction amount is input into the remediation cycle mapping table to obtain the correction cycle reduction amount. The current pollutant reduction-type remediation cycle is reduced by the correction cycle reduction amount to obtain the lowered pollutant reduction-type remediation cycle. If not, the rounded result of the harmonic average of the cycle adjustment factor correction amount and the acquisition accuracy correction amount is input into the remediation cycle mapping table to obtain the correction cycle reduction amount. The cycle increase is achieved by superimposing the current pollutant reduction remediation cycle increase with the correction cycle increase. If the abundance of soil microbial communities in the target farmland area is less than or equal to the lower reference limit, the abundance control is entered into the remediation cycle mapping table to obtain the cycle regulation factor. It is then determined whether the cycle regulation factor is greater than the set value of the cycle regulation factor. If so, the rounded result of the harmonic average of the cycle regulation factor compensation and the acquisition accuracy correction is entered into the remediation cycle mapping table to obtain the correction cycle gain. The current pollutant reduction remediation cycle is coupled with the correction cycle gain to obtain the increased pollutant reduction remediation cycle. If not, the rounded result of the harmonic average of the cycle regulation factor control and the acquisition accuracy correction is entered into the remediation cycle mapping table to obtain the correction cycle reduction. The difference between the current pollutant reduction remediation cycle and the correction cycle reduction is processed to obtain the decreased pollutant reduction remediation cycle.

[0047] As further detailed, the specific steps for determining whether to perform dynamic adjustment of the repair cycle are as follows:

[0048] If the abundance of soil microbial communities in the target farmland area is within the abundance reference range, then no dynamic adjustment of the remediation cycle will be performed. This can accurately identify that the soil microbial community is in a "functionally adapted state"—that is, the pollutant degradation efficiency corresponding to the current abundance just matches the remediation target requirements, and there is no need to adjust the cycle to adapt to the intensity of microbial activity. This avoids unnecessary interference with the stability of the existing remediation system caused by remediation cycle adjustments (such as frequent adjustments leading to gaps in the connection of remediation measures and interruptions in the pollutant degradation process), and can maintain a dynamic balance between the remediation cycle and microbial abundance. The abundance reference range represents the open interval formed by the lower limit and upper limit of the abundance reference.

[0049] If the abundance of soil microbial community in the target farmland area is greater than or equal to the upper limit of abundance reference, the abundance correction amount is input into the remediation cycle mapping table to obtain the cycle adjustment factor. It is then determined whether the cycle adjustment factor is greater than the set value of the cycle adjustment factor. This can accurately identify scenarios where the soil microbial community is in a "supersaturated functional state" (i.e., the current abundance far exceeds the minimum requirement to support the remediation target, and the pollutant degradation efficiency is significantly higher than the conventional level). At the same time, the positive deviation of abundance is used to quantify the intensity of the need to shorten the remediation cycle. This provides a scientific basis for whether to compress the remediation cycle to adapt to the high degradation efficiency of microorganisms and avoid "over-remediation" (such as excessive consumption of soil organic matter by microorganisms leading to a decline in soil fertility) or waste of resources (such as continuous investment in unnecessary monitoring and maintenance costs) due to excessively long cycles. This ensures that the remediation system can dynamically respond to the efficiency improvement potential brought about by the excess of microbial abundance, while avoiding blindly adjusting the cycle and destroying the soil ecological balance. This achieves synergistic optimization of pollutant reduction remediation in the three dimensions of "efficiency-cost-ecology". The abundance correction amount represents the positive difference between the abundance of soil microbial community in the target farmland area and the upper limit of abundance reference.

[0050] If so, the harmonic average of the period adjustment factor control value and the accuracy correction value is rounded up and input into the remediation cycle mapping table to obtain the corrected cycle reduction value. The current pollutant reduction remediation cycle is then reduced by the corrected cycle reduction value to obtain the lowered pollutant reduction remediation cycle. This scientifically compresses the remediation cycle based on the need for cycle shortening (based on the positive difference of the period adjustment factor) and the need to ensure data accuracy (based on the accuracy deviation of the acquisition) in scenarios of excessive soil microbial abundance. This fully releases the potential of high degradation efficiency of microorganisms (such as quickly achieving pollutant reduction targets and avoiding excessive consumption of soil organic matter and ecological imbalance caused by excessive cycle length). At the same time, the harmonic average takes into account the constraint of data accuracy on cycle adjustment (ensuring that the remediation effect can still be verified by accurate data after the cycle is shortened). The rounding operation strengthens the effectiveness of cycle reduction and avoids resource waste due to insufficient adjustment. Ultimately, it achieves the synergistic unity of improved pollutant reduction efficiency, cost savings in remediation, and soil ecological protection, providing a precise and efficient cycle adaptation solution for remediation decisions. The period adjustment factor control value represents the positive difference between the period adjustment factor and the set value of the period adjustment factor.

[0051] If not, the harmonic average of the periodic adjustment factor correction and the acquisition accuracy correction is rounded down and input into the remediation cycle mapping table to obtain the corrected cycle increase. The current pollutant reduction remediation cycle is then superimposed with the corrected cycle increase to obtain the adjusted pollutant reduction remediation cycle. This allows for a moderate extension of the remediation cycle, taking into account the actual scenario where soil microbial abundance exceeds the reference upper limit but the need for cycle shortening is weak (based on the negative difference of the periodic adjustment factor), as well as the accuracy deviation of multi-source data acquisition. This avoids incomplete remediation due to excessive cycle shortening (e.g., terminating remediation before pollutants reach a safe threshold, leading to subsequent rebound risks). The harmonic average also takes into account the constraints of data accuracy on cycle adjustment (ensuring that the extended cycle can be fully tracked based on reliable data). At the same time, the rounding operation avoids excessive cycle extension leading to resource waste. Ultimately, this achieves a synergistic balance between the thoroughness of pollutant reduction, the reasonableness of remediation costs, and soil ecological safety. The periodic adjustment factor correction represents the negative difference between the periodic adjustment factor and the set value of the periodic adjustment factor.

[0052] In this embodiment, by dynamically determining whether the abundance of soil microbial communities in the target farmland area is within the reference range, exceeds the upper limit, or falls below the lower limit, and combining multi-dimensional parameters such as abundance correction, periodic adjustment factor, and acquisition accuracy correction, the remediation cycle is scientifically adjusted upward or downward through harmonic averaging and mapping table lookup. This accurately adapts to the dynamic relationship between microbial abundance and pollutant degradation efficiency, effectively avoiding over-remediation or incomplete remediation, optimizing remediation resource allocation, and ensuring the synergistic optimization of pollutant reduction efficiency, cost control, and soil ecological safety. This provides efficient, accurate, and eco-friendly decision support for the collaborative remediation of farmland pollutants.

[0053] As the third module of the digital twin monitoring system for collaborative remediation of farmland pollutants, the multi-source data integration and quantification module is used to obtain the integration matching degree parameter in the multi-source data integration process. Based on the integration matching degree parameter, the consistency and fusion efficiency of the multi-source data for collaborative remediation of pollutants in the target farmland area in the spatiotemporal scale are quantified to obtain the multi-source data integration matching degree of the target farmland area.

[0054] It should be noted that the integrated matching parameters include the acquisition accuracy of multi-source data for the target farmland area to be compared, the pollutant migration path offset, and the overlap of the spatial boundaries of the multi-source data collection. Specifically, the acquisition accuracy of multi-source data for the target farmland area to be compared indicates that if acquisition accuracy adjustment was performed, the accuracy of the target farmland area's multi-source data after the adjustment was recorded as the acquisition accuracy of the target farmland area to be compared; otherwise, the accuracy of the re-acquired target farmland area's multi-source data is recorded as the acquisition accuracy of the target farmland area to be compared. The pollutant migration path offset refers to the degree of spatial deviation between the actual pollutant migration path and the predicted path. This is obtained by using high-resolution remote sensing imagery or UAV aerial photography data, combined with GIS (Geographic Information System) technology to draw the actual pollutant diffusion path, and then performing spatial overlay analysis with hydrological models such as SWAT simulation paths. The overlap of the spatial boundaries of the multi-source data collection refers to the degree of overlap of the multi-source data within the spatial collection area. By using GIS software to overlay and analyze the spatial collection ranges of different data sources (such as the distribution of sampling points, monitoring sections, and the coverage of remote sensing images), the spatial overlap area is calculated, and the overlap degree of the spatial boundaries of multi-source data collection is obtained.

[0055] By combining the results of the accuracy correction factor analysis of the ratio of the accuracy of multi-source data acquisition to the accuracy set value of the target farmland area, the accuracy impact value is obtained; where the ratio analysis refers to the division operation.

[0056] The offset influence value is obtained by combining the results of the analysis of the ratio of pollutant migration path offset to offset set value by offset correction factor.

[0057] The overlap correction factor is used to combine the results of the relative deviation ratio between the overlap degree of the multi-source data acquisition spatial boundary and the overlap degree setting value to obtain the overlap degree influence value; the result of the relative deviation ratio refers to the ratio of the absolute value of the difference between the overlap degree of the multi-source data acquisition spatial boundary and the overlap degree setting value to the overlap degree setting value.

[0058] The accuracy impact value, offset impact value, and overlap impact value are coupled to obtain the multi-source data integration matching degree of the target farmland area. Here, the coupling process refers to the addition operation.

[0059] It is important to understand that the overlap correction factor, overlap setpoint, offset correction factor, offset setpoint, acquisition accuracy correction factor, and acquisition accuracy setpoint are all obtained from the digital twin monitoring database. Specifically, there is a correlation between the acquisition accuracy of multi-source data for the target farmland area to be compared, the offset of pollutant migration paths, and the overlap of the spatial boundaries of multi-source data collection, as follows: The acquisition accuracy of multi-source data for the target farmland area to be compared directly determines the ability of the monitoring data to reflect the actual distribution of pollutants. The higher the accuracy, the lower the accuracy in capturing subtle changes in the pollutant migration path. The offset of the pollutant migration path characterizes the deviation between the actual trajectory of the pollutant and the expected path. The larger the offset, the more significantly the pollutant migration process is affected by external factors (such as rainfall, topography, soil structure, etc.). The overlap of the spatial boundaries of multi-source data collection reflects the degree of consistency of data from different sources in spatial coverage. If the overlap is within the reference range, it indicates that the foundation of data fusion is better, and the more effectively the offset of the pollutant migration path can be identified and corrected. Meanwhile, there is a positive correlation between the accuracy of multi-source data acquisition, pollutant migration path offset, and the overlap of spatial boundaries of multi-source data collection in the target farmland area to be compared, and the degree of integration and matching of multi-source data in the target farmland area. Specifically: when the spatial boundaries of multi-source data collection are highly overlapped, it means that different sensors, remote sensing platforms, or ground monitoring equipment repeatedly collect a large amount of similar or identical data in the same area. This duplication not only causes data redundancy but also leads to a waste of storage, transmission, and processing resources; the higher the accuracy of multi-source data acquisition in the target farmland area to be compared, the greater the error between the multi-source data and the actual pollutant distribution or environmental state; the greater the pollutant migration path offset, the greater the spatial deviation between the actual pollutant migration path and the predicted path.

[0060] The fourth module of the farmland pollutant co-remediation digital twin monitoring system is the integration matching degree control module. It is used to determine whether to perform integration matching degree control based on the integration matching degree of multi-source data of the target farmland area. If yes, the integration matching degree control is performed to carry out farmland pollutant co-remediation digital twin monitoring. If no, the farmland pollutant co-remediation digital twin monitoring is carried out directly. Integration matching degree control includes dynamic adjustment of lag time interval and dynamic adjustment of spatial gradient threshold.

[0061] Furthermore, if the integration matching degree of multi-source data in the target farmland area is not greater than the integration matching degree reference value, then no integration matching degree control will be performed; if the integration matching degree of multi-source data in the target farmland area is greater than the integration matching degree reference value, then it is determined whether to perform dynamic adjustment of the lag time interval based on the pollutant migration time deviation. If so, then it is determined whether to perform dynamic adjustment of the spatial gradient threshold after the dynamic adjustment of the lag time interval; if not, then it is determined directly whether to perform dynamic adjustment of the spatial gradient threshold.

[0062] It should be understood that the core logic of dynamically adjusting the runoff response lag time interval based on pollutant migration time deviation lies in the close causal relationship and dynamic coupling between the two in the process of farmland pollutant runoff migration. The runoff response lag time interval essentially depicts the time difference between "rainfall / irrigation triggering runoff" and "runoff carrying pollutants actually migrating." The rationality of its setting directly determines the timeliness of monitoring and controlling the pollutant migration process, and the pollutant migration time deviation is a key indicator reflecting the matching degree between this setting and the actual migration process. Dynamically adjusting the runoff response lag time interval based on pollutant migration time deviation allows the lag time setting to adapt to the actual rate and pattern of pollutant migration in real time, eliminating the drawback of "disconnect between preset values ​​and actual processes." This ensures that measures such as runoff response monitoring and pollutant interception can accurately match the time nodes of pollutant migration—avoiding pollution control failures due to excessively long lag times and ineffective control due to excessively short lag times. Ultimately, this improves the accuracy and effectiveness of farmland pollutant runoff loss prevention and control, ensuring the stable achievement of remediation goals.

[0063] like Figure 3 The diagram shows the flowchart of the dynamic adjustment of the lag time interval of the digital twin monitoring system for collaborative remediation of farmland pollutants provided in this application embodiment. The specific logic is as follows: If the pollutant migration time deviation is greater than the time deviation reference value, the time deviation correction amount and the integration matching degree correction amount are input into the lag time interval mapping table to obtain the interval adjustment factor. It is determined whether the interval adjustment factor is greater than the interval adjustment factor set value. If so, the interval adjustment factor comparison amount is input into the lag time interval mapping table to obtain the interval reduction amount. Based on the current runoff response lag time interval and the interval reduction amount, a reduction process is performed to obtain the lowered runoff response lag time interval. If not, the interval adjustment factor correction amount is input into the lag time interval mapping table to obtain the interval increase amount. Based on the current runoff response lag time interval and the interval increase amount, an addition process is performed to obtain the uppered runoff response lag time interval. If the pollutant migration time deviation is less than or equal to the time deviation reference value, the dynamic adjustment of the lag time interval is not performed.

[0064] As a further explanation, the specific steps for determining whether to perform dynamic adjustment of the lag time interval are as follows:

[0065] If the pollutant migration time deviation exceeds the time deviation reference value, the time deviation correction and integration matching degree correction are input into the lag time interval mapping table to obtain the interval adjustment factor. It is then determined whether the interval adjustment factor exceeds the set value. This approach accurately identifies scenarios where the pollutant migration time deviates significantly from the preset lag interval (potentially leading to a disconnect between control measures and actual migration) and where there are deviations in the spatiotemporal matching degree of multi-source data. Furthermore, based on the degree of time deviation exceeding the limit and the data integration deviation, the intensity of the need to adjust the runoff response lag time interval is quantified, informing whether subsequent adjustments to the lag interval (such as shortening the interval) are necessary. The timing of the lag adjustment provides a scientific basis for eliminating time deviations and correcting data integration contradictions. It ensures that the adjustment of the lag interval can simultaneously adapt to the pollutant migration pattern and the quality requirements of multi-source data, while avoiding insufficient correlation of the adjusted data due to a single consideration of time deviation. Ultimately, it lays the foundation for the accurate timeliness and data support reliability of farmland pollutant runoff migration control, and reduces the risk of pollution omissions or ineffective control caused by time misalignment. The time deviation correction amount represents the positive difference between the pollutant migration time deviation and the time deviation reference value, and the integration matching degree correction amount represents the difference between the integration matching degree of multi-source data in the target farmland area and the integration matching degree reference value.

[0066] If so, the interval adjustment factor control value is input into the lag time interval mapping table to obtain the interval reduction amount. Based on the current runoff response lag time interval and the interval reduction amount, a reduction process is performed to obtain the lowered runoff response lag time interval. This allows for the scientific compression of the runoff response lag time interval, based on the precise quantification of the intensity of the demand for "shortening the lag interval" in scenarios where pollutant migration time deviation exceeds the limit and multi-source data integration matching degree is biased. This quickly eliminates the control lag problem caused by "actual pollutant migration time being much shorter than the preset lag interval" (such as avoiding the situation where pollutants have migrated with the runoff but interception measures have not been implemented). (This reduces the risk of pollution spread due to delayed implementation of measures), while ensuring that the adjusted lag interval is compatible with the integration and matching degree of multi-source data (such as ensuring that the timestamps of hydrological data and pollutant concentration data are accurately aligned). This fully releases the role of "shortening the interval" in improving the timeliness of prevention and control, while avoiding the disruption of the spatiotemporal correlation of data due to blind adjustments. Ultimately, it achieves precise synchronization of pollutant runoff migration monitoring and control timing, reduces the risk of pollution leakage caused by time misalignment, and ensures the high efficiency of collaborative remediation of farmland pollutants. The interval adjustment factor control value represents the positive difference between the interval adjustment factor and the set value of the interval adjustment factor.

[0067] If not, the interval adjustment factor correction amount is input into the lag time interval mapping table to obtain the interval increase amount. Based on the current runoff response lag time interval and the interval increase amount, the result is processed to obtain the adjusted runoff response lag time interval. In actual scenarios where the pollutant migration time exceeds the deviation reference value but the adjustment demand is weak (based on the negative difference of the interval adjustment factor), the runoff response lag time interval can be appropriately extended. This avoids "premature activation of control measures" due to excessively shortening the interval (such as the early operation of interception facilities before the pollutants have migrated, resulting in resource waste). It also adapts to the multi-source data integration matching degree deviation by extending the interval (such as coordinating the lag interval with the data timestamp matching deviation to avoid the break of the spatiotemporal correlation of the data after adjustment). At the same time, it ensures that the adjusted interval can still cover the actual time range of pollutant migration, preventing monitoring gaps caused by insufficient intervals. Ultimately, it achieves the coordinated adaptation of the timing of pollutant runoff migration control with the actual process and data quality, balances the timeliness of prevention and control with the efficiency of resource utilization, and reduces the risk of ineffective control or control lag. The interval adjustment factor correction amount represents the negative difference between the interval adjustment factor and the set value of the interval adjustment factor.

[0068] If the pollutant migration time deviation is less than or equal to the time deviation reference value, no dynamic adjustment of the lag time interval will be performed. This can accurately identify scenarios where the current runoff response lag time interval is already in sync with the actual pollutant migration process. This avoids unnecessary interval adjustments that could interfere with the stability of the existing pollutant monitoring and control system (such as frequent adjustments leading to confusion in the timing of interception measures and breakage of timestamp correlations between multi-source data). It also maintains a dynamic balance between the lag time interval, pollutant migration patterns, and the need for multi-source data integration. This ensures that within this deviation range, control measures (such as runoff interception and pollutant monitoring) can accurately synchronize with the pollutant migration rhythm (avoiding resource waste due to early activation and pollution leakage due to delayed activation). At the same time, it reduces the waste of manpower and time costs caused by ineffective control, providing a stable time dimension adaptation framework for the prevention and control of farmland pollutant runoff migration. This ensures the continuity of monitoring data and the effectiveness of control measures during the remediation process, ultimately achieving a synergistic unity between the accuracy of pollutant control and the efficiency of resource utilization.

[0069] In this embodiment, by dynamically determining whether the pollutant migration time deviation exceeds the time deviation reference value, and combining the time deviation correction amount with the integration matching degree correction amount to query the lag time interval mapping table, the interval adjustment factor is scientifically determined. Based on the comparison result with the set value, the lag time interval is precisely adjusted downward, upward, or kept unchanged. Thus, when the time deviation exceeds the limit, the control lag is eliminated by shortening the interval to avoid pollution leakage. When the time deviation is reasonable, the interval is extended to optimize resource allocation and prevent data redundancy. At the same time, it is ensured that the adjusted lag interval is always compatible with the integration matching degree of multi-source data. Ultimately, the precise synchronization of pollutant runoff migration monitoring and control timing is achieved, improving the timeliness, data correlation, and decision reliability of farmland pollutant co-remediation, and effectively reducing the risk of pollution diffusion or ineffective control caused by time misalignment.

[0070] It should be understood that the core logic of dynamically adjusting the spatial gradient of interflow infiltration rate based on the overlap of spatial boundaries of multi-source data acquisition lies in the deep coupling relationship between the two in the process of monitoring and simulating interflow in farmland, namely, the "data reliability-parameter accuracy". The spatial gradient of interflow infiltration rate is a key parameter for characterizing the differences in vertical / horizontal infiltration capacity of soil moisture in different regions. Its accuracy directly determines the accuracy of interflow migration path and flux simulation. The overlap of spatial boundaries of multi-source data acquisition (such as the degree of overlap of spatial coverage of different data sources such as soil moisture monitoring points, soil texture sampling areas, and hydrological observation sections) is the core indicator reflecting the synergy and effectiveness of multi-source data in the spatial dimension. Its level directly affects the reliability of the data in characterizing soil infiltration characteristics. By dynamically adjusting the spatial gradient of interflow infiltration rate based on the overlap of spatial boundaries of multi-source data acquisition, the infiltration rate parameter can always match the data reliability. This avoids the loss of details caused by an overly coarse gradient under high overlap, and also prevents the amplification of errors caused by an overly fine gradient under low overlap. Ultimately, this improves the accuracy of interflow migration simulation and provides precise parameter support for farmland water resource management and pollutant leaching risk assessment.

[0071] As further explained in detail, the specific steps for determining whether to perform dynamic adjustment of the spatial gradient threshold are as follows:

[0072] If the overlap of the spatial boundaries of multi-source data acquisition is within the overlap reference range, then no dynamic adjustment of the spatial gradient threshold is performed. This allows for accurate identification of a "cooperative and effective state" in the spatial dimension of multi-source data—that is, under the current overlap, the spatial coverage overlap of different data sources (such as soil texture sampling areas, moisture content monitoring points, and hydrological observation sections) is sufficient to support the accurate characterization of the spatial heterogeneity of soil infiltration characteristics by the spatial gradient threshold of interflow infiltration rate. Furthermore, there is no situation where excessive overlap leads to data redundancy requiring gradient refinement, or excessively low overlap leads to insufficient data support requiring gradient simplification. This avoids unnecessary gradient threshold adjustments that could interfere with the stability of the existing interflow migration simulation system (such as frequent adjustments leading to parameter...). (The data matching relationship is broken, and the continuity of simulation results is damaged), but it can maintain a dynamic balance between the spatial gradient threshold and the reliability of multi-source data. It ensures that within the overlap range, the gradient threshold can reduce parameter errors by relying on cross-validation of multi-source data, and will not increase data processing costs or cause parameter distortion due to excessive adjustment. At the same time, it reduces the waste of human and time resources caused by ineffective regulation. It provides a suitable data support framework for the stable application of the spatial gradient of interflow infiltration rate, ensures the accuracy of interflow migration path and flux simulation, and lays a reliable parameter foundation for the optimal allocation of farmland water resources and accurate assessment of pollutant leaching risk. The overlap reference range represents the open interval formed by the lower limit of the overlap reference and the upper limit of the overlap reference.

[0073] If the overlap of the spatial boundaries of multi-source data acquisition is less than or equal to the lower limit of overlap reference, the overlap control quantity and the integration matching degree correction quantity are input into the spatial gradient mapping table to obtain the gradient control factor. It is then determined whether the gradient control factor is greater than the set value of the gradient control factor. The overlap control quantity represents the negative difference between the overlap of the spatial boundaries of multi-source data acquisition and the lower limit of overlap reference.

[0074] If so, the gradient control factor control value is input into the spatial gradient mapping table to obtain the spatial gradient threshold reduction value. The difference between the current interflow infiltration rate spatial gradient threshold and the spatial gradient threshold reduction value is processed to obtain the reduced interflow infiltration rate spatial gradient threshold. This provides a scientific basis to avoid parameter distortion caused by insufficient data support. It ensures that the gradient adjustment can simultaneously meet the dual requirements of data spatial coverage quality and internal integration quality, and avoids the break in data correlation after adjustment due to a single consideration of overlap. Ultimately, it provides a parameter framework that matches the data reliability for interflow migration simulation, reduces simulation errors caused by insufficient data support and inaccurate gradient thresholds, and ensures the accuracy of farmland water resource management and pollutant leaching risk assessment. The gradient control factor control value represents the positive difference between the gradient control factor and the gradient control factor set value.

[0075] If not, the gradient control factor correction amount is input into the spatial gradient mapping table to obtain the spatial gradient threshold gain amount. The current interflow infiltration rate spatial gradient threshold is then superimposed with the spatial gradient threshold gain amount to obtain the adjusted interflow infiltration rate spatial gradient threshold. Taking into account the actual scenario of data integration matching deviation, the interflow infiltration rate spatial gradient threshold is appropriately expanded. This avoids parameter distortion caused by "insufficient data support but forced gradient refinement" due to excessive threshold reduction (such as dividing the gradient too finely in areas lacking data verification, causing the infiltration rate assignment to be out of touch with the actual soil characteristics). It also adapts to the limitations of data coverage under low overlap by adjusting the threshold (such as using a wider gradient interval to cover data blank areas and reduce simulation errors). At the same time, it ensures that the adjusted threshold can still be coordinated with the multi-source data integration matching degree (avoiding the destruction of the intrinsic correlation of data due to threshold adjustment). Ultimately, the interflow infiltration rate spatial gradient threshold is dynamically adapted to data reliability and integration quality, ensuring the stability of interflow migration path and flux simulation. The gradient control factor correction amount represents the negative difference between the gradient control factor and the gradient control factor setting value.

[0076] It should be added that determining whether to perform dynamic adjustment of the spatial gradient threshold also includes:

[0077] If the overlap of the spatial boundaries of multi-source data acquisition is greater than or equal to the upper limit of the overlap reference, the overlap compensation amount and the integration matching degree correction amount are input into the spatial gradient mapping table to obtain the gradient adjustment factor. It is then determined whether the gradient adjustment factor is greater than the set value of the gradient adjustment factor. The overlap compensation amount represents the positive difference between the overlap of the spatial boundaries of multi-source data acquisition and the upper limit of the overlap reference.

[0078] If so, the gradient adjustment factor correction amount is input into the spatial gradient mapping table to obtain the spatial gradient threshold expansion amount. The current interflow infiltration rate spatial gradient threshold is then superimposed with the spatial gradient threshold expansion amount to obtain the adjusted interflow infiltration rate spatial gradient threshold. This approach can be used to specifically expand the interflow infiltration rate spatial gradient threshold in scenarios where the overlap of the spatial boundaries of multi-source data acquisition is lower than the reference lower limit and the gradient adjustment requirement is strong (based on the positive difference of the gradient adjustment factor), while also taking into account the deviation in data integration matching. This effectively adapts to the limitations of insufficient data spatial coverage under low overlap (e.g., using a wider gradient interval to cover data gaps or contradictory areas, avoiding data support issues). (The lack of support leads to parameter distortion caused by overly fine gradient division). The adjustment intensity is quantified by positive difference to ensure that the threshold expansion range matches the degree of data defects (avoiding simulation errors due to insufficient adjustment or loss of effective data information due to excessive adjustment). At the same time, it ensures that the adjusted threshold is consistent with the multi-source data integration logic. Ultimately, the spatial gradient threshold of interflow infiltration rate is precisely matched with data reliability and adjustment needs, improving the stability and accuracy of interflow migration path and flux simulation. This provides parameter support that meets the actual data support capabilities for the optimal allocation of farmland water resources and accurate assessment of pollutant leaching risk. The gradient adjustment factor correction amount represents the positive difference between the gradient adjustment factor and the gradient adjustment factor set value.

[0079] If not, the gradient adjustment factor comparison value is input into the spatial gradient mapping table to obtain the spatial gradient threshold reduction value. The difference between the current interflow infiltration rate spatial gradient threshold and the spatial gradient threshold reduction value is processed to obtain the reduced interflow infiltration rate spatial gradient threshold. This avoids the loss of effective data information due to "overly coarse gradient division" caused by excessive threshold expansion (such as masking the influence of soil texture differences on infiltration rate in some areas and reducing the adaptability of parameters to spatial heterogeneity). At the same time, by reducing the threshold, parameter details can be preserved as much as possible within the limited range of data support (such as maintaining a finer gradient in local areas with relatively reliable data, balancing data limitations and parameter accuracy requirements). At the same time, it ensures that the adjusted threshold is consistent with the multi-source data integration logic. Ultimately, the interflow infiltration rate spatial gradient threshold forms a dynamic balance with data reliability and adjustment requirements. This ensures that the accuracy of interflow migration simulation is not weakened by oversimplified gradients, and avoids parameter distortion caused by forcibly refining the gradient. The gradient adjustment factor comparison value represents the negative difference between the gradient adjustment factor and the gradient adjustment factor set value.

[0080] In this embodiment, by dynamically determining whether the overlap of the spatial boundaries of multi-source data acquisition is within the reference range, below the lower limit, or above the upper limit, and combining parameters such as overlap ratio, compensation ratio, and integration matching degree correction ratio to query the spatial gradient mapping table, the gradient control factor or adjustment factor is scientifically determined. Based on the comparison results with the set value, the spatial gradient threshold of interflow infiltration rate is precisely adjusted downward, upward, or kept unchanged. Thus, when the overlap is insufficient, the threshold is appropriately increased to avoid parameter distortion caused by insufficient data support. When the overlap is too high, the threshold is appropriately expanded to prevent the gradient division from being too coarse due to data redundancy. At the same time, it is ensured that the adjusted gradient threshold is always compatible with the integration matching degree of multi-source data. Ultimately, a dynamic balance is achieved between the stability, accuracy, and data reliability of interflow migration path and flux simulation, providing a scientific, robust, and practically compliant parameter framework for the optimal allocation of farmland water resources and the assessment of pollutant leaching risk.

[0081] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A digital twin monitoring system for the collaborative remediation of farmland pollutants, characterized in that, It includes a multi-source sensing quantization module, an acquisition accuracy control module, a multi-source data integration quantization module, and an integration matching degree control module: The multi-source sensing and quantification module is used to collect multi-source data accuracy parameters for the collaborative remediation of farmland pollutants in the target farmland area. Based on the multi-source data accuracy parameters, the accuracy of acquiring multi-source data during the monitoring of pollutants in the target farmland area is obtained to obtain the multi-source data acquisition accuracy of the target farmland area. The acquisition accuracy control module is used to determine whether to perform acquisition accuracy control based on the acquisition accuracy of multi-source data of the target farmland area. If yes, the multi-source data integration stage is performed after the acquisition accuracy control is performed; otherwise, the multi-source data integration stage is performed directly. The acquisition accuracy control includes dynamic adjustment of sampling point spacing and dynamic adjustment of repair cycle. The multi-source data integration and quantification module is used to obtain the integration matching degree parameter in the multi-source data integration process. Based on the integration matching degree parameter, the consistency and fusion efficiency of the multi-source data of the co-remediation of pollutants in the target farmland area in the spatiotemporal scale are quantified to obtain the multi-source data integration matching degree of the target farmland area. The integration matching degree control module is used to determine whether to perform integration matching degree control based on the integration matching degree of multi-source data of the target farmland area. If yes, the integration matching degree control is performed to carry out digital twin monitoring of farmland pollutant co-remediation. If no, the digital twin monitoring of farmland pollutant co-remediation is carried out directly. The integration matching degree control includes dynamic adjustment of lag time interval and dynamic adjustment of spatial gradient threshold. The specific steps for obtaining the multi-source data acquisition accuracy for the target farmland area are as follows: The multi-source data accuracy parameters include the spatial density of farmland pollutant sampling points, the rate of change of farmland pollutant concentration, and the instantaneous pollutant flux. The spatial density influence value is obtained by combining the results of the relative deviation between the spatial density of farmland pollutant sampling points and the spatial density set value through spatial density correction factor. The results of the analysis of the ratio of the rate of change of farmland pollutant concentration to the set value of the rate of change were combined by the rate of change correction factor to obtain the rate of change influence value. The flux impact value is obtained by combining the results of the analysis of the ratio of instantaneous pollutant flux to flux setpoint by flux correction factor. The spatial density influence value, the rate of change influence value, and the flux influence value are superimposed to obtain the multi-source data acquisition accuracy of the target farmland area; If the acquisition accuracy of multi-source data for the target farmland area is not greater than the acquisition accuracy reference value, then acquisition accuracy adjustment will not be performed. If the accuracy of multi-source data acquisition for the target farmland area is greater than the reference value, then it is determined whether to perform dynamic adjustment of the sampling point spacing based on the sampling frequency of farmland pollutants. If yes, then it is determined whether to perform dynamic adjustment of the remediation cycle after the dynamic adjustment of the sampling point spacing. If no, then it is determined directly whether to perform dynamic adjustment of the remediation cycle.

2. The digital twin monitoring system for collaborative remediation of farmland pollutants according to claim 1, characterized in that, The specific steps for determining whether to perform dynamic adjustment of the sampling point spacing are as follows: If the sampling frequency of farmland pollutants is greater than or equal to the upper limit of the sampling frequency reference, the sampling frequency correction amount is input into the sampling point spacing mapping table to obtain the spacing control factor, and it is determined whether the spacing control factor is greater than the spacing control factor setting value. The sampling frequency correction amount represents the degree of deviation between the sampling frequency of farmland pollutants and the upper limit of the sampling frequency reference. If so, the arithmetic mean of the spacing control factor correction and the acquisition accuracy correction is rounded up and input into the sampling point spacing mapping table to obtain the positive spacing adjustment coefficient. The positive spacing adjustment coefficient is then combined with the current farmland pollutant sampling point spacing to obtain the adjusted farmland pollutant sampling point spacing. The spacing control factor correction represents the positive deviation of the spacing control factor from the set value, and the acquisition accuracy correction represents the deviation of the multi-source data acquisition accuracy of the target farmland area from the acquisition accuracy reference value. If not, the arithmetic mean of the spacing control factor control quantity and the accuracy correction quantity is rounded down and input into the sampling point spacing mapping table to obtain the negative spacing adjustment coefficient. The negative spacing adjustment coefficient is combined with the current farmland pollutant sampling point spacing to obtain the adjusted farmland pollutant sampling point spacing. The spacing control factor control quantity represents the degree of negative deviation between the spacing control factor and the set value of the spacing control factor.

3. The digital twin monitoring system for collaborative remediation of farmland pollutants according to claim 2, characterized in that, The determination of whether to perform dynamic adjustment of sampling point spacing also includes: If the sampling frequency of farmland pollutants is within the sampling frequency reference range, then the dynamic adjustment of the sampling point spacing will not be performed. The sampling frequency reference range refers to the open interval formed by the lower limit of the sampling frequency reference and the upper limit of the sampling frequency reference. If the sampling frequency of farmland pollutants is less than or equal to the lower limit of the sampling frequency reference, the sampling frequency compensation amount is input into the sampling point spacing mapping table to obtain the spacing adjustment factor, and it is determined whether the spacing adjustment factor is greater than the spacing adjustment factor setting value. The sampling frequency compensation amount represents the degree of negative deviation between the sampling frequency of farmland pollutants and the lower limit of the sampling frequency reference. If so, the arithmetic mean of the spacing adjustment factor correction amount and the acquisition accuracy correction amount is rounded up and input into the sampling point spacing mapping table to obtain the negative spacing calibration coefficient. The current farmland pollutant sampling point spacing and the negative spacing calibration coefficient are combined to obtain the adjusted farmland pollutant sampling point spacing. The spacing adjustment factor correction amount represents the positive deviation of the spacing adjustment factor from the set value of the spacing adjustment factor. If not, the arithmetic mean of the spacing adjustment factor control quantity and the accuracy correction quantity is rounded down and input into the sampling point spacing mapping table to obtain the positive spacing calibration coefficient. The current farmland pollutant sampling point spacing is combined with the positive spacing calibration coefficient to obtain the adjusted farmland pollutant sampling point spacing. The spacing adjustment factor control quantity represents the degree of negative deviation between the spacing adjustment factor and the set value of the spacing adjustment factor.

4. The digital twin monitoring system for collaborative remediation of farmland pollutants according to claim 1, characterized in that, The specific steps for determining whether to perform dynamic adjustment of the repair cycle are as follows: If the abundance of soil microbial community in the target farmland area is within the abundance reference interval, then the dynamic adjustment of the remediation cycle will not be performed. The abundance reference interval refers to the open interval formed by the lower limit of the abundance reference and the upper limit of the abundance reference. If the abundance of soil microbial community in the target farmland area is greater than or equal to the upper limit of abundance reference, the abundance correction amount is input into the remediation cycle mapping table to obtain the cycle adjustment factor, and it is determined whether the cycle adjustment factor is greater than the set value of the cycle adjustment factor. The abundance correction amount represents the degree of positive deviation between the abundance of soil microbial community in the target farmland area and the upper limit of abundance reference. If so, the rounded result of the harmonic average of the periodic adjustment factor control amount and the accuracy correction amount is input into the repair cycle mapping table for querying to obtain the correction cycle reduction amount. The current pollutant reduction repair cycle is reduced by the correction cycle reduction amount to obtain the reduced pollutant reduction repair cycle. The periodic adjustment factor control amount represents the degree of positive deviation between the periodic adjustment factor and the set value of the periodic adjustment factor. If not, the harmonic average of the periodic adjustment factor correction amount and the accuracy correction amount is rounded down and input into the repair cycle mapping table to obtain the correction cycle increase amount. The current pollutant reduction repair cycle is superimposed with the correction cycle increase amount to obtain the adjusted pollutant reduction repair cycle. The periodic adjustment factor correction amount represents the degree of negative deviation between the periodic adjustment factor and the set value of the periodic adjustment factor.

5. The digital twin monitoring system for collaborative remediation of farmland pollutants according to claim 4, characterized in that, The determination of whether to perform dynamic adjustment of the repair cycle also includes: If the abundance of soil microbial community in the target farmland area is less than or equal to the lower limit of abundance reference, the abundance control amount is entered into the remediation cycle mapping table to obtain the cycle regulation factor, and it is determined whether the cycle regulation factor is greater than the set value of the cycle regulation factor. The abundance control amount represents the degree of negative deviation between the abundance of soil microbial community in the target farmland area and the upper limit of abundance reference. If so, the harmonic average of the periodic regulation factor compensation amount and the acquisition accuracy correction amount is rounded down and input into the repair cycle mapping table to obtain the corrected cycle gain amount. The current pollutant reduction repair cycle is coupled with the corrected cycle gain amount to obtain the adjusted pollutant reduction repair cycle. The periodic regulation factor compensation amount represents the degree of positive deviation between the periodic regulation factor and the set value of the periodic regulation factor. If not, the harmonic average of the periodic regulation factor control amount and the accuracy correction amount is rounded up and input into the remediation cycle mapping table to obtain the correction cycle reduction amount. The difference between the current pollutant reduction remediation cycle and the correction cycle reduction amount is processed to obtain the reduced pollutant reduction remediation cycle. The periodic regulation factor control amount represents the degree of negative deviation between the periodic regulation factor and the set value of the periodic regulation factor.

6. The digital twin monitoring system for collaborative remediation of farmland pollutants according to claim 1, characterized in that, The integration matching parameters include the accuracy of multi-source data acquisition for the target farmland area to be compared, the offset of pollutant migration paths, and the overlap of the spatial boundaries of multi-source data collection. By combining the results of the analysis of the ratio of the acquisition accuracy to the acquisition accuracy set value of the multi-source data of the target farmland area under comparison with the acquisition accuracy correction factor, the acquisition accuracy influence value is obtained. The offset influence value is obtained by combining the results of the analysis of the ratio of pollutant migration path offset to offset set value by offset correction factor. The overlap degree influence value is obtained by combining the results of the relative deviation ratio between the overlap degree of the spatial boundary of multi-source data acquisition and the overlap degree set value through the overlap degree correction factor. The accuracy impact value, offset impact value, and overlap impact value are coupled and processed to obtain the multi-source data integration matching degree of the target farmland area. If the integration matching degree of multi-source data for the target farmland area is not greater than the integration matching degree reference value, then no integration matching degree adjustment will be performed; If the integration matching degree of multi-source data for the target farmland area is greater than the integration matching degree reference value, then it is determined whether to perform dynamic adjustment of the lag time interval based on the pollutant migration time deviation. If yes, then it is determined whether to perform dynamic adjustment of the spatial gradient threshold after the dynamic adjustment of the lag time interval. If no, then it is determined directly whether to perform dynamic adjustment of the spatial gradient threshold.

7. The digital twin monitoring system for collaborative remediation of farmland pollutants according to claim 6, characterized in that, The specific steps for determining whether to perform dynamic adjustment of the lag time interval are as follows: If the pollutant migration time deviation is greater than the time deviation reference value, the time deviation correction amount and the integration matching degree correction amount are input into the lag time interval mapping table to obtain the interval adjustment factor. It is then determined whether the interval adjustment factor is greater than the interval adjustment factor set value. The time deviation correction amount represents the degree of positive deviation between the pollutant migration time deviation and the time deviation reference value, and the integration matching degree correction amount represents the degree of deviation between the integration matching degree of the multi-source data of the target farmland area and the integration matching degree reference value. If so, the interval adjustment factor reference value is input into the lag time interval mapping table to obtain the interval reduction amount. Based on the current runoff response lag time interval and the interval reduction amount, the reduction processing is performed to obtain the downgraded runoff response lag time interval. The interval adjustment factor reference value represents the degree of positive deviation between the interval adjustment factor and the interval adjustment factor setting value. If not, the interval adjustment factor correction amount is input into the lag time interval mapping table to obtain the interval increase amount. Based on the current runoff response lag time interval and the interval increase amount, the addition processing is performed to obtain the adjusted runoff response lag time interval. The interval adjustment factor correction amount represents the degree of negative deviation between the interval adjustment factor and the interval adjustment factor setting value. If the pollutant migration time deviation is less than or equal to the time deviation reference value, the dynamic adjustment of the lag time interval will not be implemented.

8. The digital twin monitoring system for collaborative remediation of farmland pollutants according to claim 6, characterized in that, The specific steps for determining whether to perform dynamic adjustment of the spatial gradient threshold are as follows: If the overlap of the spatial boundaries of multi-source data acquisition is within the overlap reference interval, then the dynamic adjustment of the spatial gradient threshold will not be performed. The overlap reference interval represents the open interval formed by the lower limit of the overlap reference and the upper limit of the overlap reference. If the overlap of the spatial boundary of multi-source data acquisition is less than or equal to the lower limit of overlap reference, the overlap comparison quantity and the integration matching degree correction quantity are input into the spatial gradient mapping table to obtain the gradient control factor. It is then determined whether the gradient control factor is greater than the gradient control factor set value. The overlap comparison quantity represents the degree of negative deviation between the overlap of the spatial boundary of multi-source data acquisition and the lower limit of overlap reference. If so, the gradient control factor control amount is input into the spatial gradient mapping table to obtain the spatial gradient threshold reduction amount. The difference between the current soil interflow infiltration rate spatial gradient threshold and the spatial gradient threshold reduction amount is processed to obtain the reduced soil interflow infiltration rate spatial gradient threshold. The gradient control factor control amount represents the degree of positive deviation between the gradient control factor and the gradient control factor set value. If not, the gradient control factor correction amount is input into the spatial gradient mapping table to obtain the spatial gradient threshold gain amount. The current interflow infiltration rate spatial gradient threshold and the spatial gradient threshold gain amount are superimposed to obtain the adjusted interflow infiltration rate spatial gradient threshold. The gradient control factor correction amount represents the degree of negative deviation between the gradient control factor and the set value of the gradient control factor.

9. The digital twin monitoring system for collaborative remediation of farmland pollutants according to claim 8, characterized in that, The determination of whether to perform dynamic adjustment of the spatial gradient threshold also includes: If the overlap of the spatial boundary of multi-source data acquisition is greater than or equal to the upper limit of the overlap reference, the overlap compensation amount and the integration matching degree correction amount are input into the spatial gradient mapping table to obtain the gradient adjustment factor. It is then determined whether the gradient adjustment factor is greater than the set value of the gradient adjustment factor. The overlap compensation amount represents the degree of positive deviation between the overlap of the spatial boundary of multi-source data acquisition and the upper limit of the overlap reference. If so, the gradient adjustment factor correction amount is input into the spatial gradient mapping table to obtain the spatial gradient threshold expansion amount. The current soil interflow infiltration rate spatial gradient threshold is superimposed with the spatial gradient threshold expansion amount to obtain the adjusted soil interflow infiltration rate spatial gradient threshold. The gradient adjustment factor correction amount represents the degree of positive deviation between the gradient adjustment factor and the gradient adjustment factor set value. If not, the gradient adjustment factor reference value is input into the spatial gradient mapping table to obtain the spatial gradient threshold reduction value. The difference between the current interflow infiltration rate spatial gradient threshold and the spatial gradient threshold reduction value is processed to obtain the reduced interflow infiltration rate spatial gradient threshold. The gradient adjustment factor reference value represents the degree of negative deviation between the gradient adjustment factor and the set value of the gradient adjustment factor.

Images