Automatic partitioning method for water supply and drainage pipe network based on topological distance clustering
By amplifying the topological distances of pressure reducing valves, water pumps, and administrative boundary pipelines in the water supply and drainage network partitioning, and combining the Floyd algorithm and clustering methods, the problem of inaccurate partitioning in existing technologies is solved, and more accurate and reliable partitioning results are achieved.
Patent Information
- Application Number
- CN202410846615.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-27
- Publication Date
- 2024-11-19
AI Technical Summary
The existing water supply and drainage network zoning method fails to effectively consider pressure reducing valves, water pumps and pipelines that cross administrative zones, resulting in zoning results that do not match reality and uneven zoning scale.
By amplifying the topological distances of pressure-reducing valves, water pumps, and pipelines crossing administrative divisions when constructing the topological distance matrix, and combining the Floyd algorithm and clustering methods, tight and sparse connections of nodes are achieved, and clustering partitioning is performed.
It improves the accuracy and reliability of water supply and drainage network zoning, ensures that the zoning results conform to the actual situation, and reduces the differences in zoning scale.
Smart Images

Figure FT_1 
Figure FT_2 
Figure FT_3
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of municipal engineering and urban water supply and drainage pipe networks, and in particular relates to an automatic zoning method for a water supply and drainage pipe network based on topological distance clustering. Background Art
[0002] With the expansion of urban areas, the scale of water supply and drainage networks is also expanding. It is necessary to accurately zone the water supply and drainage networks to facilitate the measurement of water supply and drainage, thereby reducing water supply network leakage and achieving mixed connection and infiltration diagnosis of different drainage areas. However, due to the large volume and complex topological structure of the water supply and drainage network, manual division of water supply and drainage network zoning faces a series of challenges. In recent years, the concept of community-based water supply and drainage network zoning has been proposed, which takes modularity as a key indicator and determines the zoning by maximizing modularity. The main problem is that the scale of different zonings will vary greatly, which does not meet the uniformity requirements. Therefore, how to detect the inherent community structure of the water supply and drainage network during the zoning process and ensure the size of different zonings is a current problem.
[0003] In addition, the current community-based water supply and drainage network zoning method cannot consider the impact of pressure reducing valves, pumping stations and pipelines crossing administrative divisions on the zoning, resulting in zoning results that are inconsistent with reality. In view of this, the present invention proposes a water supply and drainage network zoning method based on topological distance clustering, which effectively improves the accuracy of water supply and drainage network division by amplifying the pipeline topological distance including pressure reducing valves, pumping stations and administrative boundaries. Summary of the invention
[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method for automatically partitioning a water supply and drainage network based on topological distance clustering, by which the water supply and drainage network is accurately partitioned. The technical solution of the present invention is as follows: A method for automatic partitioning of water supply and drainage pipe networks based on topological distance clustering includes the following steps: Step 1, mapping the pipe network into an undirected graph; Step 2, construct an adjacency matrix based on the undirected graph to obtain the connection relationship between nodes; Step 3: In the zoning process, pressure reducing valves (PRVs), water pumps, and pipelines crossing administrative zones are considered as zone boundaries. When constructing the topological distance matrix, the topological distances of pressure reducing valves, water pumps, and pipelines crossing administrative zones are magnified. In the magnification process, considering that the distance from any point in the pipe network to the farthest node will not exceed the total number of pipe sections, the topological distances of the pressure reducing valves, water pumps, and pipelines crossing administrative zones are specified as the total number of pipelines, and the topological distances of other pipelines are specified as 1. Step 4: Based on steps 1 to 3, the Floyd algorithm is selected to calculate the final shortest path matrix, i.e., the topological distance matrix, in which each element represents the number of pipelines in the shortest path between nodes; Step 5: Based on the topological distance matrix constructed in step 4, low-dimensional points are mapped to high dimensions, and a clustering method is selected to cluster the high-dimensional points. By clustering nodes with smaller topological distances, densely connected nodes are formed in each cluster partition, while maintaining sparse connections between nodes in different cluster partitions, thereby achieving cluster partitioning.
[0005] The clustering method is one of K-means, fuzzy clustering and hierarchical clustering algorithms.
[0006] Compared with the existing technology, the present invention has the following advantages and beneficial effects: The automatic water supply and drainage network partitioning method based on topological distance clustering proposed in the present invention is of great significance for correctly dividing the water supply and drainage network partitions. Compared with the community-based partitioning method, the latter fails to fully consider the impact of pressure reducing valves, water pumps, and pipelines crossing administrative partitions on the partitioning results. The method of the present invention fully considers the importance of pressure reducing valves, water pumps, and pipelines crossing administrative boundaries as key factors in the partitioning process. Therefore, when constructing the topological distance matrix, the present invention improves the accuracy of the partitioning by amplifying the topological distance of these key pipelines. BRIEF DESCRIPTION OF THE DRAWINGS
[0007] Figure 1 is a flow chart of the present invention; Figure 2 is the topological distance matrix obtained by applying it to the small pipe network; Figure 3 The zoning results of geographical clustering and topological distance clustering applied to small pipe networks; Figure 4 The zoning results of the two methods applied to the C-Town pipe network; Figure 5 is the topological distance matrix obtained by applying it to the drainage network; Figure 6 This is the zoning result applied to the drainage network. DETAILED DESCRIPTION
[0008] The present invention is further described below in conjunction with the accompanying drawings and embodiments, but the protection scope of the present invention is not limited by the embodiments.
[0009] like Figure 1 The method for automatically partitioning a water supply and drainage network based on topological distance clustering shown in FIG. 1 includes the following steps: Step 1, mapping the pipe network into an undirected graph; Step 2, construct an adjacency matrix based on the undirected graph to obtain the connection relationship between nodes; Step 3: In the zoning process, pressure reducing valves (PRVs), water pumps, and pipelines crossing administrative zones are considered as zone boundaries. When constructing the topological distance matrix, the topological distances of pressure reducing valves, water pumps, and pipelines crossing administrative zones are magnified. In the magnification process, considering that the distance from any point in the pipe network to the farthest node will not exceed the total number of pipe sections, the topological distances of the pressure reducing valves, water pumps, and pipelines crossing administrative zones are specified as the total number of pipelines, and the topological distances of other pipelines are specified as 1. Step 4: Based on steps 1 to 3, the Floyd algorithm is selected to calculate the final shortest path matrix, i.e., the topological distance matrix, in which each element represents the number of pipelines in the shortest path between nodes; Step 5: Based on the topological distance matrix constructed in step 4, low-dimensional points are mapped to high dimensions, and a clustering method is selected to cluster the high-dimensional points. By clustering nodes with smaller topological distances, densely connected nodes are formed in each cluster partition, while maintaining sparse connections between nodes in different cluster partitions, thereby achieving cluster partitioning. Example 1
[0010] Figure 3 a is the partition result based on topological distance clustering applied to the small pipe network. The above steps 1 to 5 are used to Figure 3 The small pipe network shown in a is divided into two parts: Figure 3 There are a total of 7 pipelines in the pipeline network of a. The topological distance of the pipeline containing the pressure reducing valve is specified as 7, and the distance of other pipelines is specified as 1. That is, the topological distance between node 1 and node 5 is 7, thus constructing the following Figure 2 The topological distance matrix shown; according to Figure 2 A topological distance matrix is used to map low-dimensional points to high-dimensional points. A clustering method is selected to cluster high-dimensional points. By clustering nodes with smaller topological distances, tightly connected nodes are formed in each cluster partition, while sparse connections are maintained between nodes in different cluster partitions, thereby partitioning the small pipe network. This embodiment achieves accurate partitioning results by amplifying the topological distance including the pressure reducing valve.
[0011] Comparative Example 1
[0012] Figure 3 b is the partitioning result of the geographical clustering-based partitioning method applied to the small pipe network. Figure 3 a and Figure 3 b It can be seen that the partitioning method based on geographic clustering has divided nodes 4 and 8 into incorrect groups, resulting in nodes 1, 2, 3, and 8 in the same group and nodes 5, 6, 7, and 4 in another group. In contrast, the clustering method based on topological distance correctly clusters node 4 with nodes 1, 2, and 3, and node 8 with nodes 5, 6, and 7. Example 2
[0013] Figure 6 To apply the zoning results to the drainage network, the above steps 1 to 5 are used to Figure 6 The drainage network shown is divided into zones: Figure 6 There are 49 pipelines in the pipeline network. The topological distance of pipelines crossing administrative boundaries is specified as 49, and the distance of other pipelines is specified as 1. That is, the topological distance between node 18 and node 19 is 49, and the topological distance between node 31 and node 40 is 49, thus constructing the following Figure 5 The topological distance matrix shown; according to Figure 5 The topological distance matrix is used to map low-dimensional points to high-dimensional points. A clustering method is selected to cluster high-dimensional points. By clustering nodes with small topological distances, closely connected nodes are formed in each cluster partition, while sparse connections are maintained between nodes in different cluster partitions, thereby partitioning the drainage network. Figure 6 It can be seen that the present invention takes into account the impact of pipelines crossing administrative boundaries on the partitioning results, thereby achieving accurate partitioning results. Figure 5 The topological distance matrix in the paper is used to effectively partition the drainage network, making the results of the network partition more in line with the actual situation and improving its reliability and practicality.
[0014] Comparative Example 2
[0015] Figure 4 The topological distance method ( Figure 4 a) and geographic clustering-based methods ( Figure 4 b) Results when applied to the C-Town pipe network. Figure 4 The results of a show that the topological distance clustering method shows a highly accurate partitioning effect when there are pumping stations and administrative boundary pipelines in the pipeline network. In contrast, the traditional method based on geographic clustering ( Figure 4 b) The pipeline containing the water pump was not considered as the boundary during the partition. These results clearly show the accuracy of the proposed method in the real pipe network partition. Especially when there are complex situations such as pumping stations and administrative boundary pipelines in the pipe network, the topological distance clustering method can effectively realize accurate pipe network partitioning, while the traditional method may lead to partitioning results that do not conform to the actual class.
[0016] It can be seen that, considering the influence of pressure reducing valves, pumping stations and administrative boundaries on the partitioning results, the topological distance of the three is magnified to ensure the accuracy of the partitioning. The present invention fully considers the influence of these key factors on the pipe network partitioning, and adjusts their effects during the processing process to achieve more accurate and reliable partitioning results.
Claims
1. A method for automatic partitioning of water supply and drainage network based on topological distance clustering, characterized in that: The following steps are included: Step 1: Map the pipe network into an undirected graph; Step 2: construct an adjacency matrix based on the undirected graph to obtain the connection relationship between nodes; Step 3: Specify the topological distance of the pipelines including the pressure reducing valve, the water pump, and the pipelines crossing the administrative division as the total number of pipelines, and the topological distance of other pipelines as 1; Step 4: Based on steps 1 to 3, the Floyd algorithm is selected to calculate the final shortest path matrix, i.e., the topological distance matrix. Each element of the matrix represents the number of pipes in the shortest path between nodes. Step 5: Based on the topological distance matrix constructed in step 4, map the low-dimensional points to high-dimensional ones, select a clustering method to cluster the high-dimensional points, and cluster the nodes with smaller topological distances so that closely connected nodes are formed in each cluster partition. At the same time, sparse connections are maintained between nodes in different cluster partitions, thereby achieving cluster partitioning.
