Multi-target partition optimization method and system for multi-water-source large water supply pipe network

Abstract

The invention discloses a multi-target partition optimization method and system for a multi-water-source large water supply pipe network. The method comprises the steps of setting optimization basic parameters, identifying a main pipe, constructing a partition scheme, constructing a multi-target function, constructing partition constraint conditions and optimizing a model solving method. According to the multi-target water supply pipe network zoning method, pressure and metering management can be carried out conveniently, and meanwhile transformation economy is considered. The zoning method takes reduction of pipe network management difficulty as a core, pipe network operation transformation economy, zoning internal pressure uniformity and pipe network water quality safety are taken into consideration, a large water supply pipe network transmission and distribution separation pattern is constructed to clarify a pipeline transmission and distribution function, a candidate zoning scheme set is formed in combination with a spatial clustering algorithm, and a large water supply pipe network is obtained. And finally, comparing all optimization targets in the candidate schemes, and carrying out iterative updating by adopting a non-dominated sorting genetic algorithm to obtain a Pareto leading edge of the zoning scheme, so as to realize refined leakage control of the large-scale water supply pipe network.

Description

Technical Field

[0001] The present invention relates to the technical field of water supply network zoning optimization, and in particular to a method and system for zoning optimization of a large-scale water supply network with multiple water sources. Background Art

[0002] The water supply system is one of the most important municipal facilities in the city, and plays a vital role in ensuring the daily life of urban residents and industrial production. With the rapid advancement of my country's urbanization process in recent years, the scale of urban water supply pipelines has continued to expand, but it also faces a series of management and technical challenges such as corrosion and aging of pipelines, increased risks of hydraulic and water quality of pipelines, and increased difficulty in leakage control.

[0003] Under the traditional management model, the operation and monitoring of water supply networks often rely on manual experience. This method is not only time-consuming and labor-intensive, but also difficult to adapt to the management needs of large-scale and complex pipe network systems. Therefore, how to achieve efficient and safe operation of water supply pipe network systems by improving the level of management technology has become an important issue that water supply companies need to solve urgently.

[0004] The regionalization of the water supply network, i.e., the zoning of the water supply network, is an efficient technology for managing and monitoring the water supply system. By dividing the water supply network into several relatively independent transmission and distribution areas, and combining flow monitoring, pressure control, water quality testing and other means, efficient management goals such as independent water volume measurement, pressure optimization and regulation, and water quality safety control can be achieved. This zoning management method can significantly reduce the difficulty of network management and improve the overall operating efficiency of the water supply system.

[0005] At present, the methods of water supply network zoning are mainly divided into two categories: manual experience zoning and automatic zoning. Although the manual experience zoning method is simple and flexible, it relies on expert experience, is time-consuming and inefficient, and is usually only applicable to small pipe network systems. The automatic zoning method based on computer technology, pipe network model and optimization technology has the advantages of fast speed, high accuracy and strong adaptability, and is more suitable for zoning management of large-scale and complex pipe network systems, especially modern intelligent water management.

[0006] The commonly used automatic partitioning methods include cluster analysis, community discovery, path analysis, breadth-first search, depth-first search, etc. At the same time, some scholars have combined neural network algorithms for partitioning, but there are few practical applications. Different automatic partitioning methods have their own characteristics, but there are still certain limitations in the application of ultra-large pipe networks. For example, the path analysis method based on graph theory takes a long time to process ultra-large pipe networks, the community discovery method has high requirements for data quality, the cluster analysis method lacks in-depth analysis of the current water supply path, and the existing partitioning methods have not yet incorporated energy consumption accounting into the partitioning process.

[0007] Based on the above problems, the present application proposes a multi-objective efficient automatic zoning optimization method and system for large multi-source water supply pipe networks. This method takes the separation of water transmission and distribution as the core concept, combines clustering analysis, water transmission path optimization, and multi-objective optimization algorithms, aiming to achieve rapid optimization zoning of ultra-large-scale water supply pipe networks and provide technical support for the intelligent, refined, and efficient management of water supply pipe networks. Summary of the Invention

[0008] The purpose of the present invention is to provide a multi-objective zoning optimization method and system for large multi-source water supply pipe networks to solve the problems raised in the above technical background.

[0009] To achieve the above object, the present invention adopts the following technical solutions:

[0010] In the first aspect, the present application provides a multi-objective zoning optimization method for large multi-source water supply pipe networks, including the following steps:

[0011] S1. Select a water supply pipe network and establish a basic parameter module. The basic parameter module includes the maximum and minimum number of zones, node service head, minimum identified diameter of the main pipe, minimum length of the branch pipe, and ideal distribution pipe diameter.

[0012] S2. Based on the principle of separating water transmission and distribution and taking into account the connectivity between multiple water sources, identify the water transmission pipelines in the large water supply pipe network, including:

[0013] Select the pipelines with a diameter greater than a preset first threshold as the diameter-based main pipes, select the pipelines with a flow rate greater than a preset second threshold as the flow rate-based main pipes, and take the union of the diameter-based main pipes and the flow rate-based main pipes as the basic main pipes.

[0014] If the basic main pipes cannot connect all water source nodes in series, analyze the connectivity between each node of the basic main pipes and the water source nodes, and use the Dijkstra algorithm to calculate the shortest paths from different nodes of the basic main pipes to the water source nodes to determine the supplementary main pipes that need to be added. Among them, the shortest path is defined as the path with the minimum sum of weights of any connected nodes, and the calculation formula for the weight is:

[0015] w i = 10.67C i -1.852 D i -4.871 l i

[0016] In the formula, C i represents the Hazen-Williams coefficient of pipe section i; D i represents the diameter of pipe section i, in m; l i represents the length of pipe section i, in m;

[0017] The combination of the basic main pipe and the supplementary main pipe constitutes the main pipe;

[0018] S3. After the main pipe is identified, a zoning scheme is constructed, including:

[0019] Separate the main pipe from the original pipe network so that it no longer undertakes the water distribution function; search for all branch pipes connected to the main pipe along the main pipe to determine the candidate water distribution pipe set; randomly select a water distribution pipe from the candidate water distribution pipe set as the zoned water distribution pipe to form an initial scheme set of water distribution pipe selection; according to the selected water distribution pipe and the preset zoned inlet position, use the spatial clustering algorithm to group the pipe network nodes to form zoned boundaries, thereby constructing multiple zoning schemes to realize the zoning of nodes;

[0020] S4. Construct a multi-objective function, which includes the standard deviation of the average pressure in the zone, the average water age of the nodes, the energy consumption of the pipe network operation, and the renovation cost;

[0021] S5. According to the multi-objective function, use the non-dominated sorting genetic algorithm NSGA-II to optimize and solve the zoning scheme.

[0022] In the above content, the separation of water transmission and distribution in the water supply pipe network refers to a layout method of the water supply system in which the water transmission and distribution functions in the water supply pipe network are separated and designed. This layout method aims to optimize the operation efficiency and water quality management of the water supply system by clearly defining the functions of the water transmission pipe and the water distribution pipe. Specifically, the water transmission pipe is responsible for transporting water from the water source (such as a reservoir, a pumping station, etc.) to other parts of the water supply system, usually a large-diameter pipe for long-distance and high-flow water transmission; the water distribution pipe is responsible for distributing the treated water to each water use point or user, usually a small-diameter pipe for short-distance and low-flow water distribution.

[0023] In the above content, the shortest path is the path calculated based on the pipe weight. Specifically, it refers to the path with the smallest sum of the weights of all the pipes (or nodes) passed from the starting point to the ending point. The "weight" here is a value calculated according to certain attributes of the pipe (such as the Hazen-Williams coefficient, pipe diameter, pipe section length, etc.), which is used to measure the performance or cost of the pipe during the water transmission process.

[0024] In a preferred embodiment, in the step S1, the setting of the basic parameter module satisfies the following rules:

[0025] The DMA (District Metering Area) zoning scale is set to 500 - 5000 households;

[0026] The node service head is determined according to local specifications;

[0027] The minimum identifiable pipe diameter of the main pipe shall not be less than 400 mm to ensure reasonable water supply energy consumption. After the main pipe is identified, large-diameter pipes shall be selected as distribution pipes first to reduce the energy consumption loss within the district.

[0028] In the above content, the scale of the DMA district is usually set to cover 500 - 5000 households. A smaller district scale will increase the management difficulty for water supply enterprises. Therefore, to facilitate management, the range of the number of districts is limited. This preset range of the number of districts aims to ensure that each DMA district is neither too large, increasing the management difficulty, nor too small, resulting in too high management costs.

[0029] In a preferred embodiment, step S3 specifically includes the following steps:

[0030] S3.1, According to the actual situation of the water supply network, set the ideal diameter of the distribution pipe. Pipes with an actual diameter larger than the ideal diameter of the distribution pipe shall be used as the ideal distribution pipes, and the main pipe shall be separated from the original network so that it no longer undertakes the distribution function;

[0031] S3.2, If the number of ideal distribution pipes in the area is sufficient, only these pipes shall be selected as candidate distribution pipes; otherwise, all distribution pipes in the area shall be included in the candidate distribution pipe set;

[0032] S3.3, Randomly select N dp distribution pipes from the candidate distribution pipe set as the district distribution pipes, and repeat this random process to generate an initial set of distribution pipe selection schemes; where N dp shall meet the following conditions:

[0033]

[0034] In the formula: N dp represents the number of distribution pipes selected in the pipe network; represents the preset minimum number of districts; represents the preset maximum number of districts;

[0035] S3.4, After the distribution pipes are selected, for each district scheme, according to the distribution pipes and the positions of the district entrances, use the spatial clustering algorithm to assign each node to different districts to determine the node grouping;

[0036] During the clustering process, based on the entrance node, cluster the non-main pipe adjacent nodes according to the topological structure of the water supply network; after clustering, add the water demand of the adjacent nodes to the water demand of the corresponding entrance node;

[0037] If the water demand of the entrance node exceeds the preset district limit water demand, the clustering of this entrance node shall stop, where the calculation formula of the district limit water demand is as follows:

[0038]

[0039] In the formula: represents the maximum restricted water demand of the sub - area; represents the water demand of node i at time t; T represents the total number of time steps of the extended simulation; N represents the total number of nodes in the pipe network; represents the minimum number of sub - areas;

[0040] S3.5. After completing node clustering, all nodes within all sub - areas are assigned to the corresponding inlet nodes, and the nodes associated with each inlet node form an independent sub - area. The pipe segments between different groups of nodes form the boundary pipe segments of the sub - areas;

[0041] The boundary pipe segments of the sub - areas and the unselected water distribution pipes are truncated by valves to achieve independence between each sub - area.

[0042] In a preferred embodiment, in step S4, taking the standard deviation of the average pressure of the sub - areas as the objective function, it includes the following steps:

[0043] The uniformity of the pressure within the sub - area is characterized by the average value of the standard deviation of the node pressures within the sub - area through an extended simulation of 24 hours. Among them, the smaller the average value, the better the uniformity; assuming that there are a total of K sub - areas after sub - area division, taking the ratio SDR of the average standard deviation of the pressures of each sub - area of the pipe network after sub - area division to the average standard deviation of the pressures of each sub - area before sub - area division as the objective function, the formula is as follows:

[0044]

[0045] In the formula: n k represents the total number of nodes within the k - th sub - area; T represents the total number of simulation time steps during the extended simulation; represents the pressure of node i at time t within the k - th sub - area, with the unit of m; represents the average pressure of all nodes at time t within the k - th sub - area, with the unit of m; represents the pressure of node i at time t within the k - th sub - area before sub - area division, with the unit of m; represents the average pressure of all nodes at time t within the k - th sub - area before sub - area division, with the unit of m.

[0046] In a preferred embodiment, in step S4, taking the average water age of the nodes as the objective function, it includes the following steps:

[0047] The average water age of the nodes through a 24 - hour extended simulation of the pipe network is used to characterize the water quality of the pipe network. Assuming that there are a total of N nodes in the pipe network, the ratio WAR of the average water age of the nodes before and after sub - area division is used as the evaluation index for the water quality of the sub - areas, and the formula is as follows:

[0048]

[0049] In the formula: represents the water age of node i at time t; represents the water age of node i at time t without zoning.

[0050] In a preferred embodiment, in step S4, taking the energy consumption and renovation cost of the pipe network operation as the objective function, the following steps are included:

[0051] To ensure the economy of the renovation and the long-term operation energy consumption after renovation, make the number of newly added valves and monitoring devices have less cost, and the internal operation energy consumption of the pipe network after zoning is smaller. Taking the minimization of the pipe network operation and renovation cost C as the goal, the formula is as follows:

[0052]

[0053] In the formula, C is the pipe network operation and renovation cost, and K s represents the total number of SCADA devices to be installed; K v represents the total number of valves to be installed; represents the nth s cost of the monitoring device, and the device cost is related to the pipe diameter; represents the nth v cost of the valve, and the valve cost is related to the pipe diameter; y represents the designed service life of the pipe network after renovation, in years; m represents the total number of pipes in the pipe network; ρ represents the density of water, in kg / m 3 ; g represents the local acceleration of gravity, in m 2 / s; represents the flow rate of pipe l at time t after zoning, in m 3 / s; represents the head loss of pipe l at time t after zoning, in m; η represents the empirical coefficient, the average efficiency of all pumps in the water supply pipe network, and the value range is 0.6 to 0.9.

[0054] In a preferred embodiment, step S5 includes the following steps:

[0055] S5.1, randomly generate the first-generation population P0, the population consists of N individuals, and the chromosome of each individual corresponds to a water distribution pipe screening scheme, and the water distribution pipe screening scheme is expressed by a binary number with a length equal to the number of candidate zoning inlets;

[0056] In the water distribution pipe screening scheme, "1" means that a certain water distribution pipe is selected, and a sensor must be installed on this water distribution pipe to monitor its flow rate and the pressure at the zoning inlet; "0" means that a certain water distribution pipe is not selected, and a valve must be installed on this water distribution pipe to cut it off;

[0057] S5.3. Perform a time-delay simulation on the zoning scheme for each individual, calculate the objective function and evaluate the constraint conditions, sort the individuals based on the non-dominated sorting method, and select individuals according to the sorting result until the population upper limit is reached; if the population upper limit is exceeded, exclude some individuals according to the crowding distance to form the next-generation parent population P1.

[0058] S5.4. The parent population P1 generates the next-generation offspring population Q1 through selection, crossover, and mutation. Repeat the iterative process to obtain P2, and continuously iterate until the preset algorithm termination condition is met to obtain the final-generation population P. last , population P last is the Pareto approximate optimal solution.

[0059] In the second aspect, the present application provides a multi-source large-scale water supply pipe network multi-objective zoning optimization system for implementing the multi-source large-scale water supply pipe network multi-objective zoning optimization method described in the first aspect. The system includes:

[0060] A basic parameter module establishment unit for selecting a water supply pipe network and establishing a basic parameter module. The basic parameter module includes the maximum and minimum number of zones, node service head, minimum identified pipe diameter of the main pipe, minimum length of the branch pipe, and ideal distribution pipe diameter.

[0061] A main pipe identification unit that, based on the principle of separating water transmission and distribution and taking into account the connectivity between multiple water sources, identifies the water transmission pipelines in the large-scale water supply pipe network, specifically including:

[0062] A basic main pipe determination module for selecting pipes with a diameter greater than a preset first threshold as the diameter-based main pipes, selecting pipes with a flow rate greater than a preset second threshold as the flow rate-based main pipes, and taking the union of the diameter-based main pipes and the flow rate-based main pipes as the basic main pipes.

[0063] A supplementary main pipe determination module. If the basic main pipes cannot connect all water source nodes in series, analyze the connectivity between each node of the basic main pipes and the water source nodes, and use the Dijkstra algorithm to calculate the shortest paths from different nodes of the basic main pipes to the water source nodes to determine the supplementary main pipes that need to be added; among them, the shortest path is defined as the path with the smallest sum of weights of any connected nodes, and the calculation formula for the weight is:

[0064] w i = 10.67C i -1.852 D i -4.871 l i

[0065] In the formula, C i represents the Hazen-Williams coefficient of pipe section i; D iDenote the pipe diameter of pipe section i, with the unit of m; l i Denote the length of pipe section i, with the unit of m;

[0066] The main pipe determination module is used to form the main pipe by combining the set of the basic main pipe and the supplementary main pipe;

[0067] The partition scheme construction unit is used to construct the partition scheme after the main pipe is identified, specifically including:

[0068] The main pipe separation module is used to separate the main pipe from the original pipe network so that it no longer undertakes the water distribution function;

[0069] The candidate water distribution pipe set determination module searches for all branch pipes connected to the main pipe along the main pipe to determine the candidate water distribution pipe set;

[0070] The water distribution pipe selection module randomly selects a water distribution pipe from the candidate water distribution pipe set as the partition water distribution pipe to form an initial scheme set of water distribution pipe selection;

[0071] The node grouping and partition boundary formation module groups the pipe network nodes by using a spatial clustering algorithm according to the selected water distribution pipe and the preset partition entrance position to form a partition boundary, thereby constructing multiple partition schemes to realize the partition of nodes;

[0072] The multi-objective function construction unit is used to construct a multi-objective function including the standard deviation of the average pressure in the partition, the average water age of the nodes, the operation energy consumption of the pipe network, and the renovation cost;

[0073] The optimization solution unit optimizes and solves the partition scheme by using the non-dominated sorting genetic algorithm NSGA-II according to the multi-objective function to obtain the optimal or approximately optimal partition scheme.

[0074] In the third aspect, the present application also provides an electronic device, which includes: at least one processor, a memory, and an input / output unit; wherein, the memory is used to store a computer program, and the processor is used to call the computer program stored in the memory to execute the steps of the method described in any one of the first aspects.

[0075] In the fourth aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it realizes the steps of the method described in any one of the first aspects.

[0076] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0077] The present application discloses a multi-objective zoning optimization method and system for a large multi-source water supply pipe network. The method includes setting optimization basic parameters, identifying main pipes, constructing a zoning scheme, constructing a multi-objective function, zoning constraint conditions, and an optimization model solving method. The present application proposes a multi-objective water supply pipe network zoning method that is convenient for implementing pressure and metering management while taking into account the economy of renovation. The zoning method takes reducing the management difficulty of the pipe network as the core, taking into account the economy of pipe network operation and renovation, the pressure uniformity within the zone, and the water quality safety of the pipe network. By constructing a clear separation pattern of water transmission and distribution in the large water supply pipe network to clarify the water transmission and distribution functions of the pipes, and combining with the spatial clustering algorithm to form a set of candidate zoning schemes, and finally comparing the optimization objectives in the candidate schemes and using the non-dominated sorting genetic algorithm to iteratively update to obtain the Pareto front of the zoning scheme, so as to achieve refined control of water leakage in the large water supply pipe network. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required to be used in the description of the specific embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0079] Figure 1 is a schematic flow chart of a multi-objective zoning optimization method for a large multi-source water supply pipe network in a preferred embodiment of the present invention;

[0080] Figure 2 is a topological structure diagram of the water supply pipe network of a certain city shown in a preferred embodiment of the present invention;

[0081] Figure 3 is a schematic diagram of identifying main pipelines shown in a preferred embodiment of the present invention;

[0082] Figure 4 is a topological schematic diagram of the pipe network before and after stripping the main pipes shown in a preferred embodiment of the present invention;

[0083] Figure 5 is the objective function value of the zoning scheme set in a preferred embodiment of the present invention;

[0084] Figure 6 is a schematic diagram of the zoning layout of a certain scheme in a preferred embodiment of the present invention;

[0085] Figure 7 is a schematic structural diagram of a multi-objective zoning optimization system for a large multi-source water supply pipe network in a preferred embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0086] To make the above and other features and advantages of the present invention clearer, the present invention will be further described below with reference to the accompanying drawings. It should be understood that the specific embodiments given herein are for the purpose of explaining to those skilled in the art and are merely exemplary, not restrictive.

[0087] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of the present invention, "a plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0088] Embodiment 1

[0089] As Figure 1 shown, this embodiment provides a multi-objective zoning optimization method for a large multi-source water supply pipe network, which specifically includes the following steps:

[0090] Step S1. Set basic parameters: Select a water supply pipe network and establish a basic parameter module to ensure that the zoning meets the management requirements of users and enterprises. The basic parameter module includes the maximum and minimum number of zones, the node service head, the minimum recognized diameter of the main pipe, the minimum length of the branch pipe, and the ideal distribution pipe diameter.

[0091] Step S1.1. Select the water supply pipe network system of a certain city as the case analysis object. The service area of this region is about 568 km 2 , including 4 water treatment plants, the maximum daily water consumption is 2.37 million m 3 / d, the total length of pipes above DN200 is about 7056 km, and the topological structure of the water supply pipe network is as Figure 2 shown.

[0092] Step S1.2. Preset the range of the number of zones as 30 - 100; set the node service head as 16 m; the minimum total water demand and the maximum total water demand of the zone are 47,400 m 3 / d and 157,900 m 3 / d.

[0093] Step S1.3. According to the actual pipe characteristics of the pipe network model, set the minimum recognized diameter of the main pipe as 800 mm; the ideal distribution pipe diameter as 200 mm.

[0094] Step S2. Identify the main pipelines: Based on the principle of separating water transmission and distribution and taking into account the connectivity between multiple water sources, identify the water transmission pipelines in the large water supply pipe network.

[0095] Step S2.1. The basic main pipelines include the diameter-based main pipelines and the flow-based main pipelines.

[0096] Step S2.2: Set the pipes with a pipe diameter greater than or equal to 800 mm as the basic main pipes of the pipe diameter, refer to Figure 3 (a) in

[0097] Step S2.3: Take the pipes with a flow rate higher than 421 m 3 / h (the flow rate when the economic flow velocity of a 500-mm pipe is 0.9 m / s) as the basic main pipes of the flow rate, refer to Figure 3 (b) in

[0098] Step S2.4: Take the union of the basic main pipes of the pipe diameter and the basic main pipes of the flow rate as the basic main pipes. The identification of the basic main pipes is shown in Figure 3 (c) in

[0099] Step S2.5: For a multi-source water supply pipe network, if the basic main pipes cannot connect all the water source nodes in series, analyze the connectivity between each node of the basic main pipes and the water source nodes, and use the Dijkstra algorithm to calculate the shortest paths from different nodes to the water source nodes to determine the supplementary main pipes that need to be added; among them, the shortest path is defined as the path with the minimum sum of weights of any connected nodes, and the calculation formula of the weight is:

[0100] w i = 10.67C i -1.852 D i -4.871 l i

[0101] In the formula, C i represents the Hazen-Williams coefficient of pipe section i; D i represents the pipe diameter of pipe section i, with the unit of m; l i represents the length of pipe section i, with the unit of m.

[0102] In the above content, the shortest path is the path calculated based on the pipe weight. Specifically, it refers to the path with the minimum sum of weights of all the pipes (or nodes) passed from the starting point to the ending point. Here, the "weight" is a numerical value calculated according to certain attributes of the pipe (such as the Hazen-Williams coefficient, pipe diameter, pipe section length, etc.), and is used to measure the performance or cost of the pipe during the water conveyance process.

[0103] Step S3: Construct a zoning scheme: After the main pipes are identified, separate the main pipes from the original pipe network so that they no longer undertake the water distribution function; find all the branch pipes connected to the main pipes along the main pipes to determine the candidate water distribution pipe set; randomly select a water distribution pipe from the candidate water distribution pipe set as the zoned water distribution pipe to form an initial scheme set of the selected water distribution pipes; according to the selected water distribution pipes and the preset zoned inlet positions, use the spatial clustering algorithm to group the pipe network nodes to form zoned boundaries, so as to construct multiple zoning schemes and realize the zoning of the nodes.

[0104] Step S3.1: According to the actual situation of the water supply network, set the ideal pipe diameter of the distribution pipe. The pipes with an actual diameter larger than the ideal pipe diameter of the distribution pipe are used as the ideal distribution pipes, and the main pipe is "stripped" from the original network so that it no longer undertakes the water distribution function. The schematic diagram of the main pipe after "stripping" is as shown in Figure 4 . Among them, Figure 4 in (a) is the topological schematic diagram of the network before the main pipe is stripped, Figure 4 in (b) is the topological schematic diagram of the network after the main pipe is stripped.

[0105] Step S3.2: If there are sufficient ideal distribution pipes in the area, only these pipes are used as candidate distribution pipes; otherwise, all the distribution pipes in the area form a candidate distribution pipe set.

[0106] Step S3.3: After the candidate distribution pipe set is completed, randomly select N dp distribution pipes from it as the distribution pipes for the sub-districts. Repeating this random process can obtain the initial scheme set of the distribution pipe selection. Since only one distribution pipe is used for water distribution in one sub-district, the following conditions need to be met:

[0107]

[0108] In the formula: N dp —— the number of distribution pipes in the network;

[0109] —— the minimum number of sub-districts;

[0110] —— the maximum number of sub-districts.

[0111] Step S3.4: After the distribution pipes are selected, for each sub-district scheme, spatial clustering is performed according to the distribution pipes and the positions of the sub-district entrances to determine the node grouping; in the present invention, the spatial clustering method is used instead of the shortest path method to cluster the nodes of the water supply network, so as to reduce the clustering time and improve the grouping efficiency.

[0112] Among them, during clustering, based on the entrance node, the adjacent nodes of the non-main pipes are clustered according to the topological structure of the water supply network. After clustering, the water demand of the adjacent nodes is added to the water demand of the corresponding entrance node.

[0113] During the clustering process, if the water demand of the entrance node exceeds the preset sub-district limit water demand, the clustering of this entrance node stops. Among them, the calculation formula of the sub-district limit water demand is as follows:

[0114]

[0115] In the formula: represents the maximum limit water demand of the sub-district; represents the water demand of node i at time t; T represents the total number of time steps for the extended simulation; N represents the total number of nodes in the pipe network; represents the minimum number of zones.

[0116] Step S3.5: After completing node clustering, all nodes within each zone are assigned to the corresponding inlet node, and the nodes associated with each inlet node form an independent zone. After node grouping, the pipe segments between different groups of nodes are the boundary pipe segments of the zones. These boundary pipe segments of the zones and the unselected water distribution pipes are cut off by valves to achieve independence between each zone.

[0117] Step S4: Construct a multi-objective function, including the standard deviation of average pressure, average node water age, pipe network operation energy consumption, and renovation cost.

[0118] To ensure high-quality water supply service for users after zoning, while taking into account the balance among background leakage in the pipe network, water quality safety, water supply energy consumption, and renovation economy, this application uses multiple indicators to evaluate the zoning scheme.

[0119] (1) Standard deviation of average pressure:

[0120] The pressure uniformity within a zone is characterized by the mean value of the standard deviation of node pressures within the zone for a 24-hour extended simulation. The smaller the mean value, the better the uniformity. Assuming there are a total of K zones after zoning, the ratio SDR of the average pressure standard deviation of each zone in the zoned pipe network to that of each zone before zoning is used as the objective function, as follows:

[0121]

[0122] In the formula: n k —— the total number of nodes in the k-th zone;

[0123] T—— the total number of simulation time steps during the extended simulation;

[0124] —— the pressure of node i at time t in the k-th zone, m;

[0125] —— the average pressure of all nodes in the k-th zone at time t, m;

[0126] —— the pressure of node i at time t in the k-th zone before zoning, m;

[0127] —— the average pressure of all nodes in the k-th zone at time t before zoning, m.

[0128] (2) Average node water age:

[0129] The water quality index uses the node average water age simulated for 24 hours in the pipe network delay to characterize the water quality of the pipe network. Assuming there are N nodes in the pipe network, the ratio of the node average water age before and after zoning, WAR, is used as the zoning water quality evaluation index:

[0130]

[0131] In the formula: ——The water age of node i at time t;

[0132] ——The water age of node i at time t before zoning.

[0133] (3) Pipe network operation energy consumption and renovation cost:

[0134] The renovation of the pipe network zoning requires an increase in construction costs and may change the operation energy consumption. To ensure the economy of the renovation and the long-term operation energy consumption after renovation, as much as possible, the cost of the newly added valves and monitoring equipment is small, and the internal operation energy consumption of the pipe network after zoning is small. The pipe network operation and renovation cost C is as follows:

[0135]

[0136] In the formula: K s ——The total number of SCADA devices to be installed;

[0137] K v ——The total number of valves to be installed;

[0138] ——The nth s monitoring equipment cost, and the equipment cost is related to the pipe diameter;

[0139] ——The nth v valve cost, and the valve cost is related to the pipe diameter;

[0140] y——The designed service life of the pipe network after renovation, in years;

[0141] m——The total number of pipes in the pipe network;

[0142] ρ——The density of water, kg / m 3 ;

[0143] g——The local acceleration of gravity, m 2 / s;

[0144] ——The flow rate of pipe l at time t after zoning, m 3 / s;

[0145] ——The head loss of pipe l at time t after zoning, m;

[0146] η —— Empirical coefficient, the average efficiency of all pumps in the pipe network during water supply, usually taken as 0.6 - 0.9.

[0147] Step S5, optimizing the zoning plan: In order to obtain the optimal zoning plan, the present invention takes the number of zones and the position of the distribution pipes as independent variables, and the standard deviation of the average pressure in the zones, the average water age of the nodes, the energy consumption of the pipe network operation, and the renovation cost as the objective functions, and uses the non-dominated sorting genetic algorithm (NSGA-II) to optimize and solve the zoning plan.

[0148] To synchronously optimize the multi-objective functions in the zoning process, the present invention uses the non-dominated sorting genetic algorithm (NSGA-II) to solve the zoning plan. This application uses the genetic algorithm package in the Platypus library in Python and combines it with the zoning algorithm to implement the multi-objective optimization zoning program.

[0149] In the program, the specific settings are as follows: the number of individuals is 50, the number of iterations is 100, the crossover probability is 0.8, the mutation probability is 0.2, and the rest of the parameters adopt the default values.

[0150] The NSGA-II algorithm maintains the diversity of the solution set through the concepts of non-dominated sorting and crowding distance. At the same time, it uses the selection, crossover, and mutation operation operators to form a new generation of solutions, uses non-dominated sorting to distinguish the advantages and disadvantages, and selects the solution set according to the crowding distance.

[0151] The iterative and optimization process of the algorithm includes the following steps:

[0152] Step S5.1, generating the initial population: First, randomly generate the first-generation population P0. The population consists of N individuals, and the chromosome of each individual corresponds to a distribution pipe screening plan. The distribution pipe screening plan is expressed by a binary number whose length is equal to the number of candidate zone inlets.

[0153] Step S5.2, distribution pipe screening plan: In the distribution pipe screening plan, "1" means that a certain distribution pipe is selected, and sensors need to be installed on this distribution pipe to monitor its flow rate and the pressure at the zone inlet; "0" means that a certain distribution pipe is not selected, and a valve needs to be installed on this distribution pipe to cut it off.

[0154] Step S5.3, evaluating and selecting the zoning plan: Perform a time-delay simulation on the zoning plan of each individual, calculate the objective function and evaluate the constraint conditions, sort the individuals based on the non-dominated sorting method, and select individuals according to the sorting results until the population upper limit. If it exceeds the population upper limit, some individuals will be excluded according to the crowding distance to form the next-generation parent population P1.

[0155] Step S5.4, Iteration Process: The parent population P1 undergoes selection, crossover, and mutation to obtain the next-generation offspring population Q1. The iteration process is repeated to obtain P2, and the process is continuously looped and iterated until the preset algorithm termination condition is met, resulting in the final population P last , population P last is the Pareto approximate optimal solution.

[0156] After optimization, the obtained Pareto front contains a total of 22 solutions. The range of the sum ratio of the standard deviation of pressure of these solutions is between 2.62 and 3.45, the range of the ratio of the average water age of nodes is between 0.68 and 0.98, the range of the converted cost of operation and renovation is between 87.226 million yuan / year and 92.948 million yuan / year, the number of zoning schemes is between 52 and 96, and the number of truncated pipe segments is between 1213 and 1684.

[0157] The 22 solutions are divided into 4 groups according to the operation and renovation costs of the pipe network and are respectively marked with 4 colors. As Figure 5 , Figure 6 shown, the solution sets within each group generally show an inverse proportional function trend.

[0158] In summary, a multi-source large-scale water supply pipe network multi-objective zoning optimization method provided by this embodiment can efficiently realize the automatic optimization zoning of large-scale pipe networks, avoid the complexity caused by series zoning, reduce the difficulty of pressure, flow balance analysis, and water quality monitoring and management in each zone, improve the technical management efficiency, enhance the water supply security, and show certain advantages in reducing the water age of the pipe network. At the same time, it effectively controls the renovation and long-term operation costs, and has a reference and guiding role for large-scale pipe network zoning renovation projects.

[0159] Embodiment 2

[0160] As Figure 7 shown, this embodiment provides a multi-source large-scale water supply pipe network multi-objective zoning optimization system for implementing the multi-source large-scale water supply pipe network multi-objective zoning optimization method described in Embodiment 1. The system includes:

[0161] A basic parameter module establishment unit for selecting a water supply pipe network and establishing a basic parameter module, which includes the maximum and minimum number of zones, node service head, minimum identified pipe diameter of the main pipe, minimum length of the branch pipe, and ideal pipe diameter for water distribution;

[0162] A main pipe identification unit for identifying the water conveyance pipelines in a large-scale water supply pipe network based on the principle of separating water transmission and distribution and taking into account the connectivity between multiple water sources. Specifically, it includes:

[0163] The basic main pipeline determination module is used to select pipelines with a diameter greater than a preset first threshold as the diameter-based main pipelines, select pipelines with a flow rate greater than a preset second threshold as the flow rate-based main pipelines, and take the union of the diameter-based main pipelines and the flow rate-based main pipelines as the basic main pipelines;

[0164] The supplementary main pipeline determination module, if the basic main pipelines cannot connect all water source nodes in series, analyzes the connectivity between each node of the basic main pipelines and the water source nodes, and uses the Dijkstra algorithm to calculate the shortest paths from different nodes of the basic main pipelines to the water source nodes to determine the supplementary main pipelines that need to be added; among them, the shortest path is defined as the path with the smallest sum of weights of any connected nodes, and the calculation formula of the weight is:

[0165] w i = 10.67C i -1.852 D i -4.871 l i

[0166] In the formula, C i represents the Hazen-Williams coefficient of pipe section i; D i represents the diameter of pipe section i, with the unit of m; l i represents the length of pipe section i, with the unit of m;

[0167] The main pipeline determination module is used to form the main pipelines from the set of the basic main pipelines and the supplementary main pipelines;

[0168] The partition scheme construction unit is used to construct the partition scheme after the main pipelines are identified, specifically including:

[0169] The main pipeline separation module is used to separate the main pipelines from the original pipe network so that they no longer undertake the water distribution function;

[0170] The candidate distribution pipe set determination module searches for all branch pipes connected to the main pipelines along the main pipelines to determine the candidate distribution pipe set;

[0171] The distribution pipe selection module randomly selects distribution pipes from the candidate distribution pipe set as the partition distribution pipes to form an initial scheme set of distribution pipe selection;

[0172] The node grouping and partition boundary formation module groups the pipe network nodes using a spatial clustering algorithm according to the selected distribution pipes and the preset partition entrance positions to form partition boundaries, thereby constructing multiple partition schemes to realize the partitioning of nodes;

[0173] The multi-objective function construction unit is used to construct a multi-objective function including the standard deviation of the average pressure in the partition, the average water age of the nodes, the energy consumption of the pipe network operation, and the renovation cost;

[0174] Optimization solution unit, according to the multi-objective function, uses the non-dominated sorting genetic algorithm NSGA-II to optimize and solve the partition scheme to obtain the optimal or approximately optimal partition scheme.

[0175] It can be understood that the various units described in this multi-source large-scale water supply pipe network multi-objective partition optimization system correspond to Figure 1 each step in the described multi-source large-scale water supply pipe network multi-objective partition optimization method. Therefore, the operations, features, and beneficial effects described above for the multi-source large-scale water supply pipe network multi-objective partition optimization method also apply to the multi-source large-scale water supply pipe network multi-objective partition optimization system and the modules included therein, and will not be elaborated here.

[0176] This embodiment also provides an electronic device, including: one or more processors, and a storage device, on which one or more programs are stored; the one or more programs are executed by the one or more processors, so that the one or more processors implement a multi-source large-scale water supply pipe network multi-objective partition optimization method as described in Embodiment 1.

[0177] This embodiment also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of a multi-source large-scale water supply pipe network multi-objective partition optimization method as described in Embodiment 1.

[0178] The specific embodiments of the present invention have been described in detail above, but they are only examples, and the present invention is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications and substitutions to the present invention are also within the scope of the present invention. Therefore, equivalent transformations and modifications made without departing from the spirit and scope of the present invention should all be covered within the scope of the present invention.

Claims

1. A multi-objective zoning optimization method for a large water supply network with multiple water sources, characterized in that: The steps include: S1, select the water supply network and establish a basic parameter module, which includes the maximum and minimum number of partitions, node service head, minimum identification diameter of the main pipe, minimum length of the branch pipe and ideal distribution pipe diameter; S2, based on the principle of separation of transmission and distribution, taking into account the connectivity between multiple water sources, identifies water pipelines in large water supply networks, including: Select the pipes with diameters greater than a preset first threshold as the diameter-based trunk pipes, select the pipes with flow rates greater than a preset second threshold as the flow-based trunk pipes, and take the union of the diameter-based trunk pipes and the flow-based trunk pipes as the basic trunk pipes; If the basic trunk pipe cannot connect all water source nodes in series, the connectivity between each node of the basic trunk pipe and the water source node is analyzed, and the Dijkstra algorithm is used to calculate the shortest path from different nodes of the basic trunk pipe to the water source node to determine the additional supplementary trunk pipe that needs to be added; the shortest path is defined as the path with the smallest sum of the weights of any connected nodes, and the weight calculation formula is: w i =10.67C i -1.852 D i -4.871 l i In the formula, C i represents the Hezen-Williams coefficient of pipe section i; D i Indicates the diameter of pipe section i, in m; l i It represents the length of pipe segment i, in m; The collection of basic trunk pipes and supplementary trunk pipes constitutes the trunk pipe; S3, after the trunk is identified, build a partitioning scheme, including: The trunk pipe is separated from the original pipe network so that it no longer has the function of water distribution; all branches connected to the trunk pipe are found along the trunk pipe to determine the candidate water distribution pipe set; water distribution pipes are randomly selected from the candidate water distribution pipe set as partition water distribution pipes to form an initial water distribution pipe selection scheme set; according to the selected water distribution pipes and the preset partition entrance positions, the network nodes are grouped using a spatial clustering algorithm to form partition boundaries, thereby constructing multiple partition schemes to achieve node partitioning; S4, construct a multi-objective function, which includes the standard deviation of the average pressure of the partition, the average water age of the node, the energy consumption of the pipe network operation and the renovation cost; S5, based on the multi-objective function, the non-dominated sorting genetic algorithm NSGA-II is used to optimize the partitioning scheme.

2. According to claim 1, a multi-objective zoning optimization method for a large-scale water supply network with multiple water sources is characterized in that: In step S1, the setting of the basic parameter module satisfies the following rules: The DMA partition size is set to 500 to 5,000 households; The node service head is determined according to local regulations; The minimum identified diameter of the main pipe shall not be less than 400mm to ensure reasonable water supply energy consumption. After the main pipe is identified, large-diameter pipes shall be selected as distribution pipes first to reduce energy consumption losses within the partition.

3. The multi-objective zoning optimization method for a large-scale water supply network with multiple water sources according to claim 1 is characterized in that: The step S3 specifically includes the following steps: S3.1, according to the actual situation of the water supply network, set the ideal water distribution pipe diameter, take the pipe with an actual diameter larger than the ideal water distribution pipe diameter as the ideal water distribution pipe, separate the main pipe from the original pipe network so that it no longer bears the water distribution function; S3.2, if there are sufficient ideal water distribution pipes in the area, only these pipes are selected as candidate water distribution pipes; otherwise, all water distribution pipes in the area are included in the candidate water distribution pipe set; S3.3, randomly select N from the candidate distribution pipe set dp The root water distribution pipe is used as the partition water distribution pipe, and the random process is repeated to generate the initial solution set for water distribution pipe selection; where N dp The following conditions must be met: Where: N dp Indicates the number of water distribution pipes selected in the pipe network; Indicates the preset minimum number of partitions; Indicates the preset maximum number of partitions; S3.4, after the water distribution pipes are selected, for each partitioning scheme, a spatial clustering algorithm is used to assign each node to different partitions according to the locations of the water distribution pipes and partition entrances to determine the node grouping; In the clustering process, the inlet node is used as the benchmark and the non-main pipe adjacent nodes are clustered according to the topological structure of the water supply network; after the clustering is completed, the water demand of the adjacent nodes is added to the water demand of the corresponding inlet node; If the water demand of the entry node exceeds the preset partition limit water demand, the entry node stops clustering, wherein the calculation formula of the partition limit water demand is as follows: Where: Indicates the maximum restricted water demand of the zone; represents the water demand of node i at time t; T represents the total number of delay simulation time; N represents the total number of nodes in the pipe network; Indicates the minimum number of partitions; S3.5, after completing the node clustering, all nodes in the partition are assigned to the corresponding entry nodes, and the nodes associated with each entry node constitute an independent partition, and the pipe segments between different groups of nodes form the partition boundary pipe segments; The zone boundary pipe sections and unselected water distribution pipes are cut off by valves to achieve independence between each zone.

4. The multi-objective zoning optimization method for a large-scale water supply network with multiple water sources according to claim 1 is characterized in that: In step S4, the standard deviation of the average pressure of each partition is used as the objective function, and the following steps are included: The pressure uniformity within the partition is characterized by the mean of the standard deviation of the node pressure within the partition by delay simulation for 24 hours. The smaller the mean, the better the uniformity. Assuming that there are K partitions after partitioning, the ratio of the average pressure standard deviation of each partition of the pipeline network after partitioning to the average pressure standard deviation of each partition before partitioning (SDR) is used as the objective function. The formula is as follows: Where: n k represents the total number of nodes in the kth partition; T represents the total number of simulation moments during the delay simulation; represents the pressure of node i in the kth partition at time t, in m; represents the average pressure of all nodes in the kth partition at time t, in m; It represents the pressure of node i in the kth partition at time t when there is no partition, in m; It represents the average pressure of all nodes in the kth partition at time t when there is no partition, in m.

5. The multi-objective zoning optimization method for a large-scale water supply network with multiple water sources according to claim 1 is characterized in that: In step S4, the average water age of the nodes is used as the objective function, and the following steps are included: The average water age of the nodes in the pipe network is simulated for 24 hours using a time-delay simulation to characterize the water quality of the pipe network. Assuming that there are N nodes in the pipe network, the ratio of the average water age of the nodes before and after the partition, WAR, is used as the water quality evaluation index for the partition. The formula is as follows: Where: represents the water age of node i at time t; Represents the water age of node i at time t when there is no partition.

6. The multi-objective zoning optimization method for a large-scale water supply network with multiple water sources according to claim 1 is characterized in that: In step S4, the energy consumption of the pipeline network and the renovation cost are used as the objective function, and the following steps are included: In order to ensure the economy of energy consumption during the transformation and long-term operation after the transformation, the number and cost of newly added valves and monitoring equipment are relatively small, and the internal operation energy consumption of the pipe network after zoning is relatively small, with the goal of minimizing the pipe network operation and transformation cost C. The formula is as follows: In the formula, C is the cost of pipe network operation and transformation, K is s Indicates the total number of SCADA devices to be installed; K v Indicates the total number of valves to be installed; Indicates the nth s The cost of a monitoring device, which is related to the diameter of the pipeline; Indicates the nth v The cost of each valve is related to the pipe diameter; y represents the design service life of the pipe network after transformation, in years; m represents the total number of pipes in the pipe network; ρ represents the density of water, in kg / m 3 ; g represents the local gravitational acceleration, in m 2 / s; Represents the flow rate of pipeline l at time t after partitioning, in m 3 / s; It represents the head loss of pipeline l at time t after partitioning, in m; η represents the empirical coefficient, which is the average efficiency of all pumps in the pipe network during water supply, and its value range is 0.6 to 0.

9.

7. The multi-objective zoning optimization method for a large-scale water supply network with multiple water sources according to claim 1 is characterized in that: The step S5 comprises the following steps: S5.1, randomly generate the first generation population P0, which consists of N individuals. The chromosome of each individual corresponds to a water distribution pipe screening scheme, which is expressed by a string of binary numbers whose length is equal to the number of candidate partition entrances; S5.2, in the water distribution pipe screening scheme, "1" means that a water distribution pipe is selected, and a sensor must be installed on the water distribution pipe to monitor its flow rate and zone inlet pressure; "0" means that a water distribution pipe is not selected, and a valve must be installed on the water distribution pipe to cut off the water distribution pipe; S5.3, perform a delayed simulation on the partitioning scheme of each individual, calculate the objective function and evaluate the constraints, sort the individuals based on the non-dominated sorting method, and select individuals according to the sorting results until the population upper limit; If the population limit is exceeded, some individuals will be excluded according to the crowding distance to form the next generation parent population P1; S5.4, the parent population P1 obtains the next generation population Q1 through selection, crossover and mutation, repeats the iterative process to obtain P2, and iterates continuously until the preset algorithm termination condition is met to obtain the final generation population P last , population P last This is the Pareto approximate optimal solution.

8. A system for executing the multi-objective zoning optimization method for a large-scale water supply network with multiple water sources as claimed in any one of claims 1 to 7, characterized in that: include: The basic parameter module establishment unit is used to select the water supply network and establish the basic parameter module, which includes the maximum and minimum number of partitions, node service head, minimum identification diameter of the main pipe, minimum length of the branch pipe and ideal water distribution pipe diameter; The trunk pipe identification unit is based on the principle of separation of transmission and distribution, taking into account the connectivity between multiple water sources, and identifies the water pipelines in large water supply networks. Specifically, it includes: A basic trunk pipe determination module is used to select a pipe with a diameter greater than a preset first threshold as a diameter basic trunk pipe, select a pipe with a flow rate greater than a preset second threshold as a flow basic trunk pipe, and take the union of the diameter basic trunk pipe and the flow basic trunk pipe as the basic trunk pipe; In the supplementary trunk pipe determination module, if the basic trunk pipe cannot connect all water source nodes in series, the connectivity between each node of the basic trunk pipe and the water source node is analyzed, and the Dijkstra algorithm is used to calculate the shortest path from different nodes of the basic trunk pipe to the water source node to determine the supplementary trunk pipe that needs to be added; the shortest path is defined as the path with the smallest sum of weights of any connected nodes, and the weight calculation formula is: w i =10.67C i -1.852 D i -4.871 l i In the formula, C i represents the Hezen-Williams coefficient of pipe section i; D i Indicates the diameter of pipe section i, in m; l i It represents the length of pipe segment i, in m; A trunk pipe determination module, used for forming a trunk pipe from a collection of basic trunk pipes and supplementary trunk pipes; The partition scheme construction unit is used to construct the partition scheme after the trunk pipe identification is completed, which specifically includes: The main pipe separation module is used to separate the main pipe from the original pipe network so that it no longer bears the water distribution function; The candidate water distribution pipe set determination module searches for all branch pipes connected to the main pipe along the main pipe to determine the candidate water distribution pipe set; The water distribution pipe selection module randomly selects water distribution pipes from the candidate water distribution pipe set as the zone water distribution pipes to form an initial water distribution pipe selection scheme set; The node grouping and partition boundary formation module uses a spatial clustering algorithm to group the network nodes according to the selected water distribution pipes and the preset partition entrance locations to form partition boundaries, thereby constructing multiple partition schemes to achieve node partitioning; Multi-objective function construction unit, used to construct multi-objective functions including the standard deviation of average pressure in different zones, average water age of nodes, energy consumption of pipe network operation and renovation cost; The optimization solving unit uses a non-dominated sorting genetic algorithm NSGA-II to optimize the partitioning scheme according to the multi-objective function to obtain the optimal or approximately optimal partitioning scheme.

9. An electronic device, characterized in that: include: one or more processors; a storage device having one or more programs stored thereon; The one or more programs are executed by the one or more processors, so that the one or more processors implement a multi-objective zoning optimization method for a large-scale water supply network with multiple water sources as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the multi-objective zoning optimization method for a large-scale water supply network with multiple water sources are implemented as described in any one of claims 1-7.

Images