A method and system for intelligent water quality testing in water plants

Abstract

This application provides an intelligent water quality detection method and system for water plants, relating to the field of water quality detection technology. The method includes: analyzing the detection points on the main water supply line to determine the water supply detection points; optimizing the detection control of the water supply detection points according to a multi-parameter discrimination scheme to generate a parameter set; performing anomaly detection on the water supply detection points according to the parameter set to obtain detection results; predicting anomalies in the detection results through a water quality anomaly tracking channel to generate anomaly prediction points; analyzing the linkage anomalies of the anomaly prediction points to determine the pollution propagation network; and completing the water quality detection through anomaly response. This application can solve the technical problem of poor prediction accuracy of pollutant diffusion trends in existing technologies, achieving the technical effect of improving the prediction accuracy of pollution propagation paths.

Description

Technical Field

[0001] This application relates to the field of water quality testing technology, and in particular to an intelligent water quality testing method and system for water plants. Background Technology

[0002] With the increasing complexity of urban water supply systems and the rising demands of residents for water quality safety, the need for intelligent water quality testing at water plants has significantly increased. Traditional methods relying on manual timed sampling and single-point sensor monitoring can no longer meet the needs for high-frequency response and precise management of water quality that changes over a wide range and dynamically.

[0003] Currently, existing water quality testing technologies typically rely on centralized data acquisition platforms such as physicochemical parameter sensors, rapid microbial detection modules, and SCADA systems to achieve real-time monitoring of key indicators such as turbidity, residual chlorine, pH, and conductivity. However, these technologies still have limitations in practical applications. Most existing water quality monitoring systems lack intelligent data correlation analysis capabilities, cannot perform linked judgments on multi-point detection data, struggle to predict pollutant propagation trends, and are unable to achieve efficient anomaly tracing.

[0004] In summary, existing technologies suffer from the technical problem that the lack of dynamic detection point optimization and multi-parameter intelligent fusion judgment mechanism leads to the identification of water quality anomalies being limited to local single-point over-limit alarms, further affecting the accurate prediction of pollutant diffusion trends and the rapid location of pollution sources. Summary of the Invention

[0005] The purpose of this application is to provide an intelligent water quality detection method and system for water plants, in order to solve the technical problem that the lack of dynamic detection point optimization and multi-parameter intelligent fusion judgment mechanism in the existing technology leads to the identification of water quality anomalies being limited to local single-point over-limit alarms, which further affects the accurate prediction of pollutant diffusion trends and the rapid location of pollution sources.

[0006] In view of the above problems, this application provides an intelligent water quality detection method and system for water plants.

[0007] Firstly, this application provides an intelligent water quality detection method for water plants, implemented through an intelligent water quality detection system for water plants, comprising: analyzing the detection points on the main water supply line to determine the water supply detection points; optimizing the detection control of the water supply detection points according to a multi-parameter discrimination scheme to generate a parameter set; performing anomaly detection on the water supply detection points according to the parameter set to obtain detection results; predicting anomalies in the detection results through a water quality anomaly tracking channel to generate anomaly prediction points; analyzing the linkage anomalies of the anomaly prediction points to determine the pollution propagation network; and completing water quality detection through anomaly response.

[0008] Preferably, the intelligent water quality detection method for water plants further includes: introducing a multi-parameter discrimination scheme to construct a multi-parameter scheme set; performing parameter clustering on the multi-parameter scheme set based on scene attention to obtain parameter aggregation; performing parameter optimization on the parameter aggregation based on parameter detection results to obtain parameter optimization aggregation; and traversing and combining the parameter optimization aggregation to obtain parameter groups.

[0009] Preferably, the intelligent water quality detection method for water plants further includes: extracting a first parameter group according to the parameter group, and extracting a first parameter based on the first parameter group; performing anomaly detection on the water supply detection point according to the first parameter to obtain a first detection result; if the first detection result is greater than or equal to a predetermined first type of anomaly, adding the water supply detection point to the anomaly detection point; and continuing to confirm the anomaly of the anomaly detection point according to the first parameter group to obtain a detection result.

[0010] Preferably, the intelligent water quality detection method for water plants further includes: extracting a second parameter based on the first parameter group; confirming the anomaly of the abnormal detection point based on the second parameter to obtain a second detection result; if the second detection result is greater than or equal to a predetermined second type of anomaly, retaining the abnormal detection point to obtain an abnormal detection point retention result; obtaining a first anomaly percentage of the number of parameters corresponding to the abnormal detection point retention result in the first parameter group; if the first anomaly percentage is greater than a preset anomaly percentage, adding the abnormal detection point to the abnormal detection result, and combining it with the normal detection result to obtain the detection result.

[0011] Preferably, the intelligent water quality detection method for water plants further includes: using a water flow sample set from the main water supply line and a detection result sample set as input data, and an anomaly detection point sample set as output data, supervising the training of the water quality anomaly tracking channel to obtain water quality anomaly tracking learning coefficients; if the water quality anomaly tracking learning coefficients satisfy the water quality anomaly tracking learning constraints, generating the water quality anomaly tracking channel; and based on the main water supply line and the detection results, performing anomaly prediction according to the water quality anomaly tracking channel to generate the anomaly prediction points.

[0012] Preferably, the intelligent water quality detection method for water plants further includes: extracting a first abnormal detection point and a second abnormal detection point based on the abnormal detection points; determining whether the first abnormal detection point and the second abnormal detection point exist simultaneously in the abnormal prediction points to obtain a first linkage point; traversing and judging the abnormal detection points based on the first linkage point to obtain linkage points; and determining the pollution propagation network based on the linkage points.

[0013] Preferably, the intelligent water quality detection method for water plants further includes: locating abnormal sources in the pollution propagation network based on the main water supply line to obtain the abnormal source locations; responding to abnormalities based on the pollution propagation network; continuously monitoring the abnormal source locations; and completing the water quality detection.

[0014] Preferably, the intelligent water quality detection method for water plants further includes: performing grid processing on the main water supply line to obtain a water supply grid; performing anomaly prediction on the water supply grid based on the anomaly detection point sample set of the main water supply line to obtain anomaly prediction grid; dividing the detection density according to the anomaly prediction grid to obtain multiple detection density grids, and setting detection points based on the multiple detection density grids to generate the water supply detection points.

[0015] Preferably, the intelligent water quality detection method for water plants further includes: the multi-parameter discrimination scheme includes a physicochemical parameter discrimination scheme, an organic matter parameter discrimination scheme, an inorganic ion parameter discrimination scheme, a particle and turbidity parameter discrimination scheme, a microbial index parameter discrimination scheme, and an auxiliary index parameter discrimination scheme.

[0016] Secondly, this application also provides an intelligent water quality detection system for water plants, used to execute an intelligent water quality detection method for water plants as described in the first aspect, comprising: a water supply detection point determination module, used to analyze the detection points of the main water supply line and determine the water supply detection points; a parameter group generation module, used to optimize the detection control of the water supply detection points according to a multi-parameter discrimination scheme and generate a parameter group; a detection result acquisition module, used to perform anomaly detection on the water supply detection points according to the parameter group and obtain detection results; an anomaly prediction point generation module, used to perform anomaly prediction on the detection results through a water quality anomaly tracking channel and generate anomaly prediction points; and a water quality detection module, used to analyze the linkage anomalies of the anomaly prediction points, determine the pollution propagation network, and complete the water quality detection through anomaly response.

[0017] The technical solution provided in this application has at least the following technical effects or advantages: by realizing the technical objectives of intelligent deployment of detection points based on gridded analysis of water supply trunk lines, parameter group optimization and anomaly tracking fusion analysis, it achieves the technical effects of improving the accuracy of water quality anomaly identification and realizing the linkage between pollution propagation path prediction and source point location response.

[0018] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating an intelligent water quality testing method for water plants according to this application.

[0021] Figure 2 This is a schematic diagram of the structure of an intelligent water quality detection system for a water plant according to this application.

[0022] Explanation of reference numerals in the attached diagram: Module 1 for determining water supply testing points, Module 2 for generating parameter groups, Module 3 for obtaining testing results, Module 4 for generating anomaly prediction points, and Module 5 for water quality testing. Detailed Implementation

[0023] This application provides an intelligent water quality monitoring method and system for water plants, solving the technical problem that existing technologies, due to the lack of dynamic monitoring point optimization and multi-parameter intelligent fusion judgment mechanisms, limit the identification of water quality anomalies to local single-point over-limit alarms, further affecting the accurate prediction of pollutant diffusion trends and rapid location of pollution sources. It achieves the technical goals of intelligent deployment of monitoring points based on gridded analysis of water supply trunk lines, parameter group optimization, and anomaly tracking and fusion analysis, thereby improving the accuracy of water quality anomaly identification and realizing the linkage between pollution propagation path prediction and source location response.

[0024] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0025] Example 1, please refer to the appendix. Figure 1 This application provides an intelligent water quality detection method for water plants, applied to an intelligent water quality detection system for water plants, specifically including the following steps:

[0026] S1: Analyze the detection points of the main water supply line to determine the water supply detection points.

[0027] Furthermore, this application also includes: performing gridding processing on the main water supply trunk line to obtain a water supply grid; performing anomaly prediction on the water supply grid based on the anomaly detection point sample set of the main water supply trunk line to obtain anomaly prediction grid; dividing the detection density according to the anomaly prediction grid to obtain multiple detection density grids, and setting detection points based on the multiple detection density grids to generate the water supply detection points.

[0028] Specifically, analyzing the detection points along the main water supply trunk line and determining the water supply detection points involves dividing the entire water supply pipeline into multiple zones based on its spatial structure and flow characteristics, using mathematical or spatial modeling methods. This creates multiple water supply grids with independent geographical attributes and hydraulic characteristics. The main water supply trunk line refers to the core water supply channel connecting water plants and major branch networks, undertaking the main transportation tasks of the entire water supply system. The water supply grid is a series of continuous or discontinuous unit areas divided along the main water supply trunk line according to spatial intervals or water pressure zones. Each water supply grid represents a water supply unit to be analyzed, used for subsequent anomaly analysis and detection point placement.

[0029] Next, anomaly predictions are made for the water supply grid based on a sample set of historical anomaly detection points along the main water supply line, identifying potential risk areas. The anomaly detection point sample set refers to the collection of monitoring points that have previously experienced events such as water quality exceeding standards, abnormal residual chlorine, or sudden changes in turbidity. By using time-series data from historical anomaly detection points, predictions can be trained to assess the probability of similar anomalies occurring in different water supply grids in the future.

[0030] Then, the detection density is divided according to the anomaly prediction grid to achieve optimal resource allocation and maximize detection coverage. Detection density division refers to dividing the entire water supply trunk line into three types of detection zones: high-density, medium-density, and low-density. High-density grids correspond to areas with higher predicted risks and will have more monitoring points deployed there; low-density grids correspond to areas with lower risks and will have only a few monitoring points set up for basic monitoring.

[0031] Finally, based on multiple detection density grids, detection points are set up to generate water supply detection points. This refers to deploying specific sensors or monitoring devices to specific locations within the water supply network according to a density division scheme. After setup, a well-structured and fully functional set of water supply detection points will be obtained, providing basic data support for subsequent water quality testing, anomaly identification, and source tracing analysis.

[0032] S2: Optimize the detection and control of the water supply detection points according to the multi-parameter discrimination scheme, and generate parameter groups.

[0033] Furthermore, this application also includes: introducing a multi-parameter discrimination scheme to construct a multi-parameter scheme set; performing parameter clustering on the multi-parameter scheme set based on scene attention to obtain parameter aggregation; performing parameter optimization on the parameter aggregation based on parameter detection results to obtain parameter optimization aggregation; and traversing and combining the parameter optimization aggregation to obtain parameter groups.

[0034] Furthermore, this application also includes: the multi-parameter discrimination scheme includes a physicochemical parameter discrimination scheme, an organic parameter discrimination scheme, an inorganic ion parameter discrimination scheme, a particle and turbidity parameter discrimination scheme, a microbial indicator parameter discrimination scheme, and an auxiliary indicator parameter discrimination scheme.

[0035] Specifically, optimizing the detection and control of water supply monitoring points based on a multi-parameter discrimination scheme to generate parameter sets refers to formulating and screening schemes using different categories of water quality detection parameters at identified water supply monitoring points to optimize monitoring efficiency and anomaly detection accuracy. Water supply monitoring points refer to sensor installation points deployed at key locations in the main water supply network to acquire water quality data in real time. Optimizing detection and control involves finding the optimal combination of multiple detection parameters and strategies for a specific scenario, achieving optimal allocation of monitoring resources and maximum effectiveness of detection results while ensuring water quality safety.

[0036] A multi-parameter discrimination scheme is introduced to construct a multi-parameter scheme set, designing various discrimination schemes based on water quality indicators. Each scheme includes a set of functionally related detection parameters, enabling rapid response to specific types of pollution or anomalies. The multi-parameter scheme set is a collection of different parameter combinations, covering the detection needs of various water quality change scenarios.

[0037] Next, parameter clustering is performed on the multi-parameter scheme set based on scenario attention. For different detection points and their functional characteristics, such as raw water inlet, treatment plant outlet, or end-user, the corresponding water quality monitoring priorities are assessed. Scenario attention measures the sensitivity of a detection scenario to changes in a certain type of parameter. For example, at a water treatment plant outlet, residual chlorine and coliform bacteria are of higher concern; while in industrial area pipe networks, ammonia nitrogen and conductivity may be more important. Parameter clustering, after considering scenario attention, groups highly correlated parameter combinations into the same category to facilitate subsequent overall optimization.

[0038] Subsequently, parameter optimization is performed on the aggregated parameters based on the parameter detection results. The overall performance of each parameter group is evaluated through actual detection results in terms of anomaly identification, response speed to water quality changes, and false alarm rate. The parameter detection results include indicators such as monitoring sensitivity, false alarm rate, and cost-effectiveness ratio. By comparing the parameter detection results, parameter combinations that are more suitable for the target detection points can be screened, resulting in the optimized parameter aggregation, which is a further selection of high-quality combinations based on the parameter aggregation.

[0039] Next, the parameter optimization aggregation is traversed and combined to obtain parameter sets. By combining and recombining multiple optimization results in different ways, the effects of parameter combinations with different quantities and arrangements are explored. The parameter sets are the final list of parameters used for actual detection.

[0040] The multi-parameter discrimination scheme includes schemes for physicochemical parameters, organic matter parameters, inorganic ion parameters, particulate matter and turbidity parameters, microbiological indicators, and auxiliary indicators. Physicochemical parameters, such as temperature, pH, and conductivity, mainly reflect the physical and acid-base state of the water body; organic matter parameters, such as COD, BOD, and TOC, reflect the degree of organic pollution in the water; inorganic ion parameters, such as ammonia nitrogen, nitrate, and total nitrogen, reveal abnormalities in the nitrogen cycle; particulate matter and turbidity parameters, such as turbidity and suspended solids, reflect the clarity of the water body; microbiological indicators, such as coliform bacteria and heterotrophic bacteria, are used to indicate the risk of biological pollution; auxiliary indicators, such as color and odor, are used for the early detection of sensory abnormalities. Table 1 shows the data of the multi-parameter discrimination scheme for water quality.

[0041] Table 1: Data Table of Multi-Parameter Water Quality Judgment Scheme

[0042]

[0043] S3: Perform anomaly detection on the water supply detection points according to the parameter group and obtain the detection results.

[0044] Furthermore, this application also includes: extracting a first parameter group according to the parameter group, and extracting a first parameter based on the first parameter group; performing anomaly detection on the water supply detection point according to the first parameter to obtain a first detection result; if the first detection result is greater than or equal to a predetermined first type of anomaly, adding the water supply detection point to the anomaly detection point; and continuing to confirm the anomaly of the anomaly detection point according to the first parameter group to obtain a detection result.

[0045] Furthermore, this application also includes: extracting a second parameter based on the first parameter group; confirming the anomaly of the anomaly detection point based on the second parameter to obtain a second detection result; if the second detection result is greater than or equal to a predetermined second type of anomaly degree, retaining the anomaly detection point to obtain an anomaly detection point retention result; obtaining a first anomaly percentage of the number of parameters corresponding to the anomaly detection point retention result in the first parameter group; if the first anomaly percentage is greater than a preset anomaly percentage, adding the anomaly detection point to the anomaly detection result, and combining it with the normal detection result to obtain a detection result.

[0046] Specifically, anomaly detection is performed on water supply monitoring points based on parameter sets. Water quality data is collected at each water supply monitoring point, and anomalies are determined by identifying whether water quality indicators exceed the normal range, thus obtaining the detection results.

[0047] This process involves dividing the parameters from multiple category combinations within a parameter group into multiple parameter combinations. Each parameter combination contains parameters from each category. Then, a parameter combination is randomly selected from these multiple parameter combinations for initial screening, serving as the first parameter group. From this first parameter group, any parameter is extracted as the first parameter, such as turbidity, residual chlorine, or pH.

[0048] Subsequently, anomaly detection is performed on the water supply monitoring points based on the first parameter to obtain the first detection result. At each monitoring point, a preliminary anomaly assessment is performed using the first parameter to determine if there are any instances of the corresponding indicator exceeding limits, sudden changes, or abnormal trends. This first detection result represents the initial screening of suspected anomalies, which are then used for more precise verification in subsequent stages. If the first detection result is greater than or equal to the predetermined anomaly level corresponding to the first parameter, the water supply monitoring point detected as abnormal by the first parameter is marked as an anomaly monitoring point.

[0049] Next, any parameter other than the first parameter is extracted from the first parameter group and used as the second parameter. Anomaly detection is performed on the anomaly detection points based on the second parameter to obtain the second detection result, ensuring the accuracy of the anomaly judgment. If the second detection result is greater than or equal to the predetermined second type of anomaly degree corresponding to the second parameter, the anomaly detection point is retained, forming the anomaly detection point retention result.

[0050] The retained anomaly detection points are the set of points confirmed as anomalies after the second screening and requiring further response. Next, the percentage of parameters corresponding to the retained anomaly detection points in the first parameter group is obtained, called the first anomaly percentage. This percentage indicates that if most parameters show an anomaly for a given point after testing with multiple parameters, then the anomaly does indeed exist, and the anomaly detection point is added to the anomaly detection results. Finally, the complete detection results are obtained by combining the normal detection results of all points with those of the anomaly detection points.

[0051] S4: The detection results are used to predict anomalies through the water quality anomaly tracking channel, and anomaly prediction points are generated.

[0052] Furthermore, this application also includes: using a water flow sample set and a detection result sample set from the main water supply line as input data, and an anomaly detection point sample set as output data, supervising the training of the water quality anomaly tracking channel to obtain water quality anomaly tracking learning coefficients; if the water quality anomaly tracking learning coefficients satisfy the water quality anomaly tracking learning constraints, generating the water quality anomaly tracking channel; and based on the main water supply line and the detection results, performing anomaly prediction according to the water quality anomaly tracking channel to generate the anomaly prediction points.

[0053] Specifically, anomaly prediction is performed on water supply monitoring points through a water quality anomaly tracking channel, generating anomaly prediction points. These points are used to predict the spread of potential anomalies and to infer areas of the pipeline network where anomalies may occur but have not yet been detected. Anomaly prediction points refer to water supply points that are predicted to have a future possibility of anomalies based on model calculations.

[0054] The main water supply trunk line's flow sample set and the test result sample set are used as input data. The main water supply trunk line is the main channel in the water supply network that starts from the water plant and connects to branch pipelines in various areas; the flow sample set refers to the hydraulic information such as flow velocity, flow direction, pressure, and water volume collected at different times and locations, used to simulate the flow path of water in the pipeline network; the test result sample set refers to the historical data set marked as normal or abnormal in previous test processes.

[0055] Next, the water quality anomaly tracking channel is trained under supervised supervision using the anomaly detection point sample set as output data. The anomaly detection point sample set refers to the collection of historically confirmed locations of anomalous events, serving as the target label for the channel's training and guiding it to identify potential propagation paths and risk areas. Subsequently, if the water quality anomaly tracking learning coefficients satisfy the learning constraints, it indicates that the channel has met the performance requirements for training, and it can be formally established as a water quality anomaly tracking channel. The water quality anomaly tracking learning coefficients are a set of weights or parameters formed during the training process to describe the mathematical relationship between water flow and anomaly propagation. Constraints may include indicators such as low error rate and goodness of fit; only when these are satisfied is the water quality anomaly tracking channel considered to have reliable predictive ability.

[0056] Once the water quality anomaly tracking channel is established, it can predict anomalies based on the structural information and monitoring results of the main water supply line. The channel will analyze normal points that may be affected by upstream anomalies and calculate the probability of them developing into anomalies within a certain period, thus generating anomaly prediction points. Even if a point is currently within the normal range, it has been marked as a potential future risk area, allowing for early monitoring or warning.

[0057] S5: Analyze the linkage anomalies of the predicted anomaly points, determine the pollution propagation network, and complete water quality testing through anomaly response.

[0058] Furthermore, this application also includes: extracting a first anomaly detection point and a second anomaly detection point based on the anomaly detection point; determining whether the first anomaly detection point and the second anomaly detection point exist synchronously in the anomaly prediction point to obtain a first linkage point; traversing and judging the anomaly detection point based on the first linkage point to obtain linkage points; and determining the pollution transmission network based on the linkage points.

[0059] Furthermore, this application also includes: locating abnormal sources in the pollution propagation network based on the main water supply line, and obtaining the abnormal source locations; responding to abnormalities based on the pollution propagation network, continuously monitoring the abnormal source locations, and completing water quality testing.

[0060] Specifically, the process involves analyzing the interconnected anomalies at predicted locations to identify the pollution propagation network. Through anomaly response, water quality testing is conducted. By comparing the predicted and actual anomaly points, the existence of anomaly linkages between multiple locations is identified, and a pollution diffusion path is constructed. This process then reverses the process to pinpoint the pollution source and activate the response mechanism. Interconnected anomalies refer to multiple locations exhibiting related anomalies in time or space, potentially caused by the same pollution event. The pollution propagation network is a pollution diffusion map constructed based on water flow paths and anomaly temporal relationships, used to analyze how pollution spreads from the source to multiple areas. Anomaly response refers to actions taken after identifying pollution propagation to prevent further deterioration, including alarm triggering, area isolation, water quality retesting, and pollution source tracing.

[0061] Based on the anomaly detection points, first and second anomaly detection points are randomly extracted, and the confirmed anomaly detection points are further grouped or sorted. Next, it is determined whether the first and second anomaly detection points exist simultaneously in the anomaly prediction points, thus obtaining the first linkage point. The anomaly prediction point is the location of possible anomalies inferred by the model. If the actual anomaly detection results coincide with the predicted points, especially if two or more anomaly points appear simultaneously in the predicted points, a correlated anomaly relationship is formed, which is called the first linkage point. The first linkage point is an important relay point for analyzing the pollution path and can reflect the degree of matching between prediction and reality.

[0062] Next, based on the first linked point, the abnormal detection points are traversed and judged. Further, the entire set of detection data is searched to see if other points also have similar temporal or water flow path associations with the first linked point, identifying more points that may be affected by the same pollution source, ultimately forming a more complete set of linked points. Linked points refer to a set of multiple interconnected detection locations affected by the same pollution source and experiencing anomalies sequentially within a certain timeframe.

[0063] Then, based on the linked points, the pollution propagation network is determined. These linked points are connected within the water supply network using factors such as water flow direction, distance, and time windows to construct possible pollution propagation paths. The pollution propagation network includes the pollution origin, pathway points, and impact points, reflecting how pollutants spread with the water flow and at which nodes they are detected. The pollution propagation network is one of the core data structures in the water quality monitoring system, used for visualizing pollution spread, predicting the next propagation direction, and performing source tracing analysis.

[0064] Next, anomaly sources are located in the pollution propagation network based on the main water supply trunk line. Using the hydraulic model and pollution path of the main pipeline, possible pollution sources are deduced. The main water supply trunk line is the most important water transmission path in the water supply system. Source location is based on methods such as backward flow direction, time regression, and pollutant concentration gradients. For example, if multiple downstream points have anomalies at similar time intervals, while their common upstream node showed anomalies slightly earlier, then the upstream node is likely the source of the anomaly.

[0065] Finally, anomaly response based on the pollution propagation network refers to activating the corresponding emergency response mechanism after confirming the pollution path and source. This includes high-frequency water quality monitoring of the anomaly source, closing pollution branch valves, and issuing water supply warnings to relevant areas. Continuous monitoring involves deploying more sensors or increasing the frequency of existing sensors around the anomaly point to ensure that the pollution does not spread further. This completes a closed-loop water quality monitoring system, encompassing anomaly detection, prediction, diffusion identification, source location, and response.

[0066] In summary, the intelligent water quality detection method for water plants provided in this application has the following technical effects: by realizing the technical objectives of intelligent deployment of detection points based on gridded analysis of water supply trunk lines, parameter group optimization and anomaly tracking fusion analysis, it achieves the technical effects of improving the accuracy of water quality anomaly identification and realizing the linkage between pollution propagation path prediction and source point location response.

[0067] Example 2: Based on the same inventive concept as the intelligent water quality detection method for water plants in the foregoing examples, this application also provides an intelligent water quality detection system for water plants. Please refer to the appendix. Figure 2 The system includes: a water supply monitoring point determination module 1, used to analyze the monitoring points of the main water supply line and determine the water supply monitoring points; a parameter group generation module 2, used to optimize the monitoring control of the water supply monitoring points according to a multi-parameter discrimination scheme and generate a parameter group; a detection result acquisition module 3, used to perform anomaly detection on the water supply monitoring points according to the parameter group and obtain the detection results; an anomaly prediction point generation module 4, used to predict anomalies in the detection results through a water quality anomaly tracking channel and generate anomaly prediction points; and a water quality detection module 5, used to analyze the linkage anomalies of the anomaly prediction points, determine the pollution propagation network, and complete the water quality detection through anomaly response.

[0068] Furthermore, the intelligent water quality detection system for water plants is also used for: introducing a multi-parameter discrimination scheme to construct a multi-parameter scheme set; performing parameter clustering on the multi-parameter scheme set based on scene attention to obtain parameter aggregation; performing parameter optimization on the parameter aggregation based on parameter detection results to obtain parameter optimization aggregation; and traversing and combining the parameter optimization aggregation to obtain parameter groups.

[0069] Furthermore, the intelligent water quality detection system for water plants is also used for: extracting a first parameter group according to the parameter group, and extracting a first parameter based on the first parameter group; performing anomaly detection on the water supply detection point according to the first parameter to obtain a first detection result; if the first detection result is greater than or equal to a predetermined first type of anomaly, adding the water supply detection point to the anomaly detection point; and continuing to confirm the anomaly of the anomaly detection point according to the first parameter group to obtain a detection result.

[0070] Furthermore, the intelligent water quality detection system for water plants is also used for: extracting a second parameter based on the first parameter group; confirming the anomaly of the abnormal detection point based on the second parameter to obtain a second detection result; if the second detection result is greater than or equal to a predetermined second type of anomaly, retaining the abnormal detection point to obtain an abnormal detection point retention result; obtaining the first anomaly percentage of the number of parameters corresponding to the abnormal detection point retention result in the first parameter group; if the first anomaly percentage is greater than a preset anomaly percentage, adding the abnormal detection point to the abnormal detection result, and combining it with the normal detection result to obtain the detection result.

[0071] Furthermore, the intelligent water quality monitoring system for water plants is also used for: supervising the training of the water quality anomaly tracking channel using a water flow sample set and a test result sample set as input data and an anomaly detection point sample set as output data to obtain water quality anomaly tracking learning coefficients; generating the water quality anomaly tracking channel if the water quality anomaly tracking learning coefficients satisfy the water quality anomaly tracking learning constraints; and generating the anomaly prediction points based on the main water supply line and the test results and the water quality anomaly tracking channel.

[0072] Furthermore, the intelligent water quality monitoring system for water plants is also used for: extracting a first abnormal detection point and a second abnormal detection point based on the abnormal detection points; determining whether the first abnormal detection point and the second abnormal detection point exist synchronously in the abnormal prediction points to obtain a first linkage point; traversing and judging the abnormal detection points based on the first linkage point to obtain linkage points; and determining the pollution propagation network based on the linkage points.

[0073] Furthermore, the intelligent water quality detection system for water plants is also used for: locating abnormal sources in the pollution propagation network based on the main water supply line, and obtaining the abnormal source locations; responding to abnormalities based on the pollution propagation network, continuously monitoring the abnormal source locations, and completing water quality detection.

[0074] Furthermore, the intelligent water quality detection system for water plants is also used for: performing grid processing on the main water supply line to obtain a water supply grid; performing anomaly prediction on the water supply grid based on the abnormal detection point sample set of the main water supply line to obtain anomaly prediction grid; dividing the detection density according to the anomaly prediction grid to obtain multiple detection density grids, and setting detection points based on the multiple detection density grids to generate the water supply detection points.

[0075] Furthermore, the intelligent water quality detection system for water plants is also used in the following ways: the multi-parameter discrimination scheme includes a physicochemical parameter discrimination scheme, an organic matter parameter discrimination scheme, an inorganic ion parameter discrimination scheme, a particle and turbidity parameter discrimination scheme, a microbial index parameter discrimination scheme, and an auxiliary index parameter discrimination scheme.

[0076] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The intelligent water quality detection method and specific examples in the first embodiment of the above description are also applicable to the intelligent water quality detection system of the water plant in this embodiment. Through the foregoing detailed description of the intelligent water quality detection method of the water plant, those skilled in the art can clearly understand the intelligent water quality detection system of the water plant in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0077] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0078] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for intelligent water quality testing in water plants, characterized in that, include: Analyze the monitoring points along the main water supply line to determine the water supply monitoring locations; Based on a multi-parameter discrimination scheme, the detection and control of the water supply detection points are optimized to generate parameter sets. Anomaly detection is performed on the water supply detection points based on the parameter set, and the detection results are obtained, including: Based on the parameter set, extract the first parameter set, and extract the first parameter based on the first parameter set; Anomaly detection is performed on the water supply detection point based on the first parameter to obtain a first detection result; If the first detection result is greater than or equal to the predetermined first type of anomaly, the water supply detection point will be added to the anomaly detection point. Based on the first parameter group, the abnormal detection points are further confirmed to obtain detection results, including: Extract the second parameter based on the first parameter group; Based on the second parameter, the abnormal detection point is confirmed to be abnormal, and a second detection result is obtained; If the second detection result is greater than or equal to the predetermined second type of anomaly, the anomaly detection point is retained to obtain the anomaly detection point retention result. Obtain the first abnormal percentage of the number of parameters corresponding to the retained result of the abnormal detection point in the first parameter group. If the first abnormal percentage is greater than the preset abnormal percentage, add the abnormal detection point to the abnormal detection result and combine it with the normal detection result to obtain the detection result. The detection results are used to predict anomalies through a water quality anomaly tracking channel, generating anomaly prediction points. Analyze the linkage anomalies at the predicted anomaly points to determine the pollution propagation network, and complete water quality testing through anomaly response.

2. The intelligent water quality detection method for water plants as described in claim 1, characterized in that, The detection and control optimization of the water supply detection points is performed based on a multi-parameter discrimination scheme to generate a parameter set, including: A multi-parameter discrimination scheme is introduced to construct a multi-parameter scheme set; The multi-parameter solution set is clustered based on scene attention to obtain parameter aggregation; Based on the parameter detection results, parameter optimization is performed on the parameter aggregation to obtain the parameter-optimized aggregation; The parameter optimization aggregation is traversed and combined to obtain parameter groups.

3. The intelligent water quality detection method for water plants as described in claim 1, characterized in that, The detection results are used to predict anomalies through a water quality anomaly tracking channel, generating anomaly prediction points, including: Using the water flow sample set and detection result sample set of the main water supply line as input data and the abnormal detection point sample set as output data, the water quality anomaly tracking channel is supervised and trained to obtain the water quality anomaly tracking learning coefficient. If the water quality anomaly tracking learning coefficients satisfy the water quality anomaly tracking learning constraints, the water quality anomaly tracking channel is generated. Based on the main water supply line and the detection results, anomaly prediction is performed according to the water quality anomaly tracking channel to generate the anomaly prediction points.

4. The intelligent water quality detection method for water plants as described in claim 1, characterized in that, Analyze the linkage anomalies of the predicted anomaly locations to determine the pollution propagation network, including: Based on the anomaly detection points, extract the first anomaly detection point and the second anomaly detection point; Determine whether the first anomaly detection point and the second anomaly detection point exist synchronously in the anomaly prediction point to obtain the first linkage point; Based on the first linkage point, the abnormal detection points are traversed and judged to obtain the linkage points; The pollution transmission network is determined based on the aforementioned linkage points.

5. The intelligent water quality detection method for water plants as described in claim 1, characterized in that, Water quality testing is completed through anomaly response, including: Based on the main water supply line, the abnormal source location of the pollution propagation network is determined, and the abnormal source location is obtained. Anomalies are detected based on the pollution propagation network, and the abnormal source locations are continuously monitored to complete water quality testing.

6. The intelligent water quality detection method for water plants as described in claim 1, characterized in that, Analysis of monitoring points along the main water supply line was conducted to determine the water supply monitoring points, including: The main water supply line is gridded to obtain a water supply grid. Anomaly prediction is performed on the water supply grid based on the sample set of anomaly detection points of the main water supply line to obtain an anomaly prediction grid. The detection density is divided according to the anomaly prediction grid to obtain multiple detection density grids, and the detection points are set based on the multiple detection density grids to generate the water supply detection points.

7. The intelligent water quality detection method for water plants as described in claim 1, characterized in that, The multi-parameter discrimination scheme includes physicochemical parameter discrimination scheme, organic parameter discrimination scheme, inorganic ion parameter discrimination scheme, particle and turbidity parameter discrimination scheme, microbial indicator parameter discrimination scheme, and auxiliary indicator parameter discrimination scheme.

8. An intelligent water quality testing system for water plants, characterized in that, The steps for implementing the intelligent water quality detection method for water plants according to any one of claims 1 to 7 include: The water supply monitoring point determination module is used to analyze the monitoring points of the main water supply line and determine the water supply monitoring points. The parameter group generation module is used to optimize the detection and control of the water supply detection points according to the multi-parameter discrimination scheme and generate parameter groups. The detection result acquisition module is used to perform anomaly detection on the water supply detection point according to the parameter group and obtain the detection result. An anomaly prediction point generation module is used to predict anomalies in the detection results through the water quality anomaly tracking channel and generate anomaly prediction points. The water quality detection module is used to analyze the linkage anomalies at the predicted anomaly points, determine the pollution propagation network, and complete the water quality detection through anomaly response.

Images