Water quality monitoring method and system based on Internet of Things technology

Abstract

The invention relates to the technical field of water quality monitoring, in particular to a water quality monitoring method and system based on the Internet of Things technology. The method comprises the following steps: acquiring multi-source water quality monitoring data of a monitoring area; analyzing the multi-source water quality monitoring data, and determining whether illegal pollution discharge possibility exists or not; if yes, water pipe layout information and area information of the monitoring area are obtained; analyzing the water pipe layout information, and determining water flow path association; determining pollution dissipation path information based on the water flow path association according to the multi-source water quality monitoring data; analyzing the regional information, and determining factory distribution characteristics; and determining an illegal pollution discharge point according to the water flow path association, the plant area distribution characteristics and the dissipation path information. The spatial consistency of the final positioning result and the physical pipe network layout is ensured, and the false alarm caused by the model error is eliminated.

Description

Technical Field

[0001] The present application relates to the technical field of water quality monitoring, and in particular, to a water quality monitoring method and system based on Internet of Things technology. Background Art

[0002] With the rapid development of Internet of Things technology, the water quality monitoring field has gradually realized the upgrade transformation from single-node detection to distributed intelligent monitoring network. The traditional water quality monitoring system constructs a pollutant identification system based on fixed-point data collection by deploying fixed detection devices at the discharge outlets of factories or key nodes of municipal pipe networks.

[0003] However, the existing technology still has significant defects in dealing with complex urban sewage discharge scenarios. For example, in the working condition of shared pipe networks, the traditional upstream and downstream comparison detection method cannot accurately trace and quickly locate illegal sewage discharge behaviors. Summary of the Invention

[0004] The present application provides a water quality monitoring method and system based on Internet of Things technology to solve the above problems.

[0005] In a first aspect, the present application provides a water quality monitoring method based on Internet of Things technology, the method comprising: Obtaining multi-source water quality monitoring data of a monitoring area; analyzing the multi-source water quality monitoring data to determine whether there is a possibility of illegal sewage discharge; If so, obtaining the water pipe layout information and area information of the monitoring area; analyzing the water pipe layout information to determine the water flow path association; Based on the water flow path association, determining pollution dissipation path information according to the multi-source water quality monitoring data; Analyzing the area information to determine the factory distribution characteristics; Determining illegal sewage discharge points according to the water flow path association, the factory distribution characteristics and the dissipation path information.

[0006] Through this solution, multi-source water quality monitoring data of the monitoring area is obtained to eliminate the modeling error problem caused by inconsistent data time bases. Analyzing the multi-source water quality monitoring data to determine whether there may be illegal sewage discharge helps to avoid false triggering by a single parameter. If so, obtaining the water pipe layout information and regional information of the monitoring area helps to eliminate the modeling defect of the step change in flow velocity caused by sudden changes in pipe diameter. Analyzing the water pipe layout information to determine the water flow path association helps to improve the adaptability of complex pipe network topologies. Based on the water flow path association, according to the multi-source water quality monitoring data, determining the pollution dissipation path information helps to realize the mapping between the dissipation path and the hidden pipe space. Analyzing the regional information to determine the distribution characteristics of factory areas helps to improve the detection rate of hidden pipes. According to the water flow path association, the distribution characteristics of factory areas, and the dissipation path information, determining the illegal sewage discharge points ensures the spatial consistency between the final positioning result and the physical pipe network layout, and eliminates false alarms caused by model errors.

[0007] Optionally, the analyzing the water pipe layout information to determine the water flow path association includes: Analyzing the water pipe layout information to determine the water pipe intersection points, pipe diameters, and laying slopes; Parsing the multi-source water quality monitoring data to determine the data sources; According to the data sources, determining the water quality monitoring points; Analyzing the real-time flow velocity of each water quality monitoring point in the multi-source water quality monitoring data; According to the water pipe intersection points, the pipe diameters, the laying slopes, and the real-time flow velocity, constructing a dynamic weighted directed graph model; According to the dynamic weighted directed graph model, determining the water flow path association.

[0008] Optionally, the determining the pollution dissipation path information based on the water flow path association and according to the multi-source water quality monitoring data includes: Based on the water flow path association, obtaining a historical relevant pollution data set; parsing the historical relevant pollution data set to determine the historical pollution type, initial pollution concentration, and dissipation time; According to the historical pollution type, the initial pollution concentration, and the dissipation time, establishing a pollution diffusion model; Based on the pollution diffusion model, according to the multi-source water quality monitoring data, determining the pollution dissipation path information.

[0009] Optionally, the analyzing the regional information to determine the distribution characteristics of factory areas includes: Analyzing the regional information to determine the factory area coordinates; According to the factory area coordinates, determining the physical distance between each water quality monitoring point and each factory area; Based on the physical distance, establish a topological relationship matrix between the water quality monitoring points and the coordinates of the factory area; Determine the distribution characteristics of the factory area according to the topological relationship matrix.

[0010] Optionally, the determining of the illegal sewage discharge points according to the water flow path association, the factory area distribution characteristics, and the dissipation path information includes: Based on the water flow path association and the pollution dissipation path information, construct a pollution source tracing probability graph model; According to the distribution characteristics of the factory area, determine the spatial correlation degree between each factory area and the pollution dissipation path; Obtain the sewage discharge permit information of the factory area, and determine the sewage discharge probability of the factory area according to the sewage discharge permit information of the factory area and the spatial correlation degree; Take the sewage discharge probability of the factory area as the prior probability and input it into the pollution source tracing probability graph model to obtain an output result; Calculate the output result through Bayesian inference to determine the posterior probability of each water quality monitoring point; Compare the posterior probability with the preset threshold of each water quality monitoring point, and select the water quality monitoring points that exceed the preset threshold as candidate illegal sewage discharge points; According to the water flow path association and the pollution dissipation path information, conduct path consistency verification on the candidate illegal sewage discharge points. If the verification passes, determine the candidate illegal sewage discharge points as illegal sewage discharge points.

[0011] Optionally, after determining that the candidate illegal sewage discharge points are illegal sewage discharge points if the verification passes, it further includes: Obtain the factory area building layout plan and the municipal pipe network topology map; Analyze the municipal pipe network topology map and the factory area building layout plan to determine the public pipe network structure and the factory area pipe network structure; According to the public pipe network structure and the factory area pipe network structure, determine the spatial position relationship between the factory area and the public pipe network; According to the spatial position relationship, determine the potential hidden pipe laying path; According to the change of the pollutant concentration gradient on the potential hidden pipe laying path in the pollution dissipation path information, verify the possibility of the existence of the hidden pipe; Based on the possibility of the existence of the hidden pipe, determine the hidden pipe sewage discharge point according to the potential hidden pipe laying path.

[0012] Optionally, the determining of the potential hidden pipe laying path according to the spatial position relationship includes: Analyze the public pipe network structure to determine the public pipe network nodes; Analyze the factory area pipe network structure to determine the spatial coordinates of the factory area drainage facilities; Calculate the Euclidean distance between each spatial coordinate and the nearest common pipe network node to obtain a distance set; Obtain plant area information, analyze the plant area information, and determine the plant area drainage demand; Construct a path optimization function according to the distance set and the plant area drainage demand; Calculate the potential buried pipe laying path according to the path optimization function.

[0013] Optionally, based on the possibility of the existence of the buried pipe, determining the buried pipe sewage discharge point according to the potential buried pipe laying path includes: Obtain historical cases of detected buried pipes, analyze the historical cases of detected buried pipes, and determine a path feature data set; Parse the path feature data set to determine the path length, burial depth, and number of turning angles in each historical case of detected buried pipes; Construct a logistic regression model according to the path length, the burial depth, and the number of turning angles; Input the possibility of the existence of the buried pipe into the logistic regression model to determine whether the confidence level of at least one potential buried pipe laying path is higher than the confidence requirement; If so, determine the potential buried pipe laying path corresponding to the confidence level higher than the confidence requirement as the buried pipe sewage discharge point.

[0014] Optionally, the method further includes: Determine the pollution point coordinates and the administrative region to which the sewage discharge point belongs according to the plant area coordinates; Determine the pollution diffusion range according to the pollution dissipation path information; When the position of the sewage discharge point is determined, generate a warning message according to the pollution point coordinates and the pollution diffusion range; Route the warning message to the corresponding supervision terminal according to the administrative region to which the sewage discharge point belongs, and record the response timestamp of the supervision terminal to form a closed-loop management log.

[0015] In a second aspect, the present application provides a water quality monitoring system based on Internet of Things technology, and the system includes: A data analysis module, configured to obtain multi-source water quality monitoring data of a monitoring area; analyze the multi-source water quality monitoring data to determine whether there is a possibility of illegal sewage discharge; An information analysis module, configured to, if so, obtain the water pipe layout information and area information of the monitoring area; analyze the water pipe layout information to determine the water flow path association; A dissipation analysis module, configured to determine pollution dissipation path information based on the water flow path association and according to the multi-source water quality monitoring data; A feature determination module, configured to analyze the area information to determine the plant area distribution characteristics; A sewage discharge determination module, configured to determine illegal sewage discharge points according to the water flow path association, the factory area distribution characteristics, and the dissipation path information.

[0016] Optionally, when the information analysis module analyzes the water pipe layout information to determine the water flow path association, it is used for: analyzing the water pipe layout information to determine the water pipe intersection points, pipe diameters, and laying slopes; parsing the multi-source water quality monitoring data to determine the data sources; determining the water quality monitoring points according to the data sources; analyzing the multi-source water quality monitoring data, the real-time flow velocity of each water quality monitoring point; constructing a dynamic weighted directed graph model according to the water pipe intersection points, the pipe diameters, the laying slopes, and the real-time flow velocity; determining the water flow path association according to the dynamic weighted directed graph model.

[0017] Optionally, when the dissipation analysis module determines the pollution dissipation path information based on the water flow path association and the multi-source water quality monitoring data, it is used for: obtaining a historical relevant pollution data set based on the water flow path association; parsing the historical relevant pollution data set to determine the historical pollution type, the initial pollution concentration, and the dissipation time; establishing a pollution diffusion model according to the historical pollution type, the initial pollution concentration, and the dissipation time; determining the pollution dissipation path information based on the pollution diffusion model and the multi-source water quality monitoring data.

[0018] Optionally, when the feature determination module analyzes the area information to determine the factory area distribution characteristics, it is used for: analyzing the area information to determine the factory area coordinates; determining the physical distance between each water quality monitoring point and each factory area according to the factory area coordinates; establishing a topological relationship matrix between the water quality monitoring points and the factory area coordinates based on the physical distance; determining the factory area distribution characteristics according to the topological relationship matrix.

[0019] Optionally, when the sewage discharge determination module determines the illegal sewage discharge points according to the water flow path association, the factory area distribution characteristics, and the dissipation path information, it is used for: constructing a pollution source tracing probability graph model based on the water flow path association and the pollution dissipation path information; determining the spatial association degree between each factory area and the pollution dissipation path according to the factory area distribution characteristics; obtaining the factory area sewage discharge permit information, and determining the factory area sewage discharge probability according to the factory area sewage discharge permit information and the spatial association degree; using the factory area sewage discharge probability as the prior probability and inputting it into the pollution source tracing probability graph model to obtain an output result; calculating the output result through Bayesian inference to determine the posterior probability of each water quality monitoring point; comparing the posterior probability with the preset threshold of each water quality monitoring point, and screening the water quality monitoring points that exceed the preset threshold as candidate illegal sewage discharge points; verifying the path consistency of the candidate illegal sewage discharge points according to the water flow path association and the pollution dissipation path information, and if the verification passes, determining the candidate illegal sewage discharge points as illegal sewage discharge points.

[0020] Optionally, the water quality monitoring system based on the Internet of Things technology further includes a hidden pipe determination module, which is used to: obtain the plant building layout plan and the municipal pipe network topology map; analyze the municipal pipe network topology map and the plant building layout plan to determine the public pipe network structure and the plant pipe network structure; determine the spatial position relationship between the plant and the public pipe network according to the public pipe network structure and the plant pipe network structure; determine the potential hidden pipe laying path according to the spatial position relationship; verify the possibility of the existence of a hidden pipe according to the change of the pollutant concentration gradient on the potential hidden pipe laying path based on the pollution dissipation path information; and determine the hidden pipe sewage discharge point based on the possibility of the existence of the hidden pipe according to the potential hidden pipe laying path.

[0021] Optionally, when the hidden pipe determination module determines the potential hidden pipe laying path according to the spatial position relationship, it is used to: analyze the public pipe network structure to determine the public pipe network nodes; analyze the plant pipe network structure to determine the spatial coordinates of the plant drainage facilities; calculate the Euclidean distance between each spatial coordinate and the nearest public pipe network node to obtain a distance set; obtain the plant information, analyze the plant information to determine the plant drainage demand; construct a path optimization function according to the distance set and the plant drainage demand; and calculate the potential hidden pipe laying path according to the path optimization function.

[0022] Optionally, when the hidden pipe determination module determines the hidden pipe sewage discharge point based on the possibility of the existence of the hidden pipe according to the potential hidden pipe laying path, it is used to: obtain the historical hidden pipe cases that have been detected, analyze the historical hidden pipe cases that have been detected to determine the path feature data set; analyze the path feature data set to determine the path length, burial depth, and number of turning angles in each historical hidden pipe case that has been detected; construct a logistic regression model according to the path length, the burial depth, and the number of turning angles; input the possibility of the existence of the hidden pipe into the logistic regression model to determine whether the confidence level of at least one potential hidden pipe laying path is higher than the confidence requirement; if so, determine the potential hidden pipe laying path corresponding to the confidence level higher than the confidence requirement as the hidden pipe sewage discharge point.

[0023] Optionally, the water quality monitoring system based on the Internet of Things technology further includes a log formation module, which is used to: determine the pollution point coordinates and the administrative region to which the sewage discharge point belongs according to the plant coordinates; determine the pollution diffusion range according to the pollution dissipation path information; when the position of the sewage discharge point is determined, generate a warning message according to the pollution point coordinates and the pollution diffusion range; route the warning message to the corresponding supervision terminal according to the administrative region to which the sewage discharge point belongs, and record the response time stamp of the supervision terminal to form a closed-loop management log. Description of the Drawings

[0024] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0025] Figure 1 A schematic diagram of an application scenario provided by an embodiment of the present application; Figure 2 A flowchart of a water quality monitoring method based on Internet of Things technology provided by an embodiment of the present application; Figure 3 A schematic diagram of the structure of a water quality monitoring system based on Internet of Things technology provided by an embodiment of the present application. Detailed implementation manners

[0026] To make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0027] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, unless otherwise specified.

[0028] The following will further describe the embodiments of the present application in detail with reference to the accompanying drawings of the specification.

[0029] The prior art still has significant deficiencies in dealing with complex urban sewage discharge scenarios. For example, in the working condition of shared pipe networks, the traditional upstream and downstream comparison detection method cannot accurately trace and quickly locate illegal sewage discharge behaviors.

[0030] Based on this, the present application provides a water quality monitoring method and system based on Internet of Things technology, which acquires multi-source water quality monitoring data in the monitoring area and eliminates the modeling error problem caused by inconsistent data time bases. Analyzing the multi-source water quality monitoring data to determine whether there is a possibility of illegal sewage discharge helps to avoid false triggering by a single parameter. If so, acquiring the water pipe layout information and area information in the monitoring area helps to eliminate the modeling defect caused by the step change in flow velocity due to sudden change in pipe diameter. Analyzing the water pipe layout information to determine the water flow path association helps to improve the adaptability to complex pipe network topologies. Based on the water flow path association, according to the multi-source water quality monitoring data, determining the pollution dissipation path information helps to realize the spatial mapping between the dissipation path and the hidden pipe. Analyzing the area information to determine the factory distribution characteristics helps to improve the discovery rate of hidden pipes. According to the water flow path association, factory distribution characteristics and dissipation path information, determining the illegal sewage discharge point ensures the spatial consistency between the final positioning result and the physical pipe network layout and eliminates false alarms caused by model errors.

[0031] Figure 1 FIG. is a schematic diagram of an application scenario provided by the present application. When performing water quality monitoring, the method provided by the present application is applied.

[0032] Specifically, the method provided by the present application is applied to any server. The server interacts with the sensor group, acquires multi-source water quality monitoring data in the monitoring area through the sensor group, analyzes the multi-source water quality monitoring data, and determines whether there is a possibility of illegal sewage discharge. If so, it acquires the water pipe layout information and area information in the monitoring area, analyzes the water pipe layout information, and determines the water flow path association. Based on the water flow path association, according to the multi-source water quality monitoring data, it determines the pollution dissipation path information and realizes the spatial mapping between the dissipation path and the hidden pipe. Analyzing the area information to determine the factory distribution characteristics improves the discovery rate of hidden pipes. According to the water flow path association, factory distribution characteristics and dissipation path information, it determines the illegal sewage discharge point, ensures the spatial consistency between the final positioning result and the physical pipe network layout, and eliminates false alarms caused by model errors.

[0033] The specific implementation manner can refer to the following embodiments.

[0034] Figure 2 FIG. is a flowchart of a water quality monitoring method based on Internet of Things technology provided by an embodiment of the present application. The method of this embodiment can be applied to the server in the above scenario. As Figure 2 shown, the method includes: S201. Acquire multi-source water quality monitoring data in the monitoring area; analyze the multi-source water quality monitoring data to determine whether there is a possibility of illegal sewage discharge; The monitoring area can be a continuous spatial range including factory discharge outlets, public pipe network nodes, and natural water body monitoring points.

[0035] Multi-source water quality monitoring data can be a heterogeneous data set of three types of sensor groups.

[0036] Illegal sewage discharge may be an abnormal event of pollutant concentration.

[0037] Specifically, deploy three types of sensor groups at the monitoring area, namely the sensor group at the factory discharge outlet, the monitoring nodes of the public pipe network, and the monitoring points of the natural water body; then, receive the multi-source water quality monitoring data of each group through the Internet of Things gateway. Establish a moving baseline for each monitoring point. When the real-time data deviates too much from the baseline, a primary alarm is triggered; then, based on the law of conservation of fluid matter, establish a preset proportional relationship between the change rate of conductivity and the change rate of flow rate; furthermore, implement multi-parameter correlation analysis. When the direction of change of the sensor group at the factory discharge outlet deviates from the preset proportional relationship with the conductivity, a secondary alarm is generated; subsequently, according to the classification and regression tree algorithm in statistical learning theory, generate a decision tree model from the historical pollution event database through the principle of minimizing the Gini index; finally, determine whether there is a possibility of illegal sewage discharge according to the composite conditions such as turbidity and flow velocity of the three types of monitoring points.

[0038] Among them, deploy the first type of sensor group at the factory discharge outlet, including an online COD analyzer, an electromagnetic flowmeter, and an explosion-proof pH meter; deploy the second type of sensor group at the monitoring nodes of the public pipe network, equipped with a Doppler flowmeter, a pressure transmitter, and a temperature-compensated conductivity sensor; deploy the third type of sensor group at the monitoring points of the natural water body, using a solar-powered turbidity sensor, a dissolved oxygen probe, and an ammonia nitrogen analysis module.

[0039] S202. If it exists, obtain the pipe layout information and area information of the monitoring area; analyze the pipe layout information to determine the water flow path association; The pipe layout information can be the public pipe network topology structure represented by a weighted directed graph.

[0040] The area information can be a structured data set covering the geographical coordinate boundary of the factory area, the declared drainage volume, and the location of the pipe network access point.

[0041] The water flow path association can be a set of the main path and branch paths of pollutant diffusion.

[0042] Specifically, if there is a possibility of illegal sewage discharge, vector data of the pipeline network in the monitoring area is imported from the municipal GIS through the OGC WFS service interface and converted into water pipe layout information. Then, the set of boundary coordinate points of the factory area in the factory area CAD drawing is extracted. The set of coordinate points represents the discrete position data of the geographical outline of the factory area, usually represented in a two-dimensional or three-dimensional coordinate system. Based on these boundary coordinate points, a minimum bounding rectangle is calculated: this rectangle can completely enclose all boundary coordinate points and has the smallest area; the initial rectangle boundary is parallel to the coordinate axes, and the inclusiveness is optimized by iteratively adjusting the rectangle direction and size to ensure that the rectangle vertices are determined by the extreme points in the factory area boundary coordinate points. Furthermore, starting from the center point of the minimum bounding rectangle of the factory area, each vertex in the set of pipeline network vertices is used as the target position for calculation. Based on the physical topology of the pipeline network, the shortest feasible path of the cumulative distance is searched in the physical path connection from the starting position to each target position, and the cumulative distance value is the required shortest path distance. Thus, a factory area - monitoring point association matrix is established to determine the regional information. The specific method for establishing the factory area - monitoring point association matrix is as follows: create a matrix data structure, where the rows of the matrix represent the factory areas (uniquely identified by the center points of their minimum bounding rectangles), the columns of the matrix represent the pipeline network vertices (especially the positions of the water quality monitoring points among them), and each element in the matrix stores the shortest path distance value from the corresponding factory area center point to the pipeline network vertex (monitoring point). This matrix quantifies the connection distance or spatial association strength between the factory areas and each monitoring point in the pipeline network topological space. The center point coordinates of the minimum bounding rectangle of the factory area are spatially matched with the set of pipeline network vertices in the water pipe layout information to determine the water flow path association.

[0043] S203. Based on the water flow path association, determine the pollution dissipation path information according to the multi-source water quality monitoring data; The pollution dissipation path information can be the spatio-temporal migration trajectory of pollutants in the pipeline network.

[0044] Specifically, the dissipation coefficient is trained with on-site measured data and the water flow velocity and pipeline curvature to establish a pollutant concentration attenuation model; furthermore, the moving time window algorithm is used to dynamically update the pollutant characteristic index; subsequently, based on the Gaussian process regression algorithm, starting from the moment when the pollution peak appears, combined with the real-time water flow velocity field for backward diffusion simulation, the pollution dissipation path information with an overly high probability density is generated.

[0045] S204. Analyze the regional information to determine the factory area distribution characteristics; The factory area distribution characteristics can be a topological matrix that quantitatively characterizes the spatial relationship between each factory area and the monitoring points.

[0046] Specifically, a multi-dimensional matrix is constructed based on the number of factory areas and the number of monitoring nodes. Furthermore, by calculating the reciprocal of the product of the shortest water flow path length and the pipe diameter between the factory areas and the monitoring nodes, the matrix elements representing the pipe network path connectivity between each factory area and each monitoring node are obtained. Subsequently, taking the central point coordinates of the minimum circumscribed rectangle of the factory area generated in the previous step as the reference position for spatial analysis, based on the water flow path association and the pollutant diffusion trajectory calculated from multi-source water quality monitoring data, which is manifested as a continuous spatial path or a discrete high-probability point set in the pipe network, the boundary coordinates of the monitoring area are defined as the reference range for density calculation. Taking the central point coordinates of each factory area as the core position, a spatial influence function is defined, which describes the attenuation law of the influence of the factory area on the surrounding space, usually manifested as a distribution pattern centered on the factory area center and smoothly attenuating with the increase of distance. In the monitoring area, for each spatial position (or discrete path point) covered by the pollution dissipation path, calculate the superimposed spatial influence intensity value it receives from the central points of all factory areas. This intensity value reflects the comprehensive spatial proximity of this position to all factory areas. The spatial influence intensity values covering the entire monitoring area are standardized, and the intensity values are mapped onto the geographical space grid to form a continuous spatial density distribution surface. The density distribution surface is visualized to generate a heat map. The darker (or warmer) the color in the heat map, the higher the comprehensive spatial correlation of this position with all factory areas, that is, the higher the potential spatial overlap with the pollution dissipation path. The generated heat map is superimposed and analyzed with the spatial range of the pollution dissipation path, and the spatial density value corresponding to the position of the central point of each factory area in the heat map is calculated, and a spatial density threshold is set. The factory areas with spatial density values exceeding this threshold are marked as suspicious objects, and the factory areas with excessive overlapping areas are screened as suspicious objects, so as to determine the distribution characteristics of the factory areas.

[0047] S205. Determine the illegal sewage discharge points according to the water flow path association, the distribution characteristics of the factory areas and the dissipation path information.

[0048] The illegal sewage discharge point can be the geographical coordinate point of illegal sewage discharge.

[0049] Specifically, a scoring system including the water flow path association, the distribution characteristics of the factory areas, and the dissipation path information is constructed. Furthermore, the threshold is comprehensively determined based on the lower limit of the statistical confidence interval in the USEPA multi-pollutant source tracing study and the risk acceptable level parameter specified in the Technical Guidelines for Risk Assessment of Polluted Sites. Finally, when the comprehensive score of the factory area exceeds the threshold and there is a positive detection of hidden pipes, it is determined as an illegal sewage discharge point.

[0050] Through this solution, multi-source water quality monitoring data of the monitoring area is obtained to eliminate the modeling error problem caused by inconsistent data time bases. Analyze the multi-source water quality monitoring data to determine whether there may be illegal sewage discharge, which helps to avoid false triggering by a single parameter. If so, obtain the water pipe layout information and regional information of the monitoring area, which helps to eliminate the modeling defect of the step change in flow velocity caused by sudden changes in pipe diameter. Analyze the water pipe layout information to determine the water flow path association, which helps to improve the adaptability of complex pipe network topologies. Based on the water flow path association, according to the multi-source water quality monitoring data, determine the pollution dissipation path information, which helps to realize the mapping between the dissipation path and the hidden pipe space. Analyze the regional information to determine the distribution characteristics of the factory area, which helps to improve the discovery rate of hidden pipes. According to the water flow path association, the distribution characteristics of the factory area, and the dissipation path information, determine the illegal sewage discharge points to ensure the spatial consistency between the final positioning result and the physical pipe network layout, and eliminate false alarms caused by model errors.

[0051] In some embodiments, analyze the water pipe layout information to determine the water pipe intersection points, pipe diameters, and laying slopes; parse the multi-source water quality monitoring data to determine the data sources; according to the data sources, determine the water quality monitoring points; analyze the multi-source water quality monitoring data, the real-time flow velocity of each water quality monitoring point; according to the water pipe intersection points, pipe diameters, laying slopes, and real-time flow velocity, construct a dynamic weighted directed graph model; according to the dynamic weighted directed graph model, determine the water flow path association.

[0052] The water pipe intersection points can be the physical connection position coordinates of different sewage pipes.

[0053] The pipe diameter can be the measured physical parameter of the inner diameter of the sewage pipe.

[0054] The laying slope can be the inclination angle formed by the pipeline during laying with the horizontal plane.

[0055] The data source can be the type of equipment that generates the multi-source water quality monitoring data.

[0056] The water quality monitoring points can be the specific physical locations where water quality sensors are deployed.

[0057] The real-time flow velocity can be the instantaneous movement rate of the water flow.

[0058] The dynamic weighted directed graph model can be a mathematical model used to represent the pollutant diffusion path.

[0059] Specifically, extract the water pipe step information from GIS data to identify the coordinate set of several water pipe intersection points; then, parse the pipeline attribute database to obtain the pipe diameter and laying slope of each section of the water pipe. Receive multi-source water quality monitoring data from the discharge port sensor group, public pipe network monitoring nodes, and natural water body monitoring points; furthermore, extract the device ID, geographic coordinates, and deployment level information in the metadata of the data packet header; subsequently, match according to the device ID with the preset monitoring point registration table to determine the data source. Then, based on the data source, construct a set of spatial positions of monitoring points according to the longitude and latitude coordinates associated with the device ID; furthermore, screen the water quality monitoring points according to the requirements of spatial coverage density. Subsequently, extract the original data of the flow velocity sensor at each water quality monitoring point, then perform sliding window filtering processing, calculate the hydrodynamic force of the filtered reference flow velocity value and environmental parameters, and introduce a compensation factor for the slope to gravity based on the non-linear influence of the pipe wall friction resistance to dynamically adapt to the flow regime transition caused by the change in pipe diameter, and output the actual flow velocity value. Furthermore, use the water pipe intersection points and water quality monitoring points as graph nodes together, connect adjacent nodes along the water flow direction to form a set of directed edges; determine the basic weight according to the reciprocal of the pipe diameter and the laying slope; furthermore, determine the reference flow velocity based on the median value of the historical flow velocity of the water pipe section; subsequently, generate a dynamic correction factor through the ratio of the real-time flow velocity to the reference flow velocity; then, construct a dynamic weight directed graph model through the basic weight and the dynamic correction factor. Finally, based on the dynamic weight directed graph model, set the starting monitoring point as the current node, its cumulative resistance value is 0, and the cumulative resistance values of other nodes are set to infinity; execute in a loop, follow the water flow direction to extract all downstream adjacent nodes of the current node, calculate the resistance increment from the current node to the adjacent node through the basic weight, dynamic correction factor, and pipe section length. If the current node cumulative resistance + resistance increment < the existing cumulative resistance of the adjacent node, then update the cumulative resistance value of the adjacent node, record the current node as the optimal precursor node, and mark the current node as optimized. Select the node with the smallest cumulative resistance from the unoptimized nodes as the new current node. When all water quality monitoring points have completed the status marking, it is the end of the loop; calculate the minimum resistance path between any two water quality monitoring points in this loop; output a path sequence containing the numbers of the water pipe intersection points passed, the pipe diameters and slope parameters of the water pipe sections, forming a water flow path association.

[0060] Through this solution, analyzing the water pipe layout information to determine the water pipe intersection points, pipe diameters, and laying gradients helps to overcome the influence of ignoring sudden changes in pipe diameters on the step change in flow velocity. Analyzing multi-source water quality monitoring data to determine the data sources helps to eliminate the problem of missing quantification of spatial distribution characteristics. Based on the data sources, determining the water quality monitoring points to block the loopholes of bypassing the monitoring points through hidden pipes enhances the spatial binding force on illegal sewage discharge behaviors and increases the capture probability of abnormal diffusion paths. Analyzing the multi-source water quality monitoring data and the real-time flow velocity of each water quality monitoring point helps to eliminate the problem of inaccurate water flow models in areas with sudden changes in pipe diameters. Based on the water pipe intersection points, pipe diameters, laying gradients, and real-time flow velocity, constructing a dynamic weight directed graph model to quantify the probability of pollutant transmission path selection and replacing the fixed parameter assumptions of the linear model. Based on the dynamic weight directed graph model, determining the water flow path association to reduce the false connection judgment rate and providing a spatial mapping basis for hidden pipe detection.

[0061] In some embodiments, based on the water flow path association, obtaining a historical related pollution data set; analyzing the historical related pollution data set to determine the historical pollution type, initial pollution concentration, and dissipation time; based on the historical pollution type, initial pollution concentration, and dissipation time, establishing a pollution diffusion model; based on the pollution diffusion model, according to the multi-source water quality monitoring data, determining the pollution dissipation path information.

[0062] The historical related pollution data set can be a set of historical pollution event data that has spatial relevance to the currently monitored water flow path.

[0063] The historical pollution type can be the category of historical pollution events.

[0064] The initial pollution concentration can be the measured value of the pollutant concentration when the pollution event is first detected by the sensor group.

[0065] The dissipation time can be the duration required for the pollutant concentration to drop to the initial concentration.

[0066] The pollution diffusion model can be a partial differential equation model that describes the spatial attenuation law of pollutants along the pipe network path.

[0067] Specifically, from the output path sequence of the dynamic weight directed graph model, extract the number, pipe diameter, and layout slope of each water pipe intersection point to form a structured water flow path association table; then, correlate multi-source water quality monitoring data in the time dimension, and extract the COD, BOD, and NH3-N concentration data packets that match the monitoring point numbers in the water flow path association table to construct a spatio-temporally aligned historical pollution dataset. Furthermore, based on the historical relevant pollution dataset, perform K-means clustering analysis on the COD and BOD concentration ratios to generate historical pollution types; subsequently, locate the concentration mutation points of each pollution event in the historical relevant pollution dataset, align the multi-parameter monitoring sequences of historical pollution events in the time dimension, perform noise filtering on the original data to generate a time-concentration normalization curve; set a sliding analysis window forward on the time axis of the pollution event, and set the window length based on the event duration. Calculate the concentration gradient change rate and fluctuation significance in the window in real time. When both the gradient change rate > the preset mutation threshold, the fluctuation significance > 3 times the standard deviation of the historical fluctuation range, and the duration ≥ the minimum effective mutation window are satisfied, take the concentration value corresponding to the center point of the first time window that meets the mutation condition as the initial concentration of the pollution event. At this time, the corresponding timestamp and position coordinates should also be output for subsequent use; furthermore, with the concentration dropping to the initial concentration as the termination condition, combined with the pipe segment flow velocity in the water flow path association table, calculate the dissipation time according to the exponential decay model. Then, call the preset dissipation coefficient according to the historical pollution type; subsequently, use the initial pollution concentration as the initial value of the model and load it into the grid unit of the pipe segment where the pollution source is located; based on the pipe diameter and slope in the water flow path association table, extract the following inherent physical properties of the pipe segment (pipe diameter, laying slope, pipe wall roughness) from the water flow path association table. Convert the slope value into the gravitational potential energy gradient, calculate the cross-sectional characteristic parameters according to the pipe diameter, and calculate the reference flow velocity based on the pipe fluid motion state; correct the dissipation coefficient according to the real-time water temperature and dissipation time; thus construct a pollution diffusion model. Input the multi-source water quality monitoring data into the pollution diffusion model and update the parameters according to the factory discharge port data, public pipe network node data, and natural water body monitoring point data; subsequently, compare the residual between the predicted concentration distribution of the model and the actual monitoring data; furthermore, when the residual exceeds the limit, inversely adjust the real-time flow velocity weight factor in the dynamic weight directed graph model; thus determine the final pollution dissipation path information.

[0068] Through this solution, based on the association of water flow paths, historical relevant pollution datasets are obtained, cross-regional data interference is eliminated, and training samples with regional feature matching are provided for model construction. The historical relevant pollution datasets are analyzed to determine historical pollution types, initial pollution concentrations, and dissipation times, eliminating the problem that the threshold warning mechanism cannot distinguish legal / illegal sewage discharges, improving the investigation efficiency, and reducing the prediction error of the dissipation time. Based on the historical pollution types, initial pollution concentrations, and dissipation times, a pollution diffusion model is established to quantify the non-standard features of the hidden pipe illegal discharge paths, improving the coverage rate of hidden pipe path identification. Based on the pollution diffusion model, according to multi-source water quality monitoring data, the pollution dissipation path information is determined, providing a traceable diffusion path evidence chain to support the missing judicial appraisal requirements for illegal sewage discharges.

[0069] In some embodiments, the regional information is analyzed to determine the factory coordinates; according to the factory coordinates, the physical distances between the water quality monitoring points and each factory are determined; based on the physical distances, a topological relationship matrix between the water quality monitoring points and the factory coordinates is established; according to the topological relationship matrix, the factory distribution characteristics are determined.

[0070] The factory coordinates can be the longitude and latitude values of the geographical location where the sewage discharging unit is located.

[0071] A factory can be a continuous geographical area occupied by an industrial or commercial entity with an independent sewage discharge permit.

[0072] The physical distance can be the straight-line Euclidean distance between the water quality monitoring point and the factory.

[0073] The topological relationship matrix can be a two-dimensional data structure used to represent the spatial association intensity.

[0074] Specifically, based on the regional information, the factory coordinates within the target area are retrieved from the municipal sewage discharge permit database. Then, according to the factory coordinates, the physical distances between each water quality monitoring point and each factory are calculated using the planar Euclidean distance formula. Furthermore, based on the physical distances, a two-dimensional matrix is constructed according to the total number of monitoring points and the number of valid factories; the inverse distance weighting method is used to quantify the spatial association intensity, thereby determining the matrix elements; subsequently, Softmax normalization is performed on each matrix element; thus, the final topological relationship matrix between the water quality monitoring points and the factory coordinates is established. Then, DBSCAN clustering is performed based on the topological relationship matrix to determine the boundary polygon of the factory cluster; finally, according to the number of factories and the cluster area, the spatial density is calculated to determine the factory distribution characteristics.

[0075] Through this solution, analyze the regional information, determine the coordinates of the factory area, and eliminate the spatial misalignment between the coordinates of the factory area and the pipeline network topology map. According to the coordinates of the factory area, determine the physical distance between the water quality monitoring points and each factory area, and reveal the spatial proximity between the water quality monitoring points and the factory area. Based on the physical distance, establish a topological relationship matrix between the water quality monitoring points and the coordinates of the factory area, quantify the downstream correlation relationship between the factory area and the water quality monitoring points, and improve the accuracy of abnormal path recognition. According to the topological relationship matrix, determine the distribution characteristics of the factory area, which helps to indicate the behavior of bypassing the monitoring points by non-standard pipelines.

[0076] In some embodiments, based on the water flow path association and pollution dissipation path information, construct a pollution source tracing probability map model; according to the distribution characteristics of the factory area, determine the spatial correlation degree between each factory area and the pollution dissipation path; obtain the factory area pollution discharge permit information, and according to the factory area pollution discharge permit information and the spatial correlation degree, determine the factory area pollution discharge probability; use the factory area pollution discharge probability as the prior probability and input it into the pollution source tracing probability map model to obtain the output result; calculate the output result through Bayesian inference to determine the posterior probability of each water quality monitoring point; compare the posterior probability with the preset threshold of each water quality monitoring point, and screen out the water quality monitoring points that exceed the preset threshold as candidate illegal pollution discharge points; according to the water flow path association and pollution dissipation path information, conduct path consistency verification on the candidate illegal pollution discharge points. If the verification is passed, determine the candidate illegal pollution discharge points as illegal pollution discharge points.

[0077] The pollution source tracing probability map model can be a directed graph structure used to quantify the probability distribution of the pollution diffusion path. The pollution dissipation path can be the dynamic trajectory information of the pollutant migrating in the pipeline network. The spatial correlation degree can be used to characterize the spatial proximity and topological correlation intensity between the geographical location of the factory area and the pollution dissipation path. The factory area pollution discharge permit information can be the legal pollution discharge qualification data of the factory area. The factory area pollution discharge probability can be a prior probability value generated by comprehensively evaluating the spatial correlation degree and the pollution discharge permit information. The prior probability can be a preset initial probability distribution. The output result can be the intermediate state data generated after the pollution source tracing probability map model is calculated by the Bayesian network. Bayesian inference can be a statistical inference method based on conditional probability and prior probability. The posterior probability can be the updated probability value for each water quality monitoring point to be determined as an illegal pollution discharge associated node. The preset threshold can be a preset dynamic discrimination boundary value. The candidate illegal pollution discharge points can be the set of water quality monitoring points whose posterior probability exceeds the preset threshold.

[0078] Verification passed can be the final confirmation state that the pipe section where the candidate point is located meets the dark pipe feature determination conditions.

[0079] Specifically, based on the water flow path association and pollution dissipation path information, each water quality monitoring point is regarded as an independent node, and the weight of the directed edge is determined according to the product of the water flow transmission time difference and the pollutant concentration gradient between the upstream and downstream monitoring points. Furthermore, according to the independent nodes and the directed edge weights, a pollution source tracing probability graph model is constructed. The specific method for establishing the pollution source tracing probability graph model is as follows: The water quality monitoring points are set as pollution diffusion observation nodes; the factory coordinates are set as potential pollution source nodes, and the directed edge weights are determined based on the water flow transmission time and the pollutant concentration gradient; according to the factory distribution characteristics, the column vector where the target factory is located in the topological relationship matrix is extracted, and the cosine similarity is calculated with the diffusion direction vector of the pollution dissipation path; then, combined with the radial distance from the factory to the center point of the path, the spatial correlation degree between each factory and the pollution dissipation path is calculated according to the formula. The permitted emission amount and the list of pollutant types of each factory are extracted from the pollutant discharge permit information in the environmental supervision department database; furthermore, according to the factory pollutant discharge permit information and the spatial correlation degree, a probability mapping function is established to determine the factory pollutant discharge probability. Subsequently, after normalizing the factory pollutant discharge probability, it is loaded as the prior probability distribution to the corresponding nodes of the pollution source tracing probability graph model. Then, the real-time data of the water quality monitoring points are converted into pollution diffusion matching degree parameters, which are used as the observed variables of the Bayesian network; furthermore, the matching degree between the observed data of each monitoring point and the pollution diffusion path is calculated through Bayesian inference, and the posterior probability is output; the specific process is to align the observed data of the monitoring points with the pollution source tracing probability graph model on the time axis and the spatial grid according to the prior probability, the observed data of each monitoring point, and the pollution source tracing probability graph model, and then construct an evaluation function for the compliance degree of the pollution diffusion path with linear weights through the concentration gradient matching degree, the flow velocity change synchronization rate, and the pollutant attenuation consistency; the concentration gradient matching degree extracts the pollutant concentration sequences of the target water quality monitoring point and its adjacent upstream monitoring points in a continuous time window from the multi-source water quality monitoring data, calculates the concentration gradient at each time point, and then aligns the measured gradient sequence with the predicted gradient sequence of the corresponding path in the pollution source tracing probability graph model. The covariance of the measured gradient sequence and the predicted gradient sequence is divided by the product of the standard deviations of the two to obtain a correlation coefficient value, which ranges from -1 to 1, and the closer it is to 1, the higher the matching degree. The flow velocity change synchronization rate is based on the multi-source water quality monitoring data to obtain the real-time flow velocity sequences of the target monitoring point and the associated pipe segments. At the same time, the sequence of the pollutant concentration change rate is extracted, and the similarity between the flow velocity change sequence and the concentration change rate sequence is calculated through the dynamic time warping algorithm: the two sequences are stretched or compressed on the time axis to minimize the cumulative distance, and the reciprocal of the minimum cumulative distance is normalized to obtain the synchronization rate value, which ranges from 0 to 1, and the higher the value, the stronger the synchronization.For the pollutant attenuation consistency, select the pollutant concentration attenuation curves from multi-source water quality monitoring data. Use the measured data for fitting attenuation to obtain the measured attenuation coefficient. Obtain the predicted attenuation coefficient for the corresponding path from the pollution diffusion model. Divide the absolute difference between the measured attenuation coefficient and the predicted attenuation coefficient by the predicted attenuation coefficient, and then take 1 minus this relative error to obtain the consistency score, which ranges from 0 to 1. The higher the value, the better the consistency. Assign the plant sewage discharge probability as the initial weight to the associated monitoring points, and then calculate the posterior probability using the product of the prior probability, the matching degree score, and the pipe network connection reliability coefficient. Subsequently, compare the posterior probability with the preset threshold for each water quality monitoring point. When the posterior probability value of a water quality monitoring point exceeds the preset threshold of the area, mark the water quality monitoring point that exceeds the preset threshold as a candidate illegal sewage discharge point. Then, based on the spatio-temporal coupling relationship of the water flow path association and the pollution dissipation path information, conduct path consistency verification on the candidate illegal sewage discharge points to determine the illegal sewage discharge points. First, calculate the proportion of the overlapping length of the pipe segment where the candidate illegal sewage discharge point is located and the center line of the pollution dissipation path. Second, extract the coordinate set of several plants upstream of the candidate illegal sewage discharge point and verify the angle between the pollution diffusion direction and the pipe network water flow direction. Then, compare the fitting degree of the measured pollutant concentration attenuation curve of the candidate illegal sewage discharge point and the model prediction curve. Finally, based on the proportion of the overlapping length, the angle, and the fitting degree, determine the candidate illegal sewage discharge point as an illegal sewage discharge point.

[0080] Through this solution, based on the water flow path association and the pollution dissipation path information, construct a pollution source tracing probability map model, which helps to eliminate the problem of inconsistent spatio-temporal benchmarks of multi-source data. According to the plant distribution characteristics, determine the spatial correlation degree between each plant and the pollution dissipation path, and enhance the detection sensitivity of the bypass path of the hidden pipe. Obtain the plant sewage discharge permit information, and according to the plant sewage discharge permit information and the spatial correlation degree, determine the plant sewage discharge probability to avoid probability distortion caused by inconsistent multi-source data benchmarks. Use the plant sewage discharge probability as the prior probability and input it into the pollution source tracing probability map model to obtain the output result, realizing the spatio-temporal alignment of the pollutant diffusion path and the monitoring data. Calculate the output result through Bayesian inference to determine the posterior probability of each water quality monitoring point and avoid probability oscillation caused by the pipe network loop. Compare the posterior probability with the preset threshold of each water quality monitoring point, and screen the water quality monitoring points that exceed the preset threshold as candidate illegal sewage discharge points, which helps to eliminate the false negative / false positive problems caused by the fixed threshold. According to the water flow path association and the pollution dissipation path information, conduct path consistency verification on the candidate illegal sewage discharge points. If the verification passes, determine the candidate illegal sewage discharge point as an illegal sewage discharge point and exclude the misjudgment caused by the pipe network backflow.

[0081] In some embodiments, obtain the layout diagram of the factory area buildings and the topological diagram of the municipal pipe network; analyze the topological diagram of the municipal pipe network and the layout diagram of the factory area buildings to determine the public pipe network structure and the factory area pipe network structure; determine the spatial position relationship between the factory area and the public pipe network according to the public pipe network structure and the factory area pipe network structure; determine the potential hidden pipe laying path according to the spatial position relationship; verify the possibility of the existence of the hidden pipe according to the change of the pollutant concentration gradient on the potential hidden pipe laying path based on the pollution dissipation path information; based on the possibility of the existence of the hidden pipe, determine the hidden pipe sewage discharge point according to the potential hidden pipe laying path.

[0082] The layout diagram of the factory area buildings can be a vector drawing containing the contour coordinates of several buildings in the industrial factory area, the internal pipe network interface positions and elevation information. The topological diagram of the municipal pipe network can be a digital map representing the spatial connection relationship of each pipe section in the urban public drainage pipe network. The public pipe network structure can be a network entity composed of several pipe sections and pipe section connection nodes. The factory area pipe network structure can be the physical configuration of the internal drainage pipes in the factory area. The public pipe network can be the urban public drainage pipe data. The spatial position relationship can be the relative spatial distance and azimuth relationship. The potential hidden pipe laying path can be an illegal underground connection channel existing between the factory area and the public pipe network. The change of the pollutant concentration gradient can be the concentration change rate per unit distance along the pipe extension direction. The possibility of the existence of the hidden pipe can be a probability value quantified by the deviation between the measured gradient and the theoretical gradient. The hidden pipe sewage discharge point can be an illegal discharge access position determined by the path tracing algorithm.

[0083] Specifically, retrieve the factory building layout diagram and the municipal pipe network topology diagram of the target area from the database of the environmental supervision department. Then, based on the municipal pipe network topology diagram, analyze the main pipe section directions, pipe diameter mutation points, slope turning points, and monitoring point coverage blind areas, and extract the public pipe network structure; simultaneously, identify the coordinates of the access points of the internal drainage pipe network in the factory area according to the building layout diagram, mark the legal connection ports with the public pipe network, and determine the factory area pipe network structure. Furthermore, analyze the set of main pipe sections in the public pipe network structure and the set of access points in the factory area pipe network structure, and establish the spatial position relationship between the factory area and the public pipe network according to the Euclidean distance between the factory area boundary and the pipe network nodes. Subsequently, screen the factory-pipe network node pairs based on the spatial position relationship, and generate potential hidden pipe laying paths in combination with the hidden pipe laying characteristics in the historical hidden pipe case library. Then, perform dynamic verification for each potential hidden pipe laying path: First, insert virtual monitoring points at equal intervals along the potential hidden pipe laying path, and calculate the change in the pollutant concentration gradient at each virtual monitoring point; then, perform DTW dynamic time warping matching on the change in the pollutant concentration gradient and the measured data of the candidate illegal sewage discharge points to verify the possibility of the existence of the hidden pipe. Finally, if the matching degree is too high, determine the possibility of the existence of the hidden pipe; furthermore, perform end-point backtracking on the potential hidden pipe laying paths that pass the verification to determine the hidden pipe sewage discharge points: First, take a set of several factory coordinate points within the upstream range of the path end; then, calculate the perpendicular distance from each set of factory coordinate points to the center line of the potential hidden pipe laying path; determine the hidden pipe access point according to the perpendicular distance; finally, output the set of hidden pipe access point coordinates as the hidden pipe sewage discharge points.

[0084] Through this solution, obtain the factory building layout diagram and the municipal pipe network topology diagram, eliminate the coordinate drift error during manual drawing comparison, and provide an accurate spatial reference for topological analysis. Analyze the municipal pipe network topology diagram and the factory building layout diagram to determine the public pipe network structure and the factory area pipe network structure, and overcome the problem of being unable to perceive the real-time connection status of pipe sections relying on static drawings. Determine the spatial position relationship between the factory area and the public pipe network according to the public pipe network structure and the factory area pipe network structure, replace the manual drawing comparison method, and improve the processing efficiency. Determine the potential hidden pipe laying paths according to the spatial position relationship, which helps to accurately capture the typical characteristics of hidden pipe projects. Verify the possibility of the existence of the hidden pipe according to the change in the pollutant concentration gradient on the potential hidden pipe laying path based on the pollution dissipation path information, which helps to reduce the path misjudgment rate. Based on the possibility of the existence of the hidden pipe, determine the hidden pipe sewage discharge points according to the potential hidden pipe laying path, respond to the accuracy defect of insufficient resolution of the municipal pipe network diagram, and ensure that the positioning result meets the engineering acceptance standard.

[0085] In some embodiments, the public pipe network structure is parsed to determine the public pipe network nodes; the plant pipe network structure is parsed to determine the spatial coordinates of the plant drainage facilities; the Euclidean distance between each spatial coordinate and the nearest public pipe network node is calculated to obtain a distance set; the plant information is obtained, the plant information is analyzed to determine the plant drainage requirements; a path optimization function is constructed based on the distance set and the plant drainage requirements; and a potential buried pipe laying path is calculated based on the path optimization function.

[0086] The public pipe network nodes can be the physical connection points between pipe segments in the municipal drainage system. The plant drainage facilities can be the physical facilities directly involved in sewage discharge within the plant. The spatial coordinates can be the three-dimensional position data of the plant drainage facilities. The Euclidean distance can be the straight-line distance between two coordinate points in three-dimensional space. The distance set can be a numerical set composed of the Euclidean distances between each spatial coordinate and the nearest public pipe network node. The plant information can be the static drainage-related data recorded in the enterprise environmental assessment report. The plant drainage requirements can be the dynamic drainage requirements quantified by analyzing the plant information. The path optimization function can be a multi-constraint mathematical model aiming to minimize the laying path cost.

[0087] Specifically, public pipe network nodes such as the intersections of main pipes, branch access points, and the center points of inspection wells are extracted from the municipal pipe network topology map. Then, the planar coordinates of plant drainage facilities such as drainage outlets, the outlets of pretreatment ponds, and emergency discharge valves are identified from the plant pipe network structure and converted into spatial coordinates consistent with the public pipe network nodes. Furthermore, a k-d tree spatial index query is performed on the spatial coordinates of each plant drainage facility to find the nearby public pipe network nodes, record the Euclidean distance between each spatial coordinate and the nearest public pipe network node, and form a distance set. Subsequently, the plant information is obtained from the enterprise environmental assessment report; furthermore, the plant information is analyzed to count the daily discharge volume, peak flow rate, and pollutant concentration of the plant's production wastewater, thereby generating the plant drainage requirements. A path optimization function is constructed based on the distance set and the dynamic flow parameters of the plant drainage requirements. Finally, constraint conditions are set according to the path optimization function, a candidate path set is generated through the topological relationship of the public pipe network nodes, and the candidate path set is hierarchically filtered; thereby determining the potential buried pipe laying path.

[0088] Through this solution, the public pipe network structure is analyzed to determine the nodes of the public pipe network, eliminating the problem of blurred node positioning caused by insufficient drawing resolution and providing an accurate set of reference coordinates for spatial relationship calculation. The factory area pipe network structure is analyzed to determine the spatial coordinates of the drainage facilities in the factory area, achieving the alignment of the local coordinate system of the factory area with the global coordinate system of the public pipe network, eliminating the calculation error of spatial relationships caused by coordinate differences, and ensuring the geometric consistency of the distance calculation between the drainage outlets in the factory area and the pipe network nodes. The Euclidean distance between each spatial coordinate and the nearest public pipe network node is calculated to obtain a distance set, quantifying the spatial accessibility relationship between the drainage points in the factory area and the public pipe network, overcoming the defect of low efficiency in manually comparing pipe network drawings, and providing key distance parameter inputs for the path optimization function. The factory area information is obtained, analyzed, and the drainage requirements of the factory area are determined, eliminating the problem of model distortion caused by only using fixed thresholds. Based on the distance set and the drainage requirements of the factory area, a path optimization function is constructed to effectively filter out invalid paths that do not conform to the law of buried pipe laying. According to the path optimization function, the potential buried pipe laying paths are calculated, realizing the intelligent convergence from a large number of possible paths to highly probable buried pipe paths and improving the path generation efficiency.

[0089] In some embodiments, historical cases of detected buried pipes are obtained, analyzed, and a path feature data set is determined; the path feature data set is parsed to determine the path length, burial depth, and number of turning angles in each historical case of detected buried pipes; a logistic regression model is constructed based on the path length, burial depth, and number of turning angles; the probability of the existence of buried pipes is input into the logistic regression model to determine whether the confidence level of at least one potential buried pipe laying path is higher than the confidence requirement; if so, the potential buried pipe laying path corresponding to the confidence level higher than the confidence requirement is determined as the buried pipe sewage discharge point.

[0090] The historical cases of detected buried pipes can be illegal buried pipe laying instances verified and recorded by the environmental law enforcement department.

[0091] The path feature data set can be a structured data set containing the path length, burial depth, and number of turning angles.

[0092] The path length can be the total length of the broken line path from the starting point to the ending point of the buried pipe. The burial depth can be the vertical distance from the top of the buried pipe to the ground surface. The number of turning angles can be the number of turning points with angle changes in the planar orientation of the buried pipe. The logistic regression model can be a binary classification model with the path length, burial depth, and number of turning angles as independent variables and the probability of the existence of buried pipes as the dependent variable. The confidence level can be the probability value of the existence of buried pipes output by the model. The confidence requirement can be a preset determination threshold.

[0093] Specifically, historical cases of detected concealed pipelines are retrieved from the environmental law enforcement database. The total length of the continuous pipeline from the starting point to the ending point of the concealed pipeline is calculated through the GIS topological analysis tool; the measured values of the soil cover depth at the top of the pipeline in the construction records are extracted, and when the data is missing, it is inverted through the geological radar detection report; the number of turning points with angle changes in the planar alignment of the pipeline is counted. These data together form the path feature dataset.

[0094] Among them, the total length of the continuous pipeline from the starting point to the ending point of the concealed pipeline is the path length, the measured value of the soil cover depth at the top of the pipeline is the burial depth, and the number of turning points with angle changes in the planar alignment of the pipeline is the number of turning angles.

[0095] Construct the path length, burial depth, and number of turning angles of each historical detected hidden pipe case as an input vector, and output a label to indicate whether there is a hidden pipe in the case; use the maximum likelihood estimation method to construct a logistic regression model. The construction process is as follows: Retrieve historical detected hidden pipe cases from the database of the environmental law enforcement department. These cases contain proven illegal hidden pipe laying instances, and each case records detailed information on the hidden pipe laying path (such as construction drawings, ground penetrating radar detection reports, on-site measurement data), which is defined as the total actual pipe length from the starting point of the hidden pipe (factory drainage facility) to the ending point (public pipe network access point). Calculation method: Through a GIS topological analysis tool (such as the path length calculation function of ArcGIS), import the vector coordinate sequence of the hidden pipe path (starting point coordinates, ending point coordinates, and intermediate turning point coordinates), and apply the Euclidean distance formula to calculate the distances between adjacent points and then accumulate to obtain the total length (unit: meters). It is defined as the vertical distance from the top of the hidden pipe to the ground surface (unit: meters). Acquisition method: First, extract the measured value from the construction record or on-site measurement report; if the data is missing, invert it based on the ground penetrating radar detection report (such as calculating the depth through the time difference of electromagnetic wave reflection). It is defined as the number of turning points with an angular change (≥30 degrees) in the planar orientation of the hidden pipe. Statistical method: Identify the direction mutation points from the path coordinate sequence (such as calculating the included angle through the dot product of adjacent line segment vectors, and if the included angle < 150 degrees, it is counted as one turning angle). Take the path length, burial depth, and number of turning angles of each case as the feature vector, and add a binary classification label (output variable): The label "1" indicates the existence of a hidden pipe (positive case), and the label "0" indicates the non-existence of a hidden pipe (negative case, such as a legal shallow buried pipe or a false alarm case); to address the feature scale difference (the path length has a large range and the burial depth has a small range), use Z-score standardization to process each feature so that the mean is 0 and the variance is 1. Combine the standardized features into an input vector X. Each case corresponds to a feature vector: X = [standardized path length, standardized burial depth, standardized number of turning angles], and the output label y is directly in binary form (0 or 1); the probability of the existence of a hidden pipe (obtained from the previous steps) is used as the model input. The probability of the existence of a hidden pipe is a probability value (range 0 - 1), which is calculated based on the pollution dissipation path information (such as verifying through the change of pollutant concentration gradient). Standardize the currently detected potential hidden pipe laying path; subsequently, input the standardized feature vector into the trained logistic regression model to output the probability of the existence of a hidden pipe; then, based on the ROC curve analysis, determine the confidence requirement by maximizing the difference between the true positive rate and the false positive rate, so as to determine whether the confidence level of at least one potential hidden pipe laying path is higher than the confidence requirement. Finally, screen the potential hidden pipe laying paths with a probability of the existence of a hidden pipe higher than the confidence requirement, sort them in descending order of confidence level, overlay the high-confidence paths with the factory geographical information, and mark the illegal discharge port positions of the potential hidden pipe laying paths as hidden pipe pollution discharge points.

[0096] Through this solution, obtaining historical cases of detected concealed pipes, analyzing the historical cases of detected concealed pipes, and determining the path feature dataset can help eliminate the defect of relying on manual inspection of pipe network drawings. Parsing the path feature dataset to determine the path length, burial depth, and number of turning angles in each historical case of detected concealed pipes can enhance the detection sensitivity for behaviors that bypass monitoring points, filter out interference from legal shallow-buried pipes, and improve the path matching degree under complex pipe network topologies. Based on the path length, burial depth, and number of turning angles, a logistic regression model is constructed to eliminate the problem of insufficient accuracy of fixed-parameter models. Inputting the possibility of the existence of concealed pipes into the logistic regression model to determine whether the confidence level of the existence of at least one potential concealed pipe laying path is higher than the confidence requirement can help eliminate false judgments caused by water flow disturbances. If so, the potential concealed pipe laying path corresponding to the confidence level higher than the confidence requirement is determined as the concealed pipe sewage discharge point, which can help eliminate the problem of low efficiency of manual inspection.

[0097] In some embodiments, according to the plant coordinates, the coordinates of the pollution source and the administrative region to which the sewage discharge point belongs are determined; according to the pollution dissipation path information, the pollution diffusion range is determined; after the position of the sewage discharge point is determined, an early warning message is generated based on the coordinates of the pollution source and the pollution diffusion range; according to the administrative region to which the sewage discharge point belongs, the early warning message is routed to the corresponding supervision terminal, and the response timestamp of the supervision terminal is recorded to form a closed-loop management log.

[0098] The coordinates of the pollution source can be the coordinates for tracing and positioning pollutants. The administrative region to which the sewage discharge point belongs can be the administrative division identification to which the sewage discharge point belongs. The pollution diffusion range can be the spatial distribution area of pollutants. The position of the sewage discharge point can be the location of the illegal sewage discharge source. The early warning message can be a structured data message. The routing of the early warning message can be a directional distribution mechanism for the early warning message. The supervision terminal can be an authorized access device with environmental supervision functions. The response timestamp can be the UTC time mark generated when the supervision terminal returns a disposal instruction. The closed-loop management log can be an immutable record set stored in the blockchain.

[0099] Specifically, the National Geographic Coding Service API is called to convert the factory coordinates into WGS84 geographic coordinates, and the coordinates of the pollution point are calculated through the spatial interpolation algorithm; then, based on the "Administrative Division Code" database, the spatial inclusion judgment function ST_Contains is used to overlay the pollution point coordinates with the preset administrative district vector boundary to determine the administrative area to which the pollution point belongs. Then, according to the water flow velocity, pipeline curvature radius and pollutant half-life in the pollution dissipation path information, a dynamic parameter matrix is constructed, and the finite volume method is used to simulate the two-dimensional pollutant diffusion; then, the iteration termination condition is set to the concentration gradient change rate, and the pollution diffusion contour map is output, and the geometric envelope of the concentration exceeding the standard area is extracted as the pollution diffusion range. After the location of the pollution discharge point is determined, the pollution point coordinates are converted into a standard address description (administrative district + road + azimuth distance), and the boundary coordinates of the area with excessively high concentration in the pollution diffusion range are extracted; thus generating early warning information. According to the administrative region to which the pollution discharge point belongs, the routing table of the environmental regulatory agency is queried to match the ID of the regulatory terminal with jurisdiction; then, the warning information is pushed to the corresponding regulatory terminal through the message queue middleware, and a digital signature and time validity identifier are attached; then, when capturing the response instruction returned by the regulatory terminal, the response timestamp and the pollution point coordinates are extracted to build an associated index; finally, the warning information, routing path, and response delay data are written into the blockchain evidence storage module to generate an unalterable closed-loop management log.

[0100] Through this solution, the coordinates of the pollution point and the administrative area to which the discharge point belongs are determined based on the coordinates of the plant area, eliminating the positioning deviation caused by inconsistent coordinate benchmarks and resolving jurisdictional disputes caused by blurred boundaries when manually comparing drawings. According to the pollution dissipation path information, the pollution diffusion range is determined to overcome the impact of inconsistent time and space benchmarks and improve the accuracy of abnormal path identification. After the location of the discharge point is determined, early warning information is generated based on the coordinates of the pollution point and the pollution diffusion range to eliminate the problem of low efficiency of manual investigation, make the early warning location readable, and facilitate rapid on-site positioning. According to the administrative area to which the discharge point belongs, the early warning information is routed to the corresponding supervision terminal, and the response timestamp of the supervision terminal is recorded to form a closed-loop management log, eliminating the defects of easy tampering of paper logs and reversible editing of electronic logs.

[0101] Figure 3 A schematic diagram of a water quality monitoring system based on Internet of Things technology is provided in one embodiment of the present application. Figure 3 As shown, the water quality monitoring system 300 based on the Internet of Things technology of this embodiment includes: a data analysis module 301, an information analysis module 302, a dissipation analysis module 303, a feature determination module 304, and a pollution determination module 305.

[0102] The data analysis module 301 is used to obtain multi-source water quality monitoring data of the monitored area; analyze the multi-source water quality monitoring data to determine whether there is a possibility of illegal sewage discharge; The information analysis module 302 is used to, if so, obtain the water pipe layout information and area information of the monitored area; analyze the water pipe layout information to determine the water flow path association; The dissipation analysis module 303 is used to determine the pollution dissipation path information based on the water flow path association and according to the multi-source water quality monitoring data; The feature determination module 304 is used to analyze the area information to determine the factory area distribution characteristics; The sewage discharge determination module 305 is used to determine the illegal sewage discharge point according to the water flow path association, the factory area distribution characteristics and the dissipation path information.

[0103] Optionally, when the information analysis module 302 analyzes the water pipe layout information to determine the water flow path association, it is used to: analyze the water pipe layout information to determine the water pipe intersection points, pipe diameters and laying gradients; parse the multi-source water quality monitoring data to determine the data sources; determine the water quality monitoring points according to the data sources; analyze the real-time flow velocities of each water quality monitoring point in the multi-source water quality monitoring data; construct a dynamic weighted directed graph model according to the water pipe intersection points, the pipe diameters, the laying gradients and the real-time flow velocities; determine the water flow path association according to the dynamic weighted directed graph model.

[0104] Optionally, when the dissipation analysis module 303 determines the pollution dissipation path information based on the water flow path association and according to the multi-source water quality monitoring data, it is used to: obtain the historical relevant pollution data set based on the water flow path association; parse the historical relevant pollution data set to determine the historical pollution type, initial pollution concentration and dissipation time; establish a pollution diffusion model according to the historical pollution type, the initial pollution concentration and the dissipation time; determine the pollution dissipation path information based on the pollution diffusion model and according to the multi-source water quality monitoring data.

[0105] Optionally, when the feature determination module 304 analyzes the area information to determine the factory area distribution characteristics, it is used to: analyze the area information to determine the factory area coordinates; determine the physical distance between each water quality monitoring point and each factory area according to the factory area coordinates; establish a topological relationship matrix between the water quality monitoring points and the factory area coordinates based on the physical distance; determine the factory area distribution characteristics according to the topological relationship matrix.

[0106] Optionally, when determining an illegal sewage discharge point according to the water flow path association, the plant distribution characteristics, and the dissipation path information, the sewage discharge determination module 305 is configured to: construct a pollution source tracing probability graph model based on the water flow path association and the pollution dissipation path information; determine the spatial association degree between each plant and the pollution dissipation path according to the plant distribution characteristics; obtain the plant sewage discharge permit information, and determine the plant sewage discharge probability according to the plant sewage discharge permit information and the spatial association degree; use the plant sewage discharge probability as a prior probability and input it into the pollution source tracing probability graph model to obtain an output result; calculate the output result through Bayesian inference to determine the posterior probability of each water quality monitoring point; compare the posterior probability with the preset threshold of each water quality monitoring point, and select the water quality monitoring points that exceed the preset threshold as candidate illegal sewage discharge points; verify the path consistency of the candidate illegal sewage discharge points according to the water flow path association and the pollution dissipation path information, and if the verification is passed, determine the candidate illegal sewage discharge point as an illegal sewage discharge point.

[0107] Optionally, the water quality monitoring system based on the Internet of Things technology further includes a hidden pipe determination module 306, which is configured to: obtain the plant building layout diagram and the municipal pipe network topology diagram; analyze the municipal pipe network topology diagram and the plant building layout diagram to determine the public pipe network structure and the plant pipe network structure; determine the spatial position relationship between the plant and the public pipe network according to the public pipe network structure and the plant pipe network structure; determine the potential hidden pipe laying path according to the spatial position relationship; verify the possibility of the existence of a hidden pipe according to the change of the pollutant concentration gradient on the potential hidden pipe laying path in the pollution dissipation path information; determine the hidden pipe sewage discharge point based on the possibility of the existence of the hidden pipe and according to the potential hidden pipe laying path.

[0108] Optionally, when determining the potential hidden pipe laying path according to the spatial position relationship, the hidden pipe determination module 306 is configured to: analyze the public pipe network structure to determine the public pipe network nodes; analyze the plant pipe network structure to determine the spatial coordinates of the plant drainage facilities; calculate the Euclidean distance between each spatial coordinate and the nearest public pipe network node to obtain a distance set; obtain the plant information, analyze the plant information, and determine the plant drainage demand; construct a path optimization function according to the distance set and the plant drainage demand; calculate and obtain the potential hidden pipe laying path according to the path optimization function.

[0109] Optionally, when determining the dark pipe sewage discharge point based on the dark pipe existence possibility and according to the potential dark pipe laying path, the dark pipe determination module 306 is configured to: obtain historical dark pipe seizure cases, analyze the historical dark pipe seizure cases, and determine a path feature data set; parse the path feature data set to determine the path length, burial depth, and number of turning times in each historical dark pipe seizure case; construct a logistic regression model based on the path length, the burial depth, and the number of turning times; input the dark pipe existence possibility into the logistic regression model to determine whether the confidence level of at least one potential dark pipe laying path is higher than the confidence requirement; if so, determine the potential dark pipe laying path corresponding to the confidence level higher than the confidence requirement as the dark pipe sewage discharge point.

[0110] Optionally, the water quality monitoring system based on the Internet of Things technology further includes a log formation module 307, configured to: determine the pollution point coordinates and the administrative region to which the sewage discharge point belongs according to the factory area coordinates; determine the pollution diffusion range according to the pollution dissipation path information; when the position of the sewage discharge point is determined, generate a warning message according to the pollution point coordinates and the pollution diffusion range; route the warning message to the corresponding supervision terminal according to the administrative region to which the sewage discharge point belongs, and record the response timestamp of the supervision terminal to form a closed-loop management log.

[0111] The system of this embodiment can be used to execute the method of any of the above embodiments, and its implementation principle and technical effects are similar, which will not be elaborated here.

Claims

1. A water quality monitoring method based on Internet of Things technology, characterized in that, Including: Obtain multi-source water quality monitoring data of the monitoring area; Analyze the multi-source water quality monitoring data to determine whether there is a possibility of illegal sewage discharge; If so, obtain the water pipe layout information and area information of the monitoring area; analyze the water pipe layout information to determine the water flow path association; Based on the water flow path association, determine the pollution dissipation path information according to the multi-source water quality monitoring data; Analyze the area information to determine the factory distribution characteristics; Determine the illegal sewage discharge points according to the water flow path association, the factory distribution characteristics and the dissipation path information.

2. The method according to claim 1, wherein The analyzing the water pipe layout information to determine the water flow path association includes: Analyze the water pipe layout information to determine the water pipe intersection points, pipe diameters and laying gradients; Parse the multi-source water quality monitoring data to determine the data sources; Determine the water quality monitoring points according to the data sources; Analyze the real-time flow velocity of each water quality monitoring point in the multi-source water quality monitoring data; Construct a dynamic weighted directed graph model according to the water pipe intersection points, the pipe diameters, the laying gradients and the real-time flow velocity; Determine the water flow path association according to the dynamic weighted directed graph model.

3. The method according to claim 1, wherein The determining the pollution dissipation path information based on the water flow path association and according to the multi-source water quality monitoring data includes: Based on the water flow path association, obtain the historical relevant pollution data set; parse the historical relevant pollution data set to determine the historical pollution type, initial pollution concentration and dissipation time; Establish a pollution diffusion model according to the historical pollution type, the initial pollution concentration and the dissipation time; Based on the pollution diffusion model, determine the pollution dissipation path information according to the multi-source water quality monitoring data.

4. The method according to claim 2, wherein The analyzing the area information to determine the factory distribution characteristics includes: Analyze the area information to determine the factory coordinates; Determine the physical distance between each water quality monitoring point and each factory according to the factory coordinates; Based on the physical distance, establish a topological relationship matrix between the water quality monitoring points and the factory coordinates; Determine the factory distribution characteristics according to the topological relationship matrix.

5. The method according to claim 1, characterized in that, The determining the illegal sewage discharge points according to the water flow path association, the factory distribution characteristics and the dissipation path information includes: Construct a pollution source tracing probability graph model based on the water flow path association and the pollution dissipation path information; Determine the spatial association degree between each factory and the pollution dissipation path according to the factory distribution characteristics; Obtain the factory sewage discharge permit information, and determine the factory sewage discharge probability according to the factory sewage discharge permit information and the spatial association degree; Take the factory sewage discharge probability as the prior probability and input it into the pollution source tracing probability graph model to obtain the output result; Calculate the output result through Bayesian inference to determine the posterior probability of each water quality monitoring point; Compare the posterior probability with the preset threshold of each water quality monitoring point, and screen out the water quality monitoring points that exceed the preset threshold as candidate illegal sewage discharge points; According to the water flow path association and the pollution dissipation path information, conduct path consistency verification on the candidate illegal sewage discharge points. If the verification passes, determine the candidate illegal sewage discharge points as illegal sewage discharge points.

6. The method according to claim 5, wherein After determining that the candidate illegal sewage discharge point is an illegal sewage discharge point if the verification is passed, it further includes: Obtain the factory building layout diagram and the municipal pipe network topology diagram; Analyze the municipal pipe network topology diagram and the factory building layout diagram to determine the public pipe network structure and the factory pipe network structure; According to the public pipe network structure and the factory pipe network structure, determine the spatial position relationship between the factory and the public pipe network; According to the spatial position relationship, determine the potential hidden pipe laying path; Verify the possibility of the existence of a hidden pipe according to the change of the pollutant concentration gradient on the potential hidden pipe laying path based on the pollution dissipation path information; Based on the possibility of the existence of a hidden pipe, determine the hidden pipe sewage discharge point according to the potential hidden pipe laying path.

7. The method according to claim 6, wherein The determining the potential hidden pipe laying path according to the spatial position relationship includes: Analyze the public pipe network structure to determine the public pipe network nodes; Analyze the factory pipe network structure to determine the spatial coordinates of the factory drainage facilities; Calculate the Euclidean distance between each spatial coordinate and the nearest public pipe network node to obtain a distance set; Obtain the factory information, analyze the factory information, and determine the factory drainage demand; Construct a path optimization function according to the distance set and the factory drainage demand; Calculate the potential hidden pipe laying path according to the path optimization function.

8. The method according to claim 6, wherein The determining the hidden pipe sewage discharge point based on the possibility of the existence of a hidden pipe and according to the potential hidden pipe laying path includes: Obtain historical detected hidden pipe cases, analyze the historical detected hidden pipe cases, and determine a path feature data set; Analyze the path feature data set to determine the path length, burial depth, and number of turning angles in each historical detected hidden pipe case; Construct a logistic regression model according to the path length, the burial depth, and the number of turning angles; Input the possibility of the existence of a hidden pipe into the logistic regression model to determine whether the confidence level of at least one potential hidden pipe laying path is higher than the confidence requirement; If so, determine the potential hidden pipe laying path corresponding to the confidence level higher than the confidence requirement as the hidden pipe sewage discharge point.

9. The method according to claim 4, characterized in that The method further includes: According to the factory coordinates, determine the pollution point coordinates and the administrative region to which the sewage discharge point belongs; According to the pollution dissipation path information, determine the pollution diffusion range; After determining the position of the sewage discharge point, generate a warning message according to the pollution point coordinates and the pollution diffusion range; Route the warning message to the corresponding supervision terminal according to the administrative region to which the sewage discharge point belongs, and record the response timestamp of the supervision terminal to form a closed-loop management log.

10. A water quality monitoring system based on Internet of Things technology, applied to the method described in any one of claims 1-9, characterized in that, It includes: A data analysis module for obtaining multi-source water quality monitoring data of the monitoring area; Analyze the multi-source water quality monitoring data to determine whether there is a possibility of illegal sewage discharge; An information analysis module for, if so, obtaining the water pipe layout information and regional information of the monitoring area; analyzing the water pipe layout information to determine the water flow path association; A dissipation analysis module for determining the pollution dissipation path information based on the water flow path association and according to the multi-source water quality monitoring data; A feature determination module for analyzing the regional information to determine the factory distribution characteristics; A sewage discharge determination module, configured to determine illegal sewage discharge points based on the water flow path association, the plant area distribution characteristics, and the dissipation path information.

Images