Karst groundwater microelement space distribution feature extraction method and system

Abstract

The invention relates to the technical field of pollution monitoring, in particular to a karst groundwater microelement spatial distribution feature extraction method and system, and the system comprises a server side and a user side. The server side comprises a first acquisition module, a determination module, a sampling module, an analysis module, a clustering module and a display module, and the clustering module is used for determining element dimensions related to the to-be-detected area according to the trace element data corresponding to the detection points and displaying the to-be-detected area according to the determined element dimensions. Based on a preset element dimension clustering strategy and the set position data corresponding to each detection point, clustering each detection point to form a detection point set class corresponding to each element dimension; and the display module is used for performing visual display of the same element dimension in a space simulation model corresponding to a preset to-be-detected area according to the detection point set class corresponding to each element dimension and based on the position data corresponding to each detection point in the detection point set class.

Description

Technical Field

[0001] The present invention relates to the technical field of pollution monitoring, and in particular to a method and system for extracting spatial distribution characteristics of trace elements in karst groundwater. Background Art

[0002] With the rapid development of environmental science, agricultural science, geology and other fields, the study of the spatial distribution of trace elements in soil, water and other natural media has become increasingly important. The existence form, concentration level and spatial distribution pattern of these trace elements directly affect the health of the ecosystem, crop growth and the quality of the human living environment. Therefore, accurately analyzing and determining the spatial distribution of trace elements in a specific area has become the key to scientific research and technological application.

[0003] However, there are several significant limitations in traditional sampling approaches that restrict a precise understanding of the spatial distribution of trace elements.

[0004] First, traditional sampling methods often rely on manual selection of sampling points, which is easily affected by human factors, such as the experience and preferences of the sampler, resulting in the selection of sampling points that may not be uniform or representative, thus affecting the reliability of the final data analysis.

[0005] Secondly, due to the lack of effective spatial sampling strategies, traditional methods find it difficult to capture the changing characteristics of trace elements at different scales, especially the changes at the microscopic level, which further reduces the accuracy of the sampling data. Summary of the invention

[0006] One of the purposes of the present invention is to provide a method and system for extracting spatial distribution characteristics of trace elements in karst groundwater, which can realize the detection of trace elements in the area to be detected and the rationalization and reliability of spatial distribution, improve the accuracy of trace element analysis, and solve the problems of low reliability and low accuracy of trace element analysis in the prior art.

[0007] In order to achieve the above-mentioned purpose, a system for extracting spatial distribution characteristics of trace elements in karst groundwater is provided, including a server and a user end;

[0008] The server includes:

[0009] The first acquisition module is used to obtain geological data, geographical data and ground building data corresponding to the area to be detected;

[0010] A determination module is used to determine the setting position data of each detection point corresponding to the area to be detected based on the geological data, geographical data and ground building data corresponding to the area to be detected and based on a preset detection point selection strategy to form a detection point set;

[0011] A sampling module is used to sample each detection point according to each detection point in the detection point set and the corresponding setting position data, and obtain the sampling data corresponding to each detection point;

[0012] The analysis module is used to identify the sampling data corresponding to each detection point, the trace elements corresponding to the detection point and the element content corresponding to each trace element, and form the trace element data corresponding to each detection point;

[0013] A clustering module is used to determine the element dimensions involved in the detection area according to the trace element data corresponding to each detection point, and cluster each detection point according to the determined element dimensions and the setting position data corresponding to each detection point to form a detection point set class corresponding to each element dimension;

[0014] The display module is used to perform a visual display of the same element dimension in a spatial simulation model corresponding to a preset area to be detected based on the detection point set class corresponding to each element dimension and the position data corresponding to each detection point in the detection point set class, and feed back the visual display content to the user end.

[0015] Technical principles and effects of this solution: In this solution, the geological data and geographic data of the area to be detected are first obtained. Through the geological data and geographic data, combined with the preset detection point selection strategy, the location distribution of each detection point in the area to be detected can be quickly and reliably determined, thereby forming a corresponding detection point set, and then according to the location of each detection point in the detection point set. Then, according to the geological data, geographic data and ground building data corresponding to the area to be detected, combined with the preset detection point selection strategy, the setting location data of each detection point corresponding to the area to be detected is determined to form a detection point set; based on the preset detection point selection strategy, the geological data, geographic data and ground building data are comprehensively considered to ensure that the selection of sampling points is scientific and reasonable. Then, according to each detection point in the detection point set and the corresponding setting position data, each detection point is sampled to obtain the sampling data corresponding to each detection point; for the sampling data corresponding to each detection point, the trace elements corresponding to the detection point and the element content corresponding to each trace element are identified to form the trace element data corresponding to each detection point; the element dimensions involved in the detection area are determined, and according to the determined element dimensions and the setting position data corresponding to each detection point, each detection point is clustered to form a detection point set class corresponding to each element dimension;

[0016] Finally, according to the detection point set class corresponding to each element dimension, based on the position data corresponding to each detection point in the detection point set class, the same element dimension is visualized in the preset spatial simulation model corresponding to the area to be detected, and the visualization display content is fed back to the user end. Based on the position data corresponding to each detection point in the detection point set class, the same element dimension is visualized in the preset spatial simulation model corresponding to the area to be detected. Feeding back the visualization display content to the user end, the user can understand the spatial distribution characteristics of trace elements through an intuitive graphical interface, improving the readability and usability of the data.

[0017] In this scheme, accurate and reliable identification of trace elements corresponding to the area to be detected is achieved through the mutual cooperation of the determination module, sampling module, analysis module and clustering module. That is, the detection of trace elements in the area to be detected and the rationalization and reliability of spatial distribution can be realized, the accuracy of trace element analysis can be improved, the problems of low reliability and low accuracy of trace element analysis in the prior art can be solved, and the accurate identification of trace elements in karst groundwater and the determination of spatial distribution can be achieved.

[0018] Furthermore, the preset detection point selection strategy includes the following steps:

[0019] S10, dividing the area to be detected into a plurality of grids to form a corresponding grid set;

[0020] S20, randomly generating a number of initial populations, where the individuals of the initial population are a preset number of detection points corresponding to all grids in the grid set;

[0021] S30, based on the preset restriction conditions, preliminarily screen and judge the individuals in the population, those that meet the restriction conditions are feasible solutions, otherwise, they are infeasible solutions and are eliminated;

[0022] The restrictions are:

[0023]

[0024] In the formula, H j is the altitude corresponding to the jth detection point in the population, H yuzhi is the preset altitude threshold, Z is the area ratio of the smallest convex polygon containing all detection points to the area to be detected, M is the area value of the area to be detected, and Z mix is the minimum area ratio threshold, (x i ,y i ) is the coordinate of the i-th polygon vertex corresponding to the smallest convex polygon containing all the detection points, n is the number of polygon vertices of the smallest convex polygon containing all the detection points, and N is the total number of detection points in the population;

[0025] S40, performing fitness calculation on the selected population according to the preset first fitness and second fitness calculation formulas;

[0026] The first fitness calculation formula is as follows:

[0027]

[0028] In the formula, F1 is the first fitness, D is the reasonableness of the distribution of detection points corresponding to the population, |O l -O u | is the Euclidean distance between the lth detection point and the uth detection point;

[0029] The second fitness calculation formula is as follows:

[0030]

[0031] In the formula, F2 is the second fitness, CB is the sampling cost corresponding to all detection points of the population, ∈ is a positive number, C num is the total sampling cost corresponding to all detection points in the population, K1 is the fixed sampling cost corresponding to a single detection point, N is the number of detection points in the population, C reach is the sum of the accessibility of all detection points in the population, k2 is the slope sensitivity coefficient, slope k is the slope corresponding to the Kth detection point in the population, slope yuzhi is the slope threshold, distance k is the distance value from the kth detection point in the population to the nearest road, W1, W2, W3, and W4 are the corresponding weighting coefficients respectively;

[0032] S50, randomly generate a random number P from 0 to 1, and determine whether the random number P is greater than a preset random threshold. If so, in this iteration, according to the first fitness corresponding to the population, select the population whose first fitness is less than or equal to the preset first fitness threshold, and select the first three populations with the smallest second fitness from the populations eliminated at this time, and save them in the standby library;

[0033] If not, in this iteration, the current population is combined with the population in the backup library to form a new population, and the population corresponding to the current population whose second fitness is greater than the preset second fitness is selected;

[0034] S60, hybridizing and mutating the selected population to obtain an offspring population;

[0035] S70, after obtaining the offspring population, continue to execute S30 until a preset number of iterations is met;

[0036] S80, outputting the offspring population as the sampling optimal solution set corresponding to each grid, and identifying the setting position data corresponding to each detection point in the corresponding sampling optimal solution set according to the sampling optimal solution set, and forming a corresponding detection point set.

[0037] Beneficial effects: In this scheme, the population is preliminarily screened for restriction conditions each time an iteration is performed, so that the collection altitude of the detection points and the area ratio of the minimum convex polygon containing all the detection points to the area to be detected are limited, and then the first fitness and the second fitness are calculated respectively, that is, the rationality of the detection point distribution of the corresponding population and the sampling cost of the detection points are calculated. The calculation of the rationality of the detection point distribution evaluates the rationality of the distribution by calculating the minimum Euclidean distance between the detection points. If the detection points are unevenly distributed, some areas may have too dense or sparse detection points, which will affect the representativeness and accuracy of the sampling. By maximizing (F1), it can be ensured that the detection points are evenly distributed in space, improving the coverage and representativeness of the sampling.

[0038] In the process of calculating the sampling cost of the detection point, by introducing a small positive number ∈, the denominator in the entire calculation formula is avoided to be zero, thereby improving the rationality of the formula setting. Secondly, the economy of the sampling scheme is evaluated by calculating the total cost (CB) of the detection points. If the number of detection points is too large or the detection points are located in difficult-to-reach places, the cost of sampling will increase. By minimizing (F2), it can be ensured that the sampling cost is reduced as much as possible while ensuring the sampling quality.

[0039] In each iteration, by introducing a random number P to switch the fitness function corresponding to each iteration, the algorithm's dependence on a single fitness function can be reduced, and the algorithm's robustness in the face of different environments and data can be improved. That is, a single fitness function is prone to fall into a local optimal solution. Random switching can introduce new search directions to help the algorithm jump out of the local optimal solution and find the global optimal solution. Different fitness functions may focus on different features. Random switching can increase the diversity of solutions and avoid converging to the local optimal solution too early. The first fitness function may focus more on local search, while the second fitness function may focus more on global search. Through random switching, a better balance can be found between exploration and exploitation, improving global search capabilities.

[0040] At the same time, through random switching, the most suitable fitness function can be used at different stages to speed up the convergence speed, better adapt to environmental changes, improve the adaptability of the algorithm, introduce more changes, and prevent the algorithm from overfitting.

[0041] Furthermore, the server also includes:

[0042] A correlation calculation module is used to calculate the trace element correlation between each grid and the adjacent grids in the other four directions according to the detection points corresponding to each grid and the corresponding setting position data, based on a preset trace element correlation calculation strategy, and select the adjacent grid with the largest trace element correlation, and classify the adjacent grid and the corresponding grid into the same type of grids;

[0043] The random selection module is used to randomly select a detection point corresponding to a grid in the same grid as the detection point of all grids in the same grid after completing the division of all grids. After the detection points of all grids in the same grid are determined, the detection points corresponding to all grids in the same grid are collected to form a final detection point set and input into the sampling module.

[0044] Beneficial effects: The correlation calculation module can effectively reduce the number of detection points and avoid repeated sampling in areas with high correlation, thereby improving sampling efficiency. Adjacent grids with the greatest correlation are selected as similar grids to ensure that the detection points in each similar grid have similar trace element contents, thereby enhancing the representativeness of the data. The random selection module reduces the number of points actually required for sampling and reduces the sampling cost.

[0045] Furthermore, the preset trace element correlation calculation strategy is:

[0046] S100, randomly selecting a grid, and based on the grid, determining each adjacent grid in other four directions;

[0047] S200, according to the selected grid and other four adjacent grids, retrieve the historical meteorological data corresponding to each grid and the ground building data corresponding to each grid from the historical database;

[0048] S300, calculating the trace element correlation between the selected grid and the other four adjacent grids based on the historical meteorological data and ground building data corresponding to the selected grid and the other four adjacent grids and based on a preset trace element correlation calculation formula;

[0049] The preset trace element correlation calculation formula is:

[0050]

[0051] In the formula, G is the correlation of trace elements, P ab is the building correlation between grid a and grid b, Q ab is the meteorological correlation between grid a and grid b, ρ ar is the building density corresponding to the rth subgrid in grid a, t aris the historical temperature value corresponding to the rth subgrid in grid a, and h is the total number of subgrids corresponding to the grid.

[0052] Beneficial effects: In this solution, the correlation between the two grids is comprehensively evaluated by combining building density and meteorological data, realizing the evaluation reference of multi-dimensional data, and improving the comprehensiveness and accuracy of the evaluation. By selecting the adjacent grid with the highest correlation, it is ensured that the detection points in each similar grid have similar environmental conditions, which enhances the representativeness of the data. It reduces the errors caused by uneven distribution of detection points and improves the reliability and accuracy of the data.

[0053] Furthermore, the sampling data includes soil sample data, water sample data, rock sample data and plant sample data corresponding to the detection points in the area to be detected.

[0054] The present invention also provides a method for extracting spatial distribution characteristics of trace elements in karst groundwater, using the above-mentioned system for extracting spatial distribution characteristics of trace elements in karst groundwater. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a logic block diagram of the system for extracting spatial distribution characteristics of trace elements in karst groundwater in Example 1 of the present invention. DETAILED DESCRIPTION

[0056] The following is further described in detail through specific implementation methods:

[0057] Embodiment 1

[0058] The spatial distribution characteristics extraction system of trace elements in karst groundwater is basically as follows Figure 1 As shown, it includes a server and a user;

[0059] The server includes:

[0060] The first acquisition module is used to obtain geological data, geographical data and ground building data corresponding to the area to be detected;

[0061] A determination module is used to determine the setting position data of each detection point corresponding to the area to be detected based on the geological data, geographical data and ground building data corresponding to the area to be detected and based on a preset detection point selection strategy to form a detection point set;

[0062] The preset detection point selection strategy includes the following steps:

[0063] S10, dividing the area to be detected into a plurality of grids to form a corresponding grid set;

[0064] S20, randomly generating a number of initial populations, where the individuals of the initial population are a preset number of detection points corresponding to all grids in the grid set;

[0065] S30, based on the preset restriction conditions, preliminarily screen and judge the individuals in the population, those that meet the restriction conditions are feasible solutions, otherwise, they are infeasible solutions and are eliminated;

[0066] The restrictions are:

[0067]

[0068] In the formula, H j is the altitude corresponding to the jth detection point in the population, H yuzhi is the preset altitude threshold, Z is the area ratio of the smallest convex polygon containing all detection points to the area to be detected, M is the area value of the area to be detected, and Z mix is the minimum area ratio threshold, (x i ,y i ) is the coordinate of the i-th polygon vertex corresponding to the smallest convex polygon containing all the detection points, n is the number of polygon vertices of the smallest convex polygon containing all the detection points, and N is the total number of detection points in the population;

[0069] S40, performing fitness calculation on the selected population according to the preset first fitness and second fitness calculation formulas;

[0070] The first fitness calculation formula is as follows:

[0071]

[0072] In the formula, F1 is the first fitness, D is the reasonableness of the distribution of detection points corresponding to the population, |O l -O u | is the Euclidean distance between the lth detection point and the uth detection point;

[0073] The second fitness calculation formula is as follows:

[0074]

[0075] In the formula, F2 is the second fitness, CB is the sampling cost corresponding to all detection points of the population, ∈ is a positive number, C num is the total sampling cost corresponding to all detection points in the population, K1 is the fixed sampling cost corresponding to a single detection point, N is the number of detection points in the population, C reach is the sum of the accessibility of all detection points in the population, k2 is the slope sensitivity coefficient, slope k is the slope corresponding to the Kth detection point in the population, slope fuzhi is the slope threshold, distance kis the distance value from the kth detection point in the population to the nearest road, W1, W2, W3, and W4 are the corresponding weighting coefficients respectively;

[0076] S50, randomly generate a random number P from 0 to 1, and determine whether the random number P is greater than a preset random threshold. If so, in this iteration, according to the first fitness corresponding to the population, select the population whose first fitness is less than or equal to the preset first fitness threshold, and select the first three populations with the smallest second fitness from the populations eliminated at this time, and save them in the standby library;

[0077] If not, in this iteration, the current population is combined with the population in the backup library to form a new population, and the population corresponding to the current population whose second fitness is greater than the preset second fitness is selected;

[0078] S60, hybridizing and mutating the selected population to obtain an offspring population;

[0079] S70, after obtaining the offspring population, continue to execute S30 until a preset number of iterations is met;

[0080] S80, outputting the offspring population as the sampling optimal solution set corresponding to each grid, and identifying the setting position data corresponding to each detection point in the corresponding sampling optimal solution set according to the sampling optimal solution set, and forming a corresponding detection point set.

[0081] The server also includes:

[0082] A correlation calculation module is used to calculate the trace element correlation between each grid and the adjacent grids in the other four directions according to the detection points corresponding to each grid and the corresponding setting position data, based on a preset trace element correlation calculation strategy, and select the adjacent grid with the largest trace element correlation, and classify the adjacent grid and the corresponding grid into the same type of grids;

[0083] The preset trace element correlation calculation strategy is:

[0084] S100, randomly selecting a grid, and based on the grid, determining each adjacent grid in other four directions;

[0085] S200, according to the selected grid and other four adjacent grids, retrieve the historical meteorological data corresponding to each grid and the ground building data corresponding to each grid from the historical database;

[0086] S300, calculating the trace element correlation between the selected grid and the other four adjacent grids based on the historical meteorological data and ground building data corresponding to the selected grid and the other four adjacent grids and based on a preset trace element correlation calculation formula;

[0087] The preset trace element correlation calculation formula is:

[0088]

[0089] In the formula, G is the correlation of trace elements, P ab is the building correlation between grid a and grid b, Q ab is the meteorological correlation between grid a and grid b, ρ ar is the building density corresponding to the rth subgrid in grid a, t ar is the historical temperature value corresponding to the rth subgrid in grid a, and h is the total number of subgrids corresponding to the grid.

[0090] The random selection module is used to randomly select a detection point corresponding to a grid in the same grid as the detection point of all grids in the same grid after completing the division of all grids. After the detection points of all grids in the same grid are determined, the detection points corresponding to all grids in the same grid are collected to form a final detection point set and input into the sampling module.

[0091] The sampling module is used to sample each detection point according to each detection point in the detection point set and the corresponding setting position data, and obtain the sampling data corresponding to each detection point; in this embodiment, the sampling data includes soil sample data, water sample data, rock sample data and plant sample data corresponding to the detection point in the area to be detected.

[0092] The analysis module is used to identify the sampling data corresponding to each detection point, the trace elements corresponding to the detection point and the element content corresponding to each trace element, and form the trace element data corresponding to each detection point;

[0093] The clustering module is used to determine the element dimensions involved in the detection area according to the trace element data corresponding to each detection point, and cluster each detection point according to the determined element dimensions and based on the set position data corresponding to each detection point to form a detection point set class corresponding to each element dimension; in this embodiment, when clustering the detection points, firstly, according to the trace element data corresponding to each detection point, the element dimensions corresponding to the area to be detected are determined, for example, including element dimension 1, element dimension 2 and element dimension 3, and then, according to each element dimension, the data detected by each detection point and the position information corresponding to the detection point are collected. By using the information, it is possible to quickly determine whether the element content corresponding to each detection point in a certain element dimension has reached the corresponding threshold value. If so, it is judged that there is a corresponding trace element under the detection point, so as to complete the judgment of all detection points in all element dimensions, thereby forming the element content corresponding to each detection point in each element dimension. For example, under element dimension 1, the element contents corresponding to detection points 1, 2 and 3 of the element dimension 1 are aa, bb and cc respectively. Then, when displayed visually, the element content distribution diagram of the corresponding element dimension will be displayed on the corresponding detection points 1, 2 and 3.

[0094] The display module is used to perform a visual display of the same element dimension in a spatial simulation model corresponding to a preset area to be detected based on the detection point set class corresponding to each element dimension and the position data corresponding to each detection point in the detection point set class, and feed back the visual display content to the user end.

[0095] This embodiment also discloses a method for extracting spatial distribution characteristics of trace elements in karst groundwater, using the above-mentioned system for extracting spatial distribution characteristics of trace elements in karst groundwater.

[0096] The above is only an embodiment of the present invention. The common sense such as the known specific structure and characteristics in the scheme is described too much here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field of the invention before the application date or priority date, can know all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. The spatial distribution feature extraction system of trace elements in karst groundwater is characterized by: Including server and user side; The server includes: The first acquisition module is used to obtain geological data, geographical data and ground building data corresponding to the area to be detected; A determination module is used to determine the setting position data of each detection point corresponding to the area to be detected based on the geological data, geographical data and ground building data corresponding to the area to be detected and based on a preset detection point selection strategy to form a detection point set; A sampling module is used to sample each detection point according to each detection point in the detection point set and the corresponding setting position data, and obtain the sampling data corresponding to each detection point; The analysis module is used to identify the sampling data corresponding to each detection point, the trace elements corresponding to the detection point and the element content corresponding to each trace element, and form the trace element data corresponding to each detection point; A clustering module is used to determine the element dimensions involved in the detection area according to the trace element data corresponding to each detection point, and cluster each detection point according to the determined element dimensions and the setting position data corresponding to each detection point to form a detection point set class corresponding to each element dimension; The display module is used to perform a visual display of the same element dimension in a spatial simulation model corresponding to a preset area to be detected based on the detection point set class corresponding to each element dimension and the position data corresponding to each detection point in the detection point set class, and feed back the visual display content to the user end.

2. The system for extracting spatial distribution characteristics of trace elements in karst groundwater according to claim 1, characterized in that: The preset detection point selection strategy includes the following steps: S10, dividing the area to be detected into a plurality of grids to form a corresponding grid set; S20, randomly generating a number of initial populations, where the individuals of the initial population are a preset number of detection points corresponding to all grids in the grid set; S30, based on the preset restriction conditions, preliminarily screen and judge the individuals in the population, those that meet the restriction conditions are feasible solutions, otherwise, they are infeasible solutions and are eliminated; The restrictions are: In the formula, H j is the altitude corresponding to the jth detection point in the population, H yuzhi is the preset altitude threshold, Z is the area ratio of the smallest convex polygon containing all detection points to the area to be detected, M is the area value of the area to be detected, and Z mix is the minimum area ratio threshold, (x i ,y i ) is the coordinate of the i-th polygon vertex corresponding to the smallest convex polygon containing all the detection points, n is the number of polygon vertices of the smallest convex polygon containing all the detection points, and N is the total number of detection points in the population; S40, performing fitness calculation on the selected population according to the preset first fitness and second fitness calculation formulas; The first fitness calculation formula is as follows: In the formula, F1 is the first fitness, D is the reasonableness of the distribution of detection points corresponding to the population, |O l -O u | is the Euclidean distance between the lth detection point and the uth detection point; The second fitness calculation formula is as follows: In the formula, F2 is the second fitness, CB is the sampling cost corresponding to all detection points of the population, ∈ is a positive number, C num is the total sampling cost corresponding to all detection points in the population, K1 is the fixed sampling cost corresponding to a single detection point, N is the number of detection points in the population, C reach is the sum of the accessibility of all detection points in the population, k2 is the slope sensitivity coefficient, slope k is the slope corresponding to the Kth detection point in the population, slope yuzhi is the slope threshold, distance k is the distance value from the kth detection point in the population to the nearest road, W1, W2, W3, and W4 are the corresponding weighting coefficients respectively; S50, randomly generate a random number P from 0 to 1, and determine whether the random number P is greater than a preset random threshold. If so, in this iteration, according to the first fitness corresponding to the population, select the population whose first fitness is less than or equal to the preset first fitness threshold, and select the first three populations with the smallest second fitness from the populations eliminated at this time, and save them in the standby library; If not, in this iteration, the current population is combined with the population in the backup library to form a new population, and the population corresponding to the current population whose second fitness is greater than the preset second fitness is selected; S60, hybridizing and mutating the selected population to obtain an offspring population; S70, after obtaining the offspring population, continue to execute S30 until a preset number of iterations is met; S80, outputting the offspring population as the sampling optimal solution set corresponding to each grid, and identifying the setting position data corresponding to each detection point in the corresponding sampling optimal solution set according to the sampling optimal solution set, and forming a corresponding detection point set.

3. The system for extracting spatial distribution characteristics of trace elements in karst groundwater according to claim 2, characterized in that: The server also includes: A correlation calculation module is used to calculate the trace element correlation between each grid and the adjacent grids in the other four directions according to the detection points corresponding to each grid and the corresponding setting position data, based on a preset trace element correlation calculation strategy, and select the adjacent grid with the largest trace element correlation, and classify the adjacent grid and the corresponding grid into the same type of grids; The random selection module is used to randomly select a detection point corresponding to a grid in the same grid as the detection point of all grids in the same grid after completing the division of all grids. After the detection points of all grids in the same grid are determined, the detection points corresponding to all grids in the same grid are collected to form a final detection point set and input into the sampling module.

4. The system for extracting spatial distribution characteristics of trace elements in karst groundwater according to claim 3, characterized in that: The preset trace element correlation calculation strategy is: S100, randomly selecting a grid, and based on the grid, determining each adjacent grid in other four directions; S200, according to the selected grid and other four adjacent grids, retrieve the historical meteorological data corresponding to each grid and the ground building data corresponding to each grid from the historical database; S300, calculating the trace element correlation between the selected grid and the other four adjacent grids based on the historical meteorological data and ground building data corresponding to the selected grid and the other four adjacent grids and based on a preset trace element correlation calculation formula; The preset trace element correlation calculation formula is: In the formula, G is the correlation of trace elements, P ab is the building correlation between grid a and grid b, Q ab is the meteorological correlation between grid a and grid b, ρ ar is the building density corresponding to the rth subgrid in grid a, t ar is the historical temperature value corresponding to the rth subgrid in grid a, and h is the total number of subgrids corresponding to the grid.

5. The system for extracting spatial distribution characteristics of trace elements in karst groundwater according to claim 4, characterized in that: The sampling data includes soil sample data, water sample data, rock sample data and plant sample data corresponding to the detection points in the area to be detected.

6. A method for extracting spatial distribution characteristics of trace elements in karst groundwater, characterized by: A system for extracting spatial distribution characteristics of trace elements in karst groundwater according to any one of claims 1 to 5 above.

Images