Underground water health risk early warning method and system based on data analysis

Abstract

The invention discloses an underground water health risk early warning method and system based on data analysis, relates to the technical field of underground water health analysis, and realizes quantitative analysis of pollutant accumulation tendency in potential slow flow enrichment area data Pos by establishing a pollutant deposition potential function Fun oriented by a deposition mechanism. Compared with a traditional method which only depends on a pollutant average concentration value to carry out a coarse granularity evaluation mode of static judgment, the method can accurately identify a pollutant deposition amplification phenomenon existing in a local flow velocity low-level area, and further quantifies a composite enrichment degree of pollutants under a micro-scale hydrodynamic structure through an enrichment bias index Ebi. And an underground water pollution deposition risk grade standard and a zoning early warning mechanism are constructed on the basis. According to the scheme, the problems of recognition deficiency and response delay of a slow flow driving type enrichment mechanism in an existing groundwater pollution evaluation means are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of groundwater health analysis, and specifically to a groundwater health risk early warning method and system based on data analysis. Background Art

[0002] In modern water environmental science, water resource security assessment and water quality early warning mechanisms have long been core research areas within environmental monitoring. In particular, groundwater pollution identification, evolution modeling, and health risk early warning are particularly complex due to the spatial concealment and slow nature of underground environmental systems. Within this framework, a specialized subfield, groundwater health risk analysis, has emerged, focusing on identifying water quality fluctuations, long-term exposure to trace pollutants, and their migration patterns that could pose health risks to the public.

[0003] The current mainstream groundwater risk monitoring and early warning methods are still based on concentration threshold judgment and regional mean analysis. The advantages of this type of method are its simple structure and mature application, but when dealing with groundwater environments with significant spatial heterogeneity, it shows a clear lack of local risk identification capabilities. For example, in scenarios where pollution is concentrated in low-flow areas, this traditional method often ignores the pollutant accumulation process caused by the slowdown or near stagnation of water flow in a local area. Since the overall water quality parameters in such areas do not necessarily show a trend of exceeding the standard, they are easily misjudged as "safe areas" in the existing monitoring mechanism.

[0004] Even more serious is that, in actual scenarios, the placement of monitoring wells usually follows average distribution or surface administrative division standards, which further leads to blind spots in the identification of micro-scale pollution points. For example, at the boundaries of industrial plants, sewage recharge zones, or areas of old stratum subsidence, the low flow state of groundwater in local spaces will significantly delay the natural dilution process of pollutants, causing pollutants in the area, such as heavy metals, fluorides, and nitrates, to remain enriched for a long time. However, such changes are not easy to trigger conventional alarm mechanisms due to the slow growth of concentrations. This also leads to deviations in the overall assessment of "groundwater health", which in turn affects downstream links such as water resource scheduling, safe water supply decisions, and public health management. Summary of the Invention

[0005] In response to the deficiencies of the existing technology, the present invention provides a groundwater health risk early warning method and system based on data analysis, which solves the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a groundwater health risk early warning method based on data analysis, comprising the following steps: S1, collect the basic environmental data set Env in the target monitoring area through the groundwater flow velocity sensor; S2. Perform spatial interpolation and isosurface fitting on the basic environmental dataset Env to construct the water velocity map data Vmd and delineate the potential slow-flow enrichment area data Pos; S3. Extract the basic environmental data set Env from the identified potential slow-flow enrichment area data Pos, analyze the change trend in the time dimension, and calculate the pollutant spatial offset rate index Dis; S4. Combine the water velocity map data Vmd, the pollutant spatial offset rate index Dis, and the basic environmental data set Env to establish the pollutant deposition potential function Fun. For each slow-flow enrichment micro-area in the potential slow-flow enrichment area data Pos, the pollutant deposition risk under slow-flow conditions is calculated to obtain the enrichment bias index Ebi. S5. Compare the enrichment bias index Ebi with the preset threshold area, construct the groundwater pollution sedimentation risk level standard for the target monitoring area, and issue an early warning based on the groundwater pollution sedimentation risk level standard.

[0007] Preferably, said S1 includes S11 and S12; S11. Collect groundwater flow velocity sensors and hydrogeological observation equipment deployed in the target monitoring area to collect parameter data related to groundwater hydrodynamics and form a groundwater dynamics dataset Dgw. The groundwater dynamic data set Dgw includes water velocity distribution data Vel, hydraulic gradient distribution data Gra, groundwater permeability data Per and water layer thickness data Thk; Among them, groundwater flow velocity sensors and hydrogeological observation equipment include groundwater micro electromagnetic flow velocity sensors, pressure sensors, high-resolution water level automatic recorders and geological profile conductivity profilers; S12. Collect concentration data of typical pollutants by deploying water quality testing equipment, obtain pollutant concentration monitoring data at the same spatial location in the target monitoring area, and form a groundwater pollution index dataset Dpc. This is then integrated with the groundwater dynamics dataset Dgw to obtain the basic environment dataset Env. The groundwater pollution index dataset Dpc includes nitrate concentration data Con1, arsenic concentration data Con2 and lead concentration data Con3; Among them, water quality testing equipment includes online ultraviolet absorption water quality analyzer, groundwater arsenic-specific electrochemical analyzer and portable heavy metal ion detector.

[0008] Preferably, said S2 includes S21; S21, extracting the water velocity distribution data Vel from the basic environmental dataset Env, constructing a two-dimensional regular spatial grid with the spatial coordinate system of the target monitoring area as a reference, using each grid point in the two-dimensional regular spatial grid as an observation point, and then executing the Kriging interpolation method to construct a continuous velocity prediction function based on the water velocity distribution data Vel at the corresponding position of the observation point, for estimating the groundwater velocity value at any observation point; The water flow velocity value is mapped to the entire two-dimensional regular space grid to form a continuously distributed two-dimensional flow velocity field, and a flow velocity isosurface model is generated based on this. The flow velocity isosurface model divides the velocity change characteristics in the area by isovelocity value lines to generate water flow velocity map data Vmd covering the target area.

[0009] Preferably, said S2 further includes S22; S22, performing statistical calculations on the groundwater velocity values ​​at all observation points in the water velocity map data Vmd to obtain a regional average velocity parameter Vavg of the target monitoring area within the target monitoring period, which is used to reflect the baseline level of groundwater velocity in the target monitoring area; Taking 30% of the regional average velocity parameter Vavg as the slow flow determination threshold, a traversal judgment is performed on each observation point in the water velocity map data Vmd, and all observation points with predicted water velocity values ​​lower than the determination threshold are screened out, and the observation points are marked as slow flow identification grid points; Based on the spatial adjacency analysis algorithm, all slow flow identification grid points are spatially clustered to extract the contiguous distribution areas, which are defined as potential slow flow enrichment area data Pos.

[0010] Preferably, said S3 includes S31; S31. Based on the potential slow-flow enrichment area data Pos, extract from the basic environmental data set Env all groundwater pollution index data sets Dpc whose spatial coordinate positions are consistent with the slow-flow identification grid points in the potential slow-flow enrichment area data Pos, then perform interpolation and noise removal preprocessing on the groundwater pollution index data sets Dpc to fill in the missing data points within the monitoring period, and clean the mutation values ​​and random noise based on the sliding median filter; After the preprocessing is completed, the time series concentration change information structure of the groundwater pollution index dataset Dpc is sorted according to the position of the slow flow identification grid point, and a pollutant concentration change trend dataset Dtc is established with the slow flow identification grid point in the slow flow enrichment area data Pos as the basic unit.

[0011] Preferably, the S3 further includes S32; S32, constructing a non-slow flow control area set Nrg by screening control grid points that are adjacent to the potential slow flow enrichment area data Pos in the spatial coordinate system but are not marked as slow flow identification grid points from the basic environmental dataset Env; Extract the groundwater pollution index dataset Dpc of each control grid point from the non-slow flow control area set Nrg, and construct the control pollutant change trend dataset DtcZ in the time series; Then calculate the change rates of the nitrate concentration data Con1, arsenic concentration data Con2, and lead concentration data Con3 in the pollutant concentration change trend dataset Dtc and the control pollutant change trend dataset DtcZ respectively, and obtain the nitrate concentration change rate difference Cs1, arsenic concentration change rate difference Cs2, and lead concentration change rate difference Cs3; The absolute value of the change rate of the three types of pollutants is then calculated, and the average value is calculated within the preset monitoring period to obtain the spatial displacement rate index Dis of the pollutants. The spatial displacement rate index Dis is used to quantify the difference between the pollutant concentration change trend in the slow-flow area and the control area, reflecting whether there is a risk of slowed migration and deposition of pollutants. The spatial offset rate indicator Dis includes the spatial offset rate Cd1 of the nitrate concentration data Con1, the spatial offset rate Cd2 of the arsenic concentration data Con2, and the spatial offset rate Cd3 of the lead concentration data Con3.

[0012] Preferably, the S4 includes S41; S41, extracting the spatial offset rate Cd1 of the nitrate concentration data Con1, the spatial offset rate Cd2 of the arsenic concentration data Con2, the spatial offset rate Cd3 of the lead concentration data Con3, the water velocity map data Vmd, the hydraulic gradient distribution data Gra, the groundwater permeability data Per, and the water layer thickness data Thk at the slow flow identification grid point, and performing normalization processing using the range normalization method to unify the parameter value range to [0,1], thereby eliminating the dimensional differences between different parameters; The normalized parameters are then used to establish the pollutant deposition potential function Fun using a linear weighted modeling method to quantitatively express the deposition potential of the three types of pollutants at the slow flow identification grid points. The pollutant deposition potential function Fun is established by the following calculation formula: ; Where, represents the pollutant deposition potential function of the i-th type of pollutant at the slow flow identification grid point position p, Vmd(p) represents the groundwater velocity value at the slow flow identification grid point position p in the water velocity map data Vmd, It represents the spatial offset rate index of the i-th type of pollutant at the slow flow identification grid point position p, represents the hydraulic gradient value of the hydraulic gradient distribution data Gra at the slow flow identification grid point position p, Represents the permeability of groundwater permeability data Per at the slow flow identification grid point position p, Represents the water layer thickness data Thk at the slow flow identification grid point position p; w1, w2, w3, w4 and w5 represent the groundwater velocity value, spatial offset rate index of the i-th type of pollutant, hydraulic gradient value, permeability and water layer thickness weight coefficient at the slow flow identification grid point position p, respectively, and w1+w2+w3+w4+w5=1. The specific value is set by the user; Among them, i∈[1,2,3], 1 represents the spatial offset rate Cd1 of nitrate concentration data Con1, 2 represents the spatial offset rate Cd2 of arsenic concentration data Con2, and 3 represents the spatial offset rate Cd3 of lead concentration data Con3.

[0013] Preferably, the S4 further includes S42; S42, based on the obtained pollutant deposition potential function representing the i-th type of pollutant at the slow flow identification grid point position p , the maximum value among the three types of pollutants is used to quantify the enrichment bias index Ebi (p) of the joint enrichment trend at the slow flow identification grid point position p; The enrichment bias index Ebi(p) is expressed by Obtain the calculation formula.

[0014] Preferably, the S5 includes S51; S51. Based on the obtained enrichment bias index Ebi(p) at the slow flow identification grid point position p, the obtained index is compared with a preset threshold interval to establish a groundwater pollution sedimentation risk level standard for the target monitoring area, and an early warning and response are generated based on the groundwater pollution sedimentation risk level standard; Among them, the threshold interval is preset in the range of [0,1]; The groundwater contamination sediment risk level standard is obtained by the following comparison method: When 0≤Ebi(p)<0.3, it indicates that the sedimentation bias risk is low and the slow flow identification grid point p is marked as L1. No warning or response is given and monitoring continues. When 0.3≤Ebi(p)<0.6, it indicates a medium risk of sediment bias, and the slow flow identification grid point p is marked as L2. The prompt response operation of increasing the sampling frequency and scheduling frequency is executed. When 0.6≤the enrichment bias index Ebi(p)≤1 at the slow flow identification grid point position p, it indicates a high risk of sediment bias, and the slow flow identification grid point position p is marked as level L3. The early warning response of the slow flow identification grid point position p and the prompt response operation of restricting groundwater use behavior in the section to which the slow flow identification grid point position p belongs are executed.

[0015] A groundwater health risk early warning system based on data analysis, including a groundwater data acquisition module, a data velocity region identification module, a low velocity concentration change calculation module, a regional sedimentation risk calculation module, and a response decision module; The groundwater data acquisition module collects the basic environmental data set Env in the target monitoring area through the groundwater flow velocity sensor; The data velocity region identification module performs spatial interpolation and isosurface fitting on the basic environmental dataset Env, constructs the water velocity map data Vmd, and delineates the potential slow flow enrichment area data Pos; The low-flow concentration change calculation module extracts the basic environmental data set Env from the identified potential slow-flow enrichment area data Pos, analyzes the change trend in the time dimension, and calculates the pollutant spatial offset rate index Dis; The regional deposition risk calculation module combines the water velocity map data Vmd, the pollutant spatial offset rate index Dis, and the basic environmental data set Env to establish the pollutant deposition potential function Fun. It calculates the deposition risk of pollutants in each slow-flow enrichment micro-region in the potential slow-flow enrichment area data Pos under slow-flow conditions and obtains the enrichment bias index Ebi. The response decision module compares the enrichment bias index Ebi with the preset threshold area, constructs the groundwater pollution deposition risk level standard for the target monitoring area, and issues early warning based on the groundwater pollution deposition risk level standard.

[0016] The present invention provides a groundwater health risk early warning method and system based on data analysis, which has the following beneficial effects: (1) By establishing a sedimentation mechanism-guided pollutant deposition potential function Fun, a quantitative analysis of pollutant accumulation tendency in the potential slow-flow enrichment area data Pos is achieved. Compared with the coarse-grained assessment method of the traditional method that relies only on the average concentration value of pollutants for static judgment, this method can accurately identify the pollutant deposition amplification phenomenon in the local low-velocity area, and then quantify the composite enrichment degree of pollutants under the micro-scale hydrodynamic structure through the enrichment bias index Ebi. Based on this, a groundwater pollution deposition risk level standard and a zoning early warning mechanism are constructed. This scheme effectively compensates for the lack of identification and response hysteresis of the "slow-flow driven enrichment mechanism" in existing groundwater pollution assessment methods.

[0017] (2) Based on the slow flow identification grid points in the potential slow flow enrichment area data Pos, the groundwater pollution index dataset Dpc at the corresponding location is extracted from the basic environmental dataset Env, and the control pollutant change trend dataset DtcZ is generated to form the pollutant spatial offset rate index Dis. This effectively realizes the detection of subtle dynamic differences in pollutant change trends between slow flow areas and conventional areas, breaking through the monitoring bottleneck of traditional static concentration evaluation that cannot identify "migration rate slowdown" as a precursor to enrichment. It provides a reliable basis for the subsequent construction of the pollutant deposition potential function Fun and the enrichment bias index Ebi. It is particularly suitable for trend perception and potential risk judgment in the early stage of enrichment when significant concentration accumulation has not yet formed.

[0018] (3) By forming the enrichment bias index Ebi(p) at the slow-flow identification grid point position p, which is used to express the enrichment deviation trend of the composite pollutant at that location, and by comparing the enrichment bias index Ebi(p) at the slow-flow identification grid point position p with the preset threshold range, a groundwater contamination deposition risk level standard for the target monitoring area is constructed, and a L1, L2, and L3 level segmentation strategy is formed to achieve continuous monitoring from low-risk status to use restriction prompt response operations under high-risk status. The above method significantly improves the quantitative identification capability of the evolution process of groundwater pollutant deposition, not only enhancing the response accuracy of micro-scale enrichment changes, but also providing a warning and executable response basis for risk governance at the regional scale, with high practicality and regulatory integration value. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a schematic diagram of the steps of the groundwater health risk early warning method based on data analysis of the present invention; Figure 2 This is a schematic diagram of a block diagram of a groundwater health risk early warning system based on data analysis according to the present invention; Figure 3 Schematic diagram of the enrichment bias index Ebi(p) and risk level for the slow flow identification grid point position p. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0021] Example 1: The present invention provides a groundwater health risk early warning method based on data analysis, please refer to Figure 1 , including the following steps: S1, collect the basic environmental data set Env in the target monitoring area through the groundwater flow velocity sensor; S2. Perform spatial interpolation and isosurface fitting on the basic environmental dataset Env to construct the water velocity map data Vmd and delineate the potential slow-flow enrichment area data Pos; S3. Extract the basic environmental data set Env from the identified potential slow-flow enrichment area data Pos, analyze the change trend in the time dimension, and calculate the pollutant spatial offset rate index Dis; S4. Combine the water velocity map data Vmd, the pollutant spatial offset rate index Dis, and the basic environmental data set Env to establish the pollutant deposition potential function Fun. For each slow-flow enrichment micro-area in the potential slow-flow enrichment area data Pos, the pollutant deposition risk under slow-flow conditions is calculated to obtain the enrichment bias index Ebi. S5. Compare the enrichment bias index Ebi with the preset threshold area, construct the groundwater pollution sedimentation risk level standard for the target monitoring area, and issue an early warning based on the groundwater pollution sedimentation risk level standard.

[0022] In this embodiment, by establishing a sedimentation mechanism-guided pollutant deposition potential function Fun, a quantitative analysis of the pollutant accumulation tendency in the potential slow-flow enrichment area data Pos is achieved. Compared with the coarse-grained assessment method of the traditional method that only relies on the average concentration value of pollutants for static judgment, this method can accurately identify the pollutant deposition amplification phenomenon in the local low-velocity area, and then quantify the composite enrichment degree of pollutants under the micro-scale hydrodynamic structure through the enrichment bias index Ebi, and on this basis, construct a groundwater pollution deposition risk level standard and a zoning early warning mechanism. This scheme effectively makes up for the lack of identification and response hysteresis of the "slow-flow driven enrichment mechanism" in existing groundwater pollution assessment methods. It is particularly suitable for the early warning needs of sensing sedimentation health risks in advance in areas where the concentration value has not yet exceeded the standard but the local dynamic anomaly is abnormal. It has stronger foresight, spatial adaptability and health response value.

[0023] Example 2: Specifically: S1 includes S11 and S12; S11. Collect groundwater flow velocity sensors and hydrogeological observation equipment deployed in the target monitoring area to collect parameter data related to groundwater hydrodynamics and form a groundwater dynamics dataset Dgw. The groundwater dynamic data set Dgw includes water velocity distribution data Vel, hydraulic gradient distribution data Gra, groundwater permeability data Per and water layer thickness data Thk; Among them, groundwater flow velocity sensors and hydrogeological observation equipment include groundwater micro electromagnetic flow velocity sensors, pressure sensors, high-resolution water level automatic recorders and geological profile conductivity profilers; The water velocity distribution data Vel is used to describe the spatial variation of groundwater velocity in different spatial units; the hydraulic gradient distribution data Gra is used to describe the vertical and horizontal head variation trend of groundwater; the groundwater permeability data Per is used to reflect the difficulty of stratum penetration by water flow; the water layer thickness data Thk is used to reflect the effective water storage thickness of the aquifer; S12. Collect concentration data of typical pollutants by deploying water quality testing equipment, obtain pollutant concentration monitoring data at the same spatial location in the target monitoring area, and form a groundwater pollution index dataset Dpc. This is then integrated with the groundwater dynamics dataset Dgw to obtain the basic environment dataset Env. The groundwater pollution index dataset Dpc includes nitrate concentration data Con1, arsenic concentration data Con2 and lead concentration data Con3; Among them, water quality testing equipment includes online ultraviolet absorption water quality analyzer, groundwater arsenic-specific electrochemical analyzer and portable heavy metal ion detector; The nitrate concentration data Con1 is used to assess the penetration level of agricultural non-point source pollution; the arsenic concentration data Con2 is used to identify the natural geochemical background or the impact of industrial emissions; and the lead concentration data Con3 is used to monitor the evolution behavior of metal pollutants in water bodies.

[0024] In this embodiment, a sensing system consisting of a groundwater micro-electromagnetic flow sensor, a pressure sensor, a high-resolution water level automatic recorder, and a geological profile conductivity profiler is used to systematically collect water flow velocity distribution data Vel, hydraulic gradient distribution data Gra, groundwater permeability data Per, and water layer thickness data Thk to form a groundwater dynamic dataset Dgw, which comprehensively characterizes the flow driving force and medium response characteristics of groundwater in different spatial units. At the same time, with the help of an online ultraviolet absorption water quality analyzer, a groundwater arsenic-specific electrochemical analyzer, and a portable heavy metal ion detector, nitrate concentration data Con1, arsenic concentration data Con2, and lead concentration data Con3 are synchronously acquired at the same spatial location to construct a groundwater pollution index dataset Dpc. By integrating the groundwater dynamics dataset Dgw and the groundwater pollution index dataset Dpc into a unified structure, a basic environmental dataset Env with spatial consistency and indicator complementarity is formed, which provides a unified data benchmark platform for subsequent slow flow identification, sedimentation modeling and risk warning, fundamentally improves data fusion and modeling accuracy, and provides quantifiable and traceable real data support for the dynamic simulation of groundwater pollution processes.

[0025] Example 3: Specifically: S2 includes S21; S21, extracting the water velocity distribution data Vel from the basic environmental dataset Env, constructing a two-dimensional regular spatial grid with the spatial coordinate system of the target monitoring area as a reference, using each grid point in the two-dimensional regular spatial grid as an observation point, and then executing the Kriging interpolation method to construct a continuous velocity prediction function based on the water velocity distribution data Vel at the corresponding position of the observation point, for estimating the groundwater velocity value at any observation point; Mapping the water velocity value to the entire two-dimensional regular space grid to form a continuously distributed two-dimensional velocity field, and generating a velocity isosurface model based on this. The velocity isosurface model divides the velocity variation characteristics in the area by isovelocity lines to generate water velocity map data Vmd covering the target area; The water flow velocity map data Vmd is used to express the spatial velocity difference distribution of groundwater in the entire area, and is a prerequisite input for the subsequent identification of slow flow areas and the establishment of pollutant enrichment models.

[0026] Said S2 also includes S22; S22, performing statistical calculations on the groundwater velocity values ​​at all observation points in the water velocity map data Vmd to obtain a regional average velocity parameter Vavg of the target monitoring area within the target monitoring period, which is used to reflect the baseline level of groundwater velocity in the target monitoring area; Taking 30% of the regional average velocity parameter Vavg as the slow flow determination threshold, a traversal judgment is performed on each observation point in the water velocity map data Vmd, and all observation points with predicted water velocity values ​​lower than the determination threshold are screened out, and the observation points are marked as slow flow identification grid points; Based on the spatial adjacency analysis algorithm, all slow flow identification grid points are spatially clustered to extract the contiguous distribution area, which is defined as the potential slow flow enrichment area data Pos; The potential slow-flow enrichment area data Pos is used to represent a set of spatial regions with a continuously low flow velocity and a tendency for pollutant deposition and enrichment under current groundwater dynamic conditions. It is the spatial input basis for subsequent pollutant spatial offset analysis and enrichment bias index calculation.

[0027] In this example, a continuous prediction function is constructed by applying kriging interpolation to the water velocity distribution data Vel in the basic environmental dataset Env. This method generates a two-dimensional velocity contour surface model covering the target monitoring area, thereby establishing a water velocity map data Vmd that expresses the spatial variation of groundwater velocity at the regional scale. Based on the water velocity map data Vmd, a slow flow threshold is constructed in combination with the regional average velocity parameter Vavg. A spatial adjacency analysis algorithm is then used to cluster the slow flow identification grid points. Ultimately, the potential slow flow enrichment area data Pos is constructed, achieving spatially explicit expression of groundwater slow flow risk and continuous region extraction. Compared with existing slow flow area demarcation methods based on single-point judgment or discrete sample inference, this method significantly improves the continuity of velocity identification and the objectivity of the discrimination threshold, avoiding local misjudgments or human intervention in slow flow identification. It also ensures that the potential slow flow enrichment area data Pos are spatially traceable and have closed boundaries, providing a logically clear and data-consistent spatial input guarantee for the subsequent construction of the pollutant spatial displacement rate indicator Dis and the execution of the pollutant deposition potential function Fun.

[0028] Example 4: Specifically: S3 includes S31; S31. Based on the potential slow-flow enrichment area data Pos, extract from the basic environmental data set Env all groundwater pollution index data sets Dpc whose spatial coordinate positions are consistent with the slow-flow identification grid points in the potential slow-flow enrichment area data Pos, then perform interpolation and noise removal preprocessing on the groundwater pollution index data sets Dpc to fill in the missing data points within the monitoring period, and clean the mutation values ​​and random noise based on the sliding median filter; After the preprocessing is completed, the time series concentration change information structure of the groundwater pollution index dataset Dpc is sorted according to the position of the slow flow identification grid point, and a pollutant concentration change trend dataset Dtc is established with the slow flow identification grid point in the slow flow enrichment area data Pos as the basic unit.

[0029] Said S3 also includes S32; S32, constructing a non-slow flow control area set Nrg by screening control grid points that are adjacent to the potential slow flow enrichment area data Pos in the spatial coordinate system but are not marked as slow flow identification grid points from the basic environmental dataset Env; Extract the groundwater pollution index dataset Dpc of each control grid point from the non-slow flow control area set Nrg, and construct the control pollutant change trend dataset DtcZ in the time series; Then calculate the change rates of the nitrate concentration data Con1, arsenic concentration data Con2, and lead concentration data Con3 in the pollutant concentration change trend dataset Dtc and the control pollutant change trend dataset DtcZ respectively, and obtain the nitrate concentration change rate difference Cs1, arsenic concentration change rate difference Cs2, and lead concentration change rate difference Cs3; The change rate is obtained by calculating using a first-order difference method; The change rate is the difference between the pollutant change rate in the slow flow identification grid point area and the control grid point area at the same time, and the unit is mg / L·day -1 ; Example instructions for obtaining the nitrate concentration change rate difference Cs1, arsenic concentration change rate difference Cs2, and lead concentration change rate difference Cs3: Set the following parameters: The monitoring period is in days, and the concentration unit is in milligrams per liter (mg / L); Nitrate concentration data Con1: Time t1 = Day 1, concentration of p at the slow-flow identification grid point (mg / L): 19.5, concentration at the control grid point (mg / L): 21.0; At time t2 (day 2), the concentration of p at the slow-flow identification grid point (mg / L) was 20.1, and the concentration at the control grid point (mg / L) was 21.8. The concentration change rate of the slow flow identification grid point is: (20.1-19.5) / 1=0.6mg / L; The rate of change of concentration at the control grid point is: (21.8-21.0) / 1=0.8 mg / L; The difference in the rate of change of nitrate concentration Csl = 0.6-0.8 = -0.2 mg / L; Arsenic concentration data Con2: Time t1 = Day 1, concentration of p at the slow-flow identification grid point (mg / L): 0.014, concentration at the control grid point (mg / L): 0.017; Time t2 = Day 2, concentration of p at the slow-flow identification grid point (mg / L): 0.015, concentration at the control grid point (mg / L): 0.021; Slow flow identification grid point concentration change rate: (0.015-0.014) / 1=0.001mg / L; The rate of change of concentration at the control grid point is (0.021-0.017) / 1=0.004 mg / L; The difference in the rate of change of arsenic concentration Cs2 = 0.001-0.004 = -0.003 mg / L; Lead concentration data Con3: Time t1 = Day 1, concentration of p at the slow-flow identification grid point (mg / L): 0.085, concentration at the control grid point (mg / L): 0.091; Time t2 = Day 2, concentration of p at the slow-flow identification grid point (mg / L): 0.086, concentration at the control grid point (mg / L): 0.094; Slow flow identification grid point concentration change rate: (0.086-0.085) / 1=0.001mg / L; The rate of change of concentration at the control grid point is (0.094-0.091) / 1=0.003 mg / L; The difference in the rate of change of lead concentration Cs3 = 0.001-0.003 = -0.002 mg / L; According to the above example, within the same time period, the pollutant concentration change rate at the slow flow identification grid point position p is lower than the pollutant change rate in the adjacent non-slow flow control area; The absolute value of the change rate of the three types of pollutants is then calculated, and the average value is calculated within the preset monitoring period to obtain the spatial displacement rate index Dis of the pollutants. The spatial displacement rate index Dis is used to quantify the difference between the pollutant concentration change trend in the slow-flow area and the control area, reflecting whether there is a risk of slowed migration and deposition of pollutants. The spatial offset rate indicator Dis includes the spatial offset rate Cd1 of the nitrate concentration data Con1, the spatial offset rate Cd2 of the arsenic concentration data Con2, and the spatial offset rate Cd3 of the lead concentration data Con3.

[0030] In this embodiment, based on the slow flow identification grid points in the potential slow flow enrichment area data Pos, the groundwater pollution index dataset Dpc at the corresponding location is extracted from the basic environmental dataset Env. The structural integrity and stability control of the multi-pollutant data are achieved through interpolation completion and sliding median filtering, and finally the pollutant concentration change trend dataset Dtc is formed. At the same time, a non-slow flow control area set Nrg is constructed through spatial coordinate system proximity analysis, and a control pollutant change trend dataset DtcZ is generated. The first-order difference method is used to calculate the difference in nitrate concentration change rate Cs1, arsenic concentration change rate Cs2, and lead concentration change rate Cs3 for the two types of areas, respectively. The pollutant spatial offset rate indicator Dis, composed of the spatial offset rate Cd1 of the nitrate concentration data Con1, the spatial offset rate Cd2 of the arsenic concentration data Con2, and the spatial offset rate Cd3 of the lead concentration data Con3, is further derived. This method effectively detects minute dynamic differences in pollutant change trends between slow-flow areas and conventional areas, breaking through the monitoring bottleneck of traditional static concentration evaluation that cannot identify the "slowed migration rate" as a precursor to enrichment. It provides a reliable basis for dynamic behavior characterization for the subsequent construction of the pollutant deposition potential function Fun and the enrichment bias index Ebi, and is particularly suitable for trend perception and potential risk judgment in the early stage when enrichment has not yet formed significant concentration accumulation.

[0031] Example 5: Please refer to Figure 1 and Figure 3 Specifically: the S4 includes S41; S41, extracting the spatial offset rate Cd1 of the nitrate concentration data Con1, the spatial offset rate Cd2 of the arsenic concentration data Con2, the spatial offset rate Cd3 of the lead concentration data Con3, the water velocity map data Vmd, the hydraulic gradient distribution data Gra, the groundwater permeability data Per, and the water layer thickness data Thk at the slow flow identification grid point, and performing normalization processing using the range normalization method to unify the parameter value range to [0,1], thereby eliminating the dimensional differences between different parameters; The normalized parameters are then used to establish the pollutant deposition potential function Fun using a linear weighted modeling method to quantitatively express the deposition potential of the three types of pollutants at the slow flow identification grid points. The pollutant deposition potential function Fun is established by the following calculation formula: ; Where, represents the pollutant deposition potential function of the i-th type of pollutant at the slow flow identification grid point position p, Vmd(p) represents the groundwater velocity value at the slow flow identification grid point position p in the water velocity map data Vmd, It represents the spatial offset rate index of the i-th type of pollutant at the slow flow identification grid point position p, represents the hydraulic gradient value of the hydraulic gradient distribution data Gra at the slow flow identification grid point position p, Represents the permeability of groundwater permeability data Per at the slow flow identification grid point position p, represents the water layer thickness data Thk at the slow flow identification grid point position p; w1, w2, w3, w4, and w5 represent the groundwater velocity value, the spatial offset rate index of the i-th type of pollutant, the hydraulic gradient value, the permeability, and the weight coefficient of the water layer thickness at the slow flow identification grid point position p, respectively, and w1+w2+w3+w4+w5=1. The specific value is set by the user. The actual physical significance of this formula is to quantitatively assess the relative potential for deposition or enrichment of specific pollutants in local micro-regions under slow groundwater flow environments, that is, to measure whether pollutants have the risk tendency of "persistent residence" and possible enrichment and deposition in micro-environments with weak hydrodynamics and complex geological conditions. The physical significance of the groundwater velocity value Vmd(p) at the slow flow identification grid point position p in the water velocity map data Vmd is that the lower the velocity, the more likely it is that pollutant migration will be slowed down, leading to easier deposition (thus, the lower the value, the higher the risk). The spatial offset rate index of the i-th type of pollutant at the slow flow identification grid point position p The physical meaning of is: the higher the value, the stronger the "retention degree" of the pollutant in the slow flow area → indicating that migration is hindered and the possibility of deposition is greater; The hydraulic gradient distribution data Gra is the hydraulic gradient value at the slow flow identification grid point position p The physical meaning of the gradient is: large gradient → accelerated flow → small sedimentation tendency; small gradient → more likely to enrich sedimentation; it can also be expressed as hydraulic driving ability; The groundwater permeability data Per is the permeability at the slow flow identification grid point position p. The physical meaning of is that: a formation with low permeability (low Per) may cause pollutants to "reside and settle" in the pores; The water layer thickness data Thk is the water layer thickness at the slow flow identification grid point position p The physical meaning of this is that thin layers have slow water exchange and are prone to forming stagnation zones, while thicker layers have greater "scouring" capabilities, which are not conducive to sedimentation. Among them, i∈[1,2,3], 1 represents the spatial offset rate Cd1 of nitrate concentration data Con1, 2 represents the spatial offset rate Cd2 of arsenic concentration data Con2, and 3 represents the spatial offset rate Cd3 of lead concentration data Con3.

[0032] Said S4 also includes S42; S42, based on the obtained pollutant deposition potential function representing the i-th type of pollutant at the slow flow identification grid point position p , the maximum value among the three types of pollutants is used to quantify the enrichment bias index Ebi (p) of the joint enrichment trend at the slow flow identification grid point position p; The enrichment bias index Ebi(p) is expressed by Obtain the calculation formula; where max represents the maximum value operation.

[0033] The S5 includes S51; S51. Based on the obtained enrichment bias index Ebi(p) at the slow flow identification grid point position p, the obtained index is compared with a preset threshold interval to establish a groundwater pollution sedimentation risk level standard for the target monitoring area, and an early warning and response are generated based on the groundwater pollution sedimentation risk level standard; Among them, the threshold interval is preset in the range of [0,1]; The groundwater contamination sediment risk level standard is obtained by the following comparison method: When 0≤Ebi(p)<0.3, it indicates that the sedimentation bias risk is low and the slow flow identification grid point p is marked as L1. No warning or response is given and monitoring continues. When 0.3≤Ebi(p)<0.6, it indicates a medium risk of sediment bias, and the slow flow identification grid point p is marked as L2. The prompt response operation of increasing the sampling frequency and scheduling frequency is executed. When 0.6≤the enrichment bias index Ebi(p)≤1 at the slow flow identification grid point position p, it indicates a high risk of sediment bias, and the slow flow identification grid point position p is marked as level L3. The early warning response of the slow flow identification grid point position p and the prompt response operation of restricting groundwater use behavior in the section to which the slow flow identification grid point position p belongs are executed.

[0034] In this example, the enrichment bias index Ebi(p) is used to quantitatively express the joint enrichment trend of groundwater pollutants and spatially map them to risk level standards. Water velocity atlas data Vmd, hydraulic gradient distribution data Gra, groundwater permeability data Per, and water layer thickness data Thk are selected, along with spatial offset rate indicators for three types of pollutants (including the spatial offset rate Cd1 of nitrate concentration data Con1, the spatial offset rate Cd2 of arsenic concentration data Con2, and the spatial offset rate Cd3 of lead concentration data Con3). After performing range normalization on each parameter, a linear weighted modeling method is introduced to construct the pollutant deposition potential function Fun. This allows for the integrated analysis of driving factors from different sources and with different physical meanings within the same evaluation framework, accurately capturing the spatially explicit characteristics of deposition trends. On this basis, the results of the deposition potential functions of the three types of pollutants are weighted and integrated to form the enrichment bias index Ebi(p) at the slow-flow identification grid point location p, which is used to express the enrichment bias trend of the composite pollutant at that location. By comparing the enrichment bias index Ebi(p) with the preset threshold interval, a groundwater pollution deposition risk level standard for the target monitoring area is constructed, and a L1, L2, and L3 level segmentation strategy is formed to achieve continuous monitoring from low-risk states to use restriction prompt response operations under high-risk states. The above method significantly improves the quantitative identification capability of the evolution process of groundwater pollutant deposition, not only enhancing the response accuracy of microscale enrichment changes, but also providing a warning and executable response basis for risk governance at the regional scale, with high practicality and regulatory integration value.

[0035] Example 6: Groundwater health risk early warning system based on data analysis, please refer to Figure 2 ,Specifically: including groundwater data acquisition module, data flow rate area identification module, low flow rate concentration change calculation module, regional sedimentation risk calculation module and response decision module; The groundwater data acquisition module collects the basic environmental data set Env in the target monitoring area through the groundwater flow velocity sensor; The data velocity region identification module performs spatial interpolation and isosurface fitting on the basic environmental dataset Env, constructs the water velocity map data Vmd, and delineates the potential slow flow enrichment area data Pos; The low-flow concentration change calculation module extracts the basic environmental data set Env from the identified potential slow-flow enrichment area data Pos, analyzes the change trend in the time dimension, and calculates the pollutant spatial offset rate index Dis; The regional deposition risk calculation module combines the water velocity map data Vmd, the pollutant spatial offset rate index Dis, and the basic environmental data set Env to establish the pollutant deposition potential function Fun. It calculates the deposition risk of pollutants in each slow-flow enrichment micro-region in the potential slow-flow enrichment area data Pos under slow-flow conditions and obtains the enrichment bias index Ebi. The response decision module compares the enrichment bias index Ebi with the preset threshold area, constructs the groundwater pollution deposition risk level standard for the target monitoring area, and issues early warning based on the groundwater pollution deposition risk level standard.

[0036] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A groundwater health risk early warning method based on data analysis, characterized by: The following steps are involved: S1, collect the basic environmental data set Env in the target monitoring area through the groundwater flow velocity sensor; S2. Perform spatial interpolation and isosurface fitting on the basic environmental dataset Env to construct the water velocity map data Vmd and delineate the potential slow-flow enrichment area data Pos; S3. Extract the basic environmental data set Env from the identified potential slow-flow enrichment area data Pos, analyze the change trend in the time dimension, and calculate the pollutant spatial offset rate index Dis; S4. Combine the water velocity map data Vmd, the pollutant spatial offset rate index Dis, and the basic environmental data set Env to establish the pollutant deposition potential function Fun. For each slow-flow enrichment micro-area in the potential slow-flow enrichment area data Pos, the pollutant deposition risk under slow-flow conditions is calculated to obtain the enrichment bias index Ebi. S5. Compare the enrichment bias index Ebi with the preset threshold area, construct the groundwater pollution sedimentation risk level standard for the target monitoring area, and issue an early warning based on the groundwater pollution sedimentation risk level standard.

2. The groundwater health risk early warning method based on data analysis according to claim 1 is characterized by: Said S1 includes S11 and S12; S11. Collect groundwater flow velocity sensors and hydrogeological observation equipment deployed in the target monitoring area to collect parameter data related to groundwater hydrodynamics and form a groundwater dynamics dataset Dgw. The groundwater dynamic data set Dgw includes water velocity distribution data Vel, hydraulic gradient distribution data Gra, groundwater permeability data Per and water layer thickness data Thk; Among them, groundwater flow velocity sensors and hydrogeological observation equipment include groundwater micro electromagnetic flow velocity sensors, pressure sensors, high-resolution water level automatic recorders and geological profile conductivity profilers; S12. Collect concentration data of typical pollutants by deploying water quality testing equipment, obtain pollutant concentration monitoring data at the same spatial location in the target monitoring area, and form a groundwater pollution index dataset Dpc. This is then integrated with the groundwater dynamics dataset Dgw to obtain the basic environment dataset Env. The groundwater pollution index dataset Dpc includes nitrate concentration data Con1, arsenic concentration data Con2 and lead concentration data Con3; Among them, water quality testing equipment includes online ultraviolet absorption water quality analyzer, groundwater arsenic-specific electrochemical analyzer and portable heavy metal ion detector.

3. The groundwater health risk early warning method based on data analysis according to claim 2 is characterized by: Said S2 includes S21; S21, extracting the water velocity distribution data Vel from the basic environmental dataset Env, constructing a two-dimensional regular spatial grid with the spatial coordinate system of the target monitoring area as a reference, using each grid point in the two-dimensional regular spatial grid as an observation point, and then executing the Kriging interpolation method to construct a continuous velocity prediction function based on the water velocity distribution data Vel at the corresponding position of the observation point, for estimating the groundwater velocity value at any observation point; The water flow velocity value is mapped to the entire two-dimensional regular space grid to form a continuously distributed two-dimensional flow velocity field, and a flow velocity isosurface model is generated based on this. The flow velocity isosurface model divides the velocity change characteristics in the area by isovelocity value lines to generate water flow velocity map data Vmd covering the target area.

4. The groundwater health risk early warning method based on data analysis according to claim 3 is characterized by: Said S2 also includes S22; S22, performing statistical calculations on the groundwater velocity values ​​at all observation points in the water velocity map data Vmd to obtain a regional average velocity parameter Vavg of the target monitoring area within the target monitoring period, which is used to reflect the baseline level of groundwater velocity in the target monitoring area; Taking 30% of the regional average velocity parameter Vavg as the slow flow determination threshold, a traversal judgment is performed on each observation point in the water velocity map data Vmd, and all observation points with predicted water velocity values ​​lower than the determination threshold are screened out, and the observation points are marked as slow flow identification grid points; Based on the spatial adjacency analysis algorithm, all slow flow identification grid points are spatially clustered to extract the contiguous distribution areas, which are defined as potential slow flow enrichment area data Pos.

5. The groundwater health risk early warning method based on data analysis according to claim 4 is characterized by: Said S3 includes S31; S31. Based on the potential slow-flow enrichment area data Pos, extract from the basic environmental data set Env all groundwater pollution index data sets Dpc whose spatial coordinate positions are consistent with the slow-flow identification grid points in the potential slow-flow enrichment area data Pos, then perform interpolation and noise removal preprocessing on the groundwater pollution index data sets Dpc to fill in the missing data points within the monitoring period, and clean the mutation values ​​and random noise based on the sliding median filter; After the preprocessing is completed, the time series concentration change information structure of the groundwater pollution index dataset Dpc is sorted according to the position of the slow flow identification grid point, and a pollutant concentration change trend dataset Dtc is established with the slow flow identification grid point in the slow flow enrichment area data Pos as the basic unit.

6. The groundwater health risk early warning method based on data analysis according to claim 5 is characterized by: Said S3 also includes S32; S32, constructing a non-slow flow control area set Nrg by screening control grid points that are adjacent to the potential slow flow enrichment area data Pos in the spatial coordinate system but are not marked as slow flow identification grid points from the basic environmental dataset Env; Extract the groundwater pollution index dataset Dpc of each control grid point from the non-slow flow control area set Nrg, and construct the control pollutant change trend dataset DtcZ in the time series; Then calculate the change rates of the nitrate concentration data Con1, arsenic concentration data Con2, and lead concentration data Con3 in the pollutant concentration change trend dataset Dtc and the control pollutant change trend dataset DtcZ respectively, and obtain the nitrate concentration change rate difference Cs1, arsenic concentration change rate difference Cs2, and lead concentration change rate difference Cs3; The absolute value of the change rate of the three types of pollutants is then calculated, and the average value is calculated within the preset monitoring period to obtain the spatial displacement rate index Dis of the pollutants. The spatial displacement rate index Dis is used to quantify the difference between the pollutant concentration change trend in the slow-flow area and the control area, reflecting whether there is a risk of slowed migration and deposition of pollutants. The spatial offset rate indicator Dis includes the spatial offset rate Cd1 of the nitrate concentration data Con1, the spatial offset rate Cd2 of the arsenic concentration data Con2, and the spatial offset rate Cd3 of the lead concentration data Con3.

7. The groundwater health risk early warning method based on data analysis according to claim 6 is characterized by: Said S4 includes S41; S41, extracting the spatial offset rate Cd1 of the nitrate concentration data Con1, the spatial offset rate Cd2 of the arsenic concentration data Con2, the spatial offset rate Cd3 of the lead concentration data Con3, the water velocity map data Vmd, the hydraulic gradient distribution data Gra, the groundwater permeability data Per, and the water layer thickness data Thk at the slow flow identification grid point, and performing normalization processing using the range normalization method to unify the parameter value range to [0,1], thereby eliminating the dimensional differences between different parameters; The normalized parameters are then used to establish the pollutant deposition potential function Fun using a linear weighted modeling method to quantitatively express the deposition potential of the three types of pollutants at the slow flow identification grid points. The pollutant deposition potential function Fun is established by the following calculation formula: ； Where, represents the pollutant deposition potential function of the i-th type of pollutant at the slow flow identification grid point position p, Vmd(p) represents the groundwater velocity value at the slow flow identification grid point position p in the water velocity map data Vmd, It represents the spatial offset rate index of the i-th type of pollutant at the slow flow identification grid point position p, represents the hydraulic gradient value of the hydraulic gradient distribution data Gra at the slow flow identification grid point position p, Represents the permeability of groundwater permeability data Per at the slow flow identification grid point position p, The water layer thickness data Thk represents the water layer thickness at the slow flow identification grid point position p; w1, w2, w3, w4, and w5 represent the groundwater velocity value at the slow flow identification grid point p, the spatial offset rate index of the i-th type of pollutant, the hydraulic gradient value, the permeability, and the weight coefficient of the water layer thickness, respectively, and w1+w2+w3+w4+w5=1. The specific values ​​are set by the user; Among them, i∈[1,2,3], 1 represents the spatial offset rate Cd1 of nitrate concentration data Con1, 2 represents the spatial offset rate Cd2 of arsenic concentration data Con2, and 3 represents the spatial offset rate Cd3 of lead concentration data Con3.

8. The groundwater health risk early warning method based on data analysis according to claim 7 is characterized by: Said S4 also includes S42; S42, based on the obtained pollutant deposition potential function representing the i-th type of pollutant at the slow flow identification grid point position p The maximum value among the three types of pollutants is used to quantify the enrichment bias index Ebi (p) of the joint enrichment trend at the slow flow identification grid point position p.

9. The groundwater health risk early warning method based on data analysis according to claim 8, characterized in that: The S5 includes S51; S51. Based on the obtained enrichment bias index Ebi(p) at the slow flow identification grid point position p, the obtained index is compared with a preset threshold interval to establish a groundwater pollution sedimentation risk level standard for the target monitoring area, and an early warning and response are generated based on the groundwater pollution sedimentation risk level standard; Among them, the threshold interval is preset in the range of [0,1]; The groundwater contamination sediment risk level standard is obtained by the following comparison method: When 0≤Ebi(p)<0.3, it indicates that the sedimentation bias risk is low and the slow flow identification grid point p is marked as L1. No warning or response is given and monitoring continues. When 0.3≤Ebi(p)<0.6, it indicates a medium risk of sediment bias, and the slow flow identification grid point p is marked as L2. The prompt response operation of increasing the sampling frequency and scheduling frequency is executed. When 0.6≤the enrichment bias index Ebi(p)≤1 at the slow flow identification grid point position p, it indicates a high risk of sediment bias, and the slow flow identification grid point position p is marked as level L3. The early warning response of the slow flow identification grid point position p and the prompt response operation of restricting groundwater use behavior in the section to which the slow flow identification grid point position p belongs are executed.

10. A groundwater health risk early warning system based on data analysis, applied to the groundwater health risk early warning method based on data analysis according to any one of claims 1 to 9, characterized in that: It includes groundwater data acquisition module, data velocity area identification module, low velocity concentration change calculation module, regional sedimentation risk calculation module and response decision module; The groundwater data acquisition module collects the basic environmental data set Env in the target monitoring area through the groundwater flow velocity sensor; The data velocity region identification module performs spatial interpolation and isosurface fitting on the basic environmental dataset Env, constructs the water velocity map data Vmd, and delineates the potential slow flow enrichment area data Pos; The low-flow concentration change calculation module extracts the basic environmental data set Env from the identified potential slow-flow enrichment area data Pos, analyzes the change trend in the time dimension, and calculates the pollutant spatial offset rate index Dis; The regional deposition risk calculation module combines the water velocity map data Vmd, the pollutant spatial offset rate index Dis, and the basic environmental data set Env to establish the pollutant deposition potential function Fun. It calculates the deposition risk of pollutants in each slow-flow enrichment micro-region in the potential slow-flow enrichment area data Pos under slow-flow conditions and obtains the enrichment bias index Ebi. The response decision module compares the enrichment bias index Ebi with the preset threshold area, constructs the groundwater pollution deposition risk level standard for the target monitoring area, and issues early warning based on the groundwater pollution deposition risk level standard.

Images