Graph-driven water pollution source tracing method based on hydrodynamics and water quality models

Abstract

The present invention relates to the technical field of water pollution source tracing, and in particular to a graph theory driven water pollution source tracing method based on hydrodynamics and water quality models, including monitoring each monitoring point; calculating correlation coefficients; constructing a graph model to determine pollution source monitoring points; simulation calculation; calculating and verifying correlation coefficients, and adjusting or issuing key pollution source prompts according to the calculation results. The present invention evaluates the accuracy of the model by comparing the Pearson correlation coefficient of the simulation results with the actual monitoring data, and adjusts the number of monitoring points or the standard value of the correlation coefficient according to the evaluation results to optimize the accuracy and efficiency of the source tracing method, and finally issues an early warning prompt of the pollution source, effectively combining the hydrodynamics and water quality models and graph theory analysis methods, improving the accuracy and response speed of water pollution source tracing, and effectively solving the problems of complex and inefficient source tracing analysis due to incomplete and uneven distribution of water quality monitoring data, and increased difficulty in source tracing caused by the combined effects of multiple pollution sources.

Description

Technical Field

[0001] The present invention relates to the technical field of water pollution source tracing, and in particular to a graph theory driven water pollution source tracing method based on hydrodynamics and water quality models. Background Art

[0002] Currently, there are many sources of water pollution, and the workload of pollution tracing is large. Without an effective pollution tracing model simulation, it is difficult to carry out pollution source investigation and tracing work in a timely and effective manner, which will affect the management and governance of rivers. Water pollution data usually come from different monitoring sites, and these data may be uneven in time and space distribution. In addition, due to the limitations of monitoring technology and frequency, it may not be possible to obtain comprehensive information on all pollutants, resulting in incomplete data and increasing the difficulty of tracing the source. In many cases, water pollution is not a single source, but a combination of multiple pollution sources, including point source pollution, non-point source pollution, atmospheric deposition, etc. Pollutants discharged from different pollution sources may undergo physical, chemical or biological reactions in the water body, resulting in mixed effects, making the tracing analysis more complicated.

[0003] In actual work, when the water environment supervision system platform sounds an alarm, it means that a certain pollutant at the monitoring point or the early warning point has exceeded the standard or is greater than the threshold, but the system cannot determine why the indicator exceeds the standard, or whether it is caused by the discharge of pollutants from a sewage outlet or a pollution source. In a river and lake system with complex water systems, the water systems are intricately connected and the sewage outlets may crisscross. Once a sudden pollution incident occurs, it is impossible to narrow down or lock the source of pollution in the first time. Summary of the invention

[0004] To this end, the present invention provides a graph theory driven water pollution source tracing method based on hydrodynamics and water quality models, so as to overcome the problems in the prior art of complex and inefficient source tracing analysis due to incomplete and uneven distribution of water quality monitoring data, as well as increased difficulty in source tracing caused by the combined effects of multiple pollution sources.

[0005] To achieve the above object, the present invention provides a graph-driven water pollution source tracing method based on hydrodynamics and water quality models, comprising:

[0006] Divide the monitoring points in the river into preset numbers, and monitor the real-time pH value, real-time ammonia nitrogen and real-time dissolved oxygen of each monitoring point;

[0007] Calculate a number of correlation coefficients corresponding to the monitoring results of each monitoring point;

[0008] A graph model is constructed based on the monitoring results, correlation coefficients and preset standard coefficients of each monitoring point, the number of edges formed by connecting each node in the graph model is obtained, and the pollution source monitoring point is determined based on the number of edges;

[0009] Obtain river terrain elevation data of pollution source monitoring points, generate input data digitally, and set hydrodynamic parameters of hydrodynamic and water quality models based on the input data;

[0010] The time series of ammonia nitrogen at the pollution source monitoring point is loaded into the hydrodynamic and water quality model for simulation calculation to obtain the simulation results within the preset monitoring time;

[0011] The Pearson correlation coefficient is used to calculate the verification correlation coefficient between the simulation results and the time series of ammonia nitrogen at the pollution source monitoring point. According to the verification correlation coefficient and the preset standard verification correlation coefficient, it is determined to adjust the preset quantity, or adjust the preset standard coefficient, or issue a key pollution source prompt.

[0012] Furthermore, the calculation of the correlation coefficient in the monitoring results of each monitoring point includes:

[0013] The monitoring results of each monitoring point were standardized and rescaled, and the pH-ammonia-nitrogen correlation coefficient between the real-time pH value and the real-time ammonia nitrogen, the dissolved oxygen-ammonia-nitrogen correlation coefficient between the real-time dissolved oxygen and the real-time ammonia nitrogen, and the pH-dissolved oxygen correlation coefficient between the real-time pH value and the real-time dissolved oxygen in the monitoring results of any two monitoring points were calculated according to the Pearson correlation coefficient calculation formula.

[0014] Furthermore, constructing a graph model according to the monitoring results of each monitoring point, the correlation coefficient and the preset standard coefficient includes:

[0015] Each monitoring point is regarded as a node in the network, and the edges in the graph model are constructed according to the calculated pH-ammonia-nitrogen correlation coefficient, dissolved oxygen-ammonia-nitrogen correlation coefficient, pH-dissolved oxygen correlation coefficient and the preset standard coefficient, and the graph model is constructed according to the nodes and edges.

[0016] Furthermore, the edges in the graph model constructed according to the calculated pH-ammonia nitrogen correlation coefficient and the preset standard coefficient include:

[0017] According to the pH-ammonia-nitrogen correlation coefficient calculated based on the real-time pH value and real-time ammonia nitrogen of any two monitoring points, when the pH-ammonia-nitrogen correlation coefficient is greater than a preset standard coefficient, an edge between the nodes corresponding to the two monitoring points is constructed.

[0018] Furthermore, the edges in the graph model constructed according to the calculated dissolved oxygen, ammonia and nitrogen correlation coefficient and the preset standard coefficient include:

[0019] According to the dissolved oxygen and ammonia nitrogen correlation coefficient calculated based on the real-time dissolved oxygen and real-time ammonia nitrogen of any two monitoring points, when the dissolved oxygen and ammonia nitrogen correlation coefficient is greater than the preset standard coefficient, an edge between the nodes corresponding to the two monitoring points is constructed.

[0020] Furthermore, the edges in the graph model constructed according to the calculated pH-dissolved oxygen correlation coefficient and the preset standard coefficient include:

[0021] According to the pH-dissolved oxygen correlation coefficient calculated based on the real-time pH value and real-time dissolved oxygen of any two monitoring points, when the pH-dissolved oxygen correlation coefficient is greater than a preset standard coefficient, an edge between the nodes corresponding to the two monitoring points is constructed.

[0022] Further, determining the pollution source monitoring point according to the number of edges includes:

[0023] Compare the number of edges of all nodes, and select the monitoring point corresponding to the node with the largest number of edges as the pollution source monitoring point.

[0024] Furthermore, the step of setting the hydrodynamic parameters of the hydrodynamic and water quality models according to the input data includes:

[0025] The terrain elevation, roughness, initial water level and boundary conditions of the hydrodynamic and water quality models are set according to the topography, water body characteristics and actual conditions of the study area.

[0026] Further, adjusting the preset standard coefficient according to the verification correlation coefficient and the preset standard verification correlation coefficient includes:

[0027] When the verification correlation coefficient is less than the standard verification correlation coefficient, the difference between the verification correlation coefficient and the standard verification correlation coefficient is calculated. If the difference between the verification correlation coefficient and the standard verification correlation coefficient is less than the preset standard deviation, the standard coefficient is adjusted according to the difference between the verification correlation coefficient and the standard verification correlation coefficient and the standard deviation.

[0028] Further, adjusting the preset quantity according to the verification correlation coefficient and the preset standard verification correlation coefficient includes:

[0029] When the verification correlation coefficient is less than the standard verification correlation coefficient, and the difference between the verification correlation coefficient and the standard verification correlation coefficient is greater than or equal to the preset standard deviation value, the preset quantity is increased and adjusted according to the difference between the verification correlation coefficient and the standard verification correlation coefficient and the standard deviation value.

[0030] Compared with the prior art, the beneficial effect of the present invention lies in that, by comparing the Pearson correlation coefficient of the simulation results with the actual monitoring data, the accuracy of the model is evaluated, and the number of monitoring points or the standard value of the correlation coefficient is adjusted according to the evaluation results to optimize the accuracy and efficiency of the tracing method, and finally an early warning prompt of the pollution source is issued. It effectively combines the graph theory analysis method with the hydrodynamic and water quality models, improves the accuracy and response speed of water pollution tracing, and can quickly identify pollution sources through real-time monitoring and verification of correlation coefficient analysis, providing a scientific basis for environmental management departments to take timely measures to control and mitigate water pollution incidents, effectively solving the problems of complex and inefficient tracing analysis due to incomplete and uneven distribution of water quality monitoring data, and increased difficulty in tracing due to the combined effects of multiple pollution sources.

[0031] Furthermore, by comprehensively applying graph analysis methods and hydrodynamic and water quality models, the accuracy and efficiency of water pollution source tracing are effectively improved. Standardization processing and calculation of correlation coefficients enhance the comparability of data. The construction of graph models intuitively reveals the relationship between water quality parameters, which helps to quickly locate potential pollution sources. Hydrodynamic and water quality models provide dynamic simulations of pollutant diffusion, providing a scientific basis for the response and treatment of pollution incidents. In addition, by dynamically adjusting the number of monitoring points or correlation coefficient thresholds, the method has the ability to self-optimize and can adapt to different water quality monitoring and pollution source tracing needs, enhancing the applicability and flexibility of the model.

[0032] Furthermore, when the difference is small, it indicates that the validation correlation coefficient of the current model has not reached the standard, but is close. In this case, it may not be necessary to make a large adjustment, but to optimize by fine-tuning the standard coefficient.

[0033] Furthermore, when the difference reaches or exceeds the preset standard deviation, it indicates that there is a significant deviation between the model prediction and the actual monitoring data, which may mean that the coverage and representativeness of the existing monitoring points are insufficient. Increasing the number of monitoring points can improve the breadth and depth of data collection, thereby better capturing the spatial variations of water quality parameters and improving the representativeness of the model and prediction accuracy.

[0034] Furthermore, by combining real-time monitoring data, graph analysis and hydrodynamic simulation, rapid identification and accurate tracing of water pollution sources are achieved. By dynamically adjusting the standard coefficient and the number of monitoring points, the method can adapt to different monitoring conditions and water quality changes, and improve the accuracy and reliability of tracing. In addition, the self-adjustment mechanism of the method enables the model to maintain efficient prediction performance in the face of complex and changing water environments, providing strong technical support for water environment management and pollution control. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1This is a flow chart of the graph theory driven water pollution source tracing method based on the hydrodynamics and water quality model of this embodiment;

[0036] Figure 2 This is a decision logic diagram for building an edge based on the pH-ammonia-nitrogen correlation coefficient in this embodiment;

[0037] Figure 3 The decision logic diagram for building an edge according to the correlation coefficient of dissolved oxygen, ammonia and nitrogen in this embodiment;

[0038] Figure 4 This is a decision logic diagram for building an edge based on the pH-dissolved oxygen correlation coefficient in this embodiment. DETAILED DESCRIPTION

[0039] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0040] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0041] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0042] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0043] See also Figure 1 As shown, it is a flow chart of the graph theory driven water pollution source tracing method based on the hydrodynamics and water quality model in this embodiment;

[0044] This embodiment provides a graph-theory driven water pollution source tracing method based on hydrodynamics and water quality models, including:

[0045] Divide the river channel based on the monitoring points and number of pollution sources under the water quality supervision system, or the monitoring points and number of sewage outlets under the complex connected water system, and collect the real-time pH value, real-time ammonia nitrogen and real-time dissolved oxygen at each monitoring point, and standardize the monitoring results of the monitoring points, involving rescaling of the data to have a distribution with a mean of 0 and a standard deviation of 1;

[0046] The calculation formula of Pearson correlation coefficient is used to calculate several correlation coefficients corresponding to the monitoring results of each monitoring point;

[0047] A graph model is constructed based on the monitoring results, correlation coefficients and preset standard coefficients of each monitoring point, the number of edges formed by connecting each node in the graph model is obtained, and the pollution source monitoring point is determined based on the number of edges;

[0048] Obtain river terrain elevation data of pollution source monitoring points and generate digital configuration files. Use CVLGr id to divide structured orthogonal grids and set the hydrodynamic parameters of the hydrodynamic and water quality models according to the actual conditions of the study area, including river terrain elevation, riverbed roughness, initial water level and boundary conditions.

[0049] The time series of ammonia nitrogen at the pollution source monitoring point is loaded into the hydrodynamic and water quality model for simulation calculation to obtain the simulation results within the preset monitoring time;

[0050] The Pearson correlation coefficient is used to calculate the verification correlation coefficient between the simulation results and the time series of ammonia nitrogen at the pollution source monitoring point. According to the verification correlation coefficient and the preset standard verification correlation coefficient, it is determined to adjust the preset quantity, or adjust the preset standard coefficient, or issue a key pollution source prompt.

[0051] By setting a preset number of monitoring points in the river, key water quality parameters such as pH, ammonia nitrogen and dissolved oxygen are monitored in real time, and the correlation coefficients between the monitoring points are calculated using these data. A graph model is constructed based on the monitoring results and correlation coefficients, and the connection strength between nodes is analyzed to identify potential pollution source monitoring points. Furthermore, the river terrain elevation data of these key monitoring points is obtained and digitized to form input data, which is used to set the parameters of the hydrodynamic and water quality models and simulate the diffusion process of ammonia nitrogen pollutants.

[0052] By comparing the Pearson correlation coefficient of the simulation results with the actual monitoring data, the accuracy of the model is evaluated, and the number of monitoring points or the standard value of the correlation coefficient is adjusted according to the evaluation results to optimize the accuracy and efficiency of the traceability method, and finally issue an early warning of the pollution source. It effectively combines graph analysis with hydrodynamic and water quality models to improve the accuracy and response speed of water pollution traceability. Through real-time monitoring and verification of correlation coefficient analysis, it can quickly identify the pollution source, provide a scientific basis for environmental management departments, and take timely measures to control and mitigate water pollution incidents.

[0053] The preset number refers to the number of monitoring points set in the river. This number depends on the size of the river, the expected distribution of pollution sources, the availability of monitoring resources, and the required monitoring density. The general setting will be determined based on previous research, experience, or pre-experimental data.

[0054] The preset standard coefficient is used to determine the correlation coefficient threshold for establishing an edge between two monitoring points when building a graph model. This standard coefficient depends on the expected correlation strength between water quality parameters and the accuracy required for pollution tracing. Usually, this coefficient is determined through expert opinion, historical data analysis, or model testing.

[0055] The standard validation correlation coefficient is an acceptance threshold used to evaluate the correlation between simulation results and actual monitoring data. This coefficient reflects the acceptable accuracy of model predictions. It is usually set based on the purpose and application scenario of the model. For example, if traceability requires very high accuracy, this coefficient may be set higher.

[0056] The standard deviation value is used to determine whether it is necessary to adjust the preset standard coefficient or quantity threshold when the verification correlation coefficient does not meet the standard verification correlation coefficient. The size of this difference depends on the requirements for model sensitivity and robustness.

[0057] Assume that there is the following water quality monitoring data, including the measured values ​​of four monitoring points (A, B, C, D) and three water quality parameters (pH value, ammonia nitrogen, dissolved oxygen):

[0058] Monitoring Points pH Ammonia nitrogen (mg / L) Dissolved oxygen (mg / L) A 7.2 0.5 8.0 B 7.0 0.6 7.5 C 7.1 0.4 8.2 D 7.3 0.7 7.8

[0059] Among them, the correlation coefficient between pH value and ammonia nitrogen, the correlation coefficient between dissolved oxygen and ammonia nitrogen, and the correlation coefficient between pH value and dissolved oxygen in the water quality data are calculated.

[0060] The correlation coefficient between pH value and ammonia nitrogen is obtained by the correlation coefficient calculation method in statistics. The correlation coefficient is a statistic that measures the strength and direction of the linear relationship between two variables. The Pearson correlation coefficient is commonly used.

[0061] The calculation formula of Pearson correlation coefficient is as follows:

[0062]

[0063] in:

[0064] r is the correlation coefficient, which ranges from -1 to 1;

[0065] n is the sample size;

[0066] xi and yi are the values ​​of the two variables at the i-th sample point; and are the means of the two variables respectively;

[0067] For the pH value and ammonia nitrogen in the water quality data, the pH value of each monitoring point can be used as xi, and the ammonia nitrogen value can be used as yi, and then substituted into the above formula for calculation. If the calculated r value is close to 1, it means that there is a strong positive correlation between pH value and ammonia nitrogen; if it is close to -1, it means that there is a strong negative correlation; if it is close to 0, it means that the linear relationship between the two is very weak or does not exist.

[0068] The same method is used to calculate the correlation coefficients between pH value and dissolved oxygen, ammonia nitrogen and dissolved oxygen in water quality data.

[0069] Step 6: Construct the edges in the graph model and calculate the degree of each node.

[0070] Through the above, the correlation coefficients between the three have been calculated at the same time, as follows:

[0071] side pH and Ammonia Nitrogen pH and dissolved oxygen Ammonia nitrogen and dissolved oxygen AB 0.8 0.4 -0.3 AC 0.5 0.6 -0.2 AD 0.7 0.5 -0.4 BC 0.6 0.5 -0.1 BD 0.9 0.7 -0.3 CD 0.4 0.3 -0.2

[0072] Construct the edges of the graph model and set the standard correlation coefficient to 0.7. Then, in the network, there will be edges only between pairs of monitoring points with a correlation coefficient exceeding 0.7.

[0073] AB, based on the correlation coefficient of pH and ammonia nitrogen of 0.8;

[0074] BD, based on the correlation coefficient between pH and ammonia nitrogen of 0.9;

[0075] BD, based on the correlation coefficient of pH and dissolved oxygen of 0.7. Note that this is repeated with the edge above. There should only be one edge in the actual network.

[0076] In this case no edges were added to the network based on the validated correlation coefficient between ammonia nitrogen and dissolved oxygen;

[0077] Calculate the degree of a node:

[0078] The degree of A is 1, which is connected to B, based on the verified correlation coefficient between pH value and ammonia nitrogen;

[0079] The degree of B is 2, which is connected with A and D, based on the verified correlation coefficients of pH and ammonia nitrogen and pH and dissolved oxygen, respectively;

[0080] The degree of C is 0, and it is not connected to any other nodes;

[0081] The degree of D is 2, which is connected to B, based on the verified correlation coefficients of pH value and ammonia nitrogen, and pH value and dissolved oxygen. Note that one degree of B is calculated repeatedly here, and it should be calculated only once;

[0082] The pH degree is 0, and the water quality parameter nodes are not directly connected;

[0083] The degree of ammonia nitrogen is 0, and the water quality parameter nodes are not directly connected;

[0084] The degree of dissolved oxygen is 0, and the water quality parameter nodes are not directly connected;

[0085] In this graph model, there is no direct connection to water quality parameter nodes, such as pH, ammonia nitrogen, and dissolved oxygen, which are ignored this time;

[0086] The degree of a node can reflect the importance and influence of the node in the network. Node B has a higher degree, which means that it may play an important role in the changes of water quality parameters, especially pH value and ammonia nitrogen, or its location is a key area for water quality changes. Monitoring point B is regarded as the key suspect of the pollution source of the pollution incident;

[0087] The river terrain elevation from monitoring point B to the alarm point is digitized to generate a data input file, using structured orthogonal grid division;

[0088] By setting the hydrodynamic parameters for the hydrodynamic and water quality models, including river channel terrain elevation, riverbed roughness, initial water level, and boundary conditions;

[0089] Run the simulation of hydrodynamic parameters, load the temperature module and water quality module, load the time series of ammonia nitrogen at monitoring point B into the water quality module, and calculate the simulation;

[0090] The time series of ammonia nitrogen concentration simulated by the hydrodynamic and water quality models at the pollution occurrence point is extracted and compared with the actual monitoring data, and the verification correlation coefficient is calculated using the Pearson correlation coefficient. If the verification correlation coefficient is above 0.8, it can be verified that monitoring point B is a key pollution source and the enforcement personnel can go there to check.

[0091] Specifically, the calculation of the correlation coefficient in the monitoring results of each monitoring point includes:

[0092] The monitoring results of each monitoring point were standardized and rescaled, and the pH-ammonia-nitrogen correlation coefficient between the real-time pH value and the real-time ammonia nitrogen, the dissolved oxygen-ammonia-nitrogen correlation coefficient between the real-time dissolved oxygen and the real-time ammonia nitrogen, and the pH-dissolved oxygen correlation coefficient between the real-time pH value and the real-time dissolved oxygen in the monitoring results of any two monitoring points were calculated according to the Pearson correlation coefficient calculation formula.

[0093] Specifically, the construction of the graph model according to the monitoring results of each monitoring point, the correlation coefficient and the preset standard coefficient includes:

[0094] Each monitoring point is regarded as a node in the network, and the edges in the graph model are constructed according to the calculated pH-ammonia-nitrogen correlation coefficient, dissolved oxygen-ammonia-nitrogen correlation coefficient, pH-dissolved oxygen correlation coefficient and the preset standard coefficient, and the graph model is constructed according to the nodes and edges.

[0095] Please continue reading Figure 2 As shown, it is a decision logic diagram for determining the construction edge according to the pH ammonia nitrogen correlation coefficient in this embodiment;

[0096] Specifically, the edges in the graph model constructed based on the calculated pH-ammonia nitrogen correlation coefficient and the preset standard coefficient include:

[0097] According to the pH-ammonia-nitrogen correlation coefficient calculated based on the real-time pH value and real-time ammonia nitrogen of any two monitoring points, when the pH-ammonia-nitrogen correlation coefficient is greater than a preset standard coefficient, an edge between the nodes corresponding to the two monitoring points is constructed.

[0098] Please continue reading Figure 3 As shown, it is a decision logic diagram for determining the construction edge according to the dissolved oxygen, ammonia and nitrogen correlation coefficient in this embodiment;

[0099] Specifically, the edges in the graph model constructed based on the calculated dissolved oxygen, ammonia and nitrogen correlation coefficients and the preset standard coefficients include:

[0100] According to the dissolved oxygen and ammonia nitrogen correlation coefficient calculated based on the real-time dissolved oxygen and real-time ammonia nitrogen of any two monitoring points, when the dissolved oxygen and ammonia nitrogen correlation coefficient is greater than the preset standard coefficient, an edge between the nodes corresponding to the two monitoring points is constructed.

[0101] Please continue reading Figure 4 As shown, it is a decision logic diagram for determining the construction edge according to the pH dissolved oxygen correlation coefficient in this embodiment;

[0102] Specifically, the edges in the graph model constructed based on the calculated pH-dissolved oxygen correlation coefficient and the preset standard coefficient include:

[0103] According to the pH-dissolved oxygen correlation coefficient calculated based on the real-time pH value and real-time dissolved oxygen of any two monitoring points, when the pH-dissolved oxygen correlation coefficient is greater than a preset standard coefficient, an edge between the nodes corresponding to the two monitoring points is constructed.

[0104] First, a preset number of monitoring points in the river are monitored in real time to obtain key water quality parameters such as pH, ammonia nitrogen and dissolved oxygen. Then, the monitoring results are standardized and three correlation coefficients between each monitoring point are calculated: the pH-ammonia nitrogen correlation coefficient between real-time pH and ammonia nitrogen, the dissolved oxygen-ammonia nitrogen correlation coefficient between real-time dissolved oxygen and ammonia nitrogen, and the pH-dissolved oxygen correlation coefficient between real-time pH and dissolved oxygen. Using these correlation coefficients and the preset standard coefficients, a graph model is constructed, with the monitoring points as nodes, and edges are constructed between pairs of monitoring points whose correlation coefficients exceed the standard coefficients, forming a network structure that reflects the relationship between water quality parameters. Through graph model analysis, key pollution source monitoring points are identified, and the terrain elevation data of these monitoring points are used to perform hydrodynamic simulation through hydrodynamic and water quality models to predict the diffusion of ammonia nitrogen. Finally, the accuracy of the model is evaluated by comparing the verification correlation coefficients of the simulation results with the actual monitoring data, and the standard threshold of the number of monitoring points or correlation coefficients is adjusted as needed.

[0105] By comprehensively applying graph theory and hydrodynamic and water quality models, the accuracy and efficiency of water pollution source tracing are effectively improved. Standardization processing and calculation of correlation coefficients enhance the comparability of data. The construction of graph models intuitively reveals the relationship between water quality parameters, which helps to quickly locate potential pollution sources. The application of hydrodynamic and water quality models provides a dynamic simulation of pollutant diffusion, providing a scientific basis for the response and treatment of pollution incidents. In addition, by dynamically adjusting the number of monitoring points or the correlation coefficient threshold, the method has the ability to self-optimize, can adapt to different water quality monitoring and pollution source tracing needs, and enhances the applicability and flexibility of the model.

[0106] Specifically, determining the pollution source monitoring point according to the number of edges includes:

[0107] Compare the number of edges of all nodes, and select the monitoring point corresponding to the node with the largest number of edges as the pollution source monitoring point.

[0108] Specifically, the hydrodynamic parameters of the hydrodynamic and water quality models are set according to the actual topography and water characteristics of the study area, including:

[0109] The terrain elevation, roughness, initial water level and boundary conditions of the hydrodynamic and water quality models are set according to the elevation data of the actual study area.

[0110] Specifically, adjusting the preset standard coefficient according to the verification correlation coefficient and the preset standard verification correlation coefficient includes:

[0111] When the verification correlation coefficient is less than the standard verification correlation coefficient, the difference between the verification correlation coefficient and the standard verification correlation coefficient is calculated. If the difference between the verification correlation coefficient and the standard verification correlation coefficient is less than the preset standard deviation, the standard coefficient is adjusted according to the difference between the verification correlation coefficient and the standard verification correlation coefficient and the standard deviation.

[0112] When the difference is small, it indicates that the validation correlation coefficient of the current model has not reached the standard, but is close. In this case, it may not be necessary to make a large adjustment, but to optimize by fine-tuning the standard coefficient.

[0113] Specifically, adjusting the preset quantity according to the verification correlation coefficient and the preset standard verification correlation coefficient includes:

[0114] When the verification correlation coefficient is less than the standard verification correlation coefficient, and the difference between the verification correlation coefficient and the standard verification correlation coefficient is greater than or equal to the preset standard deviation value, the preset quantity is increased and adjusted according to the difference between the verification correlation coefficient and the standard verification correlation coefficient and the standard deviation value.

[0115] When the difference reaches or exceeds the preset standard deviation, it indicates that there is a significant deviation between the model prediction and the actual monitoring data, which may mean that the coverage and representativeness of the existing monitoring points are insufficient. Increasing the number of monitoring points can improve the breadth and depth of data collection, thereby better capturing the spatial variations of water quality parameters and improving the representativeness of the model and prediction accuracy.

[0116] First, a set number of monitoring points in the river are monitored in real time to obtain pH, ammonia nitrogen and dissolved oxygen data, then these data are standardized and the correlation coefficients between the monitoring points are calculated. These correlation coefficients are used to build a graph model, and the pollution source monitoring points are identified by comparing the number of edges of the nodes. Furthermore, the input file of the hydrodynamic and water quality model is generated based on the terrain elevation data of the selected pollution source monitoring point, and this file is used to set hydrodynamic parameters such as terrain elevation, roughness, initial water level and boundary conditions in the hydrodynamic and water quality model. After simulating the diffusion of ammonia nitrogen, the Pearson correlation coefficient is used to analyze the verification correlation coefficient between the simulation results and the actual data. According to the difference between the verification correlation coefficient and the preset standard, the standard coefficient or the number of monitoring points is adjusted in time to optimize the prediction accuracy of the model.

[0117] By combining real-time monitoring data, graph analysis and hydrodynamic simulation, the rapid identification and accurate tracing of water pollution sources are achieved. By dynamically adjusting the standard coefficient and the number of monitoring points, the method can adapt to different monitoring conditions and water quality changes, and improve the accuracy and reliability of tracing. In addition, the self-adjustment mechanism of the method enables the model to maintain efficient prediction performance when facing complex and changing water environments, providing strong technical support for water environment management and pollution control.

[0118] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0119] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A graph-driven water pollution source tracing method based on hydrodynamics and water quality models, characterized in that: include: Divide the monitoring points in the river into preset numbers, and monitor the real-time pH value, real-time ammonia nitrogen and real-time dissolved oxygen of each monitoring point; Calculate a number of correlation coefficients corresponding to the monitoring results of each monitoring point; A graph model is constructed based on the monitoring results, correlation coefficients and preset standard coefficients of each monitoring point, the number of edges formed by connecting each node in the graph model is obtained, and the pollution source monitoring point is determined based on the number of edges; Obtain river terrain elevation data of pollution source monitoring points, generate input data digitally, and set hydrodynamic parameters of hydrodynamic and water quality models based on the input data; The time series of ammonia nitrogen at the pollution source monitoring point is loaded into the hydrodynamic and water quality model for simulation calculation to obtain the simulation results within the preset monitoring time; The Pearson correlation coefficient is used to calculate the verification correlation coefficient between the simulation result and the time series of ammonia nitrogen at the pollution source monitoring point. According to the verification correlation coefficient and the preset standard verification correlation coefficient, the preset quantity is adjusted, or the preset standard coefficient is adjusted, or a key pollution source prompt is issued; According to the verification correlation coefficient and the preset standard verification correlation coefficient, the preset standard coefficient is adjusted including: When the verification correlation coefficient is less than the standard verification correlation coefficient, the difference between the verification correlation coefficient and the standard verification correlation coefficient is calculated. If the difference between the verification correlation coefficient and the standard verification correlation coefficient is less than the preset standard deviation, the standard coefficient is adjusted according to the difference between the verification correlation coefficient and the standard verification correlation coefficient and the standard deviation; The number of presets for validation correlation coefficient adjustment based on validation correlation coefficient and preset standard validation correlation coefficient includes: When the verification correlation coefficient is less than the standard verification correlation coefficient, and the difference between the verification correlation coefficient and the standard verification correlation coefficient is greater than or equal to the preset standard deviation value, the preset quantity is increased and adjusted according to the difference between the verification correlation coefficient and the standard verification correlation coefficient and the standard deviation value.

2. The graph-driven water pollution source tracing method based on hydrodynamics and water quality model according to claim 1 is characterized in that: The calculation of the correlation coefficient in the monitoring results of each monitoring point includes: The monitoring results of each monitoring point were standardized and rescaled, and the pH-ammonia-nitrogen correlation coefficient between the real-time pH value and the real-time ammonia nitrogen, the dissolved oxygen-ammonia-nitrogen correlation coefficient between the real-time dissolved oxygen and the real-time ammonia nitrogen, and the pH-dissolved oxygen correlation coefficient between the real-time pH value and the real-time dissolved oxygen in the monitoring results of any two monitoring points were calculated according to the Pearson correlation coefficient calculation formula.

3. The graph-theory driven water pollution source tracing method based on hydrodynamics and water quality model according to claim 2 is characterized in that: The construction of the graph model according to the monitoring results of each monitoring point, the correlation coefficient and the preset standard coefficient includes: Each monitoring point is regarded as a node in the network, and the edges in the graph model are constructed according to the calculated pH-ammonia-nitrogen correlation coefficient, dissolved oxygen-ammonia-nitrogen correlation coefficient, pH-dissolved oxygen correlation coefficient and the preset standard coefficient, and the graph model is constructed according to the nodes and edges.

4. The graph-driven water pollution source tracing method based on hydrodynamics and water quality model according to claim 3 is characterized in that: The edges in the graph model constructed based on the calculated pH-ammonia-nitrogen correlation coefficient and the preset standard coefficient include: According to the pH-ammonia-nitrogen correlation coefficient calculated based on the real-time pH value and real-time ammonia nitrogen of any two monitoring points, when the pH-ammonia-nitrogen correlation coefficient is greater than a preset standard coefficient, an edge between the nodes corresponding to the two monitoring points is constructed.

5. The graph-theory driven water pollution source tracing method based on hydrodynamics and water quality model according to claim 4 is characterized in that: The edges in the graph model constructed based on the calculated dissolved oxygen, ammonia and nitrogen correlation coefficients and the preset standard coefficients include: According to the dissolved oxygen and ammonia nitrogen correlation coefficient calculated based on the real-time dissolved oxygen and real-time ammonia nitrogen of any two monitoring points, when the dissolved oxygen and ammonia nitrogen correlation coefficient is greater than the preset standard coefficient, an edge between the nodes corresponding to the two monitoring points is constructed.

6. The graph-theory driven water pollution source tracing method based on hydrodynamics and water quality model according to claim 5 is characterized in that: The edges in the graph model constructed based on the calculated pH dissolved oxygen correlation coefficient and the preset standard coefficient include: According to the pH-dissolved oxygen correlation coefficient calculated based on the real-time pH value and real-time dissolved oxygen of any two monitoring points, when the pH-dissolved oxygen correlation coefficient is greater than a preset standard coefficient, an edge between the nodes corresponding to the two monitoring points is constructed.

7. The graph-theory driven water pollution source tracing method based on hydrodynamics and water quality model according to claim 6 is characterized in that: Determining the pollution source monitoring point according to the number of edges includes: Compare the number of edges of all nodes, and select the monitoring point corresponding to the node with the largest number of edges as the pollution source monitoring point.

8. The graph-driven water pollution source tracing method based on hydrodynamics and water quality model according to claim 7 is characterized in that: The step of setting the hydrodynamic parameters of the hydrodynamic and water quality models according to the input data includes: The terrain elevation, roughness, initial water level and boundary conditions of the hydrodynamic and water quality models are set according to the topography, water body characteristics and actual conditions of the study area.

Images