Drainage pipe network pollutant tracing method and system based on multi-level monitoring and medium

Abstract

The invention provides a drainage pipe network pollutant tracing method and system based on multi-level monitoring and a medium, and the method comprises the following steps: obtaining sampling data which is periodic data; inputting the sampling data into a trained water gene diagnosis model to obtain a diagnosis result; a target result is extracted based on the diagnosis result, so that target monitoring data are obtained based on the conductivity monitoring equipment of the drainage subarea corresponding to the target result, and the target result comprises that it is determined that the drainage subarea is polluted; and identifying a final target area based on the target monitoring data, wherein the target area comprises a pollutant source area. According to the drainage pipe network pollutant traceability method and system based on multi-level monitoring and the medium, through a low-cost efficient monitoring network, a wider area can be covered, more comprehensive water quality monitoring is achieved, the water quality problem can be found and solved in time, and the water quality monitoring efficiency is improved. And the overall operation and maintenance cost is reduced while the measurement precision and reliability are ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of environmental monitoring, and in particular relates to a method, system and medium for tracing pollutants in a drainage network based on multi-level monitoring. Background Art

[0002] The traditional online monitoring scheme for pollutant source tracing in drainage networks mainly faces compatibility issues caused by the diversity of equipment types. There are many monitoring devices, and the one-time fixed monitoring method is adopted, which cannot give full play to the value of the monitoring equipment and cannot truly reflect the concentration of pollutants. The high price of equipment is a significant drawback of traditional monitoring schemes. High-quality and high-precision monitoring equipment often comes with a high price tag, which increases the investment cost of the entire project, especially for regions or projects with limited budgets, which are difficult to bear all the costs.

[0003] At the same time, there are some low-priced but unstable equipment on the market. When choosing, you need to carefully weigh the cost-effectiveness to avoid increased maintenance costs due to equipment failure in the future. Traditional monitoring solutions rely on a lot of manual operation and maintenance, which not only increases labor costs, but also may lead to low work efficiency. Installation, commissioning, inspection and maintenance work all require the participation of professionals, and as the monitoring scope expands, the amount of labor will increase significantly. At the same time, accuracy is one of the key indicators for evaluating the performance of monitoring solutions. However, traditional monitoring solutions are often restricted by multiple factors in terms of accuracy, which will also affect the accuracy and reliability of traceability analysis, and thus affect subsequent pollution control and decision-making. Summary of the invention

[0004] In view of the shortcomings of the prior art mentioned above, the purpose of the present invention is to provide a method, system and medium for tracing pollutants in a drainage network based on multi-level monitoring, so as to solve the problem of high cost of using expensive equipment when tracing the above-mentioned pollutants and the problem of instability and low efficiency caused by using low-cost equipment.

[0005] In a first aspect, the present invention provides a method for tracing the source of pollutants in a drainage network based on multi-level monitoring, the method comprising the following steps:

[0006] Acquire sampling data, wherein the sampling data is periodic data;

[0007] Inputting the sampled data into a trained water gene diagnosis model to obtain a diagnosis result;

[0008] Extracting a target result based on the diagnosis result, thereby acquiring target monitoring data based on the conductivity monitoring device of the drainage partition corresponding to the target result, wherein the target result includes determining that it is contaminated;

[0009] A final target area is identified based on the target monitoring data, wherein the target area includes a pollutant source area.

[0010] In a possible implementation of the present application, the method further includes constructing and training a water gene diagnosis model, specifically including:

[0011] Access to the city’s sewage network;

[0012] Determine drainage zones based on the urban sewage system;

[0013] Acquire water gene pool data based on each drainage partition, wherein the water gene pool data includes fluorescence spectrum data of water quality of each drainage partition;

[0014] A water gene diagnosis initial model is constructed, and the water gene diagnosis initial model is trained based on the water gene library data to obtain the water gene diagnosis model, wherein the model is trained based on a self-training framework.

[0015] In a possible implementation of the present application, the determining of drainage zones based on the urban sewage official website specifically includes:

[0016] Taking the urban sewage pipe network as a whole and the urban sewage plant as a control node, the urban sewage pipe network is divided into the drainage zones by a preset division algorithm, wherein:

[0017] The directed pipe network in the urban sewage pipe network is divided by using a first algorithm, wherein the first algorithm includes a centrality algorithm;

[0018] The undirected pipe network in the urban sewage pipe network is divided by using a second algorithm, wherein the second algorithm includes a weighted network division algorithm;

[0019] The drainage zones are obtained by dividing the areas based on the connection relationship and attributes of different nodes and combining different pipe networks.

[0020] In a possible implementation of the present application, the inputting of the sampled data into a trained water gene diagnosis model to obtain a diagnosis result specifically includes:

[0021] Performing fluorescence spectrum detection on the sampled data to obtain detection data;

[0022] Performing similarity calculation based on the detection data and the fluorescence spectrum data in the water gene library data;

[0023] The detection data with a similarity threshold greater than or equal to a preset threshold are extracted and output as a contaminated diagnosis result, otherwise, a non-contaminated diagnosis result is output.

[0024] In a possible implementation of the present application, extracting the target result based on the diagnosis result, thereby obtaining the target monitoring data based on the conductivity monitoring device of the drainage partition corresponding to the target result, specifically includes:

[0025] Extracting and obtaining the target result based on the determined contaminated diagnostic result;

[0026] The target monitoring data is obtained based on the conductivity monitoring device set at the key node corresponding to the contaminated drainage partition, wherein the key node is the intersection of the main pipe and the branch pipe.

[0027] In a possible implementation of the present application, the final target area is identified based on the target monitoring data, and the target area includes a pollutant source area, specifically including:

[0028] Based on the acquired target monitoring data, the acquired target monitoring data is input into the trained pollutant anomaly prediction model for identification to obtain the monitoring node data corresponding to different key nodes;

[0029] A correlation analysis is performed based on the monitoring node data and the sewage treatment plant pollution data to identify the target area in the drainage partition, wherein the pollutant source area is determined based on the key node with the greatest correlation.

[0030] In a possible implementation of the present application, the method also includes visualizing the drainage partitions, which specifically includes using different colors to distinguish and visualize different drainage partitions, and includes visualizing the conductivity data monitored by the conductivity monitoring device, which specifically includes establishing a time series database and using a line graph and / or a scatter plot for visualization.

[0031] In a second aspect, the present invention provides a drainage network pollutant tracing system based on multi-level monitoring, the system comprising:

[0032] An acquisition module, used for acquiring sampling data, wherein the sampling data is periodic data;

[0033] An input module, used for inputting the sampled data into a trained water gene diagnosis model to obtain a diagnosis result;

[0034] An extraction module, for extracting a target result based on the diagnosis result, thereby acquiring conductivity monitoring data from the conductivity monitoring device of the drainage partition corresponding to the target result, wherein the target result includes determining that the drainage partition is contaminated;

[0035] An identification module is used to identify a final target area based on the conductivity monitoring data, wherein the target area includes a pollutant source area.

[0036] In a third aspect, the present invention provides an electronic device, the electronic device comprising: a processor and a memory;

[0037] The memory is used to store computer programs;

[0038] The processor is used to execute the computer program stored in the memory so that the electronic device executes the above-mentioned method for tracing the source of pollutants in the drainage network based on multi-level monitoring.

[0039] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by an electronic device, implements the above-mentioned method for tracing pollutants in a drainage network based on multi-level monitoring.

[0040] As described above, the drainage network pollutant tracing method, system and medium based on multi-level monitoring described in the present invention have the following beneficial effects: through a low-cost and efficient monitoring network, it can cover a wider area and achieve more comprehensive water quality monitoring, which not only helps to timely discover and solve water quality problems, but also reduces the overall operation and maintenance costs while ensuring measurement accuracy and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 Shown is a schematic diagram of a scene of an electronic device in an embodiment of the present invention;

[0042] Figure 2 Shown is a schematic diagram of the steps of a method for tracing the source of pollutants in a drainage network based on multi-level monitoring in one embodiment of the present invention;

[0043] Figure 3 Shown is a schematic diagram of drainage zones and control nodes in an embodiment of a method for tracing the source of pollutants in a drainage network based on multi-level monitoring of the present invention;

[0044] Figure 4 Shown is a schematic diagram of drainage zones after division in one embodiment of the method for tracing the source of pollutants in a drainage network based on multi-level monitoring of the present invention;

[0045] Figure 5 Shown is a schematic diagram of the division of areas and key nodes in an embodiment of the method for tracing the source of pollutants in a drainage network based on multi-level monitoring of the present invention;

[0046] Figure 6 Shown is a schematic structural diagram of a drainage pipe network pollutant source tracing system based on multi-level monitoring in one embodiment of the present invention;

[0047] Figure 7 It is a schematic structural diagram of an electronic device of the present invention in one embodiment. DETAILED DESCRIPTION

[0048] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0049] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0050] The following embodiments of the present invention provide a method for tracing the source of pollutants in a drainage network based on multi-level monitoring, which can be applied to Figure 1 The electronic device shown. The electronic device described in the present invention may include a mobile phone 11 with a wireless charging function, a tablet computer 12, a laptop computer 13, a wearable device, a vehicle-mounted device, an augmented reality (AR) / virtual reality (VR) device, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), etc. The embodiment of the present invention does not impose any restrictions on the specific type of the electronic device.

[0051] For example, the electronic device may be a station (STAION, ST) in a WLAN with a wireless charging function, a cellular phone, a cordless phone, a Session Initiation Protocol (SIP) phone, a Wireless Local Loop (WLL) station, a Personal Digital Assistant (PDA) device, a handheld device with a wireless charging function, a computing device or other processing device, a computer, a laptop computer, a handheld communication device, a handheld computing device, and / or other devices for communicating on a wireless system and a next generation communication system, such as a mobile terminal in a 5G network, a mobile terminal in a future evolved Public Land Mobile Network (PLMN), or a mobile terminal in a future evolved Non-terrestrial Network (NTN).

[0052] The technical solutions in the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0053] Specifically, see Figure 2 In one embodiment of the invention, the method for tracing the source of pollutants in a drainage network based on multi-level monitoring of the present invention comprises the following steps:

[0054] Step S202, acquiring sampling data, wherein the sampling data is periodic data;

[0055] Step S204, inputting the sampled data into the trained water gene diagnosis model to obtain a diagnosis result;

[0056] Step S206, extracting a target result based on the diagnosis result, thereby acquiring target monitoring data based on the conductivity monitoring device of the drainage partition corresponding to the target result, wherein the target result includes determining that the drainage partition is contaminated;

[0057] Step S208, identifying the final target area based on the target monitoring data, wherein the target area includes the pollutant source area.

[0058] It should be noted that, in this embodiment, in actual application, a trained water gene diagnosis model is specifically applied. The subsequent description will explain the water gene diagnosis model in detail, wherein the water gene diagnosis model described in the present invention specifically includes a model trained with a self-training framework, such as the Noisy Student model. Specifically, after obtaining the trained water gene diagnosis model, in the specific implementation, sampling data is first obtained, specifically water quality data sampled manually. Accordingly, the sampling data is periodic data, which is sampled manually in the field, and then the present invention is used to perform water quality detection. Specifically, manual sampling is performed based on the divided drainage partitions and control nodes to detect pollutant emissions in different drainage partitions. Considering the detection cost, the manual sampling cycle can be "3-6" hours / time under normal circumstances. After obtaining the sampling data, the sampling data is input into the trained water gene diagnosis model to obtain the model output to obtain the diagnosis result, wherein the diagnosis result includes a result of determining whether it is polluted or not polluted, such as Figure 3 As shown, the sewage treatment plant is divided into "5" drainage zones A, B, C, D, and E as the terminal node, and the corresponding sampling points are a, b, c, d, and e.

[0059] Furthermore, since the target result includes the determination of contamination, the target result is extracted based on the diagnosis result, and the target monitoring data is obtained based on the conductivity monitoring equipment of the drainage partition corresponding to the target result, and finally the final target area is identified based on the target monitoring data. The target area includes the pollutant source area, wherein the control node of the drainage partition is sampled manually, and the drainage partition itself is divided into multiple pollutant source areas, and each area differentiated within the drainage partition is subjected to conductivity monitoring by the set conductivity monitoring equipment, and then the specific source of pollutants is identified by conductivity, wherein the conductivity monitoring equipment reflects the conductivity changes of water quality in real time by measuring the ability of ions in the liquid to conduct current. This real-time feedback mechanism helps us to detect water quality abnormalities in a timely manner and respond quickly. Combined with water quality test data, we can have a more comprehensive understanding of water quality conditions and provide richer information for source tracing analysis.

[0060] It should be noted that based on the above monitoring and data analysis, users can develop scientific tracing strategies, quickly locate pollution sources and assess their impact range. At the same time, they can formulate efficient governance measures based on actual conditions, eliminate pollution threats in a timely manner, and ensure the safety and sustainable development of the water environment. The closed-loop management model from monitoring to analysis and governance of the present invention provides strong support for coping with complex drainage network pollution problems. At the same time, the use of inexpensive conductivity monitoring equipment is an extremely attractive advantage. These devices effectively control costs while ensuring measurement accuracy and reliability, making large-scale deployment possible. Through low-cost and efficient monitoring networks, they can cover a wider area in actual applications and achieve more comprehensive water quality monitoring. Specifically, the present invention overcomes the problem of high cost of high-cost equipment when tracing pollutants and the problem of instability and low efficiency caused by the use of low-cost equipment by zoning and slicing the urban sewage network and using different sampling mechanisms at different nodes, which not only ensures the accuracy and stability of monitoring, but also controls the equipment investment cost.

[0061] Furthermore, in one embodiment of the invention, the method also includes constructing and training a water gene diagnosis model, specifically including: obtaining an urban sewage pipe network; determining drainage zones based on the urban sewage pipe network; obtaining water gene library data based on each drainage zone, the water gene library data including fluorescence spectrum data of water quality in each drainage zone; building an initial water gene diagnosis model, and training the initial water gene diagnosis model based on the water gene library data to obtain the water gene diagnosis model, wherein the model is trained based on a self-training framework.

[0062] It should be noted that, in this embodiment, the urban sewage network is first obtained, and the urban sewage network is taken as the whole, and the urban sewage treatment plant is taken as the control node for sewage inflow. The network is first automatically divided into multiple drainage partitions according to the network topology relationship through the topological automatic community partitioning algorithm, and the nodes in the urban sewage network (such as sewage treatment plants, pumping stations, inspection wells, etc.) are divided into different communities (or communities) according to their connection relationships and attributes to reveal the structural characteristics and functional partitions of the network. Among them, the topological automatic community partitioning algorithm is directly based on the topological structure of the network for division, without the introduction of additional information or parameters, and reduces the influence of human factors. It is suitable for various types of networks, including undirected graphs, directed graphs, and weighted graphs. It can handle large-scale networks, can efficiently clarify the relationship between sewage networks, automatically divide drainage partitions, and determine the control nodes for monitoring.

[0063] Furthermore, water gene library data is obtained based on each drainage partition, and the water gene library data includes fluorescence spectrum data of water quality in each drainage partition, wherein the fluorescence spectrum is the spectrum emitted by a substance under specific excitation light, which is highly specific and unique. Different substances have different fluorescence spectrum characteristics, which enables fluorescence spectroscopy technology to distinguish different pollutants, with fast detection speed, short experimental time, and small sample amount required, which greatly saves time and resources. The detection results are accurate and reliable, and other interference factors can be almost eliminated, so as to realize intelligent classification and identification of different pollutants. When the fluorescence spectrum of an unknown sample is input into the model, the model can quickly determine its category, thereby realizing the traceability of the pollutant.

[0064] Furthermore, an initial model for water gene diagnosis is constructed, and the initial model for water gene diagnosis is trained based on the water gene library data to obtain the water gene diagnosis model, wherein the model is trained based on a self-training framework, specifically, data from the water gene library is collected, and the fluorescence spectrum training data set is specifically image data, which includes "10" categories, namely: catering, electronics, hairdressing, agriculture, life, sewage treatment plant, laundry, breeding, medicine and printing. Different types of urban sewage discharge can be sampled in advance through manual sampling, and machine learning can be performed. The collected data is cleaned, normalized, and missing values ​​are processed to ensure the quality and consistency of the data. Features that are crucial to model construction are extracted from the original data, and necessary feature scaling and dimensionality reduction operations are performed to improve the efficiency and accuracy of the model. Specifically, the steps of machine learning include data processing and enhancement as well as self-training framework training, wherein, in order to increase the robustness of the algorithm, a variety of transformations are performed on the image data, including but not limited to flipping, cropping, scaling, rotation, etc. Taking the Rand-Augment method (including but not limited to) as an example, the random data enhancement Rand-Augment method is based on the Auto-Augment method, and multiple parameters are optimized and managed at the strategy level. These multiple parameters are simplified into very few parameters: N transformations of the image and the intensity M of each transformation. The intensity parameter M of each transformation takes a value within a certain range of integers ab, indicating the size of the augmentation distortion (Augmentation Distortion) of the original image.

[0065] Furthermore, the water gene diagnosis model described in the present invention specifically includes a model trained using a self-training framework, such as the Noisy Student model. Specifically, the Noisy Student model can represent the highest accuracy in the current image classification community. The model uses a self-training framework in the process of training the model. The self-training framework can be the following steps:

[0066] I. Train a model on an annotated dataset (Image-Net) using conventional methods and use it as a teacher model.

[0067] II. Use the teacher model to classify some unlabeled images and collect samples with classification scores greater than the specified threshold as a pseudo-labeled dataset;

[0068] III. Retrain a student model on the mixed labeled and pseudo-annotated dataset;

[0069] IV. Use the trained student model as the teacher model, repeat steps II to III, and perform multiple iterations. The final student model is the water gene diagnosis model.

[0070] Further, in one embodiment of the invention, the determining of drainage zones based on the urban sewage network specifically includes:

[0071] Taking the urban sewage pipe network as a whole and the urban sewage plant as a control node, the urban sewage pipe network is divided into the drainage zones by a preset division algorithm, wherein:

[0072] The directed pipe network in the urban sewage pipe network is divided by using a first algorithm, wherein the first algorithm includes a centrality algorithm;

[0073] The undirected pipe network in the urban sewage pipe network is divided by using a second algorithm, wherein the second algorithm includes a weighted network division algorithm;

[0074] The drainage zones are obtained by dividing the areas based on the connection relationship and attributes of different nodes and combining different pipe networks.

[0075] It should be noted that, in this embodiment, the urban sewage pipe network is divided to obtain the drainage partitions. Since the preset division algorithm adopts a topological automatic community division algorithm, the first algorithm is used to divide the directed pipe network in the urban sewage pipe network. The first algorithm includes a centrality algorithm, among which the centrality algorithm has a better effect and the number of drainage partitions can be set. According to the complexity of the pipe network, the pipe network can be divided into multiple drainage partitions; and the second algorithm is used to divide the undirected pipe network in the urban sewage pipe network. The second algorithm includes a weighted network division algorithm, such as the Louvain algorithm. The weighting factors include but are not limited to the flow rate and pipe length of the pipe. The weighted network community division algorithm is adopted, which is suitable for weighted networks and can accurately reflect the strength of the connection between nodes.

[0076] Specifically, taking the first algorithm (centrality algorithm) as an example, the operation process of dividing the directed pipe network is as follows;

[0077] (1) Define the network model: describe the topological structure of the network, including the total number of nodes N and the total number of edges E, and construct an adjacency matrix or adjacency list to represent network connections.

[0078] (2) Calculate node centrality: calculate the degree Cd(v) of each node, that is, the number of other directly connected nodes, closeness centrality: based on the geographic centrality of the node, calculate the average distance from node v to all other nodes, and betweenness centrality is a measure of the "intermediary" importance of the node in the network, that is, the number of shortest paths passing through node v.

[0079] (3) Determine seed nodes: Select k nodes with the highest centrality as potential seed nodes, where k can be a fixed number or a percentage based on the network size.

[0080] (4) Initialize the community: Create an initial community for each seed node, and the community members are only the seed node itself.

[0081] (5) Expanding the community boundary: For each seed node, consider its neighbor nodes and include them in the community based on centrality or other correlation indicators (such as similarity). A greedy strategy can be used to select the node that will most improve the quality of the community for expansion.

[0082] (6) Update node centrality: After the community is expanded, the centrality of the nodes inside and on the boundaries of the community is recalculated to reflect their importance in the new community structure.

[0083] (7) Community merging and splitting: If the connection density between two communities is higher than that within the community, consider merging the two communities. If the connection density within the community is sparse, consider splitting the community to improve the homogeneity of the community.

[0084] (8) Evaluating community quality: Use indicators such as modularity Q to evaluate the quality of community division. The modularity range is [-1, 1]. The larger the value, the higher the quality of community division.

[0085] (9) Iterative optimization: Repeat steps (5) to (8) until the modularity of the community division is no longer significantly improved or the preset number of iterations is reached. Statistical methods such as cross-validation can be used to evaluate the stability and reliability of the algorithm.

[0086] (10) Result output: Output the final community division results, including the member node list of each community and the quality indicators of the community.

[0087] Among them, after the directed pipe network is divided into a plurality of drainage partitions whose nodes are not connected to each other by the first algorithm (centrality algorithm), the drainage partitions whose number of nodes in the drainage partitions is greater than a certain set value are selected for further drainage partition detection, and the drainage partitions whose nodes are more closely connected in the drainage partitions are separated to provide a basis for the drainage pipe network monitoring point layout. Furthermore, the method also includes visualizing the drainage partitions, which specifically includes using different colors to distinguish and visualize different drainage partitions, wherein, for example Figure 4 As shown, it is a schematic diagram of the drainage zones after division, with a total of "5" and they are marked with different colors.

[0088] Furthermore, in one embodiment of the invention, the step of inputting the sampled data into a trained water gene diagnosis model to obtain a diagnosis result specifically includes:

[0089] Performing fluorescence spectrum detection on the sampled data to obtain detection data;

[0090] Performing similarity calculation based on the detection data and the fluorescence spectrum data in the water gene library data;

[0091] The detection data with a similarity threshold greater than or equal to a preset threshold are extracted and output as a contaminated diagnosis result, otherwise, a non-contaminated diagnosis result is output.

[0092] It should be noted that, in this embodiment, when the water gene diagnosis model diagnoses the sampled data, it also performs fluorescence spectrum detection on the sampled data to obtain the detection data, thereby performing similarity calculation based on the detection data and the fluorescence spectrum data in the water gene library data, and finally extracting the detection data with a similarity threshold greater than or equal to the preset threshold to output a diagnosis result that is determined to be contaminated, otherwise it outputs a diagnosis result that is determined to be uncontaminated. Among them, when the similarity threshold obtained by the fluorescence spectrum of the sampled data reaches "80%" with the data in the water gene library, it can be considered that the tested water sample is mainly contaminated by the pollution source. It should be noted that this threshold is set according to the accuracy requirements of the system, data characteristics and actual application scenarios, taking into account the types of pollution sources that may be encountered, the degree of pollution, environmental factors, etc., and the above factors will affect the interpretation and judgment of the fluorescence spectrum data.

[0093] Further, in one embodiment of the invention, the extracting of the target result based on the diagnosis result, thereby obtaining the target monitoring data based on the conductivity monitoring device of the drainage partition corresponding to the target result, specifically includes:

[0094] Extracting and obtaining the target result based on the determined contaminated diagnostic result;

[0095] The target monitoring data is obtained based on the conductivity monitoring device set at the key node corresponding to the contaminated drainage partition, wherein the key node is the intersection of the main pipe and the branch pipe.

[0096] It should be noted that, in this embodiment, the target result is extracted based on the diagnosis result of the determined pollution. When the water quality is abnormal, the Spin-glass algorithm is combined to establish the link relationship between the main and branch pipes according to the topological relationship of the drainage network, and identify the key nodes where the main and branch pipes meet. In order to reduce the monitoring cost, the monitoring equipment with mature technology on the market is used to deploy conductivity monitoring equipment at the key nodes to analyze the abnormal pollutant data, so as to obtain the target monitoring data based on the conductivity monitoring equipment set at the key nodes corresponding to the drainage partition determined to be polluted, such as Figure 5 As shown in the figure, if the monitoring data generated at the water gene sampling point a is abnormal, the drainage partition A is determined to be the source of the problem. According to the topological automatic community partitioning algorithm, the drainage partition A is further divided into "5" partition areas, namely A1, A2, A3, A4, and A5. Correspondingly, the locations of the conductivity monitoring equipment are a1, a2, a3, a4, and a5.

[0097] Further, in one embodiment of the invention, the final target area is identified based on the target monitoring data, and the target area includes the pollutant source area, specifically including:

[0098] Based on the acquired target monitoring data, the acquired target monitoring data is input into the trained pollutant anomaly prediction model for identification to obtain the monitoring node data corresponding to different key nodes;

[0099] A correlation analysis is performed based on the monitoring node data and the sewage treatment plant pollution data to identify the target area in the drainage partition, wherein the pollutant source area is determined based on the key node with the greatest correlation.

[0100] It should be noted that in this embodiment, machine learning algorithms, such as random forests, support vector machines, etc., are used to train historical data and establish a pollutant anomaly prediction model. The model can learn the laws and patterns of changes in pollutant concentrations, thereby providing an early warning before the pollutant concentration increases abnormally. The model analyzes the newly collected data in real time to determine whether there is an abnormal input of pollutants, thereby improving the accuracy and timeliness of the early warning. Specifically, the acquired target monitoring data is input into the trained pollutant anomaly prediction model for identification to obtain monitoring node data corresponding to different key nodes, thereby performing a correlation analysis based on the monitoring node data and the sewage plant pollution data to identify the target area in the drainage partition, wherein the pollutant source area is determined based on the key node with the largest correlation. After obtaining the pollutant source area, the staff can install a water quality monitoring probe in the area to monitor and determine the final pollutants, such as industrial wastewater, domestic sewage, agricultural non-point source pollution, etc.

[0101] It should be noted that the operation steps of the pollutant anomaly prediction model are as follows:

[0102] (1) Data collection and organization:

[0103] Collect historical conductivity data from monitoring points in the sewer network. This data may include timestamps, conductivity readings, flow rate and volume of sewerage, etc. Organize the data and ensure that it is in a uniform format, for example, convert all data to the same time interval (such as readings every minute or every hour).

[0104] (2) Data preprocessing: Clean the data, remove invalid or erroneous conductivity readings, fill in missing values, possibly using values ​​from the previous time point or predictions from a statistical model. Standardize the conductivity data to convert it to a dimensionless form, such as using score normalization or min-max normalization.

[0105] (3) Feature engineering: Generate new features based on the original conductivity data, such as the moving average, rate of change, periodic components, etc. of the conductivity. Consider time series analysis methods, such as Fourier transform, to extract periodic features in the conductivity data.

[0106] (4) Label definition: Define abnormal events based on historical events or expert knowledge and assign labels to data points. For example, if the conductivity exceeds the historical mean plus three standard deviations, it is marked as abnormal.

[0107] (5) Dataset partitioning: The sorted dataset is divided into training set, validation set and test set in chronological order. For example, the most recent "60%" of data is used as the training set, the previous "20%" as the validation set, and the earliest "20%" as the test set.

[0108] (7) Model selection and training: Choose a suitable machine learning algorithm based on the nature of the problem. For binary classification problems (normal / abnormal), random forest and SVM are both viable options.

[0109] (8) Model training: Use the training set data to train the model. For random forest, you may need to set the number of trees (n_estimators), the maximum tree depth (max_depth), and the minimum number of samples required for each split (min_samples_split). Taking conductivity data as an example, assuming that the random forest algorithm is used, the possible parameter settings are as follows:

[0110] n_estimators: Sets the number of decision trees in the forest, such as "100" trees.

[0111] max_depth: Set the maximum depth of each tree, such as "10", to prevent over-fitting caused by too deep trees.

[0112] min_samples_split: Sets the minimum number of samples required to split an internal node, such as "2".

[0113] min_samples_leaf: Sets the minimum number of samples required for a leaf node, such as "1".

[0114] (9) Cross-validation: Use K-fold cross-validation to evaluate the stability of the model and ensure that the model is not overfitted on a specific data subset.

[0115] (10) Model evaluation: Different evaluation metrics such as precision, recall, F1 score, and area under the receiver operating characteristic (ROC) curve (AUC) are used on the test set.

[0116] (11) Model optimization: Adjust model parameters based on the evaluation results and use methods such as grid search or random search to optimize parameters.

[0117] (12) Model interpretation and validation: Feature importance analysis is used to understand which features are most critical to the model’s predictions, which helps verify whether the model reasonably captures the relationship between conductivity and pollutant anomalies.

[0118] (13) Model deployment: Integrate the trained model into the real-time monitoring system so that it can automatically receive new conductivity data and make anomaly predictions.

[0119] (14) Monitoring and maintenance: Regularly evaluate the performance of the model in actual applications, adjust model parameters or retrain the model to adapt to new data patterns as needed.

[0120] Furthermore, in one embodiment of the invention, the method further includes visualizing the conductivity data monitored by the conductivity monitoring device, which specifically includes establishing a time series database and using a line graph and / or a scatter plot for visualization.

[0121] It should be noted that in this embodiment, first, the data of continuous monitoring of the conductivity equipment is collected, a time series database is established, the integrity and accuracy of the data are ensured, and statistical methods such as mean, median, standard deviation, etc. are used to analyze the normal fluctuation range of the conductivity data, and a reasonable threshold is set to identify abnormal data points. Specifically, real-time or regular monitoring data of conductivity (EC) and ammonia nitrogen (NH3-N) are collected, and the time range of the data is determined according to the analysis requirements. It can be data of a certain day or long-term historical data. The abnormal values ​​in the data are checked, such as concentration values ​​beyond the normal range or data points that obviously do not conform to the laws of physics, and they are removed or replaced with reasonable values. The data at different times and under different conditions are converted into comparable forms, for example, the conductivity and ammonia nitrogen concentration can be converted into daily averages, weekly averages or monthly averages. The processed data are organized into a table or database form for subsequent analysis.

[0122] Secondly, data visualization techniques, such as line graphs and scatter plots, are used to intuitively display the trend of conductivity over time, so as to identify sudden increases or decreases. These abnormal changes are often associated with sharp changes in pollutant concentrations. Specifically, chart software (such as Excel, MATLAB, Python, etc.) is used to draw trend graphs of conductivity and ammonia nitrogen concentration over time. Linear regression, time series analysis and other methods are used to further quantify trends and correlations. Observe whether the trends of conductivity and ammonia nitrogen concentration in the trend graph are consistent or correlated. If the trends of the two are similar, it may indicate that they come from the same pollution source or have undergone similar pollution processes. The Pearson correlation coefficient (r) formula is used to calculate the correlation between conductivity and ammonia nitrogen, and the strength of the correlation between conductivity and ammonia nitrogen is interpreted according to the value of the correlation coefficient. Among them, r values ​​close to "1" or "-1" indicate strong correlation, and values ​​close to "0" indicate no correlation.

[0123] The embodiment of the present application also provides a drainage network pollutant tracing system based on multi-level monitoring, and the drainage network pollutant tracing system based on multi-level monitoring can implement the drainage network pollutant tracing method based on multi-level monitoring described in the present application, but the implementation device of the drainage network pollutant tracing method based on multi-level monitoring described in the present application includes but is not limited to the structure of the drainage network pollutant tracing system based on multi-level monitoring listed in the present embodiment. All structural deformations and replacements of the prior art made according to the principles of the present application are included in the protection scope of the present application.

[0124] See also Figure 6 In one embodiment, this embodiment provides a drainage network pollutant tracing system 60 based on multi-level monitoring, the system comprising:

[0125] An acquisition module 61 is used to acquire sampled data, wherein the sampled data is periodic data;

[0126] An input module 62, used to input the sampled data into a trained water gene diagnosis model to obtain a diagnosis result;

[0127] An extraction module 63 is used to extract a target result based on the diagnosis result, so as to obtain conductivity monitoring data from the conductivity monitoring device of the drainage partition corresponding to the target result, wherein the target result includes determining that the drainage partition is contaminated;

[0128] The identification module 64 is used to identify the final target area based on the conductivity monitoring data, and the target area includes the pollutant source area.

[0129] Since the specific implementation of this embodiment corresponds to the aforementioned method embodiment, the same details will not be repeated here, and those skilled in the art should also understand that Figure 6 The division of the modules in the embodiment is merely a division of logical functions, and in actual implementation, all or part of them can be integrated into one or more physical entities, and these modules can be implemented entirely in the form of software called by processing elements, or entirely in the form of hardware, or some modules can be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware.

[0130] In the several embodiments provided by the present invention, it should be understood that the disclosed system, device or method can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules / units is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules or units, which can be electrical, mechanical or other forms.

[0131] The modules / units described as separate components may or may not be physically separated, and the components displayed as modules / units may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules / units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention. For example, the functional modules / units in the various embodiments of the present invention may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0132] Those of ordinary skill in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0133] The embodiment of the present invention also provides a computer-readable storage medium. A person of ordinary skill in the art can understand that all or part of the steps in the method for implementing the above embodiment can be completed by instructing the processor through a program, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state hard disk, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state disk (SSD)), etc.

[0134] An embodiment of the present invention further provides an electronic device, which includes a processor and a memory.

[0135] The memory is used to store computer programs.

[0136] The memory includes: ROM, RAM, disk, USB flash drive, memory card or CD and other media that can store program codes.

[0137] The processor is connected to the memory and is used to execute the computer program stored in the memory so that the electronic device executes the above-mentioned drainage network pollutant tracing method based on multi-level monitoring.

[0138] Preferably, the processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0139] like Figure 7 As shown, the electronic device of the present invention is in the form of a general computing device. The components of the electronic device may include but are not limited to: one or more processors or processing units 71, a memory 72, and a bus 73 connecting different system components (including the memory 72 and the processing unit 71).

[0140] Bus 73 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0141] Electronic devices typically include a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device, including volatile and non-volatile media, removable and non-removable media.

[0142] The memory 72 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 721 and / or cache memory 722. The electronic device may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 723 may be used to read and write non-removable, non-volatile magnetic media ( Figure 7 not shown, usually called a "hard drive"). Although Figure 7 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 73 via one or more data medium interfaces. The memory 72 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0143] A program / utility 724 having a set (at least one) of program modules 7241 may be stored, for example, in the memory 72, such program modules 7241 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 7241 generally perform the functions and / or methods of the embodiments described herein.

[0144] The electronic device may also communicate with one or more external devices (e.g., keyboards, pointing devices, displays, etc.), one or more devices that enable a user to interact with the electronic device, and / or any device that enables the electronic device to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via input / output (I / O) interface 74. Furthermore, the electronic device may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 75. Figure 7 As shown, the network adapter 75 communicates with other modules of the electronic device via the bus 73. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0145] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A method for tracing the source of pollutants in a drainage network based on multi-level monitoring, characterized in that: include: Acquire sampling data, wherein the sampling data is periodic data; Inputting the sampled data into a trained water gene diagnosis model to obtain a diagnosis result; Extracting a target result based on the diagnosis result, thereby acquiring target monitoring data based on the conductivity monitoring device of the drainage partition corresponding to the target result, wherein the target result includes determining that it is contaminated; A final target area is identified based on the target monitoring data, wherein the target area includes a pollutant source area.

2. The method for tracing the source of pollutants in a drainage network based on multi-level monitoring according to claim 1 is characterized in that: The method also includes constructing and training a water gene diagnosis model, specifically including: Access to the city’s sewage network; Determine drainage zones based on the urban sewage system; Acquire water gene pool data based on each drainage partition, wherein the water gene pool data includes fluorescence spectrum data of water quality of each drainage partition; A water gene diagnosis initial model is constructed, and the water gene diagnosis initial model is trained based on the water gene library data to obtain the water gene diagnosis model, wherein the model is trained based on a self-training framework.

3. The method for tracing the source of pollutants in a drainage network based on multi-level monitoring according to claim 2 is characterized in that: The method of determining the drainage zones based on the urban sewage treatment plant specifically includes: Taking the urban sewage pipe network as a whole and the urban sewage plant as a control node, the urban sewage pipe network is divided into the drainage zones by a preset division algorithm, wherein: The directed pipe network in the urban sewage pipe network is divided by using a first algorithm, wherein the first algorithm includes a centrality algorithm; The undirected pipe network in the urban sewage pipe network is divided by using a second algorithm, wherein the second algorithm includes a weighted network division algorithm; The drainage zones are obtained by dividing the areas based on the connection relationship and attributes of different nodes and combining different pipe networks.

4. The method for tracing the source of pollutants in a drainage network based on multi-level monitoring according to claim 3 is characterized in that: The step of inputting the sampled data into a trained water gene diagnosis model to obtain a diagnosis result specifically includes: Performing fluorescence spectrum detection on the sampled data to obtain detection data; Performing similarity calculation based on the detection data and the fluorescence spectrum data in the water gene library data; The detection data with a similarity threshold greater than or equal to a preset threshold are extracted and output as a contaminated diagnosis result, otherwise, a non-contaminated diagnosis result is output.

5. The method for tracing the source of pollutants in a drainage network based on multi-level monitoring according to claim 4 is characterized in that: The extracting of target results based on the diagnosis results, thereby obtaining target monitoring data based on the conductivity monitoring equipment of the drainage partition corresponding to the target results, specifically includes: Extracting and obtaining the target result based on the determined contaminated diagnostic result; The target monitoring data is obtained based on the conductivity monitoring device set at the key node corresponding to the contaminated drainage partition, wherein the key node is the intersection of the main pipe and the branch pipe.

6. The method for tracing the source of pollutants in a drainage network based on multi-level monitoring according to claim 5 is characterized in that: The final target area is identified based on the target monitoring data, and the target area includes the pollutant source area, specifically including: Based on the acquired target monitoring data, the acquired target monitoring data is input into the trained pollutant anomaly prediction model for identification to obtain the monitoring node data corresponding to different key nodes; A correlation analysis is performed based on the monitoring node data and the sewage treatment plant pollution data to identify the target area in the drainage partition, wherein the pollutant source area is determined based on the key node with the greatest correlation.

7. The method for tracing the source of pollutants in a drainage network based on multi-level monitoring according to claim 6 is characterized in that: The method also includes visualizing the drainage partitions, which specifically includes using different colors to distinguish and visualize different drainage partitions, and visualizing the conductivity data monitored by the conductivity monitoring equipment, which specifically includes establishing a time series database and visualizing it using a line graph and / or a scatter graph.

8. A drainage network pollutant tracing system based on multi-level monitoring, characterized in that: include: An acquisition module, used for acquiring sampling data, wherein the sampling data is periodic data; An input module, used for inputting the sampled data into a trained water gene diagnosis model to obtain a diagnosis result; An extraction module, for extracting a target result based on the diagnosis result, thereby acquiring conductivity monitoring data from the conductivity monitoring device of the drainage partition corresponding to the target result, wherein the target result includes determining that the drainage partition is contaminated; An identification module is used to identify a final target area based on the conductivity monitoring data, wherein the target area includes a pollutant source area.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for tracing the source of pollutants in a drainage network based on multi-level monitoring as described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: The electronic device includes: a processor and a memory; wherein the memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory, so that the electronic device executes the drainage network pollutant tracing method based on multi-level monitoring as described in any one of claims 1 to 7.

Images