Water network system targeting partitioning method and system

Abstract

The application discloses a water network system targeted partition method and system, collects water network data of a research area, identifies water network nodes and extracts data of each water network node, and constructs an eigenvalue matrix of each water network node; the composite distance between each two nodes is calculated, the number of near neighbors is set based on the number of eigenvalues, water network nodes that are mutual nearest neighbors are screened, a hierarchical graph is constructed with nodes as vertices and nearest neighbor relationships as edges, and a water network characteristic neighborhood graph is obtained; a cutting edge is identified based on the neighborhood relationship between each node, and the water network characteristic neighborhood graph is divided into a plurality of water network characteristic sparse graphs; the center point of each water network characteristic sparse graph is determined based on the local density of the water network node, the similarity between each water network characteristic sparse graph is calculated, the water network characteristic sparse graph is clustered, a plurality of clustering centers are obtained, and the research area is divided into a plurality of water network system targeted partitions. The application considers multi-source data, and improves the refinement and scientific level of water network management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water network partitioning, in particular to a water network system targeted partitioning method. BACKGROUND

[0002] As an important infrastructure for ensuring the rational allocation of urban and regional water resources, flood control and drought resistance, and ecological maintenance, the scientific planning and management of the water network system is crucial. With the acceleration of urbanization and the impact of climate change, the water network system is facing many challenges such as increased flood disaster risk, uneven distribution of water resources, and increased difficulty in water pollution control. In order to effectively address these challenges and improve the operational efficiency and problem-solving capabilities of the water network system, it is necessary to carry out reasonable partitioning management of the water network system.

[0003] Currently, existing water network partitioning methods have many shortcomings. Some methods only rely on geographical regions or administrative boundaries for partitioning, ignoring key factors such as the topological structure of the water network, hydraulic characteristics, and water regime changes in different regions. This simple partitioning method cannot accurately reflect the complex relationships within the water network system, leading to low resource allocation efficiency and difficulty in precise policy implementation in actual water resource allocation, flood control and disaster reduction, and water pollution control. For example, when a flood arrives, it is difficult to quickly and accurately determine the flood propagation path and potentially affected areas, making it impossible to take timely and effective flood control measures. In periods of water scarcity, it is difficult to achieve optimal allocation of water resources to meet the water demand of different regions. Therefore, there is an urgent need for a water network system targeted partitioning method that can consider multiple factors to improve the refinement and scientific level of water network management.

[0004] The present application proposes a water network system targeted partitioning method and system to solve the above problems and improve the refinement and scientific level of water network management. SUMMARY

[0005] The present application provides a water network system targeted partitioning method to solve the above problems in the prior art. On the other hand, a water network system targeted partitioning system is provided.

[0006] Technical solution: A water network system targeted partitioning method, comprising the following steps:

[0007] Step S1, collect water network data of the study area, identify water network nodes and extract data of each water network node, and reduce the dimension of the water network node data to obtain a feature value matrix of each water network node;

[0008] Step S2, calculate the complex distance between each two nodes based on the feature value matrix of each water network node, set the number of nearest neighbors based on the number of feature values, screen water network nodes that are nearest neighbors of each other, and construct a hierarchical graph with nodes as vertices and nearest neighbor relationships as edges to obtain a water network feature neighborhood graph.

[0009] Step S3, identify the cutting edge based on the neighborhood relationship between each node, and divide the water network feature field graph into several water network feature sparse graphs;

[0010] Step S4, determine the center point of each water network feature sparse graph based on the local density of the water network node, calculate the similarity between each water network feature sparse graph, and cluster the water network feature sparse graph to obtain several clustering centers, and divide the study area into several water network system target partitions.

[0011] According to an aspect of the present application, the step S1 is further:

[0012] Step S11, collect water network data of the study area, including: water network topology data, hydraulic data, geographic data, historical drought data and historical flood data;

[0013] Step S12, identify the water network node and extract the data of each water network node, the water network node including: watershed node, engineering node, monitoring node and geographic feature node;

[0014] Step S13, sequentially reduce the dimension of the data of all water network nodes to obtain the eigenvalue matrix of each water network node.

[0015] According to an aspect of the present application, the step S13 is further:

[0016] Step S13a, extract several water network node attributes based on the data of the water network node, and calculate the correlation coefficient between all water network node attributes, classify the water network node attributes with correlation coefficient greater than the threshold value into the same water network node feature attribute, and obtain the water network node feature attribute set;

[0017] Step S13b, calculate the numerical value of each water network node feature attribute by using the entropy weight method, and construct the eigenvalue matrix of each water network node.

[0018] According to an aspect of the present application, the step S2 is further:

[0019] Step S21, calculate the composite distance between each two nodes based on the eigenvalue matrix of each water network node;

[0020] Step S22, set the initial neighbor number value based on the number of eigenvalues, and screen the neighbor nodes of each water network node, gradually increase the size of the neighbor number value, and screen the neighbor nodes of each water network node based on the current neighbor number value, until all water network nodes meet at least one neighbor node, at this time the neighbor number value is the effective neighbor number value;

[0021] Step S23, filtering the near-neighbor nodes of each water network node based on the effective near-neighbor quantity value, and constructing a hierarchical graph with the nodes as vertices and the nearest neighbor relationship as edges to obtain a water network feature neighborhood graph.

[0022] According to an aspect of the present application, the step S21 is further:

[0023] Step S21a, identifying the river basin in the study area as a river basin node, the water conservancy hub, the flood drainage channel and the storage project as an engineering node, the hydrological station as a monitoring node, the river junction, the drought area and the high flood area as a geographical feature node;

[0024] Step S21b, sequentially calculating the geographical distance, the hydraulic distance, the climate distance and the management distance between each two nodes;

[0025] Step S21c, determining the weight of each distance and calculating the composite distance between each two nodes.

[0026] According to an aspect of the present application, the step S23 is further:

[0027] Step S23a, constructing a hierarchical neighborhood graph of the river basin node, the engineering node and the geographical feature node respectively with the nodes as vertices and the near-neighbor relationship as edges;

[0028] Step S23b, representing the water conservancy association, the water transfer association and the climate association by blue edges, red edges and yellow edges respectively, representing the high flood risk area node by red, representing the medium flood risk area node by orange, representing the drought area node by yellow, and representing the climate correlation by the thickness of the edge to obtain a water network feature neighborhood graph.

[0029] According to an aspect of the present application, the step S3 is further:

[0030] Step S31, setting a cutting threshold based on the number of water network node feature values, calculating the near-neighbor quantity of each water network node, marking the water network nodes with a near-neighbor quantity greater than the cutting threshold, and retaining the edges between the marked water network nodes and all their near-neighbor nodes as cutting edges, dividing the water network feature neighborhood graph into M subgraphs and a number of scattered water network nodes based on the cutting edges, M being a positive integer greater than 10;

[0031] Step S32, sequentially calculating the distance of each scattered water network node and M subgraphs, and dividing the scattered water network nodes into the nearest sub-region to obtain M water network feature sparse graphs.

[0032] According to an aspect of the present application, the step S4 is further:

[0033] Step S41, calculating the local density of each water network node, and determining the center point of each water network feature sparse graph based on the local density of the water network node.

[0034] Step S42, the distance between the center points of each two water network feature sparse graphs is calculated in sequence, that is, the distance of the two water network feature sparse graphs;

[0035] Step S43, the number of shared nearest neighbors is set, and the ratio of the number of shared natural nearest neighbors to the distance of the two water network feature sparse graphs is taken as the similarity of the two water network feature sparse graphs, and the similarity between all water network feature sparse graphs is calculated in sequence;

[0036] Step S44, the two water network feature sparse graphs with the highest similarity are screened out and clustered into the same water network partition, a cluster label is obtained, and the cluster label is taken as a water network feature sparse graph to calculate the similarity between all water network feature sparse graphs again, and the iteration is repeated until all water network feature sparse graphs belong to a certain cluster label, N cluster labels are obtained, and the study area is divided into N water network system targeted partitions, and N is a positive integer less than M.

[0037] According to another aspect of the application, a water network system targeted partition system is provided, comprising:

[0038] at least one processor; and

[0039] a memory connected in communication with the at least one processor; wherein,

[0040] the memory stores instructions executable by the processor, and the instructions are executed by the processor to implement the water network system targeted partition method of any one of the above technical solutions.

[0041] Beneficial effects: the water network system targeted partition method is adopted to realize the targeted partition of the study area, and differentiated water saving and water supply guarantee strategies are formulated according to the water resource conditions and water demand of different partitions. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 is a flowchart of the present application.

[0043] Figure 2 is a flowchart of step S1 of the present application.

[0044] Figure 3 is a flowchart of step S2 of the present application.

[0045] Figure 4 is a flowchart of step S3 of the present application.

[0046] Figure 5 is a flowchart of step S4 of the present application. DETAILED DESCRIPTION

[0047] As Figure 1As shown, the following technical scheme is proposed. According to one aspect of the present application, a water network system targeted partitioning method is provided, characterized by comprising the following steps:

[0048] Step S1, collecting water network data of a study area, identifying water network nodes and extracting data of each water network node, and reducing the dimension of the water network node data to obtain an eigenvalue matrix of each water network node;

[0049] Step S2, calculating the composite distance between each two nodes based on the eigenvalue matrix of each water network node, setting a nearest neighbor value based on the number of eigenvalues, screening water network nodes that are nearest neighbors of each other, and constructing a hierarchical graph with nodes as vertices and nearest neighbor relationships as edges to obtain a water network characteristic neighborhood graph;

[0050] Step S3, identifying cut edges based on the neighborhood relationship between each node, and dividing the water network characteristic field graph into a plurality of water network characteristic sparse graphs;

[0051] Step S4, determining the center point of each water network characteristic sparse graph based on the local density of the water network node, calculating the similarity between each water network characteristic sparse graph, and clustering the water network characteristic sparse graphs to obtain a plurality of clustering centers, and dividing the study area into a plurality of water network system targeted partitions.

[0052] The current water network system partitioning management faces three key problems: first, the data dimension is single, and traditional methods rely mainly on geographical topology or hydraulic data, ignoring key information such as climate characteristics, historical disasters, and management ownership, resulting in partitioning results that cannot reflect the comprehensive function of the water network (such as splitting the associated nodes of inter-basin water transfer projects); second, the correlation is one-sided, and only geographical distance is used to measure node relationships, which cannot reflect core attributes such as hydraulic connectivity (such as the flow impact of upstream on downstream) and climate correlation (such as the response of nodes in the same rainfall belt), and the neighborhood relationship is disconnected from the actual function; third, the partitioning logic is extensive, and traditional clustering methods (such as K-means) are prone to forcibly dividing densely connected core areas (such as reservoir groups) or omitting isolated nodes (such as small irrigation canals), resulting in a disconnect between partitioning and management needs, making it difficult to support targeted decisions such as flood control and drought resistance.

[0053] In actual implementation, traditional water network partitioning methods face three major implementation obstacles: first, high-dimensional data processing is difficult, water network node attributes involve more than 20 items (such as flow, drought frequency, elevation, etc.), direct calculation is prone to data redundancy, and subjective weighting can lead to feature bias; second, it is difficult to define node correlation, the water network system has geographical, hydraulic, and climate correlations, and a single distance indicator cannot accurately distinguish between key and secondary correlations, which can misjudge the nearest neighbor relationship (such as excluding nodes in the same irrigation area due to geographical distance); third, it is difficult to ensure the integrity of the partitioning, and it is difficult to balance the attribution of core areas and isolated nodes, and forcibly clustering can split the regulation relationship of water conservancy projects, and ignoring isolated nodes can lead to management blind spots.

[0054] The application firstly reduces dimension through multi-source data fusion and entropy weight method, integrates topological, hydraulic, geographical, historical disaster and other data, extracts core features (such as flood risk, hydraulic connectivity), constructs an objective feature value matrix, solves the problems of high-dimensional data redundancy and subjective deviation; secondly, an innovative composite distance model is used to weight and fuse geographical, hydraulic, climatic and management distances, combined with dynamic near neighbor adjustment, to construct a multi-dimensional neighborhood graph and accurately depict node association; thirdly, the core association (such as the connection between large-scale water conservancy projects) is identified by cutting edges, and the isolated nodes are distributed to the nearest subgraph to ensure the integrity of the partition; finally, the center point is determined based on the local density, the similarity is calculated by sharing the near neighbor + distance, and the functional-oriented targeted partition is obtained by iterative clustering, so as to realize the partition that meets the management needs.

[0055] According to one aspect of the application, the step S1 is further:

[0056] Step S11, collecting water network data of the study area, including: water network topological structure data, hydraulic data, geographical data, historical drought data and historical flood data;

[0057] Step S12, identifying water network nodes and extracting data of each water network node, the water network nodes including: basin nodes, engineering nodes, monitoring nodes and geographical feature nodes;

[0058] Step S13, sequentially reducing dimension of data of all water network nodes to obtain feature value matrix of each water network node.

[0059] According to one aspect of the application, the step S13 is further:

[0060] Step S13a, extracting a plurality of water network node attributes based on the data of the water network nodes, and calculating the correlation coefficients between all the water network node attributes, classifying the water network node attributes with correlation coefficients greater than a threshold value into the same water network node characteristic attribute to obtain a water network node characteristic attribute set;

[0061] From the collected topological, hydraulic, geographical, historical disaster and management data, quantitative indicators reflecting the core function and characteristics of the nodes are screened, specifically:

[0062] Defining the node type (basin / engineering / monitoring / geographical feature node) and determining the attribute extraction direction (such as highlighting the regulation capacity of engineering nodes and the disaster sensitivity of geographical feature nodes);

[0063] Screening quantifiable indicators from the corresponding data sources (excluding qualitative descriptions, such as converting the wide river into the average width of the river);

[0064] Unifying the data units (such as the unit of drought days is d, and the unit of flow is m3 / s).

[0065] Water network node attributes specifically include:

[0066] Hydraulic function attributes: multi-year average flow, water level amplitude, water delivery efficiency, river flow capacity;

[0067] Geographical environment attributes: node elevation, soil permeability, vegetation coverage, river average width;

[0068] Disaster risk attributes: annual average flood days, flood inundation area, annual average drought days, drought impact range;

[0069] Engineering regulation attributes: total reservoir capacity, gate regulation accuracy, pumping station water lifting capacity;

[0070] Monitoring feedback attributes: water level monitoring error, water quality compliance rate, flow monitoring frequency;

[0071] Management coordination attributes: administrative division attribution, cross-regional dispatch response time.

[0072] The Pearson correlation coefficient method is used to calculate the linear correlation degree of two attributes, and the specific steps are as follows:

[0073] Determine the attribute pair to be calculated (such as annual average flood days and node elevation, multi-year average flow and water delivery efficiency);

[0074] Collect two attribute data of all nodes (assuming there are 100 nodes, get two groups of 100 data: X=[x_1,x_2,...,x_{100}], Y=[y_1,y_2,...,y_100];

[0075] Substitute the Pearson correlation coefficient formula to calculate the correlation coefficient;

[0076] Judge the correlation: set the threshold value as, if the correlation coefficient is greater than the threshold value, it is considered that the two attributes are strongly correlated and can be classified into the same characteristic attribute.

[0077] In an embodiment, 22 original attributes are extracted, and after correlation calculation, a set of 8 characteristic attributes is finally formed, specifically:

[0078] Flood risk characteristics: annual average flood days, flood inundation area, node elevation, correlation: "annual average flood days-flood inundation area" r=0.85, "flood inundation area-node elevation" r=-0.78 (negative correlation is still strong correlation);

[0079] Drought risk characteristics: annual average drought days, drought impact range, soil permeability, correlation: "annual average drought days-drought impact range" r=0.91, "drought impact range-soil permeability" r=-0.73;

[0080] Hydraulic connection features: mean annual flow, conveyance efficiency, river channel flow capacity, correlation: "mean annual flow-conveyance efficiency" r=0.82, "mean annual flow-river channel flow capacity" r=0.87;

[0081] Engineering regulation features: total reservoir capacity, gate regulation accuracy, pumping station water lifting capacity, only engineering node attributes, correlation>0.75;

[0082] Geographical location features: node elevation, average width of river channel, distance from main stream, correlation: "node elevation-distance from main stream" r=0.72;

[0083] Ecological environment features: vegetation coverage, water quality compliance rate, correlation: r=0.71;

[0084] Monitoring accuracy features: water level monitoring error, flow monitoring frequency, only monitoring node attributes, correlation r=0.76;

[0085] Management coordination features: cross-regional scheduling response time, administrative partition attribution code, correlation: r=0.74.

[0086] In particular, in the feature attribute set, some original attributes (such as node elevation) may be repeatedly classified due to the simultaneous association with multiple core features (such as flood risk and geographical location), and need to be prioritized in the core associated features (such as node elevation prioritized in the flood risk feature). Step S13b, the numerical values of each water network node feature attribute are calculated by using the entropy weight method, and the feature value matrix of each water network node is constructed.

[0087] The numerical values of each water network node feature attribute are calculated by using the entropy weight method, specifically:

[0088] Standardize the original attribute data, wherein, positive attribute: x' ij = (x ij -x jmin ) / (x jmax -x jmin );

[0089] Negative attribute: x' ij = (x jmax -x ij ) / (x jmax -x jmin );

[0090] Wherein, x ij is the jth original attribute value of the ith node, x jmax / x jmin is the maximum / minimum value of the jth attribute, x' ijfor the standardized value (range 0-1);

[0091] Calculate the information entropy of the attribute, which reflects the degree of dispersion of the attribute: the smaller the entropy value, the higher the attribute dispersion, the more information contained, and the greater the weight:

[0092] According to the information entropy, calculate the weight of each original attribute in the feature attribute:

[0093] Sum the weighted sum of the standardized original attribute value and the corresponding weight to obtain the final numerical value of the node under the feature attribute (range 0-1):

[0094] Repeat the above steps for other feature attributes to obtain the numerical value of each node under all feature attributes.

[0095] In an embodiment, specifically:

[0096] Select three nodes A, B, and C in a certain watershed: the flood risk features include three original attributes:

[0097] A: 12 days of average annual flood days, 80 km2 of flood inundated area, and 25 m of node elevation;

[0098] B: 8 days of average annual flood days, 50 km2 of flood inundated area, and 35 m of node elevation;

[0099] C: 15 days of average annual flood days, 100 km2 of flood inundated area, and 18 m of node elevation;

[0100] After standardization:

[0101] A: 0.57, 0.6, 0.59;

[0102] B: 0.0, 0.0, 0.0;

[0103] C: 1.0, 1.0, 1.0;

[0104] Calculate the information entropy:

[0105] A: (0.57+10^{-6}) / (0.57+0+1+3×10^{-6})≈0.36;

[0106] B: ≈0;

[0107] C: ≈0.64;

[0108] e_1=-1 / \ln3×(0.36\ln0.36+0\ln0+0.64\ln0.64)≈0.89;

[0109] e_2≈0.89;

[0110] e_3≈0.89;

[0111] wherein e_1 represents flood days, e_2 represents inundated area, and e_3 represents elevation;

[0112] Calculate weights:

[0113] \sum_{j=1}^{3}(1-e_j)=(1-0.89)+(1-0.89)+(1-0.89)=0.33;

[0114] w_1=w_2=w_3=0.11 / 0.33≈0.33;

[0115] Calculate characteristic attribute values:

[0116] A: 0.33x0.57+0.33x0.6+0.33x0.59≈0.59;

[0117] B: 0.33x0+0.33x0+0.33x0=0;

[0118] C: 0.33x1+0.33x1+0.33x1≈1.0;

[0119] After calculating the values of all 8 characteristic attributes, a characteristic value matrix of 3 nodes is constructed.

[0120] In another embodiment of the present application, step S13 can also be:

[0121] The node function mutation recognition in step S13a includes: constructing a function state vector containing five dimensions of drainage, water storage, regulation and storage, flood discharge and water transport, each dimension representing the activation degree of the corresponding function; calculating the rainfall impact through the hyperbolic tangent function of the difference between the rainfall intensity and the critical value, and obtaining the function change amount by combining the state transition matrix; decomposing the state transition matrix into a basic matrix and the sum of the correction matrix of soil water content, pipe network load and terrain gradient; decomposing the function change into minute-level fast variables, hour-level medium variables and day-level slow variables, and obtaining the comprehensive function state by superposition.

[0122] wherein the authenticity discrimination of the function mutation includes: calculating the time integral of the function state relative to the reference state as a persistence index, calculating the complement of the correlation coefficient of the current state and the historical state as an irreversibility index, and calculating the average value of the function change of the adjacent nodes as a cascading index; combining the three indexes by weighting and mapping through an S-shaped function to obtain a mutation authenticity score; when the authenticity score is greater than a preset threshold, it is determined that it is a real function mutation, otherwise it is determined that it is a false mutation.

[0123] Optionally, further comprising: correcting the detected function state according to the authenticity score, retaining the detection result of the real mutation, and restoring the pseudo mutation to the benchmark state; predicting the mutation duration based on a logarithmic function of the difference between the corrected function state and the benchmark state; and using the corrected function state vector as a node feature attribute for subsequent water network partitioning.

[0124] Specifically, the method comprises the following steps:

[0125] Step S13a1, constructing a node function state vector;

[0126] In this embodiment, a five-dimensional function vector is defined to represent the multi-function characteristics of the water network node. Specifically, the function vector is represented as:

[0127] F(t) = [f_drain, f_store, f_regulate, f_discharge, f_transfer];

[0128] Wherein, F(t) represents the node function state vector at time t; f_drain represents the drainage function activation degree, with a value range of 0 to 1; f_store represents the water storage function activation degree, with a value range of 0 to 1; f_regulate represents the regulation and storage function activation degree, with a value range of 0 to 1; f_discharge represents the flood discharge function activation degree, with a value range of 0 to 1; f_transfer represents the water transfer function activation degree, with a value range of 0 to 1.

[0129] The initial values of each function activation degree are obtained based on the statistical data of historical 3-year operation data. For example, the initial function vector of a certain drainage pump station is [0.8, 0.1, 0.05, 0.03, 0.02], indicating that the node mainly undertakes the drainage function.

[0130] Step S13a2, calculating the state transition driven by rainfall;

[0131] In this embodiment, a rainfall impact function is introduced to simulate the influence of extreme rainfall on node function. Specifically, the state transition calculation formula is:

[0132] ΔF(t) = tanh(α × (R(t) - R_critical)) × W × F(t-1);

[0133] Wherein, ΔF(t) represents the change amount of the function vector; R(t) represents the rainfall intensity at time t, with a unit of millimeters / hour; R_critical represents the critical rainfall intensity, with a value of 50-100 millimeters / hour; α represents the response sensitivity, with a value of 0.01-0.05; W represents a 5x5 state transition matrix; tanh represents the hyperbolic tangent function, which is used to limit the response to between -1 and 1.

[0134] Step S13a3, extract the early-stage influencing factor;

[0135] In this embodiment, the fixed transition matrix is improved to a dynamic matrix considering multiple environmental factors. Specifically, the dynamic transition matrix is calculated as:

[0136] W(t) = W_base + ΔW_soil(θ) + ΔW_load(L) + ΔW_topo(G);

[0137] where W_base represents the basic transition matrix; ΔW_soil(θ) represents the soil water content correction matrix, θ is the soil volume water content, ranging from 0 to 0.4; ΔW_load(L) represents the pipe network load correction matrix, L is the pipe network load rate, ranging from 0 to 1; ΔW_topo(G) represents the topographic gradient correction matrix, G is the topographic gradient, unit: degree.

[0138] The calculation formula of the soil water content correction matrix is:

[0139] ΔW_soil(θ) = k_soil × (θ - θ_field) × M_soil

[0140] where k_soil represents the soil influence coefficient, taking a value of 0.5-2.0; θ_field represents the field water holding capacity, a typical value of 0.25; M_soil represents the soil influence template matrix.

[0141] Step S13a4, perform multi-time scale fusion;

[0142] In this embodiment, the functional state change is decomposed into three time scales, specifically:

[0143] F(t) = F_fast(t) + F_medium(t) + F_slow(t);

[0144] The calculation formula of the fast variable (minute level) is:

[0145] F_fast(t) = φ1 × exp(-t / τ1) × ΔF_rain;

[0146] where φ1 represents the fast variable weight, taking a value of 0.4-0.6; τ1 represents the fast variable time constant, taking a value of 0.1-0.5 hours; ΔF_rain represents the functional change caused by rainfall.

[0147] The calculation formula of the medium variable (hour level) is:

[0148] F_medium(t) = φ2 × (1-exp(-t / τ2)) × ΔF_load;

[0149] Wherein, φ2 represents the medium variable weight, the value is 0.2-0.4; τ2 represents the medium time constant, the value is 1-5 hours; ΔF_load represents the functional change caused by load.

[0150] The calculation formula of slow variable (day level) is:

[0151] F_slow(t)=φ3×(1 / (1+exp(-(t-τ3) / s)))×ΔF_seasonal;

[0152] Wherein, φ3 represents the slow variable weight, the value is 0.1-0.2; τ3 represents the slow time center, the value is 12-24 hours; s represents the conversion width, the value is 2-6 hours; ΔF_seasonal represents the seasonal functional change.

[0153] Step S13a5, extracting mutation feature fingerprint;

[0154] In this embodiment, a three-dimensional mutation feature vector is defined to distinguish real mutations and pseudo mutations, specifically, the feature vector is calculated as:

[0155] Persistence index: P=(∫(F(t)-F_baseline)dt) / T;

[0156] Wherein, the integral interval is T hours after the mutation starts, T takes the value of 2-6 hours; F_baseline represents the baseline functional state.

[0157] Irreversibility index: I=1-corr(F(t), F(t-T));

[0158] Wherein, corr represents the correlation coefficient calculation function; T represents the observation time window.

[0159] Cascade index: C=(Σneighbor_change) / N_neighbors;

[0160] Wherein, neighbor_change represents the functional change of adjacent nodes; N_neighbors represents the number of adjacent nodes.

[0161] Step S13a6, constructing real and pseudo mutation discriminator;

[0162] In this embodiment, the authenticity of the mutation is determined by weighted combination of the three feature indexes, specifically, the discrimination function is:

[0163] S_real=1 / (1+exp(-(w1×P+w2×I+w3×C-θ_truth)));

[0164] wherein S real represents the mutation authenticity score, ranging from 0 to 1; w1, w2, w3 represent feature weights, obtained by training historical data, and typical values are 0.4, 0.3, 0.3 respectively; and θ truth represents the authenticity threshold, taking a value of 0.5-0.7.

[0165] The weights are obtained by least square method optimization:

[0166] min∑(S real - S observed) 2 ;

[0167] Step S13a7, performing function state correction and prediction.

[0168] In the embodiment, the detected function state is corrected based on the authenticity score, specifically:

[0169] F corrected(t) = S real × F detected(t) + (1 - S real) × F baseline(t) ;

[0170] The mutation duration prediction formula is:

[0171] T duration = κ × ln(1 + ||F corrected - F baseline||) × (1 / decay_rate) ;

[0172] wherein κ represents the duration coefficient, taking a value of 10-30; ||·|| represents the vector norm; and decay_rate represents the recovery rate, taking a value of 0.05-0.2 per hour.

[0173] For example, when the 1-hour rainfall of a certain drainage pump station reaches 120 mm, the function vector changes rapidly from [0.8, 0.1, 0.05, 0.03, 0.02] to [0.2, 0.6, 0.15, 0.03, 0.02], the drainage function decreases, and the water storage function increases rapidly, through mutation feature extraction, the persistence P = 0.8, the irreversibility I = 0.7, and the cascading C = 0.6 are obtained, the discrimination score S real = 0.82, and it is confirmed that it is a real mutation, and it is predicted that the mutation will last for 8.5 hours.

[0174] Through the above steps, the node function mutation recognition accuracy reaches more than 95%, and the response time is controlled within 20 seconds.

[0175] According to an aspect of the present application, the step S2 is further:

[0176] Step S21, calculating the composite distance between each two nodes based on the eigenvalue matrix of each water network node;

[0177] Step S22, setting an initial neighbor number value based on the feature value number, and screening the neighbor nodes of each water network node, gradually increasing the size of the neighbor number value, and screening the neighbor nodes of each water network node based on the current neighbor number value, until all water network nodes meet at least one neighbor node, at this time the neighbor number value is the effective neighbor number value;

[0178] Step S23, screening the neighbor nodes of each water network node based on the effective neighbor number value, and constructing a hierarchical graph with nodes as vertices and nearest neighbor relationships as edges to obtain a water network feature neighborhood graph.

[0179] Water network targeted partitioning is the core basis of water network planning layout and system optimization, and its scientific demarcation needs to consider multiple dimensional influencing factors, the present application comprehensively considers the distribution characteristics of regional topography, water system, social and economic development level, etc., combines rainfall, evaporation and dryness, annual runoff spatial distribution law, arid region division, floodwater risk elements, comprehensive risk degree and flood partitioning, and adopts a water network targeted partitioning method based on water network feature neighborhood graph division.

[0180] If each data point has at least one natural neighbor, the entire water network data set can be considered to be in a stable distribution state, and a natural neighbor relationship between two data points is defined, and in the water network feature neighborhood graph, there will be an edge between the corresponding nodes, further, the water network natural neighborhood graph is constructed, and the following four variables need to be determined:

[0181] Nearest neighbor determination: for any data point, the nearest neighbor function is searched to determine the search times of the nearest neighbors of the data point under the condition of a given search times, and a nearest neighbor set is obtained;

[0182] Stable search state: for each data point in the data set, there is at least one data point such that the two points are nearest neighbors of each other;

[0183] Natural neighbor: if data point 1 is one of the nearest neighbors of data point 2, and data point 2 is also one of the nearest neighbors of data point 1, then data point 2 is considered to be a natural neighbor of data point 1;

[0184] Natural feature value set: is a set composed of all positive integers that satisfy the condition that for each data point in the data set, there is at least one data point such that the two points are nearest neighbors of each other.

[0185] According to one aspect of the present application, the step S21 further comprises:

[0186] Step S21a, identifying the basins in the study area as basin nodes, the water conservancy hubs, flood drainage channels and storage projects as engineering nodes, the hydrological stations as monitoring nodes, and the river junctions, arid regions and flood-prone areas as geographical feature nodes;

[0187] Step S21b, sequentially calculate the geographical distance, hydraulic distance, climate distance and management distance between each two nodes;

[0188] The geographical distance is the straight-line distance of the spatial position of the node, which is calculated by the Haversine formula based on latitude and longitude coordinates;

[0189] The hydraulic distance represents the transmission time of water flow from the upstream node to the downstream node;

[0190] The climate distance represents the difference in climate characteristics between nodes, which is quantified by the standard deviation of the average annual rainfall;

[0191] The management distance is the degree of difficulty of water conservancy management coordination between nodes, which is quantified by a 0-1 non-dimensional value, and a double-factor comprehensive assignment of management unit attribution and response time is adopted:

[0192] Management unit attribution: 0.2 is assigned to attribution to the same level of basin management bureau, 0.5 is assigned to attribution to the same level of water conservancy department, and 0.8 is assigned to attribution across levels;

[0193] Dispatch response time: the average response time of cross-regional dispatch is standardized to 0-0.2;

[0194] Final management distance: D_{man}=management unit attribution value+response time standardized value.

[0195] Step S21c, determine the weight of each distance and calculate the composite distance between each two nodes.

[0196] Score the importance of the four types of distances, build a judgment matrix, calculate the maximum eigenvalue, random consistency index and consistency ratio of the judgment matrix, and finally determine the weight vector;

[0197] Calculate the composite distance:

[0198] D AB =w1D geo，AB +w2D hyd，AB +w3D cli，AB +w4D man，AB ;

[0199] Where D AB is the composite distance between nodes A and B.

[0200] According to one aspect of the present application, the step S23 is further:

[0201] Step S23a, respectively construct the basin node, engineering node and geographical feature node hierarchical neighborhood graph with nodes as vertices and near-neighbor relationships as edges;

[0202] Step S23b, the water-related association, water transfer association and climate association are respectively represented by blue edges, red edges and yellow edges, the high flood risk area nodes are represented by red, the medium flood risk area nodes are represented by orange, the drought area nodes are represented by yellow, the climate correlation is represented by the thickness of the edge, and the water network feature neighborhood graph is obtained.

[0203] The sub-graphs are constructed according to the node types, including the basin node hierarchical graph, the engineering node hierarchical graph and the geographical feature node hierarchical graph, and the color-shape-thickness multi-dimensional visual variables are used to intuitively display the node types, risk levels and association characteristics, specifically: the node color represents the disaster risk level, wherein red: flood risk characteristic value > 0.8 (high risk); orange: 0.5 ≤ flood risk characteristic value ≤ 0.8 (medium risk); yellow: drought risk characteristic value > 0.7 (drought area); green: other (low risk);

[0204] The node shape represents the node type, wherein the circle is the basin node; the square is the engineering node; the triangle is the monitoring node; and the rhombus is the geographical feature node.

[0205] The edge color represents the association type, wherein blue: water-related association (water collection / water transfer) between basin nodes; red: water transfer association between engineering nodes; and yellow: climate association between geographical feature nodes.

[0206] The thickness of the edge represents the climate correlation, wherein thick edge (line width 3pt): climate distance < 10mm (high correlation); medium edge (2pt): 10mm ≤ climate distance ≤ 30mm (medium correlation); and thin edge (1pt): climate distance > 30mm (low correlation).

[0207] According to one aspect of the present application, the step S21 of calculating the hydraulic distance can further include:

[0208] The hydraulic conduction time delay is calculated, including:

[0209] The basic time delay is calculated according to the river length between nodes, the basic wave speed and the real-time flow, and is asymmetrically corrected through the exponential decay relationship between the potential energy gradient and the flow; the hydraulic state is divided into steady state, gradual change state and sudden change state according to the flow rate, and a smooth transition function is used to realize the continuous transition between different states; the modified basic time delay is used for the steady state, the time delay adjusted by the logarithmic function is used for the gradual change state, and the time delay of the exponential decay is used for the sudden change state; the comprehensive time delay is obtained by weighted combination of the time delays of the three states, and the historical memory mechanism is introduced to smooth the output time delay, and the smoothed time delay is taken as the hydraulic distance between nodes.

[0210] The specific implementation of the hydraulic state division is:

[0211] The steady state is determined when the flow rate change rate is less than a first threshold value, the gradual change state is determined when the flow rate change rate is between the first threshold value and a second threshold value, and the abrupt change state is determined when the flow rate change rate is greater than or equal to the second threshold value; the first threshold value ranges from 0.1 to 0.5 cubic meters per second square, and the second threshold value ranges from 2.0 to 5.0 cubic meters per second square; and the smooth transition function adopts an S-shaped function, and the smooth degree of state transition is controlled by adjusting a transition steepness parameter.

[0212] The history memory mechanism comprises:

[0213] The current calculated comprehensive time delay is weighted and averaged with the output time delay at the previous moment, wherein an update rate is dynamically adjusted according to the flow rate change rate, and the more intense the flow rate change, the higher the update rate; and near the state transition boundary, a steepness parameter of the transition function is dynamically adjusted according to the variance of the flow rate change rate, and the greater the variance, the smaller the steepness, so as to reduce the time delay jump during state switching.

[0214] Specifically, the method comprises the following steps:

[0215] Step S21b1, constructing a basic time delay matrix;

[0216] In this embodiment, the basic time delay matrix between nodes is constructed based on the simplified Saint-Venant equation by introducing a time delay difference factor, and specifically, for any two water network nodes i and j, the basic time delay calculation formula is: τ(i,j)=L(i,j) / (c0+α×Q(t));

[0217] Wherein, τ(i,j) represents the hydraulic conduction time delay from node i to node j, and the unit is hour; L(i,j) represents the river length between node i and node j, and the unit is kilometer; c0 represents the basic wave speed, and the value range is 0.5-2.0 m / s, which is determined according to the average water depth and bottom slope of the river; Q(t) represents the real-time flow at time t, and the unit is cubic meter per second; and α represents the flow influence coefficient, and the value range is 0.001-0.01, which is used to quantify the influence degree of flow on propagation speed.

[0218] Step S21b2, performing asymmetric correction;

[0219] In this embodiment, a potential energy gradient correction term is introduced to capture the asymmetric characteristics of hydraulic conduction, and specifically, the calculation formula of the correction term is: Δτ=β×(Hi-Hj)×exp(-γ×|Q(t)| / Qbase);

[0220] Wherein, Δτ represents the time delay correction quantity, the unit is hour; Hi represents the water level elevation of node i, the unit is meter; Hj represents the water level elevation of node j, the unit is meter; β represents the potential energy influence coefficient, the value range is 0.1-0.5, taking positive value when water flows from high place to low place, and taking negative value conversely; γ represents the flow attenuation coefficient, the value range is 0.5-2.0; Qbase represents the reference flow, usually taking the annual average flow value, the unit is cubic meter / second.

[0221] The corrected time delay is: τ_corrected(i,j)=τ(i,j)+Δτ;

[0222] Step S21b3, identifying the hydraulic state;

[0223] In the embodiment, the current hydraulic state is determined by calculating the flow rate of change, specifically, three hydraulic states are defined: steady state, gradual change state and sudden change state, and the state determination is based on the following:

[0224] When dQ / dt<θ1, it is determined as steady state; when θ1≤dQ / dt<θ2, it is determined as gradual change state; when dQ / dt≥θ2, it is determined as sudden change state, wherein dQ / dt represents the derivative of flow with respect to time, the unit is cubic meter / second 2 ; θ1 represents the steady state threshold, the value range is 0.1-0.5 cubic meter / second 2 ; θ2 represents the sudden change threshold, the value range is 2.0-5.0 cubic meter / second 2 .

[0225] Step S21b4, performing segmented time delay calculation;

[0226] In the embodiment, different time delay calculation methods are adopted according to the identified hydraulic state, for steady state, the corrected time delay of step S21b2 is used; for gradual change state, the time delay calculation formula is: τ_gradual=τ_base×(1+κ×ln(1+|dQ / dt|));

[0227] Wherein, τ_gradual represents the time delay of gradual change state; τ_base represents the base time delay; κ represents the gradual change adjustment coefficient, the value range is 0.2-0.8.

[0228] For sudden change state, the time delay calculation formula is: τ_sudden=τ_min+(τ_base-τ_min)×exp(-λ×t);

[0229] Wherein, τ_sudden represents the time delay of sudden change state; τ_min represents the minimum time delay, usually 0.1-0.3 times of τ_base; λ represents the attenuation rate, the value range is 0.5-2.0; t represents the time after the sudden change occurs, the unit is hour.

[0230] Step S21b5, constructing a smooth state transition function;

[0231] In this embodiment, a Sigmoid function is used to realize smooth transition between different states. Specifically, the weight function is defined as: w(x) = 1 / (1+exp(-p x(x-theta)));

[0232] wherein w(x) represents the weight function value, ranging from 0 to 1; x represents the flow rate change rate dQ / dt; p represents the transition steepness parameter, ranging from 2.0 to 10.0; theta represents the transition center point, corresponding to the state transition threshold theta1 or theta2.

[0233] Through the combination of the three weight functions, the final time delay is calculated as: tau_final = w1(dQ / dt) x tau_steady + w2(dQ / dt) x tau_gradual + w3(dQ / dt) x tau_sudden;

[0234] wherein w1, w2, and w3 respectively correspond to the weights of the steady state, the gradual state, and the sudden state, satisfying w1 + w2 + w3 = 1.

[0235] Step S21b6, applying a history memory mechanism;

[0236] In this embodiment, a time delay history buffer is introduced to prevent instantaneous jumps. Specifically, the update formula of the output time delay is: tau_output(t) = eta x tau_final(t) + (1-eta) x tau_output(t-Dt);

[0237] wherein tau_output(t) represents the output time delay at time t; eta represents the update rate, with a basic value of 0.3; Dt represents the time step, usually 5 minutes.

[0238] The update rate is dynamically adjusted according to the hydraulic state: eta = eta_base x (1+mu x |dQ / dt|);

[0239] wherein eta_base represents the basic update rate, with a value of 0.2-0.4; mu represents the sensitivity coefficient, with a value of 0.1-0.5.

[0240] Step S21b7, performing boundary adaptive correction.

[0241] In this embodiment, a buffer width adaptive mechanism is introduced near the state transition boundary. Specifically, the transition steepness parameter is dynamically adjusted as: p(t) = p_base x (1-v x Var(dQ / dt));

[0242] Wherein, p(t) represents the conversion steepness at t time; p_base represents the basic steepness value, taking 5.0; Var(dQ / dt) represents the variance of flow rate change rate in the past 1 hour; v represents the variance sensitivity coefficient, taking 0.1-0.3.

[0243] For example, in a certain water network system, node A is located in an upstream reservoir, and node B is located in a downstream urban area. Under normal circumstances, the distance between the two nodes is 30 kilometers, the basic wave speed is 1.5 meters per second, and the calculated basic time delay is 5.56 hours. When the upstream sluice is opened for flood discharge, the flow rate increases from 100 cubic meters per second to 500 cubic meters per second within 10 minutes, and dQ / dt reaches 40 cubic meters per second. The system determines that it is in a sudden state, and at this time the time delay is rapidly reduced to 1.2 hours. Through the historical memory mechanism, the output time delay is smoothly transitioned, avoiding system shock caused by sudden changes.

[0244] Through the above steps, the dynamic quantification of hydraulic conduction time delay is realized, the time delay prediction accuracy is improved to more than 85%, and the calculation time is controlled within 50 milliseconds.

[0245] According to one aspect of the present application, the step S3 is further:

[0246] Step S31, based on the number of water network node characteristic values, set the cutting threshold, calculate the number of near neighbors of each water network node, mark the water network nodes with the number of near neighbors greater than the cutting threshold, and reserve the edges of the marked water network nodes and all their near neighbor nodes as cutting edges. Based on the cutting edges, the water network characteristic neighborhood graph is divided into M subgraphs and a number of scattered water network nodes, M is a positive integer greater than 10;

[0247] Step S32, calculate the distance of each scattered water network node and M subgraphs in turn, and divide the scattered water network nodes into the nearest sub-region to obtain M water network characteristic sparse graphs.

[0248] Hierarchical clustering is an important branch of clustering algorithms, which constructs a tree-like hierarchical structure through similarity calculation between objects. Hierarchical clustering is divided into two types: agglomerative hierarchical clustering and divisive hierarchical clustering. The latter is a variant based on the former. Agglomerative hierarchical clustering mainly includes the following two steps: first, calculate the similarity between data point pairs; then, iteratively select the data point pair or cluster with the highest similarity for merging. Divisive hierarchical clustering involves three main steps: first, identify and select the cluster with the highest dissimilarity for splitting; second, find the point with the highest separation degree in the cluster to be split and divide it into a new cluster; finally, recalculate the similarity between the data points in the to-be-split cluster and the data points in the new cluster, and according to these similarities, objects are grouped into more similar clusters.

[0249] Hierarchical clustering is often combined with other algorithms to improve the performance of hierarchical clustering, but it also brings some limitations, such as sensitivity to parameter configuration and low efficiency in processing large data.

[0250] In this embodiment, a hierarchical clustering algorithm based on water network feature neighborhood graph is adopted, which includes two processes of water network feature neighborhood graph division and neighborhood subgraph merging. First, in the division process, the standard water network feature neighborhood graph is cut to obtain the subgraph located in the core of the data, and the remaining points are distributed based on the shortest distance. Second, in the neighborhood subgraph merging process, the two most similar subgraphs are dynamically merged according to the similarity matrix, thereby effectively solving the problem of density difference between clusters.

[0251] The specific process of water network feature neighborhood graph division is as follows:

[0252] Search for the nearest neighbor of each data point;

[0253] Find the stable search state of natural neighbors;

[0254] Add edges between data points with natural neighbor relationship;

[0255] Identify the cut edges and divide the water network feature neighborhood graph into water network feature sparse graph;

[0256] Assign the remaining nodes to the nearest water network feature sparse graph;

[0257] Obtain the final water network feature sparse graph.

[0258] Cut edges refer to edges connecting nodes, at least one of which has more natural neighbors than the threshold. These edges constitute the core area of the natural neighborhood graph, specifically:

[0259] cut_edge(v i ,v j )={e(v i ,v j )∣(nb(i)>(u-1)∣∣(nb(j)>(u-1)),(e(v i ,v j )∈nan_dege)};

[0260] When there is a natural neighbor relationship between nodes, an edge will be formed between them. For each pair of natural neighbor nodes, if the number of natural neighbors of one of them exceeds the threshold, this edge will be retained and called a cut edge. These cut edges define the core part of the natural neighborhood graph. Using these cut edges and water network feature sparse graphs, and distributing the remaining nodes to the nearest subgraph based on the nearest neighbor principle, the division of the natural neighborhood graph is completed.

[0261] According to one aspect of the present application, in step S31, a water network edge risk assessment is further included, specifically:

[0262] The hydraulic importance is calculated according to the ratio of the actual flow to the maximum design flow, the water supply guarantee rate, and the complement of the redundancy of the alternative path; the recursive formula of the node function integrity is used to simulate the cascade propagation process after the edge failure, wherein the function loss is related to the influence weight and loss proportion of the upstream failed edge; the time-varying propagation rate is calculated according to the power function of the ratio of the current water level to the conventional water level and the exponential function of the flow reciprocal; the cascade influence is divided into a direct layer, an indirect layer, and a remote layer according to the propagation distance, and the accurate calculation, linear approximation, and statistical estimation methods are respectively used.

[0263] The specific implementation of the hierarchical cascade assessment is that, for the direct layer nodes within one hop range from the failed edge, the direct loss is obtained by accumulating the product of the function integrity and the influence weight of each node; for the indirect layer nodes two to three hops away from the failed edge, the indirect loss is obtained by multiplying the product of the node number and the average loss rate by the attenuation coefficient; for the remote layer nodes more than three hops away from the failed edge, the remote loss is estimated by using a negative exponential decay function; and the total loss approximation value is obtained by weighted summation of the three layer losses.

[0264] The identification of the vulnerable edge further includes: dynamically updating the vulnerability assessment of the edge by calculating the weighted average model value and the historical observation value; predicting the vulnerability at a future time according to the vulnerability change rate and the seasonal periodic function; and marking the edges with vulnerability exceeding a preset threshold as critical vulnerable edges, which are preferentially protected from being cut off during graph cutting.

[0265] In other words, according to one aspect of the present application, in step S31, a water network edge risk assessment is further included, specifically:

[0266] Identify the vulnerable edge in the system and calculate the influence of its failure on the stability of the partition;

[0267] When determining the cutting threshold, the distribution of the vulnerable edge is considered to avoid using a critical vulnerable edge as the only connection between partitions;

[0268] For the area containing a high-risk vulnerable edge, the cutting threshold is appropriately reduced to increase the redundant connection.

[0269] Or, step S3A, identifying the vulnerable edge of the water network system:

[0270] Calculate the hydraulic importance and failure risk of each edge;

[0271] Simulate the cascade influence of edge failure;

[0272] Mark the critical vulnerable edge and establish a risk archive;

[0273] Step S31, adjust the cutting strategy based on risk assessment:

[0274] Adjust the region containing the fragile edge based on the original cutting threshold;

[0275] Ensure that each partition has sufficient redundant connections.

[0276] In one embodiment, according to an aspect of the present application, step S31 is followed by a partition stability verification:

[0277] Perform a fragility analysis on each subgraph after cutting;

[0278] Identify key fragile edges between partitions and within partitions;

[0279] Evaluate the impact of fragile edge failure on partition functionality;

[0280] If it is found that the partition relies excessively on fragile edges, adjust the partition boundaries.

[0281] Specifically, the following steps are included:

[0282] Step S31a1, construct a hydraulic dependency graph;

[0283] In this embodiment, the hydraulic importance of an edge is defined to quantify the criticality of each edge in the water network system. Specifically, the hydraulic importance calculation formula is: I_h(e) = Q(e) / Q_max x P_supply(e) x (1 - R_alternative(e));

[0284] where I_h(e) represents the hydraulic importance of edge e, with a value range of 0 to 1; Q(e) represents the actual flow of edge e, with a unit of cubic meters per second; Q_max represents the maximum design flow of the edge, with a unit of cubic meters per second; P_supply(e) represents the water supply guarantee rate of edge e, with a value range of 0.85 to 0.99; R_alternative(e) represents the alternative path redundancy, with a calculation formula of: R_alternative(e) = N_alt / (N_alt + 1); where N_alt represents the number of alternative paths.

[0285] Step S31a2, simulate cascading failure propagation;

[0286] In this embodiment, a failure propagation model is established to predict the cascading effect triggered by single edge failure. Specifically, the propagation formula for node failure state is: F(v, t+1) = F(v, t) x (1 - Σ(w_e x Loss(e)));

[0287] Wherein, F(v, t) represents the functional integrity of node v at t time, the value range is 0 to 1, and the initial value is 1; w_e represents the influence weight of edge e on node v, which is determined according to the topological distance; Loss(e) represents the functional loss ratio caused by the failure of edge e.

[0288] The calculation formula of the loss ratio is: Loss(e) = min(1, Q_lost(e) / Q_demand(v));

[0289] Wherein, Q_lost(e) represents the flow loss caused by the failure of edge e; Q_demand(v) represents the water demand of node v.

[0290] Step S31a3, calculating the dynamic propagation rate;

[0291] In this embodiment, the time-varying propagation rate is introduced to adapt to the difference in cascade speed under different water conditions. Specifically, the cascade propagation rate calculation formula is: v_cascade(t) = v_base x (H(t) / H_normal)^β x exp(-γ / Q(t));

[0292] Wherein, v_cascade(t) represents the cascade propagation rate at t time, the unit is node / hour; v_base represents the basic propagation rate, the value is 1-5 node / hour; H(t) represents the current water level, the unit is meter; H_normal represents the normal water level, the unit is meter; β represents the water level influence index, the value is 1.5-2.5; γ represents the flow influence coefficient, the value is 10-50; Q(t) represents the current flow, the unit is cubic meter / second.

[0293] Step S31a4, analyzing multi-path cascade coupling;

[0294] In this embodiment, the mutual influence between multiple failure paths is considered. Specifically, the total loss calculation formula is:

[0295] Loss_total = Loss_direct + Σ(α_ij x Loss_i x Loss_j);

[0296] Wherein, Loss_total represents the total loss; Loss_direct represents the direct loss; Loss_i and Loss_j represent the losses of paths i and j respectively; α_ij represents the path coupling coefficient, the calculation formula is: α_ij = overlap(i, j) / min(length(i), length(j)); Wherein, overlap(i, j) represents the number of overlapping nodes of paths i and j; length represents the path length.

[0297] Step S31a5, performing key path pre-screening;

[0298] In the embodiment, the critical path set is screened based on the hydraulic importance and the topological position, and specifically, the screening criterion is K={p|I_h(p)>θ_importance AND Depth(p)<D_max};

[0299] wherein K represents the critical path set; p represents the path; θ_importance represents the importance threshold value, and takes a value of 0.6-0.8; Depth(p) represents the depth of the path in the network, i.e. the shortest distance to the source node; and D_max represents the maximum depth threshold value, and takes a value of 3-5.

[0300] Step S31a6, implementing hierarchical cascade approximation;

[0301] In the embodiment, the cascade influence is differentiated and calculated in three layers according to the propagation distance, and specifically:

[0302] The direct layer (1-hop range) adopts accurate calculation: Loss_direct=Σ(F(v,t)×w_direct(v));

[0303] The indirect layer (2-3-hop range) adopts linear approximation: Loss_indirect=k1×N_indirect×avg_loss;

[0304] The far-end layer (greater than 3 hops) adopts statistical estimation: Loss_far=k2×exp(-distance / λ)×total_nodes;

[0305] wherein w_direct(v) represents the direct influence weight; N_indirect represents the number of nodes in the indirect layer; avg_loss represents the average loss rate; k1 represents the attenuation coefficient of the indirect layer, and takes a value of 0.3-0.5; k2 represents the far-end layer coefficient, and takes a value of 0.1-0.2; and λ represents the attenuation length, and takes a value of 5-10.

[0306] The total loss is approximated as: Loss_approx=Loss_direct+k1×Loss_indirect+k2×Loss_far;

[0307] Step S31a7, performing dynamic update of vulnerability.

[0308] In the embodiment, the vulnerability evaluation of the edge is constantly corrected based on the historical failure events, and specifically, the update formula is:

[0309] V_new(e)=λ×V_calculated(e)+(1-λ)×V_observed(e);

[0310] wherein V_new(e) represents the updated vulnerability of edge e; V_calculated(e) represents the vulnerability calculated by the model; V_observed(e) represents the vulnerability actually observed; and λ represents a learning rate, and takes a value of 0.7-0.9.

[0311] The time series prediction formula of the vulnerability is: V(e, t+Δt) = V(e, t) + μ×dV / dt + ν×seasonality(t);

[0312] wherein μ represents a trend coefficient, and takes a value of 0.1-0.3; dV / dt represents a rate of change of the vulnerability; ν represents a seasonal coefficient, and takes a value of 0.05-0.15; and seasonality(t) represents a seasonal function, and is fitted by a sine function: seasonality(t) = A×sin(2π×t / T + φ); wherein A represents an amplitude; T represents a period, and is usually 365 days; and φ represents a phase.

[0313] For example, in a certain urban water network system, a water pipeline connecting a main water plant and an urban area is identified, the hydraulic importance I_h of which is 0.85, and it is found through cascade analysis that the failure of the pipeline will cause 15 downstream nodes to lose water within 2 hours, affecting 300,000 residents, and the calculation time is reduced from 45 minutes to 30 seconds by using the hierarchical approximation algorithm, while the prediction accuracy is maintained at above 92%.

[0314] Through the above steps, the accuracy of the key vulnerable edge identification is 93%, the prediction error of the cascade range is controlled within 15%, and the calculation efficiency is improved by more than 50 times.

[0315] According to an aspect of the present application, the step S4 is further:

[0316] Step S41, calculate the local density of each water network node, and determine the center point of each water network feature sparse graph based on the local density of the water network node;

[0317] The traditional k-nearest neighbor algorithm is often used to calculate the density, but it is difficult to select the optimal k value, therefore, in the present embodiment, a natural neighbor method is used to search for the nearest neighbor information, and unlike the k-nearest neighbor algorithm, the natural neighbor method can adaptively find the nearest neighbor according to the structural characteristics of the data itself, and the whole process does not require any parameters, and the natural neighbor set and the natural feature value are obtained by the natural neighbor method, and then the local density is calculated based on the natural neighborhood.

[0318] Step S42, calculate the distance between the center points of each two water network feature sparse graphs in turn, which is the distance of the two water network feature sparse graphs;

[0319] Step S43, set the number of shared nearest neighbors, take the ratio of the number of shared natural nearest neighbors and the distance of two water network feature sparse graphs as the similarity of the two water network feature sparse graphs, and calculate the similarity between all water network feature sparse graphs in turn;

[0320] Water network neighborhood subgraph merging is a step for merging water network feature sparse graphs to achieve more accurate clustering results, the input of the algorithm is the water network feature sparse graph and the expected cluster number, and the output is the clustering label, specifically:

[0321] Initialize the number of subgraphs to be the number of water network feature sparse graphs;

[0322] According to the shared natural nearest neighbors, the center distance and the similarity between subgraphs, the similarity between each pair of subgraphs is calculated;

[0323] When the number of subgraphs is greater than the expected cluster number, the following operations are performed:

[0324] Find the two subgraphs with the highest similarity;

[0325] Update the similarity between the subgraph and other subgraphs;

[0326] Subgraph number is reduced by 1;

[0327] For each data point belonging to the subgraph, assign a cluster label.

[0328] Return the final clustering label.

[0329] The shared nearest neighbors determine the number of shared nearest neighbors between two subgraphs, which is the intersection of the nearest neighbor sets of two points, and is used to evaluate the similarity between two subgraphs;

[0330] The calculation of the center point of the subgraph is specifically:

[0331] C(m)=argmaxρ i , v i ∈G NS (m), dc=|c(m)-c(n)|;

[0332] Where ρ i is the local density, and v i is the ith eigenvalue.

[0333] The center distance dc is defined as the Euclidean distance between the two center points |c(m)-c(n)|.

[0334] The ratio of the number of shared natural nearest neighbors and the center distance between the two subgraphs is the similarity between the two subgraphs.

[0335] Step S44, screening out two water network feature sparse graphs with the highest similarity, clustering them into the same water network partition, obtaining a clustering label, and taking it as a water network feature sparse graph to calculate the similarity between all water network feature sparse graphs again, and iteratively until all water network feature sparse graphs belong to a certain clustering label, obtaining N clustering labels, dividing the study area into N water network system targeting partitions, and N is a positive integer less than M.

[0336] In another embodiment of the present application, in step S44, when the water exchange oscillation of the adjacent partition boundary occurs, a suppression process is further included, specifically:

[0337] The number of times of changing the direction of the boundary flow in a unit time is taken as an oscillation index; the oscillation risk is predicted according to the water level difference on both sides of the boundary, the flow rate of change, and the number of adjacent nodes through an S-shaped function; an energy function including the product of oscillation intensity and water exchange amount and the distance of boundary position change is constructed; and the boundary position is iteratively optimized through gradient descent method combined with momentum term to minimize the total oscillation energy.

[0338] According to the water level and flow rate of change, a day is divided into multiple water regime stable periods, and the boundary optimization is independently performed for each period; the second-order Taylor expansion approximation of the energy function is used to directly obtain the optimal boundary adjustment amount by solving a linear equation set; a buffer zone with a width dynamically changing with the flow standard deviation and predictability is set at the boundary.

[0339] The buffer zone mechanism includes that the buffer zone width is composed of three parts of a basic width, a flow fluctuation term and a predictability term; the nodes in the buffer zone are not fixedly attributed to a certain partition, but are dynamically allocated according to the real-time load of each partition and the node joining cost; the temporary attribution of the nodes is determined by minimizing the sum of the load and the cost to realize the load balancing and oscillation suppression between the partitions.

[0340] Specifically, the following steps are included:

[0341] Step S44a1, extracting boundary oscillation features;

[0342] In this embodiment, an oscillation index is defined to quantify the state switching frequency of the boundary nodes, and specifically, the oscillation index calculation formula is: O(b)=(Σ|sign(Q(t))-sign(Q(t-1))|) / T;

[0343] Wherein, O(b) represents the oscillation index of the boundary b; Q(t) represents the flow through the boundary at t, a positive value represents water supply, and a negative value represents water receiving; sign represents a sign function; and T represents an observation time window, which is usually 1 hour.

[0344] The formula for predicting the oscillation risk is: Risk(b) = 1 / (1+exp(-(a1 x AH+a2 x dQ / dt+a3 x N_neighbors)));

[0345] wherein, Risk(b) represents the oscillation risk of the boundary b, ranging from 0 to 1; AH represents the water level difference on both sides of the boundary, in meters; a1 represents the water level difference coefficient, taking a value of 0.5-1.5; a2 represents the flow rate change rate coefficient, taking a value of 0.1-0.3; a3 represents the neighbor influence coefficient, taking a value of 0.2-0.4; and N_neighbors represents the number of adjacent nodes.

[0346] Step S44a2, performing a preventive boundary adjustment;

[0347] In this embodiment, the boundary position is adjusted in advance when the oscillation risk exceeds the threshold value, and specifically, the boundary adjustment formula is: b_new = b_old + δ x grad(Risk) / ||grad(Risk)||;

[0348] wherein, b_new represents the new boundary position coordinate; b_old represents the original boundary position coordinate; δ represents the adjustment step, taking a value of 0.5-2.0 kilometers; grad(Risk) represents the risk gradient vector; and ||·|| represents the vector modulus.

[0349] Step S44a3, defining an oscillation energy function;

[0350] In this embodiment, a global oscillation energy function is constructed to evaluate the overall oscillation level, and specifically: E_total = Σ(O(b) 2 x V(b))+λ x Σ(Dist(b_new, b_old));

[0351] wherein, E_total represents the total oscillation energy; V(b) represents the daily average water exchange of the boundary b, in ten thousand cubic meters; λ represents the boundary stability weight, taking a value of 0.1-0.5; and Dist represents the Euclidean distance function.

[0352] Step S44a4, performing gradient descent boundary optimization;

[0353] In this embodiment, the oscillation energy is minimized through iterative optimization, and specifically, the iteration formula is: B(k+1) = B(k)-η x ∇E_total+β x (B(k)-B(k-1));

[0354] wherein, B(k) represents the boundary set of the kth iteration; η represents the learning rate, taking a value of 0.01-0.1; ∇E_total represents the energy gradient; and β represents the momentum coefficient, taking a value of 0.5-0.9.

[0355] Step S44a5, identify the water regime mode and segment;

[0356] In the embodiment, 24 hours is divided into multiple water regime stable segments, and the segmentation criterion is |dH / dt|<threshold_H and |dQ / dt|<threshold_Q.

[0357] wherein dH / dt represents the water level change rate, the threshold threshold_H is 0.1 meter / hour; dQ / dt represents the flow rate change rate, and the threshold threshold_Q is 10 cubic meters / second. 2 .

[0358] Step S44a6, perform fast approximate optimization.

[0359] In the embodiment, the second-order Taylor expansion is used to approximate the energy function E_approx=E0+g'×ΔB+0.5×ΔB'×H×ΔB.

[0360] wherein E0 represents the current energy value; g represents the gradient vector; H represents the Hessian matrix; and ΔB represents the boundary adjustment amount.

[0361] The optimal adjustment amount is obtained by solving a linear equation set: ΔB*=-H^(-1)×g.

[0362] Step S44a7, establish a boundary buffer mechanism.

[0363] In the embodiment, a buffer with a dynamic width is set for the boundary, and specifically, the buffer width is calculated as follows:

[0364] Buffer(b,t)=β0+β1×σ(Q)+β2×predictability(t)

[0365] wherein Buffer(b,t) represents the buffer width of the boundary b at time t, and the unit is kilometer; β0 represents the basic width, and the value is 0.5-1.0 kilometer; β1 represents the flow fluctuation coefficient, and the value is 0.1-0.3; σ(Q) represents the flow standard deviation; β2 represents the predictability coefficient, and the value is 0.2-0.5; predictability(t) represents the predictability index based on historical data, and the range is 0 to 1.

[0366] The dynamic allocation formula of the nodes in the buffer is Assign(v)=argmin(Load(partition)+Cost(v,partition)).

[0367] Wherein, Assign(v) represents the partition ownership of node v; Load(partition) represents the current load of the partition; Cost(v, partition) represents the cost of node v joining the partition.

[0368] For example, at the boundary of two partitions of a certain water network system, the original oscillation frequency is 12 times per hour. After applying the method, by setting a 1.2 km wide dynamic buffer zone, the oscillation frequency is reduced to 2 times per hour, and the water volume statistical error is reduced from 30% to 3% through dynamic allocation of 5 nodes in the buffer zone according to real-time load.

[0369] Through the above steps, the boundary oscillation frequency is reduced by more than 85%, the water volume statistical error is controlled within 5%, and the system stability is significantly improved.

[0370] According to an aspect of the present application, the determination process of the water network feature sparse graph is specifically:

[0371] The edge density of the graph is defined as the ratio of the actual number of edges to the maximum possible number of edges, and the calculation formula is: p_graph=2E / (V(V-1)); wherein E is the number of edges in the graph, and V is the number of nodes in the graph;

[0372] When the edge density p_graph is less than the sparse threshold value θ_sparse, the subgraph is determined to be a water network feature sparse graph, wherein the calculation formula of the sparse threshold value θ_sparse is: θ_sparse=1 / (2×√V);

[0373] The threshold value is adaptively adjusted according to the number of nodes, and the more the number of nodes, the stricter the sparse standard;

[0374] For the subgraph formed after cutting, if its edge density is between 0.05 and 0.3, it is determined to be a reasonable sparse graph; if the edge density is less than 0.05, it is determined to be an excessively sparse graph and needs to be merged with adjacent subgraphs; if the edge density is greater than 0.3, it is determined to be a dense graph and needs to be further cut.

[0375] In a certain embodiment, a subgraph containing 50 nodes has a maximum possible number of edges of 50*49 / 2=1225, an actual number of edges of 120, and an edge density of 120 / 1225≈0.098. The sparse threshold value is 1 / (2*sqrt(50))≈0.071. Since 0.098>0.071 and 0.098∈[0.05,0.3], it is determined to be a reasonable water network feature sparse graph.

[0376] According to an aspect of the present application, the determination process of the cutting threshold value in step S31 is specifically:

[0377] Based on the number of eigenvalues F and the total number of nodes N, the calculation formula of the cutting threshold τ cut is: τ cut = ┗α × ln(N) × √F┛;

[0378] Wherein, α is the adjustment coefficient, the value range is 0.8 to 1.2, ┗┛ represents rounding down;

[0379] When the number of eigenvalues F=8, the cutting threshold calculation of the total number of nodes N in different ranges:

[0380] N∈[50,100]: τ cut = ┗1.0 × ln(75) × √8┛ = ┗12.2┛ = 12;

[0381] N∈[100,500]: τ cut = ┗1.0 × ln(300) × √8┛ = ┗16.1┛ = 16;

[0382] N∈[500,1000]: τ cut = ┗0.9 × ln(750) × √8┛ = ┗16.9┛ = 16;

[0383] N>1000: τ cut = ┗0.8 × ln(N) × √8┛, the maximum is not more than 20;

[0384] The physical meaning of the cutting threshold is: when the number of natural neighbors of a node exceeds the threshold, it means that the node is in the core position of the local high-density area, and the edge connected by the node should be reserved as the cutting edge to maintain the integrity of the core area.

[0385] In an embodiment, the study area contains 235 water network nodes and 8 characteristic attributes, and the cutting threshold is calculated as: τ cut = ┗1.0 × ln(235) × √8┛ = ┗15.4┛ = 15, that is, when the number of natural neighbors of a node is greater than 15, the edge connected by the node and its neighbors is marked as a cutting edge.

[0386] According to an aspect of the present application, the determination process of the shared nearest neighbor number in step S43 is specifically:

[0387] The shared nearest neighbor number k share is dynamically determined based on the subgraph size, and the calculation formula is:

[0388] k share = max(3, min(┗√(n_i × n_j)┛, 20));

[0389] Wherein, n_i and n_j are the number of nodes of two subgraphs to be compared;

[0390] For the shared nearest neighbor number of different size subgraph pairs:

[0391] Small subgraph pair (n_i, n_j < 20): k_share = 3 to 5;

[0392] Medium subgraph pair (n_i, n_j between 20 and 100): k_share = 6 to 10;

[0393] Large subgraph pair (n_i, n_j > 100): k_share = 10 to 20;

[0394] Subgraph pair with large size difference: use a smaller value of min(k_i, k_j), where k_i = ┗√n_i┛ and k_j = ┗√n_j┛;

[0395] The identification process of the shared nearest neighbor is as follows: find the k_share nearest neighbors of the center points of the two subgraphs respectively, calculate the intersection size of the two nearest neighbor sets, and the number of intersection elements is the number of shared nearest neighbors SNN(i, j).

[0396] In an embodiment, subgraph A contains 36 nodes, subgraph B contains 64 nodes, k_share = ┗√(36 x 64)┛ = ┗48┛ = 20 (take the upper limit), and the 20 nearest neighbors of the center point of subgraph A coincide with the 20 nearest neighbors of the center point of subgraph B. Therefore, SNN(A, B) = 8.

[0397] According to an aspect of the present application, the calculation process of the distance of the dispersed node to the subgraph in step S32 is as follows:

[0398] The distance of the dispersed node v to the subgraph G is defined as the weighted minimum value of the three distances:

[0399] d(v, G) = min(w_1 x d_center(v, G), w_2 x d_boundary(v, G), w_3 x d_average(v, G));

[0400] where d_center(v, G) is the composite distance of node v to the center point of subgraph G;

[0401] d_boundary(v, G) = min{d(v, u) | u ∈ boundary(G)}, which is the minimum composite distance of node v to the boundary node of subgraph G, and the boundary node is defined as the node connected to other subgraphs or dispersed nodes;

[0402] d_average(v, G) = (1 / |G|) x ∑_{u∈G} d(v, u), which is the average composite distance of node v to all nodes of subgraph G;

[0403] The weight settings are: w_1=0.5 (giving priority to the center distance), w_2=0.3 (giving secondary consideration to the boundary distance), and w_3=0.2 (giving auxiliary consideration to the average distance);

[0404] The assignment rule is: assign a dispersed node v to the subgraph with the minimum distance, i.e., assign(v)=argmin_Gd(v,G);

[0405] In an embodiment, the distance calculation of a dispersed node P to three subgraphs is as follows:

[0406] To subgraph G1: d_center=15.2, d_boundary=8.5, d_average=18.3, weighted distance=min(7.6, 2.55, 3.66)=2.55;

[0407] To subgraph G2: d_center=22.1, d_boundary=12.3, d_average=25.6, weighted distance=min(11.05, 3.69, 5.12)=3.69;

[0408] To subgraph G3: d_center=18.5, d_boundary=10.2, d_average=20.1, weighted distance=min(9.25, 3.06, 4.02)=3.06;

[0409] Node P is assigned to G1 (with the minimum distance of 2.55).

[0410] According to an aspect of the present application, the determination process of the number of clusters N in step S44 is as follows:

[0411] The optimal number of clusters N is automatically determined by using the silhouette coefficient method, and the calculation process is as follows:

[0412] For each possible number of clusters n (n from 2 to √M), the average silhouette coefficient is calculated:

[0413] S(n)=(1 / N_total)×Σ_is(i); where s(i)=(b(i)-a(i)) / max(a(i),b(i));

[0414] a(i) is the average distance of node i to other nodes in the same cluster, and b(i) is the average distance of node i to the nearest other cluster;

[0415] The n that makes the silhouette coefficient maximum is selected as the final number of clusters N: N=argmax_nS(n);

[0416] At the same time, set the constraint conditions: minimum cluster number: N_min = max(2, ┗M / 10┛); maximum cluster number: N_max = min(┗M / 2┛, 20); contour coefficient threshold: S(N) > 0.3, otherwise continue to adjust;

[0417] When the automatically determined N value does not meet the management requirements, manual intervention can be performed to set the expected cluster number N_expect, and the algorithm searches for the optimal value in the range of [N_expect-2, N_expect+2].

[0418] In an embodiment, the initial division obtains M = 45 subgraphs, and the contour coefficients of different cluster numbers are calculated: S(5) = 0.42 when n = 5, S(6) = 0.51 when n = 6, S(7) = 0.58 when n = 7, S(8) = 0.55 when n = 8, and S(9) = 0.48 when n = 9, so it is determined that N = 7 is the optimal cluster number.

[0419] According to an aspect of the present application, the processing process of the isolated node and the node without natural neighbors is as follows:

[0420] Identify special nodes:

[0421] Isolated node: a node whose composite distance with all other nodes is greater than the distance threshold D_isolate = 3 × D_mean, where D_mean is the average distance of all node pairs;

[0422] Node without natural neighbors: a node without mutual nearest neighbor relationship within the range of the effective number of neighbors;

[0423] The processing strategy includes:

[0424] For the isolated node, an extended search radius method is used: gradually increase the search radius r = r_0 × (1 + 0.2 × k), k is the iteration number, until at least one candidate neighbor is found; calculate the weighted distance to the subgraph to which all candidate neighbors belong, and the weight considers the subgraph size (high weight for large subgraph) and functional similarity; assign the isolated node to the subgraph with the smallest weighted distance;

[0425] For the node without natural neighbors, a relaxed mutual neighbor condition is used: the mutual nearest neighbor condition is relaxed to a one-way nearest neighbor, that is, if node A is one of the k nearest neighbors of node B (k ≤ 2 × effective number of neighbors), a weak connection from A to B is established; construct an extended neighborhood graph based on the weak connection; search for natural neighbors in the extended neighborhood graph;

[0426] If there are still nodes that cannot be assigned after processing, create an "edge zone" subgraph to accommodate these nodes, and preferentially merge the edge zone subgraph with the main subgraph with the most similar function when performing final clustering.

[0427] In one embodiment, the minimum composite distance of node Q to other nodes is 85.3, while the distance mean D_mean = 22.1, and the threshold D_isolate = 66.3, so Q is determined as an isolated node; the extended search finds 3 candidate subgraphs, the weighted distance to G1 is 72.5 (subgraph size 15, functional similarity 0.3), the weighted distance to G2 is 68.2 (subgraph size 25, functional similarity 0.4), and the weighted distance to G3 is 70.1 (subgraph size 20, functional similarity 0.35), finally Q is assigned to G2.

[0428] According to one aspect of the present application, the evaluation process of the water network system target partition effect is specifically:

[0429] A multi-dimensional evaluation index system is constructed:

[0430] An internal compactness index IC = (1 / N) x Σ_k (1 / n_k^2) x Σ_{i,j∈C_k} sim(i,j);

[0431] Wherein, N is the number of clusters, n_k is the number of nodes in the kth partition, and sim(i,j) is the node similarity;

[0432] An external separation index ES = min_{k≠l} d(C_k, C_l) / max_k diameter(C_k);

[0433] Wherein, d(C_k, C_l) is the distance between partitions, and diameter(C_k) is the partition diameter;

[0434] A functional consistency index FC = (1 / N) x Σ_k var(F_k) / var(F_total);

[0435] Wherein, var(F_k) is the variance of functional features in the kth partition, var(F_total) is the global functional feature variance, and the smaller FC indicates the more consistent functions within the partition;

[0436] A management feasibility index MA = (1 / N) x Σ_k (1-split(k));

[0437] Wherein, split(k) is the ratio of the number of management units crossed by the kth partition to the number of nodes in the partition;

[0438] The comprehensive evaluation score Score = w IC x IC + w ES x ES + w FC x (1-FC) + w MA x MA; wherein the weights w IC = 0.3, w ES = 0.2, w FC = 0.3, and w MA = 0.2; the evaluation criteria are: Score > 0.7 is excellent, 0.5-0.7 is good, 0.3-0.5 is qualified, and < 0.3 needs to be re-zoned.

[0439] In an embodiment, after a certain watershed is divided into 7 targeted partitions, the IC = 0.72 (close within the partition), the ES = 0.65 (good separation between partitions), the FC = 0.28 (high functional consistency), and the MA = 0.81 (clear management boundary) are calculated, the comprehensive score Score = 0.3 x 0.72 + 0.2 x 0.65 + 0.3 x 0.72 + 0.2 x 0.81 = 0.724 is obtained, and the evaluation is excellent, which is 42% higher than the traditional K-means method (Score = 0.51).

[0440] According to another aspect of the present application, a water network system targeted partitioning system is provided, characterized in that it comprises:

[0441] at least one processor; and

[0442] a memory in communication connection with the at least one processor; wherein,

[0443] The memory stores instructions executable by the processor, and the instructions are executed by the processor to implement the water network system targeted partitioning method of any one of the above.

[0444] The above describes the preferred embodiments of the present application, but the present application is not limited to the specific details in the above embodiments, and various equivalent transformations can be made to the technical solutions of the present application within the technical concept of the present application, and these equivalent transformations all belong to the protection scope of the present application.

Claims

1. A method for targeting sub-zones of a water network system, characterized in that, Comprising the following steps: Step S1, collecting water network data of the study area, identifying water network nodes and extracting data of each water network node, and reducing dimension of the data to obtain an eigenvalue matrix of each water network node; Step S2, calculating a composite distance between each two nodes based on the eigenvalue matrix of each water network node, setting a nearest neighbor number based on the number of eigenvalues, screening water network nodes that are nearest neighbors of each other, and constructing a hierarchical graph with nodes as vertices and nearest neighbor relationships as edges to obtain a water network characteristic neighborhood graph; Step S3, identifying cut edges based on the neighborhood relationship between each node, and dividing the water network characteristic field graph into a plurality of water network characteristic sparse graphs; Step S4, determining a center point of each water network characteristic sparse graph based on the local density of the water network node, calculating the similarity between each water network characteristic sparse graph, and clustering the water network characteristic sparse graphs to obtain a plurality of clustering centers, and dividing the study area into a plurality of water network system targeted partitions; The step S1 is further: Step S11, collecting water network data of the study area, including: water network topological structure data, hydraulic data, geographical data, historical drought data and historical flood data; Step S12, identifying water network nodes and extracting data of each water network node, the water network nodes including: basin nodes, engineering nodes, monitoring nodes and geographical feature nodes; Step S13, sequentially reducing dimension of data of all water network nodes to obtain an eigenvalue matrix of each water network node; The step S4 is further: Step S41, calculating the local density of each water network node, and determining the center point of each water network characteristic sparse graph based on the local density of the water network node; Step S42, sequentially calculating the distance between the center points of each two water network characteristic sparse graphs, which is the distance between the two water network characteristic sparse graphs; Step S43, setting a shared nearest neighbor number, taking the ratio of the number of shared natural nearest neighbors to the distance between the two water network characteristic sparse graphs as the similarity between the two water network characteristic sparse graphs, and sequentially calculating the similarity between all water network characteristic sparse graphs; Step S44, screening out the two water network characteristic sparse graphs with the highest similarity, and clustering them into the same water network partition to obtain a clustering label, and taking the clustering label as a water network characteristic sparse graph to calculate the similarity between all water network characteristic sparse graphs again, and repeating the iteration until all water network characteristic sparse graphs belong to a certain clustering label, obtaining N clustering labels, dividing the study area into N water network system targeted partitions, and N is a positive integer less than M.

2. The water network system targeting zonation method of claim 1, wherein, The step S13 is further: Step S13a, extracting a plurality of water network node attributes based on the data of the water network nodes, and calculating the correlation coefficient between all water network node attributes, and classifying water network node attributes with a correlation coefficient greater than a threshold value into the same water network node characteristic attribute to obtain a water network node characteristic attribute set; Step S13b, calculating the numerical value of each water network node characteristic attribute by using an entropy weight method, and constructing an eigenvalue matrix of each water network node.

3. The water network system targeting zonation method of claim 1, wherein, The step S2 is further: Step S21, calculating a composite distance between each two nodes based on the eigenvalue matrix of each water network node; Step S22, setting an initial number of neighbors value based on the number of feature values, and screening the neighbor nodes of each water network node, gradually increasing the size of the number of neighbors value, and screening the neighbor nodes of each water network node based on the current number of neighbors value, until all water network nodes meet at least one neighbor node, at which time the number of neighbors value is the effective number of neighbors value; Step S23, screening the neighbor nodes of each water network node based on the effective number of neighbors value, and constructing a hierarchical graph with nodes as vertices and nearest neighbor relationships as edges to obtain a water network feature neighborhood graph.

4. The water network system targeting zonation method of claim 3, wherein, The step S21 is further: Step S21a, identifying the basins in the study area as basin nodes, the water conservancy hubs, flood drainage channels and storage projects as engineering nodes, the hydrological sites as monitoring nodes, and the river junctions, drought areas and flood-prone areas as geographical feature nodes; Step S21b, calculating the geographical distance, hydraulic distance, climate distance and management distance between each two nodes in turn; Step S21c, determining the weight of each distance and calculating the composite distance between each two nodes.

5. The water network system targeting zonation method of claim 3, wherein, The step S23 is further: Step S23a, constructing a hierarchical neighborhood graph of basin nodes, engineering nodes and geographical feature nodes with nodes as vertices and neighbor relationships as edges; Step S23b, representing water-related, water transfer-related and climate-related with blue edges, red edges and yellow edges respectively, representing high flood risk area nodes with red, orange nodes with medium flood risk area nodes, and yellow nodes with drought area nodes, and representing climate correlation with edge thickness to obtain a water network feature neighborhood graph.

6. The water network system targeting zonation method of claim 1, wherein, The step S3 is further: Step S31, setting a cutting threshold based on the number of water network node feature values, calculating the number of neighbors of each water network node, marking the water network nodes with a number of neighbors greater than the cutting threshold, and retaining the edges of the marked water network nodes and all their neighbor nodes as cutting edges, and dividing the water network feature neighborhood graph into M subgraphs and several scattered water network nodes based on the cutting edges, M being a positive integer greater than 10; Step S32, calculating the distance between each scattered water network node and M subgraphs in turn, and dividing the scattered water network nodes into the nearest sub-regions to obtain M water network feature sparse graphs.

7. A water network system targeted zoning system characterized in that, It includes: At least one processor; And The memory is in communication connection with the at least one processor; wherein The memory stores instructions executable by the processor, and the instructions are used to be executed by the processor to realize the water network system targeted partitioning method of any one of claims 1 to 6.

Images