Method and system for dynamically adjusting underground water monitoring area, and medium

Abstract

The present application discloses an underground water real-time monitoring method and system, and a medium. The method comprises: acquiring historical environment data and historical monitoring data in an underground water monitoring area for spectral clustering data partitioning, and importing related data into a pollution monitoring model for predictive analysis and model training; acquiring pollution monitoring data in monitoring sub-areas in real time, importing the pollution monitoring data into the pollution monitoring model for pollution prediction, and obtaining pollution diffusion directions and pollution source prediction areas; and on the basis of the pollution source prediction areas and environment data of the monitoring sub-areas, calculating a monitoring demand degree of each monitoring sub-area, and generating a monitoring scheme on the basis of the monitoring demand degree. By means of the data spectral clustering method in the present application, accurate grouping and acquisition of training data can be effectively implemented, and model prediction precision is improved. In addition, the present application can realize dynamic adjustment of monitoring areas, effectively conserve human and material resources in underground water monitoring areas, and realize the cost reduction and efficiency improvement of underground water pollution monitoring.

Description

A method, system and medium for dynamically adjusting groundwater monitoring area

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to the Chinese patent application filed with the China Patent Office on August 7, 2023, with application number 202310982007.7 and invention name “A method, system and medium for real-time monitoring of groundwater”, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present application relates to the field of digital twins, and more specifically, to a method, system, and medium for real-time groundwater monitoring. Background Art

[0004] Groundwater is an important water resource in human life and industrial production. However, due to its hidden nature and comfort, it is susceptible to the intrusion and accumulation of various pollutants. Therefore, monitoring and analysis of groundwater pollution is very important.

[0005] Limited by traditional analytical techniques, groundwater sampling and analysis are typically performed at fixed points, resulting in a lack of comprehensive regional pollution analysis and prediction. Furthermore, existing groundwater-related model analysis techniques lack efficient and accurate analytical methods, hindering groundwater pollution analysis and prevention. Therefore, there is an urgent need for an efficient and accurate groundwater monitoring method.

[0006] Summary of the Invention

[0007] The present application overcomes the defects of the prior art and proposes a method, system and medium for real-time monitoring of groundwater.

[0008] The first aspect of the present application provides a method for real-time groundwater monitoring, comprising:

[0009] Acquire a groundwater monitoring area, and divide the groundwater monitoring area into a plurality of monitoring sub-areas;

[0010] Based on the groundwater monitoring area and monitoring sub-areas, a pollution monitoring model based on digital twins is constructed;

[0011] Acquiring historical environmental data and historical monitoring data in the groundwater monitoring area, performing data segmentation based on spectral clustering on the historical environmental data and historical monitoring data, obtaining training data after clustering, and importing the training data into the pollution monitoring model for model training;

[0012] Acquire pollution monitoring data in the monitoring sub-area in real time, import the pollution monitoring data into the pollution monitoring model to perform pollution prediction, and obtain the pollution diffusion direction and pollution source prediction area;

[0013] The monitoring requirement of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring plan is generated based on the monitoring requirement.

[0014] In this solution, the groundwater monitoring area is obtained by dividing the groundwater monitoring area into multiple monitoring sub-areas, specifically:

[0015] Obtain the area size and area map outline of the groundwater monitoring area;

[0016] Obtaining the groundwater monitoring depth, and constructing a three-dimensional groundwater structure model based on the area size, the area map outline, and the groundwater monitoring depth;

[0017] Based on the groundwater structure model, the groundwater monitoring area is divided into multiple monitoring sub-areas.

[0018] In this solution, a pollution monitoring model based on digital twins is constructed based on the groundwater monitoring area and monitoring sub-areas, specifically:

[0019] Obtain the locations of groundwater monitoring points and environmental monitoring points in each monitoring sub-area;

[0020] Obtaining groundwater and environmental monitoring point distribution information based on the groundwater monitoring point locations and the environmental monitoring point locations;

[0021] Based on the distribution information of groundwater and environmental monitoring points, groundwater monitoring areas and monitoring sub-areas, a pollution monitoring model based on digital twins is constructed.

[0022] In this solution, the historical environmental data and historical monitoring data in the groundwater monitoring area are obtained, the historical environmental data and historical monitoring data are divided into data based on spectral clustering, and training data after clustering is obtained. The training data is imported into the pollution monitoring model for model training, specifically:

[0023] Constructing a prediction and analysis module based on a neural network, and using the prediction and analysis module as a data processing module of a pollution monitoring model;

[0024] Acquire historical environmental data and historical monitoring data in each groundwater monitoring area within a preset time period, and divide the historical environmental data and historical monitoring data into N periodic environmental and monitoring data based on a preset time interval;

[0025] Based on the spectral clustering algorithm, a similarity graph with N nodes is set;

[0026] Based on the Gaussian kernel function, the data similarity between N periodic environments and the monitoring data is calculated, and the similarity data is obtained;

[0027] Take N periodic environments and monitoring data as N nodes in the similarity graph, and take the similarity data as the weight value of the edge in the similarity graph;

[0028] Converting the similarity graph into a Laplace matrix, and performing eigendecomposition based on the Laplace matrix to obtain K eigenvalues ​​and corresponding K eigenvectors;

[0029] Based on the preset clustering algorithm, K feature vectors are used as clustering input data for cluster analysis, and L groups of feature data are formed;

[0030] Based on L groups of feature data, data fusion and secondary grouping are performed on N periodic environmental and monitoring data to form corresponding L groups of environmental and monitoring data;

[0031] The L groups of environmental and monitoring data are imported as training data into the pollution monitoring model to perform pollution and environment correlation analysis and data prediction training based on deep learning, and a trained pollution monitoring model is obtained.

[0032] In this solution, the pollution monitoring data in the monitoring sub-area is obtained in real time, and the pollution monitoring data is imported into the pollution monitoring model to perform pollution prediction, and the pollution diffusion direction and pollution source prediction area are obtained, specifically:

[0033] Obtain real-time pollution monitoring data and real-time environmental data for each monitoring sub-area within a period;

[0034] Importing the real-time pollution monitoring data and real-time environmental data into the pollution monitoring model to perform regional pollution fluctuation analysis, obtain analysis results, and calculate and predict pollution situation information and pollution diffusion direction of each monitoring sub-region based on the analysis results;

[0035] The pollution situation information includes the concentration changes of preset pollutants and information on the types of pollutants;

[0036] Based on the pollution diffusion direction of each monitoring sub-area, the pollution source is predicted in combination with the groundwater structure model to obtain the pollution source prediction area.

[0037] In this solution, the monitoring requirement of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring solution is generated based on the monitoring requirement, specifically:

[0038] Randomly select a monitoring sub-area as the selected monitoring sub-area;

[0039] Based on the real-time environmental data of the selected monitoring sub-area, the environmental change index within the corresponding period is calculated based on weather, rainfall, and temperature;

[0040] Calculate the overlapping area value between the selected monitoring sub-area and the pollution source prediction area to obtain the pollution overlapping area;

[0041] Calculate the average shortest distance between the selected monitoring sub-area and the pollution source prediction area in multiple preset plane directions;

[0042] Calculating the monitoring demand in the selected monitoring sub-area based on the environmental change index, the pollution overlap area, and the average shortest distance;

[0043] Analyze all monitoring sub-areas and calculate the monitoring needs of all monitoring sub-areas.

[0044] In this solution, the step of calculating the monitoring requirement of each monitoring sub-area based on the environmental data of the pollution source prediction area and the monitoring sub-area, and generating a monitoring solution based on the monitoring requirement, further includes:

[0045] The monitoring sub-region with a monitoring demand greater than a first preset value is selected as a preferred monitoring sub-region;

[0046] The monitoring sub-region with a monitoring demand less than or equal to a first preset value and greater than a second preset value is selected as the second preferred monitoring sub-region;

[0047] The monitoring sub-area with a monitoring demand less than or equal to a second preset value is regarded as a non-key monitoring sub-area;

[0048] A monitoring plan is generated based on the preferred monitoring sub-area, the second preferred monitoring sub-area, and the non-key monitoring sub-area.

[0049] A second aspect of the present application further provides a groundwater real-time monitoring system, the system comprising: a memory and a processor, wherein the memory comprises a groundwater real-time monitoring program, and when the groundwater real-time monitoring program is executed by the processor, the following steps are implemented:

[0050] Acquire a groundwater monitoring area, and divide the groundwater monitoring area into a plurality of monitoring sub-areas;

[0051] Based on the groundwater monitoring area and monitoring sub-areas, a pollution monitoring model based on digital twins is constructed;

[0052] Acquiring historical environmental data and historical monitoring data in the groundwater monitoring area, performing data segmentation based on spectral clustering on the historical environmental data and historical monitoring data, obtaining training data after clustering, and importing the training data into the pollution monitoring model for model training;

[0053] Acquire pollution monitoring data in the monitoring sub-area in real time, import the pollution monitoring data into the pollution monitoring model to perform pollution prediction, and obtain the pollution diffusion direction and pollution source prediction area;

[0054] The monitoring requirement of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring plan is generated based on the monitoring requirement.

[0055] In this solution, the groundwater monitoring area is obtained by dividing the groundwater monitoring area into multiple monitoring sub-areas, specifically:

[0056] Obtain the area size and area map outline of the groundwater monitoring area;

[0057] Obtaining the groundwater monitoring depth, and constructing a three-dimensional groundwater structure model based on the area size, the area map outline, and the groundwater monitoring depth;

[0058] Based on the groundwater structure model, the groundwater monitoring area is divided into multiple monitoring sub-areas.

[0059] The third aspect of the present application also provides a computer-readable storage medium, which includes a groundwater real-time monitoring program. When the groundwater real-time monitoring program is executed by a processor, the steps of the groundwater real-time monitoring method as described in any one of the above items are implemented.

[0060] The present application discloses a method, system and medium for real-time monitoring of groundwater. By acquiring historical environmental data and historical monitoring data in the groundwater monitoring area, spectral clustering data division is performed, and the relevant data is imported into the pollution monitoring model for predictive analysis and model training; the pollution monitoring data in the monitoring sub-area is acquired in real time and imported into the pollution monitoring model for pollution prediction, and the pollution diffusion direction and pollution source prediction area are obtained; the monitoring demand of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring plan is generated based on the monitoring demand. The data spectral clustering method in this application can effectively realize the accurate grouping and acquisition of training data, and improve the model prediction accuracy. In addition, this application can realize dynamic adjustment of the monitoring area, effectively reduce the consumption of human and material resources in the groundwater monitoring area, and achieve cost reduction and efficiency improvement of groundwater pollution monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] FIG1 shows a flow chart of a method for real-time groundwater monitoring according to the present invention;

[0062] FIG2 shows a flowchart of obtaining monitoring sub-areas in this application;

[0063] FIG3 shows a flow chart of the pollution monitoring model construction of the present application;

[0064] FIG4 shows a block diagram of a groundwater real-time monitoring system of the present application. DETAILED DESCRIPTION

[0065] In order to more clearly understand the above-mentioned objects, features and advantages of the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features therein can be combined with each other in the absence of conflict.

[0066] In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present application is not limited to the specific embodiments disclosed below.

[0067] FIG1 shows a flow chart of a method for real-time groundwater monitoring according to the present application.

[0068] As shown in FIG1 , the first aspect of the present application provides a method for real-time groundwater monitoring, comprising:

[0069] S102, obtaining a groundwater monitoring area, and dividing the groundwater monitoring area into a plurality of monitoring sub-areas;

[0070] S104: constructing a digital twin-based pollution monitoring model based on the groundwater monitoring area and the monitoring sub-area;

[0071] S106, obtaining historical environmental data and historical monitoring data in the groundwater monitoring area, performing data segmentation based on spectral clustering on the historical environmental data and historical monitoring data, obtaining training data after clustering, and importing the training data into the pollution monitoring model for model training;

[0072] S108, obtaining pollution monitoring data in the monitoring sub-area in real time, importing the pollution monitoring data into the pollution monitoring model to perform pollution prediction, and obtaining the pollution diffusion direction and pollution source prediction area;

[0073] S110 , calculating a monitoring requirement of each monitoring sub-area based on environmental data of the pollution source prediction area and the monitoring sub-area, and generating a monitoring plan based on the monitoring requirement.

[0074] FIG2 shows a flow chart of obtaining monitoring sub-areas in this application.

[0075] According to an embodiment of the present application, the groundwater monitoring area is obtained by dividing the groundwater monitoring area into multiple monitoring sub-areas, specifically:

[0076] S202, obtaining the area size and area map outline of the groundwater monitoring area;

[0077] S204, obtaining the groundwater monitoring depth, and constructing a three-dimensional groundwater structure model based on the area size, the area map outline, and the groundwater monitoring depth;

[0078] S206: Based on the groundwater structure model, the groundwater monitoring area is divided into multiple monitoring sub-areas.

[0079] It should be noted that the groundwater monitoring depth is the maximum depth of groundwater monitoring. The groundwater structure model is a visualization-based three-dimensional map model that can visualize groundwater monitoring data and simulate the flow of groundwater at corresponding locations within the area. This allows users to intuitively understand groundwater pollution and enhance the user experience of the information-based system. The area is divided into multiple monitoring sub-areas, and the horizontal area difference of each monitoring sub-area must be within a preset range.

[0080] FIG3 shows a flow chart of the pollution monitoring model construction of the present application.

[0081] According to an embodiment of the present application, a pollution monitoring model based on a digital twin is constructed based on the groundwater monitoring area and the monitoring sub-area, specifically:

[0082] S302, obtaining the location of groundwater monitoring points and environmental monitoring points in each monitoring sub-area;

[0083] S304, obtaining groundwater and environmental monitoring point distribution information based on the groundwater monitoring point locations and the environmental monitoring point locations;

[0084] S306: Construct a digital twin-based pollution monitoring model based on the distribution information of groundwater and environmental monitoring points, groundwater monitoring areas, and monitoring sub-areas.

[0085] It should be noted that the pollution monitoring model is specifically a groundwater simulation model used for data analysis and prediction. The model contains information such as the geographical correlation between each monitoring sub-area, the location relationship of the monitoring points, the water flow direction, etc., which is used for data simulation analysis of groundwater.

[0086] The pollution monitoring model is primarily used for data analysis, calculation, and prediction, while the groundwater structure model is primarily used for intuitive and reasonable data display. The organic integration of these two models can improve the efficiency of system analysis and display, and enhance the robustness and robustness of the system. The pollution monitoring model and the groundwater structure model enable real-time data linkage and low-latency transmission.

[0087] According to an embodiment of the present application, the historical environmental data and historical monitoring data in the groundwater monitoring area are obtained, the historical environmental data and historical monitoring data are divided into data based on spectral clustering, and training data after clustering is obtained, and the training data is imported into the pollution monitoring model for model training, specifically:

[0088] Constructing a prediction and analysis module based on a neural network, and using the prediction and analysis module as a data processing module of a pollution monitoring model;

[0089] Acquire historical environmental data and historical monitoring data in each groundwater monitoring area within a preset time period, and divide the historical environmental data and historical monitoring data into N periodic environmental and monitoring data based on a preset time interval;

[0090] Based on the spectral clustering algorithm, a similarity graph with N nodes is set;

[0091] Based on the Gaussian kernel function, the data similarity between N periodic environments and the monitoring data is calculated, and the similarity data is obtained;

[0092] Take N periodic environments and monitoring data as N nodes in the similarity graph, and take the similarity data as the weight value of the edge in the similarity graph;

[0093] Converting the similarity graph into a Laplace matrix, and performing eigendecomposition based on the Laplace matrix to obtain K eigenvalues ​​and corresponding K eigenvectors;

[0094] Based on the preset clustering algorithm, K feature vectors are used as clustering input data for cluster analysis, and L groups of feature data are formed;

[0095] Based on L groups of feature data, data fusion and secondary grouping are performed on N periodic environmental and monitoring data to form corresponding L groups of environmental and monitoring data;

[0096] The L groups of environmental and monitoring data are imported as training data into the pollution monitoring model to perform pollution and environment correlation analysis and data prediction training based on deep learning, and a trained pollution monitoring model is obtained.

[0097] It should be noted that the historical monitoring data include the pH value, dissolved oxygen concentration, electrical conductivity, ammonia nitrogen content, heavy metal content, etc. of groundwater in the groundwater monitoring point, and the historical environmental data include the weather conditions, temperature, river water level, precipitation, etc. at the monitoring point. The preset time interval is set by the user. When the preset time period is fixed, the larger the preset time interval, the smaller N, and N is the number of periods after the data is divided. In the data fusion and secondary grouping of N periodic environments and monitoring data based on L groups of feature data, the L groups of feature data are cluster groups obtained based on spectral clustering, which lack a certain amount of data compared to the original data. This application is based on L groups of feature data, and the N periodic environments and monitoring data are secondary grouped. The grouping basis is L groups of feature data. The corresponding feature information in the obtained L groups of environments and monitoring data is L groups of feature data. The prediction and analysis module based on neural networks is specifically a model with training, analysis, and prediction functions.

[0098] It is worth mentioning that in this application, by dividing historical data into periodic data, and clustering and grouping N periodic environmental and monitoring data through the spectral clustering method, L groups of environmental and monitoring data are obtained and used as model training data, which can effectively realize the precise grouping and acquisition of training data. After grouping based on spectral clustering, the redundancy of training data is effectively reduced, which greatly increases the model training effect and improves the model prediction accuracy.

[0099] According to an embodiment of the present application, the pollution monitoring data in the monitoring sub-area is obtained in real time, and the pollution monitoring data is imported into the pollution monitoring model for pollution prediction, and the pollution diffusion direction and pollution source prediction area are obtained, specifically:

[0100] Obtain real-time pollution monitoring data and real-time environmental data for each monitoring sub-area within a period;

[0101] Importing the real-time pollution monitoring data and real-time environmental data into the pollution monitoring model to perform regional pollution fluctuation analysis, obtain analysis results, and calculate and predict pollution situation information and pollution diffusion direction of each monitoring sub-region based on the analysis results;

[0102] The pollution situation information includes the concentration changes of preset pollutants and information on the types of pollutants;

[0103] Based on the pollution diffusion direction of each monitoring sub-area, the pollution source is predicted in combination with the groundwater structure model to obtain the pollution source prediction area.

[0104] It should be noted that the pollution source prediction area generally includes multiple areas, which depend on the actual groundwater pollution situation.

[0105] According to an embodiment of the present application, the monitoring requirement of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring plan is generated based on the monitoring requirement, specifically:

[0106] Randomly select a monitoring sub-area as the selected monitoring sub-area;

[0107] Based on the real-time environmental data of the selected monitoring sub-area, the environmental change index within the corresponding period is calculated based on weather, rainfall, and temperature;

[0108] Calculate the overlapping area value between the selected monitoring sub-area and the pollution source prediction area to obtain the pollution overlapping area;

[0109] Calculate the average shortest distance between the selected monitoring sub-area and the pollution source prediction area in multiple preset plane directions;

[0110] Calculating the monitoring demand in the selected monitoring sub-area based on the environmental change index, the pollution overlap area, and the average shortest distance;

[0111] Analyze all monitoring sub-areas and calculate the monitoring needs of all monitoring sub-areas.

[0112] It should be noted that the environmental change index reflects the severity of changes in weather, rainfall, and temperature. The larger the index, the faster the pollution spreads in the corresponding area.

[0113] The preset plane directions are generally 4 or 8 directions, corresponding to the four directions of east, south, west and north in the map. More directions, such as 8 directions, can be set according to the analysis accuracy requirements. The average shortest distance is the average of the shortest distances between the selected monitoring sub-area and the pollution source prediction area in all directions. The greater the monitoring demand, the greater the monitoring demand of the corresponding monitoring sub-area. When the monitoring demand is lower than a preset minimum value, the pollution impact of the area is low and the data contribution to the entire groundwater pollution prediction and analysis process is low.

[0114] The calculation formula for the monitoring demand is:

[0115] Where, T X To monitor demand, S X is the average shortest distance, E X is the environmental change index, L S It is the average shortest distance between the monitoring sub-area and the pollution source prediction area.

[0116] According to an embodiment of the present application, the calculating of the monitoring requirement of each monitoring sub-area based on the environmental data of the pollution source prediction area and the monitoring sub-area, and generating a monitoring plan based on the monitoring requirement, further includes:

[0117] The monitoring sub-region with a monitoring demand greater than a first preset value is selected as a preferred monitoring sub-region;

[0118] The monitoring sub-region with a monitoring demand less than or equal to a first preset value and greater than a second preset value is selected as the second preferred monitoring sub-region;

[0119] The monitoring sub-area with a monitoring demand less than or equal to a second preset value is regarded as a non-key monitoring sub-area;

[0120] A monitoring plan is generated based on the preferred monitoring sub-area, the second preferred monitoring sub-area, and the non-key monitoring sub-area.

[0121] It should be noted that, in the monitoring scheme, the preferred monitoring sub-area will be monitored and analyzed in each cycle, and the second preferred monitoring sub-area will be monitored at intervals, that is, monitoring will be implemented in the current cycle and not in the next cycle, while non-key monitoring sub-areas will not be monitored. In addition, in each analysis cycle, the monitoring demand changes, and there are certain differences in the monitoring schemes corresponding to each analysis cycle. Through the method of the present application, it is possible to dynamically adjust the monitoring area, and further dynamically adjust the monitoring scheme, thereby effectively reducing unnecessary data collection and reducing the amount of data analysis, while ensuring the needs of forecasting and analysis, greatly improving the efficiency of system analysis. Furthermore, it is possible to effectively reduce the consumption of human and material resources in the groundwater monitoring area, and achieve cost reduction and efficiency improvement in groundwater pollution monitoring.

[0122] According to an embodiment of the present application, the present invention further includes:

[0123] Obtain the pollution source prediction area in the groundwater monitoring area within the current cycle;

[0124] Obtain industrial distribution information and farmland distribution information corresponding to the groundwater monitoring area;

[0125] If the number of pollution source prediction areas is greater than the preset number, at least one pollution source monitoring point is set in each pollution source prediction area;

[0126] Based on the industrial distribution information and farmland distribution information, analyze the industrial area distribution and agricultural area distribution in each pollution source prediction area, and further obtain the industrial activity information and agricultural activity information in each pollution source prediction area in the current cycle;

[0127] Obtain groundwater monitoring data from all pollution source monitoring points, import the groundwater monitoring data into the pollution monitoring model to perform pollution degree analysis and pollution change analysis, and obtain pollutant change information and pollution diffusion direction information for each pollution source prediction area;

[0128] Industrial and agricultural activity information is used as the first analysis data, and pollutant change information and pollution diffusion direction information is used as the second analysis data;

[0129] Calculating data linear variation characteristics of the first analysis data and the second analysis data, and calculating a linear correlation coefficient between the first analysis data and the second analysis data based on the linear variation characteristics;

[0130] Determine whether the linear correlation coefficient is greater than a preset correlation value. If so, mark the corresponding pollution source prediction area as a pollution warning area, and analyze and determine all pollution warning areas;

[0131] Generate corresponding pollution source investigation and prevention plans based on all pollution warning areas.

[0132] It should be noted that the preset number is generally 2 to 10. When the number of pollution source prediction areas reaches a certain number, it means that the groundwater pollution situation in the area is relatively complicated, there are many pollution sources, and pollution investigation is urgently needed. The industrial activity information specifically refers to the frequency and generation scale of industrial production activities in the current cycle, and the agricultural activity information specifically refers to the frequency of agricultural activities and information on activities such as fertilization and irrigation in agricultural production. Through the method of this application, it is possible to accurately analyze the pollution sources and generate an investigation and prevention plan for the current groundwater monitoring area, effectively improving the effect of groundwater pollution prevention and control.

[0133] FIG4 shows a block diagram of a groundwater real-time monitoring system of the present application.

[0134] A second aspect of the present application further provides a groundwater real-time monitoring system 4, comprising: a memory 41 and a processor 42, wherein the memory includes a groundwater real-time monitoring program, and when the groundwater real-time monitoring program is executed by the processor, the following steps are implemented:

[0135] Acquire a groundwater monitoring area, and divide the groundwater monitoring area into a plurality of monitoring sub-areas;

[0136] Based on the groundwater monitoring area and monitoring sub-areas, a pollution monitoring model based on digital twins is constructed;

[0137] Acquiring historical environmental data and historical monitoring data in the groundwater monitoring area, performing data segmentation based on spectral clustering on the historical environmental data and historical monitoring data, obtaining training data after clustering, and importing the training data into the pollution monitoring model for model training;

[0138] Acquire pollution monitoring data in the monitoring sub-area in real time, import the pollution monitoring data into the pollution monitoring model to perform pollution prediction, and obtain the pollution diffusion direction and pollution source prediction area;

[0139] The monitoring requirement of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring plan is generated based on the monitoring requirement.

[0140] According to an embodiment of the present application, the groundwater monitoring area is obtained by dividing the groundwater monitoring area into multiple monitoring sub-areas, specifically:

[0141] Obtain the area size and area map outline of the groundwater monitoring area;

[0142] Obtaining the groundwater monitoring depth, and constructing a three-dimensional groundwater structure model based on the area size, the area map outline, and the groundwater monitoring depth;

[0143] Based on the groundwater structure model, the groundwater monitoring area is divided into multiple monitoring sub-areas.

[0144] It should be noted that the groundwater monitoring depth is the maximum depth of groundwater monitoring. The groundwater structure model is a visualization-based three-dimensional map model that can visualize groundwater monitoring data and simulate the flow of groundwater at corresponding locations within the area. This allows users to intuitively understand groundwater pollution and enhance the user experience of the information-based system. The area is divided into multiple monitoring sub-areas, and the horizontal area difference of each monitoring sub-area must be within a preset range.

[0145] According to an embodiment of the present application, a pollution monitoring model based on a digital twin is constructed based on the groundwater monitoring area and the monitoring sub-area, specifically:

[0146] Obtain the locations of groundwater monitoring points and environmental monitoring points in each monitoring sub-area;

[0147] Obtaining groundwater and environmental monitoring point distribution information based on the groundwater monitoring point locations and the environmental monitoring point locations;

[0148] Based on the distribution information of groundwater and environmental monitoring points, groundwater monitoring areas and monitoring sub-areas, a pollution monitoring model based on digital twins is constructed.

[0149] It should be noted that the pollution monitoring model is specifically a groundwater simulation model used for data analysis and prediction. The model contains information such as the geographical correlation between each monitoring sub-area, the location relationship of the monitoring points, the water flow direction, etc., which is used for data simulation analysis of groundwater.

[0150] The pollution monitoring model is primarily used for data analysis, calculation, and prediction, while the groundwater structure model is primarily used for intuitive and reasonable data display. The organic integration of these two models can improve the efficiency of system analysis and display, and enhance the robustness and robustness of the system. The pollution monitoring model and the groundwater structure model enable real-time data linkage and low-latency transmission.

[0151] According to an embodiment of the present application, the historical environmental data and historical monitoring data in the groundwater monitoring area are obtained, the historical environmental data and historical monitoring data are divided into data based on spectral clustering, and training data after clustering is obtained, and the training data is imported into the pollution monitoring model for model training, specifically:

[0152] Constructing a prediction and analysis module based on a neural network, and using the prediction and analysis module as a data processing module of a pollution monitoring model;

[0153] Acquire historical environmental data and historical monitoring data in each groundwater monitoring area within a preset time period, and divide the historical environmental data and historical monitoring data into N periodic environmental and monitoring data based on a preset time interval;

[0154] Based on the spectral clustering algorithm, a similarity graph with N nodes is set;

[0155] Based on the Gaussian kernel function, the data similarity between N periodic environments and the monitoring data is calculated, and the similarity data is obtained;

[0156] Take N periodic environments and monitoring data as N nodes in the similarity graph, and take the similarity data as the weight value of the edge in the similarity graph;

[0157] Converting the similarity graph into a Laplace matrix, and performing eigendecomposition based on the Laplace matrix to obtain K eigenvalues ​​and corresponding K eigenvectors;

[0158] Based on the preset clustering algorithm, K feature vectors are used as clustering input data for cluster analysis, and L groups of feature data are formed;

[0159] Based on L groups of feature data, data fusion and secondary grouping are performed on N periodic environmental and monitoring data to form corresponding L groups of environmental and monitoring data;

[0160] The L groups of environmental and monitoring data are imported as training data into the pollution monitoring model to perform pollution and environment correlation analysis and data prediction training based on deep learning, and a trained pollution monitoring model is obtained.

[0161] It should be noted that the historical monitoring data include the pH value, dissolved oxygen concentration, electrical conductivity, ammonia nitrogen content, heavy metal content, etc. of groundwater in the groundwater monitoring point, and the historical environmental data include the weather conditions, temperature, river water level, precipitation, etc. at the monitoring point. The preset time interval is set by the user. When the preset time period is fixed, the larger the preset time interval, the smaller N, and N is the number of periods after the data is divided. In the data fusion and secondary grouping of N periodic environments and monitoring data based on L groups of feature data, the L groups of feature data are cluster groups obtained based on spectral clustering, which lack a certain amount of data compared to the original data. This application is based on L groups of feature data, and the N periodic environments and monitoring data are secondary grouped. The grouping basis is L groups of feature data. The corresponding feature information in the obtained L groups of environments and monitoring data is L groups of feature data. The prediction and analysis module based on neural networks is specifically a model with training, analysis, and prediction functions.

[0162] It is worth mentioning that in this application, by dividing historical data into periodic data, and clustering and grouping N periodic environmental and monitoring data through the spectral clustering method, L groups of environmental and monitoring data are obtained and used as model training data, which can effectively realize the precise grouping and acquisition of training data. After grouping based on spectral clustering, the redundancy of training data is effectively reduced, which greatly increases the model training effect and improves the model prediction accuracy.

[0163] According to an embodiment of the present application, the pollution monitoring data in the monitoring sub-area is obtained in real time, and the pollution monitoring data is imported into the pollution monitoring model for pollution prediction, and the pollution diffusion direction and pollution source prediction area are obtained, specifically:

[0164] Obtain real-time pollution monitoring data and real-time environmental data for each monitoring sub-area within a period;

[0165] Importing the real-time pollution monitoring data and real-time environmental data into the pollution monitoring model to perform regional pollution fluctuation analysis, obtain analysis results, and calculate and predict pollution situation information and pollution diffusion direction of each monitoring sub-region based on the analysis results;

[0166] The pollution situation information includes the concentration changes of preset pollutants and information on the types of pollutants;

[0167] Based on the pollution diffusion direction of each monitoring sub-area, the pollution source is predicted in combination with the groundwater structure model to obtain the pollution source prediction area.

[0168] It should be noted that the pollution source prediction area generally includes multiple areas, which depend on the actual groundwater pollution situation.

[0169] According to an embodiment of the present application, the monitoring requirement of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring plan is generated based on the monitoring requirement, specifically:

[0170] Randomly select a monitoring sub-area as the selected monitoring sub-area;

[0171] Based on the real-time environmental data of the selected monitoring sub-area, the environmental change index within the corresponding period is calculated based on weather, rainfall, and temperature;

[0172] Calculate the overlapping area value between the selected monitoring sub-area and the pollution source prediction area to obtain the pollution overlapping area;

[0173] Calculate the average shortest distance between the selected monitoring sub-area and the pollution source prediction area in multiple preset plane directions;

[0174] Calculating the monitoring demand in the selected monitoring sub-area based on the environmental change index, the pollution overlap area, and the average shortest distance;

[0175] Analyze all monitoring sub-areas and calculate the monitoring needs of all monitoring sub-areas.

[0176] It should be noted that the environmental change index reflects the severity of changes in weather, rainfall, and temperature. The larger the index, the faster the pollution spreads in the corresponding area.

[0177] The preset plane directions are generally 4 or 8 directions, corresponding to the four directions of east, south, west and north in the map. More directions, such as 8 directions, can be set according to the analysis accuracy requirements. The average shortest distance is the average of the shortest distances between the selected monitoring sub-area and the pollution source prediction area in all directions. The greater the monitoring demand, the greater the monitoring demand of the corresponding monitoring sub-area. When the monitoring demand is lower than a preset minimum value, the pollution impact of the area is low and the data contribution to the entire groundwater pollution prediction and analysis process is low.

[0178] The calculation formula for the monitoring demand is:

[0179] Where, T X To monitor demand, S X is the average shortest distance, E X is the environmental change index, L S It is the average shortest distance between the monitoring sub-area and the pollution source prediction area.

[0180] According to an embodiment of the present application, the calculating of the monitoring requirement of each monitoring sub-area based on the environmental data of the pollution source prediction area and the monitoring sub-area, and generating a monitoring plan based on the monitoring requirement, further includes:

[0181] The monitoring sub-region with a monitoring demand greater than a first preset value is selected as a preferred monitoring sub-region;

[0182] The monitoring sub-region with a monitoring demand less than or equal to a first preset value and greater than a second preset value is selected as the second preferred monitoring sub-region;

[0183] The monitoring sub-area with a monitoring demand less than or equal to a second preset value is regarded as a non-key monitoring sub-area;

[0184] A monitoring plan is generated based on the preferred monitoring sub-area, the second preferred monitoring sub-area, and the non-key monitoring sub-area.

[0185] It should be noted that, in the monitoring scheme, the preferred monitoring sub-area will be monitored and analyzed in each cycle, and the second preferred monitoring sub-area will be monitored at intervals, that is, monitoring will be implemented in the current cycle and not in the next cycle, while non-key monitoring sub-areas will not be monitored. In addition, in each analysis cycle, the monitoring demand changes, and there are certain differences in the monitoring schemes corresponding to each analysis cycle. Through the method of the present application, it is possible to dynamically adjust the monitoring area, and further dynamically adjust the monitoring scheme, thereby effectively reducing unnecessary data collection and reducing the amount of data analysis, while ensuring the needs of forecasting and analysis, greatly improving the efficiency of system analysis. Furthermore, it is possible to effectively reduce the consumption of human and material resources in the groundwater monitoring area, and achieve cost reduction and efficiency improvement in groundwater pollution monitoring.

[0186] According to an embodiment of the present application, the present invention further includes:

[0187] Obtain the pollution source prediction area in the groundwater monitoring area within the current cycle;

[0188] Obtain industrial distribution information and farmland distribution information corresponding to the groundwater monitoring area;

[0189] If the number of pollution source prediction areas is greater than the preset number, at least one pollution source monitoring point is set in each pollution source prediction area;

[0190] Based on the industrial distribution information and farmland distribution information, analyze the industrial area distribution and agricultural area distribution in each pollution source prediction area, and further obtain the industrial activity information and agricultural activity information in each pollution source prediction area in the current cycle;

[0191] Obtain groundwater monitoring data from all pollution source monitoring points, import the groundwater monitoring data into the pollution monitoring model to perform pollution degree analysis and pollution change analysis, and obtain pollutant change information and pollution diffusion direction information for each pollution source prediction area;

[0192] Industrial and agricultural activity information is used as the first analysis data, and pollutant change information and pollution diffusion direction information is used as the second analysis data;

[0193] Calculating data linear variation characteristics of the first analysis data and the second analysis data, and calculating a linear correlation coefficient between the first analysis data and the second analysis data based on the linear variation characteristics;

[0194] Determine whether the linear correlation coefficient is greater than a preset correlation value. If so, mark the corresponding pollution source prediction area as a pollution warning area, and analyze and determine all pollution warning areas;

[0195] Generate corresponding pollution source investigation and prevention plans based on all pollution warning areas.

[0196] It should be noted that the preset number is generally 2 to 10. When the number of pollution source prediction areas reaches a certain number, it means that the groundwater pollution situation in the area is relatively complicated, there are many pollution sources, and pollution investigation is urgently needed. The industrial activity information specifically refers to the frequency and generation scale of industrial production activities in the current cycle, and the agricultural activity information specifically refers to the frequency of agricultural activities and information on activities such as fertilization and irrigation in agricultural production. Through the method of this application, it is possible to accurately analyze the pollution sources and generate an investigation and prevention plan for the current groundwater monitoring area, effectively improving the effect of groundwater pollution prevention and control.

[0197] The third aspect of the present application also provides a computer-readable storage medium, which includes a groundwater real-time monitoring program. When the groundwater real-time monitoring program is executed by a processor, the steps of the groundwater real-time monitoring method as described in any one of the above items are implemented.

[0198] The present application discloses a method, system and medium for real-time monitoring of groundwater. By acquiring historical environmental data and historical monitoring data in the groundwater monitoring area, spectral clustering data division is performed, and the relevant data is imported into the pollution monitoring model for predictive analysis and model training; the pollution monitoring data in the monitoring sub-area is acquired in real time and imported into the pollution monitoring model for pollution prediction, and the pollution diffusion direction and pollution source prediction area are obtained; the monitoring demand of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring plan is generated based on the monitoring demand. The data spectral clustering method in this application can effectively realize the accurate grouping and acquisition of training data, and improve the model prediction accuracy. In addition, this application can realize dynamic adjustment of the monitoring area, effectively reduce the consumption of human and material resources in the groundwater monitoring area, and achieve cost reduction and efficiency improvement of groundwater pollution monitoring.

[0199] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0200] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0201] In addition, all functional units in the embodiments of the present application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the above-mentioned integrated units can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0202] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0203] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0204] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for real-time monitoring of groundwater, characterized in that: include: Acquire a groundwater monitoring area, and divide the groundwater monitoring area into multiple monitoring sub-areas; Based on the groundwater monitoring area and the monitoring sub-area, a pollution monitoring model based on digital twins is constructed; Acquire historical environmental data and historical monitoring data in the groundwater monitoring area, perform data division based on spectral clustering on the historical environmental data and historical monitoring data, obtain training data after cluster division, and import the training data into the pollution monitoring model for model training; Acquire pollution monitoring data in the monitoring sub-area in real time, import the pollution monitoring data into the pollution monitoring model for pollution prediction, and obtain the pollution diffusion direction and pollution source prediction area; The monitoring requirement of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring plan is generated based on the monitoring requirement.

2. A method for real-time groundwater monitoring according to claim 1, characterized in that: The obtaining of the groundwater monitoring area involves dividing the groundwater monitoring area into multiple monitoring sub-areas, specifically: Obtain the area size and area map outline of the groundwater monitoring area; Obtaining the groundwater monitoring depth, and constructing a three-dimensional groundwater structure model based on the area size, the area map outline, and the groundwater monitoring depth; Based on the groundwater structure model, the groundwater monitoring area is divided into multiple monitoring sub-areas.

3. A method for real-time monitoring of groundwater according to claim 1, characterized in that: The construction of a pollution monitoring model based on digital twins based on the groundwater monitoring area and the monitoring sub-area is specifically as follows: Obtain the location of groundwater monitoring points and environmental monitoring points in each monitoring sub-area; Based on the locations of the groundwater monitoring points and the environmental monitoring points, the distribution information of the groundwater and environmental monitoring points is obtained; Based on the distribution information of groundwater and environmental monitoring points, groundwater monitoring areas and monitoring sub-areas, a pollution monitoring model based on digital twins is constructed.

4. A method for real-time monitoring of groundwater according to claim 1, characterized in that: The method of obtaining historical environmental data and historical monitoring data in the groundwater monitoring area, performing data division based on spectral clustering on the historical environmental data and historical monitoring data, and obtaining training data after cluster division, and importing the training data into the pollution monitoring model for model training, is specifically as follows: Constructing a prediction and analysis module based on a neural network, and using the prediction and analysis module as a data processing module of a pollution monitoring model; Acquire historical environmental data and historical monitoring data in each groundwater monitoring area within a preset time period, and divide the historical environmental data and historical monitoring data into N periodic environmental and monitoring data based on a preset time interval; Based on the spectral clustering algorithm, a similarity graph with N nodes is set; Based on the Gaussian kernel function, the data similarity between N periodic environments and the monitoring data is calculated, and the similarity data is obtained; Take N periodic environments and monitoring data as N nodes in the similarity graph, and take the similarity data as the weight value of the edge in the similarity graph; The similarity graph is converted into a Laplace matrix, and eigendecomposition is performed based on the Laplace matrix to obtain K eigenvalues ​​and corresponding K eigenvectors; Based on the preset clustering algorithm, K feature vectors are used as clustering input data for clustering analysis, and L groups of feature data are formed; Based on L groups of characteristic data, data fusion and secondary grouping are performed on N periodic environmental and monitoring data, and corresponding L groups of environmental and monitoring data are formed; The L groups of environmental and monitoring data are imported as training data into the pollution monitoring model to perform pollution and environmental correlation analysis and data prediction training based on deep learning, and a trained pollution monitoring model is obtained.

5. A method for real-time monitoring of groundwater according to claim 4, characterized in that: The real-time acquisition of pollution monitoring data in the monitoring sub-area, importing the pollution monitoring data into the pollution monitoring model for pollution prediction, and obtaining the pollution diffusion direction and pollution source prediction area are specifically: Obtain real-time pollution monitoring data and real-time environmental data for each monitoring sub-area within a period; Importing the real-time pollution monitoring data and real-time environmental data into the pollution monitoring model to perform regional pollution fluctuation analysis, obtain analysis results, and calculate and predict the pollution situation information and pollution diffusion direction of each monitoring sub-area based on the analysis results; The pollution situation information includes the concentration change of preset pollutants and the type of pollutants; Based on the pollution diffusion direction of each monitoring sub-area, the pollution source is predicted in combination with the groundwater structure model to obtain the pollution source prediction area.

6. A method for real-time monitoring of groundwater according to claim 5, characterized in that: The monitoring requirement of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring plan is generated based on the monitoring requirement, specifically: A monitoring sub-area is randomly selected as the selected monitoring sub-area; According to the real-time environmental data of the selected monitoring sub-area, the environmental change index within the corresponding period is calculated based on weather, rainfall and temperature; Calculate the overlap area between the selected monitoring sub-area and the pollution source prediction area to obtain the pollution overlap area; Calculate the average shortest distance between the selected monitoring sub-area and the pollution source prediction area in multiple preset plane directions; Calculate the monitoring demand in the selected monitoring sub-area based on the environmental change index, the pollution overlap area, and the average shortest distance; Analyze all monitoring sub-areas and calculate the monitoring needs of all monitoring sub-areas.

7. A method for real-time monitoring of groundwater according to claim 1, characterized in that: The step of calculating the monitoring requirement of each monitoring sub-area based on the environmental data of the pollution source prediction area and the monitoring sub-area, and generating a monitoring plan based on the monitoring requirement, further includes: Taking the monitoring sub-region where the monitoring demand is greater than the first preset value as the preferred monitoring sub-region; The monitoring sub-region with a monitoring requirement less than or equal to a first preset value and greater than a second preset value is taken as a second preferred monitoring sub-region; The monitoring sub-area with a monitoring demand less than or equal to a second preset value is regarded as a non-key monitoring sub-area; A monitoring plan is generated based on the preferred monitoring sub-area, the second preferred monitoring sub-area, and the non-key monitoring sub-area.

8. A groundwater real-time monitoring system, characterized in that: The system includes: a memory and a processor. The memory includes a groundwater real-time monitoring program. When the groundwater real-time monitoring program is executed by the processor, the following steps are implemented: Acquire a groundwater monitoring area, and divide the groundwater monitoring area into multiple monitoring sub-areas; Based on the groundwater monitoring area and the monitoring sub-area, a pollution monitoring model based on digital twins is constructed; Acquire historical environmental data and historical monitoring data in the groundwater monitoring area, perform data division based on spectral clustering on the historical environmental data and historical monitoring data, obtain training data after cluster division, and import the training data into the pollution monitoring model for model training; Acquire pollution monitoring data in the monitoring sub-area in real time, import the pollution monitoring data into the pollution monitoring model for pollution prediction, and obtain the pollution diffusion direction and pollution source prediction area; The monitoring requirement of each monitoring sub-area is calculated based on the environmental data of the pollution source prediction area and the monitoring sub-area, and a monitoring plan is generated based on the monitoring requirement.

9. A groundwater real-time monitoring system according to claim 8, characterized in that: The obtaining of the groundwater monitoring area involves dividing the groundwater monitoring area into multiple monitoring sub-areas, specifically: Obtain the area size and area map outline of the groundwater monitoring area; Obtaining the groundwater monitoring depth, and constructing a three-dimensional groundwater structure model based on the area size, the area map outline, and the groundwater monitoring depth; Based on the groundwater structure model, the groundwater monitoring area is divided into multiple monitoring sub-areas.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a groundwater real-time monitoring program, and when the groundwater real-time monitoring program is executed by a processor, the steps of the groundwater real-time monitoring method according to any one of claims 1 to 7 are implemented.

Images