CN113807012A - Water supply network division method based on connection strengthening - Google Patents

Water supply network division method based on connection strengthening Download PDF

Info

Publication number
CN113807012A
CN113807012A CN202111074419.8A CN202111074419A CN113807012A CN 113807012 A CN113807012 A CN 113807012A CN 202111074419 A CN202111074419 A CN 202111074419A CN 113807012 A CN113807012 A CN 113807012A
Authority
CN
China
Prior art keywords
node
water supply
nodes
supply network
feature vector
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111074419.8A
Other languages
Chinese (zh)
Inventor
戎科臻
荣泽坤
付明磊
王海英
郑乐进
郑剑锋
吴德
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Laison Technology Co ltd
Original Assignee
Hangzhou Laison Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hangzhou Laison Technology Co ltd filed Critical Hangzhou Laison Technology Co ltd
Priority to CN202111074419.8A priority Critical patent/CN113807012A/en
Publication of CN113807012A publication Critical patent/CN113807012A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/08—Fluids
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/14—Pipes

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Geometry (AREA)
  • Computer Hardware Design (AREA)
  • Artificial Intelligence (AREA)
  • Water Supply & Treatment (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Primary Health Care (AREA)
  • Marketing (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Human Resources & Organizations (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Tourism & Hospitality (AREA)
  • Public Health (AREA)
  • Algebra (AREA)
  • Computing Systems (AREA)
  • Fluid Mechanics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

A water supply network division method based on connection strengthening comprises the following steps: step 1: for a water supply network containing n nodes, collecting and standardizing feature vectors H of all nodes to obtain an initial feature vector set H of the nodes0(ii) a Step 2: calculating the feature vector of any two nodes
Figure DDA0003261597180000011
And
Figure DDA0003261597180000012
connection strength e ofijObtaining a node connection strength relation set e; and step 3: updating a node feature vector set H according to the node connection strength e0To obtain H1(ii) a And 4, step 4: vector set H according to node characteristics1Classifying the nodes; and 5: DMA (direct memory Access) for dividing water supply network into m independent metering areas according to node classification results1,DMA2,…,DMAm. The invention realizes the integrated independent metering subarea of the water supply network based on the pipe network connection strength, thereby improving the rationality and reliability of the independent metering subarea of the water supply network.

Description

Water supply network division method based on connection strengthening
Technical Field
The invention relates to the field of municipal engineering and urban water supply networks, in particular to a water supply network division method based on connection strengthening.
Background
The water supply network is used for transporting water resources in cities, so as to provide residential water, commercial water, industrial water and the like, however, the water supply network generally has serious leakage due to the problems of long-term overhaul, pressure imbalance, improper management and the like. The water supply network is divided into a plurality of areas, and the flow meters are arranged at the boundaries of the areas to realize independent metering of each area, so that great management convenience is provided for timely discovery and control of leakage of the water supply network. However, the water supply network is complex and diverse in indexes, and manual division is likely to cause unreasonable partitioning, so that leakage discovery and control capacity is affected, and how to reasonably divide the water supply network becomes one of the key problems to be solved urgently in the industry.
In the prior art, a method related to water supply network division is limited, a Chinese invention patent CN201811181755.0 discloses an independent metering partition water quantity component analysis method in an urban water supply system, a Chinese invention patent CN201911300313.8 discloses a water supply network DMA automatic partition method based on graph division, a Chinese invention patent CN201810864065.9 discloses a water supply network auxiliary DMA partition method and system based on graph theory, a CN202110525659.9 discloses a water supply network independent metering partition method based on graph convolution, and a CN202110168579.2 discloses a water supply network independent metering partition method. However, the above method has two problems: firstly, the connection relation between nodes is not considered in the partitioning process based on the characteristics of the nodes of the water supply network, so that the rationality of the partitioning process is reduced; second, the method of independent metering zoning for a water supply network taking node characteristics into account in isolation reduces zoning reliability.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a water supply network partitioning method based on connection strengthening, which aims to solve the problem of partition rationality and reliability loss caused by independent water supply network metering and partitioning based on node characteristics in the prior art background, and realize integrated independent water supply network metering and partitioning based on pipe network connection strength, thereby improving the rationality and reliability of independent water supply network metering and partitioning.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a water supply network division method based on connection strengthening comprises the following steps:
step 1: for a water supply network containing n nodes, collecting and standardizing feature vectors h of all nodes to obtain an initial feature vector set of the nodes
Figure BDA0003261597160000021
Step 2: calculating the feature vector of any two nodes
Figure BDA0003261597160000022
And
Figure BDA0003261597160000023
connection strength e ofijObtaining a node connection strength relation set e;
and step 3: updating a node feature vector set H according to the node connection strength e0To obtain
Figure BDA0003261597160000024
And 4, step 4: according to node characteristicsVector set H1Classifying the nodes;
and 5: DMA (direct memory Access) for dividing water supply network into m independent metering areas according to node classification results1,DMA2,…,DMAm。
Further, in the step 2, for any two nodes i and j in the water supply network, a node feature vector is obtained
Figure BDA0003261597160000025
And
Figure BDA0003261597160000026
establishing an initial random parameter matrix W and a linear mapping relation a, and calculating the connection strength of the nodes i and j through an activation function ReLU
Figure BDA0003261597160000027
Still further, in the step 3, for any node i in the water supply network, a node group N adjacent to the node i through a pipeline is countediTo the feature vector of node i
Figure BDA0003261597160000028
Is updated to obtain
Figure BDA0003261597160000029
All the nodes are updated in sequence, and then an updated node feature vector set is obtained
Figure BDA00032615971600000210
Further, in the step 4, a set H of vectors is obtained according to the node feature1Classifying the nodes, specifically comprising the following steps:
step 4.1: establishing a multilayer forward propagation neural network, wherein the input layer of the network is a node feature vector set H1The middle layer is a random parameter set Q to be trained, and the output layer is a node classification result;
step 4.2: obtaining the node classification result under the current parameter matrix W and the network parameter set Q andaccording to each node i and N adjacent to each node iiA node j1,j2,…,
Figure BDA0003261597160000031
Characteristic difference of (2) calculating merit function
Figure BDA0003261597160000032
Step 4.3: calculating partial derivatives of the evaluation function J about the parameter matrix W and the network parameter set Q, and performing gradient descent optimization on the parameter matrix W and the network parameter set Q according to a chain derivative rule;
step 4.4: and (4.2) repeating the step 4.2 and the step 4.3 until the evaluation function J converges or an expected result is obtained, and outputting the node classification result at the moment.
The invention has the following beneficial effects: and calculating the connection strength between the nodes according to the network node characteristic correlation, updating the node characteristics according to the connection strength between the nodes, and dividing the water supply network according to the updated node characteristics. The method for enhancing the relevance of the water supply network nodes by utilizing connection strengthening has great reference value for water supply network division.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
fig. 2 is a schematic diagram illustrating a network connection strengthening process according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a partition evaluation score optimization process according to an embodiment of the present invention;
fig. 4 is a schematic diagram illustrating a partitioning result of a water supply network according to an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
Referring to fig. 1 to 4, a method for dividing a water supply network based on connection reinforcement includes the following steps:
step 1: establishing an initial feature vector set of the water supply network node;
for the A city water supply network, the network comprises 388 water demand nodes, and the longitude and the latitude of each node are collected,Forming a feature vector h by using the characteristics of dimension, altitude, water consumption and the like, and standardizing to obtain an initial feature vector set of nodes
Figure BDA0003261597160000033
Figure BDA0003261597160000034
Table 1 is a standardized initial characteristic table of the water supply network node in city a.
Figure BDA0003261597160000035
Figure BDA0003261597160000041
TABLE 1
Step 2: calculating a node connection strength relation set;
for any two nodes i and j in the water supply network in the city A, obtaining the characteristic vector of the node
Figure BDA0003261597160000042
And
Figure BDA0003261597160000043
establishing an initial random parameter matrix W and a linear mapping relation a, and calculating the connection strength of the nodes i and j through an activation function ReLU
Figure BDA0003261597160000044
The connection strengthening process is shown in fig. 2, and the connection strength of the water supply network node in city a is obtained and is shown in table 2. Table 2 is a table of connection strengths of water supply network nodes in city a.
Figure BDA0003261597160000045
TABLE 2
And step 3: updating a water supply network node characteristic set;
for water supply net of city AAny node i in the network counts a node group N adjacent to the node i through a pipelineiTo the feature vector of node i
Figure BDA0003261597160000046
Is updated to obtain
Figure BDA0003261597160000047
All nodes are updated, and then an updated node feature vector set is obtained
Figure BDA0003261597160000048
As shown in table 3. And table 3 is a characteristic table of the updated water supply network node in city a.
Figure BDA0003261597160000049
TABLE 3
And 4, step 4: training to realize water supply network node classification;
vector set H according to node characteristics1Classifying the water supply network nodes in the city A, which specifically comprises the following steps:
step 4.1: establishing a multilayer forward propagation neural network, wherein the network input layer is a characteristic vector set H of the water supply network nodes in the city A1The middle layer is a random parameter set Q to be trained, and the output layer is a classification result of the water supply network nodes in the city A;
step 4.2: obtaining the node classification results under the current parameter matrix W and the network parameter set Q, and according to each node i and the adjacent N thereofiA node j1,j2,…,
Figure BDA0003261597160000051
Calculating the evaluation score of the initial subarea of the water supply network in the market A according to the characteristic difference
Figure BDA0003261597160000052
Step 4.3: calculating partial derivatives of the evaluation function J about the parameter matrix W and the network parameter set Q, and performing gradient descent optimization on the parameter matrix W and the network parameter set Q according to a chain derivative rule, wherein the initial partition evaluation score optimization process of the urban A water supply network is shown in FIG. 3;
step 4.4: and repeating the step 4.2 and the step 4.3 until the evaluation function J converges or obtains an expected result, wherein the parameters are optimized for 41 times in total, and the node classification result at the moment is output and is shown in the table 4. And table 4 is a classification final result table of the water supply network nodes in city a.
Figure BDA0003261597160000053
TABLE 4
And 5: outputting a water supply network partition result;
DMA (direct memory Access) for dividing water supply network into 5 independent metering areas according to node classification results1,DMA2,…,DMA5As shown in fig. 4.
The embodiments described in this specification are merely illustrative of implementations of the inventive concepts, which are intended for purposes of illustration only. The scope of the present invention should not be construed as being limited to the particular forms set forth in the examples, but rather as being defined by the claims and the equivalents thereof which can occur to those skilled in the art upon consideration of the present inventive concept.

Claims (4)

1. A water supply network division method based on connection strengthening is characterized by comprising the following steps:
step 1: for a water supply network containing n nodes, collecting and standardizing feature vectors h of all nodes to obtain an initial feature vector set of the nodes
Figure FDA0003261597150000011
Step 2: calculating the feature vector of any two nodes
Figure FDA0003261597150000012
And
Figure FDA0003261597150000013
connection strength e ofijObtaining a node connection strength relation set e;
and step 3: updating a node feature vector set H according to the node connection strength e0To obtain
Figure FDA0003261597150000014
And 4, step 4: vector set H according to node characteristics1Classifying the nodes;
and 5: DMA (direct memory Access) for dividing water supply network into m independent metering areas according to node classification results1,DMA2,…,DMAm。
2. The method as claimed in claim 1, wherein in step 2, for any two nodes i and j in the water supply network, a node feature vector is obtained
Figure FDA0003261597150000015
And
Figure FDA0003261597150000016
establishing an initial random parameter matrix W and a linear mapping relation a, and calculating the connection strength of the nodes i and j through an activation function ReLU
Figure FDA0003261597150000017
3. The method for dividing water supply network based on connection strengthening as claimed in claim 1 or 2, wherein in the step 3, for any node i in the water supply network, the node group N adjacent to the node i through the pipeline is countediTo the feature vector of node i
Figure FDA0003261597150000018
Is updated to obtain
Figure FDA0003261597150000019
All the nodes are updated in sequence, and then an updated node feature vector set is obtained
Figure FDA00032615971500000110
4. The method as claimed in claim 1 or 2, wherein in the step 4, the vector set H is determined according to the node feature vectors1Classifying the nodes, specifically comprising the following steps:
step 4.1: establishing a multilayer forward propagation neural network, wherein the input layer of the network is a node feature vector set H1The middle layer is a random parameter set Q to be trained, and the output layer is a node classification result;
step 4.2: obtaining node classification results under the current parameter matrix W and the network parameter set Q and according to each node i and the adjacent N thereofiA node
Figure FDA00032615971500000111
Characteristic difference of (2) calculating merit function
Figure FDA0003261597150000021
Step 4.3: calculating partial derivatives of the evaluation function J about the parameter matrix W and the network parameter set Q, and performing gradient descent optimization on the parameter matrix W and the network parameter set Q according to a chain derivative rule;
step 4.4: and (4.2) repeating the step 4.2 and the step 4.3 until the evaluation function J converges or an expected result is obtained, and outputting the node classification result at the moment.
CN202111074419.8A 2021-09-14 2021-09-14 Water supply network division method based on connection strengthening Pending CN113807012A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111074419.8A CN113807012A (en) 2021-09-14 2021-09-14 Water supply network division method based on connection strengthening

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111074419.8A CN113807012A (en) 2021-09-14 2021-09-14 Water supply network division method based on connection strengthening

Publications (1)

Publication Number Publication Date
CN113807012A true CN113807012A (en) 2021-12-17

Family

ID=78895240

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111074419.8A Pending CN113807012A (en) 2021-09-14 2021-09-14 Water supply network division method based on connection strengthening

Country Status (1)

Country Link
CN (1) CN113807012A (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101806396A (en) * 2010-04-24 2010-08-18 上海交通大学 Method for generating pressure distribution chart of urban water supply pipeline network
US20170032246A1 (en) * 2015-07-31 2017-02-02 Canon Kabushiki Kaisha Method for training an artificial neural network
CN109871976A (en) * 2018-12-20 2019-06-11 浙江工业大学 A kind of prediction technique of power quality containing distributed power distribution network based on cluster and neural network
CN111931903A (en) * 2020-07-09 2020-11-13 北京邮电大学 Network alignment method based on double-layer graph attention neural network
CN112085124A (en) * 2020-09-27 2020-12-15 西安交通大学 Complex network node classification method based on graph attention network
CN112883663A (en) * 2021-02-07 2021-06-01 浙江工业大学 Independent metering and zoning method for water supply pipe network
CN112966706A (en) * 2020-12-11 2021-06-15 天津大学 Multilayer network representation learning method based on graph attention machine mechanism
CN113284006A (en) * 2021-05-14 2021-08-20 杭州莱宸科技有限公司 Independent water supply network metering and partitioning method based on graph convolution

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101806396A (en) * 2010-04-24 2010-08-18 上海交通大学 Method for generating pressure distribution chart of urban water supply pipeline network
US20170032246A1 (en) * 2015-07-31 2017-02-02 Canon Kabushiki Kaisha Method for training an artificial neural network
CN109871976A (en) * 2018-12-20 2019-06-11 浙江工业大学 A kind of prediction technique of power quality containing distributed power distribution network based on cluster and neural network
CN111931903A (en) * 2020-07-09 2020-11-13 北京邮电大学 Network alignment method based on double-layer graph attention neural network
CN112085124A (en) * 2020-09-27 2020-12-15 西安交通大学 Complex network node classification method based on graph attention network
CN112966706A (en) * 2020-12-11 2021-06-15 天津大学 Multilayer network representation learning method based on graph attention machine mechanism
CN112883663A (en) * 2021-02-07 2021-06-01 浙江工业大学 Independent metering and zoning method for water supply pipe network
CN113284006A (en) * 2021-05-14 2021-08-20 杭州莱宸科技有限公司 Independent water supply network metering and partitioning method based on graph convolution

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
刘俊;周鹏;: "谱聚类在给水管网分区优化中的应用", 土木建筑与环境工程, no. 06 *
周立典;信昆仑;黄慰忠;薛骁;: "供水管网DMA分区优化方法及软件实现", 给水排水, no. 02 *
王彤;冯雪峰;赵明;刘翔翔;: "基于快速网络社区算法的供水管网区块化研究", 中国给水排水, no. 05 *
黄盼盼;赵明;王少坡;张清周;邱春生;丛建松;: "基于水力模型的供水管网区块化优化研究", 水电能源科学, no. 05 *

Similar Documents

Publication Publication Date Title
CN104183119B (en) Based on the anti-arithmetic for real-time traffic flow distribution forecasting method pushed away of section OD
CN101232180A (en) Power distribution system load obscurity model building device and method
CN103606266B (en) Based on the road grid traffic improving countermeasure efficiency evaluation method of DEA
CN109214549B (en) Water supply pipe network auxiliary DMA partition method and system based on graph theory
CN105976048A (en) Power transmission network extension planning method based on improved artificial bee colony algorithm
CN105761192A (en) Intelligent method and intelligent integrated system for village-town area land use planning
CN102571954A (en) Complex network clustering method based on key influence of nodes
CN103971184A (en) Power transmission line path generation method based on spatial GIS (Geographic Information System)
Tan New-town policy and development in China
CN103428804B (en) Find mapping scheme and network code location method between network-on-chip task and node
SG194531A1 (en) A method for the construction of a water distribution model
CN103810260A (en) Complex network community discovery method based on topological characteristics
CN108984972A (en) Water quality mathematical model optimization method based on big data and parallel computation
CN115455621A (en) Heat supply system topological structure optimization method based on bionic principle
CN107391856A (en) Optimize the method for water meter and valve location in a kind of water supply network subregion
CN113807012A (en) Water supply network division method based on connection strengthening
CN117235950B (en) Natural gas pipe network steady-state simulation method, medium and equipment based on Newton iteration method
CN109033603B (en) Intelligent substation secondary system simulation method based on source flow path chain
CN106681795A (en) Virtual network mapping method utilizing local topological attributes of nodes and available resource capacity values
Wing On the issue of plan shape complexity: plan shape indices revisited
CN115860171A (en) Building time-by-time power consumption prediction method based on dynamic and static mixed data analysis
CN113806925B (en) OD matrix estimation method based on Bayesian hierarchical optimization
CN109902846A (en) A kind of simplified method and system of water supply network
Van Zyl Proposed guideline for modelling water demand by suburb
WO2022052146A1 (en) Heavy overload check method for load transfer decision of open-loop power grid

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination