Water network system networking method and system

Abstract

The present invention discloses a method and system for networking a water network system, collects digital elevation grid data of a study area, identifies rivers in the study area, divides the study area into several subnets, each of which contains several sub-basins; constructs evaluation models for each subnet and optimizes model parameters and hyperparameters, calculates the comprehensive evaluation results of the water resources-ecological environment-economic and social synergistic state of each subnet; sorts the characteristic values ​​and evaluation index values ​​of each subnet, constructs a set of feasible networking schemes, constructs complementary indicators, constructs a set of matching networking schemes and calculates the cost of engineering construction, constructs an optimization scheduling model, calculates the optimal scheduling mode and networking benefits of each networking scheme; sets a networking cost range, determines the networking scheme and inputs a pre-constructed water network system optimization networking decision model, and calculates the networking scheme of the water network system. The present invention can coordinate the overall benefits of the water network system and the local benefits of the subnets, and take into account the relationship between networking costs and comprehensive benefits.

Description

Technical Field

[0001] The invention relates to a water network system networking method. Background Art

[0002] The water network system is a comprehensive system based on natural rivers and lakes, with diversion and drainage projects as channels, storage projects as nodes, and intelligent regulation as a means. It integrates the functions of optimizing water resources allocation, flood control and disaster reduction in river basins, and protection of water ecosystems. It is an effective measure to solve the uneven spatial distribution of water resources, improve the water resources guarantee rate in the receiving areas, alleviate the contradiction between water supply and demand in water-scarce areas, and achieve rational allocation of water resources. It is an important way to promote economic development and comprehensive development and utilization of water resources in water-scarce areas.

[0003] The water network system has a multi-level characteristic, and it is difficult to build a multi-level water network directly in one go. It is necessary to connect smaller-level networks under the large network structure of the water network system to further improve the system's water resources-ecological environment-economic and social coordinated balance. The networking process involves subnetwork division and its feature identification and quantitative description. When evaluating, how to consider the multi-dimensional attribute characteristics of each subnetwork in the global water resources-ecological environment-economic and social balance evaluation model of the water network system, clarify its own water balance status, so as to guide the complementary advantages between different subnetworks during networking, and how to define the scope of networking to take into account the cost-effectiveness of networking. These problems are all problems that cannot be answered or solved by existing technologies.

[0004] The present invention proposes a water network system networking method and system to solve the above-mentioned problems currently existing, realize the complementary advantages between subnets under the overall structure of the water network system, and propose a water network system networking solution that coordinates the overall benefits of the water network system and the local benefits of the subnets, and takes into account the relationship between networking costs and comprehensive benefits. Summary of the invention

[0005] The purpose of the invention is to provide a water network system networking method to solve the above problems existing in the prior art. On the other hand, a water network system networking system is provided.

[0006] Technical solution: A method for networking a water network system, comprising the following steps:

[0007] Step S1, collect digital elevation raster data of the study area, identify rivers in the study area, divide the study area into several sub-basins, calculate the similarity between the sub-basins, classify similar sub-basins into the same subnet, and divide the study area into several subnets;

[0008] Step S2: construct evaluation models for each subnet respectively and optimize model parameters and hyperparameters, calculate characteristic values ​​of water resources supply, water disaster security and water ecological health for each sub-basin in each subnet respectively and input them into the corresponding subnet evaluation model to obtain comprehensive evaluation results of the coordinated state of water resources-ecological environment-economy and society for each subnet;

[0009] Step S3, sort the characteristic values ​​and evaluation index values ​​of each subnet, construct a feasible networking solution set based on the subnet sorting, construct a complementary index, calculate the matching degree between each subnet and set a matching threshold, obtain a matching networking solution set and calculate the engineering construction cost of each matching networking solution, construct an optimization scheduling model, and calculate the optimal scheduling mode of each networking solution and its corresponding subnet networking benefit;

[0010] Step S4, setting the networking cost range, determining the networking plan based on the subnet networking benefits and networking costs of the networking plan, and inputting the pre-built water network system optimization networking decision model to calculate the optimal plan, i.e., the water network system networking plan.

[0011] According to one aspect of the present application, step S1 further comprises:

[0012] Step S11, collect digital elevation grid data of the study area, identify rivers in the study area, identify grid anomalies, and divide the study area into n sub-basins, where n is a positive integer;

[0013] Step S12: Calculate the water resource supply, water disaster protection and water ecological health index values ​​of each sub-basin respectively, calculate the similarity between the sub-basins based on the index values ​​of the sub-basins, classify similar sub-basins into one subnet, and obtain m subnets, that is, divide the study area into m subnets, where m is a positive integer.

[0014] According to one aspect of the present application, the step S11 is further:

[0015] Step S11a, read the digital elevation grid data of the study area, construct a monitoring window, monitor the height relationship of adjacent windows, obtain the fastest descent vector, and determine the descent direction of each grid, that is, the water flow direction;

[0016] Step S11b, each grid extends the vector along its descending direction until it intersects with the river, then the grid belongs to the first river intersecting with it, and n sub-basins are obtained, where n is a positive integer;

[0017] Step S11c, scanning all grids, identifying grids that are different from all surrounding grids and belong to the same river as abnormal grids, calculating the descending gradients of the abnormal grid and its surrounding 8 grids respectively, and the river to which the grid with the largest descending gradient belongs is the river to which the abnormal grid belongs;

[0018] Step S11d: Divide the study area into n sub-basins.

[0019] According to one aspect of the present application, step S12 further comprises:

[0020] Step S12a, respectively set a water resource supply index, b water disaster security index and c water ecological health index, calculate all index values ​​of all sub-basins in sequence, project each sub-basin into a point coordinate in a+b+c dimensional space based on the index value, calculate the Euclidean distance between each point, identify similar sub-basins and classify them into the same category, a, b, c are positive integers;

[0021] Step S12b: For all similar sub-basins of the same type, extract the grids from the most downstream to the most upstream, calculate the path similarity between the two sub-basins based on the distance between the grids in each sub-basin, classify the similar sub-basins into one subnet, and obtain m subnets, that is, divide the study area into m subnets, where m is a positive integer.

[0022] According to one aspect of the present application, step S2 further comprises:

[0023] Step S21, constructing a deep reinforcement learning model based on dual DQN optimization;

[0024] Step S22: weight each sub-basin based on the river length and flow rate of the sub-basin in each sub-network, and calculate the characteristic values ​​of a water resource supply, b water disaster security and c water ecological health of each sub-network;

[0025] Step S23: Based on the subnet feature data, train the deep reinforcement learning model, construct a mapping relationship between the a+b+c feature values ​​of each subnet and the weight of the subnet evaluation index, and optimize the parameters and hyperparameters of the deep reinforcement learning model;

[0026] Step S24, extract the characteristic values ​​of a water resource supply, b water disaster protection and c water ecological health of each subnet and input them into the optimized deep reinforcement learning model to obtain the weight of the subnet evaluation index, so as to configure the parameters of the respective subnet water resources-ecological environment-economic and social coordinated status evaluation model, and extract and input the evaluation index values ​​of each subnet to obtain the comprehensive evaluation results of the status of each subnet.

[0027] According to one aspect of the present application, step S23 is further:

[0028] Step S23a: for an existing subnet, based on its a+b+c characteristic values, use an expert scoring method to determine the weight of each evaluation index of the subnet;

[0029] Step S23b: training a deep reinforcement learning model to establish a mapping relationship between a+b+c feature values ​​of each subnet and the weight of the subnet evaluation index;

[0030] Step S23c: Taking the minimum cumulative error between the deep reinforcement learning model simulation results and the expert scoring results as the objective function, construct a parameter and hyperparameter optimization model for the model, and solve the model to obtain the optimal parameter and hyperparameter solution.

[0031] According to one aspect of the present application, step S3 further comprises:

[0032] Step S31, respectively sort the characteristic values ​​and evaluation index values ​​of each subnet in forward and reverse order, obtain the forward and reverse order of the attributes of each subnet, sort each subnet in turn in forward order, and select the reverse order of other subnets that are consistent with the forward order of the selected subnet as the networking scheme, and obtain a set of feasible networking schemes;

[0033] Step S32: for each feasible networking scheme, construct a complementarity index, calculate the matching degree of all subnets in the feasible networking scheme, set a matching degree threshold, calculate the matching degree between each subnet in all feasible networking schemes in turn, extract the feasible networking schemes with matching degrees greater than the matching degree threshold to obtain a matching networking scheme set;

[0034] Step S33, calculate the engineering construction cost of each matching networking solution, input the matching networking solution into the pre-built optimization scheduling model, calculate the optimal scheduling mode and subnet networking benefits corresponding to each networking solution, and calculate the scheduling cost according to the optimal scheduling mode.

[0035] According to one aspect of the present application, step S33 is further:

[0036] Step S33a, constructing an optimization scheduling model;

[0037] Step S33b, calculating the engineering cost of each networking solution in the matching networking solution set;

[0038] Step S33c: input the matching networking scheme into the optimization scheduling model to obtain the optimal scheduling mode, calculate the scheduling cost of each matching networking scheme according to the optimal scheduling mode, and simultaneously calculate the subnet networking benefits after the optimized scheduling of each networking scheme to obtain the improvement value of the comprehensive evaluation of each subnet.

[0039] According to one aspect of the present application, step S4 is further:

[0040] Step S41: Calculate the total cost of networking based on the networking engineering construction cost and the scheduling cost, and determine a set of networkable solutions from a set of matching networking solutions with the total cost of networking as a constraint;

[0041] Step S42, constructing a water network system optimization networking decision model, the objective function is: the water network system networking benefit efficiency is maximized, and the water network system networking benefit efficiency is calculated by the tangent slope of the networking benefit to networking cost growth curve;

[0042] Step S43, input the networking solutions in the networking solution set into the water network system optimization networking decision model respectively, perform decision optimization, and obtain the optimal solution as the water network system networking solution.

[0043] According to another aspect of the present application, a water network system networking system is provided, comprising:

[0044] at least one processor; and

[0045] a memory communicatively connected to at least one of the processors; wherein,

[0046] The memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement the water network system networking method described in any of the above technical solutions.

[0047] Beneficial effects: The water network system networking method can coordinate the overall benefits of the water network system and the local benefits of the subnet, and take into account the relationship between networking costs and comprehensive benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of the present invention.

[0049] Figure 2 It is a flow chart of step S1 of the present invention.

[0050] Figure 3 It is a flow chart of step S2 of the present invention.

[0051] Figure 4 It is a flow chart of step S3 of the present invention.

[0052] Figure 5 It is a flow chart of step S4 of the present invention. DETAILED DESCRIPTION

[0053] like Figure 1 According to one aspect of the present application, a water network system networking method is provided, characterized in that it includes the following steps:

[0054] Step S1, collect digital elevation raster data of the study area, identify rivers in the study area, divide the study area into several sub-basins, calculate the similarity between the sub-basins, classify similar sub-basins into the same subnet, and divide the study area into several subnets;

[0055] Step S2: construct evaluation models for each subnet respectively and optimize model parameters and hyperparameters, calculate characteristic values ​​of water resources supply, water disaster security and water ecological health for each sub-basin in each subnet respectively and input them into the corresponding subnet evaluation model to obtain comprehensive evaluation results of the coordinated state of water resources-ecological environment-economy and society for each subnet;

[0056] Step S3, sort the characteristic values ​​and evaluation index values ​​of each subnet, construct a feasible networking solution set based on the subnet sorting, construct a complementary index, calculate the matching degree between each subnet and set a matching threshold, obtain a matching networking solution set and calculate the engineering construction cost of each matching networking solution, construct an optimization scheduling model, and calculate the optimal scheduling mode of each networking solution and its corresponding subnet networking benefit;

[0057] Step S4, setting the networking cost range, determining the networking plan based on the subnet networking benefits and networking costs of the networking plan, and inputting the pre-built water network system optimization networking decision model to calculate the optimal plan, i.e., the water network system networking plan.

[0058] Several sub-basins, i.e. sub-networks, are then divided into sub-basins to analyze their water resources supply, water disaster protection, and water ecological health. Indicators are formulated for these three goals, and the corresponding current internal conditions of the basin are determined based on the indicator values.

[0059] Then, based on the internal conditions of each subnet, adjacent subnets are selected for matching, and complementary matching is performed based on the satisfaction of each subnet in terms of the three goals of water resource supply, water disaster protection, and water ecological health, as well as the size of the gap, to construct a set of networking solutions, and select the optimal solution based on economic benefits to network the water network system. According to one aspect of the present application, step S1 is further:

[0060] Step S11, collect digital elevation grid data of the study area, identify rivers in the study area, identify grid anomalies, and divide the study area into n sub-basins, where n is a positive integer;

[0061] Digital elevation grids provide high-precision terrain data, which enables the structure and morphology of river networks to be accurately identified. Compared with traditional manual measurements or low-resolution terrain data, digital elevation grids can reflect the subtle ups and downs of the surface in more detail. By analyzing elevation data, the hydrological process of the basin can be considered, and river flow, runoff paths and hydrological characteristics can be better simulated.

[0062] Step S12: Calculate the water resource supply, water disaster protection and water ecological health index values ​​of each sub-basin respectively, calculate the similarity between the sub-basins based on the index values ​​of the sub-basins, classify similar sub-basins into one subnet, and obtain m subnets, that is, divide the study area into m subnets, where m is a positive integer.

[0063] According to one aspect of the present application, the step S11 is further:

[0064] Step S11a, read the digital elevation grid data of the study area, construct a monitoring window, monitor the height relationship of adjacent windows, obtain the fastest descent vector, and determine the descent direction of each grid, that is, the water flow direction;

[0065] The steepest descent vector, that is, the direction from the high point in space to the low point, is also the direction of water flow;

[0066] Step S11b, each grid extends the vector along its descending direction until it intersects with the river, then the grid belongs to the first river intersecting with it, and n sub-basins are obtained, where n is a positive integer;

[0067] Step S11c, scanning all grids, identifying grids that are different from all surrounding grids and belong to the same river as abnormal grids, calculating the descending gradients of the abnormal grid and its surrounding 8 grids respectively, and the river to which the grid with the largest descending gradient belongs is the river to which the abnormal grid belongs;

[0068] Step S11d: Divide the study area into n sub-basins.

[0069] In this embodiment, after the sub-basins are divided, the division of the geographical dimension is actually completed. The sub-networks are geographically continuous and there will be no disconnection. Now that the sub-basins have been divided, it is necessary to divide the sub-basins into different sub-networks based on the similarities between the sub-basins and divide the study area into several sub-networks.

[0070] According to one aspect of the present application, step S12 further comprises:

[0071] Step S12a, respectively set a water resource supply index, b water disaster security index and c water ecological health index, calculate all index values ​​of all sub-basins in sequence, project each sub-basin into a point coordinate in a+b+c dimensional space based on the index value, calculate the Euclidean distance between each point, identify similar sub-basins and classify them into the same category, a, b, c are positive integers;

[0072] Step S12b: For all similar sub-basins of the same type, extract the grids from the most downstream to the most upstream, calculate the path similarity between the two sub-basins based on the distance between the grids in each sub-basin, classify the similar sub-basins into one subnet, and obtain m subnets, that is, divide the study area into m subnets, where m is a positive integer.

[0073] According to one aspect of the present application, step S2 further comprises:

[0074] Step S21, constructing a deep reinforcement learning model based on dual DQN optimization;

[0075] Double DQN is a reinforcement learning algorithm that can solve the problem of overestimation of Q values ​​in traditional Q-learning algorithms. The basic idea of ​​double DQN is to use two Q networks to separately select actions and calculate Q values. Double DQN reduces the overestimation error of Q values ​​and enhances the stability and convergence speed of the model. Compared with traditional DQN and simple Q-Learning methods, double DQN not only solves the overestimation problem that may occur during the strategy update process, but also can more effectively capture dynamic changes in complex environments. Therefore, in this embodiment, double DQN is used to optimize the deep reinforcement learning model.

[0076] Step S22: weight each sub-basin based on the river length and flow rate of the sub-basin in each sub-network, and calculate the characteristic values ​​of a water resource supply, b water disaster security and c water ecological health of each sub-network;

[0077] Step S23: Based on the subnet feature data, train the deep reinforcement learning model, construct a mapping relationship between the a+b+c feature values ​​of each subnet and the weight of the subnet evaluation index, and optimize the parameters and hyperparameters of the deep reinforcement learning model;

[0078] Step S24, extract the characteristic values ​​of a water resource supply, b water disaster protection and c water ecological health of each subnet and input them into the optimized deep reinforcement learning model to obtain the weight of the subnet evaluation index, so as to configure the parameters of the respective subnet water resources-ecological environment-economic and social coordinated status evaluation model, and extract and input the evaluation index values ​​of each subnet to obtain the comprehensive evaluation results of the status of each subnet.

[0079] According to one aspect of the present application, step S23 is further:

[0080] Step S23a: for an existing subnet, based on its a+b+c characteristic values, use an expert scoring method to determine the weight of each evaluation index of the subnet;

[0081] Step S23b: training a deep reinforcement learning model to establish a mapping relationship between a+b+c feature values ​​of each subnet and the weight of the evaluation index of the subnet;

[0082] Step S23c: Taking the minimum cumulative error between the deep reinforcement learning model simulation results and the expert scoring results as the objective function, construct a parameter and hyperparameter optimization model for the model, and solve the model to obtain the optimal parameter and hyperparameter solution.

[0083] According to one aspect of the present application, step S3 further comprises:

[0084] Step S31, respectively sort the characteristic values ​​and evaluation index values ​​of each subnet in forward and reverse order, obtain the forward and reverse order of the attributes of each subnet, sort each subnet in turn in forward order, and select the reverse order of other subnets that are consistent with the forward order of the selected subnet as the networking scheme, and obtain a set of feasible networking schemes;

[0085] Step S32: for each feasible networking solution, construct a complementarity index, calculate the matching degree of all subnets in the feasible networking solution, set a matching degree threshold, calculate the matching degree between each subnet in all feasible networking solutions in turn, extract the feasible networking solutions with a matching degree greater than the matching degree threshold to obtain a matching networking solution set; in a certain embodiment, specifically:

[0086] For the i-th subnetwork, the sorting matrix of its various indicators is recorded as S i =[S i,1 ,S i,2 ,…,S i,k ,…S i,n ], k is the number of indicators and n is the number of indicators. Similarly, for the j-th subnetwork, the ranking matrix of each indicator is recorded as S j =[S j,1 ,S j,2 ,…,S j,k ,…S j,n ]

[0087] Then the complementarity index between the i-th subnet and the j-th subnet is:

[0088]

[0089] When the match is complete, H=0. The smaller H is, the higher the matching degree is. i and j are positive integers.

[0090] By setting a matching degree threshold, the matching degree between the two subnets is calculated respectively to determine whether they meet the matching networking scheme, and a matching networking scheme set is obtained.

[0091] Step S33, calculate the engineering construction cost of each matching networking solution, input the matching networking solution into the pre-built optimization scheduling model, calculate the optimal scheduling mode and subnet networking benefits corresponding to each networking solution, and calculate the scheduling cost according to the optimal scheduling mode.

[0092] According to one aspect of the present application, step S33 is further:

[0093] Step S33a, constructing an optimization scheduling model;

[0094] Step S33b, calculating the engineering cost of each networking solution in the matching networking solution set;

[0095] Step S33c: input the matching networking scheme into the optimization scheduling model to obtain the optimal scheduling mode, calculate the scheduling cost of each matching networking scheme according to the optimal scheduling mode, and simultaneously calculate the subnet networking benefits after the optimized scheduling of each networking scheme to obtain the improvement value of the comprehensive evaluation of each subnet.

[0096] According to one aspect of the present application, step S4 is further:

[0097] Step S41: Calculate the total cost of networking based on the networking engineering construction cost and the scheduling cost, and determine a set of networkable solutions from a set of matching networking solutions with the total cost of networking as a constraint;

[0098] Step S42, constructing a water network system optimization networking decision model, the objective function is: the water network system networking benefit efficiency is maximized, and the water network system networking benefit efficiency is calculated by the tangent slope of the networking benefit to networking cost growth curve;

[0099] Step S43, input the networking solutions in the networking solution set into the water network system optimization networking decision model respectively, perform decision optimization, and obtain the optimal solution as the water network system networking solution.

[0100] The present invention firstly identifies the rivers in the study area through the digital elevation raster data of the study area and divides the study area into several sub-basins, decomposes the study area at the geographical level, and then collects the respective water resources supply, water disaster security and water ecological health indicators according to the divided sub-areas, and calculates the equilibrium state of each sub-area to support the subsequent networking calculation;

[0101] After extracting all subnets that do not meet the equilibrium state, they are sorted according to the different strengths of water resources supply, water disaster security and water ecological health, and the order of water resources supply, water disaster security and water ecological health needs of each subnet is obtained from large to small and from small to large. Based on the complementarity between the subnets, a set of feasible networking solutions is obtained;

[0102] Then, a complementarity index is constructed, the matching degree of all subnets in the feasible networking scheme is calculated, the matching degree between each subnet in all feasible networking schemes is calculated, and the feasible networking schemes with matching degree greater than the matching degree threshold are extracted to obtain a set of matching networking schemes;

[0103] After obtaining the set of matching networking solutions, it is necessary to calculate the cost and benefits of each matching networking solution, construct a growth curve of networking benefits versus networking costs, and screen out the optimal networking solution based on this.

[0104] According to another aspect of the present application, a water network system networking system is provided, characterized in that it includes:

[0105] at least one processor; and

[0106] a memory communicatively connected to at least one of the processors; wherein,

[0107] The memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement any of the above-mentioned water network system networking methods.

[0108] The preferred embodiments of the present invention are described in detail above; however, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all belong to the protection scope of the present invention.

Claims

1. A method for networking a water network system, It is characterized in that The steps include: Step S1, collect digital elevation raster data of the study area, identify rivers in the study area, divide the study area into several sub-basins, calculate the similarity between the sub-basins, classify similar sub-basins into the same subnet, and divide the study area into several subnets; Step S2: construct evaluation models for each subnet respectively and optimize model parameters and hyperparameters, calculate characteristic values ​​of water resources supply, water disaster security and water ecological health for each sub-basin in each subnet respectively and input them into the corresponding subnet evaluation model to obtain comprehensive evaluation results of the coordinated state of water resources-ecological environment-economy and society for each subnet; Step S3, sort the characteristic values ​​and evaluation index values ​​of each subnet, construct a feasible networking solution set based on the subnet sorting, construct a complementary index, calculate the matching degree between each subnet and set a matching threshold, obtain a matching networking solution set and calculate the engineering construction cost of each matching networking solution, construct an optimization scheduling model, and calculate the optimal scheduling mode of each networking solution and its corresponding subnet networking benefit; Step S4, setting a networking cost range, determining a networking plan based on the subnet networking benefits and networking costs of the networking plan, and inputting a pre-built water network system optimization networking decision model to calculate the optimal plan, i.e., the water network system networking plan; The step S1 is further as follows: Step S11, collect digital elevation grid data of the study area, identify rivers in the study area, identify grid anomalies, and divide the study area into n sub-basins, where n is a positive integer; Step S12, respectively calculating the water resources supply, water disaster security and water ecological health index values ​​of each sub-basin, calculating the similarity between the sub-basins based on the index values ​​of the sub-basins, classifying similar sub-basins into one subnet, and obtaining m subnets, that is, dividing the study area into m subnets, where m is a positive integer; The step S11 is further as follows: Step S11a, read the digital elevation grid data of the study area, construct a monitoring window, monitor the height relationship of adjacent windows, obtain the fastest descent vector, and determine the descent direction of each grid, that is, the water flow direction; Step S11b, each grid extends the vector along its descending direction until it intersects with the river, then the grid belongs to the first river intersecting with it, and n sub-basins are obtained, where n is a positive integer; Step S11c, scanning all grids, identifying grids that are different from all surrounding grids and belong to the same river as abnormal grids, calculating the descending gradients of the abnormal grid and its surrounding 8 grids respectively, and the river to which the grid with the largest descending gradient belongs is the river to which the abnormal grid belongs; Step S11d, dividing the study area into n sub-basins; The step S12 is further as follows: Step S12a, respectively set a water resource supply index, b water disaster security index and c water ecological health index, calculate all index values ​​of all sub-basins in sequence, project each sub-basin into a point coordinate in a+b+c dimensional space based on the index value, calculate the Euclidean distance between each point, identify similar sub-basins and classify them into the same category, a, b, c are positive integers; Step S12b: for all similar sub-basins of the same type, extract the grids from the most downstream to the most upstream, calculate the path similarity between the two sub-basins based on the distance between the grids in each sub-basin, classify the similar sub-basins into one subnet, and obtain m subnets, that is, divide the study area into m subnets, where m is a positive integer; The step S2 is further as follows: Step S21, constructing a deep reinforcement learning model based on dual DQN optimization; Step S22: weight each sub-basin based on the river length and flow rate of the sub-basin in each sub-network, and calculate the characteristic values ​​of a water resource supply, b water disaster security and c water ecological health of each sub-network; Step S23: Based on the subnet feature data, train the deep reinforcement learning model, construct a mapping relationship between a+b+c feature values ​​of each subnet and the weight of the subnet evaluation index, and optimize the parameters and hyperparameters of the deep reinforcement learning model; Step S24, extracting the characteristic values ​​of a water resource supply, b water disaster security and c water ecological health of each subnet and inputting them into the optimized deep reinforcement learning model to obtain the subnet evaluation index weights, so as to configure the parameters of the respective subnet water resources-ecological environment-economic and social coordinated state evaluation model, extracting and inputting the evaluation index values ​​of each subnet into the respective subnet water resources-ecological environment-economic and social coordinated state evaluation model to obtain the comprehensive evaluation results of the state of each subnet; The step S23 is further as follows: Step S23a: for an existing subnet, determine the weight of each evaluation index of the subnet using an expert scoring method based on its a+b+c characteristic values; Step S23b: training a deep reinforcement learning model to establish a mapping relationship between a+b+c eigenvalues ​​of each subnet and the weight of the subnet evaluation index; Step S23c, taking the minimum cumulative error between the deep reinforcement learning model simulation results and the expert scoring results as the objective function, constructing a parameter and hyperparameter optimization model for the model, and solving the model to obtain the optimal parameter and hyperparameter solution; The step S3 is further as follows: Step S31, respectively sort the characteristic values ​​and evaluation index values ​​of each subnet in forward and reverse order, obtain the forward and reverse order of the attributes of each subnet, sort each subnet in turn in forward order, and select the reverse order of other subnets that are consistent with the forward order of the selected subnet as the networking scheme, and obtain a set of feasible networking schemes; Step S32: for each feasible networking scheme, construct a complementarity index, calculate the matching degree of all subnets in the feasible networking scheme, set a matching degree threshold, calculate the matching degree between each subnet in all feasible networking schemes in turn, extract the feasible networking schemes with matching degrees greater than the matching degree threshold to obtain a matching networking scheme set; Step S33, calculating the engineering construction cost of each matching networking solution, inputting the matching networking solution into the pre-built optimization scheduling model, calculating the optimal scheduling mode and subnet networking benefits corresponding to each networking solution, and calculating the scheduling cost according to the optimal scheduling mode; The step S33 is further as follows: Step S33a, constructing an optimization scheduling model; Step S33b, calculating the engineering cost of each networking solution in the matching networking solution set; Step S33c: input the matching networking scheme into the optimization scheduling model to obtain the optimal scheduling mode, calculate the scheduling cost of each matching networking scheme according to the optimal scheduling mode, and simultaneously calculate the subnet networking benefits after the optimized scheduling of each networking scheme to obtain the improvement value of the comprehensive evaluation of each subnet.

2. The water network system networking method according to claim 1, It is characterized in that The step S4 is further as follows: Step S41: Calculate the total cost of networking based on the networking engineering construction cost and the scheduling cost, and determine a set of networkable solutions from a set of matching networking solutions with the total cost of networking as a constraint; Step S42, constructing a water network system optimization networking decision model, the objective function is: the water network system networking benefit efficiency is maximized, and the water network system networking benefit efficiency is calculated by the tangent slope of the networking benefit to networking cost growth curve; Step S43, input the networking solutions in the networking solution set into the water network system optimization networking decision model respectively, perform decision optimization, and obtain the optimal solution as the water network system networking solution.

3. A water network system networking system, It is characterized in that include: at least one processor; as well as a memory communicatively connected to at least one of the processors; wherein, The memory stores instructions that can be executed by the processor, and the instructions are used to be executed by the processor to implement the water network system networking method described in claim 1 or 2.

Images