Independent metering zoning method for water supply pipe network

Abstract

The invention discloses an independent metering and partitioning method for a water supply pipe network, and relates to the technical field of water supply pipe network management.The independent metering and partitioning method for the water supply pipe network comprises the steps that intelligent water meters are installed at user access points, a big data analysis platform is constructed, and water consumption time and water consumption data are subjected to standardization and clustering analysis; according to the method, the water consumption behavior characteristics of the users in different independent metering partitions can be accurately determined, subarea division is more scientific and reasonable through fine classification based on the user behaviors, it is ensured that the water consumption modes of the users in all the partitions are similar, the pertinence of water supply management can be improved, the water consumption demand can be more accurately predicted, and the user experience is improved. Water supply resource allocation is optimized, unnecessary energy consumption and water resource waste are reduced, in the time-sharing water supply plan, operation of the water pumps can be accurately regulated and controlled according to peak and valley periods of all the subareas, and energy conservation is achieved while the requirement for water consumption of users is met.

Description

Technical Field

[0001] The present invention relates to the technical field of water supply network management, and specifically provides a method for independent metering and zoning of water supply networks. Background Art

[0002] With the acceleration of urbanization and the continuous improvement of people's living standards, the scale and complexity of urban water supply network systems are increasing day by day. As an important part of urban infrastructure, the stability, reliability, and service quality of water supply are directly related to the daily life of residents and the normal operation of the city. In today's society, the rational utilization and efficient allocation of water resources have become the focus of global attention. How to reduce water resource waste and energy consumption while meeting the water use needs of users is one of the important challenges faced by the water supply industry.

[0003] Traditional water supply network zoning methods often focus on physical factors such as geographical areas and network topologies, lacking in-depth analysis and consideration of users' water use behavior characteristics during the water supply planning and management process. This approach results in a lack of pertinence in water supply services and an inability to accurately meet the diverse needs of different user groups. During peak water use periods, some areas may experience difficulties in water use due to insufficient water supply pressure, while other areas may face increased risks of water resource waste and pipeline leakage due to excessive water pressure. At the same time, it is difficult to accurately guarantee the special water quality requirements of different users, and the water-saving potential of users cannot be fully tapped, which is not conducive to the sustainable utilization of water resources.

[0004] In summary, the existing water supply network zoning and water supply management technologies have shown limitations when faced with the growing user demands and resource protection pressures. Therefore, there is an urgent need for an innovative method for independent metering and zoning of water supply networks that can fully consider users' water use behavior characteristics, achieve precision and personalization in water supply services, improve water resource utilization efficiency and the comprehensive benefits of water supply systems, and meet the needs of modern urban development, providing strong guarantees for the sustainable development of cities. Summary of the Invention

[0005] The purpose of the present invention is to make up for the deficiencies of the existing technology and provide a method for independent metering and zoning of water supply networks. It can use intelligent water meters and big data analysis technologies to deeply study the water use behavior characteristics of users in different independently metered zones. According to the analysis results, it can conduct refined classification and clustering of users in each zone, and formulate personalized water supply service strategies for different types of user groups, which can better meet the diverse needs of users, improve user satisfaction, and at the same time achieve energy conservation, emission reduction, and sustainable development goals by optimizing the allocation of water supply resources.

[0006] To solve the above technical problems, the present invention provides the following technical solution: A method for independent metering and zoning of water supply networks, and the specific steps of the method are as follows:

[0007] S100. Install an intelligent water meter at the user access point of the water supply network to collect data on the water usage time T and water consumption V of users.

[0008] S200. Build a big data analysis platform to standardize and analyze the data collected by the intelligent water meter, determine the water usage behavior characteristics of users in different independent metering zones, conduct refined clustering of users according to the water usage behavior characteristics, and classify them in combination with the clustering results.

[0009] S300. Divide the water supply network into multiple independent metering zones according to the user classification results in combination with the water supply network layout, that is, conduct the division of independent metering zones of the water supply network based on the undirected graph G(V, E). Plan the layout of the water supply network as an undirected graph G(V, E), where the node set V represents the nodes in the pipe network, the edge set E represents the pipes connecting these nodes, and each edge is assigned a weight, and the weight value w ij Through Calculation, where L ij Is the pipe length connecting data nodes i and j, and D ij Is the pipe diameter of this pipe. Based on the user classification results, divide the nodes where users with similar water usage behavior characteristics are located into the same subgraph, and divide the entire graph G into multiple independent subgraphs, and each subgraph is an independent metering zone;

[0010] Users within each independent metering zone have similar water usage behavior characteristics;

[0011] S400. Develop targeted personalized water supply service strategies according to the described similar water usage behavior characteristics. The personalized water supply service strategies include time-of-use water supply plans, differential water pressure regulation, customized water quality guarantee measures, and personalized water-saving suggestions;

[0012] Through real-time monitoring by the intelligent water meter and user feedback, continuously optimize and adjust the water supply service strategy to ensure the service effect;

[0013] S500. Real-time monitor the water supply parameters of each zone within the water supply network layout and feed the monitoring data back to the big data analysis platform to optimize and adjust the analysis of user water usage behavior and the water supply service strategy.

[0014] Furthermore, in S200, the water usage time T and water consumption V are respectively standardized to And Where T i And V i Are the original water usage time and water consumption data of the i-th user, And Are respectively the mean values of the water usage time and water consumption of all users, and σT and σ V are the standard deviations of water usage time and water consumption respectively. The standardized water usage time and water consumption data are denoted as x and y respectively. Cluster analysis is performed on the standardized data. The process of the cluster analysis is as follows: The objective function of the clustering is where the value of C reflects the total distance from all data points to their respective cluster centers, x i and y i are the standardized water usage time and water consumption data of the i-th user. μ and ν are the average water usage time and water consumption of the current cluster center respectively, and n is the number of users. When continuously adjusting and updating μ, ν and the cluster to which the data points belong, making C≈0 indicates that the distance from each data point to its cluster center is close, and the water usage behavior characteristics of users within the same cluster are similar;

[0015] Initially, randomly select k data points as the initial cluster centers, and assign each user's data point to the cluster where the nearest cluster center is located. That is, for different data nodes i and j of each user i and j, calculate the new center coordinates of each cluster. For the j-th cluster (j = 1, 2,..., k), its new cluster center's average water usage time μ j and average water consumption ν j are: where n j is the number of data points in the j-th cluster, x ij and y ij are the standardized water usage time and water consumption coordinates of the i-th data point in the j-th cluster. Repeat the process of assignment and update of the cluster center until the cluster center no longer changes. At this time, k cluster results are obtained, that is, C 1 , C 2 , …, C k , and each cluster represents a type of water usage behavior pattern, thus completing the classification of users.

[0016] Furthermore, when the S200 determines the initial cluster center, calculate the distance between all data points. The distance is where x i , y i and x j , y j are the standardized water usage time and water consumption coordinates of the i-th and j-th data points respectively. Select the two data points with the farthest distance, that is, the two data points with the largest d ij value as the first two initial cluster centers, denoted as c 1 and c 2 , and their corresponding coordinates are (x c1 , y c1 ) and (x c2 , y c2For the remaining data points, calculate the distance from each point to the selected cluster center. where k = 1, 2, select the point with the maximum distance to the selected cluster center as the next cluster center, i.e., c 3 , whose coordinates are (x c3 , y c3 ), satisfying D i3 = max(D i1 , D i2 ) and i = 3, 4, …, n. Repeat this process until k initial cluster centers are selected.

[0017] Furthermore, when the S300 divides the independent metering area, not only the user classification results and the pipe network layout are considered, but also the connectivity reliability evaluation index CRI is introduced to optimize the zoning scheme and measure the ability of the zone to maintain water supply in the face of emergencies. The connectivity reliability evaluation index CRI measures the reliability of the zone by calculating the connectivity probability P ij between any two nodes i and j, and statistically analyzes the historical failure probability p k of the pipe segment k. For the path from node i to node j being path ij , its path connectivity probability is the product of the normal working probabilities of all pipe segments on this path, i.e., And there are multiple paths from node i to node j. Therefore, the connectivity probability P ij between nodes i and j comprehensively considers all paths and conducts N simulations. In each simulation, the number of times m that nodes i and j can be connected is statistically analyzed ij . Then

[0018] After obtaining the connectivity probability P ij between all node pairs in the zone, set the zone connectivity threshold CRI yz , and calculate the connectivity reliability index CRI as where m is the number of nodes in the zone, is the combination number of selecting two nodes from m nodes, realizing the independent metering area division of the water supply pipe network that not only meets the similarity of user water use behavior characteristics but also has reliability.

[0019] Furthermore, the relationship between the connectivity reliability index and the zone connectivity threshold CRI yz is as follows:

[0020] When the connectivity reliability index CRI of the zone < CRI yz , it means that there is a risk in the reliability of water supply in this independent metering area in the face of emergencies;

[0021] When the connected reliability index CRI of the partition < CRI yz it indicates that when the independent metering partition faces emergencies, the nodes within the partition can maintain connectivity and supply water stably.

[0022] Furthermore, the process of the S400 time-sharing water supply plan is as follows:

[0023] Based on the water usage behavior characteristic data of each partition user determined by S200, sort out the water usage time data, and count the number of users n using water in each hour period h , and at the same time set the judgment of peak and valley periods as P high and P low When this period is designated as the peak water usage period; when it is designated as the valley period, where N is the number of users after partitioning.

[0024] According to the peak and valley period division results of each partition, formulate corresponding pump operation scheduling plans at the water supply pump station, that is, increase the number of operating pumps ΔM during the peak period high which is where h ∈ peak represents the set of hour periods in the peak period, n avg is the average number of water-using users per hour in this partition, k 1 is the adjustment coefficient, is the ceiling function, and increase the number of operating pumps to M during the peak period 0 +ΔM high where M 0 is the initial number of operating pumps to maintain stable water supply; similarly, during the valley period, gradually reduce the number of operating pumps ΔM low , and where h ∈ valley represents the set of hour periods in the valley period, is the floor function, and reduce the pump speed.

[0025] Furthermore, the specific process of the S400 differential water pressure regulation is as follows: Install pressure sensors at the starting node, end node and nodes on different floors of each independent metering partition to collect the water pressure data P s of the pipe network in real time, where s represents the s-th node, and transmit it to the big data analysis platform. The big data analysis platform sets a specific water pressure target value P for each partition according to the water usage behavior characteristics of each partition user and the pipe network layout t,s , where t represents the partition number, s represents the node number in this partition, set the pressure adjustment coefficient as k p , and the initial opening degree of the regulating valve in the intelligent water meter is K 0, the adjusted opening is K. When P s >P t,s , K = K 0 -k p ×(P s -P t,s ); When P s <P t,s , K = K 0 +k p ×(P t,s -P s ), for precise adjustment of water pressure.

[0026] Furthermore, the S400 customized water quality guarantee measures are as follows: Analyze the water usage characteristics of users in each zone to determine the water quality guarantee requirements for different zones. For zones with high water quality requirements, deeply treat the raw water in the water supply plant. In the water supply network, set up water quality monitoring points at intervals to form a water quality monitoring network, and collect water quality parameters in real time to ensure the water use safety and health of users;

[0027] The personalized water-saving suggestions identify user groups with high water consumption according to the zoning results, analyze their daily water use time distribution, water use equipment usage, and water consumption data, determine the average and peak values of their water use frequency and single water use volume, and formulate personalized water-saving suggestion plans. The suggestions include reasonably arranging water use time, dispersing water use periods, and using water-saving appliances.

[0028] Compared with the prior art, this method for independent metering zoning of a water supply network has the following beneficial effects:

[0029] 1. The method for independent metering zoning of the water supply network of the present invention can accurately determine the water use behavior characteristics of users in different independent metering zones by installing intelligent water meters at user access points and constructing a big data analysis platform, and performing standardization and clustering analysis on water use time and water consumption data. This fine classification based on user behavior makes the zoning more scientific and reasonable, ensuring that the water use patterns of users in each zone are similar. This not only helps improve the pertinence of water supply management, but also can more accurately predict water use demand, optimize the allocation of water supply resources, reduce unnecessary energy consumption and water resource waste. Moreover, in the time-of-use water supply plan, it can accurately control the operation of water pumps according to the peak and valley periods of each zone, achieving energy conservation while meeting the water use of users. Also, in terms of differential water pressure regulation, it can provide stable and appropriate water pressure according to the characteristics of the zone and user needs, avoiding pipe damage and water use inconvenience caused by water pressure problems.

[0030] Second, for the method of independent metering and zoning of the water supply network of the present invention, on the one hand, a connectivity reliability evaluation index is introduced to optimize the zoning scheme, greatly enhancing the response ability of the water supply network in the face of emergencies. By accurately calculating the connectivity probability between nodes and setting thresholds, it is ensured that the zoning can still maintain a certain water supply capacity in case of failures, guaranteeing the continuity and safety of urban water supply. On the other hand, customized water quality guarantee measures and personalized water-saving suggestions fully consider the special needs of users in different zones, conduct in-depth treatment for zones with high water quality requirements, and real-time monitor through a water quality monitoring network, ensuring the health of users' water use.

[0031] Other advantages, objectives, and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following-described drawings are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0033] Figure 1 It is an operation diagram of a method for independent metering and zoning of a water supply network;

[0034] Figure 2 It is a step flowchart of user classification in Embodiment 2. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention objective, the following will, in conjunction with the accompanying drawings and preferred embodiments, describe in detail the specific implementation manners, structures, features, and their effects according to the present invention.

[0036] Embodiment 1

[0037] As Figure 1 shown, this embodiment demonstrates the specific application process of a method for independent metering and zoning of a water supply network. By installing intelligent water meters at user access points to collect water use data and using a big data analysis platform for data processing and analysis, it realizes the accurate grasp of users' water use behavior characteristics, and accordingly completes the reasonable zoning of the water supply network and the formulation and implementation of personalized water supply service strategies. In the whole process, various factors such as the pipe network layout, users' water use patterns, and water supply reliability are comprehensively considered, effectively improving the efficiency and quality of water supply management.

[0038] First, enter the intelligent water meter data acquisition stage (S100). Install intelligent water meters at each user access point in the water supply network. This intelligent water meter can not only accurately measure the water consumption V and water usage time T of users, but also has data acquisition and transmission functions, can record the water usage time and instantaneous flow data in real time, and send the data to the big data analysis platform through wireless communication technology.

[0039] Then, enter the big data analysis platform construction stage (S200). The intelligent water meter continuously collects the water usage data of users. For user i, the original water usage time is T i , and the water consumption is V i . After the big data analysis platform receives these data, it processes them according to the standardization formula. The standardization formula for water usage time is The standardization formula for water consumption is where and are the means of the water usage times and water consumptions of all users respectively, σ T and σ V are the standard deviations of water usage time and water consumption respectively. Through standardization processing, the water usage data of different users are converted to the same dimension for subsequent clustering analysis. Use the clustering algorithm to analyze the standardized data. The objective function of clustering is Initially, first calculate the distances between all data points Select the two data points with the farthest distance as the first two initial clustering centers c 1 and c 2 , and their corresponding coordinates are (x c 1, y c1 ) and (x c 2, y c2 ). For the remaining data points, calculate the distance from each point to the selected clustering centers (where k = 1, 2), select the point with the largest distance to the selected clustering centers as the next clustering center, and repeat this process until k initial clustering centers are selected. Then, assign the data points of each user to the cluster where the nearest clustering center is located, and calculate the new center coordinates of each cluster. For the jth cluster (j = 1, 2,..., k), its new clustering center water usage time mean Water consumption mean where n j is the number of data points in the jth cluster, x ij and y ij are the standardized water usage time and water consumption coordinates of the ith data point in the jth cluster. Repeat the process of assigning and updating the clustering centers until the clustering centers no longer change. At this time, k clustering results C 1 , C2 , …, C k , each cluster represents a type of water use behavior pattern, thus completing the classification of users.

[0040] Subsequently, enter the stage of independent metering zone division and optimization (S300). According to the user classification results and combined with the layout of the water supply network, divide the zones. Construct the water supply network layout as an undirected graph G(V, E), where the node set V represents the nodes in the network, and the edge set E represents the pipes connecting these nodes. For each edge e ij , according to the pipe length L between the nodes i and j it connects ij and the pipe diameter D ij , according to the weight formula Calculate the weight. Based on the user classification results, divide the nodes where users with similar water use behavior characteristics are located into the same subgraph, and divide the entire graph G into multiple independent subgraphs. Each subgraph is an independent metering zone. At the same time, introduce the connected reliability evaluation index CRI to optimize the zoning scheme, and count the historical failure probability p of the pipe segment k k , for the path path from node i to node j ij , its path connection probability Due to the existence of multiple paths, through N simulations, count the number of times m that can be connected between nodes i and j ij , then the connection probability between nodes i and j After obtaining the connection probability P between all node pairs within the zone ij , set the zone connection threshold CRI yz , and calculate the connected reliability index where m is the number of nodes within the zone, is the combination number of selecting two nodes from m nodes. For example, for a preliminarily divided zone, it is found through calculation that its CRI value is lower than the set threshold CRI yz , which indicates that there is a risk in the water supply reliability of this zone in the face of emergencies. Through further analysis and adjustment of the pipe network structure, add the connections of some key pipe segments or standby pipes, recalculate the CRI value, and make it meet the reliability requirements to ensure that the zone can maintain a relatively stable water supply capacity under various working conditions.

[0041] Next, enter the stage of formulating and implementing personalized water supply service strategies (S400). Based on the data of the water use behavior characteristics of users in each zone obtained from the cluster analysis, sort out the water use time data, and count the number of users n using water in each hour period h , form the water use time - user number distribution curve of each zone, and set the peak ratio threshold P high and the trough ratio threshold P low , when When that period is determined as the peak water consumption period; when <P low When it is determined as the low valley period, where N is the number of users after zoning. According to the division results of the peak and low valley periods of each zone, a corresponding pump operation scheduling plan is formulated at the water supply pumping station. Let the initial number of operating pumps be M 0 , and the number of pumps to be increased during the peak period is where h ∈ peak represents the set of hourly periods during the peak period, n avg is the average number of water - using users per hour in this zone, k 1 is a coefficient determined according to the hydraulic characteristics of the pipe network and water supply demand, is the ceiling function. During the peak period, the number of operating pumps is increased to M 0 +ΔM high . Similarly, during the low valley period, the number of operating pumps is gradually reduced by ΔM low , and the where h ∈ valley represents the set of hourly periods during the low valley period, is the floor function, and the pump speed is reduced; at the starting node, end node of each independently metered zone, and nodes on different floors, pressure sensors are installed to collect the water pressure data P s of the pipe network in real - time. The big data analysis platform sets a specific water pressure target value P t,s for each zone according to the water - using behavior characteristics of users in each zone and the pipe network layout, and sets the pressure adjustment coefficient as k p . The initial opening of the intelligent regulating valve is K 0 , and the adjusted opening is K. When P s >P t,s , K = K 0 -k p ×(P s -P t,s ); when P s <P t,s , K = K 0 +k p ×(P t,s -P s) In this way, the water pressure is precisely regulated to ensure that the water pressure in each zone is stable within the set range, meeting the water pressure requirements of different users. At the same time, it prevents problems such as pipeline damage, water leakage, and water-using equipment failures caused by too high or too low water pressure; analyze the water-using characteristics of users in each zone to determine the water quality guarantee requirements for different zones. For zones with high water quality requirements, the raw water is deeply treated in the water supply plant. In the water supply pipeline network, water quality monitoring points are set at intervals to form a water quality monitoring network, and water quality parameters are collected in real time to ensure the water use safety and health of users; according to the zoning results, identify user groups with large water consumption, analyze their daily water use time distribution, water-using equipment usage, and water consumption data, determine the average and peak values of their water use frequency and single water consumption, and formulate personalized water-saving suggestion plans. The suggestions include reasonably arranging water use time, dispersing water use periods, and using water-saving appliances.

[0042] Finally, enter the operation monitoring and feedback adjustment stage (S500). Through real-time monitoring by intelligent water meters and user feedback, continuously collect the water supply data and user opinions of each zone. Feed the monitoring data back to the big data analysis platform to optimize and adjust the user water use behavior analysis and water supply service strategy. The big data analysis platform re-evaluates the water use demand of the zone according to the new data, adjusts the water pump operation scheduling plan, and increases the water supply capacity during peak hours to ensure the stability of user water use. At the same time, according to user feedback, optimize the personalized water-saving suggestions and water quality guarantee measures to further improve the quality of water supply service and user satisfaction.

[0043] In summary, this embodiment details the implementation process of the independent metering zoning method for water supply pipeline networks. Through data collection by intelligent water meters and processing and analysis by the big data analysis platform, accurate user classification and scientific pipeline network zoning are achieved, and targeted personalized water supply service strategies are formulated. During the implementation process, factors such as water use behavior characteristics, pipeline network layout, and water supply reliability are fully considered. Through continuous monitoring and optimization adjustment, the efficiency and quality of water supply management are effectively improved, the water use requirements of different users are met, and the rational utilization of water resources and the stable operation of the water supply system are realized.

[0044] Embodiment 2

[0045] As Figure 2 shown, the specific process of clustering analysis and user classification in the independent metering zoning method for water supply pipeline networks in this embodiment lays a foundation for subsequent accurate zoning and formulation of personalized water supply service strategies.

[0046] First, smart water meters are installed at each user access point in the water supply network. They are responsible for collecting data on the user's water usage time T and water consumption V. These data carry key information about the user's water usage behavior. However, due to the huge differences in the water usage habits, living or production patterns of different users, the original data shows complex and diverse characteristics in terms of numerical range and change trend. For example, the water usage time of residential users is usually closely related to daily life schedules, showing a pattern of morning and evening peaks; while the water usage time of commercial users is mostly concentrated during business hours, and the water consumption also varies greatly depending on the type of business; the water usage time and consumption of industrial users may show specific periodic or batch changes according to the production process arrangement. To eliminate the interference of these differences on subsequent analysis, for the water usage time T, let the original water usage time of the i-th user be T i , the mean of the water usage time of all users is and the standard deviation is σ T , the mean is obtained by summing up the water usage time data of all users and dividing by the total number of users. It reflects the average level of the water usage time of the entire user group. The standard deviation σ T is calculated through a series of complex statistical calculations and is used to measure the degree of dispersion of the data relative to the mean. The standardization formula its function is to transform the water usage time data of different users into a space with a unified standard and comparable scale. In this way, the water usage time data, which originally had a large numerical difference due to individual user differences, can more clearly show its relative position and distribution characteristics in the standardized space, enabling subsequent clustering analysis to be carried out on a relatively fair and unified basis. Similarly, for the water consumption V, the original water consumption of the i-th user is V i , calculate the mean and standard deviation σ V of the water consumption of all users, and use for standardization processing. This process is also to make the water consumption data of different users comparable, so as to better explore the hidden water usage behavior patterns behind the data.

[0047] Then, after completing the standardization, it enters the key stage of clustering analysis. The clustering algorithm adopted conducts iterative optimization around the objective function , where x i and y iis the standardized water usage time and water consumption data of the \(i\)-th user. \(\mu\) and \(\nu\) are the mean water usage time and water consumption of the current cluster center respectively, and \(n\) is the number of users. The essence of the objective function \(C\) is a quantitative expression of the sum of the distances from all data points to their respective cluster centers. When the value of \(C\) is smaller, it means that the distances from each data point to its corresponding cluster center are closer, indicating that the users within the same cluster are more similar in water usage behavior characteristics, and the differences between different clusters are more obvious, thus achieving an ideal clustering effect. In the step of determining the initial cluster centers, it is necessary to calculate the distances between all data points according to the distance formula Calculate the distances between all data points. Here, \(x\) i , \(y\) i and \(x\) j , \(y\) j are the standardized water usage time and water consumption coordinates of the \(i\)-th and \(j\)-th data points respectively. Each data point needs to calculate the distance with all other data points to find the two data points with the farthest distance as the first two initial cluster centers, denoted as \(c\) 1 and \(c\) 2 , and their corresponding coordinates are \((x\) c1 , \(y\) c1 ) and \((x\) c2 , \(y\) c2 ). Taking them as starting points can ensure that different types of water usage behavior patterns can be captured as much as possible. After determining the first two initial cluster centers, for the remaining data points, calculate the distance from each point to the selected cluster centers (where \(k = 1, 2\)), and select the point with the largest distance to the selected cluster centers as the next cluster center, that is, \(c\) 3 , and its coordinates are \((x\) c3 , \(y\) c3 ), satisfying \(D\) i3 =\(\max(D\) i1 , \(D\) i2 ) and \(i = 3, 4, \cdots, n\). This process is repeated continuously until \(k\) initial cluster centers are selected. In this process, each time a new cluster center is selected, it is to find the data point with the largest difference from the existing cluster centers to enrich the diversity and representativeness of the clustering.

[0048] Finally, after determining the initial cluster centers, start to assign the data points of each user to the cluster where the nearest cluster center is located. For the \(j\)-th cluster (\(j = 1, 2, \cdots, k\)), it is necessary to calculate its new center coordinates. The new mean water usage time of the cluster center The mean water consumption where \(n\) j is the number of data points in the \(j\)-th cluster, \(x\) ij and \(y\) ijis the standardized water usage time and water consumption coordinate of the i-th data point in the j-th cluster. This calculation process is based on the mean calculation principle in statistics. By summing up the standardized water usage time and water consumption of all data points within the cluster and dividing by the number of data points, the central coordinate representing the overall water usage behavior characteristics of the cluster is obtained. The process of reassigning and updating the cluster center is repeated. In each iteration, data points may be reassigned to different clusters due to the update of the cluster center, and the cluster center will be recalculated based on the newly assigned data points. This iterative process continues until specific stopping conditions are met. The stopping condition is that the coordinates of the cluster center are small after consecutive iterations. When these stopping conditions are met, it means that the clustering results have tended to be stable. At this time, k clustering results C 1 , C 2 , …, C k are obtained. Each cluster represents a type of water usage behavior pattern, thus successfully completing the classification of users.

[0049] In summary, the precise clustering analysis and classification of water supply network users are achieved, laying a solid foundation for the subsequent formulation of independent metering zoning and personalized water supply service strategies for the water supply network, and strongly promoting the refinement process of water supply network management.

[0050] The above description is only a preferred embodiment of the present invention and does not impose any form of limitation on the present invention. Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to equivalent embodiments by using the disclosed technical content within the scope of the technical solution of the present invention. However, as long as it does not depart from the content of the technical solution of the present invention, any brief modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A method for independent metering and zoning of a water supply network, characterized in that: The specific steps of this method are: S100, installing a smart water meter at a user access point of the water supply network to collect data on the user's water use time T and water consumption V; S200, building a big data analysis platform to standardize and analyze the data collected by the smart water meter to determine the water use behavior characteristics of users in different independent metering zones, perform refined clustering of users based on the water use behavior characteristics, and classify them based on the clustering results; S300. According to the user classification results and the layout of the water supply network, the water supply network is divided into multiple independent metering zones, that is, the independent metering zones of the water supply network are divided based on the undirected graph G(V,E), and the layout of the water supply network is planned as an undirected graph G(V,E), where the node set V represents the nodes in the network, the edge set E represents the pipes connecting these nodes, and each edge is assigned a weight, and the weight value w ij pass Calculate, where L ij is the length of the pipeline connecting data nodes i and j, D ij is the diameter of the pipeline. Based on the user classification results, the nodes of users with similar water use behavior characteristics are divided into the same subgraph, and the entire graph G is divided into multiple independent subgraphs, each of which is an independent metering partition; Users in each independent metering area have similar water use behavior characteristics; S400, formulating a targeted personalized water supply service strategy based on the similar water use behavior characteristics, wherein the personalized water supply service strategy includes a time-sharing water supply plan, differentiated water pressure regulation, customized water quality assurance measures, and personalized water-saving suggestions; Through real-time monitoring of smart water meters and user feedback, water supply service strategies are continuously optimized and adjusted to ensure service effectiveness; S500 monitors the water supply parameters of each zone within the water supply network layout in real time, and feeds the monitoring data back to the big data analysis platform to optimize and adjust user water use behavior analysis and water supply service strategies.

2. A method for independent metering and zoning of a water supply network according to claim 1, characterized in that: The step S200 performs standardization processing on the water use time T and the water use amount V respectively, and the standardization is and Where T i and V i is the original water use time and water consumption data of the i-th user, and They are the mean of water use time and water consumption of all users, σ T and σ V are the standard deviations of water use time and water use amount, respectively. The standardized water use time and water use amount data are recorded as x and y, respectively. Cluster analysis is performed on the standardized data. The cluster analysis process is: The objective function of clustering is Among them, the value of C reflects the sum of the distances from all data points to the cluster center to which they belong, x i and i is the standardized water use time and water consumption data of the i-th user, μ and ν are the mean water use time and water consumption of the current cluster center, and n is the number of users. When μ and ν and the cluster to which the data point belongs are continuously adjusted and updated, C≈0, indicating that the distance between each data point and its cluster center is close, and the water use behavior characteristics of users in the same cluster are similar; Initially, k data points are randomly selected as the initial cluster centers, and each user's data point is assigned to the cluster with the nearest cluster center. That is, each user i and j is a different data node i and j. The new center coordinates of each cluster are calculated. For the jth cluster (j = 1, 2, ..., k), the new cluster center water usage time mean μ j and the mean water consumption ν j for: where n j is the number of data points in the jth cluster, x ij and ij is the standardized water use time and water consumption coordinates of the ith data point in the jth cluster. The process of allocating and updating cluster centers is repeated until the cluster center no longer changes. At this time, k clustering results are obtained, namely C1, C2, …, C k ,Each cluster represents a type of water use behavior pattern, thus completing the classification of users.

3. A method for independent metering and zoning of a water supply network according to claim 2, characterized in that: When determining the initial cluster center, S200 calculates the distance between all data points, and the distance is where x i ,y i and x j ,y j The coordinates of the standardized water use time and water consumption of the i-th and j-th data points are selected, and the one with the longest distance, i.e., d ij The two data points with the largest values ​​are taken as the first two initial cluster centers, denoted as c1 and c2, and their corresponding coordinates are (x c1 ,y c1 ) and (x c2 ,y c2 ), for the remaining data points, calculate the distance from each point to the selected cluster center Where k = 1, 2, select the point with the largest distance to the selected cluster center as the next cluster center, that is, c3, whose coordinates are (x c3 ,y c3 ), satisfying D i3 =max(D i1 ,D i2 ) and i = 3, 4, ..., n, and repeat the process until k initial cluster centers are selected.

4. A method for independent metering and zoning of a water supply network according to claim 1, characterized in that: When S300 divides the independent metering zones, it not only considers the user classification results and the pipe network layout, but also introduces the connectivity reliability evaluation index CRI to optimize the zone scheme and measure the ability of the zone to maintain water supply in the face of emergencies. The connectivity reliability evaluation index CRI is calculated by calculating the connectivity probability P between any two nodes i and j. ij To measure the reliability of the partition, the historical failure probability p of pipe segment k is calculated. k , the path from node i to node j is path ij , the path connectivity probability is the product of the normal working probabilities of all pipe sections on the path, that is, Moreover, there are multiple paths from node i to node j, so the connectivity probability between nodes i and j is P ij Perform N simulations based on all paths. In each simulation, count the number of times nodes i and j can be connected. ij ,but The connectivity probability P between all node pairs in the partition is obtained. ij Then, set the partition connectivity threshold CR I yz , and calculate the connectivity reliability index CRI as Where m is the number of nodes in the partition, The number of combinations of two nodes is selected from m nodes to realize the independent metering zoning of the water supply network that satisfies the similarity of users' water use behavior characteristics and has reliability.

5. A method for independent metering and zoning of a water supply network according to claim 4, characterized in that: The connectivity reliability index and the partition connectivity threshold CRI yz The relationship is: When the connectivity reliability index CRI of the partition is less than CRI yz When the water supply reliability of the independent metering area is at risk in the face of emergencies, When the connectivity reliability index CRI of the partition is less than CRI yz It means that when the independent metering partition faces an emergency, the nodes within the partition can maintain connectivity and stabilize water supply.

6. A method for independent metering and zoning of a water supply network according to claim 1, characterized in that: The process of the S400 time-sharing water supply plan is: According to the water use behavior characteristic data of users in each partition determined in S200, the water use time data is sorted out, and the number of users n who use water in each hour period is counted. h , and set the peak and valley time judgment as P high and P low ,when When When , it is defined as the valley period, where N is the number of users after partitioning; According to the results of the peak and valley period division of each zone, a corresponding water pump operation scheduling plan is formulated at the water supply pump station, that is, the number of operating pumps ΔM is increased during the peak period. high for Where h∈peak represents the set of hourly periods during the peak period, n avg is the average number of water users per hour in the zone, k1 is the adjustment coefficient, To round up the function, increase the number of running pumps to M0+ΔM during peak hours high , M0 is the number of running pumps at the beginning to maintain a stable water supply; similarly, during the off-peak period, the number of running pumps ΔM is gradually reduced low , Where h∈valley represents the set of small time periods during the valley period, It is a round-down function and reduces the pump speed.

7. A method for independent metering and zoning of a water supply network according to claim 1, characterized in that: The specific process of the S400 differentiated water pressure regulation is: install pressure sensors at the starting node, end node and nodes on different floors of each independent metering partition to collect the water pressure data P of the pipe network in real time. s , s represents the sth node, and transmits it to the big data analysis platform. The big data analysis platform sets a specific water pressure target value P for each partition based on the water use behavior characteristics of users in each partition and the layout of the pipe network. t,s , t represents the partition number, s represents the node number in the partition, and the pressure adjustment coefficient is set to k p , the initial opening of the regulating valve in the smart water meter is K0, and the opening after adjustment is K. When P s >P t,s When K = K0-k p ×(P s -P t,s ); when P s <P t,s When K = K0 + k p ×(P t,s -P s ) to precisely adjust the water pressure.

8. The method for independent metering and zoning of a water supply network according to claim 1, characterized in that: The S400 customized water quality assurance measures are: analyzing the water use characteristics of users in each zone, determining the water quality assurance requirements of different zones, deeply treating the raw water in the water supply plant for zones with high water quality requirements, setting water quality monitoring points at intervals in the water supply network to form a water quality monitoring network, collecting water quality parameters in real time, and ensuring the safety and health of users' water use; The personalized water-saving suggestions identify user groups with high water consumption based on the zoning results, analyze their daily water use time distribution, water equipment usage and water consumption data, determine their water use frequency and the average and peak values ​​of single water consumption, and formulate personalized water-saving suggestion plans. The suggestions include reasonably arranging water use time, dispersing water use time periods, and using water-saving appliances.

Images