CN118133587A - Monitoring point optimal arrangement method and device based on drainage pipe network partition dimension reduction - Google Patents
Monitoring point optimal arrangement method and device based on drainage pipe network partition dimension reduction Download PDFInfo
- Publication number
- CN118133587A CN118133587A CN202410559620.2A CN202410559620A CN118133587A CN 118133587 A CN118133587 A CN 118133587A CN 202410559620 A CN202410559620 A CN 202410559620A CN 118133587 A CN118133587 A CN 118133587A
- Authority
- CN
- China
- Prior art keywords
- hydraulic
- data
- partition
- pipe network
- optimal
- 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
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 172
- 238000005192 partition Methods 0.000 title claims abstract description 120
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000009467 reduction Effects 0.000 title claims abstract description 32
- 238000004088 simulation Methods 0.000 claims abstract description 88
- 239000011159 matrix material Substances 0.000 claims abstract description 66
- 238000001514 detection method Methods 0.000 claims abstract description 23
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 21
- 238000000638 solvent extraction Methods 0.000 claims abstract description 21
- 239000007788 liquid Substances 0.000 claims description 35
- 238000012512 characterization method Methods 0.000 claims description 19
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 19
- 230000004044 response Effects 0.000 claims description 18
- 238000012795 verification Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 9
- 230000003993 interaction Effects 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 238000010206 sensitivity analysis Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 238000010606 normalization Methods 0.000 description 4
- 230000035945 sensitivity Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004836 empirical method Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000012530 fluid Substances 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a monitoring point optimizing arrangement method and device based on drainage pipe network partition dimension reduction, comprising the following steps: collecting pipe network data and actual monitoring data corresponding to a drainage pipe network from a geographic information system; constructing an SWMM hydraulic model based on pipe network data, calibrating by using actual monitoring data, and performing hydraulic simulation on rainfall data by using the calibrated SWMM hydraulic model to obtain hydraulic simulation data; partitioning the drainage pipe network according to the hydraulic simulation data and by adopting a community detection algorithm to obtain a partitioning result; based on hydraulic simulation data and partition results, performing node dimension reduction on nodes in each partition by using non-negative matrix factorization, and extracting key nodes which can most reflect hydraulic changes in the partition according to the total contribution of the comprehensive factors to form a monitoring point optimal arrangement scheme; and outputting the final optimal arrangement scheme of the monitoring points when the optimal arrangement scheme of the monitoring points is verified to be reasonable, so that the cost is reduced, and the quality and coverage of the monitoring data can be maintained and even improved.
Description
Technical Field
The invention belongs to the technical field of optimal arrangement of monitoring points of a drainage pipe network, and particularly relates to an optimal arrangement method and device of monitoring points based on drainage pipe network partition dimension reduction.
Background
In urban infrastructure management, efficient operation of the drainage network is critical. With the rise of digital management modes, dynamic monitoring and real-time management of urban drainage systems by using a large number of monitoring instrument devices have become a trend. However, in this mode, the current monitoring point arrangement suffers from the following major drawbacks:
The first and monitoring points are too many. In early stages, monitoring equipment was typically installed at many nodes of the network in order to fully understand the water quality and water volume laws. This results in a large number of devices, increasing costs and complexity of maintenance.
Second, efficiency and cost are difficult to trade off. Although the data coverage rate can be improved by adding the monitoring points, the equipment and the operation cost are greatly improved. In addition, the collection and processing of large amounts of data also builds pressure on management systems.
Third, non-optimization of the monitoring point arrangement. Traditional monitoring point arrangement is often not based on an empirical method, is not optimized, so that monitoring resource allocation may be uneven, and the potential of a monitoring network cannot be fully exerted.
Over time and data accumulate, the data of the drainage network is gradually accumulated, which provides the possibility to optimize the monitoring point arrangement using statistical and operational theories. The aim is to comprehensively reflect the information of the drainage pipe network through as few monitoring points as possible, so that the cost is obviously reduced while the monitoring efficiency is maintained. At present, optimization methods of monitoring points are mainly divided into two types: based on statistical theory and based on operational theory. Although there have been many studies on optimizing monitoring points of water supply systems, the study on water drainage systems in this respect is relatively few and the technical methodology is still imperfect. The method shows that the optimization of the monitoring points of the drainage system is a problem to be solved, and has important practical application value.
The patent application with publication number CN113221440A discloses a drainage system monitoring point optimizing arrangement and real-time global inversion method, which screens an optimal monitoring point set by carrying out principal component analysis on simulation result data of a 2D hydraulic model, and then the constructed rainwater well liquid level inversion monitoring model inverts real-time monitoring data of the optimal monitoring point set to obtain the liquid level of the rainwater well. However, in the technical scheme, no partition is performed when the optimal monitoring points are determined, global optimization is performed based on the whole drainage system, so that more optimal monitoring points are caused in some areas, no monitoring points are caused in other areas, the distribution of the whole optimal monitoring points is unbalanced, and a monitoring blind area exists. In areas without monitoring points, the data will indeed lead to an inability to effectively monitor the flow or fluid level conditions in the area, making it difficult to find potential problems and take action in an emergency situation. Meanwhile, the distribution of monitoring points is unbalanced, so that data acquisition nonuniformity can be caused, understanding of the whole system behavior and accuracy of a model are affected, and optimization of the system and formulation of a management scheme are affected.
Disclosure of Invention
In view of the above, the invention aims to provide the monitoring point optimizing arrangement method and the device based on the drainage pipe network partition dimension reduction, which not only can effectively reduce the number of required monitoring points and reduce the cost, but also can maintain and even improve the quality and coverage of monitoring data.
In order to achieve the above object, an embodiment of the present invention provides a method for optimizing arrangement of monitoring points based on partition dimension reduction of a drainage pipe network, including the following steps:
Collecting pipe network data corresponding to a drainage pipe network from a geographic information system and acquiring actual monitoring data of the drainage pipe network, wherein the actual monitoring data comprise flow data and liquid level data of each pump station in the pipe network, flow data and liquid level data of actual monitoring points and rainfall data;
Constructing an SWMM hydraulic model based on pipe network data, calibrating by using actual monitoring data, and performing hydraulic simulation on rainfall data by using the calibrated SWMM hydraulic model to obtain hydraulic simulation data, wherein the hydraulic simulation data comprises flow data and liquid level data;
Partitioning the drainage pipe network according to the hydraulic simulation data and by adopting a community detection algorithm to obtain a partitioning result;
Node dimension reduction is carried out on nodes in each partition by using non-negative matrix factorization based on hydraulic simulation data and partition results, key nodes which can most reflect hydraulic changes in the partition are extracted according to the total contribution of comprehensive factors to serve as optimal monitoring points, and all the optimal monitoring points form a monitoring point optimal arrangement scheme;
And verifying whether the optimal arrangement scheme of the monitoring points is reasonable or not, and outputting the final optimal arrangement scheme of the monitoring points passing verification.
Preferably, constructing and calibrating the SWMM hydraulic model based on the pipe network data includes:
model construction and initialization: constructing an SWMM hydraulic model according to pipe network data, and initializing parameters of the SWMM hydraulic model;
Iterative calibration model parameters: the SWMM hydraulic model with current parameters is utilized for simulation to obtain hydraulic simulation data, the hydraulic simulation data is compared with actual monitoring data, nash efficiency coefficients are calculated to evaluate model performance, model parameters are adjusted one by one according to the Nash efficiency coefficients, sensitivity analysis is carried out to determine final model parameters to be adjusted, and the final model parameters are adjusted;
Model output: model calibration was considered successful when the Nash efficiency coefficient reached the standard and a calibrated SWMM hydraulic model was output.
Preferably, the hydraulic simulation of rainfall data using a calibrated SWMM hydraulic model includes:
and inputting rainfall data with different intensities into a calibrated SWMM hydraulic model, and performing multiple simulation to obtain hydraulic simulation data, wherein the selected rainfall data comprises rainfall events with different intensities such as light rain, medium rain, heavy rain and the like.
Preferably, partitioning the drainage pipe network according to the hydraulic simulation data and by adopting a community detection algorithm comprises the following steps:
Carrying out standardized processing on the hydraulic simulation data, constructing a network model based on the topological structure of the drainage pipe network, wherein nodes in the network model represent nodes of the drainage pipe network, edges represent pipelines, represent water flow paths among the nodes, and constructing a weighted adjacency matrix of the network model according to the standardized processed hydraulic simulation data;
And carrying out community detection partition on the drainage network represented by the network model by adopting a Leiden algorithm, calculating modularity according to a weighted adjacent matrix in the Leiden algorithm, analyzing hydraulic consistency in the partition and hydraulic independence between the partitions, and outputting a partition result when judging that the hydraulic response of the partition is reasonable according to the hydraulic consistency and the hydraulic independence.
Preferably, the weighted adjacency matrix for constructing the network model from the normalized hydraulic simulation data is denoted as W,Each element/>The weight of the edge between the nodes i and j is represented, the weight is the combination of the node flow and the liquid level, and the calculation formula is as follows:
；
Wherein the method comprises the steps of Is a parameter between 0 and 1 for adjusting the relative importance of flow data and level data in the weights,/>Is the standardized node flow data,/>Is the node liquid level data after standardized treatment;
the formula used for calculating modularity according to the weighted adjacency matrix is:
；
Wherein, Representing modularity of the network,/>And/>The intensities of nodes i and j, respectively, i.e., the sum of the weights of all edges connected to these nodes,/>Is the sum of the weights of all edges in the network,/>And/>Representing the community in which nodes i and j are located,Is an indication function, when nodes i and j belong to the same community, the value is 1, otherwise, the value is 0,/>Is a resolution parameter used to control the granularity of the network partition.
Preferably, the parameters in the weighted adjacency matrix calculation formula are adjusted when the hydraulic response of the partition is judged to be unreasonable according to the hydraulic consistency and the hydraulic independenceAnd resolution parameters/>, in a module calculation formulaAnd then, the community detection partition is carried out again until the hydraulic response of the partition is reasonable.
Preferably, the hydraulic consistency is judged by taking at least one of the standard deviation of the flow and the liquid level of each node in the partition as a hydraulic consistency characterization value, wherein the smaller the hydraulic consistency characterization value is, the higher the hydraulic consistency in the partition is;
the hydraulic independence represents interaction among the subareas by taking the ratio of the boundary flow to the total subarea flow as a hydraulic independence characterization value, wherein the smaller the hydraulic independence characterization value is, the smaller the water flow interaction among the subareas is, and the higher the hydraulic independence is;
And when the hydraulic consistency characterization value and the hydraulic independence characterization value are smaller than the respective corresponding threshold values, the hydraulic response is reasonable, otherwise, the hydraulic response is considered to be unreasonable.
Preferably, non-negative matrix factorization is used for nodes in each partition based on hydraulic simulation data and partition results, and key nodes which can most reflect hydraulic changes in the partition are extracted according to total contribution of comprehensive factors to serve as optimal monitoring points, and the method comprises the following steps:
Obtaining hydraulic simulation data of each partition based on the hydraulic simulation data and the partition result, and after carrying out standardization processing on the hydraulic simulation data, constructing an original matrix composed of data points corresponding to each node at different time points for the hydraulic simulation data of each partition;
Carrying out non-negative matrix factorization on the original matrix of each partition by adopting different factor numbers to obtain a weight matrix and a feature matrix, taking the product of the weight matrix and the feature matrix as a reconstruction matrix, and selecting the factor number corresponding to the minimum error as the optimal factor number of the non-negative matrix factorization by comparing the error of the reconstruction matrix corresponding to the original matrix and each factor number;
And screening an optimal feature matrix corresponding to the optimal factor number for each partition, wherein the optimal feature matrix reflects the contribution degree of each node in the partition on each factor, calculating the total contribution degree of the comprehensive factors according to the contribution degree of each factor, and screening the nodes with the total contribution degree of the comprehensive factors larger than a contribution degree threshold as key nodes which can most reflect the hydraulic change in the partition to which the nodes belong, wherein the key nodes are optimal monitoring points.
Preferably, verifying whether the optimal arrangement scheme of the monitoring points is reasonable, and outputting the final optimal arrangement scheme of the monitoring points passing verification, comprising:
And sorting the optimal monitoring points of each partition in the optimal arrangement scheme of the monitoring points in a descending order according to the total contribution degree of the comprehensive factors, screening at least one optimal monitoring point meeting the preset standard at the forefront of the sorting from the sorting according to the preset standard, judging the optimal monitoring point to pass verification, and forming a final optimal arrangement scheme of the monitoring points by all the optimal monitoring points passing verification.
In order to achieve the above object, the embodiment of the present invention further provides a monitoring point optimizing arrangement device based on drainage pipe network partition dimension reduction, including:
the data acquisition module is used for collecting pipe network data corresponding to the drainage pipe network from the geographic information system and acquiring actual monitoring data of the drainage pipe network, wherein the actual monitoring data comprise flow data and liquid level data of each pump station in the pipe network, flow data and liquid level data of an actual monitoring point and rainfall data;
The model construction and simulation module is used for constructing an SWMM hydraulic model based on pipe network data, calibrating by utilizing actual monitoring data, and carrying out hydraulic simulation on rainfall data by utilizing the calibrated SWMM hydraulic model to obtain hydraulic simulation data, wherein the hydraulic simulation data comprises flow data and liquid level data;
The community detection partitioning module is used for partitioning the drainage pipe network according to the hydraulic simulation data and by adopting a community detection algorithm to obtain a partitioning result;
The arrangement scheme generating module is used for carrying out node dimension reduction on nodes in each partition by using non-negative matrix factorization based on hydraulic simulation data and partition results, extracting key nodes which can most reflect hydraulic changes in the partition according to the total contribution degree of the comprehensive factors as optimal monitoring points, and forming a monitoring point optimal arrangement scheme by all the optimal monitoring points;
The scheme verification module is used for verifying whether the optimal arrangement scheme of the monitoring points is reasonable or not and outputting the final optimal arrangement scheme of the monitoring points which passes verification.
Compared with the prior art, the invention has the beneficial effects that at least the following steps are included:
The optimal monitoring points in each area are determined through partition drop and non-negative matrix factorization of the drainage pipe network, so that the number of unnecessary monitoring points can be effectively reduced, the distribution balance of the optimal monitoring points is improved, and the operation and maintenance monitoring cost of the whole drainage pipe network is reduced while the monitoring efficiency is ensured.
Based on SWMM hydraulic model and environmental condition, through community detection partition dimension reduction, the monitoring network corresponding to the monitoring point optimal arrangement scheme is more suitable for various complicated practical application scenes, the change of the whole drainage pipe network can be better covered and reflected in time, and the overall response capability of the system is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a monitoring point optimizing arrangement method based on the partition dimension reduction of a drainage pipe network, which is provided by the embodiment;
fig. 2 is a schematic structural diagram of a monitoring point optimizing arrangement device based on the partition dimension reduction of a drainage pipe network according to an embodiment.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the detailed description is presented by way of example only and is not intended to limit the scope of the invention.
As shown in fig. 1, an embodiment provides a method for optimizing arrangement of monitoring points based on partition dimension reduction of a drainage pipe network, which includes the following steps:
S1, collecting pipe network data corresponding to a drainage pipe network from a geographic information system and acquiring actual monitoring data of the drainage pipe network.
In an embodiment, collecting pipe network data corresponding to a drainage pipe network from a geographic information system (Geography Information Syetem, GIS) includes: pipe network topology data, pipeline numbers, pipeline lengths, pipe diameters, node numbers, node coordinates, node types, pipeline starting point numbers, pipeline ending point numbers, pipeline starting point bottom elevations, pipeline ending point bottom elevations, pump station nodes and the like, soil types, ground elevation data and the like. The nodes refer to key components such as pump stations, water outlets, inspection wells, diversion wells and the like.
The actual monitoring data of the drainage pipe network is time series data of historical monitoring, and specifically comprises flow data and liquid level data of each pump station in the pipe network, flow data and liquid level data of actual monitoring points and rainfall data. The monitoring point is a set point of the monitoring instrument equipment in the drainage pipe network.
S2, constructing an SWMM hydraulic model based on pipe network data, calibrating by using actual monitoring data, and performing hydraulic simulation on rainfall data by using the calibrated SWMM hydraulic model to obtain hydraulic simulation data.
In an embodiment, a SWMM hydraulic model file is built based on Infoworks ICM software that can input the SWMM. The data input to Infoworks ICM software is pipe network data, and comprises pipe properties and node properties of a target drainage pipe network, wherein the pipe properties comprise pipe length, pipe diameter, pipe number, pipe starting point number, pipe ending point number and the like, and the node properties comprise node number, node type, pipe starting point bottom elevation, pipe ending point bottom elevation, node coordinates and the like.
And obtaining an inp file based on Infoworks ICM software to obtain the constructed SWMM hydraulic model. The SWMM hydraulic model is then calibrated by using the actual monitoring data of the actual monitoring points, which specifically comprises:
Initializing model parameters of a SWMM hydraulic model: the model parameters include fixed parameters such as slope and water impermeability (% Imperv) which are not required to be adjusted and are determined by the underlying surface data. The model parameters also comprise adjustable parameters to be adjusted, and initial values corresponding to the adjustable parameters comprise: the attenuation coefficient (Decay) is 3, the Manning coefficient of the impermeable area (N-Imperv) is 0.015, the Manning coefficient of the permeable area (N-perv) is 0.06, the water-impermeable pothole water storage (Dstore-Imperv) is 3.0, the water-permeable pothole water storage (Dstore-Perv) is 0, and the pipeline roughness is 0.012857.
Iterative calibration model parameters: and simulating by using the SWMM hydraulic model with the current parameters to obtain hydraulic simulation data, comparing the hydraulic simulation data with actual monitoring data, calculating Nash efficiency coefficients to evaluate the model performance, adjusting the model parameters one by one according to the Nash efficiency coefficients, performing sensitivity analysis to determine final model parameters to be adjusted, and adjusting the final model parameters.
The first iteration is performed by adopting an SWMM hydraulic model with initial parameters, and the other iterations are performed by adopting an SWMM hydraulic model with updated parameters. And when simulating each round, inputting the historical rainfall data into the SWMM hydraulic model for simulation to obtain hydraulic simulation data, wherein the hydraulic simulation data comprises flow data and liquid level data. Model performance is assessed by comparing the hydraulic simulation data with actual monitoring data and calculating a Nash efficiency coefficient (NSE), wherein the Nash efficiency coefficient is calculated as follows:
；
Wherein, Monitoring point observation flow representing the ith time step,/>Node hydraulic simulation flow representing the ith time step,/>Represents the average of the observed flow at all monitoring points, N being the total time step.
After the Nash efficiency coefficient is obtained, the model parameters are firstly adjusted one by one, and the sensitivity of the parameters is analyzed by using a global sensitivity analysis method so as to know the influence degree of the parameters on the model output. The obtained parameter sensitivity analysis results are as follows: the sensitivity of the pipe roughness is at most much greater than other parameters, the sensitivity of the key parameters is ordered as follows: pipeline roughness > attenuation coefficient > Manning coefficient of permeable region > Manning coefficient of impermeable region > water permeable pothole water storage > water impermeable pothole water storage. And then, determining final model parameters to be adjusted as pipeline roughness and attenuation coefficient according to the sensitivity analysis result, and adjusting the two parameters.
Model output: model calibration was considered successful when the Nash efficiency coefficient reached the standard and a calibrated SWMM hydraulic model was output. In each round of iteration, calculating to obtain a Nash efficiency coefficient, comparing the Nash efficiency coefficient with a coefficient threshold, considering that the model calibration is successful when the Nash efficiency coefficient is larger than or equal to the coefficient threshold, outputting a calibrated SWMM hydraulic model, and carrying out the iterative calibration of the next round when the Nash efficiency coefficient is smaller than the coefficient threshold.
After the calibrated SWMM hydraulic model is obtained, hydraulic simulation is carried out on rainfall data by using the calibrated SWMM hydraulic model to obtain hydraulic simulation data, and the method specifically comprises the following steps: and inputting the historical rainfall data with different intensities into a calibrated SWMM hydraulic model, and performing multiple simulation to obtain hydraulic simulation data to form a data set. In order to ensure that the simulation situation covers possible events as much as possible, the selected historical rainfall data comprises rainfall events with different intensities such as light rain, medium rain, heavy rain and the like, and the number of rainfall events with each intensity is more than 200. If the number of the historical rainfall events can not meet the event number requirement of each intensity, a Chicago design rainfall formula is used, and rainfall time is generated to complement according to the required rainfall intensity and duration.
In a specific example, 200 different rainfall events were screened for each intensity from the historical observations. The rainfall events take the time sequence rainfall data consisting of the rainfall intensity every 10 minutes as the input of a SWMM hydraulic model, and the obtained hydraulic simulation data comprise liquid level data and flow data of different time points of each node.
And S3, partitioning the drainage pipe network according to the hydraulic simulation data by adopting a community detection algorithm to obtain a partitioning result.
In an embodiment, the community detection algorithm may employ a Leiden algorithm. Partitioning the drainage pipe network based on the Leiden algorithm comprises the following steps:
(a) The hydraulic simulation data is subjected to normalization processing, wherein the normalization processing generally refers to normalization processing of liquid level data and flow data of each node contained in the hydraulic simulation data. Meanwhile, a network model is built based on the topological structure of the drainage pipe network, nodes in the network model represent nodes of the drainage pipe network, the nodes comprise key components such as a pump station, a water outlet, an inspection well, a diversion well and the like, and edges represent pipelines to represent water flow paths among the nodes;
(b) Constructing a weighted adjacency matrix of a network model according to the standardized hydraulic simulation data Matrix/>Is/>Representing the weight of the edge between nodes i and j. The weight is the combination of the node flow and the liquid level, and is specifically expressed as follows:
；
Wherein the method comprises the steps of Is a parameter between 0 and 1 for adjusting the relative importance of flow data and level data in the weights,/>Is the standardized node flow data,/>Is the standardized node liquid level data.
(C) Carrying out community detection partition on a drainage network represented by a network model by adopting a Leiden algorithm, and calculating modularity according to a weighted adjacent matrix in the Leiden algorithm, wherein the calculating formula of the modularity is as follows:
；
Wherein, Representing modularity of the network,/>And/>The intensities of nodes i and j, respectively, i.e., the sum of the weights of all edges connected to these nodes,/>Is the sum of the weights of all edges in the network,/>Is an indication function,/>And/>Representing the community in which the nodes i and j are located, when the nodes i and j belong to the same community, the value is 1, otherwise, the value is 0,/>Is a resolution parameter for controlling granularity of network partitioning, increasing/>The penalty term in the modularity formula will be increased, tending to generate more smaller partitions, and vice versa.
Based on the calculated modularity, the process of community detection partition on the drainage network represented by the network model by adopting the Leiden algorithm is as follows: after the communities are initialized first, starting from a random node in the network model, attempting to move the node to its neighboring communities, and if the modularity M increases, then performing the movement. Traversing all nodes, repeating the steps of module calculation and movement until the increase of the module is no longer remarkable or a certain iteration number is reached. And then, regarding all the current communities formed by the nodes as supernodes, wherein the edge weights among the current supernodes are the sum of the weights of all the nodes in the communities after the nodes move, performing community detection on the supernodes again by using a Leiden algorithm, and repeatedly performing local movement and community merging until the community structure is stable, namely after the modularity is not obviously improved any more, outputting community labels of the nodes, namely, a partition result.
(D) Analyzing the hydraulic consistency in the partition and the hydraulic independence between the partitions, and outputting a partition result when the hydraulic response of the partition is judged to be reasonable according to the hydraulic consistency and the hydraulic independence.
In the embodiment, the distribution and the change trend of the flow and the liquid level of each node in the partition are observed by analyzing the hydraulic consistency in the partition and the hydraulic independence among the partitions. The hydraulic consistency means that each node in the partition shows similar modes and trends in flow and liquid level changes, and can be judged by taking at least one of standard deviations of the flow and the liquid level of each node in the partition as a hydraulic consistency characterization value, wherein the smaller the hydraulic consistency characterization value is, the higher the hydraulic consistency in the partition is. In practice, when the hydraulic consistency characterization value is selected, a monitoring target is mainly considered, when the monitoring target is the flow, the standard deviation of the flow is selected, when the monitoring target is the liquid level, the standard deviation of the liquid level is selected, the standard deviation of the parameters corresponding to the monitoring equipment is selected, when the monitoring target is the flow and the liquid level at the same time, the standard deviation of the flow and the liquid level is selected to be comprehensively considered, and when judging whether the hydraulic response is reasonable or not, the standard deviation is required to be compared with the respective threshold value.
Hydraulic independence means that the hydraulic behavior in one partition should be independent of the other partition to the extent possible, i.e. the interaction between partitions is minimal. The interaction between the subareas can be represented by taking the ratio of the boundary flow to the total subarea flow as a hydraulic independence characterization value, wherein the smaller the hydraulic independence characterization value is, the smaller the interaction of water flows between the subareas is, and the higher the hydraulic independence is.
Judging that the hydraulic response of the subarea is reasonable based on hydraulic consistency and hydraulic independence comprises: and when the hydraulic consistency characterization value and the hydraulic independence characterization value are smaller than the respective corresponding threshold values, the hydraulic response is reasonable, otherwise, the hydraulic response is considered to be unreasonable.
(E) Adjusting parameters in a weighted adjacency matrix calculation formula when the hydraulic response of the judgment partition according to the hydraulic consistency and the hydraulic independence is not reasonableAnd resolution parameters/>, in a module calculation formulaAnd then, the community detection partition is carried out again until the hydraulic response of the partition is reasonable.
And S4, carrying out node dimension reduction on nodes in each partition by using non-negative matrix factorization based on the hydraulic simulation data and the partition result, extracting key nodes which can reflect the hydraulic change most in the partition according to the total contribution degree of the comprehensive factors as optimal monitoring points, and forming a monitoring point optimal arrangement scheme by all the optimal monitoring points.
In an embodiment, constructing a monitoring point optimal arrangement scheme includes:
Firstly, obtaining hydraulic simulation data of each partition based on the hydraulic simulation data and partition results, and carrying out standardization processing on the hydraulic simulation data, wherein the standardization processing is to carry out normalization processing on flow data and liquid level data contained in the hydraulic simulation data, and construct an original matrix composed of data points corresponding to each node at different time points for the hydraulic simulation data of each partition.
And then, carrying out non-negative matrix factorization on the original matrix of each partition by adopting different factor numbers to obtain a weight matrix and a feature matrix, taking the product of the weight matrix and the feature matrix as a reconstruction matrix, and selecting the factor number corresponding to the minimum matrix error as the optimal factor number of the non-negative matrix factorization by comparing the matrix error of the reconstruction matrix corresponding to each factor number with the original matrix.
In a specific example, different factor numbers of 2-8 can be selected, and matrix errors under the factor numbers can be calculated and the optimal factor number can be screened.
And finally, screening an optimal feature matrix corresponding to the optimal factor number for each partition, wherein the optimal feature matrix reflects the contribution degree of each node in the partition on each factor, calculating the total contribution degree of the comprehensive factors according to the contribution degree of each factor, and screening the nodes with the total contribution degree of the comprehensive factors larger than a contribution degree threshold as key nodes which can most reflect the hydraulic change in the partition to which the nodes belong, wherein the key nodes are optimal monitoring points, so that a monitoring point optimal arrangement scheme is formed.
The contribution threshold value adopts a statistical error, namely the contribution threshold value is the sum of the average value and twice standard deviation of the total contribution of the comprehensive factors of all nodes.
And S5, verifying whether the optimal arrangement scheme of the monitoring points is reasonable, and outputting the final optimal arrangement scheme of the monitoring points passing verification.
The optimal arrangement scheme of the monitoring points obtained through the step S4 is obtained based on optimization of an algorithm under ideal conditions, but in practical application, preset standards of practical application exist, and the preset standards can be the number of the monitoring points based on a budget nucleometer and can also be objective reasons for whether the selected nodes have the monitoring points which cannot be installed. Therefore, when the optimal arrangement scheme of the monitoring points is obtained, whether the optimal arrangement scheme of the monitoring points is reasonable or not is verified, and the method specifically comprises the following steps:
And sorting the optimal monitoring points of each partition in the optimal arrangement scheme of the monitoring points in a descending order according to the total contribution degree of the comprehensive factors, screening at least one optimal monitoring point meeting the preset standard at the forefront of the sorting from the sorting according to the preset standard, judging the optimal monitoring point to pass verification, and forming a final optimal arrangement scheme of the monitoring points by all the optimal monitoring points passing verification. And the optimal monitoring points which do not meet the preset standard are directly abandoned.
As shown in fig. 2, the embodiment further provides a monitoring point optimizing arrangement device 20 based on the drainage pipe network partition dimension reduction, which includes: the system comprises a data acquisition module 21, a model construction and simulation module 22, a community detection partition module 23, an arrangement scheme generation module 24 and a scheme verification module 25, wherein the data acquisition module 21 is used for collecting pipe network data corresponding to a drainage pipe network from a geographic information system and acquiring actual monitoring data of the drainage pipe network; the model construction and simulation module 22 is used for constructing an SWMM hydraulic model based on pipe network data, calibrating the SWMM hydraulic model by using actual monitoring data, and performing hydraulic simulation on rainfall data by using the calibrated SWMM hydraulic model to obtain hydraulic simulation data; the community detection partitioning module 23 is used for partitioning the drainage pipe network according to the hydraulic simulation data and by adopting a community detection algorithm to obtain a partitioning result; the arrangement scheme generating module 24 is used for carrying out node dimension reduction on nodes in each partition by using non-negative matrix factorization based on hydraulic simulation data and partition results, extracting key nodes which can most reflect hydraulic changes in the partition according to the total contribution degree of the comprehensive factors as optimal monitoring points, and forming a monitoring point optimal arrangement scheme by all the optimal monitoring points; the scheme verification module 25 is used for verifying whether the optimized arrangement scheme of the monitoring points is reasonable or not, and outputting the final optimized arrangement scheme of the monitoring points passing verification.
It should be noted that, when the monitoring point optimizing arrangement device based on the partition dimension reduction of the drainage pipe network provided in the foregoing embodiment performs the monitoring point optimizing arrangement, the division of the foregoing functional modules should be used to illustrate, where the foregoing functional allocation may be completed by different functional modules according to the need, that is, the internal structure of the terminal or the server is divided into different functional modules, so as to complete all or part of the functions described above. In addition, the embodiment of the method for constructing the optimized arrangement of the monitoring points based on the dimension reduction of the drainage pipe network partition provided by the embodiment belongs to the same conception, and the detailed implementation process of the method is detailed in the embodiment of the optimized arrangement of the monitoring points based on the dimension reduction of the drainage pipe network partition, and is not repeated here.
The foregoing detailed description of the preferred embodiments and advantages of the invention will be appreciated that the foregoing description is merely illustrative of the presently preferred embodiments of the invention, and that no changes, additions, substitutions and equivalents of those embodiments are intended to be included within the scope of the invention.
Claims (10)
1. A monitoring point optimizing arrangement method based on drainage pipe network partition dimension reduction is characterized by comprising the following steps:
Collecting pipe network data corresponding to a drainage pipe network from a geographic information system and acquiring actual monitoring data of the drainage pipe network, wherein the actual monitoring data comprise flow data and liquid level data of each pump station in the pipe network, flow data and liquid level data of actual monitoring points and rainfall data;
Constructing an SWMM hydraulic model based on pipe network data, calibrating by using actual monitoring data, and performing hydraulic simulation on rainfall data by using the calibrated SWMM hydraulic model to obtain hydraulic simulation data, wherein the hydraulic simulation data comprises flow data and liquid level data;
Partitioning the drainage pipe network according to the hydraulic simulation data and by adopting a community detection algorithm to obtain a partitioning result;
Node dimension reduction is carried out on nodes in each partition by using non-negative matrix factorization based on hydraulic simulation data and partition results, key nodes which can most reflect hydraulic changes in the partition are extracted according to the total contribution of comprehensive factors to serve as optimal monitoring points, and all the optimal monitoring points form a monitoring point optimal arrangement scheme;
And verifying whether the optimal arrangement scheme of the monitoring points is reasonable or not, and outputting the final optimal arrangement scheme of the monitoring points passing verification.
2. The method for optimizing and arranging monitoring points based on the partition dimension reduction of the drainage pipe network according to claim 1, wherein the method for constructing and calibrating the SWMM hydraulic model based on pipe network data comprises the following steps:
model construction and initialization: constructing an SWMM hydraulic model according to pipe network data, and initializing parameters of the SWMM hydraulic model;
Iterative calibration model parameters: the SWMM hydraulic model with current parameters is utilized for simulation to obtain hydraulic simulation data, the hydraulic simulation data is compared with actual monitoring data, nash efficiency coefficients are calculated to evaluate model performance, model parameters are adjusted one by one according to the Nash efficiency coefficients, sensitivity analysis is carried out to determine final model parameters to be adjusted, and the final model parameters are adjusted;
Model output: model calibration was considered successful when the Nash efficiency coefficient reached the standard and a calibrated SWMM hydraulic model was output.
3. The method for optimizing arrangement of monitoring points based on drainage pipe network partition dimension reduction according to claim 1, wherein the hydraulic simulation of rainfall data by using a calibrated SWMM hydraulic model comprises the following steps:
And inputting rainfall data with different intensities into a calibrated SWMM hydraulic model, and performing multiple simulation to obtain hydraulic simulation data, wherein the selected rainfall data comprises rainfall events with different intensities of light rain, medium rain and heavy rain.
4. The method for optimizing arrangement of monitoring points based on drainage pipe network partition dimension reduction according to claim 1, wherein the method for partitioning the drainage pipe network according to hydraulic simulation data by adopting a community detection algorithm comprises the following steps:
Carrying out standardized processing on the hydraulic simulation data, constructing a network model based on the topological structure of the drainage pipe network, wherein nodes in the network model represent nodes of the drainage pipe network, edges represent pipelines, represent water flow paths among the nodes, and constructing a weighted adjacency matrix of the network model according to the standardized processed hydraulic simulation data;
And carrying out community detection partition on the drainage network represented by the network model by adopting a Leiden algorithm, calculating modularity according to a weighted adjacent matrix in the Leiden algorithm, analyzing hydraulic consistency in the partition and hydraulic independence between the partitions, and outputting a partition result when judging that the hydraulic response of the partition is reasonable according to the hydraulic consistency and the hydraulic independence.
5. The method for optimizing the arrangement of monitoring points based on the partition dimension reduction of the drainage pipe network according to claim 4, wherein a weighted adjacency matrix of a network model constructed according to the standardized hydraulic simulation data is denoted by W,Each element/>The weight of the edge between the nodes i and j is represented, the weight is the combination of the node flow and the liquid level, and the calculation formula is as follows:
；
Wherein the method comprises the steps of Is a parameter between 0 and 1, for adjusting the relative importance of flow data and level data in the weights,Is the standardized node flow data,/>Is the node liquid level data after standardized treatment;
the formula used for calculating modularity according to the weighted adjacency matrix is:
；
Wherein, Representing modularity of the network,/>And/>The intensities of nodes i and j, respectively, i.e., the sum of the weights of all edges connected to these nodes,/>Is the sum of the weights of all edges in the network,/>And/>Representing the community in which nodes i and j are located,/>Is an indication function, when nodes i and j belong to the same community, the value is 1, otherwise, the value is 0,/>Is a resolution parameter used to control the granularity of the network partition.
6. The optimal arrangement method of monitoring points based on the partition dimension reduction of the drainage pipe network according to claim 5, wherein when the hydraulic response of the partition is judged to be unreasonable according to the hydraulic consistency and the hydraulic independence, parameters in a weighted adjacent matrix calculation formula are adjustedAnd resolution parameters/>, in a module calculation formulaAnd then, the community detection partition is carried out again until the hydraulic response of the partition is reasonable.
7. The optimal arrangement method of monitoring points based on the partition dimension reduction of the drainage pipe network according to claim 4 or 6, wherein the hydraulic consistency is judged by taking at least one of the standard deviation of the flow and the liquid level of each node in the partition as a hydraulic consistency characterization value, and the smaller the hydraulic consistency characterization value is, the higher the hydraulic consistency in the partition is indicated;
the hydraulic independence represents interaction among the subareas by taking the ratio of the boundary flow to the total subarea flow as a hydraulic independence characterization value, wherein the smaller the hydraulic independence characterization value is, the smaller the water flow interaction among the subareas is, and the higher the hydraulic independence is;
And when the hydraulic consistency characterization value and the hydraulic independence characterization value are smaller than the respective corresponding threshold values, the hydraulic response is reasonable, otherwise, the hydraulic response is considered to be unreasonable.
8. The optimal arrangement method of monitoring points based on the partition dimension reduction of the drainage pipe network according to claim 1, wherein non-negative matrix factorization is used for nodes in each partition based on hydraulic simulation data and partition results, and key nodes which can most reflect hydraulic changes in the partition are extracted as optimal monitoring points according to total contribution of comprehensive factors, and the method comprises the following steps:
Obtaining hydraulic simulation data of each partition based on the hydraulic simulation data and the partition result, and after carrying out standardization processing on the hydraulic simulation data, constructing an original matrix composed of data points corresponding to each node at different time points for the hydraulic simulation data of each partition;
Carrying out non-negative matrix factorization on the original matrix of each partition by adopting different factor numbers to obtain a weight matrix and a feature matrix, taking the product of the weight matrix and the feature matrix as a reconstruction matrix, and selecting the factor number corresponding to the minimum error as the optimal factor number of the non-negative matrix factorization by comparing the error of the reconstruction matrix corresponding to the original matrix and each factor number;
And screening an optimal feature matrix corresponding to the optimal factor number for each partition, wherein the optimal feature matrix reflects the contribution degree of each node in the partition on each factor, calculating the total contribution degree of the comprehensive factors according to the contribution degree of each factor, and screening the nodes with the total contribution degree of the comprehensive factors larger than a contribution degree threshold as key nodes which can most reflect the hydraulic change in the partition to which the nodes belong, wherein the key nodes are optimal monitoring points.
9. The optimal arrangement method of monitoring points based on the partition dimension reduction of the drainage pipe network according to claim 1, wherein verifying whether the optimal arrangement scheme of the monitoring points is reasonable and outputting the final optimal arrangement scheme of the monitoring points passing verification comprises the following steps:
And sorting the optimal monitoring points of each partition in the optimal arrangement scheme of the monitoring points in a descending order according to the total contribution degree of the comprehensive factors, screening at least one optimal monitoring point meeting the preset standard at the forefront of the sorting from the sorting according to the preset standard, judging the optimal monitoring point to pass verification, and forming a final optimal arrangement scheme of the monitoring points by all the optimal monitoring points passing verification.
10. Monitoring point optimal placement device based on drainage pipe network subregion dimension reduction, its characterized in that includes:
the data acquisition module is used for collecting pipe network data corresponding to the drainage pipe network from the geographic information system and acquiring actual monitoring data of the drainage pipe network, wherein the actual monitoring data comprise flow data and liquid level data of each pump station in the pipe network, flow data and liquid level data of an actual monitoring point and rainfall data;
The model construction and simulation module is used for constructing an SWMM hydraulic model based on pipe network data, calibrating by utilizing actual monitoring data, and carrying out hydraulic simulation on rainfall data by utilizing the calibrated SWMM hydraulic model to obtain hydraulic simulation data, wherein the hydraulic simulation data comprises flow data and liquid level data;
The community detection partitioning module is used for partitioning the drainage pipe network according to the hydraulic simulation data and by adopting a community detection algorithm to obtain a partitioning result;
The arrangement scheme generating module is used for carrying out node dimension reduction on nodes in each partition by using non-negative matrix factorization based on hydraulic simulation data and partition results, extracting key nodes which can most reflect hydraulic changes in the partition according to the total contribution degree of the comprehensive factors as optimal monitoring points, and forming a monitoring point optimal arrangement scheme by all the optimal monitoring points;
The scheme verification module is used for verifying whether the optimal arrangement scheme of the monitoring points is reasonable or not and outputting the final optimal arrangement scheme of the monitoring points which passes verification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410559620.2A CN118133587A (en) | 2024-05-08 | 2024-05-08 | Monitoring point optimal arrangement method and device based on drainage pipe network partition dimension reduction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410559620.2A CN118133587A (en) | 2024-05-08 | 2024-05-08 | Monitoring point optimal arrangement method and device based on drainage pipe network partition dimension reduction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118133587A true CN118133587A (en) | 2024-06-04 |
Family
ID=91248247
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410559620.2A Pending CN118133587A (en) | 2024-05-08 | 2024-05-08 | Monitoring point optimal arrangement method and device based on drainage pipe network partition dimension reduction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118133587A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130108912A (en) * | 2012-03-26 | 2013-10-07 | 대림산업 주식회사 | Method for setting a site of sensor in water distribution pipe network |
| CN113221440A (en) * | 2021-04-07 | 2021-08-06 | 浙江大学 | Optimal arrangement and real-time global inversion method for monitoring points of drainage system |
| WO2023284060A1 (en) * | 2021-07-12 | 2023-01-19 | 浙江大学 | Geographic three-dimensional information-based method for analyzing uncertainty of flow rate of sewage pipe network |
| CN117236673A (en) * | 2023-11-16 | 2023-12-15 | 水利部交通运输部国家能源局南京水利科学研究院 | Urban water network multi-scale flood control and drainage combined optimization scheduling method and system |
-
2024
- 2024-05-08 CN CN202410559620.2A patent/CN118133587A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130108912A (en) * | 2012-03-26 | 2013-10-07 | 대림산업 주식회사 | Method for setting a site of sensor in water distribution pipe network |
| CN113221440A (en) * | 2021-04-07 | 2021-08-06 | 浙江大学 | Optimal arrangement and real-time global inversion method for monitoring points of drainage system |
| WO2023284060A1 (en) * | 2021-07-12 | 2023-01-19 | 浙江大学 | Geographic three-dimensional information-based method for analyzing uncertainty of flow rate of sewage pipe network |
| CN117236673A (en) * | 2023-11-16 | 2023-12-15 | 水利部交通运输部国家能源局南京水利科学研究院 | Urban water network multi-scale flood control and drainage combined optimization scheduling method and system |
Non-Patent Citations (3)
| Title |
|---|
| 杨祺琪;张书亮;戴强;靳竺霖;: "基于SWMM和改进差分进化算法的雨水管网优化方法", 中国给水排水, no. 17, 1 September 2016 (2016-09-01) * |
| 苟德菲;杨辉;周午阳;孙志民;: "基于InfoWorks ICM的广州市H区流域排水系统模型验证", 净水技术, no. 04, 25 April 2020 (2020-04-25) * |
| 陈前虎;周明;邹澄昊;张泸少;王贤萍;: "海绵化改造小区调蓄池优化布局及其雨水径流控制效果――以浙江省嘉兴市烟波苑小区为例", 水土保持通报, no. 03, 15 June 2020 (2020-06-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chacon-Hurtado et al. | Rainfall and streamflow sensor network design: a review of applications, classification, and a proposed framework | |
| Lü et al. | The streamflow estimation using the Xinanjiang rainfall runoff model and dual state-parameter estimation method | |
| Fan et al. | Probabilistic prediction for monthly streamflow through coupling stepwise cluster analysis and quantile regression methods | |
| CN106022518A (en) | Pipe damage probability prediction method based on BP neural network | |
| Kişi | Wavelet regression model as an alternative to neural networks for monthly streamflow forecasting | |
| Fan et al. | A stepwise-cluster forecasting approach for monthly streamflows based on climate teleconnections | |
| CN116363601A (en) | Data acquisition and analysis method and system for pollution monitoring equipment | |
| CN110134907B (en) | Rainfall missing data filling method and system and electronic equipment | |
| CN112084608B (en) | Method for identifying risk pipelines and nodes by adopting parameter uncertainty analysis model | |
| CN116796799A (en) | Method for creating small-river basin flood rainfall threshold model in area without hydrologic data | |
| CN113344470B (en) | Intelligent power supply system management method based on block chain | |
| Haverkamp et al. | Subwatershed spatial analysis tool: Discretization of a distributed hydrologic model by statistical criteria 1 | |
| CN116823067B (en) | Method and device for determining water quality cleaning state of pipe network and electronic equipment | |
| Pedersen et al. | All models are wrong, but are they useful? Assessing reliability across multiple sites to build trust in urban drainage modelling | |
| CN117688844A (en) | Urban waterlogging real-time simulation method and system based on deep neural network | |
| Gao et al. | A framework for automatic calibration of SWMM considering input uncertainty | |
| CN118133587A (en) | Monitoring point optimal arrangement method and device based on drainage pipe network partition dimension reduction | |
| Hadiani | Analysis of rainfall-runoff neuron input model with artificial neural network for simulation for availability of discharge at Bah Bolon Watershed | |
| Nasrin et al. | Modelling impact of extreme rainfall on sanitary sewer system by predicting rainfall derived infiltration/inflow | |
| CN115267950B (en) | Initial rainwater quantification method and device, storage medium and electronic equipment | |
| CN115829163A (en) | Multi-mode integration-based runoff prediction method and system for middle and lower reaches of Yangtze river | |
| CN113609683B (en) | Rainwater pipe network sub-catchment area dividing method based on basin analysis and Thiessen polygons | |
| WO2022264422A1 (en) | Rainy weather water infiltration rate estimation device, rainy weather water infiltration rate estimation method, and program | |
| Shaw et al. | A methodology for preserving channel flow networks and connectivity patterns in large‐scale distributed hydrological models | |
| CN114818221A (en) | Pipe network operation situation analysis method based on pipe network hydrodynamic model and data driving model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |