Water quality pollution tracing method and terminal

Abstract

The application discloses a water quality pollution tracing method and a terminal, establishes a water quality pollution knowledge graph of a target river basin, constructs a tracing network based on the water quality pollution knowledge graph, monitoring sites and upstream and downstream relationships, acquires a pollutant index of each monitoring site, determines the monitoring site as an abnormal monitoring site when the pollutant index exceeds a preset concentration, generates a water quality pollution event corresponding to the abnormal monitoring site, traces the water quality pollution event based on the tracing network, and enables the tracing network to adaptively infer each abnormal monitoring site by combining the water quality pollution knowledge graph, so that the tracing network is changed from locating a polluted water section to locating a specific pollution source, and the tracing precision is improved. Even if a complex lake water environment is polluted, high-precision tracing can be achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water environment pollution monitoring, and in particular to a water quality pollution tracing method and terminal. BACKGROUND

[0002] Water environment pollution tracing, that is, evaluating all possible pollution sources in a water quality pollution event, finding out the location and information of the pollution source, and mainly inferring the process as follows: identifying a water quality pollution event, extracting water quality pollution characteristics, tracing the pollution source path, locating the pollution source position, and evaluating the pollution source impact range.

[0003] The existing water environment pollution tracing methods are similar and follow the traditional investigation approach:

[0004] (1) The water quality pollution categories, pollution sources, and water network context involved in the water environment are sorted out, wherein the water quality pollution categories are mainly derived from the summary of historical pollution events and the investigation of potential pollution events, the pollution source information is mainly some sewage pipes, enterprise factories, and residential areas around the water environment, and the water network context is mainly the relationship between the inflow and outflow of waterways;

[0005] (2) The waterway network is divided to obtain corresponding segment nodes (pipe network nodes, water collection units, etc.), and in theory, the more detailed the segmentation is, the more nodes there are, and the higher the pollution source positioning accuracy is. Due to the cost limitation, reasonable division is often made based on the information sorted out in (1);

[0006] (3) Monitoring devices are arranged at each node to obtain real-time water quality monitoring data, and the data interval time is determined according to the monitoring device;

[0007] (4) The alarm information received in a certain period is located to the node with excessive pollution concentration;

[0008] (5) The segment with abnormal pollution concentration is traced along the upstream node, the pollution inflow and outflow of each node segment are calculated, and the pollution source is roughly located;

[0009] (6) The pollution source information of the segment is investigated to perform fine positioning.

[0010] The above-mentioned traditional water environment pollution tracing scheme fixes the pollution mode of the monitoring site, is suitable for relatively closed environments such as pipes or relatively simple river basins such as surrounding environments. This type of way tends to trace simple pollution events, and for complex water environments such as the openness and invasiveness of lake water environments, multiple pollution points are often mixed, and the tracing result is very rough, which makes the credibility of the tracing result not high. SUMMARY

[0011] The technical problem solved by the present application is to provide a water pollution tracing method and terminal, which can improve the tracing accuracy.

[0012] To solve the above technical problems, one technical solution adopted by the present application is:

[0013] A water pollution tracing method comprises the following steps:

[0014] Establishing a water pollution knowledge graph of a target river basin;

[0015] Obtaining monitoring stations of the target river basin and upstream and downstream relationships of the monitoring stations, and constructing a tracing network based on the water pollution knowledge graph, the monitoring stations and the upstream and downstream relationships;

[0016] Obtaining a pollutant index of each monitoring station, and determining whether the pollutant index exceeds a preset concentration, if yes, determining the monitoring station as an abnormal monitoring station, and generating a water pollution event corresponding to the abnormal monitoring station;

[0017] Tracing the water pollution event based on the tracing network to obtain a tracing result.

[0018] To solve the above technical problems, another technical solution adopted by the present application is:

[0019] A water pollution tracing terminal comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the following steps when executing the computer program:

[0020] Establishing a water pollution knowledge graph of a target river basin;

[0021] Obtaining monitoring stations of the target river basin and upstream and downstream relationships of the monitoring stations, and constructing a tracing network based on the water pollution knowledge graph, the monitoring stations and the upstream and downstream relationships;

[0022] Obtaining a pollutant index of each monitoring station, and determining whether the pollutant index exceeds a preset concentration, if yes, determining the monitoring station as an abnormal monitoring station, and generating a water pollution event corresponding to the abnormal monitoring station;

[0023] Tracing the water pollution event based on the tracing network to obtain a tracing result.

[0024] The beneficial effects of the present application are that the water quality pollution knowledge graph of the target basin is established, the tracing network is constructed based on the water quality pollution knowledge graph, the monitoring sites and the upstream and downstream relationship, the pollutant index of each monitoring site is acquired, when it exceeds the preset concentration, the monitoring site is determined as an abnormal monitoring site, and the water quality pollution event corresponding to the abnormal monitoring site is generated, the water quality pollution event is traced based on the tracing network, the tracing network can realize adaptive inference for each abnormal monitoring site by combining the water quality pollution knowledge graph, and the existing pollution water section positioning is changed to specific pollution source positioning, so that the tracing accuracy is improved, and even the complex lake water environment pollution can realize high-precision tracing. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 A step flow chart of a water quality pollution tracing method according to an embodiment of the present application;

[0026] Figure 2 A structure schematic diagram of a water quality pollution tracing terminal according to an embodiment of the present application;

[0027] Figure 3 A water quality pollution knowledge graph according to an embodiment of the present application;

[0028] Figure 4 A tracing network according to an embodiment of the present application;

[0029] Figure 5 A pollution contribution index probability of the tracing network according to an embodiment of the present application. DETAILED DESCRIPTION

[0030] To make the technical content, the achieved purposes and effects of the present application clear, the following will be described in detail in combination with the embodiments and the accompanying drawings.

[0031] Please refer to Figure 1 The present application provides a water quality pollution tracing method, which comprises the following steps:

[0032] Establishing a water quality pollution knowledge graph of a target basin;

[0033] Acquiring monitoring sites of the target basin and upstream and downstream relationships of the monitoring sites, and constructing a tracing network based on the water quality pollution knowledge graph, the monitoring sites and the upstream and downstream relationships;

[0034] Acquiring a pollutant index of each monitoring site, and determining whether the pollutant index exceeds a preset concentration, if yes, determining the monitoring site as an abnormal monitoring site, and generating a water quality pollution event corresponding to the abnormal monitoring site;

[0035] Trace the water quality pollution event based on the traceability network to obtain a traceability result.

[0036] From the above description, the beneficial effects of the present application are that the water quality pollution knowledge graph of the target river basin is established, the traceability network is constructed based on the water quality pollution knowledge graph, the monitoring site and the upstream and downstream relationship, the pollutant index of each monitoring site is obtained, when it exceeds the preset concentration, the monitoring site is determined as an abnormal monitoring site, and the water quality pollution event corresponding to the abnormal monitoring site is generated, the water quality pollution event is traced based on the traceability network, the traceability network can realize adaptive inference of each abnormal monitoring site by combining the water quality pollution knowledge graph, and the existing pollution water section positioning is changed to specific pollution source positioning, thereby improving the traceability accuracy, and even the complex lake water environment pollution can realize high-precision traceability.

[0037] Further, the establishment of the water quality pollution knowledge graph of the target river basin comprises:

[0038] Collecting water environment related literature, water quality historical records of the target river basin, surrounding environment information of the target river basin and pollution source information;

[0039] According to the water environment related literature, the water quality historical records of the target river basin, the surrounding environment information of the target river basin and the pollution source information, data extraction is performed to establish the water quality pollution knowledge graph of the target river basin.

[0040] From the above description, the water quality pollution knowledge graph is a total spectrum of knowledge that can affect water quality change, and by constructing the water quality pollution knowledge graph of the target river basin, subsequent accurate and efficient water quality pollution traceability can be realized.

[0041] Further, the data extraction according to the water environment related literature, the water quality historical records of the target river basin, the surrounding environment information of the target river basin and the pollution source information to establish the water quality pollution knowledge graph of the target river basin comprises:

[0042] The water environment related literature, the water quality historical records of the target river basin, the surrounding environment information of the target river basin and the pollution source information are preprocessed to obtain preprocessed data;

[0043] The preprocessed data is structured data extraction in the form of triplets to obtain triplet data;

[0044] The triplet data is knowledge fusion to obtain fused data;

[0045] The Neo4j graph database is used to construct the water quality pollution knowledge graph of the target river basin according to the fused data.

[0046] From the above description, since the collected data is complex and there may be multiple versions before and after, preprocessing can filter out data that conforms to local reality and helps to understand the water quality impact situation, and then data extraction and knowledge fusion are performed to make the established knowledge graph fully and truly reflect the factors affecting the water quality situation.

[0047] Further, the constructing a traceability network based on the water pollution knowledge graph, the monitoring sites and the upstream and downstream relationships comprises:

[0048] Building a basic network according to the monitoring sites and the upstream and downstream relationships;

[0049] According to the water pollution knowledge graph, the pollution path matching is performed on each monitoring site in the basic network, and the pollution mode of each monitoring site is obtained;

[0050] According to the pollution mode and the basic network, a traceability network is constructed.

[0051] From the above description, the pollution mode of each site is matched, and the traceability network is constructed based thereon, so that the pollution source can be accurately and quickly located during subsequent traceability, and reliable traceability is realized.

[0052] Further, the traceability of the water pollution event based on the traceability network comprises:

[0053] Obtaining the traceability network corresponding to the abnormal monitoring site;

[0054] Using the inflow diagnosis algorithm to determine the pollutant inflow amount of the abnormal monitoring site on each path according to the traceability network;

[0055] Determining the sum of the pollutant inflow amounts on the paths as the total inflow amount, and determining the proportion of the pollutant inflow amount on each path to the total inflow amount as the pollution index contribution rate on each path;

[0056] Collecting the actual pollution concentration of the abnormal monitoring site, and determining whether the difference between the total inflow amount and the actual pollution concentration is less than a first preset threshold, if yes, determining that the data of the abnormal monitoring site is highly matched, if not, determining that the data of the abnormal monitoring site is not matched;

[0057] For the abnormal monitoring site with highly matched data, the water pollution event is traced based on the pollution index contribution rate, and a traceability result is obtained;

[0058] For the abnormal monitoring site where the data does not match, it is judged whether the monitoring data of the abnormal monitoring site is abnormal, if yes, a data anomaly description is generated, and the data anomaly description is output, if not, unknown pollution source investigation is performed on the abnormal monitoring site to obtain a traceability result.

[0059] As can be seen from the above description, considering data errors and the complexity of the water environment, it is judged whether the data matches according to the difference between the total inflow amount and the actual pollution concentration. When the data is highly matched, it indicates that there is no external interference, and the water pollution event can be traced based on the pollution index contribution rate, that is, accurate pollutant tracing can be realized. When the data is not matched, unknown pollution source investigation is performed on the abnormal monitoring site to obtain a traceability result, thereby improving the comprehensiveness of water pollution tracing.

[0060] Further, the traceability of the water pollution event based on the pollution index contribution rate to obtain a traceability result comprises:

[0061] It is judged whether the pollution index contribution rate is greater than or equal to a second preset threshold, if yes, the path corresponding to the pollution index contribution rate is determined as a pollution path;

[0062] The pollution path is added to a candidate pollution source list;

[0063] The relative contribution rate of each pollution path in the candidate pollution source list is calculated according to the pollution index contribution rate;

[0064] According to the relative contribution rate, the pollution paths in the candidate pollution source list are sorted in descending order to obtain a sorted candidate pollution source list;

[0065] A main pollution source is determined from the sorted candidate pollution source list;

[0066] The basic information, excitation detail information and pollution impact range of the main pollution source are determined according to the traceability network;

[0067] A traceability result corresponding to the water pollution event is generated according to the basic information, excitation detail information and pollution impact range of the main pollution source.

[0068] As can be seen from the above description, in actual pollution analysis, all pollution sources or pollution sources with too low relative contribution rates do not need to be concerned. A main pollution source is determined from the sorted candidate pollution source list, the basic information, excitation detail information and pollution impact range of the main pollution source are determined according to the traceability network, and then a traceability result corresponding to the water pollution event is generated, thereby realizing efficient and reliable water pollution tracing.

[0069] Further, the unknown pollution source of the abnormal monitoring station is investigated to obtain a tracing result, which comprises:

[0070] A pollution concentration curve of a main monitoring index of the abnormal monitoring station is obtained, and a total inflow curve of each path is generated according to the total inflow amount;

[0071] The pollution concentration curve and the total inflow curve of each path are analyzed to obtain a first trend similarity;

[0072] Whether the current pollution mode of the abnormal monitoring station is reasonable or not is determined according to the first trend similarity and a third preset threshold;

[0073] A pollution concentration curve of another monitoring index of the abnormal monitoring station is obtained;

[0074] The pollution concentration curve of the other monitoring index and the pollution concentration curve are analyzed to obtain a second trend similarity;

[0075] The other monitoring index with the maximum second trend similarity is determined as a primary associated index;

[0076] If the result of whether the current pollution mode is reasonable or not is reasonable, and the primary associated index exists, it is determined that the investigation result is that the abnormality is caused by endogenous fluctuation;

[0077] If the result of whether the current pollution mode is reasonable or not is reasonable, and the primary associated index does not exist, it is determined that the investigation result is that the abnormality is caused by uneven distribution of water body concentration around the monitoring equipment;

[0078] If the result of whether the current pollution mode is reasonable or not is unreasonable, and the primary associated index exists, the primary associated index is analyzed to obtain the investigation result;

[0079] If the result of whether the current pollution mode is reasonable or not is unreasonable, and the primary associated index does not exist, it is determined that the investigation result is that the monitoring equipment needs to be paid attention to;

[0080] The tracing result is manually determined according to the investigation result.

[0081] As can be seen from the above description, when an interference factor intervenes to cause obvious abnormality, it will not only be reflected on the unique index. Therefore, the other monitoring index with the maximum second trend similarity is determined as the primary associated index, and then the investigation result is determined based on the result of whether the current pollution mode is reasonable or not and the primary associated index. The tracing result is manually determined according to the investigation result, unknown pollution source investigation is realized, and the comprehensiveness of water quality pollution tracing is improved.

[0082] Further, whether the current pollution mode of the abnormal monitoring station is reasonable or not is determined according to the first trend similarity and a third preset threshold, which comprises:

[0083] determining whether the first trend similarity is lower than a third preset threshold value, if yes, determining that the result of the rationality of the current pollution mode of the abnormal monitoring station is irrational, and if no, determining that the result of the rationality of the current pollution mode of the abnormal monitoring station is rational.

[0084] From the above description, it can be seen that the rationality of the current pollution mode is obtained by judging the trend similarity, which is beneficial to accurately investigate the unknown pollution source.

[0085] Further, after the water quality pollution event is traced based on the tracing network to obtain a tracing result, the method further comprises:

[0086] obtaining a disposal result corresponding to the tracing result;

[0087] determining a quantitative index corresponding to the disposal result;

[0088] adding the quantitative index to the water quality pollution knowledge graph to obtain an updated water quality pollution knowledge graph;

[0089] updating the tracing network according to the updated water quality pollution knowledge graph.

[0090] From the above description, it can be seen that after the water quality pollution event is disposed, the disposal result is obtained, the disposal result is quantified, and then is used to update the water quality pollution knowledge graph and the tracing network, the pollution mode is continuously explored and updated in the tracing process, the tracing network accuracy can be continuously improved, the tracing system can be improved, and the tracing accuracy can be improved.

[0091] Please refer to Figure 2 Another embodiment of the present application provides a water quality pollution tracing terminal, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements each step in the above-mentioned water quality pollution tracing method when executing the computer program.

[0092] The above-mentioned water quality pollution tracing method and terminal of the present application can be applied to simple water environment pollution and complex water environment pollution, and is particularly suitable for lake water environment pollution. The following will be described through specific embodiments:

[0093] Embodiment one

[0094] Please refer to Figure 1 , Figures 3-5 The water quality pollution tracing method of the present embodiment comprises the following steps:

[0095] S1, establishing a water quality pollution knowledge graph of a target river basin, specifically comprising:

[0096] S11, collect water environment related literature, water quality historical records of the target basin, surrounding environment information of the target basin, and pollution source information;

[0097] In an optional implementation, other data related to water pollution of the target basin can also be collected to further improve the comprehensiveness of the water pollution knowledge graph.

[0098] S12, data extraction is performed according to the water environment related literature, water quality historical records of the target basin, surrounding environment information of the target basin, and pollution source information, and a water pollution knowledge graph of the target basin is established, specifically including:

[0099] S121, the water environment related literature, water quality historical records of the target basin, surrounding environment information of the target basin, and pollution source information are preprocessed to obtain preprocessed data;

[0100] In an optional implementation, the preprocessing includes data screening and data cleaning.

[0101] S122, the preprocessed data is structured data extraction in the form of triples to obtain triple data;

[0102] The triple form includes entities, relationships, and attribute data, for example, the triple data can be domestic sewage - increase - total phosphorus or domestic sewage - form - direct discharge of pipeline.

[0103] In an optional implementation, when the data form and text expression are complex, manual structured data extraction can be performed to obtain triple data.

[0104] S123, knowledge fusion is performed on the triple data to obtain fused data;

[0105] Specifically, the triple data is knowledge fused, and descriptions of the same entity from multiple data sources are integrated or disambiguated to obtain fused data.

[0106] S124, a Neo4j graph database is used to construct a water pollution knowledge graph of the target basin according to the fused data, and the water pollution knowledge graph is a total spectrum of knowledge that can affect water quality changes, such as Figure 3 As shown in the figure, the inflow of the upstream station, domestic sewage discharge, direct inflow of rainfall, and agricultural breeding non-point source pollution caused by rainfall scouring all affect the water pollution index.

[0107] S2, the monitoring stations of the target basin and the upstream and downstream relationships of the monitoring stations are obtained, and a traceability network is constructed based on the water pollution knowledge graph, the monitoring stations, and the upstream and downstream relationships, specifically including:

[0108] S21, acquire monitoring sites of the target river basin and upstream and downstream relationships of the monitoring sites;

[0109] The upstream and downstream relationships are the inflow and outflow relationships between waterways, and the present application takes each monitoring site as an analysis unit to carry out pollution tracing.

[0110] S22, build a basic network according to the monitoring sites and the upstream and downstream relationships, such as the basic inflow and outflow network of monitoring sites A, B and C in the figure;

[0111] S23, match a pollution path of each monitoring site in the basic network according to the water quality pollution knowledge graph, to obtain a pollution mode of each monitoring site;

[0112] In an optional implementation, the pollution path of each monitoring site in the basic network is matched according to the water quality pollution knowledge graph in combination with field investigation information, to obtain the pollution mode of each monitoring site, so as to improve the matching accuracy.

[0113] As shown in the figure, the pollution mode of site C is mainly influenced by upstream inflow (site B), rainfall scouring inflow and residential area sewage inflow, among which domestic sewage is greatly affected by traditional customs and tourism, and is listed as a consideration factor in the future. Figure 4

[0114] S24, build a tracing network according to the pollution mode and the basic network.

[0115] S3, acquire a pollutant index of each monitoring site, and determine whether the pollutant index exceeds a preset concentration, if yes, determine the monitoring site as an abnormal monitoring site, and generate a water quality pollution event corresponding to the abnormal monitoring site;

[0116] S4, trace the water quality pollution event based on the tracing network, to obtain a tracing result, specifically including:

[0117] S41, acquire a tracing network corresponding to the abnormal monitoring site;

[0118] Specifically, the tracing network at the moment when the abnormal monitoring site occurs the water quality pollution event is acquired.

[0119] ​In an optional embodiment, the monitoring data corresponding to the time when the water pollution event occurs at the abnormal monitoring site is obtained simultaneously, including monitoring site data and plant discharge data, etc. It is also necessary to consider the lag effect between network nodes. For example, the current abnormal monitoring site C obtains monitoring data from 13:00 to 15:00. For monitoring site B, the lag time is obtained by dividing the waterway distance between B and C by the average water flow speed. Assuming that the lag time is 2 hours, monitoring site B obtains monitoring data from 15:00 to 17:00.

[0120] S42, determining the amount of pollutant influx of the abnormal monitoring site on each path according to the tracing network using a influx diagnosis algorithm;

[0121] The influx diagnosis algorithm includes a hydrodynamic model, time series trend matching, fuzzy decision, machine learning, or deep learning, etc.

[0122] In an optional embodiment, the amount of pollutant influx of the abnormal monitoring site on each path is determined according to the tracing network using a influx diagnosis algorithm based on the monitoring data.

[0123] S43, determining the sum of the amount of pollutant influx on each path as the total influx, and determining the proportion of the amount of pollutant influx on each path to the total influx as the pollution indicator contribution rate on each path.

[0124] As shown in FIG. 4, for example, monitoring site B needs to calculate the total influx of the amount of pollutant influx of site A and the sewage treatment plant. Figure 4

[0125] S44, collecting the actual pollution concentration of the abnormal monitoring site, and determining whether the difference between the total influx and the actual pollution concentration is less than a first preset threshold value. If yes, it is determined that the data of the abnormal monitoring site is highly matched. If no, it is determined that the data of the abnormal monitoring site is not matched. In this way, the accuracy of tracing is ensured by considering the data error and the complexity of the water environment.

[0126] In an optional embodiment, the first preset threshold value is determined according to historical data rules, and the degree of pollution source refinement. The lower the threshold value, the more influencing factors need to be considered.

[0127] S45, for the abnormal monitoring site with highly matched data, tracing the water pollution event based on the pollution indicator contribution rate to obtain a tracing result, specifically including:

[0128] S451, determining whether the pollution indicator contribution rate is greater than or equal to a second preset threshold value. If yes, the path corresponding to the pollution indicator contribution rate is determined as the pollution path.​

[0129] In an optional implementation, the second preset threshold is set according to historical pollution events and actual needs.

[0130] In an optional implementation, if the path corresponding to the pollution index contribution rate is a confluence of an upstream node, the complete pollution path is obtained by tracing upwards.

[0131] For example, assuming that the second preset threshold is 20%, the pollution of the abnormal monitoring site C is traced, all paths with a contribution rate of 20% are traced, and the town domestic sewage is traced upwards to the folk festival and daily life; the upstream site B needs to be traced upwards further, the sewage treatment plant is taken as the complete pollution path, and the site A is traced to obtain the complete pollution path of the reservoir.

[0132] S452, the pollution path is added to a candidate pollution source list;

[0133] S453, the relative contribution rate of each pollution path in the candidate pollution source list is calculated according to the pollution index contribution rate.

[0134] Specifically, the pollution index contribution rates of each pollution path are multiplied to obtain the relative contribution rate of each pollution path.

[0135] For example, as shown in FIG. 5, the relative contribution rate of the pipe network leakage is the product of the pollution index contribution rates of all paths between the sewage treatment plant and the monitoring site C, that is, 100% × 80% × 65% = 52%, and the relative contribution rate of the folk festival is 24%, the relative contribution rate of the reservoir is 13%, and the relative contribution rate of daily life is 6% by analogy. Figure 5

[0136] S454, the pollution paths in the candidate pollution source list are sorted in descending order according to the relative contribution rates to obtain a sorted candidate pollution source list.

[0137] S455, a main pollution source is determined from the sorted candidate pollution source list.

[0138] In an optional implementation, a preset number of pollution paths are selected from the sorted candidate pollution source list as the main pollution source, and the preset number is set according to actual conditions.

[0139] In another optional implementation, a pollution path with a relative contribution rate reaching a fourth preset threshold is selected from the sorted candidate pollution source list as the main pollution source, and the fourth preset threshold is set according to actual conditions.

[0140] ​S456, determining basic information, triggering detail information and pollution influence range of the main pollution source according to the traceability network;

[0141] In an optional implementation, the basic information includes pollution position and pollution amount, etc., refers to first-level element analysis of the monitoring site, i.e., elements directly flowing into the node, such as the sewage treatment plant and urban domestic sewage directly connected with the monitoring site B and C as shown in the figure.

[0142] The triggering detail information, i.e., the cause of the pollution source, includes natural factors or human factors, etc., refers to the analysis of the upper nodes of the first-level elements, i.e., all upper elements affecting the first-level elements, such as the urban domestic sewage being affected by daily life, folk festivals and tourist peak seasons; the sewage treatment plant being affected by equipment shutdown, pipe network leakage, etc.

[0143] The pollution influence range includes pollution transmission path and station pollution index increase, etc.

[0144] S457, generating traceability results corresponding to the water pollution event according to the basic information, triggering detail information and pollution influence range of the main pollution source.

[0145] In an optional implementation, the traceability results corresponding to the water pollution event are displayed in the form of a pollution event report.

[0146] S46, for the abnormal monitoring site with unmatched data, determining whether the monitoring data of the abnormal monitoring site is abnormal, if yes, performing S461, and if not, performing S462.

[0147] The abnormality includes data missing or invalid value, etc.

[0148] S461, generating data abnormality description and outputting the data abnormality description.

[0149] S462, performing unknown pollution source investigation on the abnormal monitoring site to obtain traceability results, specifically including:

[0150] S4621, obtaining pollution concentration curves of main monitoring indexes of the abnormal monitoring site, and generating total inflow curves of each path according to the total inflow;

[0151] The main monitoring index refers to an abnormal index with pollution.

[0152] S4622, performing similarity analysis on the pollution concentration curves and the total inflow curves of each path to obtain a first trend similarity;

[0153] In an alternative embodiment, the pollution concentration curve and the respective path total inflow curve are analyzed for similarity using a Pearson correlation coefficient algorithm to obtain a first trend similarity.

[0154] For example, as shown in FIG. 6, assuming that a total phosphorus pollution event occurs at monitoring site C, the total inflow curves of the three pollution paths (rainfall, surface runoff, and urban domestic sewage) of site C are summed up at each time point to obtain a total inflow curve, which is then analyzed for similarity with the total phosphorus pollution concentration curve of site C. Figure 5

[0155] S4623, determining whether the current pollution mode of the abnormal monitoring site is reasonable or not according to the first trend similarity and a third preset threshold value;

[0156] Specifically, it is determined whether the first trend similarity is lower than the third preset threshold value. If yes, it is determined that the current pollution mode of the abnormal monitoring site is unreasonable. If no, it is determined that the current pollution mode of the abnormal monitoring site is reasonable.

[0157] S4624, obtaining a pollution concentration curve of other monitoring indicators of the abnormal monitoring site;

[0158] The other monitoring indicators are monitoring indicators other than the main monitoring indicators.

[0159] S4625, analyzing the pollution concentration curve of the other monitoring indicators for similarity with the pollution concentration curve to obtain a second trend similarity;

[0160] For example, as shown in FIG. 6, assuming that a total phosphorus pollution event occurs at monitoring site C, the total inflow curves of the three pollution paths (rainfall, surface runoff, and urban domestic sewage) of site C are summed up at each time point to obtain a total inflow curve, which is then analyzed for similarity with the total phosphorus pollution concentration curve of site C. Figure 5

[0161] S4626, determining the other monitoring indicator with the largest second trend similarity as a primary associated indicator;

[0162] S4627, if the result of whether the current pollution mode is reasonable or not is reasonable and the primary associated indicator exists, determining that the investigation result is an endogenous fluctuation impact;

[0163] After the staff understands the investigation result, the staff analyzes the cause in combination with the water quality pollution knowledge graph.

[0164] S4628, if the result of whether the current pollution mode is reasonable or not is reasonable and the primary associated indicator does not exist, determining that the investigation result is an abnormal cause that needs to be concerned about the uneven distribution of water body concentration around the monitoring equipment;

[0165] ​​S4629, if the reasonableness result is unreasonable and the primary correlation index exists, analyzing the primary correlation index to obtain an investigation result;

[0166] The method for analyzing the primary correlation index is the same as S41-S46, and will not be described here.

[0167] S46210, if the reasonableness result is unreasonable and the primary correlation index does not exist, determining that the investigation result is that the running state of the monitoring equipment needs to be paid attention to;

[0168] S46211, manually determining a traceability result according to the investigation result.

[0169] Specifically, according to the investigation result, a staff member combines a basin environment and a correlation index, consults literature, consults experts and scholars, analyzes whether there is a pollution mode missing, lists factors that may cause current site pollution, conducts on-site visits and inquiries, and investigates suspicious pollution sources on site. When conditions permit, samples can be taken to assist in chemical analysis. For occasional non-natural sources such as residents' illegal dumping and illegal discharge, disposal registration is performed according to regulations. For regular and condition-triggered natural sources, further tracing is required.

[0170] After finding a candidate pollution source, data demonstration needs to be further performed in combination with related theories. For closed and constrained pollution sources such as drainage outlets, temporary monitoring instruments are deployed. For open pollution sources such as construction, sampling analysis or on-site testing is required. Based on data, the pollution amount and pollution value fluctuation trend are matched to demonstrate the correctness of the pollution path.

[0171] In an optional implementation, after manually determining a traceability result, if data demonstration is reasonable, the pollution source is locked, and is updated into the water quality pollution knowledge graph. Quantifiable factors can be directly added to the knowledge graph. Non-quantifiable factors need to be traced back to quantifiable factors. For example, when evaluating the pollution caused by the inflow of farmland fertilizers, it is necessary to trace back one level, such as rainfall, field height, and crop type. The intelligent algorithm is a commonly used multiple regression method.

[0172] S5, obtaining a disposal result corresponding to the traceability result;

[0173] S6, determining a quantifiable index corresponding to the disposal result;

[0174] Specifically, if the disposal result includes a series of prevention and control measures such as deploying monitoring sites, video monitoring, and sewage treatment, it is refined into a quantifiable index.

[0175] In an alternative embodiment, for the presence of pollutant monitoring values, direct addition is added, for the absence of pollutant monitoring values, it is necessary to refine into the number of events, such as 1 times of wastewater dumping, etc., and the difference between the inflow total amount and the actual pollution concentration is taken as the initial weight thereof, and subsequent occurrence of the same event can continuously update the event weight.

[0176] S7, adding the quantitative index to the water quality pollution knowledge graph to obtain an updated water quality pollution knowledge graph;

[0177] From the pollution mode, the pollution traceability can be focused on the pain points and key points of each water section, and the same event can be avoided from repeated investigation. From the pollution traceability, the traceability accuracy can be gradually increased, and with the increase of the quantitative index, the first preset threshold value will be allowed to be lower and lower, meeting the fine management demand.

[0178] S8, updating the traceability network according to the updated water quality pollution knowledge graph.

[0179] Specifically, the traceability network matches a new pollution mode from the updated water quality pollution knowledge graph to obtain a new traceability network, realizes dynamic growth, improves the traceability network accuracy, and perfects the traceability system.

[0180] Embodiment two

[0181] Please refer to Figure 2 The water quality pollution traceability terminal of the embodiment comprises a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the processor realizes each step in the water quality pollution traceability method in embodiment one when executing the computer program.

[0182] In summary, the water quality pollution traceability method and terminal provided by the application establish a water quality pollution knowledge graph of a target river basin, construct a traceability network based on the water quality pollution knowledge graph, monitoring sites and upstream and downstream relationships, obtain the pollutant index of each monitoring site, determine the monitoring site as an abnormal monitoring site when the pollutant index exceeds a preset concentration, and generate a water quality pollution event corresponding to the abnormal monitoring site. The water quality pollution event is traced based on the traceability network, the traceability network can realize adaptive inference of each abnormal monitoring site by combining the water quality pollution knowledge graph, the specific pollution source positioning is changed from the existing pollution water section positioning, and thus the traceability accuracy is improved. Even complex lake water environment pollution can realize high-precision traceability. In addition, the pollution mode of each site is matched, and the traceability network is constructed based thereon, so that the pollution source can be accurately and quickly positioned during subsequent traceability, and reliable traceability is realized.

[0183] The above merely illustrates the embodiments of the present application, and is not intended to limit the patent scope of the present application. Any equivalent transformation or direct or indirect application in the related technical field based on the content of the present application specification and drawings is also included in the patent protection scope of the present application.

Claims

1. A method for tracing the source of water pollution, characterized in that: Including steps: Establish a water pollution knowledge map for the target watershed; Obtaining the monitoring sites of the target watershed and the upstream and downstream relationships of the monitoring sites, and constructing a traceability network based on the water pollution knowledge graph, the monitoring sites, and the upstream and downstream relationships; Obtaining pollutant indicators at each monitoring site and determining whether the pollutant indicators exceed a preset concentration; if so, determining the monitoring site as an abnormal monitoring site and generating a water pollution event corresponding to the abnormal monitoring site; Tracing the water pollution incident based on the tracing network to obtain a tracing result; The tracing of the water pollution incident based on the tracing network to obtain the tracing result includes: Obtaining the traceability network corresponding to the abnormal monitoring site; Determine the pollutant inflow amount of the abnormal monitoring site on each path according to the traceability network using an inflow diagnosis algorithm; The sum of the pollutant inflows on the various paths is determined as the total inflow amount, and the ratio of the pollutant inflow on each path to the total inflow amount is determined as the pollution index contribution rate on each path; Collecting the actual pollution concentration of the abnormal monitoring site, and determining whether the difference between the total inflow amount and the actual pollution concentration is less than a first preset threshold, if so, determining that the data of the abnormal monitoring site is highly matched; if not, determining that the data of the abnormal monitoring site is not matched; For the abnormal monitoring site whose data does not match, determine whether the monitoring data of the abnormal monitoring site is abnormal. If so, generate a data abnormality description and output the data abnormality description. If not, conduct an unknown pollution source investigation on the abnormal monitoring site to obtain a traceability result. The unknown pollution source investigation of the abnormal monitoring site is performed to obtain the traceability results including: Obtaining pollution concentration curves of main monitoring indicators of the abnormal monitoring site, and generating total inflow curves for each path based on the total inflow; Performing similarity analysis on the pollution concentration curve and the total inflow curve of each path to obtain a first trend similarity; Determining whether the current pollution pattern of the abnormal monitoring site is reasonable or not based on the first trend similarity and a third preset threshold; Obtaining pollution concentration curves of other monitoring indicators at the abnormal monitoring site; Performing similarity analysis on the pollution concentration curve of the other monitoring indicators and the pollution concentration curve to obtain a second trend similarity; Determine the other monitoring indicators with the greatest similarity to the second trend as primary correlation indicators; If the rationality result is reasonable and the primary correlation indicator exists, then the screening result is determined to be affected by endogenous fluctuations; If the reasonableness result is reasonable and the primary correlation indicator does not exist, then the investigation result is determined to be the abnormal cause of uneven concentration distribution in the water body around the monitoring equipment, which requires attention; If the rationality result is unreasonable and the primary correlation indicator exists, the primary correlation indicator is analyzed to obtain a screening result; If the rationality result is unreasonable and the primary related indicator does not exist, then the troubleshooting result is determined to require attention to the operating status of the monitoring equipment; The traceability results are manually determined based on the investigation results.

2. A water pollution source tracing method according to claim 1, characterized in that: The establishment of a water quality pollution knowledge graph for the target watershed includes: Collect relevant literature on water environment, historical records of water quality in the target watershed, information on the surrounding environment of the target watershed, and information on pollution sources; Data is extracted based on the water environment-related literature, the water quality history records of the target river basin, the surrounding environment information of the target river basin, and the pollution source information to establish a water quality pollution knowledge map for the target river basin.

3. A water pollution source tracing method according to claim 2, characterized in that: The data extraction based on the water environment related literature, the water quality history records of the target river basin, the surrounding environment information of the target river basin and the pollution source information to establish the water quality pollution knowledge map of the target river basin includes: Preprocessing the water environment related literature, the water quality history records of the target watershed, the surrounding environment information of the target watershed, and the pollution source information to obtain preprocessed data; Performing structured data extraction on the preprocessed data in triple form to obtain triple data; Performing knowledge fusion on the triple data to obtain fused data; A Neo4j graph database is used to construct a water quality pollution knowledge graph of the target watershed based on the fused data.

4. A water pollution source tracing method according to claim 1, characterized in that: The constructing of a traceability network based on the water pollution knowledge graph, the monitoring sites, and the upstream and downstream relationships includes: Building a basic network based on the monitoring sites and the upstream and downstream relationships; Matching pollution paths for each monitoring site in the basic network according to the water pollution knowledge graph to obtain pollution patterns for each monitoring site; A traceability network is constructed according to the pollution pattern and the basic network.

5. A water pollution source tracing method according to claim 1, characterized in that: Also includes: For the abnormal monitoring sites with highly matched data, the water pollution incident is traced to its source based on the contribution rate of the pollution index to obtain a tracing result.

6. A method for tracing the source of water pollution according to claim 5, characterized in that: The tracing of the water pollution incident based on the pollution index contribution rate to obtain the tracing result includes: Determining whether the pollution index contribution rate is greater than or equal to a second preset threshold, and if so, determining the path corresponding to the pollution index contribution rate as a polluted path; adding the pollution path to a candidate pollution source list; Calculate the relative contribution rate of each pollution path in the candidate pollution source list according to the pollution index contribution rate; sorting the pollution paths in the candidate pollution source list in descending order according to the relative contribution rates to obtain a sorted candidate pollution source list; determining a major pollution source from the ranked list of candidate pollution sources; Determine the basic information, triggering details and pollution impact range of the main pollution sources based on the traceability network; The source tracing result corresponding to the water pollution incident is generated based on the basic information of the main pollution source, the triggering detailed information and the pollution impact range.

7. A method for tracing the source of water pollution according to claim 1, characterized in that: The determination of whether the current pollution pattern of the abnormal monitoring site is reasonable or not based on the first trend similarity and the third preset threshold value includes: Determine whether the first trend similarity is lower than a third preset threshold. If so, determine whether the rationality of the current pollution mode of the abnormal monitoring site is unreasonable. If not, determine whether the rationality of the current pollution mode of the abnormal monitoring site is reasonable.

8. A method for tracing the source of water pollution according to claim 1, characterized in that: The tracing of the water pollution incident based on the tracing network and obtaining the tracing result includes: Obtaining a disposal result corresponding to the traceability result; Determining quantitative indicators corresponding to the disposal results; Adding the quantitative indicators to the water pollution knowledge graph to obtain an updated water pollution knowledge graph; The traceability network is updated according to the updated water pollution knowledge graph.

9. A water pollution tracing terminal, 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 computer program, each step of the water pollution source tracing method described in any one of claims 1 to 8 is implemented.

Images