Intelligent water data management method and system

Abstract

The application provides a smart water data management method and system, comprising: determining adjacent leakage points through distribution density change trend, identifying pressure gradient between adjacent leakage points, and determining single-point water output correction proportion; verifying the single-point water output correction proportion with the instantaneous flow value in the leakage point related data set to obtain a flow deviation value, identifying the leakage point type according to the flow deviation value and flow fluctuation characteristics, and adjusting the leakage contribution degree evaluation weight table according to the leakage point type; summarizing total leakage water quantity from the classified real leakage water quantity components, analyzing the change rate in combination with historical total leakage water quantity data, evaluating whether the change rate reflects the total water quantity decline phenomenon under the increase of the number of leakage points, and obtaining a total leakage water quantity verification result; generating a pipe network health evaluation report by integrating the partition pressure data and the distribution density change trend according to the total leakage water quantity verification result, and evaluating the overall accuracy improvement degree through the cumulative value of the real leakage water quantity components in the report.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a smart water data management method and system. Background Technology

[0002] With the acceleration of urbanization, the healthy operation of water supply networks is directly related to residents' quality of life and sustainable urban development. However, network leakage remains a core challenge affecting water supply efficiency, necessitating the use of scientific methods to improve the accuracy of leakage assessment in order to reduce water waste. Currently, although many methods attempt to estimate total water loss by counting the number of leakage points, these methods often ignore the mutual influence and specific differences between leakage points, leading to significant deviations between assessment results and actual conditions. Especially when leakage points are densely distributed spatially, simply relying on numerical accumulation or averaging estimates fails to accurately reflect the dynamic changes in water loss.

[0003] A deeper technical challenge lies in the core factor of hydraulic interference between leakage points. Hydraulic interference refers to the phenomenon where the actual water output of adjacent leakage points is not independent due to the interaction of water pressure and flow rate, but is reduced by the influence of surrounding leakage points. This phenomenon makes calculating the total leakage volume exceptionally complex. Especially in high-density areas, when multiple small-scale leakage points cluster together, the water output of a single point may be significantly reduced due to pressure dispersion, but the increase in the overall number can mislead people into believing that the total leakage volume has increased. Specifically, in a section of a city's water supply network, suppose multiple small cracks cause leakage points to appear in a short period. Although these points are numerous, the water output of each point is far lower than expected due to hydraulic interference from adjacent locations. If this interference effect is not considered during the assessment, judging the leakage as severe solely based on the increase in the number of points may overestimate the actual loss, thus affecting resource allocation and maintenance priority.

[0004] Therefore, accurately identifying the impact of hydraulic disturbances on the outflow of a single point when the density of leakage points changes, and scientifically calculating the total leakage volume accordingly, has become a key issue in the health assessment of smart water management networks. Summary of the Invention

[0005] This invention provides a smart water data management method, the method comprising:

[0006] The instantaneous flow rate, zone pressure data, spatial coordinate information, historical total water loss data, and leakage contribution evaluation weight table of each leakage point are obtained. Noise interference is removed through preliminary filtering, and the flow fluctuation characteristics of each leakage point are extracted to obtain the relevant dataset of leakage points.

[0007] Based on the leak point related dataset, the spatial distribution density around each leak point is identified. The spatial coordinate information is grouped and clustered to obtain leak point cluster groups. By identifying the areas with a high concentration of leak points in the leak point cluster groups, high-density areas are determined. By comparing the changing trends of the spacing between each leak point in the high-density areas, the trend of distribution density change is determined.

[0008] The distribution density change trend is used to determine adjacent leakage points, identify the pressure gradient between adjacent leakage points, and determine the single-point water output correction ratio.

[0009] The single-point water output correction ratio is verified with the instantaneous flow rate value in the dataset related to the leakage point to obtain the flow rate deviation value. The leakage point type is identified based on the flow rate deviation value and the flow rate fluctuation characteristics. The leakage contribution evaluation weight table is adjusted according to the leakage point type.

[0010] Using the adjusted leakage contribution evaluation weight table, combined with the single-point water output correction ratio and the spatial distribution density, the weighted contribution value of each leakage point is extracted, and the weighted contribution value is classified to distinguish between the real leakage signal and the interference signal, and the real leakage water volume component after classification is determined.

[0011] The total leakage volume is summed from the actual leakage volume components after classification, and the change rate is analyzed in combination with the historical total leakage volume data to obtain the verification result of the total leakage volume;

[0012] Based on the total leakage water volume verification results, the zone pressure data and the distribution density change trend are integrated to generate a pipeline health assessment report.

[0013] Furthermore, the process involves acquiring instantaneous flow rates at each leakage point, zone pressure data, spatial coordinate information, historical total leakage water volume data, and a leakage contribution assessment weight table. After preliminary filtering to remove noise interference, the flow fluctuation characteristics of each leakage point are extracted to obtain a leakage point-related dataset, including:

[0014] The system obtains the instantaneous flow rate value of each leakage point from the pipeline monitoring terminal, reads real-time pressure data from the zone pressure sensor, extracts the spatial coordinate information of each leakage point through the GIS system, retrieves the historical total leakage water volume record from the database, generates a leakage contribution evaluation weight table based on the pipe diameter and the location and depth of the leakage point, and uses the moving average method to smooth the instantaneous flow rate value to obtain the smoothed flow rate sequence.

[0015] The smoothed flow sequence is subjected to differential operation in time order to calculate the flow change value of each leakage point at adjacent time points, and the maximum, minimum and average values ​​of flow change per unit time are statistically analyzed to form a flow fluctuation feature vector.

[0016] Based on the change amplitude in the flow fluctuation feature vector, combined with the spatial coordinate information, partition pressure data, and the leakage contribution evaluation weight table, a weight value is applied to the flow fluctuation feature vector for weighted correction, thereby constructing a leakage point related dataset containing flow time series data, pressure distribution data, location information, and weighted fluctuation features.

[0017] Furthermore, the step of identifying the spatial distribution density around each leak point based on the leak point-related dataset, grouping and clustering the spatial coordinate information to obtain leak point cluster groups, identifying high-density areas by identifying regions with a high concentration of leak points in the cluster groups, and determining the distribution density change trend by comparing the changing trends of the spacing between leak points in the high-density areas, includes:

[0018] Based on the spatial coordinate information in the dataset related to the leakage points, the local spatial density values ​​of all leakage points are sorted to form a density distribution sequence;

[0019] The DBSCAN clustering algorithm is used as input, taking the spatial coordinates of each leak point in the density distribution sequence as input, to form a classification of core points, boundary points and noise points. The core points and all points that their density can reach are grouped into the same cluster, and the leak point cluster grouping is output.

[0020] The number of leakage points in each group of the leakage point cluster is counted, the spatial envelope area of ​​each group is calculated, the group density value is obtained by dividing the number of leakage points by the envelope area, the center coordinates and boundary coordinates of the high-density area are recorded, and the leakage points in the high-density area are weighted in combination with the leakage contribution evaluation weight table.

[0021] Furthermore, the step of determining adjacent leakage points by analyzing the distribution density change trend, identifying the pressure gradient between adjacent leakage points, and determining the single-point outflow correction ratio includes:

[0022] Based on the gradient direction of the distribution density change, search for the nearest neighbor of each leakage point along the density increasing direction, determine the spatial distance between each leakage point and its nearest neighbor, record the coordinate information and distance value of all adjacent leakage point pairs, and form an adjacent relationship mapping table.

[0023] Obtain the real-time pressure data of each pair of adjacent leakage points in the adjacent relationship mapping table, determine the pressure difference between the two points, divide the pressure difference by the pipe length between the two points to obtain the pressure change rate per unit length, and use the pressure change rate per unit length as the pressure gradient value between adjacent leakage points.

[0024] By comparing the pressure gradient value with the original flow rate value of each leakage point, the proportion of the pressure gradient value to the original flow rate value is obtained, and the pressure influence coefficient is obtained. Based on the relative positional relationship of the adjacent leakage points and the pressure influence coefficient, the degree of interference of each leakage point with the influence of the adjacent points is determined.

[0025] The original outflow of water at each leakage point is corrected based on the degree of interference, and the corrected actual outflow is obtained. The ratio of the actual outflow to the original outflow is determined, and the correction ratio of the outflow at a single point is determined.

[0026] Furthermore, the step of verifying the single-point outflow correction ratio with the instantaneous flow rate value in the dataset related to the leakage point to obtain a flow rate deviation value, identifying the leakage point type based on the flow rate deviation value and the flow rate fluctuation characteristics, and adjusting the leakage contribution evaluation weight table based on the leakage point type includes:

[0027] The corrected flow rate is obtained by multiplying the single-point outflow correction ratio by the instantaneous flow rate value of each leak point in the leak point related dataset. The difference between the corrected flow rate value and the actual monitored flow rate value of each leak point is calculated to obtain the flow deviation value.

[0028] Based on the magnitude and variation pattern of the flow deviation value, and combined with the fluctuation frequency and amplitude information in the flow fluctuation characteristics, the leakage point type is identified. The standard deviation of the flow deviation value sequence is calculated as the amplitude, and the number of times the deviation value exceeds the average deviation value is divided by the total number of time points as the frequency.

[0029] Different weight adjustment coefficients are assigned to sudden, chronic, and intermittent leakage points, and the weight values ​​of the corresponding leakage points in the leakage contribution evaluation weight table are updated with the adjusted weight coefficients.

[0030] Furthermore, by using the adjusted leakage contribution evaluation weight table, combined with the single-point outflow correction ratio and the spatial distribution density, the weighted contribution value of each leakage point is extracted. The weighted contribution values ​​are then classified to distinguish between real leakage signals and interference signals, and the classified real leakage water volume component is determined, including:

[0031] The weight value of each leakage point is obtained by using the adjusted leakage contribution evaluation weight table. The weight value is multiplied by the single-point outflow correction ratio of the corresponding leakage point, and then multiplied by the normalized spatial density factor of the area where the leakage point is located to obtain the weighted contribution value of each leakage point.

[0032] Based on the distribution characteristics of the weighted contribution values, calculate the mean and standard deviation of the weighted contribution values ​​of all leakage points to complete the signal classification of leakage points;

[0033] For the leakage point corresponding to the real leakage signal, the instantaneous flow rate value of the leakage point in the relevant dataset is extracted. The instantaneous flow rate value is multiplied by the corresponding weighted contribution value to obtain the real leakage water volume at a single point. The real leakage water volumes at all real leakage signal points are accumulated to determine the real leakage water volume components.

[0034] Furthermore, the step of summing the total leakage volume from the classified actual leakage volume components, and analyzing the rate of change in the historical total leakage volume data to obtain the total leakage volume verification result includes:

[0035] The total leakage volume for the current period is obtained by summing the water volume component values ​​of all real leakage points from the classified real leakage volume components. The historical values ​​for the corresponding period are extracted from the historical total leakage volume data, and the leakage volume change rate is calculated.

[0036] The total number of newly added leakage points and the total number of existing leakage points in the current period are counted and compared with the total number of leakage points in the same period in history. The characteristic parameters of this reverse phenomenon are recorded.

[0037] Based on the characteristic parameters of the reverse phenomenon and the rate of change of leakage water volume, the total leakage water volume verification result is output, including the verification mark, the change rate value, and the change in the number of leakage points.

[0038] Furthermore, the step of integrating the zonal pressure data and the distribution density change trend based on the total leakage water volume verification results to generate a pipeline health assessment report includes:

[0039] Based on the verification markers and change rate values ​​in the total leakage water volume verification results, the zone pressure data and the distribution density change trend are integrated, and the data is organized according to time sequence and spatial location to generate a pipeline health assessment report containing verification results, pressure data and density trend information.

[0040] The cumulative value of the actual leakage water volume component is extracted from the pipeline health assessment report and compared with the original value obtained by simply summing the instantaneous flow rate of each leakage point to obtain the overall accuracy improvement.

[0041] On the other hand, the present invention also discloses a smart water data management system, the system comprising:

[0042] The data acquisition and preprocessing module is used to acquire instantaneous flow values, zone pressure data, spatial coordinate information, historical total water loss data, and a weight table for assessing the contribution of leakage at each leakage point. It removes noise interference through preliminary filtering and extracts the flow fluctuation characteristics of each leakage point to obtain a dataset related to the leakage point.

[0043] The spatial clustering and density analysis module is used to identify the spatial distribution density around each leak point based on the leak point related dataset, group the spatial coordinate information to obtain leak point cluster groups, identify high-density areas by identifying areas with a high concentration of leak points in the leak point cluster groups, and determine the trend of distribution density change by comparing the changing trend of the distance between each leak point in the high-density area.

[0044] The pressure gradient and correction module is used to determine adjacent leakage points by the distribution density change trend, identify the pressure gradient between adjacent leakage points, and determine the single-point water output correction ratio.

[0045] The flow verification and type identification module is used to verify the single-point water output correction ratio with the instantaneous flow value in the data set related to the leakage point to obtain the flow deviation value, identify the leakage point type based on the flow deviation value and the flow fluctuation characteristics, and adjust the leakage contribution evaluation weight table based on the leakage point type.

[0046] The weighted contribution classification module is used to extract the weighted contribution value of each leakage point by means of the adjusted leakage contribution evaluation weight table, combined with the single-point water output correction ratio and the spatial distribution density, classify the weighted contribution value to distinguish between real leakage signals and interference signals, and determine the real leakage water volume component after classification.

[0047] The total leakage water volume verification module is used to summarize the total leakage water volume from the classified actual leakage water volume components, analyze the change rate in combination with the historical total leakage water volume data, and obtain the total leakage water volume verification result.

[0048] The health assessment report generation module is used to integrate the zone pressure data and the distribution density change trend based on the total leakage water volume verification results to generate a pipeline health assessment report.

[0049] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0050] This invention discloses a smart water management data management method and system. Addressing the complex business scenarios of leak point identification, distribution characteristic analysis, and water volume assessment in pipeline networks, it integrates the correlation between the spatial distribution density of leak points, pressure gradients, and flow fluctuation characteristics. The aim is to solve the core challenges of inaccurate differentiation between leak signals and interference signals, and large deviations in the total leak volume assessment. This invention performs preliminary filtering and feature extraction on leak point data, combined with spatial clustering and density change trend analysis, to accurately identify high-density areas and adjacent leak points. Then, it corrects the single-point outflow ratio through pressure gradient correction, optimizing flow deviation calculation and leak type determination. Simultaneously, it adjusts the contribution weight table, extracts weighted contribution values ​​to distinguish true leak signals, and finally summarizes the total leak volume and verifies the rate of change to generate a pipeline network health assessment report. This invention significantly improves the accuracy of leak volume components and the reliability of overall assessment, providing a scientific basis for pipeline network management. Attached Figure Description

[0051] Figure 1 This is a flowchart of a smart water data management method according to the present invention.

[0052] Figure 2 This is a schematic diagram of a smart water data management method according to the present invention.

[0053] Figure 3 This is a schematic diagram of the structure of a smart water data management system according to the present invention. Detailed Implementation

[0054] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0055] like Figures 1-3 This embodiment of a smart water data management method and system may specifically include:

[0056] Step S101: Obtain the instantaneous flow rate value, zone pressure data, spatial coordinate information, historical total leakage water volume data, and leakage contribution evaluation weight table for each leakage point. Remove noise interference through preliminary filtering and extract the flow fluctuation characteristics of each leakage point to obtain the relevant dataset of leakage points.

[0057] Instantaneous flow rates at each leak point are obtained from the pipeline monitoring terminal, and real-time pressure data is read from the zone pressure sensors. Spatial coordinate information of each leak point is extracted through the GIS system, and historical total leakage water volume records are retrieved from the database. A leakage contribution assessment weight table is generated based on pipe diameter and leak point location depth. The instantaneous flow rates are smoothed using a moving average method to remove noise interference caused by sensor errors, resulting in a smoothed flow sequence. The smoothed flow sequence is then subjected to differential operations in chronological order to calculate the flow change value of each leak point at adjacent time points. The maximum, minimum, and average values ​​of flow change per unit time are statistically analyzed to form a flow fluctuation feature vector. The fluctuation type is determined based on the amplitude of the change in the flow fluctuation feature vector. If the amplitude exceeds a preset threshold, it is marked as an abnormal fluctuation; otherwise, it is marked as a normal fluctuation. Based on the distribution of normal and abnormal fluctuations, and combined with the spatial coordinate information of each leakage point, the partition pressure data, and the leakage contribution evaluation weight table, weight values ​​are applied to the fluctuation feature vector for weighted correction, and a leakage point related dataset containing flow time series data, pressure distribution data, location information, weighted fluctuation features, and weight values ​​is constructed.

[0058] In one implementation, the pipeline monitoring terminal collects the instantaneous flow rate at leakage points in real time using ultrasonic flow meters installed at various nodes of the pipeline. Each flow meter records the water flow rate according to a preset sampling frequency. The sampling frequency is dynamically adjusted based on pipeline pressure fluctuations; when pressure fluctuations are large, the sampling frequency is increased to once per second, and when the pressure is stable, it is reduced to once per minute. Zoned pressure sensors are deployed at key nodes and branch points of the pipeline network. These sensors use piezoelectric sensors to monitor pressure changes within the pipeline in real time, converting the pressure signals into electrical signals and transmitting them to the data acquisition system.

[0059] Specifically, the GIS system extracts the spatial coordinates of each leakage point from the pipeline network topology map. Each leakage point is located in three dimensions according to latitude and longitude coordinates and pipeline burial depth, with coordinate accuracy down to the centimeter level. Historical total leakage water volume records are retrieved from the pipeline network operation and maintenance database, including cumulative leakage water volume statistics for daily, weekly, and monthly periods over past time periods. During the generation of the leakage contribution assessment weight table, a basic weight value is set according to the pipe diameter, with larger pipe diameters having higher basic weights. At the same time, the weight is adjusted based on the location and depth of the leakage point; leakage points buried deeper are assigned higher weights due to the greater difficulty of repair.

[0060] For example, when processing instantaneous flow rates, the moving average method selects continuous flow data points within a time window and calculates their arithmetic mean as the smoothed value for the current moment. The width of the time window is adaptively adjusted according to the frequency of flow fluctuations. When the sensor generates abnormal spike signals due to electromagnetic interference or mechanical vibration, the moving average method effectively filters out these noise interferences through smoothing processing, resulting in more stable and reliable flow sequence data. Differential operations quantify the trend of flow changes over time by calculating the differences in flow rates between adjacent time points. The resulting flow fluctuation feature vector contains multi-dimensional information such as fluctuation amplitude, frequency, and duration.

[0061] In one possible implementation, the preset threshold is determined based on the flow fluctuation range under normal pipeline network operation. The formula μ + 2σ is used as the threshold for judging abnormal fluctuations, where μ is the mean of historical normal flow fluctuations and σ is the standard deviation. When the flow fluctuation at a leak point exceeds this threshold, it indicates that there may be a sudden expansion or new leak at that point. The final leak point-related dataset is stored in a structured manner, aligning the data of each dimension according to timestamps to form a standardized data format suitable for subsequent analysis and processing.

[0062] Step S102: Identify the spatial distribution density around each leak point based on the relevant dataset of leak points, group and cluster the spatial coordinate information to obtain leak point cluster groups, identify high-density areas by identifying areas with a high concentration of leak points in the cluster groups, and determine the trend of distribution density change by comparing the changing trend of the distance between each leak point in the high-density area.

[0063] Based on the spatial coordinate information in the dataset related to leak points, the local spatial density values ​​of all leak points are sorted to form a density distribution sequence. The DBSCAN clustering algorithm is used, taking the spatial coordinates of each leak point in the density distribution sequence as input, setting a neighborhood radius parameter and a minimum sample point parameter. Based on the density reachability relationship between leak points, core points, boundary points, and noise points are classified. Core points and all density-reachable points are grouped into the same cluster, outputting the leak point cluster groups. The number of leak points contained in each cluster group is counted, and the spatial envelope area of ​​each cluster is calculated. The density value of the group is obtained by dividing the number of leak points by the envelope area. If the group density value exceeds a preset density threshold, the group is determined to be a high-density region, and the center coordinates and boundary coordinates of the high-density region are recorded. For the leakage points in the high-density area, they are sorted in order of distance from the center of the area to the farthest point. The spacing between adjacent leakage points is calculated to form a spacing sequence. The spacing sequence is averaged by a sliding window to obtain a smoothed spacing change curve. By comparing the increasing or decreasing trend of the average spacing within the window, the trend of distribution density change is determined. The leakage contribution evaluation weight table is used to assign weights to the leakage points in the high-density area.

[0064] In one implementation, when identifying spatial density based on spatial coordinate information in the leak point related dataset, a three-dimensional coordinate system is used to represent the location of each leak point, where the horizontal and vertical coordinates represent geographical location, and the vertical coordinate represents pipeline burial depth. The Euclidean distance is calculated considering distance components in all three dimensions, and the straight-line distance between any two leak points is obtained by taking the square root of the sum of squares. The search radius is set based on the characteristics of the pipeline network structure, taking a preset multiple of the minimum pipe segment length. This multiple is dynamically adjusted according to the density of the pipeline network, using a smaller multiple in dense areas and a larger multiple in sparse areas.

[0065] Specifically, in the calculation of local spatial density values, a spherical search space is formed with each leak point as the center and the search radius as the boundary. The number of other leak points contained within this space is counted. The volume of the spherical region is calculated using a formula. The local spatial density value at the location of the leak point is obtained by dividing the counted number of leak points by the spherical volume. This process is repeated for all leak points in the pipeline network, forming data pairs containing each leak point and its corresponding density value. These data pairs are then sorted by density value from smallest to largest to form a density distribution sequence.

[0066] It should be noted that the DBSCAN clustering algorithm uses the three-dimensional spatial coordinates of each leak point in the density distribution sequence as input data when processing spatial clustering of leak points. The neighborhood radius parameter is set considering the actual physical structure of the pipeline network, usually taking the average distance between adjacent pipe segments. The minimum number of sample points parameter is determined based on the complexity of the pipeline network, with a larger value set for complex intersection areas and a smaller value set for simple straight pipe segments. The algorithm identifies density reachability relationships, grouping spatially close leak points with similar densities into the same cluster group, while isolated, single leak points are not assigned to any group.

[0067] In one possible implementation, after clustering, density is evaluated for each group. The spatial envelope volume is calculated using a three-dimensional convex hull algorithm to find the smallest convex polyhedron that can contain all leakage points within the group, and the volume of this polyhedron is calculated. The group density value is equal to the total number of leakage points within the group divided by the spatial envelope volume, reflecting the degree of clustering of leakage points in that area. The preset density threshold is determined through historical data statistics, collecting leakage distribution data for various areas of the pipeline network over a past period, calculating the average density value and standard deviation, and using the average value plus the standard deviation as the threshold for determining high-density areas.

[0068] Preferably, the center coordinates of the high-density region are obtained by calculating the arithmetic mean of the coordinates of all leakage points within the region, representing the geometric center of the region. The boundary coordinates are determined by identifying the leakage point farthest from the center within the region, forming a description of the outer boundary of the high-density region. This method of determining the center and boundary accurately reflects the spatial distribution characteristics of the high-density region, providing a foundation for subsequent density change analysis.

[0069] For example, when analyzing the density change trend within a high-density region, the points are sorted according to their distance from the region center, forming a sequence of leak points from the inside out. The distance between adjacent leak points is calculated; the distance between the first and second leak points is used as the first distance value, the distance between the second and third is used as the second distance value, and so on, forming a distance sequence. A sliding window averaging process is used, where a fixed-length window slides across the distance sequence, and the average of all distance values ​​within the window is calculated each time, serving as the smoothed distance value at the center of the window.

[0070] For example, as the sliding window moves from the beginning to the end of the spacing sequence, the average spacing within the window is recalculated at each position. By comparing the average spacing values ​​of adjacent windows, the trend of spacing change is determined. If the average spacing of the later window is greater than that of the earlier window, it indicates that the distribution of leakage points gradually thins out from the center outwards, and the distribution density decreases. Conversely, if the average spacing gradually decreases, the distribution density increases. This trend judgment method can effectively identify density gradient changes within high-density areas. Furthermore, after determining the trend of distribution density change, the pipe diameter weight and burial depth weight in the leakage contribution evaluation weight table are used to weight each leakage point in the high-density area. Specifically, the basic weight value in the weight table is multiplied by the density contribution of each leakage point to form a weighted density distribution map. This weighted density distribution map is used to subsequently identify key areas that have a significant impact on the total leakage volume.

[0071] Understandably, determining the trend of density variation is crucial for assessing hydraulic disturbances. When density increases, it indicates that leaks are clustering in a certain direction, resulting in strong hydraulic disturbances between leaks in that direction. Conversely, when density decreases, leaks gradually disperse, and hydraulic disturbances gradually weaken. Accurately identifying the trend of density variation provides a scientific basis for subsequent leakage volume correction, improving the accuracy of total leakage volume assessment.

[0072] Step S103: Determine adjacent leakage points by the distribution density change trend, identify the pressure gradient between adjacent leakage points, and determine the single-point water output correction ratio.

[0073] Based on the gradient direction of the density change, the nearest neighbor of each leak point is searched along the density increasing direction. The spatial distance between each leak point and its nearest neighbor is determined. If the distance is less than a preset distance threshold, the two points are considered adjacent leak point pairs. The coordinate information and distance values ​​of all adjacent leak point pairs are recorded to form an adjacency relationship mapping table. Real-time pressure data of each pair of adjacent leak points in the adjacency relationship mapping table are obtained to determine the pressure difference between the two points. The pressure difference is divided by the pipe length between the two points to obtain the pressure change rate per unit length, which is used as the pressure gradient value between adjacent leak points. The pressure gradient value is compared with the original flow value of each leak point to obtain the proportion of the pressure gradient value to the original flow value, thus obtaining the pressure influence coefficient. Based on the relative positional relationship of adjacent leak points and the pressure influence coefficient, the degree of interference of each leak point with the adjacent points is determined. The original water output of each leak point is corrected using the degree of interference to obtain the corrected actual water output. The ratio of the actual water output to the original water output is determined to determine the single-point water output correction ratio.

[0074] Specifically, in one implementation, the process of determining adjacent leakage points based on the distribution density change trend requires comprehensive consideration of two dimensions: spatial location and density gradient. The gradient direction of the density trend is obtained by calculating the ratio of the density difference between adjacent regions to their distance, with the gradient direction pointing in the direction of the fastest density increase. A fan-shaped region search is performed along this gradient direction, with the current point as the vertex and the gradient direction as the central axis, expanding to both sides by a certain angle range, gradually increasing the search radius until the search radius exceeds a preset maximum range or the nearest neighbor point is found. When the distance between the searched leakage point and the current point is less than a preset distance threshold, the coordinate information, relative azimuth angle, and straight-line distance of this pair of leakage points are recorded, forming an adjacency relationship mapping table containing multi-dimensional information.

[0075] It should be noted that the preset distance threshold is determined based on the effective range of the hydraulic interference, which is related to the pipeline pressure, pipe diameter, and leakage orifice diameter. In high-pressure pipe sections, the effective range of the hydraulic interference is larger, and the distance threshold is correspondingly increased; in low-pressure pipe sections, the effective range is smaller, and the distance threshold is correspondingly decreased. By statistically analyzing the interference range under different operating conditions using historical data, a pressure-distance threshold correspondence table is established. In practical applications, the applicable distance threshold is obtained by looking up the table based on the current pressure.

[0076] Specifically, when acquiring real-time pressure data for adjacent leakage point pairs, the sampling time of the pressure sensors needs to be synchronized to ensure that the pressure difference reflects the pressure distribution at the same moment. The calculation of the pressure difference takes into account the influence of pressure fluctuations, using the average value of multiple samples as the stable pressure value. The rate of change of pressure per unit length is equal to the pressure difference divided by the pipe length, where the pipe length is calculated through the pipe network topology, taking into account the actual connection path of the pipe rather than the straight-line distance, thus accurately reflecting the actual pressure transmission characteristics in the pipe.

[0077] For example, in analyzing the relationship between pressure gradient values ​​and original flow rates, a pressure-flow response model needs to be established. This model is based on Bernoulli's equation and the continuity equation, considering the outflow characteristics of the leakage orifice. The specific model is as follows: Where Q is the output flow rate, and C is the output flow rate. d Where A is the emission coefficient, P1 is the upstream input pressure, P2 is the downstream pressure, and ρ is the fluid density.

[0078] in d is the aperture.

[0079] Calculation process: The velocity is derived from Bernoulli's equation. Multiply by A and C d Q is obtained. The boundary conditions are steady-state flow and friction is negligible, applicable to small-diameter leakage. When the pressure at the upstream leakage point decreases due to leakage, it affects the pressure distribution at the downstream leakage point, thus changing the outflow rate at the downstream point. The pressure influence coefficient is defined as the ratio of the pressure gradient value to the original flow rate value. This coefficient reflects the sensitivity of pressure changes to flow rate. The larger the coefficient, the more significantly the outflow rate at that leakage point is affected by adjacent points.

[0080] In one possible implementation, determining the degree of interference based on the relative positions of adjacent leakage points requires considering the direction of water flow and the upstream / downstream relationship of the leakage points. Leakage from an upstream leakage point reduces downstream pressure, while the impact of a downstream leakage point on the upstream is relatively small. The degree of interference is obtained by multiplying the pressure influence coefficient by a position correction factor (PCF), where PCF is a dimensionless parameter calculated as PCF = 1 + (Δh / d), where Δh is the elevation difference between the two points (meters) and d is the horizontal distance (meters). The ratio of elevation difference to horizontal distance reflects the influence of terrain slope on the interference. A larger elevation difference indicates a more pronounced gravitational effect, resulting in a corresponding increase in the degree of interference; a greater horizontal distance leads to greater pressure transmission attenuation, resulting in a corresponding decrease in the degree of interference.

[0081] Preferably, in the process of correcting the original outflow rate using the degree of interference, the correction formula is: the corrected outflow rate equals the original outflow rate multiplied by the reciprocal of the degree of interference. When the degree of interference is greater than 1, it indicates a suppressive effect, and the corrected outflow rate is less than the original value; when the degree of interference is less than 1, it indicates an enhanced effect, and the corrected outflow rate is greater than the original value. This correction mechanism can reflect the impact of the interaction between leakage points on the total leakage. For example, in actual pipe networks, when multiple leakage points are densely distributed in the same pipe section, a complex pressure interaction network is formed between the points. By calculating the comprehensive degree of interference experienced by each leakage point, a more accurate single-point outflow rate can be obtained. The single-point outflow rate correction ratio is equal to the ratio of the actual outflow rate to the original outflow rate. This ratio typically varies between 0.5 and 1.5, reflecting the significant impact of hydraulic interference on leakage assessment. When the dense distribution of adjacent leakage points leads to a local pressure reduction, the correction ratio is less than 1, indicating that the actual outflow rate at a single point is lower than the original estimate; when the leakage points are relatively independent, the correction ratio is close to 1.

[0082] Understandably, by establishing an adjacency mapping, calculating the pressure gradient, determining the degree of interference, and correcting the outflow rate, a series of processes are employed to quantitatively assess the hydraulic interference effect between leakage points. Specifically, the weighting table generated in S101 is used in S102 to weight the degree of interference, and in S103 it is applied to correct the outflow rate ratio. Determining the correction ratio provides a scientific basis for subsequent calculations of the total leakage volume, avoiding assessment bias caused by simple accumulation, and improving the accuracy and reliability of pipeline network leakage assessment.

[0083] Step S104: Verify the single-point water output correction ratio with the instantaneous flow rate value in the relevant dataset of the leakage point to obtain the flow rate deviation value. Identify the leakage point type based on the flow rate deviation value and flow rate fluctuation characteristics, and adjust the leakage contribution evaluation weight table according to the leakage point type.

[0084] The corrected flow rate is obtained by multiplying the single-point outflow correction ratio by the instantaneous flow rate value of each leak point in the relevant dataset. The difference between the corrected flow rate value and the actual monitored flow rate value of the leak point is then calculated to obtain the flow deviation value. Based on the magnitude and variation pattern of the flow deviation value, and combined with the fluctuation frequency and amplitude information in the flow fluctuation characteristics, the type of leak point is identified. The flow fluctuation characteristics are calculated by using the standard deviation of the flow deviation value sequence as the amplitude, and the number of times the deviation value exceeds the average deviation value divided by the total number of time points as the frequency. If the flow deviation value exceeds the average deviation value and the fluctuation frequency is higher than a preset frequency threshold, it is identified as a sudden leak point. If the flow deviation value is less than the average deviation value and the fluctuation frequency is lower than the preset frequency threshold, it is identified as a chronic leak point. If the flow deviation value shows periodic changes, it is identified as an intermittent leak point. Different weight adjustment coefficients are assigned to the sudden, chronic, and intermittent leakage points respectively. The weight coefficient of the sudden leakage point is the original weight multiplied by an adjustment coefficient greater than 1. The original weight comes from the initial value of the leakage contribution assessment weight table. The original weight of the chronic leakage point remains unchanged. The weight coefficient of the intermittent leakage point is determined according to the ratio of its active period to the total period. The weight value of the corresponding leakage point in the leakage contribution assessment weight table is updated with the adjusted weight coefficient.

[0085] In one implementation, the verification process of the single-point outflow correction ratio and the instantaneous flow rate value is carried out using a point-by-point comparison method. The correction ratio reflects the degree of influence of hydraulic disturbance on the outflow of a single leakage point. This ratio is multiplied by the instantaneous flow rate value centrally recorded in the relevant data set of the leakage point to obtain the corrected flow rate value considering hydraulic disturbance. The actual monitored flow rate value is obtained directly through a flow sensor, and the difference between the two is the flow deviation value, which reflects the degree of agreement between the theoretical correction and the actual situation.

[0086] Specifically, the analysis of flow deviation values ​​needs to be combined with the time dimension. The deviation values ​​over a continuous time period are statistically analyzed, and the average deviation value is calculated as a benchmark. The fluctuation frequency included in the flow fluctuation characteristics is calculated by the number of flow changes per unit time, and the fluctuation amplitude is determined by the peak-to-valley difference. When the deviation value is consistently higher than the average deviation value and the fluctuation frequency exceeds a preset frequency threshold, it indicates that the leakage point is in an unstable state and is identified as a sudden leakage point. These types of leakage points typically correspond to the expansion of pipe cracks or new damage.

[0087] It should be noted that chronic leaks are characterized by stable flow deviations that are below the average value, with low fluctuation frequency, corresponding to slow corrosion or material aging of the pipeline leading to stable leakage. Intermittent leaks exhibit a clear periodic variation in deviation values, usually related to periodic changes in pipeline pressure or peak and trough water usage. The ratio of active periods to total periods is calculated by statistically analyzing the proportion of time when the deviation value exceeds the threshold within the total monitoring time.

[0088] Preferably, the weight adjustment coefficient is set considering the different contributions of different types of leakage points to the total leakage. Sudden leakage points, due to their drastic and unpredictable flow variations, are assigned an adjustment coefficient greater than 1, typically between 1.2 and 1.5, giving them a higher weight in the total leakage assessment. Chronic leakage points have stable and predictable flow, maintaining their original weight. Intermittent leakage points are dynamically adjusted based on the proportion of active periods; the higher the proportion, the larger the weight coefficient. The formula for calculating the weight coefficient w is w = 0.4 + p, where p is the proportion of active periods.

[0089] For example, when p is 0.3, w is 0.7; when p is 0.7, w is 1.1. Through this differentiated weight adjustment mechanism, the leakage contribution assessment weight table can more accurately reflect the actual contribution of various leakage points to the total leakage of the pipeline network, thus improving the accuracy of leakage assessment.

[0090] Step S105: Using the adjusted leakage contribution evaluation weight table, combined with the single-point outflow correction ratio and spatial distribution density, extract the weighted contribution value of each leakage point, classify the weighted contribution value to distinguish between the real leakage signal and the interference signal, and determine the classified real leakage water volume component.

[0091] The weight value of each leakage point is obtained through the adjusted leakage contribution evaluation weight table. This weight value is multiplied by the single-point outflow correction ratio of the corresponding leakage point, and then multiplied by the normalized spatial density factor of the area where the leakage point is located. The normalized spatial density factor is obtained by dividing the number of leakage points in the area by the area of ​​the region and then by the average density of the entire pipe network; it is a dimensionless parameter. This yields the weighted contribution value of each leakage point. Based on the distribution characteristics of the weighted contribution values, the mean and standard deviation of the weighted contribution values ​​of all leakage points are calculated. If the weighted contribution value of a leakage point is within the range of the mean plus or minus the standard deviation, it is determined to be a real leakage signal; otherwise, it is determined to be an interference signal, thus completing the signal classification of the leakage points. For the leakage points corresponding to the real leakage signals, their instantaneous flow values ​​in the leakage point related dataset are extracted. This instantaneous flow value is multiplied by the corresponding weighted contribution value to obtain the single-point real leakage water volume. The single-point real leakage water volumes of all real leakage signal points are accumulated to determine the real leakage water volume component.

[0092] In one implementation, when extracting the weight value of each leakage point using the adjusted leakage contribution assessment weight table, the weight table includes weight coefficients determined according to the leakage point type. The product of the weight value and the single-point outflow correction ratio reflects the actual leakage contribution after considering hydraulic disturbances. Multiplying this by the spatial distribution density value reflects the clustering effect of leakage points in high-density areas. The product of these three parameters forms a weighted contribution value that comprehensively considers multiple factors.

[0093] Specifically, the distribution characteristics analysis of the weighted contribution values ​​employs statistical methods. The central tendency of the overall leakage level is obtained by calculating the arithmetic mean of the weighted contribution values ​​of all leakage points. The standard deviation reflects the dispersion of the contribution values ​​at each leakage point; a larger standard deviation indicates more significant differences between leakage points. The interval formed by adding and subtracting the standard deviation from the mean represents the reasonable range of normal leakage, and this range is determined based on the statistical regularities of a large amount of historical data.

[0094] It's important to note that the distinction between genuine leakage signals and interference signals lies in their statistical characteristics. The weighted contribution values ​​of genuine leakage signals typically exhibit a normal distribution, concentrating around the mean. Interference signals, on the other hand, are often caused by sensor malfunctions, data transmission errors, or sudden changes in the external environment, and their weighted contribution values ​​deviate significantly from the normal range. When the weighted contribution value of a leakage point exceeds the mean plus or minus the standard deviation, it indicates that the data at that point may be affected by abnormal factors and needs to be marked as interference and removed.

[0095] Preferably, in the process of extracting the actual leakage water volume component, the instantaneous flow rate value is taken from the preprocessed measured data in the relevant dataset of the leakage point. The instantaneous flow rate value is multiplied by the corresponding weighted contribution value to obtain the actual leakage water volume at a single point, considering multiple factors such as weight, correction, and density. This calculation method avoids the errors caused by simple summation and can more accurately reflect the actual contribution of each leakage point to the total leakage volume.

[0096] For example, within a specific section of a pipeline network, the actual leakage volume of that section is obtained by summing the individual actual leakage volumes of all leakage points identified as genuine leakage signals. This component eliminates the influence of interference signals, accurately reflects the actual leakage situation of the pipeline network, and provides reliable data support for pipeline network maintenance decisions.

[0097] Step S106: Summarize the total leakage volume from the classified actual leakage volume components, analyze the change rate by combining the historical total leakage volume data, evaluate whether the change rate reflects the phenomenon of total water volume decrease under the increase of the number of leakage points, and obtain the verification result of total leakage volume.

[0098] The total leakage volume for the current period is obtained by summing the water volume components of all actual leakage points from the classified actual leakage volume components. Historical values ​​for the corresponding period are extracted from the historical total leakage volume data. The leakage volume change rate is calculated by subtracting the historical value from the current total leakage volume and then dividing by the historical value. The sum of the number of newly added leakage points and the number of existing leakage points in the current period is counted and compared with the total number of leakage points in the same historical period. If the number of leakage points increases and the corrected leakage volume change rate is negative, it is determined that a reverse phenomenon has occurred: the number of leakage points increases but the corrected total leakage volume decreases. This phenomenon reflects the single-point water output reduction effect caused by hydraulic disturbance. The characteristic parameters of this reverse phenomenon are recorded. Based on the characteristic parameters of the reverse phenomenon and the rate of change of leakage water volume, if a reverse phenomenon exists and the absolute value of the rate of change exceeds a preset threshold, the verification result is marked as hydraulic interference-dominated; if no reverse phenomenon exists or the absolute value of the rate of change is less than the preset threshold, the verification result is marked as independent leakage. The output includes the verification mark, the value of the rate of change, and the change in the number of leakage points, as well as the total leakage water volume verification result.

[0099] In one implementation, when summarizing the actual leakage water volume components, the component values ​​of each partition need to be aligned with timestamps and then accumulated. The accumulation process uses a partition-by-partition traversal method to ensure that the water volume contribution of each actual leakage point is included in the total. Historical total leakage water volume data is usually stored in a time-series database and archived at different granularities such as daily, weekly, and monthly. When extracting historical values, the historical period with the same statistical time period is selected. For example, if the current period is the first week of a certain month, the data from the first week of the same month in the past is extracted for comparison.

[0100] Specifically, the calculation of the corrected leakage rate reflects the evolving trend of network leakage. A positive rate indicates an increase in the total leakage after correction, while a negative rate indicates a decrease. This comparative analysis of the rate of change with the change in the number of leakage points reveals the existence of hydraulic interference effects. Under normal circumstances, an increase in the number of leakage points leads to a corresponding increase in the total leakage after correction. However, in the presence of significant hydraulic interference, newly added leakage points reduce the outlet pressure of existing leakage points, resulting in a decrease in the outlet flow at a single point. This leads to the opposite phenomenon: an increase in the number of leakage points but a decrease in the total leakage after correction.

[0101] It should be noted that the characteristic parameters of the reverse phenomenon include the increase in the number of leakage points, the decrease in total water volume, and the ratio between the two. These parameters together constitute the basis for judging the intensity of hydraulic disturbance. When the increase in the number of leakage points is large and the decrease in total water volume is significant, it indicates that the hydraulic disturbance effect is dominant. The preset threshold is usually set as a multiple of the standard deviation of the historical rate of change to distinguish between normal fluctuations and abnormal changes.

[0102] Preferably, the classification of verification results adopts a binary judgment method. The hydraulic interference-dominated type indicates that the mutual influence between leakage points in the pipeline network is significant, requiring full consideration of interference correction in leakage assessment; the independent leakage type indicates that each leakage point is relatively independent, and the total leakage can be assessed using the traditional summation method. The change rate values ​​and changes in the number of leakage points included in the verification results provide a quantitative basis for the formulation of pipeline maintenance strategies.

[0103] For example, a water supply network zone added 5 new leakage points within a month, increasing the total number of leakage points from 10 to 15, an increase of 50%. However, the total leakage water volume decreased from 100 cubic meters per day to 90 cubic meters, a change rate of -10%. This inverse phenomenon of increased leakage points but decreased total water volume verifies the existence of hydraulic disturbance effect, and the system marks the verification result as hydraulic disturbance dominant.

[0104] Step S107: Based on the total leakage water volume verification results, integrate the zonal pressure data and distribution density change trends to generate a pipeline health assessment report. The overall accuracy improvement is assessed by the cumulative value of the actual leakage water volume components in the report.

[0105] Based on the verification markers and change rate values ​​in the total leakage water volume verification results, the zone pressure data and distribution density change trends are integrated, and the data is organized according to time sequence and spatial location to generate a pipeline health assessment report containing verification results, pressure data, and density trend information. The cumulative value of the actual leakage water volume component is extracted from the pipeline health assessment report and compared with the original value obtained by simply summing the instantaneous flow rates at each leakage point. The overall accuracy improvement is obtained by subtracting the absolute value of the original value from the cumulative value of the actual leakage water volume component and dividing by the original value.

[0106] In one implementation, the pipeline health assessment report is generated using a structured data organization method. The total leakage volume verification result serves as the core content of the report, including the verification marker type and specific rate of change values. Zonal pressure data is spatially arranged according to the pipeline topology, with the pressure value of each zonal associated with the location of the corresponding leakage point. Distribution density change information indicates the trend of leakage point density changes over time or space, presented in a chart format to show the spatial distribution characteristics of the density gradient. The density gradient refers to the gradient value of density change, calculated using the formula D=N / A, where N is the number of leakage points and A is the area of ​​the region. This data is derived from the integration of zonal pressure data.

[0107] Specifically, the report is organized into three levels: overview, detailed data, and assessment conclusions. In the detailed data section, the zone pressure data is arranged according to the zone number of the pipeline network topology. Each zone records the average pressure value, pressure fluctuation range, and the number of leakage points within that zone. Information on density variation is presented as a density gradient heatmap, marking the main directions of density increase and decrease, and associating them with corresponding high-density area numbers. By overlaying the pressure data with density trend information, the spatial correspondence between areas of abnormal pressure and areas with clustered leakage points can be intuitively identified. The overview section includes basic pipeline network information and the assessment period; the detailed data section lists the location, flow rate, correction ratio, and other parameters for each leakage point; and the assessment conclusions section presents the total leakage volume and verification results.

[0108] It should be noted that the overall accuracy improvement is assessed by comparing the difference in leakage before and after correction. The original value is obtained by simply summing the instantaneous flow rates at each leakage point, without considering the influence of hydraulic interference. The sum of the actual leakage components is the corrected actual leakage. The difference between the two reflects the degree of improvement in assessment accuracy after considering hydraulic interference, providing a quantitative basis for optimizing pipeline network leakage control strategies.

[0109] This invention provides a smart water management data management system, mainly comprising:

[0110] The data acquisition and preprocessing module is used to acquire instantaneous flow values, zone pressure data, spatial coordinate information, historical total water loss data, and a weight table for assessing the contribution of leakage at each leakage point. It removes noise interference through preliminary filtering and extracts the flow fluctuation characteristics of each leakage point to obtain a dataset related to the leakage point.

[0111] The spatial clustering and density analysis module is used to identify the spatial distribution density around each leak point based on the leak point related dataset. It groups and clusters the spatial coordinate information to obtain leak point cluster groups. By identifying the areas with a high concentration of leak points in the cluster groups, it determines the high-density areas. By comparing the changing trends of the distance between each leak point in the high-density areas, it determines the trend of distribution density changes.

[0112] The pressure gradient and correction module is used to determine adjacent leakage points by the direction of distribution density change, identify the pressure gradient between adjacent leakage points, and determine the correction ratio of single-point water output.

[0113] The flow verification and type identification module is used to verify the single-point water output correction ratio with the instantaneous flow value in the relevant dataset of the leakage point to obtain the flow deviation value. Based on the flow deviation value and flow fluctuation characteristics, the leakage point type is identified, and the leakage contribution evaluation weight table is adjusted according to the leakage point type.

[0114] The weighted contribution classification module is used to evaluate the weight table of the adjusted leakage contribution, combine the single-point water output correction ratio and spatial distribution density, extract the weighted contribution value of each leakage point, classify the weighted contribution value to distinguish between the real leakage signal and the interference signal, and determine the real leakage water volume component after classification.

[0115] The total leakage volume verification module is used to summarize the total leakage volume from the classified actual leakage volume components, analyze the change rate by combining historical total leakage volume data, evaluate whether the change rate reflects the phenomenon of total water volume decrease under the increase of the number of leakage points, and obtain the total leakage volume verification result.

[0116] The health assessment report generation module is used to integrate zone pressure data and distribution density changes based on the total leakage water volume verification results to generate a pipeline health assessment report. The overall accuracy improvement is assessed by the cumulative value of the actual leakage water volume components in the report.

[0117] The above embodiments are merely one of the preferred embodiments of the present invention and should not be used to limit the scope of protection of the present invention. Any modifications or refinements made to the main design concept and spirit of the present invention that are not of substantial significance, but solve the same technical problem as the present invention, should be included within the scope of protection of the present invention.

Claims

1. A smart water data management method, characterized in that, The method includes: The instantaneous flow rate, zone pressure data, spatial coordinate information, historical total water loss data, and leakage contribution evaluation weight table of each leakage point are obtained. Noise interference is removed through preliminary filtering, and the flow fluctuation characteristics of each leakage point are extracted to obtain the relevant dataset of leakage points. Based on the leak point related dataset, the spatial distribution density around each leak point is identified. The spatial coordinate information is grouped and clustered to obtain leak point cluster groups. By identifying the areas with a high concentration of leak points in the leak point cluster groups, high-density areas are determined. By comparing the changing trends of the spacing between each leak point in the high-density areas, the trend of distribution density change is determined. By identifying adjacent leakage points based on the distribution density change trend, determining the pressure gradient between adjacent leakage points, and determining the single-point outflow correction ratio, including: Based on the gradient direction of the distribution density change, search for the nearest neighbor of each leakage point along the density increasing direction, determine the spatial distance between each leakage point and its nearest neighbor, record the coordinate information and distance value of all adjacent leakage point pairs, and form an adjacent relationship mapping table. Obtain the real-time pressure data of each pair of adjacent leakage points in the adjacent relationship mapping table, determine the pressure difference between the two points, divide the pressure difference by the pipe length between the two points to obtain the pressure change rate per unit length, and use the pressure change rate per unit length as the pressure gradient value between adjacent leakage points. By comparing the pressure gradient value with the original flow rate value of each leakage point, the proportion of the pressure gradient value to the original flow rate value is obtained, and the pressure influence coefficient is obtained. Based on the relative positional relationship of the adjacent leakage points and the pressure influence coefficient, the degree of interference of each leakage point with the influence of the adjacent points is determined. The original outflow of each leakage point is corrected using the aforementioned level of interference to obtain the corrected actual outflow. The ratio of the actual outflow to the original outflow is determined, and the correction ratio for the outflow at a single point is determined. The single-point water output correction ratio is verified with the instantaneous flow rate value in the dataset related to the leakage point to obtain the flow rate deviation value. The leakage point type is identified based on the flow rate deviation value and the flow rate fluctuation characteristics. The leakage contribution evaluation weight table is adjusted according to the leakage point type. By using the adjusted leakage contribution evaluation weight table, combined with the single-point water output correction ratio and the spatial distribution density, the weighted contribution value of each leakage point is extracted, and the weighted contribution value is classified to distinguish between the real leakage signal and the interference signal, and the real leakage water volume component after classification is determined. The total leakage volume is summed from the actual leakage volume components after classification, and the change rate is analyzed in combination with the historical total leakage volume data to obtain the verification result of the total leakage volume; Based on the total leakage water volume verification results, the zone pressure data and the distribution density change trend are integrated to generate a pipeline health assessment report.

2. The intelligent water data management method according to claim 1, characterized in that, The process involves acquiring instantaneous flow rates at each leakage point, zone pressure data, spatial coordinate information, historical total leakage water volume data, and a leakage contribution assessment weight table. After preliminary filtering to remove noise interference, the flow fluctuation characteristics of each leakage point are extracted to obtain a leakage point-related dataset, including: The system obtains the instantaneous flow rate value of each leakage point from the pipeline monitoring terminal, reads real-time pressure data from the zone pressure sensor, extracts the spatial coordinate information of each leakage point through the GIS system, retrieves the historical total leakage water volume record from the database, generates a leakage contribution evaluation weight table based on the pipe diameter and the location and depth of the leakage point, and uses the moving average method to smooth the instantaneous flow rate value to obtain the smoothed flow rate sequence. The smoothed flow sequence is subjected to differential operation in time order to calculate the flow change value of each leakage point at adjacent time points, and the maximum, minimum and average values ​​of flow change per unit time are statistically analyzed to form a flow fluctuation feature vector. Based on the change amplitude in the flow fluctuation feature vector, combined with the spatial coordinate information, partition pressure data, and the leakage contribution evaluation weight table, a weight value is applied to the flow fluctuation feature vector for weighted correction, thereby constructing a leakage point related dataset containing flow time series data, pressure distribution data, location information, and weighted fluctuation features.

3. The intelligent water data management method according to claim 1, characterized in that, The step of identifying the spatial distribution density around each leak point based on the leak point related dataset, grouping and clustering the spatial coordinate information to obtain leak point cluster groups, identifying high-density areas by identifying areas with a high concentration of leak points in the cluster groups, and determining the trend of distribution density change by comparing the changing trends of the spacing between leak points in the high-density areas, includes: Based on the spatial coordinate information in the dataset related to the leakage points, the local spatial density values ​​of all leakage points are sorted to form a density distribution sequence; The DBSCAN clustering algorithm is used as input, taking the spatial coordinates of each leak point in the density distribution sequence as input, to form a classification of core points, boundary points and noise points. The core points and all points that their density can reach are grouped into the same cluster, and the leak point cluster grouping is output. The number of leakage points in each group of the leakage point cluster is counted, the spatial envelope area of ​​each group is calculated, the group density value is obtained by dividing the number of leakage points by the envelope area, the center coordinates and boundary coordinates of the high-density area are recorded, and the leakage points in the high-density area are weighted in combination with the leakage contribution evaluation weight table.

4. The intelligent water data management method according to claim 1, characterized in that, The step of verifying the single-point water outflow correction ratio with the instantaneous flow rate value in the dataset related to the leakage point to obtain a flow rate deviation value, identifying the leakage point type based on the flow rate deviation value and the flow rate fluctuation characteristics, and adjusting the leakage contribution evaluation weight table according to the leakage point type includes: The corrected flow rate is obtained by multiplying the single-point outflow correction ratio by the instantaneous flow rate value of each leak point in the leak point related dataset. The difference between the corrected flow rate value and the actual monitored flow rate value of each leak point is calculated to obtain the flow deviation value. Based on the magnitude and variation pattern of the flow deviation value, and combined with the fluctuation frequency and amplitude information in the flow fluctuation characteristics, the leakage point type is identified. The standard deviation of the flow deviation value sequence is calculated as the amplitude, and the number of times the flow deviation value exceeds the average deviation value is divided by the total number of time points as the frequency. Different weight adjustment coefficients are assigned to sudden, chronic, and intermittent leakage points, and the weight values ​​of the corresponding leakage points in the leakage contribution evaluation weight table are updated with the adjusted weight coefficients.

5. The intelligent water data management method according to claim 1, characterized in that, The adjusted leakage contribution evaluation weight table, combined with the single-point outflow correction ratio and the spatial distribution density, extracts the weighted contribution value of each leakage point, classifies the weighted contribution value to distinguish between real leakage signals and interference signals, and determines the classified real leakage water volume component, including: The weight value of each leakage point is obtained by using the adjusted leakage contribution evaluation weight table. The weight value is multiplied by the single-point outflow correction ratio of the corresponding leakage point, and then multiplied by the normalized spatial density factor of the area where the leakage point is located to obtain the weighted contribution value of each leakage point. Based on the distribution characteristics of the weighted contribution values, calculate the mean and standard deviation of the weighted contribution values ​​of all leakage points to complete the signal classification of leakage points; For the leakage point corresponding to the real leakage signal, the instantaneous flow rate value of the leakage point in the relevant dataset is extracted. The instantaneous flow rate value is multiplied by the corresponding weighted contribution value to obtain the real leakage water volume at a single point. The real leakage water volumes at all real leakage signal points are accumulated to determine the real leakage water volume components.

6. The intelligent water data management method according to claim 1, characterized in that, The total leakage volume is summed from the classified actual leakage volume components, and the change rate is analyzed in conjunction with the historical total leakage volume data to obtain the total leakage volume verification result, including: The total leakage volume for the current period is obtained by summing the water volume component values ​​of all real leakage points from the classified real leakage volume components. The historical values ​​for the corresponding period are extracted from the historical total leakage volume data, and the leakage volume change rate is calculated. The total number of newly added leakage points and the total number of existing leakage points in the current period are counted and compared with the total number of leakage points in the same period in history, and the characteristic parameters of the reverse phenomenon are recorded. Based on the characteristic parameters of the reverse phenomenon and the rate of change of leakage water volume, the total leakage water volume verification result is output, including the verification mark, the change rate value, and the change in the number of leakage points.

7. The intelligent water data management method according to claim 1, characterized in that, The process of integrating the zonal pressure data and the distribution density change trend based on the total leakage water volume verification results to generate a pipeline health assessment report includes: Based on the verification markers and change rate values ​​in the total leakage water volume verification results, the zone pressure data and the distribution density change trend are integrated, and the data is organized according to time sequence and spatial location to generate a pipeline health assessment report containing verification results, pressure data and density trend information. The cumulative value of the actual leakage water volume component is extracted from the pipeline health assessment report and compared with the original value obtained by simply summing the instantaneous flow rate of each leakage point to obtain the overall accuracy improvement.

8. A smart water management data management system, characterized in that, The management system is based on the smart water data management method as described in any one of claims 1-7, and the system includes: The data acquisition and preprocessing module is used to acquire instantaneous flow values, zone pressure data, spatial coordinate information, historical total water loss data, and a weight table for assessing the contribution of leakage at each leakage point. It removes noise interference through preliminary filtering and extracts the flow fluctuation characteristics of each leakage point to obtain a dataset related to the leakage point. The spatial clustering and density analysis module is used to identify the spatial distribution density around each leak point based on the leak point related dataset, group the spatial coordinate information to obtain leak point cluster groups, identify high-density areas by identifying areas with a high concentration of leak points in the leak point cluster groups, and determine the trend of distribution density change by comparing the changing trend of the distance between each leak point in the high-density area. The pressure gradient and correction module is used to determine adjacent leakage points by the distribution density change trend, identify the pressure gradient between adjacent leakage points, and determine the single-point water output correction ratio. The flow verification and type identification module is used to verify the single-point water output correction ratio with the instantaneous flow value in the data set related to the leakage point to obtain the flow deviation value, identify the leakage point type based on the flow deviation value and the flow fluctuation characteristics, and adjust the leakage contribution evaluation weight table based on the leakage point type. The weighted contribution classification module is used to extract the weighted contribution value of each leakage point by using the adjusted leakage contribution evaluation weight table, combined with the single-point water output correction ratio and the spatial distribution density, and classify the weighted contribution value to distinguish between the real leakage signal and the interference signal, and determine the real leakage water volume component after classification. The total leakage water volume verification module is used to summarize the total leakage water volume from the classified actual leakage water volume components, analyze the change rate in combination with the historical total leakage water volume data, and obtain the total leakage water volume verification result. The health assessment report generation module is used to integrate the zone pressure data and the distribution density change trend based on the total leakage water volume verification results to generate a pipeline health assessment report.

Images