An underground water management and control decision method based on a digital twin model

Abstract

The application discloses a kind of groundwater management and control decision-making methods based on digital twinborn model, it is related to groundwater environment management and control technical field, including: obtaining the hydrogeology of groundwater management and control area, water supply, management and control constraint, groundwater pollution source data and the target pollutant concentration data of toxic and harmful substance online monitoring probe feedback, establish groundwater digital twinborn model and generate current period state benchmark record;Generate current representation record of pollution plume, confirm pollution plume boundary object, migration front and boundary moving direction by short-time feedback measurement task;In combination with pollution plume boundary movement record, groundwater flow direction and groundwater pollution source data, the migration state of pollution plume is inverted, and the migration prediction record of pollution plume is generated;Accordingly generate management and control action generation condition record, carry out groundwater management and control decision-making result by time-sharing simulation calculation and combined simulation recalculation output.The application improves the accuracy of toxic and harmful substance pollution plume migration prediction and groundwater management and control decision-making.

Description

Technical Field

[0001] This invention relates to the field of groundwater environment management technology, and in particular to a groundwater management decision-making method based on a digital twin model. Background Technology

[0002] Groundwater pollution control typically targets industrial sites, chemical industrial parks, mining areas, landfills, oil storage areas, contaminated sites, and areas sensitive to groundwater environments. Current control methods generally involve analyzing the concentrations of heavy metals, volatile organic compounds (VOCs), petroleum pollutants, nitrogenous pollutants, or chlorinated organic compounds in groundwater through well sampling, online water quality monitoring, groundwater level monitoring, hydrogeological data processing, and groundwater simulation model calculations. This analysis combines data on the location, type, release depth, release intensity, pollution source status, groundwater flow direction, extraction well location, recharge point location, and environmental control boundaries to determine if pollutants are spreading. With the development of online monitoring probes, groundwater simulation models, and digital twin display technologies, managers can view the location of monitoring wells, target pollutant concentrations, groundwater levels, groundwater pollution source locations, and environmental control boundaries within the model interface. This provides a data foundation for pollution plume identification, groundwater pollution source verification, extraction control, increased sampling, downstream early warning, and the formulation of control actions.

[0003] However, existing groundwater pollution control methods still have shortcomings: On the one hand, the concentration data of toxic and hazardous substances fed back by online monitoring probes are mostly used to determine whether the concentration at a single point exceeds the limit. There is a lack of a process to process the probe location, monitoring layer, computational grid, aquifer number, groundwater pollution source data, and environmental control boundaries accordingly. This makes it difficult to continuously convert concentration changes into pollution plume boundary objects, pollution plume migration fronts, and boundary movement directions. On the other hand, existing methods usually directly generate extraction, warning, or sampling actions based on the concentration exceeding the limit, without fully utilizing pollution plume boundary movement records, groundwater flow direction, aquifer parameters, groundwater pollution source release records, and groundwater pollution source treatment records within the continuous monitoring period to invert and verify the migration speed, migration direction, and downstream arrival time of toxic and hazardous substance pollution plumes in groundwater. This lack of continuity between control actions and the on-site pollution plume migration status easily affects the accuracy of groundwater control decisions and the effectiveness of environmental control boundary protection. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a groundwater management and control decision-making method based on a digital twin model, which solves the problem that the discontinuous collection of groundwater monitoring deviation objects leads to a lack of unified data constraints in the generation and combined recalculation of management and control actions.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a groundwater management decision-making method based on a digital twin model. The method includes: acquiring hydrogeological data, water intake and recharge data, management constraint data, groundwater pollution source data, and target pollutant concentration data from online monitoring probes for toxic and hazardous substances in the target groundwater management area; establishing a groundwater digital twin model and generating a current periodic state baseline record; generating a current plume characterization record based on the current periodic state baseline record; and confirming the plume boundary objects, plume migration front, and boundary movement direction through a short-time feedback measurement task. The confirmation of the plume boundary objects, plume migration front, and boundary movement direction through the short-time feedback measurement task includes: based on the plume core area objects, pollution... Short-time feedback measurement tasks are generated based on candidate objects for the plume boundary, candidate objects for the plume leading edge, probe validity status, and distances to downstream affected objects. When the online monitoring probe can adjust the sampling interval, encrypted online monitoring records are generated. When online monitoring probes are missing measurements, concentrations exceed their range, or concentrations conflict within the same computational grid, probe calibration tasks or manual sampling tasks are generated. Feedback measurement records are generated after the short-time feedback measurement tasks are completed, and the plume boundary objects, plume migration leading edge, and boundary movement direction are confirmed based on these records. Monitoring points participating in the short-time feedback measurement tasks include plume core area objects, plume boundary candidate objects, plume boundary outer verification monitoring points, groundwater pollution source downstream verification monitoring points, and environmental control boundary monitoring points. The process involves: filtering by point; generating a plume boundary movement record based on the current plume characterization record and feedback measurement record; combining groundwater flow direction, aquifer parameters, and groundwater pollution source data; and calling a plume migration inversion model trained based on historical monitoring samples to generate a plume migration prediction record. The generation of the plume boundary movement record based on the current plume characterization record and feedback measurement record includes: generating a plume boundary movement record based on the plume boundary objects confirmed in the previous monitoring period and the plume boundary objects confirmed by feedback measurement records in the current monitoring period. The plume boundary movement record includes the boundary position in the previous monitoring period, the boundary position in the current monitoring period, the boundary movement distance, the movement time interval, and the boundary movement... The system calculates the direction of movement, the type of target pollutant, and the aquifer number. When the boundary movement distance is less than the error in the installation position of the online monitoring probe, the error in the calculation grid size, or the error in the manual sampling location, no pollution plume migration velocity record is generated; only a boundary movement record to be verified is generated. Based on the pollution plume migration prediction record, groundwater pollution source data, and environmental control boundary impact record, a control action generation condition record is generated, and the control action generation condition record is marked as allowed to generate, pending compensation, pending verification, prohibited from generating, or excluded. The system reads the control action generation condition records in the allowed to generate, pending compensation, or pending verification states, generates a unit control action record, and generates groundwater control decision results through time-segmented simulation calculations and combined simulation recalculations.

[0007] As a preferred embodiment of the groundwater management decision-making method based on a digital twin model as described in this invention, the following steps are included: acquiring hydrogeological data, water intake and recharge data, management constraint data, groundwater pollution source data, and target pollutant concentration data from online monitoring probes for toxic and hazardous substances in the target groundwater management area; establishing a digital twin model of groundwater; and generating a current periodic status baseline record. This includes: determining upstream baseline monitoring points, near-source monitoring points, downstream monitoring points, verification monitoring points outside the pollution plume boundary, and environmental management boundary monitoring points based on the location of groundwater pollution sources, groundwater flow direction, aquifer distribution, well locations, and environmental management boundary locations; matching online monitoring probes for toxic and hazardous substances with monitoring wells, computational grids, aquifer numbers, target pollutant types, and groundwater pollution source data to generate a groundwater object correspondence table.

[0008] As a preferred embodiment of the groundwater management decision-making method based on a digital twin model as described in this invention, the following steps are included: acquiring hydrogeological data, water intake and recharge data, management constraint data, groundwater pollution source data, and target pollutant concentration data from online monitoring probes for toxic and hazardous substances in the target groundwater management area; establishing a digital twin model of groundwater; and generating a current periodic status baseline record. This includes: performing probe status verification, sampling time verification, monitoring layer verification, and concentration validity verification on the target pollutant concentration data from the online monitoring probes for toxic and hazardous substances; writing verified data into the pollutant concentration input record; writing data with missing probe measurements, concentrations exceeding the range, inconsistent monitoring layers, or concentration conflicts within the same computational grid into the monitoring record to be reviewed; and generating the current periodic status baseline record based on the pollutant concentration input record, simulated water level, measured water level, groundwater flow direction, groundwater pollution source correspondence, and environmental management boundary distance.

[0009] As a preferred embodiment of the groundwater management and decision-making method based on a digital twin model as described in this invention, the step of generating a current characterization record of a pollution plume based on the current periodic state baseline record includes: reading the current periodic state baseline record, generating a current grid concentration record according to the target pollutant type, aquifer number, and computational grid number, comparing the current grid concentration record with the grid concentration record of the previous monitoring period to generate a concentration change record, and determining the core area object, boundary candidate object, and leading edge candidate object of the pollution plume based on the pollutant concentration control value, boundary determination range, groundwater pollution source location, and groundwater flow direction.

[0010] As a preferred embodiment of the groundwater management decision-making method based on a digital twin model as described in this invention, the step of calling the pollution plume migration inversion model trained based on historical monitoring samples to generate pollution plume migration prediction records includes: the processor writing pollution plume boundary movement records, online monitoring concentration time series, feedback measurement records, monitoring point spatial locations, groundwater flow direction, aquifer parameters, groundwater pollution source data, and groundwater pollution source release records into the pollution plume migration inversion sample records; calling the pollution plume migration inversion model trained based on historical monitoring samples to generate pollution plume migration speed, pollution plume migration direction, pollution plume boundary prediction location, and downstream impact object arrival time; when the predicted boundary location exceeds the reachable calculation grid range within the predicted time period calculated by the groundwater simulation model, or when the migration direction is inconsistent with the calculation results of the groundwater simulation model, an inversion result pending verification record is generated, and no pollution plume migration prediction record is generated; when the target pollutant type in the groundwater pollution source corresponding record is inconsistent with the target pollutant type fed back by the online monitoring probe, a groundwater pollution source corresponding to pending verification record is generated, and the corresponding inversion result is not entered into the management action generation process.

[0011] As a preferred embodiment of the groundwater management decision-making method based on a digital twin model as described in this invention, the step of generating management action generation condition records based on pollution plume migration prediction records, groundwater pollution source data, and environmental management boundary impact records, and marking these management action generation condition records as allowed to generate, pending compensation, pending review, prohibited from generating, or excluded states, includes: generating core area concentration reduction demand records, pollution plume boundary stabilization demand records, downstream arrival time difference records, groundwater pollution source treatment demand records, and environmental boundary review demand records based on pollution plume migration prediction records and groundwater pollution source data, and generating management action generation condition records; pollution plume migration direction, target pollutant concentration status, groundwater pollution source correspondence, downstream impact object early warning records, and environmental management boundary impact records. If the record of the impact of the control boundary is consistent and does not trigger the requirements for pollution source verification, impact of adjacent control areas, or boundary verification, it is marked as allowed to be generated; if the verification of groundwater extraction control, recharge adjustment, or water intake disturbance may cause changes in groundwater flow direction, expansion of the pollution plume boundary, or impact on the water level of adjacent control areas, it is marked as pending compensation; if the correspondence of groundwater pollution sources, pollution plume boundary objects, status of online monitoring probes, or pollution plume migration prediction records need to be verified, it is marked as pending verification; if there is a lack of action objects, a lack of verification objects, a violation of prohibited action types, or control constraint data that does not allow execution, it is marked as prohibited to be generated; if the action type does not correspond to the pollution plume migration direction, downstream arrival time difference, groundwater pollution source status, or target pollutant concentration status, it is marked as excluded.

[0012] As a preferred embodiment of the groundwater management decision-making method based on a digital twin model as described in this invention, the step of reading the management action generation condition records in the allowed generation, compensation pending, or verification pending states, and generating unit management action records includes: reading the management action generation condition records in the allowed generation, compensation pending, or verification pending states, generating unit management action records, and performing water level change calculation, pollution plume response calculation, adjacent management area impact calculation, groundwater pollution source treatment impact calculation, and environmental management boundary impact calculation on the unit management action records; water level change calculation reads simulated water level, measured water level, extraction flow rate, reinjection volume, and water intake disturbance records, and determines the water level change of the target calculation grid and adjacent calculation grids after the execution of the unit management action; pollution plume response calculation reads... The system records pollution plume migration predictions, target pollutant concentrations, extraction periods, barrier monitoring points, and densified sampling points to determine changes in the core area concentration, boundary location, migration speed, and migration direction of the pollution plume after the execution of unit control actions. It also calculates the impact of groundwater pollution source treatment on the location, release intensity, release duration, and treatment status of groundwater pollution sources to determine the influence of pollution source verification or treatment review actions on the corresponding relationships of pollution sources and the core area concentration of the pollution plume. Simulation results show that when unit control actions result in groundwater pollution source corresponding relationships awaiting verification, water level impacts in adjacent control areas, or environmental control boundary impacts, the processor generates groundwater pollution source verification actions, adjacent control area protection actions, pollutant concentration review actions, action downgrade records, or safe exit records.

[0013] As a preferred embodiment of the groundwater management and decision-making method based on a digital twin model as described in this invention, the generation of groundwater management and decision-making results through time-segmented simulation calculation and combined simulation recalculation includes: generating candidate management and decision-making action combination records based on unit management and decision-making action records calculated through time-segmented simulation; performing relationship restriction verification and combination action constraint verification on the candidate management and decision-making action combination records; relationship restriction verification includes whether there are conflicting actions at the same extraction well, whether sampling tasks are repeatedly generated at the same monitoring point, whether verification tasks are repeatedly generated at the same groundwater pollution source, and whether boundary verification actions have been configured at the same environmental protection management and decision-making boundary; combination action constraint verification includes verification of pollution source correspondence, pollutant concentration control, pollution plume boundary control, downstream impact object early warning coverage, drinking water source protection boundary impact, ecologically sensitive boundary impact, and prohibited diffusion boundary impact; after verification, the groundwater simulation model is called to recalculate the water level change, target pollutant concentration change, pollution plume boundary position change, migration speed change, migration direction change, groundwater pollution source correspondence change, and environmental protection management and decision-making boundary impact after the combined actions are executed, and the groundwater management and decision-making results are generated based on the combined simulation recalculation results.

[0014] The beneficial effects of this invention are as follows: By generating a current characterization record of the pollution plume based on the current periodic state baseline record, and confirming the pollution plume boundary objects, the pollution plume migration front, and the boundary movement direction through a short-time feedback measurement task, the target pollutant concentration data is converted into invertible pollution plume location data, solving the problem of insufficient connection between concentration data and pollution plume boundaries; by calling the pollution plume migration inversion model trained based on historical monitoring samples according to the pollution plume boundary movement record and groundwater pollution source data, a pollution plume migration prediction record is generated, and a control action generation condition record and groundwater control decision result are generated accordingly, so that the control action corresponds to the migration speed, migration direction, groundwater pollution source status, and environmental control boundary impact, thereby improving the accuracy of pollution plume migration control. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of 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.

[0016] Figure 1 This is a flowchart of a groundwater management decision-making method based on a digital twin model.

[0017] Figure 2 Generate a flowchart for the current periodic state baseline record.

[0018] Figure 3 A flowchart for generating current characterization records and short-term feedback measurement tasks for pollution plumes.

[0019] Figure 4 A flowchart for generating pollution plume migration prediction and groundwater management decision-making results. Detailed Implementation

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0021] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0023] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a groundwater management decision-making method based on a digital twin model, comprising the following steps: S1. Acquire hydrogeological data, water intake and recharge data, control and constraint data, groundwater pollution source data, and target pollutant concentration data fed back by online monitoring probes for toxic and hazardous substances in the target groundwater control area, establish a digital twin model of groundwater, and generate a current periodic state baseline record.

[0024] Within the target control area, the processor reads the location of the pollution source, groundwater flow direction, aquifer distribution, existing monitoring well locations, water intake well locations, extraction well locations, drinking water source protection boundaries, ecologically sensitive boundaries, and prohibited diffusion boundaries to determine the deployment locations of online monitoring probes for toxic and hazardous substances. The deployment locations of the online monitoring probes are determined according to the following criteria: upstream baseline monitoring point, near-source monitoring point, downstream groundwater monitoring point, outer boundary verification monitoring point of the pollution plume, and environmental control boundary monitoring point. The upstream baseline monitoring point is used to collect background concentrations; the near-source monitoring point is used to collect concentration changes of the target toxic and hazardous substances; the downstream groundwater monitoring point is used to collect concentration changes after pollutant diffusion; the outer boundary verification monitoring point is used to determine whether the pollution plume boundary has shifted outwards; and the environmental control boundary monitoring point is used to determine whether the pollution plume is approaching the water intake well, drinking water source protection boundary, ecologically sensitive boundary, or prohibited diffusion boundary.

[0025] Online monitoring probes are installed in monitoring wells, intake wells, extraction wells, or dedicated monitoring boreholes. Each online monitoring probe is associated with a probe number, installation location, installation depth, monitoring layer, corresponding aquifer number, and corresponding monitoring well number or dedicated monitoring borehole number. The online monitoring probes are used to continuously collect the concentration of target toxic and hazardous substances in groundwater, while simultaneously collecting data on water level, water temperature, conductivity, pH value, oxidation-reduction potential, sampling time, and probe operating status. The target toxic and hazardous substances are determined based on the pollution sources and environmental control requirements of the target control area. Heavy metal pollutants are detected using electrochemical or ion-selective detection probes; volatile organic compounds, chlorinated organic compounds, or petroleum pollutants are detected using optical or fluorescence detection probes; nitrogen-containing pollutants are detected using ion-selective detection probes or composite water quality detection probes. The processor reads the detection range, detection limits, response time, calibration validity period, and applicable pollutant types of the online monitoring probe. If the target pollutant type is not applicable to the probe, or if the pollutant concentration control value is lower than the probe's detection limit, the probe is not considered a valid online monitoring probe. When there are multiple aquifer monitoring layers in a single monitoring well, probes are set up for each layer or separate probe monitoring records are generated. Concentration data corresponding to different aquifer numbers are not combined for calculation.

[0026] Online monitoring probes generate online monitoring records of toxic and hazardous substance concentrations according to preset sampling intervals. These records include probe number, monitoring well number, installation depth, monitoring layer, corresponding aquifer number, sampling time, target pollutant type, concentration value, water level, water temperature, conductivity, pH value, oxidation-reduction potential, probe operating status, and data source. The sampling interval is determined based on the pollution source type, the rate of change in pollutant concentration over historical monitoring periods, groundwater flow velocity, and the distance between the monitoring point and the environmental control boundary. The sampling intervals for monitoring points near the pollution source, verification monitoring points outside the pollution plume boundary, and monitoring points at the environmental control boundary are shorter than the sampling interval for the upstream baseline monitoring point. When the groundwater digital twin model determines that the pollution plume boundary is close to a water intake well, drinking water source protection boundary, ecologically sensitive boundary, or prohibited diffusion boundary, the sampling interval for the corresponding online monitoring probe is shortened, and a more frequent monitoring record is generated.

[0027] The processor acquires hydrogeological data, groundwater monitoring data, water intake and recharge data, control constraint data, groundwater pollution source data, and pollutant concentration data for the target control area. Hydrogeological data is used to determine aquifer distribution, aquitard distribution, groundwater flow direction, permeability coefficient, specific yield, storage coefficient, and boundary conditions. Groundwater monitoring data is used to determine monitoring well locations, monitoring layers, measured water levels, and manual sampling results. Water intake and recharge data is used to determine water intake well locations, water intake layers, water intake volume, recharge point locations, recharge volume, rainfall infiltration, river and lake recharge, and lateral recharge / discharge. Control constraint data is used to determine water level control lines, pollutant concentration control values, water intake well protection zones, drinking water source protection boundaries, ecologically sensitive boundaries, no-diffusion boundaries, boundary determination ranges, and early warning time requirements. Groundwater pollution source data is used to determine pollution source locations, pollution source types, pollutant types, release depths, release times, release intensities, release durations, historical treatment records, pollution source status, and the correspondence between pollution sources and monitoring wells, computational grids, and aquifers. Pollutant concentration data is used to determine the concentration detected by manual sampling, the historical online monitoring concentration, and the concentration changes over different monitoring periods. The boundary determination range is obtained from the control constraint data; if the control constraint data does not record the boundary determination range, it is determined based on the pollutant concentration control value, the probe detection limit, and the concentration change range over historical monitoring periods.

[0028] The processor establishes a correspondence between online monitoring probes, monitoring wells, computational grids, aquifers, groundwater pollution sources, intake wells, extraction wells, recharge points, downstream affected objects, and environmental control boundaries. The processor first divides the computational grid according to the spatial extent of the target control area, and then divides different aquifers at the same planar location into different computational units according to aquifer numbers. Each online monitoring probe is matched with a computational grid and an aquifer number based on its installation location, installation depth, and monitoring layer. Groundwater pollution sources are matched with corresponding computational grids and aquifer numbers based on pollution source coordinates, release depth, release time, and pollutant type. Intake wells, extraction wells, and recharge points are matched with corresponding computational grids and aquifer numbers based on well location coordinates, affected layer, and operating period. Drinking water source protection boundaries, ecologically sensitive boundaries, and prohibited diffusion boundaries are matched with corresponding computational grids based on boundary coordinates. After the correspondence is established, a groundwater object correspondence table is generated.

[0029] The groundwater object correspondence table supports real-time visualization and migration inversion. This table records the relationships between probe numbers and monitoring well numbers, probe installation layers and aquifer numbers, monitoring wells and computational grids, groundwater pollution sources and computational grids, groundwater pollution sources and target pollutant types, groundwater pollution sources and candidate objects in the plume core area, extraction wells and the plume core area or plume boundary, recharge points and recharge computational grids, potential plume migration pathways and downstream affected objects, and verification monitoring points outside the plume boundary and environmental control boundaries. Relationship status is categorized as valid, pending verification, and deferred. Data with a valid relationship status is entered into the current period's baseline record; data with a pending verification or deferred relationship status is entered into the pending verification monitoring record and does not directly participate in plume migration inversion.

[0030] When establishing a digital twin model of groundwater, the processor uses computational grids and aquifer numbers as units to write hydrogeological data, online monitoring concentration data, manual sampling and detection data, water intake and recharge data, groundwater pollution source data, and control constraint data into the model. Each computational grid in the groundwater digital twin model stores simulated water level, measured water level, groundwater flow direction, aquifer parameters, water intake, recharge volume, pollution source number, pollution source type, target pollutant type, online monitoring concentration value, manual sampling concentration value, and pollutant concentration control value. The groundwater digital twin model displays the current groundwater level, target toxic and hazardous substance concentration, groundwater pollution source location, online monitoring probe location, water intake well location, extraction well location, recharge point location, drinking water source protection boundary, ecologically sensitive boundary, and prohibited diffusion boundary according to the computational grid.

[0031] After the online monitoring probes report new concentration data, the processor writes the data into the corresponding computational grid based on the probe number, sampling time, installation depth, and monitoring layer. The concentration display value of the corresponding computational grid updates with the sampling time, and the groundwater digital twin model synchronously displays areas of increased concentration, decreased concentration, and relatively stable concentration. Calculation grids with concentrations exceeding pollutant concentration control values ​​are displayed as areas exceeding the limit; calculation grids with concentrations reaching the boundary judgment range are displayed as candidate areas for the pollution plume boundary; monitoring points with concentrations below the detection limit are saved as monitoring points on the outer side of the pollution plume. Areas with increased concentration are used to determine candidate objects for the core area of ​​the pollution plume in the next treatment stage, calculation grids reaching the boundary judgment range are used to determine candidate objects for the pollution plume boundary in the next treatment stage, and the environmental control boundary distance is used to determine short-term feedback measurement segments and early warning records for downstream affected objects. The digital twin display results are not used as the sole basis for control actions, but rather together with groundwater flow direction, groundwater pollution source location, aquifer parameters, and environmental control boundaries to form the current cycle state baseline record.

[0032] Before online monitoring probe data is entered into the current cycle's status baseline record, the processor performs data verification. Probe status verification determines whether the probe is operating normally, whether it is within its calibration validity period, and whether there are any disconnection or drift records. Sampling time verification determines whether the sampling time belongs to the current monitoring cycle and whether the sampling times of different probes correspond. Installation layer verification determines whether the probe installation depth and monitoring layer are consistent with the corresponding aquifer number. Concentration validity verification determines whether the concentration value is within the probe's detection range, whether it is below the detection limit, whether there are any abrupt changes, and whether it is significantly inconsistent with historical concentration changes at the same monitoring point. When the concentration value is below the probe's detection limit, it is not considered a concentration exceeding the limit and is only saved as monitoring data from the outer edge of the pollution plume; when the concentration value is above the probe's upper limit, a concentration exceeding the range record is generated, and manual sampling verification is triggered. When the online monitoring probe fails to return concentration data continuously, a probe missing measurement record is generated, and the corresponding calculation grid is marked as requiring supplementary measurement. When the concentration values ​​from multiple online monitoring probes within the same computational grid differ beyond the equipment's allowable error, a grid-wide concentration verification record is generated, and the concentration value from a single probe is not used as the current grid's input value. Data that fails verification is written to the monitoring record awaiting verification and does not directly participate in the current cycle's plume characterization and migration inversion. Data that passes verification is written to the pollutant concentration input record.

[0033] When there is a discrepancy between online monitoring probe data and manual sampling results, the processor reads the probe calibration record, probe drift record, manual sampling results, and sampling time difference. If the manual sampling results and online monitoring probe data belong to the same monitoring layer and the same pollutant type, and the sampling time difference is within the allowable time range, the processor generates a concentration correction record. The concentration correction record includes the concentration value before correction, the concentration value after correction, the basis for correction, the corresponding probe number, the corresponding monitoring well number, the corresponding aquifer number, and the correction effective time. The concentration value before correction is retained in the online monitoring record for toxic and hazardous substances, while the concentration value after correction is written to the pollutant concentration input record. If the manual sampling results and online monitoring probe data do not belong to the same monitoring layer or have inconsistent pollutant types, no concentration correction is performed; only a record indicating mismatched detection data is generated.

[0034] The processor generates a current-cycle state baseline record based on verified online monitoring concentration data, corrected concentration data, simulated groundwater level, measured groundwater level, groundwater flow direction, aquifer parameters, groundwater pollution source correspondence, water intake and recharge disturbance records, and environmental control boundary correspondence. This current-cycle state baseline record records the simulated and measured groundwater levels, target toxic and hazardous substance concentrations, concentration data sources, groundwater flow direction, groundwater pollution source correspondence status, water intake and recharge disturbance status, downstream impact distance, and environmental control boundary distance for each computational grid within the current monitoring cycle. This current-cycle state baseline record serves as the data foundation for current plume characterization, short-term feedback measurements, plume boundary movement record generation, input to the plume migration inversion model trained based on historical monitoring samples, and the generation of control actions.

[0035] After the current cycle state baseline record is generated, the groundwater digital twin model is displayed synchronously. The synchronous display includes groundwater level distribution, target toxic and hazardous substance concentration distribution, online monitoring probe status, groundwater pollution source location, water intake / recharge disturbance location, environmental control boundary location, and downstream affected object location. The concentration distribution results in the current cycle state baseline record are used to determine the current characterization of the pollution plume; the online monitoring probe status is used to determine whether short-term feedback measurements are needed; the groundwater pollution source correspondence is used to determine the relationship between the pollution plume core area and the groundwater pollution source; the water intake / recharge disturbance record is used to determine changes in groundwater flow direction; and the environmental control boundary correspondence is used to determine the positional relationship between the pollution plume migration front and downstream affected objects.

[0036] S2. Generate the current characterization record of the contamination plume based on the current periodic state baseline record, and confirm the boundary objects, migration front, and boundary movement direction of the contamination plume through short-time feedback measurement tasks.

[0037] The processor reads the current cycle state baseline record and extracts from it the computational grid number, aquifer number, monitoring well number, online monitoring probe number, target pollutant type, target toxic and hazardous substance concentration, concentration data source, concentration participation status in the calculation, simulated water level, measured water level, groundwater flow direction, groundwater pollution source correspondence, groundwater pollution source release record, water intake and recharge disturbance record, environmental control boundary correspondence, and distance to downstream affected objects. Downstream affected objects include downstream monitoring wells, verification monitoring points outside the pollution plume boundary, drinking water source protection boundaries, ecologically sensitive boundaries, and prohibited diffusion boundaries. The processor aggregates the concentration data within each computational grid according to the monitoring cycle number, target pollutant type, and aquifer number, ensuring that the same target pollutant type participates in the current characterization of the pollution plume within the same aquifer at the same time baseline.

[0038] The processor uses the concentration data from the current cycle's baseline record to filter valid concentration records for the current cycle. Records with valid concentration data are written into the current cycle's valid concentration record. Records marked as pending verification, requiring supplementary measurement, exceeding the measurement range, or requiring verification within the same grid are not included in the plume boundary confirmation process and are only used as candidate data for short-term feedback measurement tasks. Records with concentration values ​​below the probe's detection limit are not used as evidence of exceeding the limit and are written into the monitoring record outside the plume. Different target pollutant types are processed separately, and records of exceeding the limit, plume boundary candidates, and the current plume characterization results corresponding to different pollutant types are not merged. Concentration data corresponding to different aquifer numbers are processed separately and are not merged across aquifers.

[0039] The processor matches the current period's effective concentration record with the computational grid in the groundwater digital twin model to generate the current grid concentration record. The current grid concentration record includes the computational grid number, aquifer number, monitoring well number, probe number, target pollutant type, current concentration value, data source, sampling time, and participation status in the calculation. When a computational grid corresponds to multiple effective concentration values, the processor prioritizes the concentration value whose sampling time is closest to the current period's reference time, whose probe calibration status is valid, and whose monitoring layer is consistent. If multiple usable concentration values ​​still exist, the processor generates multi-point concentration records within the same grid, retaining the concentration values ​​and corresponding sampling times of each probe, without directly merging different probe concentration values ​​into a single concentration value. When the difference between multiple probe concentration values ​​within the same grid exceeds the equipment's allowable error, a grid-wide concentration verification record is generated. The equipment's allowable error is determined by the online monitoring probe's factory precision, on-site calibration records, or monitoring scheme configuration records.

[0040] The processor compares the current grid concentration record with the grid concentration record of the previous monitoring period to generate a concentration change record. If the current concentration value is higher than the concentration value of the previous monitoring period and the difference exceeds the equipment's allowable error, it is marked as a concentration increase; if the current concentration value is lower than the concentration value of the previous monitoring period and the difference exceeds the equipment's allowable error, it is marked as a concentration decrease; if the difference does not exceed the equipment's allowable error, it is marked as a basically stable concentration. If the previous monitoring period lacks concentration values ​​for the same computational grid, the same aquifer number, and the same target pollutant type, the processor does not generate a concentration change direction, but only generates the first valid concentration record. The concentration change record includes the computational grid number, aquifer number, target pollutant type, current concentration value, previous period concentration value, concentration change amount, concentration change direction, and data source.

[0041] The processor reads the pollutant concentration control values ​​and boundary determination ranges from the control and regulation data, and compares the current grid concentration record with the pollutant concentration control values. Grids whose current concentration exceeds the pollutant concentration control value are recorded as exceeding the limit; grids whose current concentration does not exceed the pollutant concentration control value but reaches the boundary determination range are recorded as pollutant plume boundary candidate objects; monitoring points whose current concentration is below the probe detection limit are recorded as outer-pipe monitoring records. The boundary determination range is obtained from the control and regulation data; if the control and regulation data does not record the boundary determination range, it is determined based on the pollutant concentration control value, the probe detection limit, and the concentration change range over historical monitoring periods. The exceeding-limit records, pollutant plume boundary candidate objects, and outer-pipe monitoring records collectively serve as the data source for the current characterization processing of the pollutant plume.

[0042] The processor generates candidate plume core areas based on concentration exceedance records, concentration change records, groundwater flow direction, groundwater pollution source location, groundwater pollution source release records, and aquifer number. The processor reads concentration exceedance calculation grids for the same target pollutant type and aquifer number. Calculation grids whose concentration values ​​exceed the pollutant concentration control value and are ranked at the top by concentration value within the same target pollutant type and aquifer number, or whose concentration change direction during the continuous monitoring period is marked as increasing, are marked as candidate plume core areas. If a candidate plume core area matches the groundwater pollution source location, release depth, target pollutant type, and groundwater flow direction, it is written into the plume core area object; otherwise, a groundwater pollution source verification record is generated and not directly used as the pollution source input for plume migration inversion.

[0043] The processor generates a set of candidate boundary objects for the plume based on candidate boundary objects, concentration exceedance records, monitoring records outside the plume, and groundwater flow direction. Candidate boundary objects are determined by computational grids that reach the boundary determination range, or by computational grids located between concentration exceedance computational grids and monitoring points outside the plume. The processor sorts the plume core area objects, candidate boundary objects, and monitoring points according to groundwater flow direction, and records the positional relationships between candidate boundary objects and adjacent exceedance computational grids and adjacent monitoring points outside the plume. Candidate boundary objects only represent boundary locations that need to be verified and are not considered confirmed plume boundary objects.

[0044] The processor extracts the object closest to the downstream affected object along the groundwater flow direction from the candidate object set of the plume boundary, generating plume leading edge candidate objects. Each plume leading edge candidate object includes the candidate object number, computational grid number, aquifer number, target pollutant type, corresponding downstream affected object, distance to the downstream affected object, and data source. After confirmation by a short-time feedback measurement task, the plume leading edge candidate objects are written into the plume migration leading edge. When the water intake and recharge disturbance record shows a change in the groundwater flow direction during the current monitoring period, the processor generates a flow direction verification record, and the relevant boundary candidate objects enter the short-time feedback measurement task; the boundary movement direction is not directly generated.

[0045] The processor generates concentration change pointing records based on the location changes of concentration increase records within a continuous monitoring period. These pointing records are determined by the location changes of multiple concentration increase calculation grids within the same target pollutant type and aquifer number. When a concentration increase calculation grid moves downstream towards the affected object along the groundwater flow direction, the pointing record is marked as pointing towards the downstream affected object; when a concentration increase calculation grid recedes towards the groundwater pollution source, the pointing record is marked as receding towards the groundwater pollution source; when the concentration increase calculation grids are scattered in multiple directions and cannot correspond to the groundwater flow direction, a pointing record to be verified is generated. The calculation grid corresponding to the pointing record to be verified enters the short-time feedback measurement task.

[0046] The processor generates the current characterization results of the plume based on the core area objects, the set of candidate objects for the plume boundary, the candidate objects for the plume leading edge, concentration change direction records, the distance to downstream affected objects, and the correspondence with environmental control boundaries. The current characterization results include the target pollutant type, aquifer number, core area objects, set of candidate objects for the plume boundary, candidate objects for the plume leading edge, concentration change direction records, current impact range, downstream affected objects, and recording status. The current impact range is determined based on the core area objects, the set of candidate objects for the plume boundary, groundwater flow direction, distance to downstream affected objects, and environmental control boundaries. Recording status includes "feedbackable measurement," "pending supplementary measurement," and "temporarily suspended." When a candidate object for the plume boundary can establish a positional correspondence with a monitoring record outside the plume, the recording status is marked as "feedbackable measurement"; when the number of online monitoring probes is insufficient to determine the boundary location, it is marked as "pending supplementary measurement"; when data verification fails or the correspondence with the groundwater pollution source is not established, it is marked as "temporarily suspended."

[0047] After generating the current characterization results of the plume, the processor does not directly generate control actions based on the concentration exceeding the limit at a single monitoring point. Instead, it generates a short-term feedback measurement task based on the current characterization results of the plume. The short-term feedback measurement task is used to confirm whether candidate objects for the plume boundary can be written into the plume boundary object, and to confirm whether candidate objects at the plume leading edge and the concentration change pointing record are valid. The processor determines the monitoring points participating in the short-term feedback measurement task based on the plume core area object, the set of plume boundary candidate objects, the plume leading edge candidate objects, the distance to downstream affected objects, the effective status of the probe, and the location to be measured. The monitoring points participating in the feedback measurement are selected from the plume core area object, plume boundary candidate objects, verification monitoring points outside the plume boundary, verification monitoring points downstream of the groundwater pollution source, and environmental control boundary monitoring points.

[0048] The execution method of short-time feedback measurement tasks is determined based on the data status of the monitoring points. When the online monitoring probe is operating normally and the sampling interval can be adjusted, the processor shortens the sampling interval and generates encrypted online monitoring records. When the online monitoring probe is missing data, the concentration exceeds the range, or there is a concentration conflict within the same grid, the processor generates a probe calibration task or a manual sampling task. When there are no monitoring points between the candidate object at the boundary of the pollution plume and the monitoring points on the outer side of the pollution plume, the processor generates encrypted sampling point records at the corresponding locations. The allowable time range for manual sampling tasks is determined by the sampling scheme configuration record, the online monitoring sampling interval, and the sample retention period. During the execution of short-time feedback measurement tasks, the processor records the feedback concentration value, feedback water level value, sampling time, probe operating status, and manual sampling results.

[0049] After the short-term feedback measurement task is completed, the processor generates a feedback measurement record. The feedback measurement record includes the feedback period, the probe number participating in the feedback, the monitoring well number, the computational grid number, the aquifer number, the target pollutant type, the feedback concentration value, the feedback water level value, the concentration change, the change in the candidate boundary location, the change in the distance to downstream affected objects, the probe operating status, and the manual sampling results. The feedback measurement record is used to correct the current characterization results of the pollution plume and serves as data input for pollution plume migration inversion.

[0050] The processor corrects the current characterization results of the plume based on the feedback measurement records. When the feedback concentration value consistently reaches the boundary judgment range, and the corresponding computational grid is located between the concentration exceeding the limit computational grid and the monitoring point outside the plume, the plume boundary candidate object is written into the plume boundary object. When the feedback concentration value is lower than the probe detection limit or does not reach the boundary judgment range, the corresponding boundary candidate object is written into the plume boundary outer verification record. When the feedback measurement records show that the boundary candidate object has moved downstream relative to the affected object direction of the previous monitoring cycle, the boundary movement direction is generated, and the plume leading edge candidate object is written into the plume migration leading edge. When the feedback measurement records cannot confirm the boundary position, the recording status remains pending supplementary measurement, and the encrypted sampling point record is retained.

[0051] When the number of online monitoring probes is insufficient to determine the upstream, downstream, or lateral boundaries of a pollution plume, the processor does not output a complete pollution plume boundary. Instead, it generates a pollution plume boundary supplementary monitoring record. This record includes the location of missing monitoring data, the computational grid number, the aquifer number, the target pollutant type, the reason for supplementary monitoring, and suggested sampling locations. Suggested sampling locations are determined from downstream of areas of increased concentration, outside of candidate pollution plume boundaries, downstream verification locations of groundwater pollution sources, near drinking water source protection boundaries, near ecologically sensitive boundaries, or near prohibited diffusion boundaries. This supplementary monitoring record is used for selecting monitoring points in the next round of short-time feedback measurement tasks.

[0052] The groundwater digital twin model updates the display content of the corresponding computational grid based on the corrected current characterization results of the pollution plume, and uses the core area object, boundary object, migration front, and boundary movement direction of the pollution plume as input data for migration inversion processing. The processor updates the target toxic and hazardous substance concentration, core area object, boundary object, migration front, outer monitoring points of the pollution plume, and distance to downstream affected objects in the corresponding computational grid. The current concentration displays the changes in toxic and hazardous substance concentrations within each computational grid; the pollution plume boundary object displays the current boundary of the pollution plume; the migration front displays the location of the pollution plume closest to downstream affected objects; and the distance to downstream affected objects displays the positional relationship between the pollution plume and downstream monitoring wells, drinking water source protection boundaries, ecologically sensitive boundaries, or prohibited diffusion boundaries.

[0053] The processor writes the corrected current plume characterization results into the current plume characterization record. The current plume characterization record includes the characterization record number, monitoring cycle number, target pollutant type, aquifer number, plume core area object, plume boundary object, plume migration front, boundary movement direction, current impact range, downstream affected objects, plume boundary record number awaiting supplementary measurement, feedback measurement record number, and digital twin display status. The current plume characterization record is used to generate plume boundary movement records, inputs to the plume migration inversion model trained based on historical monitoring samples, and control requirement records. When the current plume characterization results are in a pending or postponed state, the processor does not generate plume boundary movement records, but only retains short-term feedback measurement tasks, encrypted sampling point records, and monitoring records awaiting verification.

[0054] S3. Generate a pollution plume boundary movement record based on the current characterization record and feedback measurement record of the pollution plume. Combine the groundwater flow direction, aquifer parameters and groundwater pollution source data, and call the pollution plume migration inversion model trained based on historical monitoring samples to generate a pollution plume migration prediction record.

[0055] The processor reads the current plume characterization record, feedback measurement record, current cycle status baseline record, historical online monitoring concentration record, historical manual sampling and detection record, historical plume boundary objects, historical groundwater flow direction, historical groundwater pollution source data, groundwater pollution source release record, groundwater pollution source treatment record, groundwater simulation model operation record, and the corresponding relationship of environmental control boundaries. The current plume characterization record provides the plume core area objects, plume boundary objects, plume migration front, boundary movement direction, and downstream affected objects; the feedback measurement record provides the feedback concentration value, feedback water level value, candidate boundary position change, and manual sampling results after the completion of the short-term feedback measurement task; the current cycle status baseline record provides the simulated water level, measured water level, groundwater flow direction, aquifer number, target pollutant type, corresponding status of groundwater pollution source, and distance to the environmental control boundary. The processor matches the read data according to the monitoring cycle number, target pollutant type, and aquifer number. Different target pollutant types are processed separately, and data corresponding to different aquifer numbers are not merged.

[0056] The processor generates a plume boundary movement record based on the current plume characterization record and feedback measurement record. The plume boundary movement record includes the boundary position of the previous monitoring period, the boundary position of the current monitoring period, the boundary movement distance, the movement time interval, the boundary movement direction, the target pollutant type, the aquifer number, the data source, and the recording status. The boundary position of the previous monitoring period is derived from the plume boundary object already confirmed in the previous monitoring period, while the boundary position of the current monitoring period is derived from the plume boundary object confirmed by the feedback measurement record. The boundary movement distance is determined based on the distance between the two boundary positions on the same coordinate datum, the movement time interval is determined based on the sampling time of the two monitoring periods, and the boundary movement direction is determined based on the spatial change direction of the current boundary position relative to the previous boundary position. When the current plume characterization record is in a pending or suspended state, the processor does not generate a plume boundary movement record, but only retains the feedback measurement record, the encrypted sampling point record, and the monitoring record pending verification.

[0057] The processor writes the plume boundary movement record, online monitoring concentration time series, feedback measurement record, monitoring point spatial location, groundwater flow direction, aquifer parameters, groundwater pollution source data, and groundwater pollution source release record into the plume migration inversion sample record. The plume migration inversion sample record is generated unit by unit for the same target pollutant type and the same aquifer number. Each plume migration inversion sample record includes the plume boundary object of the previous monitoring period, the plume boundary object of the current monitoring period, the plume boundary object confirmed by feedback measurement, the plume migration front, target pollutant concentration change, groundwater flow direction, simulated water level change, measured water level change, groundwater pollution source location, groundwater pollution source type, release depth, release intensity, release duration, pollution source status, pollution source treatment status, monitoring period time interval, and distance to downstream affected objects. If the groundwater simulation model operation record lacks simulated water level, groundwater flow direction, aquifer parameters, boundary conditions, or groundwater pollution source correspondence, the processor generates a missing input record for the simulation model and does not use the corresponding data to verify the inversion results.

[0058] The migration speed of the contamination plume is calculated based on the distance traveled from the plume boundary and the time interval between those travels. ; in, Indicates the migration speed of the pollution plume. This indicates the distance traveled by the boundary object of a pollution plume of the same target pollutant type and under the same aquifer number between two monitoring periods. This indicates the time interval between two monitoring cycles. Or, if the sampling time for two monitoring cycles cannot be determined, the processor will not calculate. And generate a migration speed record to be verified.

[0059] When multiple boundary points exist within the same monitoring period for a pollution plume boundary object, the processor prioritizes selecting the boundary point closest to the downstream affected object along the groundwater flow direction to calculate the pollution plume migration velocity. When multiple boundary points are close to different downstream affected objects, boundary movement records and migration velocity records are generated separately. If the boundary movement distance is less than the online monitoring probe installation position error, the calculated grid size error, or the manual sampling positioning error, the processor does not output the migration velocity, but only generates a boundary movement record to be verified. If the boundary movement distance exceeds the sum of the maximum boundary movement distance in the historical monitoring period and the allowable boundary movement error, the processor generates a boundary movement abrupt change record and triggers a probe calibration task or a manual sampling verification task. The allowable boundary movement error is determined by the online monitoring probe installation position error, the calculated grid size error, and the manual sampling positioning error.

[0060] The processor calculates the boundary movement direction based on the coordinate changes of the pollution plume boundary object. The boundary movement direction is represented by the direction from the boundary position in the previous monitoring cycle to the boundary position in the current monitoring cycle. ; in, Indicates the direction of boundary movement. This represents the position vector of the boundary object of the pollution plume in the previous monitoring period. This represents the position vector of the boundary object of the pollution plume in the current monitoring period. This represents the distance between two position vectors. and The same coordinate reference is used in the digital twin model of groundwater. At that time, the processor does not calculate And generate a record of the boundary not being moved.

[0061] The processor compares the boundary movement direction with the groundwater flow direction calculated by the groundwater simulation model. If the boundary movement direction matches the groundwater flow direction, or the deviation is within the allowable deviation range, the boundary movement direction is written into the plume boundary movement record. If the boundary movement direction does not match the groundwater flow direction, and the groundwater pollution source data does not show any new release points, changes in release intensity, or changes in release depth, the processor generates a flow direction inconsistency verification record and does not use the boundary movement direction. When the groundwater pollution source data shows that there are new pollution sources, increased pollution source release intensity, prolonged release duration, or changes in pollution source treatment status during the current monitoring period, the processor writes the groundwater pollution source data into the plume migration inversion sample record and calls the groundwater simulation model to recalculate the groundwater flow direction and pollutant concentration distribution for the corresponding time period.

[0062] The processor invokes a plume migration inversion model trained on historical monitoring samples. The training samples are generated from historical online monitoring concentration records, historical manual sampling records, historical plume boundary objects, historical groundwater flow direction, historical groundwater pollution source data, historical groundwater pollution source release records, historical groundwater pollution source treatment records, and historical downstream impact object arrival records. The input fields for the training samples include the target pollutant concentration time series, monitoring point location, aquifer number, groundwater flow direction, simulated water level change, groundwater pollution source location, groundwater pollution source type, release depth, release intensity, release duration, pollution source status, and boundary movement distance. The output labels for the training samples include plume migration velocity, plume migration direction, predicted boundary location, and downstream impact object arrival time. If the number of historical samples for the same target pollutant type or the same aquifer number is insufficient, the processor does not invoke the plume migration inversion model; instead, it generates prediction records to be verified based only on the plume boundary movement records and the groundwater simulation model.

[0063] The input data for the plume migration inversion model includes online monitoring concentration time series, plume boundary movement records, feedback measurement records, spatial locations of monitoring points, groundwater flow direction, aquifer parameters, groundwater pollution source data, and groundwater pollution source release records output by the groundwater simulation model. The output of the plume migration inversion model includes groundwater pollution source correspondence records, predicted plume boundary locations, plume migration velocity records, plume migration direction records, and predicted arrival times of downstream affected objects. Groundwater pollution source correspondence records include the target pollutant type, groundwater pollution source number, groundwater pollution source location, release depth, release intensity, and the correspondence between the groundwater pollution source and objects in the plume core area. Plume migration velocity records include the migration velocity calculated based on boundary movement distance and time intervals, as well as the migration velocity status verified by groundwater flow direction, aquifer parameters, and groundwater pollution source data. Plume migration direction records include groundwater flow direction, boundary movement direction, groundwater pollution source location direction, and inversion output direction.

[0064] The processor calculates the initial arrival time of the downstream affected objects based on the distance between the leading edge of the contamination plume and the downstream affected objects, the plume migration speed, and the migration direction. ; in, This indicates the predicted time when the migrating front of the pollution plume will reach downstream affected objects. This indicates the distance along the migration direction between the leading edge of the pollution plume and the downstream affected objects. This indicates the migration speed of the contamination plume. The meaning remains the same as in the previous formula. , The processor does not perform calculations when no object is generated or the migration direction does not point to downstream affected objects. It also generates arrival time records to be verified or records of downstream affected objects not in the migration direction.

[0065] The processor verifies the output of the plume migration inversion model using a groundwater simulation model. The groundwater simulation model calculates the reachable computational grid range within the prediction period based on the simulated water level, groundwater flow direction, aquifer parameters, groundwater pollution source location, groundwater pollution source release depth, groundwater pollution source release intensity, and boundary conditions from the current periodic state baseline record. If the predicted plume boundary location output by the inversion model is within the reachable computational grid range within the prediction period, and the plume migration direction is consistent with or deviates from the groundwater flow direction calculated by the groundwater simulation model within an acceptable range, the processor marks the inversion results as usable for generating control actions. If the predicted plume boundary location exceeds the reachable computational grid range within the prediction period, or if the migration direction is inconsistent with the calculation results of the groundwater simulation model, the processor generates an inversion result verification record and triggers encrypted sampling, probe calibration, or manual sampling verification.

[0066] The processor compares the inversion results with the feedback measurement records. If the predicted boundary location does not correspond to the pollution plume boundary object confirmed by the short-term feedback measurement task, a record of predicted boundary mismatch is generated; if the predicted concentration change direction is inconsistent with the feedback concentration change direction, a record of predicted concentration change mismatch is generated; if the target pollutant type in the groundwater pollution source corresponding record is inconsistent with the target pollutant type fed back by the online monitoring probe, a groundwater pollution source corresponding record to be verified is generated. The inversion results corresponding to these records are not included in the control requirement breakdown processing; they are only saved as verification data.

[0067] After the inversion results are verified through the groundwater simulation model and feedback measurement records, the processor updates the groundwater digital twin model with records of plume migration velocity, plume migration direction, predicted plume boundary location, and predicted arrival time of downstream affected objects. The groundwater digital twin model displays the current plume boundary, predicted plume boundary, plume migration front, plume migration velocity, plume migration direction, arrival time of downstream affected objects, distance to downstream monitoring wells, distance to drinking water source protection boundaries, distance to ecologically sensitive boundaries, and distance to prohibited diffusion boundaries, all displayed on a computational grid. The current plume boundary is derived from the current plume characterization records, the predicted plume boundary is derived from the inversion model output, and the migration velocity and direction are derived from the verified plume migration velocity and direction records.

[0068] When the groundwater digital twin model shows that the migration front of a pollution plume is approaching a downstream monitoring well, drinking water source protection boundary, ecologically sensitive boundary, or prohibited diffusion boundary, the processor generates a downstream impact object early warning record and an environmental control boundary impact record. The downstream impact object early warning record includes the downstream impact object number, target pollutant type, predicted arrival time, current distance, migration speed, migration direction, and the reason for triggering the early warning. The environmental control boundary impact record includes the environmental control boundary number, boundary type, predicted boundary location, current boundary distance, and whether a boundary verification action is required.

[0069] The processor writes the verified inversion results into the plume migration prediction record. The plume migration prediction record includes the prediction record number, monitoring cycle number, target pollutant type, aquifer number, plume core area object, plume boundary object, plume migration front, boundary movement record number, migration velocity, migration direction, predicted boundary location, arrival time of downstream affected objects, groundwater simulation model verification results, feedback measurement record verification results, and digital twin display status. The plume migration prediction record serves as the data source for control requirement breakdown and control action generation. If the inversion results fail verification, the processor does not generate a plume migration prediction record; instead, it generates an inversion result pending verification record, a encrypted sampling point record, or a manual sampling verification task.

[0070] S4. Based on the pollution plume migration prediction records, groundwater pollution source data, and environmental protection control boundary impact records, generate control action generation condition records and mark the control action generation condition records as allowed to generate, pending compensation, pending verification, prohibited to generate, or excluded. Read the control action generation condition records in the allowed to generate, pending compensation, or pending verification states, generate unit control action records, and generate groundwater control decision results through time-segmented simulation calculations and combined simulation recalculations.

[0071] The processor reads pollution plume migration prediction records, current cycle status baseline records, current pollution plume characterization records, downstream impact warning records, environmental control boundary impact records, groundwater pollution source data, water intake and recharge data, and control constraint data. The pollution plume migration prediction records provide the target pollutant type, aquifer number, core area objects of the pollution plume, boundary objects of the pollution plume, the migration front, migration velocity, migration direction, predicted boundary location, and arrival time of downstream impact objects. The current cycle status baseline records provide simulated water level, measured water level, groundwater flow direction, effective status of online monitoring probes, water intake and recharge disturbance status, and distance to the environmental control boundary. Groundwater pollution source data provides the pollution source number, pollution source location, pollution source type, pollutant type, release depth, release intensity, release duration, pollution source status, historical treatment records, and pollution source correspondence. The processor matches data according to the monitoring cycle number, target pollutant type, and aquifer number. Different target pollutant types generate separate control actions; data corresponding to different aquifer numbers are not merged.

[0072] If a pollution plume migration prediction record is not generated, or if the verification results of the groundwater simulation model or feedback measurement record in the pollution plume migration prediction record fail, the processor will not generate groundwater extraction control, reinjection adjustment, or water intake / recharge adjustment actions. Instead, it will only generate actions for review sampling, probe calibration, pollution plume boundary verification, groundwater pollution source verification, or early warning of downstream impact objects. If the online monitoring probe corresponding to the pollution plume boundary object is in a state of missing measurement, concentration exceeding range, or awaiting calibration, the processor will mark the relevant control action generation conditions as pending verification and will not directly output control actions that change the groundwater flow field.

[0073] The processor does not directly convert the concentration exceedance results of a single monitoring point into extraction control actions. Instead, it generates a plume control requirement record based on the plume migration prediction record and groundwater pollution source data. This plume control requirement record includes core area concentration reduction requirement records, plume boundary stabilization requirement records, downstream arrival time difference records, groundwater pollution source treatment requirement records, and environmental boundary verification requirement records. The core area concentration reduction requirement record is generated based on the difference between the current concentration value of the object in the plume's core area and the pollutant concentration control value. The plume boundary stabilization requirement record is generated based on the positional change between the predicted boundary location and the current boundary location. The downstream arrival time difference record is generated based on the difference between the arrival time of the downstream affected object and the warning time requirement in the control constraint data. The groundwater pollution source treatment requirement record is generated based on the groundwater pollution source location, pollution source status, release intensity, release duration, and historical pollution source treatment records. The environmental boundary verification requirement record is generated based on the distance between the predicted boundary location and the environmental control boundary.

[0074] The processor generates control action generation condition records based on the pollution plume control requirements. These records include the target pollutant type, aquifer number, target object, action type, trigger data, associated core area objects of the pollution plume, associated boundary objects of the pollution plume, associated groundwater pollution sources, associated downstream affected objects, execution period, verification requirements, and record status. Action types include groundwater pollution source verification, groundwater pollution source treatment verification, pollution plume boundary verification, intensified sampling, groundwater extraction control, barrier monitoring, recharge adjustment, water intake / recharge disturbance verification, downstream affected object early warning, and environmental control boundary verification. Each action type is linked and stored with specific data from the pollution plume migration prediction record, the current periodic status baseline record, groundwater pollution source data, or online monitoring probe records.

[0075] When the target pollutant concentration in the core area of ​​a pollution plume consistently exceeds the pollutant concentration control value, and the location of the groundwater pollution source, the type of target pollutant, the release depth, and the direction of groundwater flow correspond, the processor generates a groundwater pollution source verification action generation condition record and a groundwater extraction control action generation condition record. The groundwater pollution source verification action generation condition record includes the groundwater pollution source number, groundwater pollution source location, pollution source type, target pollutant type, core area of ​​the pollution plume, and the location of the verification sampling. The groundwater extraction control action generation condition record includes the extraction well number, extraction layer, target computational grid, target pollutant type, proposed extraction period, and proposed extraction flow range. The proposed extraction flow range is determined based on the permitted flow rate of the extraction well, historical extraction operation records, the allowable water level drop of the target aquifer, the concentration reduction requirement in the core area of ​​the pollution plume, and the distance to the environmental control boundary.

[0076] When the migration velocity in the pollution plume migration prediction record is higher than the migration velocity of the previous monitoring period, or higher than the upper limit of the migration velocity in the historical monitoring period, the processor generates a record of the conditions for generating early warning actions for downstream affected objects and a record of the conditions for generating pollution plume boundary verification actions. When the distance between the leading edge of the pollution plume migration and the downstream affected object is less than the early warning distance in the control constraint data, or when the arrival time of the downstream affected object is less than the early warning time requirement, the processor generates a record of the conditions for generating early warning actions for downstream affected objects. When the predicted boundary position is close to the drinking water source protection boundary, ecologically sensitive boundary, or prohibited diffusion boundary, the processor generates a record of the conditions for generating barrier monitoring actions and a record of the conditions for generating environmental control boundary verification actions. When the predicted boundary position enters the computational grid corresponding to the prohibited diffusion boundary, the processor does not generate actions for reinjection adjustment or actions that may push the pollution plume to continue moving downstream.

[0077] The processor records control action generation conditions according to their status: allowed generation, pending compensation, pending verification, prohibited generation, or excluded. If the plume migration direction, pollutant concentration exceeding limits, groundwater pollution source correspondence, downstream impact warning records, and environmental control boundary impact records correspond, and do not trigger pollution source verification, adjacent control area impact, or boundary verification requirements, it is marked as allowed generation. If groundwater extraction control, recharge adjustment, or water intake disturbance verification may cause changes in groundwater flow direction, plume boundary expansion, or impact on water levels in adjacent control areas, it is marked as pending compensation. If groundwater pollution source correspondence, plume boundary objects, online monitoring probe status, or migration prediction records require verification, it is marked as pending verification. If there is a missing target object, a missing verification object, a prohibited action type, or control constraint data that does not allow execution, it is marked as prohibited generation. If the action type does not correspond to the plume migration direction, downstream arrival time difference, groundwater pollution source status, or target pollutant concentration status, it is marked as excluded.

[0078] The processor reads the control action generation condition records that are allowed to be generated, pending compensation, or pending review, and generates unit control action records. Unit control action records include action number, action type, affected object, target pollutant type, target monitoring well, target computational grid, associated groundwater pollution source, associated pollution plume boundary object, associated downstream impact object, adjustment amount, execution period, review requirements, and action status. Groundwater extraction control actions include extraction well location, extraction layer, extraction flow rate, extraction period, and target pollutant type. Barrier monitoring actions include barrier monitoring point, monitoring period, and target pollutant type. Downstream impact object early warning actions include early warning object, early warning trigger data, and early warning period. Enhanced sampling actions include enhanced sampling point, sampling layer, sampling period, and target pollutant type. Groundwater pollution source verification actions include groundwater pollution source location, review sampling location, review pollutant type, and review period.

[0079] The processor performs time-segmented simulation calculations on the unit control action records. These calculations include water level change calculations, plume response calculations, impact calculations on adjacent control areas, impact calculations on groundwater pollution source treatment, and environmental control boundary impact calculations. Water level change calculations read simulated water levels, measured water levels, extraction flow rates, reinjection volumes, and water intake disturbance records to determine water level changes in the target and adjacent calculation grids after the unit control action is executed. Plume response calculations read plume migration prediction records, target pollutant concentrations, extraction periods, barrier monitoring points, and densified sampling points to determine changes in plume core area concentration, plume boundary location, migration speed, and migration direction after the unit control action is executed. Groundwater pollution source treatment impact calculations read groundwater pollution source location, release intensity, release duration, and pollution source treatment status to determine the impact of pollution source verification or treatment review actions on the pollution source correspondence and plume core area concentration. Environmental control boundary impact calculations determine whether the unit control action affects drinking water source protection boundaries, ecologically sensitive boundaries, and prohibited diffusion boundaries.

[0080] The time-segmented simulation results show that if, after the execution of a unit control action, the concentration in the core area of ​​the pollution plume is lower than the concentration before the action, or the migration speed of the pollution plume is lower than the migration speed before the action, or the predicted boundary position does not continue to move downstream and will not cause the pollution plume boundary to expand, affect the water level of adjacent control areas, or affect the environmental control boundary, the processor writes the unit control action into the combinatorial action set. The combinatorial action set is the set of unit control action records that can participate in the combined simulation recalculation.

[0081] When unit control actions meet the requirements for controlling the pollution plume but generate groundwater pollution source correspondence that needs verification, impact on adjacent control areas, or pollutant concentration verification requirements, the processor generates groundwater pollution source verification actions, adjacent control area protection actions, or pollutant concentration verification actions. Groundwater pollution source correspondence is generated when it is inconsistent with the core area of ​​the pollution plume. Adjacent control area protection actions are generated when water level changes exceed the allowable range in adjacent control areas. Pollutant concentration verification actions are generated when there is inconsistency between online monitoring concentration data, manual sampling results, or inversion results. The processor pairs and saves the main control actions with pollution source verification actions, protection actions, or verification actions.

[0082] When pollution source verification, protection, or review actions fail to meet requirements, the processor performs action downgrading. Action downgrading proceeds in the following order: reducing single-well extraction flow rate, shortening extraction time, reducing the number of extraction wells, increasing density of sampling points, adding pollution plume boundary review actions, adding groundwater pollution source verification actions, and changing to downstream impact object early warning and boundary review. After each downgrading, the processor re-executes time-segmented simulation calculations. If the downgraded actions still cause the pollution plume boundary to continue approaching the environmental control boundary, or if changes in groundwater flow direction may cause the pollution plume to enter the prohibited diffusion boundary, the processor generates a safe exit record.

[0083] After a safe exit record is generated, the processor does not output control actions that directly alter the groundwater flow field, such as groundwater extraction control or reinjection adjustment. It only retains data on encrypted online monitoring, manual sampling verification, groundwater pollution source verification, pollution plume boundary verification, downstream impact warnings, and environmental control boundary verification. The safe exit record includes the exit action number, exit reason, associated pollution plume migration prediction records, associated groundwater pollution sources, associated environmental control boundaries, retained action type, and verification period, and is saved as part of the groundwater management decision-making results.

[0084] The processor generates candidate control action combination records based on the set of combinable actions. These records include main control actions, pollution source verification actions, review actions, early warning actions, action execution order, action execution time periods, and action combination status. When the calculation grids for multiple unit control actions are located in the same aquifer and their execution time periods overlap, the processor marks the corresponding candidate control action combination record as requiring relationship restriction verification. Relationship restriction verification includes checking for conflicting actions at the same extraction well, duplicate sampling tasks at the same monitoring point, duplicate verification tasks at the same groundwater pollution source, and whether boundary review actions have been configured at the same environmental control boundary. If the relationship restriction verification fails, the processor generates a combination relationship adjustment record and adjusts the action order, deletes duplicate actions, or retains review actions.

[0085] The processor performs combined action constraint verification on candidate control action combination records. Combined action constraint verification includes pollution source correspondence verification, pollutant concentration control verification, pollution plume boundary control verification, downstream impact target early warning coverage verification, drinking water source protection boundary impact verification, ecologically sensitive boundary impact verification, and prohibited diffusion boundary impact verification. Pollution source correspondence verification determines whether the combined action has covered the groundwater pollution source corresponding to the object in the core area of ​​the pollution plume. Pollutant concentration control verification determines whether the concentration in the core area of ​​the pollution plume is lower than the concentration before execution or whether it has entered the re-sampling control range. Pollution plume boundary control verification determines whether the pollution plume boundary continues to move towards downstream impact targets. Downstream impact target early warning coverage verification determines whether objects with predicted arrival times earlier than the early warning time requirement have already generated early warning actions or re-verification actions.

[0086] After the combined action constraint verification is passed, the processor calls the groundwater simulation model to perform a combined simulation recalculation. The combined simulation recalculation uses the simulated water level, measured water level, groundwater flow direction, target pollutant concentration, groundwater pollution source data, current plume characterization record, and plume migration prediction record from the current period state baseline record as initial data to recalculate the changes in water level, target pollutant concentration, plume boundary position, migration velocity, migration direction, groundwater pollution source correspondence, and environmental control boundary impacts after the combined action execution. The combined simulation recalculation does not directly add the calculation results of individual unit control actions. After the combined simulation recalculation is completed, the processor generates a combined simulation recalculation record.

[0087] The processor generates groundwater management decision results based on the combined simulation recalculation records. These results include a list of actions to be performed, execution periods, extracted control parameters, densified sampling points, verification targets, warning targets, plume control results, groundwater pollution source verification results, impact results of drinking water source protection boundaries, impact results of ecologically sensitive boundaries, impact results of prohibited diffusion boundaries, and model version number. When multiple candidate combinations of management actions meet the constraints, the processor sorts them in the following order: the magnitude of concentration reduction in the plume core area, the plume boundary not moving downstream, the complete verification of groundwater pollution source correspondence, minimal impact on adjacent management areas, complete coverage of verification actions, and the fewest number of actions performed. The processor then outputs the groundwater management decision results ranked highest.

[0088] After the groundwater management decision results are generated, the groundwater digital twin model displays the corresponding extraction wells, reinjection points, densified sampling points, groundwater pollution source verification locations, pollution plume boundary verification locations, downstream impact object early warning locations, and environmental management boundary verification locations according to the computational grid. The processor associates and stores the groundwater management decision results with pollution plume migration prediction records, current periodic state baseline records, current pollution plume characterization records, groundwater pollution source data, and combined simulation recalculation records, and records the trigger data, simulation results, and constraint verification results corresponding to each output action.

[0089] In summary, this invention generates a current characterization record of the pollution plume based on the current periodic state baseline record, and confirms the pollution plume boundary objects, the pollution plume migration front, and the boundary movement direction through a short-time feedback measurement task. This converts the target pollutant concentration data into invertible pollution plume location data, solving the problem of insufficient connection between concentration data and the pollution plume boundary. By calling the pollution plume migration inversion model trained based on historical monitoring samples according to the pollution plume boundary movement record, a pollution plume migration prediction record is generated. Based on this, a record of control action generation conditions and groundwater control decision results are generated, so that the control actions correspond to the migration speed, migration direction, and environmental control boundary impact, improving the adaptability of the decision.

[0090] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A groundwater management decision-making method based on a digital twin model, characterized in that, include: Acquire hydrogeological data, water intake and recharge data, control and constraint data, groundwater pollution source data, and target pollutant concentration data fed back by online monitoring probes for toxic and hazardous substances in the target groundwater control area; establish a digital twin model of groundwater; and generate a current periodic state baseline record. Generate a current characterization record of the plume based on the current periodic state baseline record, and confirm the plume boundary objects, plume migration front, and boundary movement direction through a short-time feedback measurement task; The process of confirming the contamination plume boundary object, the contamination plume migration front, and the boundary movement direction through short-time feedback measurement tasks includes: Short-time feedback measurement tasks are generated based on objects in the core area of ​​the plume, candidate objects at the plume boundary, candidate objects at the leading edge of the plume, the effective status of the probe, and the distance to downstream affected objects. The online monitoring probe can generate encrypted online monitoring records when the sampling interval is adjusted. When the online monitoring probe is missing a measurement, the concentration exceeds the range, or there is a concentration conflict in the same calculation grid, a probe calibration task or a manual sampling task is generated. After the short-time feedback measurement task is completed, a feedback measurement record is generated. Based on the feedback measurement record, the boundary objects of the pollution plume, the migration front of the pollution plume, and the direction of boundary movement are confirmed. The monitoring points participating in the short-term feedback measurement task were selected from the core area of ​​the pollution plume, the candidate objects of the pollution plume boundary, the verification monitoring points outside the pollution plume boundary, the verification monitoring points downstream of the groundwater pollution source, and the environmental protection control boundary monitoring points. Based on the current characterization record and feedback measurement record of the pollution plume, a pollution plume boundary movement record is generated. Combined with groundwater flow direction, aquifer parameters and groundwater pollution source data, a pollution plume migration inversion model trained based on historical monitoring samples is called to generate a pollution plume migration prediction record. The process of generating a plume boundary movement record based on the current plume characterization record and feedback measurement record includes: Based on the pollution plume boundary objects confirmed in the previous monitoring cycle and the pollution plume boundary objects confirmed by the feedback measurement records in the current monitoring cycle, a pollution plume boundary movement record is generated. The pollution plume boundary movement record includes the boundary position of the previous monitoring cycle, the boundary position of the current monitoring cycle, the boundary movement distance, the movement time interval, the boundary movement direction, the target pollutant type, and the aquifer number; When the boundary movement distance is less than the installation position error of the online monitoring probe, the calculation grid size error, or the manual sampling positioning error, no pollution plume migration speed record is generated; only a boundary movement record to be verified is generated. Based on the pollution plume migration prediction records, groundwater pollution source data, and environmental protection control boundary impact records, control action generation condition records are generated, and these records are marked as allowed to generate, pending compensation, pending review, prohibited from generating, or excluded. Read the control action generation condition records that are allowed to be generated, pending compensation, or pending review, generate unit control action records, and generate groundwater control decision results through time-segmented simulation calculations and combined simulation recalculations.

2. The groundwater management and decision-making method based on a digital twin model as described in claim 1, characterized in that, The process of acquiring hydrogeological data, water intake and recharge data, control constraint data, groundwater pollution source data, and target pollutant concentration data from online monitoring probes of toxic and hazardous substances in the target groundwater control area, establishing a digital twin model of groundwater, and generating a current periodic state baseline record includes: Based on the location of groundwater pollution sources, groundwater flow direction, aquifer distribution, well locations, and environmental control boundary locations, the following monitoring points are determined: upstream baseline monitoring point, near pollution source monitoring point, downstream groundwater flow monitoring point, outer verification monitoring point of pollution plume boundary, and environmental control boundary monitoring point. The online monitoring probes for toxic and hazardous substances are matched with monitoring wells, computational grids, aquifer numbers, target pollutant types, and groundwater pollution source data to generate a groundwater object correspondence table.

3. The groundwater management and decision-making method based on a digital twin model as described in claim 1 or 2, characterized in that, The process of acquiring hydrogeological data, water intake and recharge data, control constraint data, groundwater pollution source data, and target pollutant concentration data from online monitoring probes of toxic and hazardous substances in the target groundwater control area, establishing a digital twin model of groundwater, and generating a current periodic state baseline record includes: For the target pollutant concentration data fed back by the online monitoring probes for toxic and hazardous substances, the probe status, sampling time, monitoring layer, and concentration validity are verified. Data that passes verification is written into the pollutant concentration input record. Data with missing probe readings, concentrations exceeding the range, inconsistent monitoring layers, or concentration conflicts within the same calculation grid are written into the monitoring record pending verification. The processor generates the current cycle status benchmark record based on the pollutant concentration input record, simulated water level, measured water level, groundwater flow direction, groundwater pollution source correspondence, and environmental control boundary distance.

4. The groundwater management and decision-making method based on a digital twin model as described in claim 3, characterized in that, The process of generating the current characterization record of the pollution plume based on the current periodic state baseline record includes: Read the current period status baseline record, generate the current grid concentration record according to the target pollutant type, aquifer number and calculation grid number, compare the current grid concentration record with the grid concentration record of the previous monitoring period, generate the concentration change record, and determine the core area object, boundary candidate object and leading edge candidate object of the pollution plume according to the pollutant concentration control value, boundary determination range, groundwater pollution source location and groundwater flow direction.

5. The groundwater management and decision-making method based on a digital twin model as described in claim 4, characterized in that, The process of calling the pollution plume migration inversion model trained based on historical monitoring samples to generate pollution plume migration prediction records includes: The processor writes the plume boundary movement record, online monitoring concentration time series, feedback measurement record, monitoring point spatial location, groundwater flow direction, aquifer parameters, groundwater pollution source data and groundwater pollution source release record into the plume migration inversion sample record, calls the plume migration inversion model trained based on historical monitoring samples, and generates plume migration speed, plume migration direction, plume boundary predicted location and downstream impact object arrival time. When the predicted boundary location exceeds the reachable computing grid range within the predicted time period calculated by the groundwater simulation model, or when the migration direction is inconsistent with the calculation results of the groundwater simulation model, an inversion result record for review is generated, and no pollution plume migration prediction record is generated. When the target pollutant type in the record corresponding to the groundwater pollution source is inconsistent with the target pollutant type fed back by the online monitoring probe, a groundwater pollution source record to be verified is generated, and the corresponding inversion result is not included in the generation and processing of control actions.

6. The groundwater management and decision-making method based on a digital twin model as described in claim 5, characterized in that, The process of generating control action generation condition records based on pollution plume migration prediction records, groundwater pollution source data, and environmental control boundary impact records, and marking these records as allowed to generate, pending compensation, pending review, prohibited from generating, or excluded includes: Based on the pollution plume migration prediction records and groundwater pollution source data, the following records are generated: core area concentration reduction demand records, pollution plume boundary stabilization demand records, downstream arrival time difference records, groundwater pollution source treatment demand records, and environmental boundary verification demand records. In addition, control action generation condition records are generated. If the migration direction of the pollution plume, the concentration status of the target pollutant, the correspondence of the groundwater pollution source, the early warning record of the downstream affected object, and the impact record of the environmental protection control boundary can correspond, and the pollution source review, the impact of the adjacent control area, or the boundary review requirement are not triggered, it is marked as allowed to generate; When groundwater extraction control, recharge adjustment, or water intake disturbance review may cause changes in groundwater flow direction, expansion of pollution plume boundaries, or impact on water levels in adjacent control areas, it is marked as pending compensation. When the correspondence between groundwater pollution sources, the boundary objects of pollution plumes, the status of online monitoring probes, or the records of pollution plume migration prediction need to be verified, they should be marked as pending verification. If there is a missing target, a missing verification target, a prohibited action type, or control constraint data that does not allow execution, mark it as prohibited from generation; If the action type does not correspond to the migration direction of the pollution plume, the downstream arrival time difference, the status of the groundwater pollution source, or the concentration status of the target pollutant, it is marked as excluded.

7. The groundwater management and decision-making method based on a digital twin model as described in claim 6, characterized in that, The process of reading the control action generation condition record for the allowed generation, pending compensation, or pending review status includes the following control action record for the generating unit: Read the control action generation condition records that are allowed to be generated, pending compensation, or pending review, generate unit control action records, and perform water level change calculation, pollution plume response calculation, adjacent control area impact calculation, groundwater pollution source treatment impact calculation, and environmental control boundary impact calculation on the unit control action records; The water level change calculation reads the simulated water level, measured water level, extraction flow rate, reinjection volume, and water intake disturbance records to determine the water level changes of the target calculation grid and adjacent calculation grids after the unit control action is executed. The pollution plume response calculation reads the pollution plume migration prediction records, target pollutant concentrations, extraction periods, barrier monitoring points, and densified sampling points to determine the changes in the core area concentration, boundary location, migration speed, and migration direction of the pollution plume after the unit's control actions are implemented. The impact calculation of groundwater pollution source treatment reads the location, release intensity, release duration and treatment status of groundwater pollution sources, and determines the impact of pollution source verification or treatment review actions on the corresponding relationship of pollution sources and the concentration in the core area of ​​the pollution plume; Simulation results show that when unit control actions cause groundwater pollution source correspondence to be verified, water level in adjacent control areas to be affected, or environmental control boundaries to be affected, the processor generates groundwater pollution source verification actions, adjacent control area protection actions, pollutant concentration verification actions, action downgrade records, or safe exit records.

8. The groundwater management and decision-making method based on a digital twin model as described in claim 7, characterized in that, The groundwater management decision results generated through time-segmented simulation calculations and combined simulation recalculations include: Candidate control action combination records are generated based on the unit control action records calculated through time-segmented simulation, and the execution relationship restriction and combination action constraint verification of the candidate control action combination records are performed. Relationship restriction verification includes whether there are conflicting actions for the same extraction well, whether the same monitoring point generates repeated sampling tasks, whether the same groundwater pollution source generates repeated verification tasks, and whether the same environmental protection control boundary has been configured with boundary review actions. The combined action constraint verification includes pollution source correspondence verification, pollutant concentration control verification, pollution plume boundary control verification, downstream impact object early warning coverage verification, drinking water source protection boundary impact verification, ecologically sensitive boundary impact verification, and prohibited diffusion boundary impact verification. After verification, the groundwater simulation model is called to recalculate the changes in water level, target pollutant concentration, pollution plume boundary position, migration velocity, migration direction, groundwater pollution source correspondence, and environmental control boundary impact after the combined actions are executed. Based on the combined simulation recalculation results, groundwater control decision results are generated.

Images