A water quality detection method based on stratified sampling and related device

Abstract

The present invention discloses a water quality detection method and related device based on stratified sampling, and relates to the technical field of water quality detection. The method comprises: pre-processing regional data of a water body area to be detected; generating a water quality parameter classification map based on a parameter distribution characteristic map; performing clustering region division based on a local Moran index generated by performing spatial autocorrelation analysis on the water quality parameter classification map and the regional data; determining a stratified sampling scheme based on the water quality parameter classification map and the pre-processed regional data; performing algae distribution analysis based on water body sample monitoring data using an algae index analysis model; performing water quality pollution analysis based on a pollution source-water quality association model using water body sample monitoring data combined with rainfall data; and performing a comprehensive water quality assessment based on algae distribution information and water quality pollution analysis data. The present invention fully guarantees the comprehensiveness and accuracy of water quality detection in the water body area to be detected, and enables water quality detection to achieve a more ideal effect.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of water quality detection, and in particular to a water quality detection method based on stratified sampling and related devices. Background Art

[0002] The prevention and control of water pollution is one of the major key issues of environmental protection. Therefore, it is necessary to continuously test water quality and timely discover water pollution problems. In water quality testing and processing, in order to improve the accuracy of water quality testing, stratified sampling methods are usually used to obtain water sample monitoring data for water quality testing. In the current water quality testing methods, the regional division of the water body area to be tested mainly adopts the qualitative analysis method, but the qualitative analysis method is too dependent on the cognitive level of experts, which makes the division of the water body area lack sufficient objectivity, resulting in insufficient reliability and stability of the regional division of the water body area. For the stratified sampling scheme, the sampling points are currently determined only by the distribution characteristics of water quality parameters, but the accuracy of the sampling points determined by this method is not high, which makes the determined stratified sampling scheme insufficiently representative, resulting in insufficient objectivity and representativeness of the water body monitoring data obtained by the stratified sampling scheme, resulting in excessive errors in water quality testing. As for the water quality analysis of water quality sample monitoring data, there is a lack of correlation analysis between pollution sources and water quality, and a lack of specific situation analysis on the distribution of algae in the water body. As a result, the water quality testing in the water area to be tested is insufficiently comprehensive and accurate, making it impossible for the water quality testing to achieve the expected results. Summary of the invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides a water quality detection method and related devices based on stratified sampling, so that the comprehensiveness and accuracy of water quality detection in the water body area to be tested are fully guaranteed, and the water quality detection can achieve a more ideal effect.

[0004] In order to solve the above technical problems, the present invention provides a water quality detection method based on stratified sampling, the method comprising:

[0005] Acquire regional data of a water body area to be measured, and preprocess the regional data to obtain preprocessed regional data;

[0006] Obtaining several parameter distribution characteristic maps of the water body area to be tested, and generating a water quality parameter classification map based on the several parameter distribution characteristic maps;

[0007] Performing spatial autocorrelation analysis based on the water quality parameter classification map and pre-processed regional data to obtain a local Moran's index, and performing clustering regional division on the water body area to be tested based on the local Moran's index to obtain a plurality of sampling unit areas;

[0008] Determining corresponding stratified sampling schemes in a plurality of sampling unit areas based on the water quality parameter classification diagram and preprocessed regional data;

[0009] Acquire water sample monitoring data in each sampling unit area based on the corresponding stratified sampling scheme, and analyze algae distribution using an algae index analysis model based on the water sample monitoring data to obtain algae distribution information;

[0010] Constructing a pollution source-water quality correlation model, and performing water quality pollution analysis based on the pollution source-water quality correlation model using the water body sample monitoring data combined with rainfall data to obtain water quality pollution analysis data;

[0011] A comprehensive water quality assessment is performed based on the algae distribution information and water pollution analysis data to obtain a comprehensive water quality assessment result.

[0012] Optionally, the step of obtaining a plurality of parameter distribution characteristic maps of the water body area to be measured, and generating a water quality parameter classification map based on the plurality of parameter distribution characteristic maps, includes:

[0013] Acquire water body reflectivity data, generate water quality parameter characteristic bands based on the water body reflectivity data, and generate a plurality of parameter distribution characteristic graphs based on the water quality parameter characteristic bands;

[0014] Based on a number of parameter distribution characteristic maps, non-water quality pixels are removed to obtain a number of parameter distribution characteristic maps after the non-water quality pixels are removed;

[0015] Based on the unsupervised classification method, a water quality parameter classification map is generated using several parameter distribution characteristic maps after non-water quality pixels are removed.

[0016] Optionally, the spatial autocorrelation analysis is performed based on the water quality parameter classification map and the pre-processed regional data to obtain a local Moran's index, and the water body area to be tested is clustered and divided based on the local Moran's index to obtain a plurality of sampling unit areas, including:

[0017] Performing global spatial autocorrelation analysis based on the water quality parameter classification map and pre-processed regional data to obtain a global Moran index;

[0018] Constructing a spatial weight matrix based on the spatial information encoder, and constructing an adjacency matrix based on the spatial weight matrix using a k-nearest neighbor algorithm;

[0019] Performing incremental spatial autocorrelation analysis based on the global Moran's index combined with the proximity matrix to obtain a local Moran's index;

[0020] Performing spatial clustering based on the local Moran index to obtain a spatial correlation local clustering graph;

[0021] A preset zoning index is obtained, an elevation grid is generated based on the preset zoning index, and the elevation grid is used to perform clustering area division on the water body area to be measured based on the spatial correlation local clustering map to obtain a plurality of sampling unit areas.

[0022] Optionally, determining a corresponding stratified sampling scheme in a plurality of sampling unit areas based on the water quality parameter classification map and the preprocessed regional data includes:

[0023] Generate a pixel value cumulative frequency distribution map based on the water quality parameter classification map, generate an initial sampling point based on the pixel value cumulative frequency distribution map, and perform point error analysis based on the initial sampling point combined with a number of parameter distribution characteristic maps to obtain a sampling point error map;

[0024] Based on the Voronoi diagram, distribution characteristic aggregation analysis is performed on the water quality parameter classification diagram and the pre-processed regional data to obtain distribution characteristic aggregation information;

[0025] Based on the distribution feature aggregation information, the initial sampling points are encrypted using a particle swarm clustering method to obtain encrypted sample data;

[0026] The target sampling points for each sampling unit area are generated based on the sampling point error map and the sample point encryption data, the sampling water body depth is obtained based on the water quality parameter classification map, and the corresponding stratified sampling plan is generated based on the target sampling points and the sampling water body depth.

[0027] Optionally, the algae distribution analysis is performed based on the water sample monitoring data using an algae index analysis model to obtain algae distribution information, including:

[0028] Acquire historical algae index sample data, set a time series of the historical algae index sample data, establish a linear regression model based on the time series, and construct an algae index analysis model based on the linear regression model;

[0029] Based on the algae index analysis model, the water sample monitoring data is used to perform algae distribution area analysis and algae type analysis to obtain algae distribution area information and algae type information;

[0030] Based on the algae type information and the algae distribution area information, data simulation analysis is performed using the algae growth trend data to obtain algae growth simulation analysis data;

[0031] Based on the algae type information and the algae distribution area information combined with the algae growth simulation analysis data, an algae distribution analysis is performed to obtain algae distribution information.

[0032] Optionally, the construction of a pollution source-water quality correlation model, and performing water quality pollution analysis based on the pollution source-water quality correlation model using the water sample monitoring data in combination with rainfall data to obtain water quality pollution analysis data, include:

[0033] Setting an objective function based on a preset regression tree combined with historical water pollution trend information, and setting a penalty term based on the objective function;

[0034] Based on the greedy algorithm, the gradient boosting tree generated by the XGboost algorithm is divided into nodes to obtain the gradient boosting tree after node division;

[0035] Constructing an association regression model, and constructing a pollution source-water quality association model based on the association regression model and the gradient boosting tree after node division combined with the objective function and the penalty term;

[0036] Based on the water sample monitoring data, water quality pollution information is output using the pollution source-water quality association model, rainfall runoff flow information is obtained based on rainfall data, and water quality pollution analysis is performed based on the water quality pollution information and rainfall runoff flow information to obtain water quality pollution analysis data.

[0037] Optionally, the comprehensive water quality assessment is performed based on the algae distribution information and the water pollution analysis data to obtain a comprehensive water quality assessment result, including:

[0038] The algae distribution information and water pollution analysis data of each sampling unit area are compared with the preset expected threshold value to obtain a comparison result, the water quality grade of each sampling unit area is divided based on the comparison result, and a comprehensive water quality assessment result is generated based on the water quality grade and the comparison result.

[0039] In addition, the present invention also provides a water quality detection device based on stratified sampling, the device comprising:

[0040] Regional data preprocessing module: used to obtain regional data of the water body area to be tested, preprocess the regional data, and obtain preprocessed regional data;

[0041] Water quality parameter classification map generation module: used to obtain several parameter distribution characteristic maps of the water body area to be tested, and generate a water quality parameter classification map based on the several parameter distribution characteristic maps;

[0042] Sampling unit area division module: used for performing spatial autocorrelation analysis based on the water quality parameter classification map and the pre-processed regional data to obtain the local Moran index, and performing cluster area division on the water body area to be tested based on the local Moran index to obtain a number of sampling unit areas;

[0043] A stratified sampling scheme determination module is used to determine a corresponding stratified sampling scheme in a plurality of sampling unit areas based on the water quality parameter classification diagram and the pre-processed regional data;

[0044] Algae distribution analysis module: used to obtain water sample monitoring data in each sampling unit area based on the corresponding stratified sampling scheme, and to analyze the algae distribution using the algae index analysis model based on the water sample monitoring data to obtain algae distribution information;

[0045] Water pollution analysis module: used to construct a pollution source-water quality correlation model, and to perform water pollution analysis based on the pollution source-water quality correlation model using the water sample monitoring data combined with rainfall data to obtain water pollution analysis data;

[0046] Water quality comprehensive assessment module: used to conduct a comprehensive water quality assessment based on the algae distribution information and water pollution analysis data to obtain a comprehensive water quality assessment result.

[0047] In addition, the present invention also provides an electronic device, which includes a processor and a memory, wherein the memory is used to store instructions, and the processor is used to call the instructions in the memory so that the electronic device executes the above-mentioned water quality detection method based on stratified sampling.

[0048] In addition, the present invention also provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the above-mentioned water quality detection method based on stratified sampling.

[0049] In an embodiment of the present invention, the local Moran index generated by the spatial autocorrelation analysis of the water quality parameter classification map and the pre-processed regional data is used to perform cluster regional division on the water body area to be tested, and a number of sampling unit areas are obtained, so that the accuracy and objectivity of the regional division are higher, and the reliability of the regional division is guaranteed. The target sampling points of each sampling unit area are generated based on the sampling point error map and the sample point encryption data, so that the accuracy of the determined sampling points is higher, the lack of representativeness of the stratified sampling scheme is avoided, and the spatial representativeness of the sampling points is effectively improved. Based on the water body sample monitoring data, the algae index analysis model is used to analyze the algae distribution, and based on the pollution source-water quality association model, the water body sample monitoring data is combined with rainfall data to analyze water pollution. The pollution source-water quality association model can more accurately analyze water pollution information, enhance the relevance of water quality and pollution source to water quality analysis, and use the algae index analysis model to achieve high-precision algae distribution analysis. Thereby, the comprehensiveness and accuracy of water quality detection in the water body area to be tested are fully guaranteed, and the water quality detection achieves a more ideal effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 Schematic diagram of a water quality detection method based on stratified sampling in an embodiment of the present invention;

[0052] Figure 2 Schematic diagram of the structure of a water quality detection device based on stratified sampling in an embodiment of the present invention;

[0053] Figure 3 It is a schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] Embodiment 1

[0056] See also Figure 1 , Figure 1 It is a flow chart of a water quality detection method based on stratified sampling in an embodiment of the present invention.

[0057] like Figure 1 As shown, a water quality detection method based on stratified sampling, the method comprising:

[0058] S11: Acquire regional data of the water body area to be measured, preprocess the regional data, and obtain preprocessed regional data;

[0059] In the specific implementation process of the present invention, regional data of the water body area to be measured is obtained, and the regional data includes the spatial size, water depth and remote sensing image data of the water body area to be measured. The regional data is preprocessed, and outlier removal processing is performed on the regional data to obtain regional data after outlier removal processing, and noise reduction processing is performed on the regional data after outlier removal processing. After the noise reduction processing is completed, the preprocessing of the regional data is completed, and the preprocessed regional data can be obtained.

[0060] S12: obtaining a plurality of parameter distribution characteristic maps of the water body area to be measured, and generating a water quality parameter classification map based on the plurality of parameter distribution characteristic maps;

[0061] In the specific implementation process of the present invention, the method of obtaining several parameter distribution characteristic maps of the water body area to be tested and generating a water quality parameter classification map based on the several parameter distribution characteristic maps includes: obtaining water body reflectivity data, generating water quality parameter characteristic bands based on the water body reflectivity data, and generating several parameter distribution characteristic maps based on the water quality parameter characteristic bands; removing non-water quality pixels based on the several parameter distribution characteristic maps to obtain several parameter distribution characteristic maps after the non-water quality pixels are removed; generating a water quality parameter classification map based on an unsupervised classification method using several parameter distribution characteristic maps after the non-water quality pixels are removed.

[0062] Specifically, the water body reflectance data can be obtained by processing the remote sensing image data in the regional data through radiation correction and atmospheric correction. Based on the water body reflectance data, the characteristic band of water quality parameters is generated. Since the change of the water quality parameter to be measured will reflect the reflection characteristics of the water body, the characteristic band of water quality parameters is obtained by performing characteristic band analysis on the water body reflectance data. For example, chlorophyll is most sensitive to the red band in the water body reflectance data. Its characteristic band analysis can be performed to obtain its characteristic band. Based on the water quality parameter characteristic band, several parameter distribution characteristic graphs are generated. The band reflectance ratio is generated by using the water quality parameter characteristic band through the band ratio model. The reflection peak area is generated by using the water quality parameter characteristic band through the peak area model. Several parameter distribution characteristic graphs are generated by using the corresponding water quality parameter characteristic band through the reflection peak area and the band reflectance ratio. Non-water quality pixels are removed based on several parameter distribution characteristic graphs, non-water quality pixels are extracted from several parameter distribution characteristic graphs, and the extracted non-water quality pixels are removed based on the mask removal method to obtain several parameter distribution characteristic graphs after the non-water quality pixels are removed. Based on the unsupervised classification method, several parameter distribution characteristic maps after the non-water quality pixels are removed are used to generate a water quality parameter classification map. The unsupervised classification classifies by the difference in category features in the feature space. No prior category standard is required for unsupervised classification. At the same time, it can reduce human errors and improve classification accuracy. A covariance matrix is ​​constructed. According to the covariance matrix, several parameter distribution characteristic maps after the non-water quality pixels are removed are classified by unsupervised classification combined with preset clustering evaluation indicators to obtain a water quality parameter classification map.

[0063] S13: performing spatial autocorrelation analysis based on the water quality parameter classification map and the pre-processed regional data to obtain a local Moran's index, and performing clustering regional division on the water body area to be tested based on the local Moran's index to obtain a plurality of sampling unit areas;

[0064] In the specific implementation process of the present invention, the spatial autocorrelation analysis is performed based on the water quality parameter classification map and the preprocessed regional data to obtain the local Moran's index, and the water body area to be tested is clustered based on the local Moran's index to obtain a number of sampling unit areas, including: performing a global spatial autocorrelation analysis based on the water quality parameter classification map and the preprocessed regional data to obtain the global Moran's index; constructing a spatial weight matrix based on the spatial information encoder, and constructing an adjacency matrix based on the spatial weight matrix using the k-neighbor algorithm; performing incremental spatial autocorrelation analysis based on the global Moran's index combined with the adjacency matrix to obtain the local Moran's index; performing spatial clustering based on the local Moran's index to obtain a spatially associated local clustering map; obtaining a preset zoning index, generating an elevation grid based on the preset zoning index, and using the elevation grid to cluster the water body area to be tested based on the spatially associated local clustering map to obtain a number of sampling unit areas.

[0065] Specifically, a global spatial autocorrelation analysis is performed based on the water quality parameter classification map and the preprocessed regional data. The global spatial autocorrelation analysis is an overall analysis of the study area, reflecting the overall spatial clustering pattern of the entire area. A spatial weight matrix is ​​constructed based on the water quality parameter classification map and the preprocessed regional data, and the weights of adjacent elements and non-adjacent elements in the spatial weight matrix are defined. The aggregation of all spatial weights in the spatial weight matrix is ​​calculated, the average of all element weights in the spatial matrix is ​​calculated, and the first deviation of the adjacent element weights from the average value and the second deviation of the non-adjacent element weights from the average value are calculated. The global Moran's index is calculated based on the aggregation, first deviation and second deviation of all spatial weights to obtain the global Moran's index, which can reflect the correlation of global spatial units. Based on the spatial information encoder, a spatial weight matrix is ​​constructed. The spatial information encoder includes a sine and cosine function and a fully connected neural network to construct a spatial coordinate matrix. The spatial coordinate matrix is ​​input into the spatial information encoder to obtain a spatial weight matrix, and an adjacency matrix is ​​constructed based on the spatial weight matrix using the k-nearest neighbor algorithm. A spatial graph is constructed based on the spatial weight matrix using the k-nearest neighbor algorithm. The adjacency matrix is ​​constructed through the spatial graph. The k-nearest neighbor algorithm does not rely on the specific distribution and feature selection of the data, so it can handle multiple categories of problems and has better data adaptability. Based on the global Moran's index and the proximity matrix, an incremental spatial autocorrelation analysis is performed. The incremental spatial autocorrelation analysis can calculate the local Moran's index by calculating the increasing spatial distance, calculate the standard score of the global Moran's index, obtain the weight of each element of the adjacency matrix and the average value of all element weights, calculate the deviation of each element weight from the average value of all element weights, calculate the local Moran's index according to the weight of each element of the adjacency matrix, the average value of all element weights, the deviation and the standard score, and obtain the local Moran's index. The calculation expression of the local Moran's index is:

[0066]

[0067] Where I is the local Moran index, f is the standard score, and z i is the deviation of the weight of the ith element in the adjacency matrix from the average value, z j is the deviation of the jth element weight from the average value in the adjacency matrix, n is the number of elements in the adjacency matrix, ω i, is the difference between the weight of the ith element and the weight of the jth element, and j and i are the element numbers in the adjacency matrix. Spatial clustering is performed based on the local Moran index, that is, spatial clustering is performed according to the local Moran index by preset local association indicators, and the preset local association indicators are different spatial aggregation modes to obtain a spatial association local clustering map. The preset zoning index is obtained in the local database, and the preset zoning index may include the regional grid division size and the elevation of the grid. The elevation grid is generated based on the preset zoning index, and the data is interpolated according to the preset zoning index to obtain the initial elevation grid. The initial elevation grid is trimmed based on the resampling technology to obtain the final elevation grid. Based on the spatial association local clustering map, the elevation grid is used to perform clustering area division on the water body area to be tested, and the fitting metric of the linear regression model and the preset clustering pedigree map are used in combination with the spatial association local clustering map and the elevation grid to perform attribute clustering to obtain the optimal number of clusters. The water body area to be tested is clustered according to the optimal number of clusters to obtain a number of sampling unit areas, thereby improving the accuracy of cluster area division and reducing the complexity of cluster area division. Based on the local Moran's index generated by spatial autocorrelation analysis of the water quality parameter classification map and the pre-processed regional data, the water body area to be tested is clustered and divided into several sampling unit areas, which makes the regional division more accurate and objective and ensures the reliability of regional division.

[0068] S14: determining a corresponding stratified sampling scheme in a plurality of sampling unit areas based on the water quality parameter classification diagram and the preprocessed regional data;

[0069] In the specific implementation process of the present invention, the corresponding stratified sampling scheme is determined in several sampling unit areas based on the water quality parameter classification map and the preprocessed regional data, including: generating a pixel value cumulative frequency distribution map based on the water quality parameter classification map, generating initial sampling points based on the pixel value cumulative frequency distribution map, and performing point error analysis based on the initial sampling points in combination with several parameter distribution characteristic maps to obtain a sampling point error map; performing distribution feature aggregation analysis on the water quality parameter classification map and the preprocessed regional data based on the Voronoi map to obtain distribution feature aggregation information; encrypting the initial sampling points using a particle swarm clustering method based on the distribution feature aggregation information to obtain encrypted sample data; generating target sampling points for each sampling unit area based on the sampling point error map and the encrypted sample data, obtaining the sampling water body depth based on the water quality parameter classification map, and generating a corresponding stratified sampling scheme based on the target sampling points and the sampling water body depth.

[0070] Specifically, a pixel value cumulative frequency distribution map is generated based on the water quality parameter classification map, and pixel value statistics are performed on the water quality parameter classification map, and a pixel value cumulative frequency distribution map is obtained based on the pixel value statistics. An initial sampling point is generated based on the pixel value cumulative frequency distribution map, and the pixel value cumulative frequency distribution map is divided at equal distances to obtain a number of representative pixels, and the position of each representative pixel is obtained. Density clustering is performed according to the position of each representative pixel to obtain the initial sampling point. Point error analysis is performed based on the initial sampling point in combination with a number of parameter distribution characteristic maps, and each pixel in each parameter distribution characteristic map is subtracted from the initial sampling point based on the nearest principle to obtain a number of adjacent space point error maps, and a number of adjacent space point error maps are superimposed to obtain a sampling point error map. Distribution feature aggregation analysis is performed on the water quality parameter classification map and the preprocessed regional data based on the Voronoi diagram, and polygon data calculation is performed using the water quality parameter classification map and the preprocessed regional data through the Voronoi diagram, and the polygon data calculation result is compared with the preset calculation threshold to obtain distribution feature aggregation information. Based on the distribution feature aggregation information, the initial sampling points are encrypted by using a particle swarm clustering method, a mapping relationship between the initial sampling points and the particles is obtained, a binary particle swarm is initialized according to the mapping relationship between the initial sampling points and the particles, the fitness of each particle in the binary particle swarm is calculated, the individual historical optimal position of the particle and the global historical optimal position of the particle swarm are determined, the binary particle position is updated according to the individual historical optimal position and the global historical optimal position, and after the binary particle position is updated, it is determined whether a termination condition is met. If the termination condition is met, a maximum number of iterations is used as the algorithm termination condition, and the sample encryption is completed. If the termination condition is not met, a new round of iterative update is performed again to obtain sample encryption data. Sample encryption by using a particle swarm algorithm can ensure the accuracy of the clustered distribution of the sampling points, improve the representativeness of the sampling points, and also improve the efficiency of the sampling point adjustment. The target sampling points of each sampling unit area are generated based on the sampling point error map and the sample point encryption data, the initial sampling points are corrected by the sampling point error map, and the initial sampling points are adjusted by the sample point encryption data to obtain the target sampling points of each sampling unit area, the sampling water body depth is obtained based on the water quality parameter classification map, the sampling depth is set by the water quality pixel in the water quality parameter classification map, and a corresponding stratified sampling scheme is formed based on the target sampling points and the sampling water body depth. The target sampling points of each sampling unit area are generated based on the sampling point error map and the sample point encryption data, so that the accuracy of the determined sampling points is higher, the lack of representativeness of the stratified sampling scheme is avoided, and the spatial representativeness of the sampling points is effectively improved.

[0071] S15: acquiring water sample monitoring data in each sampling unit area based on the corresponding stratified sampling scheme, and performing algae distribution analysis based on the water sample monitoring data using an algae index analysis model to obtain algae distribution information;

[0072] In the specific implementation process of the present invention, the algae distribution analysis is performed based on the water sample monitoring data using an algae index analysis model to obtain algae distribution information, including: obtaining historical algae index sample data, setting a time series of the historical algae index sample data, establishing a linear regression model based on the time series, and constructing an algae index analysis model based on the linear regression model; based on the algae index analysis model, the water sample monitoring data is used to perform algae distribution area analysis and algae type analysis to obtain algae distribution area information and algae type information; based on the algae type information and algae distribution area information, algae growth trend data is used to perform data simulation analysis to obtain algae growth simulation analysis data; based on the algae type information and algae distribution area information, the algae distribution analysis is performed in combination with the algae growth simulation analysis data to obtain algae distribution information.

[0073] Specifically, water sample monitoring data in each sampling unit area is obtained based on the corresponding stratified sampling scheme, and the corresponding equipment is controlled according to the corresponding stratified sampling scheme to obtain water sample monitoring data in each sampling unit area. The corresponding underwater camera equipment can be controlled to obtain water image data according to the sampling water depth in the stratified sampling scheme, and the corresponding sensor equipment is controlled to obtain water component data according to the target sampling point in the stratified sampling scheme. The historical algae index sample data is obtained in the local database, and the time series of the historical algae index sample data is set. A linear regression model is established based on the time series, that is, a linear regression model is established using a linear regression formula according to the number of moments in the time series, and an algae index analysis model is constructed based on the linear regression model. The dynamic characteristic distribution of algae distribution is calculated according to the linear regression model, and the algae index analysis model is constructed according to the dynamic characteristic distribution of algae distribution. Based on the algae index analysis model, the water sample monitoring data is used to perform algae distribution area analysis and algae type analysis, and the water image data in the water sample monitoring data is subjected to color feature value extraction to obtain the target color feature value, and the target color feature value is input into the algae index analysis model to perform algae distribution area analysis to obtain algae distribution area information, and the target color feature value is subjected to similarity analysis using the color contrast feature value in the dynamic feature distribution in the algae index analysis model to obtain algae type information. The algae index analysis model can realize high-precision algae distribution situation analysis. Based on the algae type information and algae distribution area information, the algae growth trend data is used to perform data simulation analysis, and the algae type information can be used to determine the type of algae, the algae distribution area information can be used to determine the distribution quantity of the corresponding algae, and the algae growth trend data can be used to determine the reproduction rate of the corresponding algae. The algae growth trend information can be obtained in the algae database, and the algae type information, the algae distribution area information and the algae growth trend data are imported into the algae growth simulation model to obtain the algae growth simulation analysis data. Based on the algae type information and the algae distribution area information combined with the algae growth simulation analysis data, an algae distribution analysis is performed. Through the algae type information and the algae distribution area information combined with the algae growth simulation analysis data, the reproduction rate and reproduction number of the corresponding type of algae in unit time can be known, thereby obtaining the algae distribution information.

[0074] S16: constructing a pollution source-water quality correlation model, and performing water quality pollution analysis based on the pollution source-water quality correlation model using the water body sample monitoring data combined with rainfall data to obtain water quality pollution analysis data;

[0075] In the specific implementation process of the present invention, the pollution source-water quality association model is constructed, and the water quality pollution analysis is performed based on the pollution source-water quality association model using the water body sample monitoring data combined with rainfall data to obtain water quality pollution analysis data, including: setting an objective function based on a preset regression tree combined with historical water quality pollution trend information, and setting a penalty item based on the objective function; performing node division on the gradient boosting tree generated by the XGboost algorithm based on a greedy algorithm to obtain a gradient boosting tree after node division; constructing an associated regression model, and constructing a pollution source-water quality association model based on the associated regression model and the gradient boosting tree after node division combined with the objective function and the penalty item; outputting water quality pollution information based on the water body sample monitoring data using the pollution source-water quality association model, obtaining rainfall runoff flow information based on rainfall data, and performing water quality pollution analysis based on the water quality pollution information and rainfall runoff flow information to obtain water quality pollution analysis data.

[0076] Specifically, an objective function is set based on a preset regression tree combined with historical water pollution trend information, the historical water pollution trend information is input into the preset regression tree to obtain a water pollution prediction value, an objective function is set according to the water pollution prediction value, a penalty item is set based on the objective function, and in order to avoid the risk of overfitting, the number of leaf nodes of the gradient boosting tree is limited and a penalty item is set. Based on the greedy algorithm, the gradient boosting tree generated by the extreme gradient boosting (eXtreme Gradient Boosting, XGboost) algorithm is node-partitioned. The XGboost algorithm is a boosting tree model that integrates many tree models to form a very strong classifier. The integration is performed by using several classification and regression trees, wherein the classification and regression trees are continuously feature-split, and the classification and regression trees are grown according to the continuous feature splitting to obtain the gradient boosting tree. The greedy algorithm is used to divide the gradient boosting tree into left and right subtrees, and the branches of the tree are iteratively generated from the leaves. The split points are screened out by iteratively generating the branches of the tree from the leaves, and the gradient boosting tree is node-partitioned according to the screened split points to obtain the gradient boosting tree after node division. The greedy algorithm is suitable for tree structures with large computational complexity, so the greedy algorithm can improve the efficiency and accuracy of node division. Construct an association regression model, construct a mechanism model based on historical water quality monitoring data, run the mechanism model, generate water quality prediction and evaluation data using water sample monitoring data through the mechanism model, input the water quality prediction and evaluation data as data input features into the trained deep learning model, obtain mechanism water quality prediction and evaluation data, obtain historical water quality pollutant data, generate a fusion decision model using the decision tree model through the historical water quality pollutant data, and construct an association regression model through the mechanism water quality prediction and evaluation data and the fusion decision model. Based on the association regression model and the gradient boosting tree after node division combined with the objective function and penalty term, a pollution source-water quality association model is constructed, root cause node analysis is performed on the gradient boosting tree after node division through the association regression model, and the gradient boosting tree after node division is optimized according to the root cause node analysis results to obtain an optimized gradient boosting tree, regularize the penalty term to obtain a regularized penalty term, generate an optimized objective function according to the objective function and the regularized penalty term, and generate a pollution source-water quality association model according to the optimized objective function combined with the optimized gradient boosting tree combined with the association regression model. Based on the water sample monitoring data, the water quality pollution information is outputted using the pollution source-water quality association model. The water quality pollution information may include the type of water quality pollutants, the amount of pollutants, and the transparency of the water body, etc. The rainfall runoff flow information is obtained based on the rainfall data, and the rainfall runoff flow information is obtained through the rainfall runoff accumulation function according to the rainfall data; based on the water quality pollution information and the rainfall runoff flow information, water quality pollution analysis is performed, the rainfall runoff pollution load is determined through the water quality pollution information and the rainfall runoff flow information, and water quality pollution analysis data is generated according to the rainfall runoff pollution load combined with the water quality pollution information.The pollution source-water quality correlation model can more accurately analyze water pollution information and enhance the correlation between water quality and pollution sources in water quality analysis.

[0077] S17: Perform a comprehensive water quality assessment based on the algae distribution information and water pollution analysis data to obtain a comprehensive water quality assessment result.

[0078] In the specific implementation process of the present invention, the comprehensive water quality assessment is performed based on the algae distribution information and water pollution analysis data to obtain a comprehensive water quality assessment result, including: comparing the algae distribution information and water pollution analysis data of each sampling unit area with a preset expected threshold to obtain a comparison result, dividing the water quality grade of each sampling unit area based on the comparison result, and generating a comprehensive water quality assessment result based on the water quality grade and the comparison result.

[0079] Specifically, the algae distribution information and water pollution analysis data of each sampling unit area are compared with the preset expected threshold value to obtain a comparison result, that is, the algae distribution information and water pollution analysis data of the sampling unit area are calculated by difference with the preset expected threshold value to obtain the corresponding difference value, and the water quality level of each sampling unit area is divided based on the comparison result. Each difference value has a corresponding water quality level stored in the database, and the water quality level includes severely polluted, slightly polluted, qualified water quality and excellent water quality detection, and a comprehensive water quality assessment result is generated based on the water quality level and the comparison result. A corresponding water quality treatment plan is generated according to the comprehensive water quality assessment result, and water quality treatment is performed on the water body area to be tested according to the water quality treatment plan.

[0080] In an embodiment of the present invention, the local Moran index generated by the spatial autocorrelation analysis of the water quality parameter classification map and the pre-processed regional data is used to perform cluster regional division on the water body area to be tested, and a number of sampling unit areas are obtained, so that the accuracy and objectivity of the regional division are higher, and the reliability of the regional division is guaranteed. The target sampling points of each sampling unit area are generated based on the sampling point error map and the sample point encryption data, so that the accuracy of the determined sampling points is higher, the lack of representativeness of the stratified sampling scheme is avoided, and the spatial representativeness of the sampling points is effectively improved. Based on the water body sample monitoring data, the algae index analysis model is used to analyze the algae distribution, and based on the pollution source-water quality association model, the water body sample monitoring data is combined with rainfall data to analyze water pollution. The pollution source-water quality association model can more accurately analyze water pollution information, enhance the relevance of water quality and pollution source to water quality analysis, and use the algae index analysis model to achieve high-precision algae distribution analysis. Thereby, the comprehensiveness and accuracy of water quality detection in the water body area to be tested are fully guaranteed, and the water quality detection achieves a more ideal effect.

[0081] Embodiment 2

[0082] See also Figure 2 , Figure 2 It is a schematic diagram of the structural composition of a water quality detection device based on stratified sampling in an embodiment of the present invention.

[0083] like Figure 2 As shown, a water quality detection device based on stratified sampling, the device comprises:

[0084] The regional data preprocessing module 21 is used to obtain the regional data of the water body area to be measured, and preprocess the regional data to obtain the preprocessed regional data;

[0085] The water quality parameter classification diagram generation module 22 is used to obtain a plurality of parameter distribution characteristic diagrams of the water body area to be measured, and generate a water quality parameter classification diagram based on the plurality of parameter distribution characteristic diagrams;

[0086] The sampling unit area division module 23 is used to perform spatial autocorrelation analysis based on the water quality parameter classification map and the pre-processed regional data to obtain a local Moran index, and to perform cluster area division on the water body area to be tested based on the local Moran index to obtain a plurality of sampling unit areas;

[0087] A stratified sampling scheme determination module 24 is used to determine a corresponding stratified sampling scheme in a plurality of sampling unit areas based on the water quality parameter classification diagram and the pre-processed regional data;

[0088] Algae distribution analysis module 25: used to obtain water sample monitoring data in each sampling unit area based on the corresponding stratified sampling scheme, and perform algae distribution analysis based on the water sample monitoring data using an algae index analysis model to obtain algae distribution information;

[0089] Water pollution analysis module 26: used to construct a pollution source-water quality correlation model, and to perform water pollution analysis based on the pollution source-water quality correlation model using the water sample monitoring data combined with rainfall data to obtain water pollution analysis data;

[0090] The water quality comprehensive assessment module 27 is used to conduct a comprehensive water quality assessment based on the algae distribution information and the water pollution analysis data to obtain a comprehensive water quality assessment result.

[0091] In the specific implementation process of the present invention, the specific implementation method of the device item can refer to the implementation method of the above-mentioned method item, which will not be repeated here.

[0092] In an embodiment of the present invention, the local Moran index generated by the spatial autocorrelation analysis of the water quality parameter classification map and the pre-processed regional data is used to perform cluster regional division on the water body area to be tested, and a number of sampling unit areas are obtained, so that the accuracy and objectivity of the regional division are higher, and the reliability of the regional division is guaranteed. The target sampling points of each sampling unit area are generated based on the sampling point error map and the sample point encryption data, so that the accuracy of the determined sampling points is higher, the lack of representativeness of the stratified sampling scheme is avoided, and the spatial representativeness of the sampling points is effectively improved. Based on the water body sample monitoring data, the algae index analysis model is used to analyze the algae distribution, and based on the pollution source-water quality association model, the water body sample monitoring data is combined with rainfall data to analyze water pollution. The pollution source-water quality association model can more accurately analyze water pollution information, enhance the relevance of water quality and pollution source to water quality analysis, and use the algae index analysis model to achieve high-precision algae distribution analysis. Thereby, the comprehensiveness and accuracy of water quality detection in the water body area to be tested are fully guaranteed, and the water quality detection achieves a more ideal effect.

[0093] A computer-readable storage medium provided in an embodiment of the present invention stores a computer program on the computer-readable storage medium, and when the program is executed by a processor, a water quality detection method based on stratified sampling of any one of the above embodiments is implemented. Wherein, the computer-readable storage medium includes but is not limited to any type of disk (including floppy disk, hard disk, optical disk, CD-ROM, and magneto-optical disk), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (EraSable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic card or optical card. That is, the storage device includes any medium that stores or transmits information in a readable form by a device (for example, a computer, a mobile phone), which can be a read-only memory, a disk or an optical disk, etc.

[0094] Embodiment 3

[0095] See also Figure 3 , Figure 3 It is a schematic diagram of the structure of an electronic device in an embodiment of the present invention.

[0096] The embodiment of the present invention further provides an electronic device, such as Figure 3As shown, the electronic device includes a memory 31, a processor 33, and a computer program 32 stored in the memory 31 and executable on the processor 33. Those skilled in the art will appreciate that Figure 3 The electronic device shown does not constitute a limitation on all devices, and may include more or fewer components than shown, or combine certain components. The memory 31 can be used to store the computer program 32 and various functional modules, and the processor 33 runs the computer program 32 stored in the memory 31, thereby executing various functional applications and data processing of the device. The memory can be an internal memory or an external memory, or include both internal and external memories. The internal memory can include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, or a random access memory. The external memory can include a hard disk, a floppy disk, a ZIP disk, a U disk, a magnetic tape, etc. The processor 33 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general processor may be a microprocessor, a single chip microcomputer or the processor 33 may also be any conventional processor, etc. The processor and memory disclosed in the present invention include but are not limited to these types of processors and memories. The processor and memory disclosed in the present invention are only examples and not limitations.

[0097] As an embodiment, the electronic device includes: one or more processors 33, a memory 31, and one or more computer programs 32, wherein the one or more computer programs 32 are stored in the memory 31 and are configured to be executed by the one or more processors 33, and the one or more computer programs 32 are configured to execute the water quality detection method based on stratified sampling in any of the above-mentioned embodiments. For the specific implementation process, please refer to the above-mentioned embodiments, which will not be repeated here.

[0098] In an embodiment of the present invention, the local Moran index generated by the spatial autocorrelation analysis of the water quality parameter classification map and the pre-processed regional data is used to perform cluster regional division on the water body area to be tested, and a number of sampling unit areas are obtained, so that the accuracy and objectivity of the regional division are higher, and the reliability of the regional division is guaranteed. The target sampling points of each sampling unit area are generated based on the sampling point error map and the sample point encryption data, so that the accuracy of the determined sampling points is higher, the lack of representativeness of the stratified sampling scheme is avoided, and the spatial representativeness of the sampling points is effectively improved. Based on the water body sample monitoring data, the algae index analysis model is used to analyze the algae distribution, and based on the pollution source-water quality association model, the water body sample monitoring data is combined with rainfall data to analyze water pollution. The pollution source-water quality association model can more accurately analyze water pollution information, enhance the relevance of water quality and pollution source to water quality analysis, and use the algae index analysis model to achieve high-precision algae distribution analysis. Thereby, the comprehensiveness and accuracy of water quality detection in the water body area to be tested are fully guaranteed, and the water quality detection achieves a more ideal effect.

[0099] In addition, the above is a detailed introduction to a water quality detection method based on stratified sampling and related devices provided in an embodiment of the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for a general technician in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A water quality detection method based on stratified sampling, characterized in that: The method comprises: Acquire regional data of a water body area to be measured, and preprocess the regional data to obtain preprocessed regional data; Obtaining several parameter distribution characteristic maps of the water body area to be tested, and generating a water quality parameter classification map based on the several parameter distribution characteristic maps; Performing spatial autocorrelation analysis based on the water quality parameter classification map and pre-processed regional data to obtain a local Moran's index, and performing clustering regional division on the water body area to be tested based on the local Moran's index to obtain a plurality of sampling unit areas; Determining corresponding stratified sampling schemes in a plurality of sampling unit areas based on the water quality parameter classification diagram and preprocessed regional data; Acquire water sample monitoring data in each sampling unit area based on the corresponding stratified sampling scheme, and analyze algae distribution using an algae index analysis model based on the water sample monitoring data to obtain algae distribution information; Constructing a pollution source-water quality correlation model, and performing water quality pollution analysis based on the pollution source-water quality correlation model using the water body sample monitoring data combined with rainfall data to obtain water quality pollution analysis data; Perform a comprehensive water quality assessment based on the algae distribution information and water pollution analysis data to obtain a comprehensive water quality assessment result; Wherein, the method of obtaining a plurality of parameter distribution characteristic maps of the water body area to be measured and generating a water quality parameter classification map based on the plurality of parameter distribution characteristic maps comprises: obtaining water body reflectance data, generating a water quality parameter characteristic band based on the water body reflectance data, and generating a plurality of parameter distribution characteristic maps based on the water quality parameter characteristic band; removing non-water quality pixels based on the plurality of parameter distribution characteristic maps to obtain a plurality of parameter distribution characteristic maps after the non-water quality pixels are removed; generating a water quality parameter classification map based on an unsupervised classification method using a plurality of parameter distribution characteristic maps after the non-water quality pixels are removed; The method of determining a corresponding stratified sampling scheme in a plurality of sampling unit areas based on the water quality parameter classification map and the preprocessed regional data comprises: generating a pixel value cumulative frequency distribution map based on the water quality parameter classification map, generating an initial sampling point based on the pixel value cumulative frequency distribution map, and performing point error analysis based on the initial sampling point in combination with a plurality of parameter distribution characteristic maps to obtain a sampling point error map; performing distribution feature aggregation analysis on the water quality parameter classification map and the preprocessed regional data based on a Voronoi diagram to obtain distribution feature aggregation information; performing sample point encryption on the initial sampling point using a particle swarm clustering method based on the distribution feature aggregation information to obtain sample point encryption data; generating a target sampling point for each sampling unit area based on the sampling point error map and the sample point encryption data, obtaining a sampling water body depth based on the water quality parameter classification map, and generating a corresponding stratified sampling scheme based on the target sampling point and the sampling water body depth.

2. The water quality detection method based on stratified sampling according to claim 1 is characterized in that: The spatial autocorrelation analysis is performed based on the water quality parameter classification map and the pre-processed regional data to obtain the local Moran index, and the water body area to be tested is clustered and divided based on the local Moran index to obtain a plurality of sampling unit areas, including: Performing global spatial autocorrelation analysis based on the water quality parameter classification map and pre-processed regional data to obtain a global Moran index; Constructing a spatial weight matrix based on the spatial information encoder, and constructing an adjacency matrix based on the spatial weight matrix using a k-nearest neighbor algorithm; Performing incremental spatial autocorrelation analysis based on the global Moran's index combined with the adjacency matrix to obtain a local Moran's index; Performing spatial clustering based on the local Moran index to obtain a spatial correlation local clustering graph; A preset zoning index is obtained, an elevation grid is generated based on the preset zoning index, and the elevation grid is used to perform clustering area division on the water body area to be measured based on the spatial correlation local clustering map to obtain a plurality of sampling unit areas.

3. The water quality detection method based on stratified sampling according to claim 1 is characterized in that: The algae distribution analysis is performed based on the water sample monitoring data using an algae index analysis model to obtain algae distribution information, including: Acquire historical algae index sample data, set a time series of the historical algae index sample data, establish a linear regression model based on the time series, and construct an algae index analysis model based on the linear regression model; Based on the algae index analysis model, the water sample monitoring data is used to perform algae distribution area analysis and algae type analysis to obtain algae distribution area information and algae type information; Based on the algae type information and the algae distribution area information, data simulation analysis is performed using the algae growth trend data to obtain algae growth simulation analysis data; Based on the algae type information and the algae distribution area information combined with the algae growth simulation analysis data, an algae distribution analysis is performed to obtain algae distribution information.

4. The water quality detection method based on stratified sampling according to claim 1 is characterized in that: The construction of the pollution source-water quality correlation model, and the use of the water sample monitoring data combined with rainfall data to perform water quality pollution analysis based on the pollution source-water quality correlation model to obtain water quality pollution analysis data include: Setting an objective function based on a preset regression tree combined with historical water pollution trend information, and setting a penalty term based on the objective function; Based on the greedy algorithm, the gradient boosting tree generated by the XGboost algorithm is divided into nodes to obtain the gradient boosting tree after node division; Constructing an association regression model, and constructing a pollution source-water quality association model based on the association regression model and the gradient boosting tree after node division combined with the objective function and the penalty term; Based on the water sample monitoring data, water quality pollution information is output using the pollution source-water quality association model, rainfall runoff flow information is obtained based on rainfall data, and water quality pollution analysis is performed based on the water quality pollution information and rainfall runoff flow information to obtain water quality pollution analysis data.

5. The water quality detection method based on stratified sampling according to claim 1 is characterized in that: The comprehensive water quality assessment is performed based on the algae distribution information and the water pollution analysis data to obtain a comprehensive water quality assessment result, including: The algae distribution information and water pollution analysis data of each sampling unit area are compared with the preset expected threshold value to obtain a comparison result, the water quality grade of each sampling unit area is divided based on the comparison result, and a comprehensive water quality assessment result is generated based on the water quality grade and the comparison result.

6. A water quality detection device based on stratified sampling, characterized in that: The device comprises: Regional data preprocessing module: used to obtain regional data of the water body area to be tested, preprocess the regional data, and obtain preprocessed regional data; Water quality parameter classification map generation module: used to obtain several parameter distribution characteristic maps of the water body area to be tested, and generate a water quality parameter classification map based on the several parameter distribution characteristic maps; Sampling unit area division module: used for performing spatial autocorrelation analysis based on the water quality parameter classification map and the pre-processed regional data to obtain the local Moran index, and performing cluster area division on the water body area to be tested based on the local Moran index to obtain a number of sampling unit areas; A stratified sampling scheme determination module is used to determine a corresponding stratified sampling scheme in a plurality of sampling unit areas based on the water quality parameter classification diagram and the pre-processed regional data; Algae distribution analysis module: used to obtain water sample monitoring data in each sampling unit area based on the corresponding stratified sampling scheme, and to analyze the algae distribution using the algae index analysis model based on the water sample monitoring data to obtain algae distribution information; Water pollution analysis module: used to construct a pollution source-water quality correlation model, and to perform water pollution analysis based on the pollution source-water quality correlation model using the water sample monitoring data combined with rainfall data to obtain water pollution analysis data; Water quality comprehensive assessment module: used to conduct a comprehensive water quality assessment based on the algae distribution information and water pollution analysis data to obtain a comprehensive water quality assessment result; Wherein, the method of obtaining a plurality of parameter distribution characteristic maps of the water body area to be measured and generating a water quality parameter classification map based on the plurality of parameter distribution characteristic maps comprises: obtaining water body reflectance data, generating a water quality parameter characteristic band based on the water body reflectance data, and generating a plurality of parameter distribution characteristic maps based on the water quality parameter characteristic band; removing non-water quality pixels based on the plurality of parameter distribution characteristic maps to obtain a plurality of parameter distribution characteristic maps after the non-water quality pixels are removed; generating a water quality parameter classification map based on an unsupervised classification method using a plurality of parameter distribution characteristic maps after the non-water quality pixels are removed; The method of determining a corresponding stratified sampling scheme in a plurality of sampling unit areas based on the water quality parameter classification map and the preprocessed regional data comprises: generating a pixel value cumulative frequency distribution map based on the water quality parameter classification map, generating an initial sampling point based on the pixel value cumulative frequency distribution map, and performing point error analysis based on the initial sampling point in combination with a plurality of parameter distribution characteristic maps to obtain a sampling point error map; performing distribution feature aggregation analysis on the water quality parameter classification map and the preprocessed regional data based on a Voronoi diagram to obtain distribution feature aggregation information; performing sample point encryption on the initial sampling point using a particle swarm clustering method based on the distribution feature aggregation information to obtain sample point encryption data; generating a target sampling point for each sampling unit area based on the sampling point error map and the sample point encryption data, obtaining a sampling water body depth based on the water quality parameter classification map, and generating a corresponding stratified sampling scheme based on the target sampling point and the sampling water body depth.

7. An electronic device, comprising a processor and a memory, characterized in that: The memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the electronic device executes the water quality detection method based on stratified sampling as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed on an electronic device, the electronic device executes the water quality detection method based on stratified sampling according to any one of claims 1 to 5.

Images