Water supply network abnormality identification method, device, electronic equipment and storage medium

Abstract

The embodiment of the present invention discloses a method, device, electronic device and storage medium for identifying abnormalities in a water supply network, and relates to the technical field of identifying abnormalities in a water supply network, wherein the method includes: collecting pressure data from each monitoring point and preprocessing it, using the Pearson correlation coefficient to construct a dynamic pressure correlation matrix based on the pressure data, and performing abnormality judgment and repair on the monitoring equipment; calculating the standard deviation of the pressure difference of each monitoring point in adjacent time periods to identify the water hammer effect; constructing a residual distribution benchmark based on historical monitoring data through a machine learning algorithm, calculating the real-time pressure residual of each monitoring point, calculating a comprehensive abnormality index based on the residual distribution benchmark, and performing abnormality judgment and early warning; using the Kriging interpolation method and the semivariogram function to generate and display the spatial distribution of pressure residuals based on the real-time pressure residuals. The present invention solves the problems of the prior art in that data verification is not systematic, time-space correlation is missing, the abnormality classification dimension is single, it is easy to misjudge, and the response efficiency is low.

Description

Technical Field

[0001] The present invention relates to the technical field of water supply network anomaly identification, and in particular to a water supply network anomaly identification method, device, electronic equipment and storage medium. Background Art

[0002] With the acceleration of urbanization, the scale of water supply networks continues to expand, and water demand is becoming increasingly complex and volatile. This poses unprecedented challenges to the stable operation and safe management of water supply networks. Abnormal events such as pipe bursts, water hammer, and valve misoperation are common in water supply networks. These events not only waste water resources but can also cause water supply interruptions, impacting residents' lives and industrial production.

[0003] However, traditional water supply network anomaly identification technologies exhibit significant limitations when addressing the complexities of modern water supply network management. Traditional anomaly identification methods often rely on single-line time series analysis or simple threshold judgments. These methods lack systematic verification of the integrity and accuracy of monitoring data, making it difficult to effectively identify data anomalies caused by equipment failures (such as sampling pipe blockages and RTU failures), which can easily lead to misjudgments in subsequent pipeline network event analysis.

[0004] At the same time, these methods often overlook the spatiotemporal coupling between the geographic location of monitoring points and pressure data, making it difficult to accurately distinguish between equipment failures and actual pipeline network events, resulting in a high rate of misjudgment of abnormal events. Furthermore, traditional methods are also very limited in their classification of abnormal events, failing to fully capture the complex abnormal patterns within the pipeline network. They also lack effective means of locating abnormal areas, making it difficult for operations and maintenance personnel to quickly locate and take action when faced with an abnormal event.

[0005] Therefore, there is an urgent need for a water supply network anomaly identification method that can effectively integrate spatiotemporal coupling analysis and multi-feature fusion technology to improve the accuracy, real-time and reliability of water supply network anomaly identification. Summary of the Invention

[0006] The embodiments of the present invention provide a method for identifying anomalies in a water supply network to address the problems of the prior art, such as unsystematic data verification, lack of spatiotemporal correlation, single anomaly classification dimension, easy misjudgment, and low response efficiency. The technical solution is as follows:

[0007] According to one aspect of the present invention, a method for identifying abnormalities in a water supply network includes: collecting pressure data from each monitoring point and preprocessing it, using the Pearson correlation coefficient to construct a dynamic pressure correlation matrix based on the pressure data, and performing abnormal judgment and repair on the monitoring equipment; the pressure data includes a timestamp, a pressure value, the geographical coordinates of the monitoring point, the number and the water supply system to which it belongs; calculating the standard deviation of the pressure difference of each of the monitoring points in adjacent time periods to identify the water hammer effect; constructing a residual distribution benchmark based on historical monitoring data through a machine learning algorithm, calculating the real-time pressure residual of each of the monitoring points, calculating a comprehensive abnormality index based on the residual distribution benchmark and the real-time pressure residual, and performing abnormality judgment and early warning based on the comprehensive abnormality index; using Kriging interpolation and semivariogram function to generate a pressure residual spatial distribution surface based on the real-time pressure residual, and using color gradient mapping to convert the spatial distribution surface into a graphical interface and display it.

[0008] In one embodiment, collecting pressure data from each monitoring point and preprocessing it is achieved by the following steps: collecting pressure data from each monitoring point in the water supply network, unifying the timestamps and matching the geographic coordinates, and setting the valid range of the pressure value; marking the monitoring points where the pressure value exceeds the valid range as over-threshold points, and marking the monitoring points where data is missing as missing points based on the continuity of the timestamps; calculating the data integrity of each monitoring point based on the number of over-threshold points and missing points, and triggering an abnormality warning if the data integrity is less than a preset threshold.

[0009] In one embodiment, a dynamic pressure correlation matrix is ​​constructed based on the pressure data using the Pearson correlation coefficient, and abnormal judgment and repair of the monitoring equipment are achieved through the following steps: a set sliding window is used to traverse the pressure time series according to the pressure data, the spatial proximity relationship of the monitoring points is quantified based on the Euclidean distance, and multiple neighboring points of the monitoring points are determined; a dynamic pressure correlation matrix is ​​constructed based on the pressure values ​​in the sliding window of each monitoring point using the Pearson correlation coefficient, and the pressure correlation coefficient of each monitoring point and multiple neighboring points is calculated based on the dynamic pressure correlation matrix, and the maximum value of the coefficient is taken; if the maximum value of the coefficient is less than a preset threshold, it is determined that the pressure fluctuation correlation of the monitoring point is abnormal, triggering an abnormal warning and repair of the monitoring equipment, and deleting the relevant data of the monitoring point from the pressure data.

[0010] In one embodiment, the water hammer effect is identified by calculating the pressure difference standard deviation of each monitoring point in adjacent time periods through the following steps: calculating the absolute value of the pressure difference of each monitoring point in adjacent time periods, calculating the pressure difference standard deviation based on the absolute value of the pressure difference, setting a pressure difference standard deviation threshold, and triggering a water hammer event alarm if the pressure difference standard deviation exceeds the threshold; the pressure difference standard deviation is used to reflect the severity of pressure fluctuations.

[0011] In one embodiment, constructing a residual distribution benchmark based on historical monitoring data through a machine learning algorithm is achieved by the following steps: predicting the pipeline network pressure through a machine learning algorithm based on historical monitoring data to obtain a predicted value, calculating the pressure residuals of each monitoring point at different time points based on the predicted value and the true value; calculating the residual mean and standard deviation of each monitoring point based on the pressure residual, and constructing a residual distribution benchmark under normal conditions based on the residual mean and standard deviation.

[0012] In one embodiment, a comprehensive anomaly index is calculated based on the residual distribution benchmark and the real-time pressure residual, and abnormality determination and early warning are performed based on the comprehensive anomaly index by the following steps: the pressure anomaly index of the monitoring point is calculated by the following formula:

[0013] ;

[0014] According to the hydraulic importance of each monitoring point in the water supply network, the pressure anomaly index of each monitoring point whose pressure anomaly index is greater than 0 is weightedly summed to obtain a comprehensive anomaly index; a first-level warning threshold and a second-level warning threshold are set, and the status is determined to be normal when the comprehensive anomaly index is not greater than the first-level warning threshold; when the comprehensive anomaly index is greater than the first-level warning threshold and not greater than the second-level warning threshold, a medium-risk alarm is triggered and manual verification is started; when the comprehensive anomaly index is greater than the second-level warning threshold, a high-risk alarm is triggered and an emergency response mechanism is started.

[0015] In one embodiment, the spatial distribution surface of pressure residuals is generated according to the real-time pressure residuals using the Kriging interpolation method and the semivariogram function through the following steps: for the monitoring points where the pressure anomaly index is greater than 0, the spatial autocorrelation of the monitoring points is quantified according to the pressure data of the monitoring points using the Kriging interpolation method, and the spatial variation pattern of the pressure residuals of the monitoring points is analyzed by the semivariogram function based on the real-time pressure residuals of the monitoring points; a theoretical model is determined according to the residual distribution benchmark, and a mathematical model characterizing the spatial correlation of pressure residuals is constructed by fitting the semivariogram function with the theoretical model, and a continuous spatial distribution surface of pressure residuals is generated in combination with the results of the Kriging interpolation method.

[0016] According to one aspect of the present invention, a water supply network anomaly identification device includes: a data acquisition and equipment maintenance module, which is used to collect pressure data from each monitoring point and preprocess it, and use the Pearson correlation coefficient to construct a dynamic pressure correlation matrix based on the pressure data to judge and repair anomalies of the monitoring equipment; the pressure data includes a timestamp, a pressure value, the geographical coordinates of the monitoring point, the number and the water supply system to which it belongs; an anomaly identification and pressure analysis module, which is used to calculate the standard deviation of the pressure difference of each of the monitoring points in adjacent time periods to identify the water hammer effect; an anomaly calculation and risk warning module, which constructs a residual distribution benchmark through a machine learning algorithm based on historical monitoring data, which is used to calculate the real-time pressure residual of each of the monitoring points, calculates a comprehensive anomaly index based on the residual distribution benchmark and the real-time pressure residual, and performs anomaly judgment and warning based on the comprehensive anomaly index; a spatial distribution generation and display module, which is used to generate a pressure residual spatial distribution surface based on the real-time pressure residual using the Kriging interpolation method and the semivariogram function, and converts the spatial distribution surface into a graphical interface and displays it using color gradient mapping.

[0017] According to one aspect of the present invention, an electronic device includes at least one processor and at least one memory, wherein the memory stores computer-readable instructions; the computer-readable instructions are executed by one or more of the processors, so that the electronic device implements the water supply network abnormality identification method as described above.

[0018] According to one aspect of the present invention, a storage medium stores computer-readable instructions thereon, wherein the computer-readable instructions are executed by one or more processors to implement the water supply network anomaly identification method as described above.

[0019] The beneficial effects brought about by the technical solution provided by the present invention are:

[0020] In the above technical solution, the present invention first collects pressure data containing information such as timestamps, pressure values, geographic coordinates, etc. from each monitoring point and preprocesses it to ensure the integrity and accuracy of the data, and constructs a dynamic pressure correlation matrix to judge the abnormality of the monitoring equipment and repair it, thereby solving the problem of data distortion. Secondly, by calculating the standard deviation of the pressure difference of each monitoring point in adjacent time periods, the severity of the pressure fluctuation is captured, and then sudden abnormal events such as water hammer effect are identified. Based on historical monitoring data, a machine learning algorithm is used to construct a residual distribution benchmark to provide a scientific and quantitative basis for subsequent abnormal judgment, calculate the real-time pressure residual, and calculate the comprehensive abnormality index in combination with the residual distribution benchmark. Through multi-dimensional features Fusion, including time series fluctuation characteristics and spatial weight characteristics, realizes accurate anomaly judgment and early warning, effectively distinguishes equipment failures from real pipe network events, and avoids misjudgment. Finally, Kriging interpolation method and semivariogram function are used to generate the spatial distribution surface of pressure residual, and it is converted into a graphical interface through color gradient mapping to intuitively display the abnormal area, helping operation and maintenance personnel to quickly locate abnormal points, improve emergency response speed, and significantly improve the accuracy and efficiency of water supply network anomaly identification, realizing full process control from data quality assurance to accurate identification of abnormal events, and effectively solving the problems of existing technologies such as unsystematic data verification, lack of spatiotemporal correlation, single anomaly classification dimension, easy misjudgment and low response efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0022] Figure 1 is a flow chart showing a method for identifying abnormalities in a water supply network according to an exemplary embodiment;

[0023] Figure 2 is a flow chart of a method for identifying abnormalities in a water supply network in an exemplary embodiment;

[0024] Figure 3 is a schematic diagram of a water supply network abnormality visualization interface in an exemplary embodiment;

[0025] Figure 4 is a block diagram of a device for identifying abnormalities in a water supply network according to an exemplary embodiment;

[0026] Figure 5 is a hardware structure diagram of an electronic device according to an exemplary embodiment;

[0027] Figure 6It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0028] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0029] It will be understood by those skilled in the art that, unless expressly stated otherwise, the singular forms "a", "an", "said" and "the" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of the present disclosure refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, "connected" or "coupled" as used herein may include wireless connections or wireless couplings. The term "and / or" used herein includes all or any units and all combinations of one or more associated listed items.

[0030] The present invention provides a method for identifying anomalies in a water supply network. Through a closed-loop processing process involving data integrity verification, spatiotemporal correlation analysis, fusion of temporal fluctuation and spatial coupling features, and visualization of abnormal areas, this method achieves hierarchical management and control of the entire process, from data quality assurance to abnormal event identification and location. This method addresses the problems of existing technologies, such as unsystematic data verification, lack of spatiotemporal correlation, single anomaly classification dimension, prone to misjudgment, and low response efficiency. The method is suitable for use in a water supply network anomaly identification device, which can be an electronic device. The water supply network anomaly identification method in the embodiments of the present invention can be applied to a variety of scenarios, such as identifying anomalies in urban water supply systems.

[0031] See also Figure 1 , an embodiment of the present invention provides a method for identifying abnormalities in a water supply network, which is applicable to electronic equipment.

[0032] In the following method embodiments, for ease of description, the execution subject of each step of the method is taken as an electronic device as an example for illustration, but this does not constitute a specific limitation.

[0033] like Figure 1 As shown, the method may include the following steps:

[0034] Step 110 , collecting pressure data from each monitoring point and pre-processing it, using the Pearson correlation coefficient to construct a dynamic pressure correlation matrix based on the pressure data, and performing abnormality judgment and repair on the monitoring equipment.

[0035] In one possible implementation, pressure data is collected from each monitoring point in the water supply network, the timestamps are unified and matched with the geographic coordinates, the valid range of the pressure value is set, the monitoring points with pressure values ​​exceeding the valid range are marked as over-threshold points, the monitoring points with missing data are marked as missing points based on the continuity of the timestamps, the data integrity of each monitoring point is calculated based on the number of over-threshold points and missing points, and if the data integrity is less than the preset threshold, an abnormality warning is triggered.

[0036] The pressure data includes timestamp, pressure value, geographical coordinates of the monitoring point, number and water supply system, etc., which are not limited here.

[0037] In one possible implementation, a set sliding window is used to traverse the pressure time series according to the pressure data, the spatial proximity relationship of the monitoring points is quantified based on the Euclidean distance, and multiple neighboring points of the monitoring point are determined. The Pearson correlation coefficient is used to construct a dynamic pressure correlation matrix according to the pressure values ​​in the sliding window of each monitoring point. The pressure correlation coefficient of each monitoring point and multiple neighboring points is calculated based on the dynamic pressure correlation matrix, and the maximum value of the coefficient is taken. If the maximum value of the coefficient is less than a preset threshold, it is determined that the pressure fluctuation correlation of the monitoring point is abnormal, triggering an abnormal warning and repair of the monitoring equipment, and deleting the relevant data of the monitoring point from the pressure data.

[0038] In the above process, the embodiment of the present invention comprehensively collects key data of the water supply network, ensuring the integrity and accuracy of the data. By unifying the timestamp and geographic coordinates, and setting the effective range of pressure values, the data quality is effectively improved. The dynamic pressure correlation matrix is ​​constructed by combining geographic distance and time series correlation, which effectively identifies data distortion caused by equipment failure, avoids misjudgment of real pipeline network events, improves data quality, and provides a more accurate data basis for subsequent data analysis and anomaly identification.

[0039] Step 120 , calculating the standard deviation of the pressure difference at each monitoring point in adjacent time periods to identify the water hammer effect.

[0040] In one possible implementation, the absolute value of the pressure difference at each monitoring point in adjacent time periods is calculated, the pressure difference standard deviation is calculated based on the absolute value of the pressure difference, and a pressure difference standard deviation threshold is set. If the pressure difference standard deviation exceeds the threshold, a water hammer event alarm is triggered.

[0041] Among them, the pressure difference standard deviation is used to reflect the severity of pressure fluctuations.

[0042] Step 130: Based on the historical monitoring data, a residual distribution benchmark is constructed through a machine learning algorithm, the real-time pressure residual of each monitoring point is calculated, a comprehensive anomaly index is calculated based on the residual distribution benchmark and the real-time pressure residual, and anomaly judgment and warning are performed based on the comprehensive anomaly index.

[0043] In one possible implementation, the pipeline network pressure is predicted based on historical monitoring data through a machine learning algorithm to obtain a predicted value. The pressure residuals at different time points of each monitoring point are calculated based on the predicted value and the true value. The residual mean and standard deviation of each monitoring point are calculated based on the pressure residuals. The residual distribution benchmark under normal conditions is constructed based on the residual mean and standard deviation.

[0044] In one possible implementation, the pressure anomaly index of the monitoring point is calculated using the following formula:

[0045] .

[0046] Furthermore, according to the hydraulic importance of each monitoring point in the water supply network, the pressure anomaly index of each monitoring point with a pressure anomaly index greater than 0 is weightedly summed to obtain a comprehensive anomaly index, and a first-level warning threshold and a second-level warning threshold are set. When the comprehensive anomaly index is not greater than the first-level warning threshold, the state is judged to be normal. When the comprehensive anomaly index is greater than the first-level warning threshold and not greater than the second-level warning threshold, a medium-risk alarm is triggered and manual verification is started. When the comprehensive anomaly index is greater than the second-level warning threshold, a high-risk alarm is triggered and the emergency response mechanism is started.

[0047] Among them, when the pressure anomaly index is less than or equal to 0, the pressure anomaly index of the monitoring point is 0 and does not participate in the subsequent comprehensive anomaly index calculation. When the pressure anomaly index is greater than 0, the pressure of the monitoring point is abnormal and participates in the comprehensive anomaly index calculation.

[0048] In the above process, the embodiment of the present invention realizes multi-dimensional anomaly judgment by combining time series fluctuation characteristics and spatial weight characteristics. Through the pressure difference standard deviation and residual distribution benchmark, it effectively identifies events such as water hammer effect and pipe burst, improves the accuracy and efficiency of anomaly identification, and realizes timely warning of abnormal events in the water supply network.

[0049] Step 140 , using Kriging interpolation and semivariogram to generate a pressure residual spatial distribution surface based on the real-time pressure residual, and using color gradient mapping to convert the spatial distribution surface into a graphical interface and display it.

[0050] In one possible implementation, for monitoring points with a pressure anomaly index greater than 0, the Kriging interpolation method is used to quantify the spatial autocorrelation of the monitoring points based on the pressure data of the monitoring points. The spatial variation pattern of the pressure residuals of the monitoring points is analyzed through the semivariogram based on the real-time pressure residuals of the monitoring points. The theoretical model is determined according to the residual distribution benchmark. The semivariogram is fitted with the theoretical model to construct a mathematical model to characterize the spatial correlation of the pressure residuals. The results of the Kriging interpolation method are combined to generate a continuous spatial distribution surface of the pressure residuals.

[0051] In the above process, the embodiment of the present invention uses Kriging interpolation and semivariogram to realize spatial autocorrelation modeling of pressure residuals, and converts complex spatial data into an intuitive graphical interface through color gradient mapping, which facilitates operation and maintenance personnel to quickly locate abnormal areas, improves the accuracy and efficiency of abnormal area positioning, and reduces the cost of pipeline accident handling.

[0052] Through the above process, the present invention realizes efficient and accurate identification of water supply network anomalies through four main steps: data collection and preprocessing, dynamic pressure correlation matrix construction and equipment anomaly judgment, pipe network event identification and early warning, and abnormal area visualization and positioning. It combines spatiotemporal coupling modeling and multi-feature fusion technology to effectively solve the problems of unsystematic data verification, fuzzy anomaly classification and low response efficiency in the existing technology, and provides strong support for the safe operation and maintenance of water supply networks.

[0053] In an exemplary embodiment, the process of identifying abnormalities in a water supply network using the method for identifying abnormalities in a water supply network of the present invention is demonstrated.

[0054] like Figure 2 As shown, the following steps may be included:

[0055] Step S1: data collection and processing.

[0056] Step S1.1, data acquisition.

[0057] Specifically, monitoring devices, such as pressure sensors, deployed at key locations in the water supply network collect real-time pressure data at a frequency of minutes. This data includes timestamps, pressure values, the geographic coordinates of the monitoring point, its number, and the water supply system to which it belongs. For example, a city's water supply network may have 100 pressure sensors installed, each transmitting its measured pressure data to a data center every minute.

[0058] Step S1.2, data processing.

[0059] Spatiotemporal alignment: Due to time synchronization errors among various monitoring devices, the collected data are first time-aligned. Using a unified time base with minute-level accuracy, the data of all monitoring points are organized according to timestamps to ensure that data at the same moment can be compared in the same time dimension. At the same time, the geographic coordinates of the monitoring points are matched to associate the spatial location information with the pressure data to prepare for subsequent spatial analysis.

[0060] Invalid Data Removal: A valid range for pressure values ​​is set. For example, based on the normal operating pressure range of the pipeline network, the lower limit is set to 0.1 MPa and the upper limit is set to 0.6 MPa. Pressure data outside this range is marked as invalid and removed. Timestamp continuity is also checked, and data points with missing or discontinuous timestamps are also marked and removed. This processing improves data integrity and accuracy, providing a reliable foundation for subsequent analysis.

[0061] Step S2: Identify faults of monitoring equipment.

[0062] Step S2.1, data integrity anomaly identification.

[0063] Data integrity analysis: Count the number of missing data at each monitoring point over a period of time (e.g., one day). For example, calculate the proportion of missing data at each monitoring point to the total data duration within one day.

[0064] Data integrity anomaly determination: If the data missing ratio of a monitoring point exceeds a certain threshold (such as 30%), it is determined that the monitoring point has a data integrity anomaly, and the data of the monitoring point with the anomaly will not be included in the subsequent steps.

[0065] Step S2.2, data accuracy anomaly identification.

[0066] Multi-point spatiotemporal correlation analysis: Quantify the spatial proximity of monitoring points using Euclidean distance and identify multiple neighbors for each monitoring point. For example, for a monitoring point A, the five closest monitoring points are selected as neighbors. A dynamic pressure correlation matrix is ​​constructed using the Pearson correlation coefficient based on the mean pressure values ​​within a sliding window (e.g., a 24-hour window) for each monitoring point. The pressure correlation coefficient between each monitoring point and its neighbors is calculated, and the maximum value is taken.

[0067] The dynamic pressure correlation matrix reflects the temporal and spatial correlations between pressures at different monitoring points in a water supply network. The values ​​in the matrix provide a visual indicator of which monitoring points have high correlations and which have weaker correlations. This is crucial for understanding the operational status of water supply networks and the mechanisms by which anomalies propagate.

[0068] Furthermore, the maximum value of the correlation coefficient with the target point is selected for comparison because it represents the strongest correlation between the target point and its neighboring points. If this maximum value is lower than the preset threshold, it indicates that the correlation between the pressure fluctuations of the target point and its neighboring points is abnormal, which may indicate a problem with the monitoring equipment.

[0069] Data Accuracy Abnormality Determination: The pressure correlation coefficient threshold is set to 0.8. If the maximum pressure correlation coefficient at a monitoring point is less than 0.8, the pressure fluctuation correlation at that monitoring point is considered abnormal, indicating a possible equipment failure. This triggers a monitoring equipment abnormality alert, and the data from that monitoring point is disregarded in subsequent steps to avoid interference with subsequent analysis.

[0070] Step S3: Pipeline network event identification.

[0071] Step S3.1, water hammer event identification.

[0072] Single-point fluctuation anomaly analysis: Calculate the absolute value of the pressure difference at each monitoring point in adjacent time periods, and calculate the pressure difference standard deviation based on the absolute value of the pressure difference.

[0073] Water hammer event determination: Water hammer events are manifested as violent fluctuations in pressure, with sudden and high-frequency characteristics. A pressure difference standard deviation threshold is set. For example, based on historical data and pipeline network characteristics, the threshold is set to 0.05MPa. If the pressure difference standard deviation of a monitoring point exceeds 0.05MPa, it is considered that there is an abnormal pressure fluctuation at the monitoring point, and a water hammer event is determined to have occurred, triggering a water hammer event warning.

[0074] Step S3.2: Identification of pipe burst and other events.

[0075] Pressure prediction and anomaly analysis: Based on historical monitoring data, pipeline network pressure is predicted using machine learning algorithms (such as time series prediction models) to obtain predicted values. The predicted values ​​and actual measured values ​​at each monitoring point are subtracted to obtain pressure residuals. The residual mean and standard deviation are calculated based on the pressure residuals as the distribution characteristics of the residuals under normal conditions. A residual distribution benchmark under normal operating conditions is established to provide a quantitative basis for real-time anomaly judgment.

[0076] Furthermore, the real-time pressure residual of each monitoring point is calculated, that is, the difference between the actual pressure value and the predicted value, and then the pressure anomaly index of each monitoring point is calculated using the following formula:

[0077] .

[0078] Furthermore, the pressure anomaly index of each monitoring point is calculated. If the pressure anomaly index is less than or equal to 0, the pressure anomaly index of the monitoring point is 0 and does not participate in the subsequent comprehensive anomaly index calculation. Otherwise, it indicates pressure anomaly. The set multiple in the formula is typically 2 or 3, which is used to quantify the degree of abnormal deviation. The degree of abnormality is represented by the multiple of the residual exceeding the normal standard deviation.

[0079] Determination of events such as pipe bursts: Considering the differences in the hydraulic importance of monitoring points in the pipeline network, the pressure anomaly index of each monitoring point is weighted and summed to obtain a comprehensive anomaly index. Based on historical pipe burst event data and pipeline network operating conditions, the first-level warning threshold and the second-level warning threshold are set, for example, the first-level warning threshold is 5 and the second-level warning threshold is 10. When the comprehensive anomaly index is no greater than 5, the status is determined to be normal; when the comprehensive anomaly index is greater than 5 and no greater than 10, a medium-risk alarm is triggered and manual verification is initiated; when the comprehensive anomaly index is greater than 10, a high-risk alarm is triggered, and it is determined that a serious event such as a pipe burst may occur.

[0080] Step S4: abnormality determination and output.

[0081] Specifically, the results of monitoring equipment fault identification, water hammer event identification, and pipe burst event identification are summarized and output as event identification results. If any of these types of events are determined to have occurred, the abnormality determination phase is entered, and it is determined that an abnormality exists in the water supply network.

[0082] Step S5: Visualization of abnormal alarm.

[0083] Step S5.1, constructing a spatial model based on spatiotemporal data.

[0084] Specifically, a spatial model of the water supply network is constructed based on spatiotemporal data, combined with the geographic coordinates and pressure data of the monitoring points. For example, the network is presented as a map, with each monitoring point marked on the map and annotated according to its pressure value and anomalies.

[0085] Step S5.2: Visualize the abnormal area.

[0086] like Figure 3 As shown in the figure, the spatial distribution surface of pressure residuals is converted into a graphical interface using color gradient mapping. Abnormal areas are marked with a dark color (such as dark gray), visually demonstrating their location and extent within the pipeline network. This visual interface allows operations and maintenance personnel to quickly locate abnormal points and take timely action to address them, reducing the cost of handling pipeline network incidents.

[0087] Through the above process, the embodiment of the present invention first ensures data quality through pre-processing such as minute-level pressure data collection and spatiotemporal alignment, and invalid data elimination. Secondly, data integrity analysis and multi-point spatiotemporal correlation analysis are used to accurately identify monitoring equipment failures and avoid misjudgments. Then, through pressure difference standard deviation calculation and dynamic residual analysis, events such as water hammer and pipe burst are effectively identified, and a graded early warning is achieved by combining the comprehensive anomaly index weighted by hydraulic importance. Subsequently, the event identification results are summarized and anomalies are determined to ensure timely response. Finally, a spatial model is constructed based on spatiotemporal data, and the color gradient mapping technology is used to visualize the spatial distribution of pressure residuals, intuitively presenting abnormal areas, and assisting operation and maintenance personnel in quickly locating and processing them. This embodiment verifies the remarkable effectiveness of this method in improving the accuracy and efficiency of anomaly identification in water supply networks, and provides strong support for safe operation and maintenance of pipeline networks.

[0088] In one application scenario, abnormal pressure fluctuations suddenly occur at pressure monitoring point A of a city's water supply network, and the pressure data lose synchronization with the pressure data of other nearby monitoring points. The abnormality identification method of an embodiment of the present invention is applied to identify abnormalities in the monitoring equipment.

[0089] Specifically, the minute-level pressure data of monitoring point A and its four adjacent monitoring points are first obtained from the SCADA system, including timestamps, pressure values, geographic coordinates and other information. The collected data are aligned in time and space to ensure that the timestamps are consistent and match the geographic coordinates. Then, data missing points and exceeding threshold points are marked (the pressure range is set to 0.05MPa to 0.8MPa).

[0090] Furthermore, a sliding window with a length of 24 hours was used to traverse the pressure time series, and the Pearson correlation coefficient between monitoring point A and its neighboring points in each window was calculated. A dynamic pressure correlation matrix was constructed, and the maximum correlation coefficient between monitoring point A and its neighboring points was extracted. It was found that it was significantly lower than the preset threshold of 0.6, triggering the "monitoring equipment data accuracy abnormality" alarm, prompting the operation and maintenance personnel to check the equipment at monitoring point A.

[0091] Finally, after on-site inspection by the operation and maintenance personnel, it was found that the sampling tube at monitoring point A was blocked, resulting in data distortion. After the repair, the pressure fluctuations at monitoring point A and the adjacent points resumed synchronization, and the correlation coefficient rose to above 0.99, verifying the effectiveness of the accuracy verification method based on spatiotemporal correlation.

[0092] In another application scenario, in the pressurization area of ​​a municipal pump station, the pressure monitoring point B experiences a sharp pressure fluctuation during the peak water usage period, and a water hammer event is suspected. The abnormality recognition method of an embodiment of the present invention is applied to identify the water hammer event.

[0093] Specifically, the minute-level pressure data of monitoring point B is first obtained from the SCADA system, and the data is temporally and spatially aligned and invalid data is eliminated. The absolute value of the pressure difference between adjacent time points of monitoring point B within the analysis period is calculated, and the standard deviation of the pressure difference is calculated based on these differences.

[0094] Furthermore, it was found that the standard deviation of the pressure difference at monitoring point B was significantly higher than the preset threshold of 0.01 MPa, triggering a "water hammer event" alarm and prompting the operation and maintenance personnel to check the pump station equipment.

[0095] Finally, after on-site verification, it was found that the failure of the valve after the pump caused a sudden change in water flow rate, triggering a water hammer effect. After replacing the faulty valve and expanding the main water supply pipeline, the pressure fluctuation tended to stabilize and the standard deviation of the pressure difference returned to the normal range, verifying the effectiveness of the water hammer event identification method based on dynamic pressure difference analysis.

[0096] In another application scenario, in a certain water supply network area, the pressure monitoring data shows abnormal fluctuations, and a pipe burst is suspected. The abnormality identification method of an embodiment of the present invention is applied to identify pressure residuals.

[0097] Specifically, we first obtain minute-level pressure data from multiple monitoring points in the pipeline network from the SCADA system, perform spatiotemporal alignment on the data, and eliminate invalid data. We then build a pressure prediction model based on historical monitoring data, predict the pressure value of each monitoring point, and calculate the real-time pressure residual of each monitoring point, that is, the difference between the actual pressure value and the predicted value.

[0098] Furthermore, a weighted summation is performed based on the hydraulic importance of each monitoring point to obtain a comprehensive anomaly index, and a first-level warning threshold of 50 and a second-level warning threshold of 200 are set.

[0099] A pipe burst occurred at 17:17. The comprehensive anomaly index was 139 at 17:18. Because 50<139≤200, a medium-risk alarm for pipe burst and other events was triggered. The comprehensive anomaly index was 327 at 17:19. Because 327>200, a high-risk alarm for pipe burst and other events was triggered. After the water supply was cut off and the valve was closed in the burst section, the comprehensive anomaly index fell back to the normal range as the pipeline network pressure recovered. The comprehensive anomaly index effectively reflected the abnormal evolution law of pipeline network pressure during the development and treatment of the pipe burst, verifying the effectiveness of the pipe burst event identification method based on multi-feature fusion.

[0100] In another application scenario, in a water supply network area, multiple abnormal points are identified by the abnormality identification method of an embodiment of the present invention, and they need to be visualized so that operation and maintenance personnel can quickly locate them.

[0101] Specifically, the geographic coordinates of the real-time abnormal monitoring points and the corresponding real-time pressure residuals are extracted, the spatial autocorrelation is quantified using the Kriging interpolation method, a pressure residual spatial distribution model is constructed, and a continuous pressure residual spatial distribution surface is generated based on the Kriging interpolation results, which is then visualized through a geographic information system (GIS) platform.

[0102] Furthermore, the preset color gradient mapping rule is used to transform the spatial distribution of pressure residuals into an intuitive graphical interface.

[0103] Finally, the operation and maintenance personnel can quickly locate concentrated areas of abnormal pressure fluctuations through the visual interface, such as high-risk areas for pipe bursts, activate the emergency response mechanism based on the visualization results, and carry out pipeline network repair work.

[0104] Through the above embodiments, we can clearly see the effectiveness and practicality of the water supply network anomaly identification method and system based on spatiotemporal coupling of pressure data and multi-feature fusion proposed in the present invention in practical applications.

[0105] Through the above process, the practical application effect of the water supply network anomaly identification method provided by the embodiment of the present invention is fully demonstrated. Through the dynamic pressure correlation matrix, the accuracy anomalies of the monitoring equipment data are accurately identified, and the misjudgment caused by equipment failure is avoided. Using the pressure difference standard deviation analysis, the water hammer event is successfully captured, and the effectiveness of the dynamic pressure difference analysis method is verified. Through multi-feature fusion and comprehensive anomaly index calculation, the pipe burst event is accurately identified, and the graded early warning of the event is realized. Finally, through the Kriging interpolation method and visualization technology, the distribution of abnormal areas is intuitively displayed, assisting operation and maintenance personnel to quickly locate and handle abnormal points. These embodiments jointly demonstrate the significant advantages of the present invention in improving the accuracy and efficiency of water supply network anomaly identification, and provide strong technical support for the safe operation and maintenance of water supply networks.

[0106] The following are device embodiments of the present invention, which can be used to implement the water supply network anomaly identification method involved in the present invention. For details not disclosed in the device embodiments of the present invention, please refer to the method embodiments of the water supply network anomaly identification method involved in the present invention.

[0107] See also Figure 4 , an embodiment of the present invention provides a water supply network abnormality identification device 800.

[0108] The water supply network anomaly identification device 800 includes but is not limited to: a data acquisition and equipment maintenance module 810, an anomaly identification and pressure analysis module 830, an anomaly calculation and risk warning module 850 and a spatial distribution generation and display module 870.

[0109] Among them, the data acquisition and equipment maintenance module 810 is used to collect pressure data from each monitoring point and pre-process it, use the Pearson correlation coefficient to construct a dynamic pressure correlation matrix based on the pressure data, and perform abnormal judgment and repair on the monitoring equipment; the pressure data includes timestamp, pressure value, geographical coordinates of the monitoring point, number and water supply system to which it belongs.

[0110] The abnormality identification and pressure analysis module 830 is used to calculate the standard deviation of the pressure difference at each monitoring point in adjacent time periods to identify the water hammer effect.

[0111] The anomaly calculation and risk warning module 850 constructs a residual distribution benchmark based on historical monitoring data through a machine learning algorithm, which is used to calculate the real-time pressure residual of each monitoring point, calculate the comprehensive anomaly index based on the residual distribution benchmark and the real-time pressure residual, and perform anomaly judgment and warning based on the comprehensive anomaly index.

[0112] The spatial distribution generation and display module 870 is used to generate a pressure residual spatial distribution surface based on the real-time pressure residual using the Kriging interpolation method and the semivariogram function, and convert the spatial distribution surface into a graphical interface and display it using color gradient mapping.

[0113] It should be noted that the water supply network abnormality identification provided in the above embodiment is only illustrated by the division of the above-mentioned functional modules. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the water supply network abnormality identification device will be divided into different functional modules to complete all or part of the functions described above.

[0114] In addition, the water supply network anomaly identification device and the water supply network anomaly identification method provided in the above embodiments belong to the same concept, and the specific manner in which each module performs operations has been described in detail in the method embodiment and will not be repeated here.

[0115] Figure 5 The following is a schematic diagram of the structure of an electronic device according to an exemplary embodiment.

[0116] It should be noted that the electronic device is only an example adapted to the present invention and cannot be considered to provide any limitation on the scope of use of the present invention. The electronic device cannot be interpreted as needing to rely on or must have Figure 5 One or more components of exemplary electronic device 2000 are shown.

[0117] The hardware structure of the electronic device 2000 may vary greatly due to different configurations or performances, such as Figure 5As shown, the electronic device 2000 includes a power supply 210 , an interface 230 , at least one memory 250 , and at least one central processing unit (CPU) 270 .

[0118] Specifically, the power supply 210 is used to provide operating voltage for various hardware devices on the electronic device 2000 .

[0119] The interface 230 includes at least one wired or wireless network interface 231 for interacting with external devices. Of course, in other examples adapted by the present invention, the interface 230 may further include at least one serial-to-parallel conversion interface 233, at least one input / output interface 235, and at least one USB interface 237, etc. Figure 5 As shown, this does not constitute a specific limitation.

[0120] The memory 250 serves as a carrier for resource storage and can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon include an operating system 251, application 253 and data 255, etc. The storage method can be temporary storage or permanent storage.

[0121] Among them, the operating system 251 is used to manage and control the various hardware devices and application programs 253 on the electronic device 2000, so as to enable the central processing unit 270 to calculate and process the massive data 255 in the memory 250. It can be WindowsServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0122] The application 253 is a computer-readable instruction that performs at least one specific task based on the operating system 251, and may include at least one module ( Figure 5 (not shown), each module may include computer-readable instructions for the electronic device 2000. For example, the water supply network anomaly identification device may be considered as an application 253 deployed on the electronic device 2000.

[0123] The data 255 may be signal information, etc., and is stored in the memory 250 .

[0124] The central processing unit 270 may include one or more processors and is configured to communicate with the memory 250 via at least one communication bus to read computer-readable instructions stored in the memory 250, thereby performing operations and processing on the massive amount of data 255 in the memory 250. For example, the water supply network anomaly identification method can be implemented by the central processing unit 270 reading a series of computer-readable instructions stored in the memory 250.

[0125] In addition, the present invention can also be implemented through hardware circuits or hardware circuits combined with software. Therefore, the implementation of the present invention is not limited to any specific hardware circuits, software, or combination thereof.

[0126] See also Figure 6 In an embodiment of the present invention, an electronic device 4000 is provided. The electronic device 4000 may include: a desktop computer, a laptop computer, a server, etc. with sensor recognition capabilities.

[0127] exist Figure 6 In the embodiment, the electronic device 4000 includes at least one processor 4001 and at least one memory 4003.

[0128] The data exchange between the processor 4001 and the memory 4003 can be achieved through at least one communication bus 4002. The communication bus 4002 may include a path for transmitting data between the processor 4001 and the memory 4003. The communication bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The communication bus 4002 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0129] Optionally, the electronic device 4000 may further include a transceiver 4004, which may be used for data exchange between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the number of transceivers 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0130] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the present disclosure. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0131] The memory 4003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program instructions or codes in the form of instructions or data structures and can be accessed by the electronic device 4000, but is not limited to these.

[0132] Computer-readable instructions are stored in the memory 4003 , and the processor 4001 can read the computer-readable instructions stored in the memory 4003 through the communication bus 4002 .

[0133] The computer-readable instructions are executed by one or more processors 4001 to implement the water supply network anomaly identification method in the above-mentioned embodiments.

[0134] In addition, an embodiment of the present invention provides a storage medium having computer-readable instructions stored thereon. The computer-readable instructions are executed by one or more processors to implement the water supply network anomaly identification method as described above.

[0135] A computer program product is provided in an embodiment of the present invention. The computer program product includes computer-readable instructions, which are stored in a storage medium. One or more processors of an electronic device read the computer-readable instructions from the storage medium, load and execute the computer-readable instructions, so that the electronic device implements the water supply network abnormality identification method as described above.

[0136] Compared with the related art, the present invention has the following beneficial effects:

[0137] 1. The present invention can significantly improve the reliability of water supply network anomaly identification; through a layered and progressive data verification system, including a double-layer filtering mechanism of integrity verification and accuracy verification, it can effectively eliminate distorted data caused by equipment failure.

[0138] 2. The present invention has the ability to accurately distinguish between equipment failures and real pipeline network events; through spatiotemporal coupling modeling, it integrates the spatiotemporal correlation of geographic coordinates and pressure data, and uses the dynamic pressure correlation matrix and Euclidean distance to screen neighboring points.

[0139] 3. The present invention can enhance the multidimensionality and accuracy of pipeline network event identification; through multi-feature fusion technology, combined with the pressure fluctuation time series characteristics, spatial correlation characteristics and hydraulic importance weights, a comprehensive anomaly index model is constructed.

[0140] 4. The present invention has the advantages of quickly locating abnormal areas and improving response efficiency; through pressure residual space modeling and visualization technology based on Kriging interpolation, the distribution of abnormal areas is intuitively presented, assisting operation and maintenance personnel in making quick decisions.

[0141] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0142] The above descriptions are only partial embodiments of the present invention. It should be pointed out that ordinary technicians in this technical field can make several improvements and modifications without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for identifying abnormalities in a water supply network, characterized in that: The method comprises: Pressure data is collected from each monitoring point and pre-processed, and a dynamic pressure correlation matrix is ​​constructed based on the pressure data using the Pearson correlation coefficient to identify and repair abnormalities in the monitoring equipment; the pressure data includes a timestamp, pressure value, geographic coordinates of the monitoring point, number, and water supply system to which it belongs; Calculating the standard deviation of the pressure difference at each of the monitoring points in adjacent time periods to identify the water hammer effect; Based on historical monitoring data, a machine learning algorithm is used to predict the pipe network pressure to obtain a predicted value, and the pressure residuals at different time points of each monitoring point are calculated based on the predicted value and the true value; the residual mean and standard deviation of each monitoring point are calculated based on the pressure residuals, and a residual distribution benchmark under normal conditions is constructed based on the residual mean and standard deviation; The real-time pressure residual of each monitoring point is calculated, and the pressure anomaly index of the monitoring point is calculated by the following formula: ； According to the hydraulic importance of each monitoring point in the water supply network, the pressure anomaly index of each monitoring point whose pressure anomaly index is greater than 0 is weightedly summed to obtain a comprehensive anomaly index; setting a first-level warning threshold and a second-level warning threshold, and determining that the state is normal when the comprehensive anomaly index is not greater than the first-level warning threshold; When the comprehensive abnormality index is greater than the first-level warning threshold and not greater than the second-level warning threshold, a medium-risk alarm is triggered and manual verification is initiated; when the comprehensive abnormality index is greater than the second-level warning threshold, a high-risk alarm is triggered and an emergency response mechanism is initiated; A pressure residual spatial distribution surface is generated according to the real-time pressure residual using Kriging interpolation and semivariogram, and the spatial distribution surface is converted into a graphical interface and displayed using color gradient mapping.

2. The water supply network abnormality identification method according to claim 1, characterized in that: The pressure data collected from each monitoring point and pre-processed include: Collect pressure data from each monitoring point in the water supply network, unify the timestamps and match them with the geographic coordinates, and set the effective range of the pressure value; Marking the monitoring points where the pressure value exceeds the valid range as over-threshold points, and marking the monitoring points where data is missing as missing points based on the continuity of the timestamps; The data integrity of each monitoring point is calculated based on the number of the exceeding threshold points and the missing points. If the data integrity is less than a preset threshold, an abnormality warning is triggered.

3. The water supply network abnormality identification method according to claim 1, characterized in that: The method of constructing a dynamic pressure correlation matrix based on the pressure data using the Pearson correlation coefficient to perform abnormality judgment and repair on the monitoring equipment includes: Using a set sliding window to traverse the pressure time series according to the pressure data, quantifying the spatial proximity relationship of the monitoring points based on the Euclidean distance, and determining multiple neighboring points of the monitoring point; A dynamic pressure correlation matrix is ​​constructed based on the pressure values ​​within the sliding window of each monitoring point using the Pearson correlation coefficient, and the pressure correlation coefficient between each monitoring point and a plurality of adjacent points is calculated based on the dynamic pressure correlation matrix, and the maximum coefficient is taken; If the maximum value of the coefficient is less than a preset threshold, the pressure fluctuation correlation of the monitoring point is determined to be abnormal, an abnormal warning of the monitoring equipment is triggered and repaired, and the relevant data of the monitoring point is deleted from the pressure data.

4. The water supply network abnormality identification method according to claim 1, characterized in that: Calculating the pressure difference standard deviation of each monitoring point in adjacent time periods to identify the water hammer effect includes: Calculate the absolute value of the pressure difference at each monitoring point in adjacent time periods, calculate the pressure difference standard deviation based on the absolute value of the pressure difference, set a pressure difference standard deviation threshold, and trigger a water hammer event alarm if the pressure difference standard deviation exceeds the threshold; the pressure difference standard deviation is used to reflect the severity of pressure fluctuations.

5. The water supply network abnormality identification method according to claim 1, characterized in that: The method of using the Kriging interpolation method and the semivariogram function to generate a pressure residual spatial distribution surface according to the real-time pressure residual comprises: For the monitoring points where the pressure anomaly index is greater than 0, the spatial autocorrelation of the monitoring points is quantified based on the pressure data of the monitoring points using the Kriging interpolation method, and the spatial variation pattern of the pressure residual of the monitoring points is analyzed using the semivariogram based on the real-time pressure residual of the monitoring points; A theoretical model is determined based on the residual distribution benchmark, and a mathematical model characterizing the spatial correlation of pressure residuals is constructed by fitting the semivariogram with the theoretical model. A continuous pressure residual spatial distribution surface is generated in combination with the results of the Kriging interpolation method.

6. A water supply network abnormality identification device, characterized in that: The device is applied to the water supply network abnormality identification method according to any one of claims 1 to 5, and the device comprises: The data acquisition and equipment maintenance module is used to collect and pre-process pressure data from each monitoring point, construct a dynamic pressure correlation matrix based on the pressure data using the Pearson correlation coefficient, and perform abnormality judgment and repair on the monitoring equipment. The pressure data includes the timestamp, pressure value, geographic coordinates of the monitoring point, number, and water supply system to which it belongs; Anomaly identification and pressure analysis module, used to calculate the standard deviation of the pressure difference of each monitoring point in adjacent time periods to identify the water hammer effect; Anomaly calculation and risk warning module, which constructs a residual distribution benchmark based on historical monitoring data through a machine learning algorithm, is used to calculate the real-time pressure residual of each monitoring point, calculates a comprehensive anomaly index based on the residual distribution benchmark and the real-time pressure residual, and performs anomaly judgment and warning based on the comprehensive anomaly index; The spatial distribution generation and display module is used to generate a pressure residual spatial distribution surface according to the real-time pressure residual using the Kriging interpolation method and the semivariogram function, and convert the spatial distribution surface into a graphical interface and display it using color gradient mapping.

7. An electronic device, characterized in that: include: at least one processor and at least one memory, wherein: The memory has computer-readable instructions stored thereon; The computer-readable instructions are executed by one or more of the processors, so that the electronic device implements the water supply network abnormality identification method according to any one of claims 1 to 5.

8. A storage medium having computer-readable instructions stored thereon, characterized in that: The computer-readable instructions are executed by one or more processors to implement the water supply network anomaly identification method according to any one of claims 1 to 5.

Images