A water pollution source tracing analysis method and system based on multidimensional data fusion analysis

Abstract

This invention relates to the field of water pollution source tracing technology, and proposes a water pollution source tracing analysis method and system based on multi-dimensional data fusion analysis. The method includes: constructing a monitoring network based on the discharge paths of rainwater and sewage to acquire real-time monitoring data; determining whether the real-time monitoring data is abnormal; if an anomaly occurs, issuing an alarm for the monitoring node with the anomaly; upon receiving the alarm information, performing a fusion analysis of the pollution sources of rainwater and sewage based on a pollution source tracing model based on discharge characteristics analysis and a pollution source tracing model based on three-dimensional fluorescence; positively verifying the results of the fusion analysis through a verification model to obtain the source tracing analysis results; conducting on-site source tracing and evidence collection based on the source tracing analysis results; and predicting the impact of the alarm points based on a rainwater management model and taking emergency measures. This technical solution solves the problems of poor accuracy in discharge source tracing, difficulty in obtaining discharge evidence, and untimely handling of sudden pollution events in existing technologies.

Description

Technical Field

[0001] The present invention relates to the technical field of water pollution source tracing, and in particular to a water pollution source tracing analysis method and system based on multidimensional data fusion analysis. Background Art

[0002] The main methods of tracing the source of water pollution at present include three-dimensional fluorescence atlas method, isotope method, etc. The main problems of the above-mentioned water pollution tracing technologies are: since it is impossible to find out the characteristics of abnormal pollution discharge time and discharge law, the actual evidence collection still requires personnel to squat for a long time to collect pollution discharge evidence after mastering the results of three-dimensional fluorescence atlas tracing and isotope tracing, which is difficult to help efficient environmental supervision and evidence collection.

[0003] A single three-dimensional fluorescence atlas method or isotope method can only obtain suspected pollution sources. Since it is impossible to understand the time and pattern of abnormal pollution discharge, the tracing effect is higher in the field of industrial wastewater. When domestic sewage or water bodies lack three-dimensional fluorescence atlases and have no obvious characteristics, the tracing accuracy is lower.

[0004] The pollution source tracing model based on pollution emission characteristic analysis is mainly applicable to pollution source tracing scenarios with certain pollution emission patterns, and its accuracy is poor for pollution source tracing without obvious pollution emission patterns.

[0005] The discharge path of rainwater and sewage is "pollution source-drainage network-river (lake) / sewage treatment plant". When pollution occurs at the pollution source or drainage network, it is difficult to predict the impact of pollutants on the receiving water terminal, resulting in untimely water pollution treatment, affecting the healthy operation of the sewage treatment process of the sewage treatment plant and affecting the stable compliance of river water bodies. Summary of the invention

[0006] The present invention proposes a water pollution source tracing analysis method and system based on multi-dimensional data fusion analysis, which solves the problems that the single three-dimensional fluorescence spectrum method and isotope method in the related technology have low tracing accuracy when the three-dimensional fluorescence spectrum of domestic sewage or water body has no obvious characteristics; the pollution source tracing model based on pollution discharge characteristic analysis has poor accuracy in tracing the pollution discharge without obvious pollution discharge rules; and the problem of difficulty in obtaining evidence due to poor tracing accuracy.

[0007] The technical solution of the present invention is as follows:

[0008] The present invention proposes a water pollution source tracing analysis method based on multidimensional data fusion analysis, comprising:

[0009] S1. Construct a monitoring network based on the discharge path of rainwater and sewage, and obtain real-time monitoring data of rainwater and sewage through the monitoring network;

[0010] S2, determine whether the real-time monitoring data is abnormal; if the real-time monitoring data is abnormal, an alarm is issued to the monitoring node where the abnormality occurs; otherwise, continue to execute S2; wherein the monitoring node that issues an alarm is the alarm point;

[0011] S3. After receiving the alarm information, the pollution source of rainwater and sewage is analyzed by fusion according to the pollution source tracing model based on the pollution characteristics analysis and the pollution source tracing model based on three-dimensional fluorescence, and the result of the fusion analysis is positively verified by the accounting verification model to obtain the tracing analysis result. At the same time, the impact of the alarm point on the downstream is predicted according to the rainwater management model and emergency measures are taken;

[0012] S4. Conduct on-site traceability and evidence collection based on the traceability analysis results.

[0013] Furthermore, in S1, a monitoring network is constructed according to the discharge path of rainwater and sewage, and real-time monitoring data of rainwater and sewage are obtained through the monitoring network, including:

[0014] S11. Construct the spatial location topology of the watershed based on the discharge path of rainwater and sewage;

[0015] S12. Construct a monitoring network based on the spatial topological relationship of the watershed, and deploy water quality monitoring equipment, water quantity monitoring equipment and three-dimensional fluorescence monitoring equipment at each monitoring node based on the monitoring network.

[0016] Furthermore, in S3, after receiving the alarm information, the following steps are also included:

[0017] Manually review whether the alarm information is true; if the alarm information is true, continue to execute subsequent steps; otherwise, mark the alarm information as a false alarm, stop executing subsequent steps, and return to S2.

[0018] Furthermore, in S3, after receiving the alarm information, the pollution source of rainwater and sewage is analyzed by fusion according to the pollution source tracing model based on the pollution characteristics analysis and the pollution source tracing model based on three-dimensional fluorescence, and the results of the fusion analysis are positively verified by the accounting verification model, and the tracing analysis results include:

[0019] S31. Analyze the real-time monitoring data of the alarm points according to the pollution source tracing model based on the pollution discharge characteristics analysis to obtain the first group of suspected pollution sources;

[0020] S32. Analyze the polluted water samples at the warning points according to the pollution source tracing model based on three-dimensional fluorescence to obtain a second group of suspected pollution sources;

[0021] S33, performing weighted calculation based on the first group of suspected pollution sources and their corresponding traceability similarities, and the second group of suspected pollution sources and their corresponding traceability similarities, to obtain the corrected traceability similarities of the first group of suspected pollution sources and the second group of suspected pollution sources;

[0022] S34. According to the first group of suspected pollution sources, the second group of suspected pollution sources, the corrected source tracing similarity, and the suspected pollution discharge time, based on the spatial position topological relationship of the watershed, a calculation verification model is constructed, and a forward simulation is performed on the first group of suspected pollution sources and the second group of suspected pollution sources to obtain the predicted time when the pollution discharge behavior of each suspected pollution source reaches the alarm point at the suspected pollution discharge time, and a weighted calculation is performed according to the predicted time and the alarm time of the alarm point to obtain the final source tracing similarity of the first group of suspected pollution sources and the second group of suspected pollution sources;

[0023] S35. Obtain the source tracing analysis result according to the final source tracing similarity, the first group of suspected pollution sources, the second group of suspected pollution sources and the suspected pollution discharge time.

[0024] Furthermore, in S31, according to the pollution source tracing model based on the pollution emission characteristic analysis, the real-time monitoring data of the alarm point is analyzed to obtain the first group of suspected pollution sources, which specifically includes:

[0025] S311, according to the pollution source tracing model based on the pollution characteristics analysis, feature extraction is performed on the real-time monitoring data of each monitoring node in the monitoring network, and a pollution characteristics portrait of each monitoring node is constructed according to the extracted features;

[0026] S312, calculating the similarity between the pollution discharge characteristic portrait of the alarm point and the pollution discharge characteristic portrait of each monitoring node belonging to the upstream according to the Mahalanobis distance;

[0027] S313, determine the suspected pollution source PY based on similarity i , traceability similarity PS i , suspected sewage discharge time PT i ; Where i represents the i-th suspected pollution source;

[0028] S314, traceability similarity PS i Sort from large to small and take the traceability similarity PS i The largest top n suspected pollution sources PY i As the first group of suspected pollution sources.

[0029] Furthermore, in S32, the polluted water samples at the alarm points are analyzed according to the pollution source tracing model based on three-dimensional fluorescence to obtain a second group of suspected pollution sources, including:

[0030] S321. Establish a three-dimensional fluorescence image database of watershed pollution sources based on a pollution source tracing model based on three-dimensional fluorescence;

[0031] S322, collecting polluted water samples at the alarm points, and analyzing the three-dimensional fluorescence spectrum of the polluted water samples;

[0032] S323, based on the convolutional neural network algorithm, the three-dimensional fluorescence spectrum of the polluted water sample is compared with the three-dimensional fluorescence image database of the pollution source in the watershed to obtain the suspected pollution source YY i , traceability similarity YS i ;

[0033] S324, traceability similarity YY i Sort from large to small and take the traceability similarity YS i The largest n suspected pollution sources YY i As the second group of suspected pollution sources.

[0034] Further, in S33, a weighted calculation is performed based on the first group of suspected pollution sources and their corresponding traceability similarities, and the second group of suspected pollution sources and their corresponding traceability similarities, to obtain the corrected traceability similarities of the first group of suspected pollution sources and the second group of suspected pollution sources, including:

[0035] If a suspected pollution source exists in both the first group of suspected pollution sources and the second group of suspected pollution sources, the weighted calculation formula for the traceability similarity of the suspected pollution source is as follows:

[0036] Y i =(PS i +YS i )×(k1 / 2);

[0037] Among them, Y i represents the corrected traceability similarity, k1 represents the first enhancement coefficient, and the value range of k1 is [1, 1.5];

[0038] If a suspected pollution source only exists in the first group of suspected pollution sources or the second group of suspected pollution sources, the weighted calculation formula for the traceability similarity of the suspected pollution source is as follows:

[0039] Y i =MAX(PS i , YS i )×m1;

[0040] Wherein, m1 represents the first weakening coefficient, and the value range of m1 is [0.5, 1].

[0041] Furthermore, in S34, a weighted calculation is performed based on the predicted time and the alarm time of the alarm point to obtain the final traceability similarity between the first group of suspected pollution sources and the second group of suspected pollution sources, including:

[0042] If the predicted time of a suspected pollution source is consistent with the alarm time of the alarm point, the calculation formula for the final traceability similarity of the suspected pollution source is as follows:

[0043] Y i '=Y i×k2;

[0044] Among them, Y i ' represents the final traceability similarity, k2 represents the second enhancement coefficient, and the value range of k2 is [1, 1.5];

[0045] If the predicted time of a suspected pollution source is inconsistent with the alarm time of the alarm point, the calculation formula for the final traceability similarity of the suspected pollution source is as follows:

[0046] Y i '=Y i ×m2;

[0047] Wherein, m2 represents the second weakening coefficient, and the value range of m2 is [0.5, 1].

[0048] Furthermore, in S3, the impact of the warning point on the downstream is predicted based on the rainwater management model and emergency measures are taken, including:

[0049] S3.1. According to the rainwater management model, simulate and predict the concentration and time of pollutants from the alarm point to the downstream receiving terminal; when it is predicted that the pollutants flow through the sewage pipe network, the receiving terminal is the sewage treatment plant, and enter S3.2; when it is predicted that the pollutants flow through the rainwater pipe network, the receiving terminal is the river outlet, and enter S3.3;

[0050] S3.2. Based on the predicted concentration and time of pollutants arriving at the sewage treatment plant, as well as the influent water quality requirements of the sewage treatment plant, assess whether the predicted pollutants will affect the biological treatment effect of the sewage treatment process; if the predicted pollutants will affect the biological treatment effect of the sewage treatment process, take emergency measures in advance in the regulating tank of the sewage treatment plant;

[0051] S3.3. Based on the predicted concentration and time of pollutants reaching the river outlet, as well as the water quality requirements of the receiving river, assess whether the predicted pollutants will affect the river water quality standards; if the predicted pollutants will affect the river water quality standards, take emergency measures in advance to intercept the predicted pollutants and prevent them from entering the river.

[0052] The present invention also proposes a water pollution source tracing analysis system based on multidimensional data fusion analysis, which is used to implement the water pollution source tracing analysis method based on multidimensional data fusion analysis as mentioned above. The system includes:

[0053] The data module is used to construct a monitoring network according to the discharge path of rainwater and sewage, and obtain real-time monitoring data of rainwater and sewage through the monitoring network;

[0054] The alarm module is connected to the data module and is used to determine whether the real-time monitoring data is abnormal; if the real-time monitoring data is abnormal, an alarm is issued to the monitoring node where the abnormality occurs; otherwise, the real-time monitoring data continues to be determined whether it is abnormal;

[0055] The source tracing analysis module is connected to the alarm module. After receiving the alarm information, it is used to conduct a fusion analysis of the pollution sources of rainwater and sewage based on the pollution source tracing model based on the pollution discharge characteristics analysis and the pollution source tracing model based on three-dimensional fluorescence, and to positively verify the results of the fusion analysis through the accounting verification model to obtain the tracing analysis results. At the same time, according to the rainwater management model, it predicts the impact of the alarm point on the downstream and takes emergency measures;

[0056] The information notification module is connected to the traceability analysis module and is used to conduct on-site traceability and evidence collection based on the traceability analysis results.

[0057] The working principle and beneficial effects of the present invention are:

[0058] The present invention performs a fusion analysis on the pollution sources of rainwater and sewage according to a pollution source tracing model based on pollution discharge characteristic analysis and a pollution source tracing model based on three-dimensional fluorescence, thereby solving the problem of insufficient accuracy of a single tracing method; by constructing a monitoring network according to the discharge path of rainwater and sewage, and constructing an accounting verification model according to the discharge path of rainwater and sewage, and positively verifying the results of the fusion analysis through the accounting verification model, it is possible to comprehensively monitor the discharge path of rainwater and sewage, thereby solving the problem of large limitations on the applicable scenarios of water pollution source tracing methods, improving the tracing accuracy, and expanding the scope of application of water pollution source tracing analysis methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0060] Figure 1 A flow chart of a water pollution source tracing analysis method based on multidimensional data fusion analysis provided in Example 1 of the present invention;

[0061] Figure 2 A flow chart of obtaining a first group of suspected pollution sources according to a pollution source tracing model based on pollution discharge characteristic analysis provided in Example 1 of the present invention;

[0062] Figure 3 A flow chart of obtaining a second group of suspected pollution sources according to a pollution source tracing model based on three-dimensional fluorescence provided in Example 1 of the present invention;

[0063] Figure 4 A flowchart of pollution prediction, early warning and emergency linkage provided by Example 1 of the present invention;

[0064] Figure 5A logic diagram of a water pollution source tracing analysis method based on multidimensional data fusion analysis provided in Example 1 of the present invention;

[0065] Figure 6 A schematic diagram of the structure of a water pollution source tracing analysis system based on multi-dimensional data fusion analysis provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0067] Example 1

[0068] like Figure 1~Figure 5 As shown, this embodiment proposes a water pollution source tracing analysis method based on multi-dimensional data fusion analysis.

[0069] In this embodiment, a water pollution source tracing analysis method based on multi-dimensional data fusion analysis includes:

[0070] S1. Construct a monitoring network based on the discharge path of rainwater and sewage, and obtain real-time monitoring data of rainwater and sewage through the monitoring network.

[0071] S11. Construct the spatial location topological relationship of the watershed based on the discharge paths of rainwater and sewage.

[0072] Specifically, according to the discharge path of rainwater and sewage, the relationship between the upstream and downstream nodes of the basin's "pollution source-drainage network-river (lake) / sewage treatment plant" (abbreviated as "source-network-plant-river") is clarified, and the spatial position topological relationship of the basin is constructed.

[0073] S12. Construct a monitoring network based on the spatial topological relationship of the watershed, and deploy water quality monitoring equipment, water quantity monitoring equipment and three-dimensional fluorescence monitoring equipment at each monitoring node based on the monitoring network.

[0074] Specifically, a monitoring network is constructed based on the topological relationship of the spatial position of the watershed, and water quality monitoring equipment and water volume monitoring equipment are deployed at important pollution sources, drainage pipe networks, and river sections; three-dimensional fluorescence monitoring equipment is deployed at river sections, sewage treatment plant inlets, or important drainage pipe network nodes, and polluted water samples can be collected during source tracing analysis to analyze the three-dimensional fluorescence water quality map of the water body in the laboratory. Among them, important pollution sources can be determined based on the level of pollution concentration or the amount of sewage discharged; drainage pipe network nodes are the nodes where the main trunk and branch trunks in the drainage pipe network meet and converge, as well as the ends of the pipe network.

[0075] S2. Determine whether the real-time monitoring data is abnormal.

[0076] Specifically, if the real-time monitoring data is abnormal, an alarm is issued to the monitoring node where the abnormality occurs; otherwise, S2 is continued. The real-time monitoring data may include information such as water quality, water volume, and time; the abnormality may include that the real-time monitoring data exceeds a preset threshold or that the real-time monitoring data suddenly changes. The monitoring node that issues an alarm is the alarm point.

[0077] S3. After receiving the alarm information, the pollution sources of rainwater and sewage are fused and analyzed according to the pollution source tracing model based on pollution discharge characteristics analysis and the pollution source tracing model based on three-dimensional fluorescence, and the results of the fusion analysis are positively verified through the accounting verification model to obtain the tracing analysis results. At the same time, the impact of the alarm point on the downstream is predicted according to the rainwater management model and emergency measures are taken.

[0078] Specifically, after receiving the alarm information, the alarm information is manually reviewed to determine whether it is true. If the alarm information is true, the subsequent steps are continued; otherwise, the alarm information is marked as a false alarm, the subsequent steps are stopped, and the process returns to S2. For example, when the manual review is in real time, the alarm information in which the monitoring data exceeds the preset threshold less or the magnitude of the sudden change in the real-time monitoring data is smaller can be marked as a false alarm based on the empirical value.

[0079] If the alarm is true, continue to perform the following steps:

[0080] S31. According to the pollution source tracing model based on the pollution emission characteristic analysis, the real-time monitoring data of the alarm points are analyzed to obtain the first group of suspected pollution sources.

[0081] S311. According to the pollution source tracing model based on the pollution emission characteristic analysis, feature extraction is performed on the real-time monitoring data of each monitoring node in the monitoring network, and a pollution emission characteristic portrait of each monitoring node is constructed based on the extracted features.

[0082] S312. Calculate the similarity between the pollution discharge characteristic portrait of the alarm point and the pollution discharge characteristic portraits of each upstream monitoring node based on the Mahalanobis distance.

[0083] S313, determine the suspected pollution source PY based on similarity i , traceability similarity PS i , suspected sewage discharge time PT i .

[0084] Among them, i represents the i-th suspected pollution source.

[0085] S314, traceability similarity PS i Sort from large to small and take the traceability similarity PS i The largest top n suspected pollution sources PYi As the first group of suspected pollution sources.

[0086] S32. According to the pollution source tracing model based on three-dimensional fluorescence, the polluted water samples at the alarm points are analyzed to obtain a second group of suspected pollution sources.

[0087] S321. According to the pollution source tracing model based on three-dimensional fluorescence, a three-dimensional fluorescence image database of watershed pollution sources is established.

[0088] S322. Collect polluted water samples from the alarm points and analyze the three-dimensional fluorescence spectrum of the polluted water samples.

[0089] S323, based on the convolutional neural network algorithm, the three-dimensional fluorescence spectrum of the polluted water sample is compared with the three-dimensional fluorescence image database of the pollution source in the watershed to obtain the suspected pollution source YY i , traceability similarity YS i .

[0090] S324, traceability similarity YY i Sort from large to small and take the traceability similarity YS i The largest n suspected pollution sources YY i As the second group of suspected pollution sources.

[0091] S33. Perform a weighted calculation based on the first group of suspected pollution sources and their corresponding traceability similarities, and the second group of suspected pollution sources and their corresponding traceability similarities, to obtain the corrected traceability similarities of the first group of suspected pollution sources and the second group of suspected pollution sources.

[0092] Specifically, if a suspected pollution source exists in both the first group of suspected pollution sources and the second group of suspected pollution sources, the weighted calculation formula for the traceability similarity of the suspected pollution source is as follows:

[0093] Y i =(PS i +YS i )×(k1 / 2);

[0094] Among them, Y i represents the corrected traceability similarity, k1 represents the first enhancement coefficient, and the value range of k1 is [1, 1.5];

[0095] If a suspected pollution source only exists in the first group of suspected pollution sources or the second group of suspected pollution sources, the weighted calculation formula for the traceability similarity of the suspected pollution source is as follows:

[0096] Y i =MAX(PS i , YS i )×m1;

[0097] Wherein, m1 represents the first weakening coefficient, and the value range of m1 is [0.5, 1].

[0098] S34. According to the first group of suspected pollution sources, the second group of suspected pollution sources, the corrected source tracing similarity, and the suspected pollution discharge time, based on the spatial position topological relationship of the watershed, a calculation verification model is constructed, and a forward simulation is performed on the first group of suspected pollution sources and the second group of suspected pollution sources to obtain the predicted time when the pollution discharge behavior of each suspected pollution source arrives at the alarm point at the suspected pollution discharge time. A weighted calculation is performed based on the predicted time and the alarm time of the alarm point to obtain the final source tracing similarity of the first group of suspected pollution sources and the second group of suspected pollution sources.

[0099] Specifically, a weighted calculation is performed based on the prediction time and the alarm time of the alarm point, and the final traceability similarity between the first group of suspected pollution sources and the second group of suspected pollution sources is obtained, including:

[0100] If the predicted time of a suspected pollution source is consistent with the alarm time of the alarm point, the calculation formula for the final traceability similarity of the suspected pollution source is as follows:

[0101] Y i '=Y i ×k2;

[0102] Among them, Y i ' represents the final traceability similarity, k2 represents the second enhancement coefficient, and the value range of k2 is [1, 1.5];

[0103] If the predicted time of a suspected pollution source is inconsistent with the alarm time of the alarm point, the calculation formula for the final traceability similarity of the suspected pollution source is as follows:

[0104] Y i '=Y i ×m2;

[0105] Wherein, m2 represents the second weakening coefficient, and the value range of m2 is [0.5, 1].

[0106] S35. Obtain the source tracing analysis result according to the final source tracing similarity, the first group of suspected pollution sources, the second group of suspected pollution sources and the suspected pollution discharge time.

[0107] Furthermore, if the alarm information is true, the following steps are performed simultaneously:

[0108] S3.1. Based on the Storm Water Management Model (SWMM), simulate and predict the concentration and time of pollutants from the warning point to the downstream receiving terminal.

[0109] Specifically, when pollutants are predicted to flow through the sewage network, the receiving terminal is the sewage treatment plant and enters S3.2; when pollutants are predicted to flow through the rainwater network, the receiving terminal is the river outlet and enters S3.3.

[0110] S3.2. Based on the predicted concentration and time of pollutants arriving at the sewage treatment plant, as well as the influent water quality requirements of the sewage treatment plant, evaluate whether the predicted pollutants will affect the biological treatment effect of the sewage treatment process.

[0111] Specifically, if it is predicted that the pollutants will affect the biological treatment effect of the sewage treatment process, emergency measures will be taken in advance in the regulating tank of the sewage treatment plant; otherwise, no emergency measures will be taken.

[0112] S3.3. Based on the predicted concentration and time of pollutants reaching the river outlet and the water quality requirements of the receiving river, assess whether the predicted pollutants will affect the river water quality compliance.

[0113] Specifically, if it is predicted that pollutants will affect the water quality of the river, emergency measures will be taken in advance to intercept the predicted pollutants and prevent them from entering the river; otherwise, no emergency measures will be taken.

[0114] S4. Conduct on-site traceability and evidence collection based on the traceability analysis results.

[0115] Specifically, the source tracing analysis results include the n suspected pollution sources with the highest source tracing similarity, source tracing similarity, and suspected pollution discharge time. Personnel conduct on-site source tracing and evidence collection based on the source tracing analysis results of this plan. There is no need for personnel to stay on site for a long time to collect pollution discharge evidence, which greatly improves the efficiency of evidence collection.

[0116] The present invention performs a fusion analysis on the pollution sources of rainwater and sewage based on a pollution source tracing model based on pollution discharge characteristics analysis and a pollution source tracing model based on three-dimensional fluorescence, thereby solving the problem of insufficient accuracy of a single tracing method; by constructing a monitoring network according to the discharge path of rainwater and sewage, and constructing an accounting verification model according to the discharge path of rainwater and sewage, and by positively verifying the results obtained by the fusion analysis through the accounting verification model, the discharge path of rainwater and sewage can be fully monitored, thereby solving the problem of large limitations on the applicable scenarios of water pollution source tracing methods, improving the accuracy of tracing, and expanding the scope of application of water pollution source tracing analysis methods;

[0117] The traceability analysis results of the present invention include the n suspected pollution sources with the highest traceability similarity, traceability similarity, and suspected pollution discharge time, which can assist in efficient environmental supervision and evidence collection, and solve the problem that the pollution discharge time is difficult to find out and the pollution discharge path is difficult to identify in the existing method, resulting in the actual evidence collection still requiring personnel to squat for a long time to collect pollution discharge evidence, and the environmental traceability evidence of abnormal pollution discharge is very difficult;

[0118] The present invention also simulates and predicts the concentration and time of pollutants reaching the downstream receiving terminal from the alarm point based on the accounting verification model, which can assist in emergency treatment of sudden pollution.

[0119] Example 2

[0120] like Figure 6 As shown, based on the same concept as the above-mentioned embodiment 1, this embodiment also proposes a water pollution source tracing analysis system based on multidimensional data fusion analysis, which is used to implement a water pollution source tracing analysis method based on multidimensional data fusion analysis as described in the above-mentioned embodiment.

[0121] In this embodiment, the system includes:

[0122] The data module is used to construct a monitoring network according to the discharge path of rainwater and sewage, and obtain real-time monitoring data of rainwater and sewage through the monitoring network;

[0123] The alarm module is connected to the data module and is used to determine whether the real-time monitoring data is abnormal; if the real-time monitoring data is abnormal, an alarm is issued to the monitoring node where the abnormality occurs; otherwise, the real-time monitoring data continues to be determined whether it is abnormal;

[0124] The source tracing analysis module is connected to the alarm module. After receiving the alarm information, it is used to conduct a fusion analysis of the pollution sources of rainwater and sewage based on the pollution source tracing model based on the pollution discharge characteristics analysis and the pollution source tracing model based on three-dimensional fluorescence, and to positively verify the results of the fusion analysis through the accounting verification model to obtain the tracing analysis results. At the same time, according to the rainwater management model, it predicts the impact of the alarm point on the downstream and takes emergency measures;

[0125] The information notification module is connected to the traceability analysis module and is used to conduct on-site traceability and evidence collection based on the traceability analysis results.

[0126] Furthermore, the information notification module is also used to take emergency measures based on the simulated prediction of the concentration and time of pollutants reaching the downstream receiving terminal from the alarm point.

[0127] Furthermore, the system also includes:

[0128] The prediction and early warning module is connected to the alarm module and is used to simulate and predict the concentration and time of pollutants reaching the downstream receiving terminal from the alarm point based on the rainwater management model.

[0129] The visualization display module is connected with the prediction and warning module, the traceability analysis module, the alarm module and the data module, and is used to display the basic data of the data module, the real-time monitoring data of the monitoring network, the traceability analysis results of the traceability analysis module and the warning results of the prediction and warning module.

[0130] The pollution incident management module is connected to the information notification module and is used to archive pollution incidents and to collect statistics on pollution incidents.

[0131] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A water pollution source tracing analysis method based on multidimensional data fusion analysis, characterized in that: include: S1. Building a monitoring network according to the discharge path of rainwater and sewage, and obtaining real-time monitoring data of the rainwater and sewage through the monitoring network; Specifically include: S11. Construct the spatial location topology of the watershed based on the discharge path of rainwater and sewage; S12, constructing a monitoring network according to the spatial position topological relationship of the watershed, and deploying water quality monitoring equipment, water quantity monitoring equipment and three-dimensional fluorescence monitoring equipment at each monitoring node based on the monitoring network; S2, judging whether the real-time monitoring data is abnormal; if the real-time monitoring data is abnormal, an alarm is issued to the monitoring node where the abnormality occurs; otherwise, S2 is continued to be executed; wherein the monitoring node that issues an alarm is the alarm point; S3. After receiving the alarm information, the pollution source of the rainwater and sewage is analyzed by fusion according to the pollution source tracing model based on the pollution characteristics analysis and the pollution source tracing model based on the three-dimensional fluorescence, and the result of the fusion analysis is positively verified by the accounting verification model to obtain the tracing analysis result. At the same time, the impact of the alarm point on the downstream is predicted according to the rainwater management model and emergency measures are taken; Specifically include: S31. Analyze the real-time monitoring data of the alarm points according to the pollution source tracing model based on the pollution discharge characteristic analysis to obtain a first group of suspected pollution sources; Specifically include: S311, according to the pollution source tracing model based on pollution discharge characteristic analysis, feature extraction is performed on the real-time monitoring data of each monitoring node in the monitoring network, and a pollution discharge characteristic portrait of each monitoring node is constructed according to the extracted features; S312, calculating the similarity between the pollution discharge characteristic portrait of the alarm point and the pollution discharge characteristic portrait of each monitoring node belonging to the upstream according to the Mahalanobis distance; S313, determining the suspected pollution source PY according to the similarity i , traceability similarity PS i , suspected sewage discharge time PT i ; Where i represents the i-th suspected pollution source; S314: The traceability similarity PS i Sort from large to small and take the traceability similarity PS i The largest top n suspected pollution sources PY i As the first group of suspected pollution sources; S32, analyzing the polluted water samples at the alarm points according to the pollution source tracing model based on three-dimensional fluorescence to obtain a second group of suspected pollution sources; Specifically include: S321, establishing a three-dimensional fluorescence image database of watershed pollution sources according to the pollution source tracing model based on three-dimensional fluorescence; S322, collecting polluted water samples at the alarm points, and analyzing the three-dimensional fluorescence spectrum of the polluted water samples; S323, based on the convolutional neural network algorithm, the three-dimensional fluorescence spectrum of the polluted water sample is compared with the three-dimensional fluorescence image database of the pollution source in the watershed to obtain the suspected pollution source YY i , traceability similarity YS i ; S324, the traceability similarity YY i Sort from large to small and take the traceability similarity YS i The largest n suspected pollution sources YY i As the second group of suspected pollution sources; S33, performing weighted calculation based on the first group of suspected pollution sources and their corresponding traceability similarities, and the second group of suspected pollution sources and their corresponding traceability similarities, to obtain the corrected traceability similarities between the first group of suspected pollution sources and the second group of suspected pollution sources; Specifically include: If a suspected pollution source exists in both the first group of suspected pollution sources and the second group of suspected pollution sources, the weighted calculation formula for the traceability similarity of the suspected pollution source is as follows: Y i =(P.S. i +YS i )×(k1 / 2); Among them, Y i represents the corrected traceability similarity, k1 represents the first enhancement coefficient, and the value range of k1 is [1, 1.5]; If a suspected pollution source only exists in the first group of suspected pollution sources or the second group of suspected pollution sources, the weighted calculation formula for the traceability similarity of the suspected pollution source is as follows: Y i =MAX(PS i ,YS i )×m1; Wherein, m1 represents the first weakening coefficient, and the value range of m1 is [0.5, 1]; S34. According to the first group of suspected pollution sources, the second group of suspected pollution sources, the corrected source tracing similarity, and the suspected pollution discharge time, based on the spatial position topological relationship of the watershed, a calculation verification model is constructed, and the first group of suspected pollution sources and the second group of suspected pollution sources are forward simulated to obtain the predicted time when the pollution discharge behavior of each suspected pollution source reaches the alarm point at the suspected pollution discharge time, and a weighted calculation is performed according to the predicted time and the alarm time of the alarm point to obtain the final source tracing similarity of the first group of suspected pollution sources and the second group of suspected pollution sources; Specifically include: If the predicted time of a suspected pollution source is consistent with the alarm time of the alarm point, the calculation formula for the final traceability similarity of the suspected pollution source is as follows: AND i '=And i ×k2; Among them, Y i ' represents the final traceability similarity, k2 represents the second enhancement coefficient, and the value range of k2 is [1, 1.5]; If the predicted time of a suspected pollution source is inconsistent with the alarm time of the alarm point, the calculation formula for the final traceability similarity of the suspected pollution source is as follows: AND i '=And i ×m2; Wherein, m2 represents the second weakening coefficient, and the value range of m2 is [0.5, 1]; S35, obtaining a source tracing analysis result according to the final source tracing similarity, the first group of suspected pollution sources, the second group of suspected pollution sources and the suspected pollution discharge time; S4. Conduct on-site source tracing and evidence collection based on the source tracing analysis results.

2. The water pollution source tracing analysis method based on multidimensional data fusion analysis according to claim 1 is characterized in that: In the above S3, after receiving the alarm information, it also includes: Manually review whether the alarm information is true; if the alarm information is true, continue to execute subsequent steps; otherwise, mark the alarm information as a false alarm, stop executing subsequent steps, and return to S2.

3. The water pollution source tracing analysis method based on multidimensional data fusion analysis according to claim 1 is characterized in that: In S3, the impact of the warning point on the downstream is predicted according to the rainwater management model and emergency measures are taken, including: S3.

1. According to the rainwater management model, simulate and predict the concentration and time of pollutants from the alarm point to the downstream receiving terminal; when the predicted pollutants flow through the sewage pipe network, the receiving terminal is the sewage treatment plant, and enter S3.2; when the predicted pollutants flow through the rainwater pipe network, the receiving terminal is the river outlet, and enter S3.3; S3.

2. Based on the predicted concentration and time of the pollutants arriving at the sewage treatment plant, and the influent water quality requirements of the sewage treatment plant, assess whether the predicted pollutants will affect the biological treatment effect of the sewage treatment process; if the predicted pollutants will affect the biological treatment effect of the sewage treatment process, take emergency measures in advance in the regulating tank of the sewage treatment plant; S3.

3. Based on the predicted concentration and time of the pollutants reaching the river outlet and the water quality requirements of the receiving river, assess whether the predicted pollutants will affect the river water quality standards; if the predicted pollutants will affect the river water quality standards, take emergency measures in advance to intercept the predicted pollutants and prevent them from entering the river.

4. A water pollution source tracing analysis system based on multidimensional data fusion analysis, used to implement a water pollution source tracing analysis method based on multidimensional data fusion analysis as described in any one of claims 1 to 3 above, characterized in that: The system includes: A data module is used to construct a monitoring network according to the discharge path of rainwater and sewage, and obtain real-time monitoring data of the rainwater and sewage through the monitoring network; an alarm module, connected to the data module, for determining whether the real-time monitoring data is abnormal; if the real-time monitoring data is abnormal, an alarm is issued to the monitoring node where the abnormality occurs; otherwise, the real-time monitoring data is continuously determined to determine whether the abnormality occurs; A source tracing analysis module is connected to the alarm module and is used to perform a fusion analysis on the pollution sources of the rainwater and sewage according to the pollution source tracing model based on the pollution discharge characteristic analysis and the pollution source tracing model based on three-dimensional fluorescence after receiving the alarm information, and to positively verify the results of the fusion analysis through the accounting verification model to obtain the source tracing analysis results. At the same time, the impact of the alarm point on the downstream is predicted according to the rainwater management model and emergency measures are taken; The information notification module is connected to the traceability analysis module and is used to perform on-site traceability and evidence collection based on the traceability analysis results.

Images