A method for tracing abnormal pollution at surface water discharge outlets into rivers based on smart IoT

Abstract

The present invention discloses a method and device for tracing the source of abnormal pollution of surface water discharge outlets into rivers based on smart Internet of Things. The method includes: constructing a database according to basic data of surface water pollution; dividing all pollution sources in the database into different levels of pipeline logical relationships according to the land area and pipe network, and the upstream and downstream connection relationship of the water system, and constructing a pollution source-land plot-pipe network-monitoring station smart Internet of Things grid; pre-processing and feature analysis of all data in the smart Internet of Things grid to construct a model database; and building a tracing model and a pollution contribution model based on the model database, the tracing model is used to calculate the pollution contribution rate of each pollution source, and determine the target discharge outlet or target sewage discharge node according to the pollution contribution rate, and the pollution contribution model is used to generate and display the pollution contribution value of the target discharge outlet or target sewage discharge point in response to an input instruction. The problem of being unable to quickly and accurately determine the pollution source in surface water environmental management is solved.

Description

Technical Field

[0001] The present invention relates to the field of environmental protection, and in particular to a method for tracing the source of abnormal pollution at a surface water discharge outlet into a river based on smart Internet of Things. Background Art

[0002] Surface water is one of the most important natural resources on earth. It not only provides drinking water and agricultural irrigation water for human beings, but also supports the stable operation of the ecosystem. Surface water monitoring is the main means to evaluate the water quality and quantity of natural water bodies. By monitoring the chemical, physical and biological properties of surface water bodies, the health of water bodies can be determined. Through surface water monitoring, potential sources of pollution can be identified, which helps to take necessary measures to protect water sources. With the continuous development of monitoring technology, the accuracy and timeliness of surface water monitoring have been significantly improved. However, due to the diversity and complexity of surface water pollution sources, environmental monitoring still lags behind the emergence of environmental pollution. Some pollution is difficult to trace the source of pollution. When paying attention to the excessive water quality of river sections, how to quickly identify the source, time, intensity and scope of pollution, prevent the spread and deterioration of pollution, and provide a basis for pollution control and pollution source tracing has become a problem that technicians in this field need to solve urgently.

[0003] Traditional water ecological environment monitoring and tracing investigation is divided into off-site tracing and on-site tracing. Off-site tracing is mainly a pollution tracing method for outlets with clear discharge destinations and clear sewage channels to determine the source of sewage and the main body of the sewage discharge through data review, water balance analysis, etc., and requires precise data such as historical data of the outlet, spatial vector data, pipe network distribution, and land use data before analysis and tracing; on-site tracing is a tracing method for determining the source of sewage at the outlet and the main body of the sewage discharge through on-site surveys or the use of equipment technology and other scientific and technological means. It requires a certain amount of manpower and material resources, and it is still difficult to achieve accurate tracing for outlets with complex layouts. In actual tracing work, for example, the tracing of sewage outlets mainly relies on manual on-site inspections, which is limited by low efficiency and the inability to detect hidden drainage outlets; for example, most regional water environment monitoring surveys are limited by hardware equipment construction configuration and high operation and maintenance costs, and can only be sporadically deployed at key points. The regional water environment survey is mainly in the form of water quality monitoring stations with manual sampling and testing. There are difficulties such as insufficient station distribution density and limited personnel, and it is impossible to identify pollution sources in time, and it is impossible to present the spatiotemporal state of the water quality of the measured water body, and it is impossible to achieve efficient, large-scale, spatial dimension monitoring and evaluation and accurate tracing of pollution sources. With the development of science and technology, digital technologies such as artificial intelligence, big data, and the Internet of Things have begun to be gradually applied to ecological environment governance monitoring. The existing technical means of monitoring and alarm are limited to early warning alarms at a single site, but it is still impossible to achieve intelligent Internet of Things for surface water monitoring. There is a lack of intelligent models for effective prediction of abnormal sources, and it is difficult to lock the source of pollution.

[0004] To this end, the present invention proposes a method for tracing the abnormal pollution of surface water discharge outlets into rivers based on smart Internet of Things, aiming to solve the problem of being unable to quickly and accurately determine the source of pollution in surface water environmental management. Summary of the invention

[0005] To this end, an embodiment of the present invention provides a method for tracing the abnormal pollution of surface water discharge outlets into rivers based on a smart Internet of Things, in order to achieve accurate and rapid tracing of the pollution source.

[0006] In order to achieve the above purpose, the embodiment of the present invention provides the following technical solutions:

[0007] The present invention provides a method for tracing the source of abnormal pollution at a surface water discharge outlet into a river based on smart IoT, the method comprising:

[0008] Obtaining basic data on surface water pollution in the target monitoring area, and building a database based on the basic data on surface water pollution;

[0009] All pollution sources in the database are divided into different levels of pipeline logic relationships according to the land area and the upstream and downstream connection relationship of the pipeline network and water system, and a pollution source-land area-pipeline network-monitoring station smart Internet of Things grid is constructed;

[0010] Preprocess and feature analyze all data in the pollution source-land-pipeline network-monitoring station smart Internet of Things grid to build a model database;

[0011] Based on the model database, a source tracing model and a pollution contribution model are respectively constructed. The source tracing model is used to calculate the pollution contribution rate of each pollution source, and determine the target emission port or target pollution discharge node according to the pollution contribution rate. The pollution contribution model is used to generate and display the pollution contribution value of the target emission port or target pollution discharge point in response to input instructions.

[0012] In some embodiments, the database specifically includes:

[0013] Collect and summarize the surface water system in the target monitoring area, the deployment status of monitoring equipment at each point, and the spatial relationship between various pollution sources and the cross-section catchment range;

[0014] The statistical data of the emission categories, emission factors, emission limits, industry characteristics of pollution sources and layout of sewage discharge channels of different water pollution sources within the catchment area of ​​each monitoring point;

[0015] Statistics of existing standard stations, micro water quality monitoring stations and monitoring conditions of various pollution sources within the target monitoring area.

[0016] In some embodiments, a pollution source-land-pipeline network-monitoring station smart Internet of Things grid is constructed, specifically including:

[0017] Converting each of the surface water pollution basic data into a vectorized format to obtain a plurality of vectorized objects, and marking attribute information for each vector object;

[0018] The pollution source information, pipeline information, interception wells, inspection wells, sewage outlets and water flow direction in the surface water pollution basic data are vector-spaced to obtain point data of the pollution source, interception wells, inspection wells and sewage outlets; the pipeline information is converted into line data; and the target area is converted into surface data;

[0019] Perform geographic coordinate projection on the point data, line data and surface data after spatial vectorization processing to form a geographic coordinate model with longitude and latitude coordinates;

[0020] In the geographic coordinate model, the pollution sources are divided into different levels of pipeline logical relationships according to the land area and the upstream and downstream connectivity of the pipeline network and water system, so as to construct a pollution source-land-pipeline network-monitoring station smart Internet of Things grid, and display the spatial distribution of each node, water flow direction and mutual connectivity.

[0021] In some embodiments, all data in the pollution source-land-pipeline network-monitoring station smart Internet of Things grid are preprocessed, specifically including:

[0022] Data cleaning and attribute classification of all sites;

[0023] Handle numerical outliers of water environmental parameters and align data in time and space;

[0024] Verify the longitude and latitude, unify the longitude and latitude of the monitoring points into the same coordinate system, eliminate the longitude and latitude sites that cannot be confirmed, and replace the longitude and latitude of the sites with scattered longitude and latitude through programming;

[0025] For stations with different monitoring start and end times, timing matching is performed through programming, stations that enter different stable monitoring states are deleted, and missing data in different intervals are filled in with the mean value according to the upper and lower timings.

[0026] In some embodiments, building a traceability model based on the model database specifically includes:

[0027] Using cluster analysis algorithms, the basic surface water pollution data of each pollution source are clustered to obtain the water quality data grouping results of different monitoring points and the trend of water quality changes over time;

[0028] Using the pre-trained random forest model, calculate the contribution of each pollution source to the water pollution of the monitored section and determine the pollution source that contributes most to water pollution;

[0029] A space-time-concentration three-dimensional convergence model is constructed based on a Bayesian network, and the space-time-concentration three-dimensional convergence model is trained using a model database to obtain the traceability model.

[0030] In some embodiments, building a pollution contribution model based on the model database specifically includes:

[0031] Obtain water quality monitoring data within the target period, collect pollutant emission concentration and emission data at each monitoring point, determine the distance between each monitoring point and the monitoring section, and the geographical location of each monitoring point within the catchment area;

[0032] Identify model variables;

[0033] Construct pollution contribution model, total contribution model and pollution proportion model.

[0034] In some embodiments, building a pollution contribution model based on the model database further includes:

[0035] The pollution contribution model, total contribution model and pollution proportion model are optimized according to pollutant types and seasonal factors.

[0036] The present invention also provides a device for tracing the source of abnormal pollution at a surface water discharge outlet into a river based on smart IoT, the device comprising:

[0037] A database construction unit, used to obtain basic data of surface water pollution in the target monitoring area and construct a database based on the basic data of surface water pollution;

[0038] The IoT grid construction unit is used to divide all pollution sources in the database into different levels of pipeline logic relationships according to the land area and the upstream and downstream connection relationship of the pipe network and the water system, and construct a pollution source-land area-pipe network-monitoring station smart IoT grid;

[0039] A data preprocessing unit, used to preprocess and feature analyze all data in the pollution source-land plot-pipeline network-monitoring station smart Internet of Things grid to build a model database;

[0040] A model building unit is used to build a source tracing model and a pollution contribution model based on the model database, respectively. The source tracing model is used to calculate the pollution contribution rate of each pollution source and determine the target emission port or target pollution discharge node according to the pollution contribution rate. The pollution contribution model is used to generate and display the pollution contribution value of the target emission port or target pollution discharge point in response to an input instruction.

[0041] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the program.

[0042] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and the computer program implements the steps of the above method when executed by a processor.

[0043] The method and device for tracing the abnormal pollution source of surface water discharge outlets into rivers based on smart Internet of Things provided by the present invention obtains basic data of surface water pollution in the target monitoring area and builds a database based on the basic data of surface water pollution; divides all pollution sources in the database into pipeline logical relationships of different levels according to the connection relationship between the land area and the pipeline network and the upstream and downstream of the water system, and builds a pollution source-land area-pipeline network-monitoring station smart Internet of Things grid; pre-processes and performs feature analysis on all data in the pollution source-land area-pipeline network-monitoring station smart Internet of Things grid to build a model database; and builds a tracing model and a pollution contribution model based on the model database, the tracing model is used to calculate the pollution contribution rate of each pollution source, and determine the target discharge outlet or target sewage node according to the pollution contribution rate, and the pollution contribution model is used to generate and display the pollution contribution value of the target discharge outlet or target sewage point in response to an input instruction.

[0044] In this way, the method and device for tracing the abnormal pollution of the sewage outlet into the river provided by the smart Internet of Things of the present invention, based on the construction of the database, pollution source-plot-pipeline network-monitoring station smart Internet of Things grid, can reflect the water quality of the river / lake in time and space within a certain period of time, and can obtain the distribution law of the pollution situation and the diffusion characteristics, trends and scope of the pollutant factors, which will bring more help to the investigation, management and tracing of water pollution; at the same time, through artificial intelligence algorithms and mathematical models, the tracing model and pollution contribution model are built, which can efficiently locate the suspicious range, automatically trace the suspected pollution and discharge enterprises, and dynamically lock the source of the sewage discharge at the outlet in real time, with high efficiency and good accuracy, and relatively low personnel and equipment cost investment. Thus, the problem of being unable to quickly and accurately determine the pollution source in surface water environment management is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0046] The structures, proportions, sizes, etc. illustrated in this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with the technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantial technical significance. Any structural modification, change in proportion or adjustment of size shall still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0047] Figure 1 This is one of the flow charts of the method for tracing the source of abnormal pollution at a surface water discharge outlet into a river based on smart IoT provided by the present invention;

[0048] Figure 2 This is the second flow chart of the method for tracing the abnormal pollution source of surface water into river sewage outlets based on smart Internet of Things provided by the present invention;

[0049] Figure 3 This is the third flow chart of the method for tracing the abnormal pollution source of surface water into river sewage outlets based on smart Internet of Things provided by the present invention;

[0050] Figure 4 This is the fourth flow chart of the method for tracing the abnormal pollution source of surface water into river sewage outlets based on smart Internet of Things provided by the present invention;

[0051] Figure 5 This is a schematic diagram of the structure of the device for tracing the source of abnormal pollution at a surface water discharge outlet into a river based on the smart Internet of Things provided by the present invention;

[0052] Figure 6 This is a structural block diagram of a computer device provided by the present invention. DETAILED DESCRIPTION

[0053] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] In a specific embodiment, Figure 1 As shown, the method for tracing the abnormal pollution source of surface water into river sewage outlets based on smart IoT provided by the present invention includes the following steps:

[0055] S110: Acquire basic data of surface water pollution in the target monitoring area, and construct a database based on the basic data of surface water pollution;

[0056] S120: Divide all pollution sources in the database into different levels of pipeline logic relationships according to the land area and the upstream and downstream connection relationship of the pipe network and the water system, and construct a pollution source-land area-pipe network-monitoring station smart Internet of Things grid;

[0057] S130: Preprocessing and feature analysis of all data in the pollution source-land plot-pipeline network-monitoring station smart Internet of Things grid to build a model database;

[0058] S140: Based on the model database, a source tracing model and a pollution contribution model are respectively constructed. The source tracing model is used to calculate the pollution contribution rate of each pollution source, and determine the target discharge port or target sewage discharge node according to the pollution contribution rate. The pollution contribution model is used to generate and display the pollution contribution value of the target discharge port or target sewage discharge point in response to an input instruction. When constructing a pollution source tracing model, a three-dimensional convergence model of space-time-concentration is constructed, and the calculation method is implemented using a random forest regression algorithm, cluster analysis, correlation analysis, threshold analysis, and Bayesian network, and finally the suspicious discharge port or sewage discharge node with large pollution contribution is calculated.

[0059] In step S110, the database specifically includes:

[0060] Collect and summarize the surface water system in the target monitoring area, the deployment status of monitoring equipment at each point, and the spatial relationship between various pollution sources and the cross-section catchment range;

[0061] The statistical data of the emission categories, emission factors, emission limits, industry characteristics of pollution sources and layout of sewage discharge channels of different water pollution sources within the catchment area of ​​each monitoring point;

[0062] Statistics of existing standard stations, micro water quality monitoring stations and monitoring conditions of various pollution sources within the target monitoring area.

[0063] In a specific usage scenario, a regional surface water pollution basic data resource information database is established. First, the regional water system, sewage discharge and monitoring conditions are collected, sorted and summarized, and the spatial position relationship between the regional surface water system, the layout of monitoring equipment at each point, various pollution sources (including different types of sewage outlets into the river, generalized outlets, etc.) and the cross-section catchment range is clarified. A resource information database is established to view the specific information of the monitoring point, and to count the emission categories, emission factors, emission limits, industry characteristics of pollution sources and the layout of sewage discharge channels of different water pollution sources within the catchment range of the monitoring point, so as to link the pollution source; statistics on the layout of monitoring points, including existing standard stations, micro water quality monitoring stations, automatic monitoring of various pollution sources, and manual monitoring, are counted to facilitate the further construction of the smart Internet of Things grid; monitoring devices are reasonably added at key nodes according to the spatial distribution of monitoring points, and micro water quality monitoring stations are deployed at important key nodes such as water transfer channels, inspection wells, and key sewage-discharging enterprises. Monitoring characteristic factors are selected according to the pollution source conditions to further encrypt the smart Internet of Things grid.

[0064] In step S120, Figure 2 As shown in the figure, the construction of the pollution source-land-pipeline network-monitoring station smart Internet of Things grid includes the following steps:

[0065] S210: Convert the basic surface water pollution data into a vectorized format to obtain multiple vectorized objects, and annotate attribute information for each vector object; that is, in the process of establishing the smart Internet of Things grid, according to the pollution source situation, the distribution and direction of sewage discharge channels, the distribution of pipeline plots, the distribution of enterprises and industry characteristics, and the distribution of water systems, the deployment of existing monitoring stations and important node encrypted monitoring micro-stations is comprehensively considered, and each physical element is pointed and vectorized, and the collected data is converted into a suitable vectorized format, and attribute information is added to each vector object, such as the type of pollution source, type of pollutant, emission volume, diameter of the pipeline, etc., for subsequent analysis.

[0066] S220: vector-spatialize the pollution source information, pipeline information, intercepting wells, inspection wells, sewage outlets and water flow direction in the basic data of surface water pollution to obtain point data of the pollution source, intercepting wells, inspection wells and sewage outlets; convert the pipeline information into line data; convert the target area into surface data; that is, vector-spatialize the pollution source information, pipeline information, intercepting wells, inspection wells, sewage outlets and effective water flow direction information in the database, and the vectorization processing is to convert the information such as pollution source, intercepting well, inspection well and sewage outlet into point data, and convert the pipeline information into line data to indicate the direction and connection relationship of the pipeline, convert certain areas into surface data as needed, and add auxiliary information to each vector object.

[0067] S230: Perform geographic coordinate projection on the point data, line data and surface data after spatial vectorization processing to form a geographic coordinate model with longitude and latitude coordinates; perform geographic coordinate projection transformation on the spatial vectorized information to form a geographic coordinate model with longitude and latitude coordinates, and process it using conventional geographic information system (GIS) software to ensure that all data use the same coordinate system.

[0068] S240: In the geographic coordinate model, the pollution sources are divided into pipeline logical relationships of different levels according to the land area and the upstream and downstream connectivity of the pipeline network and water system, so as to construct a pollution source-land area-pipeline network-monitoring station smart Internet of Things grid, and display the spatial distribution of each node, the direction of water flow and the interconnection relationship; implement river basin management for pollution sources, drainage channels and surface water systems, establish a pointer relationship library from the four parts of pollution source-land area-water system network-monitoring station, divide the pollution sources into pipeline logical relationships of different levels according to the land area and the upstream and downstream connectivity of the pipeline network and water system, construct the source-land area-network-monitoring station smart Internet of Things grid association relationship, and display the spatial distribution of each node, the direction of water flow and the interconnection relationship.

[0069] In step S130, all data in the pollution source-land-pipeline network-monitoring station smart Internet of Things grid are preprocessed, such as Figure 3 As shown, the specific steps include:

[0070] S310: Data cleaning and attribute classification of all sites; invalid data, abnormal data, and noise interference data are removed during data cleaning;

[0071] S320: Processing water environment parameter numerical abnormal values, and aligning the data in time and space; wherein the water environment parameter numerical abnormal values ​​include null values, 0 values, negative values, constant values, maximum and minimum values, etc.; aligning the data in time and space to ensure the consistency and comparability of the data;

[0072] S330: Verify the longitude and latitude, unify the longitude and latitude of the monitoring points into the same coordinate system, remove the longitude and latitude sites that cannot be confirmed, and replace the longitude and latitude of the sites with scattered longitude and latitude through programming;

[0073] S340: For stations with different monitoring start and end times, time sequence matching is performed through programming, stations that enter different stable monitoring states are deleted, and missing data in different intervals are filled with mean values ​​according to the upper and lower time sequences.

[0074] In this way, after the relevant monitoring data in the database are pre-processed by cleaning, conversion, normalization, registration and spatialization, the temporal and spatial variation characteristics of the water quality data are analyzed.

[0075] In some embodiments, Figure 4 As shown, building a traceability model based on the model database specifically includes the following steps:

[0076] S410: clustering the basic surface water pollution data of each pollution source using a cluster analysis algorithm to obtain the water quality data grouping results of different monitoring points and the trend of water quality changes over time.

[0077] Before applying random forests, cluster analysis can be used to preprocess the data. Cluster analysis can help identify the characteristics of different water bodies, the impact of pollution sources, and patterns of water quality changes. Through cluster analysis, water quality data from different monitoring points can be grouped to identify areas with similar water quality. By analyzing the emission data of different pollution sources, pollution sources with greater impact on water quality can be identified. Time series clustering of water quality data can identify patterns and trends in water quality changes. Revealing the inherent structure of water quality data can also provide valuable feature information for subsequent analysis (such as random forest models).

[0078] Furthermore, cluster analysis in water environment pollution source tracing collects and summarizes water quality monitoring data including different pollution sources, determines the characteristic variables used for clustering, and performs preprocessing steps such as data cleaning, missing value processing, and standardization to ensure data quality and reliability. Select a suitable clustering algorithm, such as K-means clustering, hierarchical clustering, density clustering, etc., and choose the most appropriate algorithm according to the specific situation. Perform cluster analysis on the data to divide the data points into different clusters, each cluster represents data points with similar characteristics; analyze the clustering results, identify the types of pollution sources represented by different clusters, and evaluate the similarities and differences between the pollution sources. Based on the clustering results, the characteristics of different pollution sources can be further analyzed to assist in source tracing.

[0079] Furthermore, before building the random forest model, correlation analysis can be performed to identify the relationship between features. By calculating the correlation between features and the target variable, features with high correlation with the target variable can be selected, thereby reducing the complexity of the model and improving performance. The Pearson correlation coefficient or Spearman rank correlation coefficient is used to calculate the correlation between each pollution source and the water quality index. Based on the correlation analysis results, the pollution sources with high correlation with the water quality index are identified, and further in-depth analysis is performed to quantify their impact on water quality.

[0080] S420: Use the pre-trained random forest model to calculate the contribution of each pollution source to the water pollution of the monitored section and determine the pollution source that contributes the most to the water pollution.

[0081] Specifically, when using the random forest algorithm to calculate the contribution of each pollution source to the environmental pollution of the surface water monitoring point, data preparation is first performed, including the collection of water quality data of the monitoring section, including the concentration of major pollutants (such as the conventional five parameters, COD, permanganate index, ammonia nitrogen, total phosphorus, total nitrogen, etc.), the collection of emission data of each pollution source, including industrial emissions, agricultural non-point source pollution, domestic sewage, etc., the collection of meteorological data (such as precipitation, temperature, wind speed, etc.) and hydrological data (such as flow, water level, etc.), and the data is organized into a table (data frame), each row represents the monitoring data at a time point, and the columns include the target variable (water quality index) and the characteristic variable (emissions of each pollution source, meteorological and hydrological conditions, etc.). The data is preprocessed, data cleaning is performed, missing values ​​(filled or deleted), outliers are processed, the integrity and accuracy of the data are ensured, the data is merged, the water quality monitoring data, pollution source data, and meteorological and hydrological data are merged into a complete data set, and time alignment is performed to ensure that all data are aligned in time for subsequent analysis. Select features related to water pollution, including the emissions of each pollution source, meteorological and hydrological conditions, etc., convert timestamps into periodic features (such as hours, days, weeks, etc.), construct interactive features between pollution sources, consider the joint effects of different pollution sources, and standardize or normalize the features to improve the performance of the model. Construct a random forest model, divide the data set into a training set and a test set, usually 70% as a training set and 30% as a test set, use the training set to train the random forest model, use the test set to make predictions, and evaluate the model performance. Use the feature importance evaluation function of the random forest model to calculate the contribution of each pollution source to the water pollution of the monitored section, use visualization tools to display the feature importance, help understand the impact of each pollution source, analyze the feature importance results, and identify the pollution source that contributes most to water pollution. According to the performance of the model and the actual situation, iterative optimization such as feature selection and model parameter adjustment is performed to improve the accuracy and reliability of the model.

[0082] The relationship between random forest algorithm, cluster analysis and correlation analysis is mainly reflected in data preprocessing, feature selection and model evaluation. By combining these three methods, we can better understand the data structure, select appropriate features, and improve the performance and explanatory power of the model. Among them, the random forest algorithm is used to analyze the contribution of each pollution source to the pollution of the monitoring point. By collecting a sample data set including the environmental pollution data of the monitoring point and the emission data of each pollution source, it is determined which pollution source emission data will be input as a feature into the random forest model, and the data is cleaned, missing value processing and standardized. Preprocessing steps; use the random forest algorithm to train the data set and establish a model; obtain the importance ranking of each feature through the random forest model, so as to understand the contribution of each pollution source to the environmental pollution of the monitoring point.

[0083] S430: constructing a space-time-concentration three-dimensional convergence model based on a Bayesian network, and training the space-time-concentration three-dimensional convergence model using a model database to obtain the traceability model.

[0084] Specifically, the space-time-concentration three-dimensional convergence model established by the Bayesian network has a clear research goal of identifying pollution sources and evaluating changes in pollutant concentrations. It determines spatial, temporal and concentration variables, sets possible nodes, such as different pollution sources (outlets into rivers, industrial sewage outlets, etc.), different pollutant concentrations, hydrological factors (flow, flow velocity, water level, etc.) and meteorological factors (rainfall, temperature, etc.); determines the connection relationship of edges, identifies the causal relationship between variables, such as rainfall may wash pollutants into the water body and affect the pollutant concentration, the impact of agricultural non-point sources on the pollutant concentration of surrounding water bodies, etc., uses statistical analysis methods to identify the relationship between variables, uses the Bayesian network structure learning algorithm to automatically learn the network structure from the data, uses the greedy algorithm to gradually add edges to maximize the likelihood of the network, uses the constraint-based method, determines the existence of edges through statistical tests, evaluates the advantages and disadvantages of different structures, and selects the structure with the highest score. After determining the connection relationship between nodes and edges, design a suitable network structure, use cross-validation and other methods to evaluate the performance of the model, and adjust the network structure when necessary. According to the prediction results of the model and the actual observation results, the network structure is continuously optimized. Based on the trained Bayesian deep learning model, a probability distribution model of water quality indicators is established. Based on the probability distribution of water quality indicators, a heat map is generated to show the water quality compliance status of different areas. The combined impact of time and space is considered, water quality data is obtained in real time, and model predictions are updated. When the water quality indicators are close to the substandard threshold, an early warning mechanism is triggered.

[0085] Establishing a Bayesian network for a space-time-concentration three-dimensional convergence model is a complex process. It involves collecting sample data sets including spatial location, time, and pollutant concentration, and performing preprocessing steps such as cleaning the data, processing missing values ​​and outliers; determining the variables in the model, including spatial location coordinates, time, and pollutant concentration; determining the relationship between variables, such as spatial location may affect pollutant concentration, and time changes may also affect concentration; designing the structure of the Bayesian network based on the relationship between variables, and determining the connection relationship between nodes and edges; considering the dependency between space, time, and concentration, and designing a suitable network structure; assigning a probability distribution to each node to represent the relationship between the node and its parent node; using sample data to learn the parameters of the Bayesian network and estimate the probability distribution parameters of each node; using the established Bayesian network to make inferences, and inferring the values ​​of other variables based on known variable values; and making model predictions to estimate the future space-time-concentration three-dimensional convergence. Based on the construction of a deep learning model, a probability spatial distribution measurement method is used to analyze the distribution of thermal regions and the main reasons that affect water quality compliance, and to achieve dynamic spatiotemporal sources.

[0086] The national standards of various water quality indicators are used as thresholds, and water quality values ​​exceeding the threshold range are considered abnormal values. By performing threshold analysis on the water quality indicators of the micro-station, abnormal values ​​exceeding the threshold are screened out, thereby accurately tracing the source from the micro-station of the pipeline network to the plot of land to the source.

[0087] While constructing the above-mentioned artificial intelligence model, a mathematical model can also be constructed, that is, a mathematical model is established by comprehensively using data such as water quality section monitoring and automatic monitoring of pollution sources and the catchment range of the section. From the three dimensions of the distance between the enterprise outlet and the section, the enterprise pollutant emission concentration and the emission volume, the "pollution contribution" ranking of the pollutant discharge unit is calculated, and the specified time period, specified section, and specified pollutant are visually displayed. The proportion of the pollutant emissions of each pollutant discharge unit within the catchment range on the pollutant concentration monitored in the section. The greater the proportion, the greater the impact on the water quality of the section. The specific function is that statistical analysis can be performed based on conditions such as section type (national control, provincial control), section, monitoring factor, pollutant discharge unit (discharge outlet), pollutant, time type (hour, day), time range, etc. The analysis results include cities, counties, pollutant discharge units (discharge outlets), river discharge outlets, section monitoring factors, discharged pollutants, distance from the section (km), distance ratio, average concentration (mg / L), concentration ratio, emission volume (kg), emission ratio, and impact ratio. The analysis results are displayed in the form of a paginated list, and the analysis results support export to Excel. Click on the pollutant discharge unit (discharge outlet) in the analysis result list, and a trend analysis page will pop up to display the concentration change trend of the corresponding section and section monitoring factors within the statistical time range, as well as the concentration and emission change trend of the corresponding pollutant discharge unit and discharge outlet, pollutants.

[0088] In some embodiments, building a pollution contribution model based on the model database specifically includes:

[0089] Obtain water quality monitoring data within the target period, collect pollutant emission concentration and emission data at each monitoring point, determine the distance between each monitoring point and the monitoring section, and the geographical location of each monitoring point within the catchment area;

[0090] Identify model variables;

[0091] Construct pollution contribution model, total contribution model and pollution proportion model.

[0092] More specifically, the model construction process includes the following steps:

[0093] S11: Data collection and collation, obtaining water quality monitoring data within a specified period, including pollutant concentrations (such as COD, ammonia nitrogen, etc.) at the monitoring section, collecting pollutant emission concentration and emission data at each monitoring point to determine the distance between each monitoring point and the monitoring section, as well as their geographical locations within the catchment area;

[0094] S12: Determine model variables, including distance variable (D): distance between the pollutant discharge unit and the monitoring section (unit: m); emission concentration (C): pollutant emission concentration of the pollutant discharge unit (unit: mg / L); emission volume (Q): pollutant emission volume of the pollutant discharge unit (unit: kg / h); monitoring concentration (M): pollutant concentration at the monitoring section (unit: mg / L);

[0095] S13: Establish a pollution contribution model. In order to calculate the contribution of each pollution source to the pollutant concentration in the monitoring section, the following formula can be used:

[0096]

[0097] in:

[0098] Contribution i is the contribution of the ith pollutant discharge unit to the pollutant concentration in the monitoring section;

[0099] C i is the pollutant emission concentration of the i-th pollutant discharge unit;

[0100] Q i is the pollutant emission of the ith pollutant emission unit;

[0101] D i is the distance between the ith pollutant discharge unit and the monitoring section.

[0102] S14: Calculate the total contribution, calculate the total contribution of pollution sources to the pollutant concentration in the monitoring section:

[0103]

[0104] S15: Calculate the proportion of pollution contribution of each pollutant discharge unit:

[0105]

[0106] S16: Ranking and visualization: Rank each pollutant discharger according to its pollution contribution ratio, and use visualization tools (such as Matplotlib, Tableau, etc.) to display the pollution contribution ratio of each pollutant discharger for intuitive display.

[0107] Further, a pollution contribution model is built based on the model database, and then it also includes:

[0108] The pollution contribution model, total contribution model and pollution proportion model are optimized according to pollutant types and seasonal factors.

[0109] That is to say, the above model can be adjusted and optimized to improve its accuracy and practicality by considering the impact of different pollutants and seasonal changes. The specific adjustment and optimization process includes the following steps:

[0110] S21: Calculate the contribution of different pollutants i ; Establish a multi-pollutant model: For different types of pollutants (such as COD, ammonia nitrogen, heavy metals, etc.), establish pollution contribution models respectively, and define independent contribution calculation formulas for each pollutant:

[0111]

[0112] in:

[0113] i represents the i-th pollutant discharge unit;

[0114] j represents the jth pollutant;

[0115] C ij It represents the emission concentration of the jth pollutant of the i-th pollutant discharge unit;

[0116] Q ij It represents the emission of the jth pollutant of the i-th pollutant discharge unit;

[0117] It represents the distance between the jth pollutant of the ith pollutant discharge unit and the monitoring section;

[0118] Introduce weight coefficients; according to the impact of different pollutants on water quality, introduce weight coefficients W for the contribution calculation of each pollutant j :

[0119] Weighted Contribution ij =W j ×Contribution ij

[0120] S22: Calculate the contribution of seasonal changes The introduction of seasonal factors means adjusting the emission concentration and emission amount according to the hydrological characteristics and pollutant emission characteristics of different seasons; for example, in summer, the concentration of some pollutants may change due to rainfall and rising water temperature, and the expression is:

[0121]

[0122] Among them, S j It is a seasonal factor and can be obtained through statistical analysis of historical data.

[0123] Use time series models: Perform time series analysis on historical monitoring data to identify seasonal trends and cyclical changes and predict future pollutant concentrations.

[0124] S23: Comprehensive model building

[0125] The final pollution contribution model can comprehensively consider multiple pollutants and seasonal factors, and calculate the total contribution TotalContribution by comprehensively considering multiple pollutants and seasonality j :

[0126]

[0127] Calculate the contribution of each pollutant:

[0128]

[0129] Furthermore, after completing the source tracing, the present invention also includes automatically identifying and pushing a list of abnormal pollution sources / a list of pollution sources with a high proportion, so as to achieve accurate source tracing.

[0130] In the above specific implementation manner, the method for tracing the abnormal pollution source of surface water discharge outlets into rivers based on smart Internet of Things provided by the present invention obtains the basic data of surface water pollution in the target monitoring area, and constructs a database based on the basic data of surface water pollution; divides all pollution sources in the database into different levels of pipeline logical relationships according to the land area and pipeline network, and the upstream and downstream connectivity of the water system, and constructs a pollution source-land area-pipeline network-monitoring station smart Internet of Things grid; pre-processes and feature analyzes all data in the pollution source-land area-pipeline network-monitoring station smart Internet of Things grid to construct a model database; and builds a tracing model and a pollution contribution model based on the model database, the tracing model is used to calculate the pollution contribution rate of each pollution source, and determine the target discharge outlet or target sewage node according to the pollution contribution rate, and the pollution contribution model is used to generate and display the pollution contribution value of the target discharge outlet or target sewage point in response to an input instruction.

[0131] In this way, the method for tracing the abnormal pollution source of the sewage outlet of the surface water into the river provided by the smart Internet of Things of the present invention is based on the construction of the database, pollution source-plot-pipeline network-monitoring station smart Internet of Things grid, which can reflect the water quality of the river / lake in time and space within a certain period of time, and can obtain the distribution law of the pollution situation and the diffusion characteristics, trends and scope of the pollutant factors, which brings more help to the investigation, management and tracing of water pollution; at the same time, through artificial intelligence algorithms and mathematical models, the tracing model and the pollution contribution model are built, which can efficiently locate the suspicious range, automatically trace the suspicious pollution and the enterprises that secretly discharge, and dynamically lock the source of the sewage discharge at the outlet in real time, and the efficiency is high, the accuracy is good, and the personnel and equipment cost investment is relatively low. Thereby solving the problem of being unable to quickly and accurately determine the pollution source in the surface water environment management.

[0132] In addition to the above method, the present invention also provides a device for tracing the source of abnormal pollution at a surface water discharge outlet into a river based on smart IoT, such as Figure 5 As shown, the device comprises:

[0133] A database construction unit 510 is used to obtain basic data of surface water pollution in a target monitoring area and construct a database based on the basic data of surface water pollution;

[0134] The Internet of Things grid construction unit 520 is used to divide all pollution sources in the database into different levels of pipeline logical relationships according to the land area and the upstream and downstream connection relationship of the pipe network and the water system, and construct a pollution source-land area-pipe network-monitoring station smart Internet of Things grid;

[0135] A data preprocessing unit 530 is used to preprocess and analyze the characteristics of all data in the pollution source-land plot-pipeline network-monitoring station smart Internet of Things grid to build a model database;

[0136] The model building unit 540 is used to build a source tracing model and a pollution contribution model based on the model database. The source tracing model is used to calculate the pollution contribution rate of each pollution source and determine the target emission port or target pollution discharge node according to the pollution contribution rate. The pollution contribution model is used to generate and display the pollution contribution value of the target emission port or target pollution discharge point in response to an input instruction.

[0137] In some embodiments, the database specifically includes:

[0138] Collect and summarize the surface water system in the target monitoring area, the deployment status of monitoring equipment at each point, and the spatial relationship between various pollution sources and the cross-section catchment range;

[0139] The statistical data of the emission categories, emission factors, emission limits, industry characteristics of pollution sources and layout of sewage discharge channels of different water pollution sources within the catchment area of ​​each monitoring point;

[0140] Statistics of existing standard stations, micro water quality monitoring stations and monitoring conditions of various pollution sources within the target monitoring area.

[0141] In some embodiments, a pollution source-land-pipeline network-monitoring station smart Internet of Things grid is constructed, specifically including:

[0142] Converting each of the surface water pollution basic data into a vectorized format to obtain a plurality of vectorized objects, and marking attribute information for each vector object;

[0143] The pollution source information, pipeline information, interception wells, inspection wells, sewage outlets and water flow direction in the surface water pollution basic data are vector-spaced to obtain point data of the pollution source, interception wells, inspection wells and sewage outlets; the pipeline information is converted into line data; and the target area is converted into surface data;

[0144] Perform geographic coordinate projection on the point data, line data and surface data after spatial vectorization processing to form a geographic coordinate model with longitude and latitude coordinates;

[0145] In the geographic coordinate model, the pollution sources are divided into different levels of pipeline logical relationships according to the land area and the upstream and downstream connectivity of the pipeline network and water system, so as to construct a pollution source-land-pipeline network-monitoring station smart Internet of Things grid, and display the spatial distribution of each node, water flow direction and mutual connectivity.

[0146] In some embodiments, all data in the pollution source-land-pipeline network-monitoring station smart Internet of Things grid are preprocessed, specifically including:

[0147] Data cleaning and attribute classification of all sites;

[0148] Handle numerical outliers of water environmental parameters and align data in time and space;

[0149] Verify the longitude and latitude, unify the longitude and latitude of the monitoring points into the same coordinate system, eliminate the longitude and latitude sites that cannot be confirmed, and replace the longitude and latitude of the sites with scattered longitude and latitude through programming;

[0150] For stations with different monitoring start and end times, timing matching is performed through programming, stations that enter different stable monitoring states are deleted, and missing data in different intervals are filled in with the mean value according to the upper and lower timings.

[0151] In some embodiments, building a traceability model based on the model database specifically includes:

[0152] Using cluster analysis algorithms, the basic surface water pollution data of each pollution source are clustered to obtain the water quality data grouping results of different monitoring points and the trend of water quality changes over time;

[0153] Using the pre-trained random forest model, calculate the contribution of each pollution source to the water pollution of the monitored section and determine the pollution source that contributes most to water pollution;

[0154] A space-time-concentration three-dimensional convergence model is constructed based on a Bayesian network, and the space-time-concentration three-dimensional convergence model is trained using a model database to obtain the traceability model.

[0155] In some embodiments, building a pollution contribution model based on the model database specifically includes:

[0156] Obtain water quality monitoring data within the target period, collect pollutant emission concentration and emission data at each monitoring point, determine the distance between each monitoring point and the monitoring section, and the geographical location of each monitoring point within the catchment area;

[0157] Identify model variables;

[0158] Construct pollution contribution model, total contribution model and pollution proportion model.

[0159] In some embodiments, building a pollution contribution model based on the model database further includes:

[0160] The pollution contribution model, total contribution model and pollution proportion model are optimized according to pollutant types and seasonal factors.

[0161] In the above specific implementation manner, the device for tracing the abnormal pollution source of surface water discharge outlets into rivers based on smart Internet of Things provided by the present invention obtains the basic data of surface water pollution in the target monitoring area, and constructs a database based on the basic data of surface water pollution; divides all pollution sources in the database into different levels of pipeline logical relationships according to the land area and pipeline network, and the upstream and downstream connection relationship of the water system, and constructs a pollution source-land area-pipeline network-monitoring station smart Internet of Things grid; pre-processes and feature analyzes all data in the pollution source-land area-pipeline network-monitoring station smart Internet of Things grid to construct a model database; and builds a tracing model and a pollution contribution model based on the model database, the tracing model is used to calculate the pollution contribution rate of each pollution source, and determine the target discharge outlet or target sewage node according to the pollution contribution rate, and the pollution contribution model is used to generate and display the pollution contribution value of the target discharge outlet or target sewage point in response to an input instruction.

[0162] In this way, the surface water into the river sewage outlet pollution abnormal tracing device provided by the present invention, based on the construction of the database, pollution source-plot-pipeline network-monitoring station smart Internet of Things grid, can reflect the water quality of rivers / lakes in time and space within a certain period of time, and can obtain the distribution law of pollution conditions and the diffusion characteristics, trends and scope of pollutant factors, which brings more help to the investigation, management and tracing of water pollution; at the same time, through artificial intelligence algorithms and mathematical models, the tracing model and pollution contribution model are built, which can efficiently locate the suspicious range, automatically trace the suspicious pollution and discharge enterprises, and dynamically lock the source of the discharge of wastewater in real time, and the efficiency is high, the accuracy is good, and the personnel and equipment cost investment is relatively low. Thereby solving the problem of being unable to quickly and accurately determine the source of pollution in surface water environment management.

[0163] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory and a network interface connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a model prediction. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The model prediction of the computer device is used to store static information and dynamic information data. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the steps in the above method embodiment are implemented.

[0164] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0165] Corresponding to the above embodiment, the embodiment of the present invention further provides a computer storage medium, which contains one or more program instructions, wherein the one or more program instructions are used to execute the above method.

[0166] The present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the above method.

[0167] In the embodiment of the present invention, the processor may be an integrated circuit chip having the ability to process signals. The processor may be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0168] The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor for execution, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The processor reads the information in the storage medium and completes the steps of the above method in combination with its hardware.

[0169] The storage medium may be a memory, which may be, for example, a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memory.

[0170] Among them, the non-volatile memory can be a read-only memory (ROM), a programmable ROM (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.

[0171] The volatile memory may be a random access memory (RAM) which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct Rambus RAM (DRRAM).

[0172] The storage media described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memory.

[0173] Those skilled in the art will appreciate that in one or more of the above examples, the functions described in the present invention can be implemented using a combination of hardware and software. When software is used, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any media that facilitates the transmission of computer programs from one place to another. Storage media can be any available media that can be accessed by general or special-purpose computers.

[0174] The above specific implementation methods further illustrate the purpose, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above are only specific implementation methods of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solutions of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for tracing the source of abnormal pollution at a surface water discharge outlet into a river based on smart IoT, characterized in that: The method comprises: Obtaining basic data on surface water pollution in the target monitoring area, and building a database based on the basic data on surface water pollution; All pollution sources in the database are divided into different levels of pipeline logic relationships according to the land area and the upstream and downstream connection relationship of the pipeline network and water system, and a pollution source-land area-pipeline network-monitoring station smart Internet of Things grid is constructed; Preprocess and feature analyze all data in the pollution source-land-pipeline network-monitoring station smart Internet of Things grid to build a model database; Based on the model database, a source tracing model and a pollution contribution model are respectively constructed, wherein the source tracing model is used to calculate the pollution contribution rate of each pollution source, and determine the target discharge port or target pollution discharge node according to the pollution contribution rate, and the pollution contribution model is used to generate and display the pollution contribution value of the target discharge port or target pollution discharge point in response to an input instruction; Wherein, building a pollution contribution model based on the model database specifically includes: Obtain water quality monitoring data within the target period, collect pollutant emission concentration and emission data at each monitoring point, determine the distance between each monitoring point and the monitoring section, and the geographical location of each monitoring point within the catchment area; Identify model variables; Construct pollution contribution model, total contribution model and pollution proportion model; Building a pollution contribution model based on the model database, and then including: The pollution contribution model, total contribution model and pollution proportion model are optimized according to pollutant types and seasonal factors.

2. The method for tracing the abnormal pollution source of surface water discharge outlets into rivers based on smart IoT according to claim 1 is characterized in that: The database specifically includes: Collect and summarize the surface water system in the target monitoring area, the deployment status of monitoring equipment at each point, and the spatial relationship between various pollution sources and the cross-section catchment range; The statistical data of the emission categories, emission factors, emission limits, industry characteristics of pollution sources and layout of sewage discharge channels of different water pollution sources within the catchment area of ​​each monitoring point; Statistics of existing standard stations, micro water quality monitoring stations and monitoring conditions of various pollution sources within the target monitoring area.

3. The method for tracing the abnormal pollution source of surface water discharge outlets into rivers based on smart IoT according to claim 1 is characterized in that: Construct a smart IoT grid of pollution source-land-pipeline network-monitoring station, including: Converting each of the surface water pollution basic data into a vectorized format to obtain a plurality of vectorized objects, and marking attribute information for each vector object; The pollution source information, pipeline information, interception wells, inspection wells, sewage outlets and water flow direction in the surface water pollution basic data are vector-spaced to obtain point data of the pollution source, interception wells, inspection wells and sewage outlets; the pipeline information is converted into line data; and the target area is converted into surface data; Perform geographic coordinate projection on the point data, line data and surface data after spatial vectorization processing to form a geographic coordinate model with longitude and latitude coordinates; In the geographic coordinate model, the pollution sources are divided into different levels of pipeline logical relationships according to the land area and the upstream and downstream connectivity of the pipeline network and water system, so as to construct a pollution source-land-pipeline network-monitoring station smart Internet of Things grid, and display the spatial distribution of each node, water flow direction and mutual connectivity.

4. The method for tracing the abnormal pollution source of surface water discharge outlets into rivers based on smart IoT according to claim 1 is characterized in that: Preprocess all data in the pollution source-land-pipeline network-monitoring station smart Internet of Things grid, specifically including: Data cleaning and attribute classification of all sites; Handle numerical outliers of water environmental parameters and align data in time and space; Verify the longitude and latitude, unify the longitude and latitude of the monitoring points into the same coordinate system, eliminate the longitude and latitude sites that cannot be confirmed, and replace the longitude and latitude of the sites with scattered longitude and latitude through programming; For stations with different monitoring start and end times, timing matching is performed through programming, stations that enter different stable monitoring states are deleted, and missing data in different intervals are filled in with the mean value according to the upper and lower timings.

5. The method for tracing the abnormal pollution source of surface water discharge outlets into rivers based on smart IoT according to any one of claims 1 to 4, characterized in that: Building a traceability model based on the model database specifically includes: Using cluster analysis algorithms, the basic surface water pollution data of each pollution source are clustered to obtain the water quality data grouping results of different monitoring points and the trend of water quality changes over time; Using the pre-trained random forest model, calculate the contribution of each pollution source to the water pollution of the monitored section and determine the pollution source that contributes most to water pollution; A space-time-concentration three-dimensional convergence model is constructed based on a Bayesian network, and the space-time-concentration three-dimensional convergence model is trained using a model database to obtain the traceability model.

6. A device for tracing the source of abnormal pollution at a surface water discharge outlet into a river based on smart IoT, characterized in that: The device comprises: A database construction unit, used to obtain basic data of surface water pollution in the target monitoring area and construct a database based on the basic data of surface water pollution; The IoT grid construction unit is used to divide all pollution sources in the database into different levels of pipeline logic relationships according to the land area and the upstream and downstream connection relationship of the pipe network and the water system, and construct a pollution source-land area-pipe network-monitoring station smart IoT grid; A data preprocessing unit, used to preprocess and feature analyze all data in the pollution source-land plot-pipeline network-monitoring station smart Internet of Things grid to build a model database; A model building unit, used to build a source tracing model and a pollution contribution model based on the model database, the source tracing model is used to calculate the pollution contribution rate of each pollution source, and determine the target discharge port or target pollution discharge node according to the pollution contribution rate, and the pollution contribution model is used to generate and display the pollution contribution value of the target discharge port or target pollution discharge point in response to an input instruction; Wherein, building a pollution contribution model based on the model database specifically includes: Obtain water quality monitoring data within the target period, collect pollutant emission concentration and emission data at each monitoring point, determine the distance between each monitoring point and the monitoring section, and the geographical location of each monitoring point within the catchment area; Identify model variables; Construct pollution contribution model, total contribution model and pollution proportion model; Building a pollution contribution model based on the model database, and then including: The pollution contribution model, total contribution model and pollution proportion model are optimized according to pollutant types and seasonal factors.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images