Water grid pattern planning method and device, and electronic equipment

Abstract

The present application relates to the technical field of hydraulic engineering, and in particular to a water grid pattern planning method and device and an electronic device. At least one water body corresponding to a target region and water body characteristics corresponding to each water body are obtained. The water body characteristics include water body position pattern characteristics, water body shape characteristics, water body structure characteristics, and water body flow characteristics. Based on the water body characteristics corresponding to each water body, a target water system connectivity result of a target water network corresponding to the target region is calculated. The target water network is composed of each water body. Based on the target water system connectivity result, a target water resource supply and demand matching degree corresponding to the target water network is calculated. Based on the target water resource supply and demand matching degree, the pattern of the target water network is planned and adjusted. The water grid pattern is accurately and effectively planned. This is helpful for realizing sustainable utilization of water resources and sustainable development of the region.

Description

Technical Field

[0001] This invention relates to the field of water conservancy engineering technology, specifically to water network planning methods, devices, and electronic equipment. Background Technology

[0002] With the acceleration of global climate change and urbanization, problems such as uneven spatial and temporal distribution of water resources and ecological degradation are becoming increasingly serious. Water network planning has become a key link in ensuring regional water resource security and sustainable development.

[0003] Traditional water network planning is mostly based on experience and static data, mainly to meet single functional needs such as water supply and flood control, which is difficult to adapt to the complex and ever-changing needs of modern water network management.

[0004] Therefore, how to accurately and effectively plan the water network pattern has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, the present invention provides a water network pattern planning method, apparatus and electronic equipment to solve the problem of how to accurately and effectively plan the water network pattern.

[0006] In a first aspect, the present invention provides a method for planning a water network pattern, the method comprising:

[0007] Obtain at least one water body corresponding to the target area and the water body characteristics of each water body; the water body characteristics include the location and pattern characteristics of the water body, the morphological characteristics of the water body, the structural characteristics of the water body, and the water flow characteristics of the water body;

[0008] Based on the water characteristics of each water body, the target water system connectivity of the target water network corresponding to the target area is calculated; the target water network is based on the composition of each water body.

[0009] Based on the target water system connectivity results, calculate the target water resource supply and demand matching degree corresponding to the target water network;

[0010] Based on the target water resource supply and demand matching degree, the pattern of the target water network is planned and adjusted.

[0011] The water network planning method provided in this application obtains at least one water body corresponding to the target area and the water body characteristics of each water body; based on the water body characteristics of each water body, it calculates the target water system connectivity results of the target water network corresponding to the target area. This allows for a quantitative assessment of the connectivity between various water bodies in the target water network. This helps determine the smoothness of water resource flow throughout the entire water network, as well as the possibility and ease of water resource allocation between different areas. By analyzing water system connectivity, it is possible to identify potential weak links or obstacles in the water network. Based on the target water system connectivity results, it calculates the target water resource supply and demand matching degree corresponding to the target water network. This clarifies the balance between water resource supply and demand in the target water network. This helps managers understand the water resource supply and demand situation in different areas, thereby enabling more scientific allocation and scheduling of water resources, achieving optimal resource allocation, and avoiding the unreasonable phenomenon of water resource surplus in some areas while severe shortages occur in others. Accurate calculation results of the water resource supply and demand matching degree provide important quantitative basis for water resource management decisions. Based on the target water resource supply and demand matching degree, the pattern of the target water network is planned and adjusted. This allows for targeted improvements to the structure and function of the water network. Rational planning and adjustment of the water network pattern contributes to the sustainable use of water resources and regional sustainable development. By optimizing the water network pattern, it is possible to better protect the ecological environment, maintain the balance of the aquatic ecosystem, and achieve coordinated allocation and utilization of water resources among economic, social, and ecological needs, thereby promoting the long-term stable development of the target region.

[0012] In one optional implementation, based on the water body characteristics corresponding to each water body, the target water system connectivity result of the target water network corresponding to the target area is calculated, including:

[0013] Based on the characteristics of each water body, each water body is taken as a node and the channels between each water body are taken as edges, thereby constructing the target water system network model.

[0014] Based on the target water system network model, construct the target adjacency matrix corresponding to each water body;

[0015] Based on the target adjacency matrix, calculate the target connectivity index, target shortest flow path, and target flow efficiency corresponding to the target water network.

[0016] Based on the target connectivity index, the target shortest flow path, and the target flow efficiency, the target water system connectivity result corresponding to the target water network is determined.

[0017] The water network planning method provided in this application constructs a target water system network model based on the characteristics of each water body, treating each water body as a node and the channels between water bodies as edges. This model can present the structure of the target area's water network in an intuitive graphical way. The target water system network model simplifies complex water network systems into combinations of water body nodes and edges, facilitating the analysis of complex water flow relationships within the water network. Based on the target water system network model, a target adjacency matrix is ​​constructed for each water body, enabling a quantitative representation of the connection relationships between water bodies in the water network. The elements in the matrix accurately reflect information such as whether there is a direct connection between any two water bodies and the strength of that connection, providing a standardized data structure for subsequent numerical calculations and analysis. Based on the target adjacency matrix, the target connectivity index, target shortest flow path, and target flow efficiency corresponding to the target water network are calculated. Calculating the target connectivity index quantitatively measures the overall connectivity of the target water network. The target shortest flow path helps find the most efficient path for water resources to flow within the water network. This is of great significance for optimizing water resource allocation schemes and reducing energy loss and time delays during water flow. By selecting the shortest flow path, the efficiency of water transport can be improved, allocation costs reduced, and water resources ensured to reach the demand areas quickly and efficiently. Target flow efficiency can further reveal the flow characteristics of water in the water network. Based on the target connectivity index, the target shortest flow path, and the target flow efficiency, the target water system connectivity results corresponding to the target water network are determined.

[0018] In one optional implementation, based on the characteristics of each water body, a target water system network model is constructed, with each water body as a node and the channels between water bodies as edges, including:

[0019] Acquire future meteorological observation data and current hydrological monitoring data for each water body;

[0020] Based on the water body characteristics of each water body, future meteorological observation data and current hydrological monitoring data, node attribute information is added to the water body nodes corresponding to each water body, and edge attribute information is added to the edges corresponding to each channel to construct an initial water system network model.

[0021] Input the water characteristics of each water body, future meteorological observation data and current hydrological monitoring data into the preset prediction model, and output the dynamic hydrological change data of each water body within a preset time period in the future.

[0022] The hydrological dynamic change data corresponding to each water body are fused with the initial water system network model to generate the target water system network model corresponding to the target water network.

[0023] The water network planning method provided in this application acquires future meteorological observation data and current hydrological monitoring data corresponding to each water body. Based on the water body characteristics, future meteorological observation data, and current hydrological monitoring data corresponding to each water body, node attribute information is added to the water body nodes corresponding to each water body, and edge attribute information is added to the edges corresponding to each channel, thus constructing an initial water system network model. Different types of data (such as location pattern data reflecting the static characteristics of water bodies and hydrological data reflecting dynamic changes) are organically combined through the addition of attribute information, which facilitates the analysis of the interaction and influence mechanisms between various factors and lays the foundation for in-depth research on the dynamic change patterns of the water system. The water body characteristics, future meteorological observation data, and current hydrological monitoring data corresponding to each water body are input into a preset prediction model, which outputs the hydrological dynamic change data corresponding to each water body within a preset future time. This helps to understand the possible changes in the water system in the future and provides forward-looking information for decision-making in water resource management, flood control and disaster reduction, and ecological protection. Accurate prediction of hydrological dynamic changes can help to rationally plan the utilization and allocation of water resources. The hydrological dynamic change data corresponding to each water body is fused with the initial water system network model to generate a target water system network model corresponding to the target water network. This enables the generated target water system network model to reflect changes in the water system in real time and dynamically. The target water system network model integrates current and future water system information, providing a more comprehensive and accurate basis for decision-making in water resource management, water ecological protection, and water conservancy project planning. Through model analysis and simulation, the effects of different decision-making schemes can be evaluated, the advantages and disadvantages of various potential measures can be compared, and the optimal solution can be selected, improving the scientific and rational nature of decision-making and promoting the sustainable use of water resources and the healthy development of the water system's ecological environment.

[0024] In one optional implementation, based on the target water system network model, a target adjacency matrix corresponding to each water body is constructed, including:

[0025] Based on the connection status between each water body, generate the initial adjacency matrix corresponding to each water body;

[0026] Based on the water body characteristics, future meteorological observation data, current hydrological monitoring data and hydrological dynamic change data of each water body in the target water system network model, a dynamic data list corresponding to each water body is generated.

[0027] Associate each dynamic data list with the initial adjacency matrix to generate a backup adjacency matrix;

[0028] Based on the water characteristics of each water body, future meteorological observation data, current hydrological monitoring data, and hydrological dynamic change data, the weight information corresponding to each edge in the backup adjacency matrix is ​​calculated, and the target adjacency matrix is ​​generated.

[0029] The water network pattern planning method provided in this application generates an initial adjacency matrix for each water body based on the connection status between them, intuitively and concisely describing the connectivity relationships between water bodies in the target water network in matrix form. Based on the water body characteristics, future meteorological observation data, current hydrological monitoring data, and hydrological dynamic change data corresponding to each water body in the target water network model, a dynamic data list for each water body is generated. This achieves deep integration of water network-related data, providing a rich information foundation for comprehensive analysis of water network behavior. Furthermore, the dynamic data list can reflect the state changes of each water body under different time and environmental conditions in real time and accurately, providing strong support for prediction and decision-making. Each dynamic data list is associated with the initial adjacency matrix to generate a backup adjacency matrix, giving the adjacency matrix a richer connotation. This allows the adjacency matrix to not only reflect the connectivity relationships between water bodies but also associate the dynamic attribute information of each water body. Based on the water body characteristics, future meteorological observation data, current hydrological monitoring data, and hydrological dynamic change data corresponding to each water body, the weight information corresponding to each edge in the backup adjacency matrix is ​​calculated to generate the target adjacency matrix. This achieves precise quantification of the connectivity relationships between water bodies. The weighting comprehensively considers various factors such as water flow resistance, water conveyance cost, and connectivity stability, enabling the matrix to more realistically reflect the actual water flow transmission situation in the water network. Furthermore, the target adjacency matrix provides core data support for scientific decision-making in water networks. In water resource planning and water network optimization, the weight information based on the matrix can accurately assess the advantages and disadvantages of different connection paths, and formulate optimal water resource allocation schemes and water conservancy project construction plans. Simultaneously, by dynamically updating the weights, changes in the water network can be reflected in real time, allowing decisions to adapt to the constantly changing water network environment and ensuring the scientific and effective management of water resources.

[0030] In one optional implementation, based on the target adjacency matrix, the target connectivity index, the target shortest flow path, and the target flow efficiency corresponding to the target water network are calculated, including:

[0031] Based on the target adjacency matrix, calculate the current connectivity index of the target water network and the predicted connectivity index within a preset time period in the future;

[0032] Generate the target connectivity index based on the current connectivity index and the predicted connectivity index;

[0033] Based on the target adjacency matrix, calculate the current shortest flow path corresponding to the target water network and the predicted shortest flow path within a preset time period in the future.

[0034] Based on the current shortest flow path and the predicted shortest flow path, generate the target shortest flow path;

[0035] Based on the current shortest flow path and the predicted shortest flow path, calculate the current flow efficiency of the target water network and the predicted flow efficiency within a preset time period in the future.

[0036] Based on the current flow efficiency and the predicted flow efficiency, a target flow efficiency is generated.

[0037] The water network planning method provided in this application, based on the target adjacency matrix, calculates the current connectivity index and the predicted connectivity index within a preset future time period for the target water network. This accurately quantifies the current connectivity of the target water network. Calculating the predicted connectivity index within the preset future time period allows for the estimation of the water network's connectivity development trend over a future period based on existing data and models. A target connectivity index is generated based on the current and predicted connectivity indices. This method comprehensively evaluates the connectivity of the target water network from both static and dynamic dimensions. This comprehensive evaluation avoids the limitations of considering only the current state while ignoring future changes, or relying solely on predictions while ignoring current realities, providing a more accurate and comprehensive reference for the long-term planning and management of water networks. Based on the target adjacency matrix, the method calculates the current shortest flow path and the predicted shortest flow path within a preset future time period for the target water network. Determining the current shortest flow path helps achieve efficient water resource allocation in the present. Predicting the predicted shortest flow path within the preset future time period considers the water flow direction of the water network under different future conditions. Based on the current and predicted shortest flow paths, a target shortest flow path is generated, providing a complete water flow path scheme for water network management. This scheme considers both the current actual situation of the water network and potential future changes. Based on the current and predicted shortest flow paths, the current water flow efficiency and the predicted water flow efficiency within a preset future timeframe for the target water network are calculated. Calculating the current water flow efficiency helps understand the water flow operation status of various parts of the water network, identify areas with excessively slow flow velocity or excessive energy loss, and take timely optimization measures to improve the overall operational efficiency of the water network. Calculating the predicted water flow efficiency allows for early prediction of potential efficiency problems the water network may face in the future. Based on the current and predicted water flow efficiencies, a target water flow efficiency is generated. This provides an important reference indicator for the long-term operation and management of the water network. Long-term maintenance and upgrade plans can be formulated based on the target water flow efficiency to ensure that the water network maintains high operational efficiency at different times, achieving efficient utilization and sustainable management of water resources.

[0038] In one optional implementation, the target water system connectivity result includes the current water system connectivity result and the future water system connectivity result within a preset time period; based on the target water system connectivity result, the target water resource supply and demand matching degree corresponding to the target water network is calculated, including:

[0039] Based on the current water system connectivity results, calculate the current water resource supply and demand matching degree corresponding to the target water network;

[0040] Based on the future water system connectivity results, calculate the future water resource supply and demand matching degree corresponding to the target water network;

[0041] Based on the current water resource supply and demand matching degree and the future water resource supply and demand matching degree, the target water resource supply and demand matching degree corresponding to the target water network is determined.

[0042] The water network planning method provided in this application calculates the current water resource supply and demand matching degree corresponding to the target water network based on the current water system connectivity results. This accurately reflects the balance between water resource supply and demand in the target water network at the current moment. Based on future water system connectivity results, it calculates the future water resource supply and demand matching degree corresponding to the target water network, enabling prediction of water resource supply and demand trends within a preset future time period. Based on the current and future water resource supply and demand matching degrees, the target water resource supply and demand matching degree corresponding to the target water network is determined. This achieves a comprehensive assessment of the water resource supply and demand situation of the target water network from the present to the future. This assessment method avoids the limitations of focusing only on a single current or future stage, comprehensively considering the dynamic changes in water network operation, and providing a more systematic and comprehensive perspective for water resource management. Furthermore, the target water resource supply and demand matching degree integrates the current actual situation and future development trends, providing a more scientific and reliable basis for water resource management decisions.

[0043] In one optional implementation, based on the future water system connectivity results, the future water resource supply and demand matching degree corresponding to the target water network is calculated, including:

[0044] Obtain future time feature data for the target area for a predetermined duration, as well as the future water resource demand for the target area.

[0045] Based on the water body characteristics of each water body, future meteorological observation data and current hydrological monitoring data, calculate the future water resource supply corresponding to the target water network.

[0046] Based on future time characteristic data, future water resource demand, future water resource supply, and future water system connectivity results, calculate the future spatial supply and demand matching degree and the future time supply and demand matching degree corresponding to the target water network.

[0047] Based on the future spatial supply and demand matching degree and the future temporal supply and demand matching degree, the future water resource supply and demand matching degree corresponding to the target water network is determined.

[0048] The water network pattern planning method provided in this application obtains future time characteristic data of a target area for a predetermined future duration, as well as the future water resource demand of the target area. Based on the water body characteristics of each water body, future meteorological observation data, and current hydrological monitoring data, the future water resource supply of the target water network is calculated, ensuring the accuracy of the obtained future water resource supply. Based on the future time characteristic data, future water resource demand, future water resource supply, and future water system connectivity results, the future spatial supply and demand matching degree and the future temporal supply and demand matching degree of the target water network are calculated, ensuring the accuracy of the calculated future spatial supply and demand matching degree and the future temporal supply and demand matching degree of the target water network. Based on the future spatial supply and demand matching degree and the future temporal supply and demand matching degree, the future water resource supply and demand matching degree of the target water network is determined, ensuring the accuracy of the determined future water resource supply and demand matching degree of the target water network.

[0049] In one optional implementation, based on future time characteristic data, future water resource demand, future water resource supply, and future water system connectivity results, the future spatial supply and demand matching degree and the future temporal supply and demand matching degree corresponding to the target water network are calculated, including:

[0050] Using each water-using area in the target water network as a regional node and the water flow channels between each water-using area as edges, weights are assigned to each edge based on the future water system connectivity results to construct a basic graph structure.

[0051] Based on the time step, the state of each regional node and the weight of each edge in the basic graph structure are adjusted according to future time feature data, future water resource demand, and future water resource supply to generate multiple spatiotemporal graphs based on time series.

[0052] Construct a multi-layer spatiotemporal graph convolutional network and use graph convolution operations to analyze the spatial node features corresponding to each region node in the basic graph structure.

[0053] By performing convolution operations on each spatiotemporal graph using temporal convolutional layers, the temporal node features corresponding to each region node in the basic graph structure are obtained.

[0054] Based on the characteristics of each spatial node, calculate the future spatial supply and demand matching degree of the target water network;

[0055] Based on the characteristics of each time point, the future time supply and demand matching degree of the target water network is calculated.

[0056] The water network planning method provided in this application uses each water-using area in the target water network as a regional node and the water flow channels between each water-using area as edges. Weights are assigned to each edge based on future water system connectivity results to construct a basic graph structure. This method can intuitively present the water flow relationships and connectivity between each water-using area in the target water network. According to the time step, the state of each regional node and the weights of each edge in the basic graph structure are adjusted based on future time feature data, future water resource demand, and future water resource supply, generating multiple spatiotemporal graphs based on time series, which can fully consider the dynamic changes in water resource supply and demand over time. A multi-layer spatiotemporal graph convolutional network is constructed, and the spatial node features corresponding to each regional node in the basic graph structure are processed through graph convolution operations. This can effectively mine the spatial correlation features between each water-using area. Temporal convolutional layers are used to perform convolution operations on each spatiotemporal graph to obtain the time node features corresponding to each regional node in the basic graph structure. This can capture the characteristics and patterns of water resource supply and demand over time. This helps to understand the evolution pattern of water resource supply and demand over time and provides important temporal information for predicting future supply and demand. By analyzing the characteristics of each time node, the matching degree of water resource supply and demand at different time scales can be assessed, providing support for the formulation of reasonable time-based scheduling strategies. Based on the characteristics of each spatial node, the future spatial supply and demand matching degree corresponding to the target water network can be calculated; this allows for the assessment of the matching of water resource supply and demand among different water-using areas within the target water network from a spatial perspective, clarifying the spatial distribution of water resource surplus and deficit in different areas. Based on the characteristics of each time node, the future time-based supply and demand matching degree corresponding to the target water network can be calculated. This allows for the assessment of the matching degree of water resource supply and demand at different times from a time perspective, helping to formulate water resource scheduling plans adapted to different time scales.

[0057] Secondly, the present invention provides a water network pattern planning device, characterized in that the device comprises:

[0058] The acquisition module is used to acquire at least one water body corresponding to the target area and the water body characteristics corresponding to each water body; the water body characteristics include water body location pattern characteristics, water body morphology characteristics, water body structure characteristics, and water body flow characteristics;

[0059] The first calculation module is used to calculate the target water system connectivity results of the target water network corresponding to the target area based on the water body characteristics of each water body; the target water network is based on the composition of each water body.

[0060] The second calculation module is used to calculate the target water resource supply and demand matching degree corresponding to the target water network based on the target water system connectivity results.

[0061] The adjustment module is used to plan and adjust the pattern of the target water network based on the target water resource supply and demand matching degree.

[0062] The water network planning device provided in this application acquires at least one water body corresponding to a target area and the water body characteristics of each water body; based on the water body characteristics of each water body, it calculates the target water system connectivity results of the target water network corresponding to the target area. This allows for a quantitative assessment of the connectivity between various water bodies in the target water network. This helps determine the smoothness of water resource flow throughout the entire water network, as well as the possibility and ease of water resource allocation between different areas. By analyzing water system connectivity, it is possible to identify potential weak links or obstacles in the water network. Based on the target water system connectivity results, it calculates the target water resource supply and demand matching degree corresponding to the target water network. This clarifies the balance between water resource supply and demand in the target water network. This helps managers understand the water resource supply and demand situation in different areas, thereby enabling more scientific allocation and scheduling of water resources, achieving optimal resource allocation, and avoiding the unreasonable phenomenon of water surplus in some areas while severe shortages occur in others.

[0063] Accurate calculations of water resource supply and demand matching provide crucial quantitative data for water resource management decisions. Based on the target water resource supply and demand matching degree, the pattern of the target water network is planned and adjusted to improve its structure and function in a targeted manner. Rational planning and adjustment of the water network pattern contributes to the sustainable use of water resources and regional sustainable development. By optimizing the water network pattern, while ensuring water demand for economic development, it is possible to better protect the ecological environment, maintain the balance of the aquatic ecosystem, achieve coordinated allocation and utilization of water resources among economic, social, and ecological needs, and promote the long-term stable development of the target region.

[0064] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the water network pattern planning method described in the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0065] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0066] Figure 1 This is a flowchart illustrating the water network pattern planning method according to an embodiment of the present invention;

[0067] Figure 2 This is a flowchart illustrating another water network pattern planning method according to an embodiment of the present invention;

[0068] Figure 3 This is a structural block diagram of a water network pattern planning device according to an embodiment of the present invention;

[0069] Figure 4 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0071] It should be noted that the water network pattern planning method provided in this application embodiment can be implemented by a water network pattern planning device. This device can be implemented as part or all of an electronic device through software, hardware, or a combination of both. The electronic device can be a server or a terminal. In this application embodiment, the server can be a single server or a server cluster composed of multiple servers. The terminal in this application embodiment can be a smartphone, personal computer, tablet computer, wearable device, or other intelligent hardware device such as an intelligent robot. The following method embodiments will use an electronic device as an example for explanation.

[0072] According to an embodiment of the present invention, a method for planning a water network pattern is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0073] This embodiment provides a water network pattern planning method, which can be used in the aforementioned electronic equipment. Figure 1 This is a flowchart of a water network pattern planning method according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps:

[0074] Step S101: Obtain at least one water body corresponding to the target area and the water body characteristics corresponding to each water body.

[0075] The characteristics of a water body include its location and pattern, morphology, structure, and flow. Location and pattern characteristics can include geographical location and altitude. Morphology characteristics can include planar and three-dimensional (elevation) features. Planar features include river length, curvature, shoreline development coefficient, and the ratio of main river area to length. Three-dimensional (elevation) features include width-to-depth ratio, channel storage capacity, storage capacity per unit area, and adjustable storage capacity per unit area. Structural characteristics reflect the relationships between rivers and lakes. Reflecting relationships between rivers and lakes includes river branching ratio, fractal dimension of the water network, river network development coefficient, river network structural stability, and water network unevenness coefficient. Reflecting relationships between rivers and lakes includes the number of river network nodes, water network looping, node connectivity, and hydrological connectivity. Flow characteristics include flow direction, velocity, discharge, water level changes, and flow path.

[0076] The electronic device can receive at least one water body corresponding to the target area input by the user, as well as the water body characteristics corresponding to each water body. It can also receive at least one water body and the water body characteristics corresponding to each water body sent by other devices.

[0077] Step S102: Based on the water characteristics of each water body, calculate the target water system connectivity result of the target water network corresponding to the target area.

[0078] The target water network is based on the composition of each water body.

[0079] Specifically, electronic devices can input the water features corresponding to each water body into a preset model to calculate the target water system connectivity results of the target water network corresponding to the target area.

[0080] This step will be explained in detail below.

[0081] Step S103: Based on the target water system connectivity results, calculate the target water resource supply and demand matching degree corresponding to the target water network.

[0082] Specifically, electronic devices can calculate the target water resource supply and demand matching degree corresponding to the target water network based on the target water system connectivity results.

[0083] This step will be explained in detail below.

[0084] Step S104: Based on the target water resource supply and demand matching degree, the pattern of the target water network is planned and adjusted.

[0085] Specifically, electronic devices can analyze the distribution of various water bodies in a target water network based on the target water resource supply-demand matching degree, checking for spatial mismatches between water supply and demand areas. For example, some urban areas with high water demand lack stable water sources nearby, while water-rich areas are far from water consumption centers, resulting in high water transmission costs and significant losses. And / or, based on the target water resource supply-demand matching degree, electronic devices can assess the connectivity of the target water network, checking for poor connections between water bodies and instances of some areas becoming "islands." Poor connectivity limits water resource allocation capacity, making effective replenishment difficult during localized water shortages. And / or, based on the target water resource supply-demand matching degree, electronic devices can examine the scale and function of reservoirs, ponds, wetlands, and other water storage bodies in the target water network, determining whether they possess sufficient storage capacity to cope with temporal and spatial changes in water resources. For example, if excess water cannot be effectively stored during the wet season, water demand will be difficult to meet during the dry season.

[0086] Then, based on the target water resource supply and demand matching degree and the target water system connectivity results, a more reasonable water resource allocation route is planned. For water-scarce areas, priority is given to diverting water from nearby surplus areas, and routes with good connectivity and high water conveyance efficiency are selected. For example, by constructing new water conveyance channels or optimizing existing river channels, the water flow path is shortened, and water conveyance losses are reduced. To address the problem of insufficient connectivity, engineering measures (such as constructing new connecting ditches and widening narrow river channels) and management measures (optimizing the scheduling of water conservancy facilities) are adopted to enhance the connectivity of the water network. Improving the overall connectivity of the water network helps to achieve flexible allocation of water resources and improve the ability to respond to emergencies. Reasonable planning and construction of water storage facilities are needed to increase the water storage capacity of reservoirs and ponds and restore and protect natural water storage spaces such as wetlands. By optimizing the layout and operation management of water storage facilities, the water storage capacity is improved, and the rational allocation of water resources during wet and dry seasons is achieved. Based on the water resource supply and demand situation, industrial layout and urban development planning are rationally adjusted. Water-intensive industries are guided to relocate to water-rich areas, agricultural planting structures are optimized, and water demand in water-scarce areas is reduced. For example, promoting the planting of drought-resistant crops and developing water-saving agriculture in water-scarce areas.

[0087] The water network planning method provided in this embodiment obtains at least one water body corresponding to the target area and the water body characteristics of each water body; based on the water body characteristics, it calculates the target water system connectivity results of the target water network corresponding to the target area. This allows for a quantitative assessment of the connectivity between various water bodies in the target water network. This helps determine the smoothness of water resource flow throughout the entire water network, as well as the possibility and ease of water resource allocation between different areas. By analyzing water system connectivity, it is possible to identify potential weak links or obstacles in the water network. Based on the target water system connectivity results, it calculates the target water resource supply and demand matching degree corresponding to the target water network. This clarifies the balance between water resource supply and demand in the target water network. This helps managers understand the water resource supply and demand situation in different areas, thereby enabling more scientific allocation and scheduling of water resources, achieving optimal resource allocation, and avoiding the unreasonable phenomenon of water surplus in some areas while severe shortages occur in others.

[0088] Accurate calculations of water resource supply and demand matching provide crucial quantitative data for water resource management decisions. Based on the target water resource supply and demand matching degree, the pattern of the target water network is planned and adjusted to improve its structure and function in a targeted manner. Rational planning and adjustment of the water network pattern contributes to the sustainable use of water resources and regional sustainable development. By optimizing the water network pattern, while ensuring water demand for economic development, it is possible to better protect the ecological environment, maintain the balance of the aquatic ecosystem, achieve coordinated allocation and utilization of water resources among economic, social, and ecological needs, and promote the long-term stable development of the target region.

[0089] This embodiment provides a water network pattern planning method, which can be used in the aforementioned electronic equipment. Figure 2 This is a flowchart of a water network pattern planning method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps:

[0090] Step S201: Obtain at least one water body corresponding to the target area and the water body characteristics corresponding to each water body.

[0091] Among them, water body characteristics include water body location and pattern characteristics, water body morphology characteristics, water body structure characteristics, and water flow characteristics.

[0092] Please refer to the above description of step S101 for details on this step; it will not be repeated here.

[0093] Step S202: Based on the water body characteristics corresponding to each water body, calculate the target water system connectivity result of the target water network corresponding to the target area.

[0094] The target water network is based on the composition of each water body.

[0095] Specifically, step S202 above may include the following steps:

[0096] Step S2021: Based on the characteristics of each water body, each water body is taken as a node and the channels between water bodies are taken as edges, thereby constructing the target water system network model.

[0097] Specifically, step S2021 above may include the following steps:

[0098] Step a1: Obtain future meteorological observation data and current hydrological monitoring data for each water body.

[0099] Future meteorological observation data may include rainfall, evaporation, temperature, wind speed, wind direction, and sunshine duration. Current hydrological monitoring data may include water level, flow rate, and water quality.

[0100] Specifically, electronic devices can obtain publicly available future weather observation data from the websites of meteorological departments in various countries and the International Climate Data Center. For research with special needs, commercial weather data services can be purchased, or customized future weather observation data can be obtained in cooperation with meteorological research institutions.

[0101] Electronic equipment can monitor water levels by setting up level gauges around the water body and measuring the vertical distance from the water surface to a reference plane to reflect changes in water volume; it can also measure flow velocity using velocity meters set up around the water body to obtain the speed of water movement; and it can measure flow rate, i.e., the amount of water passing through a cross-section per unit time, using flow meters set up around the water body. In addition, it can monitor water quality indicators such as dissolved oxygen, chemical oxygen demand, total nitrogen, and total phosphorus, which are commonly detected by water quality monitoring instruments.

[0102] Step a2: Based on the water body characteristics, future meteorological observation data and current hydrological monitoring data, add node attribute information to the water body nodes corresponding to each water body, add edge attribute information to the edges corresponding to each channel, and construct an initial water system network model.

[0103] Specifically, electronic devices can extract key features from various data types, including water body characteristics, future meteorological observation data, and current hydrological monitoring data, based on a preset autoencoder. These extracted key features are then encoded into a unified feature vector, enabling deep data fusion. For example, the location and morphological features of water bodies can be fused with future precipitation probabilities and current water level data to uncover potential correlations between the data, providing richer and more accurate information for subsequent modeling.

[0104] Then, based on the feature vectors corresponding to each water body, node attribute information is added to each corresponding water body node.

[0105] The electronic equipment calculates the velocity correction coefficient of water flow in the channel at different time periods based on wind speed and direction from future meteorological data, combined with the topography and orientation of the channel, and dynamically adjusts it over time. In addition, the electronic equipment uses the current hydrological monitoring flow data, combined with the channel's water carrying capacity (based on the channel's morphology and structural characteristics), to determine the flow carrying capacity attribute of the side, and dynamically adjusts it over time.

[0106] Then, based on the speed correction coefficient and flow carrying capacity attribute, the electronic device adds edge attribute information to the edge corresponding to each channel to construct an initial water system network model.

[0107] Step a3: Input the water characteristics of each water body, future meteorological observation data, and current hydrological monitoring data into the preset prediction model, and output the dynamic hydrological change data of each water body within a preset time period.

[0108] Specifically, for each water body, the future meteorological observation data and current hydrological monitoring data corresponding to the water body are input into the first sub-prediction model of the preset prediction model, and the output is the first sub-hydrological dynamic change data corresponding to the water body within a preset future time. The future meteorological observation data and current hydrological monitoring data corresponding to the water body are input into the second sub-prediction model of the preset prediction model, and the output is the second sub-hydrological dynamic change data corresponding to the water body within a preset future time.

[0109] The water body features are input into the preset spatial feature model in the preset prediction model, and the spatial features corresponding to the water body are output.

[0110] The first and second sub-sub-hydrological dynamic change data, along with spatial features, are input into the meta-model, which finally outputs the hydrological dynamic change data corresponding to the water body within a preset future time period.

[0111] The first and second sub-prediction models can be any of the following: Recurrent Neural Networks (RNNs) and their variants Long Short-Term Memory Networks (LSTMs) and Gated Recurrent Units (GRUs). The first and second sub-prediction models are distinct. The preset spatial feature model can be a convolutional neural network or other networks. The meta-model can be a logistic regression model or other models. This application does not specifically limit the first sub-prediction model, the second sub-prediction model, the preset spatial feature model, or the meta-model.

[0112] Step a4: The hydrological dynamic change data corresponding to each water body is fused with the initial water system network model to generate the target water system network model corresponding to the target water network.

[0113] Specifically, the hydrological dynamic change data corresponding to each water body is used as new attribute information and associated with the attribute information corresponding to each water body to generate the target water system network model corresponding to the target water network.

[0114] For example, for each water body node, the predicted future water level change data is added as a dynamic attribute of the water body node; for the edges of the channel, the water flow velocity change data is updated as an attribute of the edge. In this way, the model can reflect the dynamic characteristics of the water body and water flow.

[0115] Step S2022: Based on the target water system network model, construct the target adjacency matrix corresponding to each water body.

[0116] Specifically, step S2022 above may include the following steps:

[0117] Step b1: Generate the initial adjacency matrix for each water body based on the connection status between them.

[0118] Specifically, the electronic device counts the total number of water bodies within the target area, let's say the total number of water bodies is n. The initial adjacency matrix is ​​determined to be n×n based on the number of water bodies. The rows and columns of the matrix correspond to each water body, so that each element in the matrix accurately represents the connection between a pair of water bodies. Then, an n×n matrix is ​​created, and all elements are initialized to 0. In this initial state, the matrix represents that there are no connections between any water bodies; the elements will be modified later based on the actual connection situation. Each pair of water bodies is traversed sequentially according to the row and column order of the matrix. For each pair of water bodies, the matrix elements are filled according to the previously determined connection relationship. If the two water bodies currently being traversed are connected, the element at the intersection of the corresponding row and column is set to 1; if the two water bodies are not connected, the element is kept at 0. For example, if water body A and water body B are connected, and water body A corresponds to the i-th row and water body B corresponds to the j-th column (i...

[0119] =j), then set the elements aij and aji in the matrix to 1 (in an undirected graph, the connectivity is symmetrical; if the direction of water flow is considered, in a directed graph, only the element in one direction may be set to 1). For the connection between water bodies themselves (i.e., the diagonal element aii of the matrix), the value is usually determined according to the research purpose. In general water system connectivity analysis, it can be set to 0, indicating that there is no actual water flow connection between water bodies (except in special cases, such as some water bodies with internal circulating water flow).

[0120] Step b2: Based on the water body characteristics, future meteorological observation data, current hydrological monitoring data, and hydrological dynamic change data corresponding to each water body in the target water system network model, generate a dynamic data list corresponding to each water body.

[0121] Specifically, electronic devices can extract information related to the static attributes of water bodies from water feature data, such as the geographical coordinates, altitude, surface area, and river network density of the water body. These attributes reflect the basic characteristics of the water body and are the foundational information for constructing dynamic data lists, helping to determine the location and scale of the water body within the river network.

[0122] Then, electronic devices combine future meteorological observation data and current hydrological monitoring data to extract attributes closely related to the dynamic changes of water bodies. For example, they calculate the trend of water balance changes based on precipitation and evaporation data; extract factors that may affect the surface flow pattern of water bodies from wind speed and direction data; and use indicators such as dissolved oxygen and chemical oxygen demand from water quality monitoring data to reflect the dynamic changes in water quality.

[0123] Furthermore, electronic devices perform in-depth analysis of hydrological dynamics data, selecting key dynamic characteristics. For example, they focus on the magnitude of water level fluctuations, the rate of change in flow rate, and the trends in water quality indicators. These key characteristics highlight the dynamic trends of water bodies over future periods, providing crucial information for subsequent decision-making and analysis.

[0124] Finally, the electronic device determines the structure of the dynamic data list based on the extracted data features. The list is typically presented in tabular form, with each row representing a time step (e.g., hourly, daily) and each column corresponding to a different attribute or characteristic of a water body. The table header clearly indicates the meaning of the data represented by each column, such as "time," "water level," "flow rate," and "dissolved oxygen," ensuring data clarity and readability. Following chronological order, information related to the processed static attributes and dynamic attributes of the water body is sequentially filled into the dynamic data list based on dynamic change characteristics. During the filling process, the accuracy and completeness of the data are ensured, with the attribute data of each water body at the same time step being filled into the corresponding cells. For example, at a certain moment, the water level, flow rate, water balance change value calculated based on meteorological data, and predicted water quality changes of a certain water body are accurately filled into the corresponding rows and columns. To better understand the data in the dynamic data list, necessary annotations and explanatory information are added. Explanations are provided for some special data values ​​or trends, indicating the data source, processing method, and possible error range. This helps users interpret the data more accurately and avoid misunderstandings and misjudgments.

[0125] Step b3: Associate each dynamic data list with the initial adjacency matrix to generate a backup adjacency matrix.

[0126] Specifically, electronic devices can assign a unique identifier (such as a water body number or geographic location code) to each water body. Both the dynamic data list and the initial adjacency matrix use this identifier to identify the corresponding water body. This water body identifier serves as the link between the two, establishing a connection. Thus, during data association, the dynamic data of a specific water body can be accurately matched with the corresponding row and column in the initial adjacency matrix.

[0127] Then, the electronic device determines the correspondence rules between the dynamic data list and the elements of the initial adjacency matrix. For example, certain key data in the dynamic data list (such as flow rate, water level change trends, etc.) can be used as weights and combined with the connection relationships in the initial adjacency matrix. If there is a connection between two water bodies (the corresponding element in the initial adjacency matrix is ​​1), the weight of this connection is adjusted according to the magnitude of their flow rate data; the larger the flow rate, the higher the weight. Alternatively, the direction of water flow can be determined based on the water level change trend. If the water level of one water body is rising and the water level of another water body is falling, and the two are connected, the connection direction and strength information can be further refined.

[0128] Based on the size and structure of the initial adjacency matrix, a framework for a backup adjacency matrix is ​​created. The backup adjacency matrix has the same number of rows and columns as the initial adjacency matrix, and also uses water bodies as row and column identifiers. In this framework, each element will carry the merged information, reflecting not only the connectivity between water bodies but also incorporating dynamic data.

[0129] Finally, according to the established association methods and corresponding rules, the data in the dynamic data list is filled into the standby adjacency matrix. For positions in the initial adjacency matrix where the element is 1 (i.e., there is a connection between water bodies), new values ​​are calculated and filled based on the dynamic data. For example, if two water bodies are connected, the flow data between them is used as the weight, and the weight value is filled into the corresponding element position in the standby adjacency matrix; if the element is 0 (no connection between water bodies), a special value (such as -1 indicating no connection and no related dynamic data association) or a value of 0 is decided based on the actual situation to maintain the sparsity and readability of the matrix. When filling matrix elements, an appropriate data fusion strategy is adopted. For cases where multiple dynamic data affect the same matrix element, methods such as weighted average and principal component analysis can be used for fusion. For example, when both flow and water level change trends affect the weight of the connection between two water bodies, the final weight value is determined by calculating a weighted average, where the weight can be set according to the importance or relevance of the data.

[0130] Step b4: Based on the water characteristics of each water body, future meteorological observation data, current hydrological monitoring data, and hydrological dynamic change data, calculate the weight information corresponding to each edge in the backup adjacency matrix and generate the target adjacency matrix.

[0131] Specifically, electronic devices can calculate the initial weight information corresponding to each edge in the backup adjacency matrix based on the water flow characteristics between water bodies. For example, water flow-related factors such as flow rate, velocity, and water level difference can be considered. Edges connecting water bodies with high flow rates can have higher weights because high flow rates mean more frequent exchange of matter and energy between water bodies, resulting in a greater impact on the water network. Flow velocity can also be used as a basis for weight adjustment; edges with faster flow rates have appropriately higher weights because faster flow rates facilitate rapid connectivity and material transport between water bodies. Simultaneously, water level difference reflects the potential energy difference between water bodies; edges with larger water level differences have appropriately higher weights, indicating that water flows more easily between them.

[0132] Then, based on future meteorological observation data, the initial weight information corresponding to each edge in the backup adjacency matrix is ​​adjusted to generate backup weight information. For example, increased precipitation may lead to rising water levels and increased flow; therefore, the weights of water body edges that may be strengthened by precipitation are increased accordingly.

[0133] Finally, the backup weight information is adjusted again based on the characteristics and functions of the water bodies to generate the target weight information corresponding to each edge in the backup adjacency matrix, thus generating the target adjacency matrix. For example, the morphological characteristics of water bodies, such as the curvature and width of river channels, affect flow resistance. Edges with less curvature and wider channels can have their weights appropriately increased because the flow is smoother. Structural characteristics of water bodies, such as the connectivity of the river network and the importance of nodes, can also be used for weight setting. Water bodies located at key connectivity nodes should have higher weights for their edges, as these water body nodes are crucial to the overall connectivity and function of the water network. Furthermore, the weights of water bodies with different functions, such as irrigation water bodies and ecological water bodies, can be adjusted according to their functional importance when calculating weights. For edges connecting water bodies that ensure agricultural irrigation, the weights can be appropriately increased during peak agricultural water use periods.

[0134] Step S2023: Based on the target adjacency matrix, calculate the target connectivity index, the target shortest flow path, and the target flow efficiency corresponding to the target water network.

[0135] Specifically, step S2023 above may include the following steps:

[0136] Step c1: Based on the target adjacency matrix, calculate the current connectivity index of the target water network and the predicted connectivity index within a preset future time period.

[0137] Specifically, based on the target adjacency matrix, each row (or column) of the matrix is ​​traversed to determine the connectivity of each water body. For a given water body, the matrix elements in its corresponding row (or column) are checked; non-zero elements indicate that the corresponding water body is connected to it, and the number of connected water bodies is counted. For example, for water body C, if its corresponding row in the target adjacency matrix has 5 non-zero elements, then the number of water bodies connected to water body C, Nconnected, is 5. Then, based on the formula... Calculate the current connectivity index corresponding to the target water network, where W i It is the weight of water body i (such as area, ecological importance, etc.), C i It is the connectivity coefficient of water body i. WCI can more comprehensively assess the overall function of the water system and highlight the connectivity of important water bodies.

[0138] The future connectivity of water networks will be affected by various factors, such as climate change and human activities. Based on future meteorological observation data and water conservancy project plans, the target adjacency matrix is ​​adjusted. For example, if increased precipitation is predicted in a certain area, it may strengthen the connection between two previously weakly connected water bodies, thus increasing the values ​​of the corresponding elements in the target adjacency matrix for those two water bodies; if a planned new water conservancy project severs the connection between two water bodies, the corresponding element is set to 0. Using the same method as calculating the current connectivity index, the connectivity index for the future preset time period is calculated based on the adjusted target adjacency matrix. A suitable connectivity index calculation method (basic connectivity index or weighted connectivity index) is selected, and the number of connected water bodies is determined according to the calculation steps (by re-traversing the adjusted matrix to determine Nconnected), the weights are calculated (if using the weighted connectivity index), and the index value is calculated using the formula. The result is the predicted connectivity index of the target water network for the future preset time period.

[0139] Step c2: Generate the target connectivity index based on the current connectivity index and the predicted connectivity index.

[0140] Specifically, the electronic device combines the current connectivity index and the predicted connectivity index to generate the target connectivity index.

[0141] Step c3: Based on the target adjacency matrix, calculate the current shortest flow path corresponding to the target water network and the predicted shortest flow path within a preset time period in the future.

[0142] Specifically, electronic devices can construct a graph data structure based on the target adjacency matrix. Each water body is treated as a node in the graph, and the connections between water bodies are treated as edges. The weight of each edge is the value of the corresponding element in the adjacency matrix. Simultaneously, a starting node and a target node are determined. The starting node can be selected based on the research objective, such as a major water source; the target node can be a water body corresponding to an area with high water demand.

[0143] First, the distance to the starting node is marked as 0, and the distances to other nodes are marked as infinity. Then, starting from the starting node, its neighboring nodes are traversed, and the distance from each neighboring node to the starting node is updated (distance equals the distance from the starting node to the current node plus the edge weight from the current node to its neighboring nodes). Each time, the undetermined node with the smallest distance is selected, and its neighboring nodes are expanded until the target node is reached or all nodes have been processed. During the traversal, the predecessor node of each water body node is recorded so that the shortest path can be obtained by backtracking. By backtracking the predecessor node of the target node, the current shortest flow path from the starting node to the target node is obtained. This path is the optimal path for water flow from the starting point to the target point under the current water network connection state and weight settings.

[0144] Furthermore, electronic devices consider the impact of future meteorological observation data on water network flow. If future precipitation increases in a certain area, it may lead to increased flow in some water bodies, and the weights of the connecting edges between these water bodies will be adjusted accordingly to make them more favorable for water flow (e.g., by decreasing weight values); conversely, if precipitation decreases, it may be necessary to increase the weights of some edges. If there are plans to construct new river channels, sluice gates, or widen existing river channels in the future, the target adjacency matrix will be adjusted according to the project plan. In addition, land use changes (such as urbanization, agricultural expansion, etc.) may affect the structure and flow conditions of the water network.

[0145] Then, similar to calculating the current shortest flow path, the electronic device selects a preset shortest path algorithm and performs calculations based on the adjusted target adjacency matrix. Similarly, it determines the starting node and the target node, and calculates the predicted shortest flow path from the starting node to the target node within a preset future timeframe, following the algorithm steps.

[0146] Step c4: Generate the target shortest flow path based on the current shortest flow path and the predicted shortest flow path.

[0147] Specifically, the electronic device combines the current shortest flow path with the predicted shortest flow path to generate the target shortest flow path.

[0148] Step c5: Based on the current shortest flow path and the predicted shortest flow path, calculate the current flow efficiency of the target water network and the predicted flow efficiency within a preset time period.

[0149] Specifically, the electronic device calculates the total resistance of the water flow through the current shortest flow path based on the weights of each connecting edge. Assuming the current shortest flow path consists of edges e1, e2, ..., en, with weights w1, w2, ..., wn respectively, the total resistance... The current flow efficiency E can be expressed as the reciprocal of the total resistance, i.e., E = 1 / R.

[0150] Similarly, the electronic device calculates the predicted water flow efficiency within a preset time period based on the predicted shortest flow path.

[0151] Step c6: Generate the target flow efficiency based on the current flow efficiency and the predicted flow efficiency.

[0152] Specifically, the electronic device combines the current flow efficiency and the predicted flow efficiency to generate the target flow efficiency.

[0153] Step S2024: Based on the target connectivity index, the target shortest flow path, and the target flow efficiency, determine the target water system connectivity result corresponding to the target water network.

[0154] Specifically, the electronic device integrates the current connectivity index in the target connectivity index, the current shortest flow path in the target shortest flow path, and the current flow efficiency in the target flow efficiency to generate the current water system connectivity result.

[0155] Specifically, the electronic device can construct a judgment matrix by comparing the relative importance of the current connectivity index, the current shortest flow path, and the current flow efficiency to the current water system connectivity in pairs. A 1-9 scale is used for the comparison, where 1 indicates that the two factors are equally important, 3 indicates that one factor is slightly more important than the other, 5 indicates that one factor is significantly more important than the other, 7 indicates that one factor is strongly more important than the other, 9 indicates that one factor is extremely more important than the other, and 2, 4, 6, and 8 are the median values ​​of adjacent judgments. Let A represent the judgment matrix, and a... ij Let A represent the ratio of the relative importance of the i-th factor to the j-th factor. Then, the judgment matrix A is as follows:

[0156]

[0157] For example, if we consider the current connectivity index to be significantly more important than the current shortest flow path, and strongly more important than the current flow efficiency, while the current shortest flow path is slightly more important than the current flow efficiency, then the judgment matrix might be:

[0158]

[0159] Calculate the largest eigenvalue λmax of the judgment matrix A and its corresponding eigenvector W. This can usually be done using methods such as the root method or the sum-product method. Taking the root method as an example, the calculation steps are as follows:

[0160] Calculate the product Mi of the elements in each row of the judgment matrix A:

[0161] M1=1×5×7=35, M2=1 / 5×1×3=3 / 5, M3=1 / 7×1 / 3×1=1 / 21.

[0162] Calculate M i nth root (where n is the order of the matrix, and here n = 3):

[0163]

[0164] Will Normalization yields the weight vector W:

[0165]

[0166] Therefore, the weight vector W = (0.730, 0.188, 0.082). T This represents the relative weights of the current connectivity index, the current shortest flow path, and the current flow efficiency with respect to the current water system connectivity.

[0167] Similarly, the electronic device integrates the future connectivity index in the target connectivity index, the future shortest flow path in the target shortest flow path, and the future flow efficiency in the target flow efficiency to generate the future water system connectivity result.

[0168] Then, the current water system connectivity results are combined with the future water system connectivity results to generate the target water system connectivity results.

[0169] Step S203: Based on the target water system connectivity results, calculate the target water resource supply and demand matching degree corresponding to the target water network.

[0170] Specifically, the target water system connectivity result includes the current water system connectivity result and the future water system connectivity result within a preset time period. The above step S203 may include the following steps:

[0171] Step S2031: Based on the current water system connectivity results, calculate the current water resource supply and demand matching degree corresponding to the target water network.

[0172] Specifically, electronic devices can acquire current time characteristic data corresponding to the target area, which may include time information such as season, month, and day. This data will affect the demand and supply of water resources. For example, water consumption usually increases in summer, while river runoff may decrease in winter.

[0173] Electronic equipment calculates the current water demand of a target area by statistically analyzing water-using sectors such as population, industry, and agriculture. For example, it calculates the domestic water demand in densely populated urban areas and the production water demand in industrial zones. Then, based on the characteristics of each water body (such as river width, depth, and flow velocity, and lake storage capacity) and current hydrological monitoring data (such as precipitation, evaporation, and runoff), it calculates the water supply that the target water network can provide. For example, by monitoring river runoff and lake water level changes, combined with relevant water body characteristics, it estimates the amount of water available for supply.

[0174] Electronic devices can analyze the water supply and demand situation at different locations within a target area. For example, some areas may have abundant water supply but low demand, while other areas may have high demand but insufficient supply. Specifically, the target area can be divided into several sub-regions, and the water supply-demand ratio for each sub-region can be calculated. For example, for sub-region i, with water supply of Si and demand of Di, the supply-demand ratio Ri = Si / Di.

[0175] Then, based on factors such as the area and population of each sub-region, these supply-demand ratios are weighted and averaged to obtain the current spatial supply-demand matching degree of the entire target region. For example, if the area of ​​sub-region i accounts for wi of the total area of ​​the target region, then the current spatial supply-demand matching degree is... Where n is the number of subregions.

[0176] Consider the impact of current time-specific data on water supply and demand. For example, during the high temperatures of summer, the demand for agricultural irrigation and domestic water increases, while rainfall may be low, river levels may drop, and water supply may be relatively reduced.

[0177] Electronic devices can analyze water supply and demand patterns under similar historical time characteristics and establish supply and demand relationship models. For example, by statistically analyzing the water supply and demand during summers over many years, they can identify the patterns of change between them.

[0178] Substituting the current water supply and demand into the model, the supply-demand matching degree at the current time can be calculated. For example, it can be comprehensively measured by the ratio of the current supply to the historical average supply for the same period, and the ratio of the current demand to the historical average demand for the same period. Assume the current supply is S. now The historical average supply for the same period is S avg The current demand is D. now The historical average demand for the same period is D avg The current supply and demand matching degree

[0179] The current water resource supply and demand matching degree corresponding to the target water network is determined by comprehensively considering the current spatial supply and demand matching degree Ms and the current temporal supply and demand matching degree Mt. Different weights can be assigned to the spatial and temporal supply and demand matching degrees according to the actual situation, for example, the weights can be set as α and β (α+β=1) respectively.

[0180] The current water resource supply and demand matching degree is then M = αMs + βMt. This method integrates the supply and demand situation across both spatial and temporal dimensions to comprehensively assess the current water resource supply and demand matching degree of the target water network. If the value of M is close to 1, it indicates a good water resource supply and demand matching; if the value of M is much greater than 1, it indicates that supply exceeds demand; if the value of M is much less than 1, it indicates that demand exceeds supply, and there is a water shortage.

[0181] Step S2032: Based on the future water system connectivity results, calculate the future water resource supply and demand matching degree corresponding to the target water network.

[0182] Specifically, step S2032 above may include the following steps:

[0183] Step d1: Obtain future time feature data for the target area for a preset duration, as well as the future water resource demand for the target area.

[0184] Specifically, electronic devices can collect time-related data about a target area over a relatively long period of time (e.g., several to several decades), including information such as seasonal changes, average monthly temperature, monthly precipitation, and holiday distribution.

[0185] Then, based on the characteristics and data features of the target region, a time series forecasting model is determined. This model can be a moving average, exponential smoothing, or ARIMA (Autoregressive Integral Moving Average) model. Furthermore, the electronic device needs to consider external factors that may influence future time characteristics. For example, climate change trends may lead to changes in temperature and precipitation patterns, and changes in policies and regulations may affect holiday arrangements or social activity patterns. These external factors are used as input variables or adjustment factors in the time series forecasting model to improve forecast accuracy. Finally, based on the time series forecasting model and considering external factors, the electronic device predicts the time characteristics within a preset time period.

[0186] Electronic devices can collect historical water consumption data from various industries and sectors in a target area, including domestic, industrial, agricultural, and ecological water use. The analysis focuses on the trends, seasonal characteristics, and relationships of this data with factors such as population growth and economic development. Based on the target area's population and economic development plans, the projected population size and economic growth rate over a predetermined timeframe are then used to predict future trends. Furthermore, if the target area has plans for industrial restructuring, such as a shift from water-intensive heavy industries to water-efficient high-tech industries, the impact of this restructuring on water demand needs to be analyzed. Based on water consumption quotas and development plans for different industries, the future water demand for each industry is re-estimated.

[0187] Electronic equipment integrates the aforementioned factors such as population, economy, and industrial structure, and uses various models and analytical methods to calculate the water resource demand of each industry and sector in the target area over a predetermined period of time. This calculation then aggregates the total future water resource demand. This demand is a dynamic, predicted value that considers changes in multiple factors, providing crucial information for water resource planning and management.

[0188] Step d2: Based on the water characteristics of each water body, future meteorological observation data, and current hydrological monitoring data, calculate the future water resource supply corresponding to the target water network.

[0189] Specifically, electronic devices can input the water body characteristics of each water body, future meteorological observation data, and current hydrological monitoring data into a preset water resource supply calculation model to calculate the future water resource supply corresponding to the target water network. This preset water resource supply calculation model can be a distributed hydrological model (such as SWAT, HEC-HMS, etc.), a lumped model (such as the Xin'anjiang model), or a numerical model based on physical processes. The preset water resource supply calculation model is obtained by calibrating and validating the parameters of the original model based on historical data. Specifically, past meteorological and hydrological observation data are input into the original model, and the model parameters are adjusted to make the model output match the actual observation data as closely as possible. For example, by adjusting soil permeability coefficients and vegetation interception parameters, the simulated river flow of the preset water resource supply calculation model is made consistent with the measured flow. The calibrated preset water resource supply calculation model can more accurately reflect the hydrological processes of the target water network.

[0190] Step d3: Based on future time characteristic data, future water resource demand, future water resource supply, and future water system connectivity results, calculate the future spatial supply and demand matching degree and the future time supply and demand matching degree corresponding to the target water network.

[0191] Specifically, step d3 above may include the following steps:

[0192] Step d31: Using each water-using area in the target water network as a regional node and the water flow channels between each water-using area as edges, weights are assigned to each edge based on the future water system connectivity results to construct a basic graph structure.

[0193] Specifically, electronic devices can perform detailed analysis of the target water network coverage area, clearly defining the boundaries and extent of each water use zone. Water use zones can be different functional areas of a city (such as residential areas, industrial areas, commercial areas, etc.), agricultural irrigation areas, ecological protection areas, etc. Based on actual water demand and management units, the target water network is divided into several relatively independent water use zones, each serving as a regional node in the basic map structure.

[0194] Next, the water flow channels between the various water-using areas in the target water network are identified. These channels include natural rivers, canals, and lake connections, and may also include man-made water pipelines and canals. These channels connect different water-using areas and form an important basis for constructing the edges in the basic graph structure.

[0195] The various indicators in the future water system connectivity results are correlated with the water flow channels between water use areas. Then, weights are assigned to each side based on the future water system connectivity results. The weights should reflect the importance of the water flow channel in the future water network connectivity and the ease of water flow. Generally, higher weights are assigned to water flow channels with better future water system connectivity, while lower weights are assigned to channels with poor connectivity or certain obstacles to water flow.

[0196] Optionally, a comprehensive score can be calculated based on indicators such as the target connectivity index, the target shortest flow path length, and the target flow efficiency. For example, the target connectivity index can be standardized and multiplied by a certain weighting coefficient (e.g., 0.4), the reciprocal of the target shortest flow path length can be standardized and multiplied by a certain coefficient (e.g., 0.3), and the target flow efficiency can be standardized and multiplied by a certain coefficient (e.g., 0.3). These three scores are then added together to obtain the weight value of the edge. Finally, with the defined water use area as the region node and the flow channel as the edge, the basic graph structure is constructed by assigning corresponding weights to each edge.

[0197] Step d32: Based on future time feature data, future water resource demand, and future water resource supply, adjust the state of each regional node and the weight of each edge in the basic graph structure according to the time step to generate multiple spatiotemporal graphs based on time series.

[0198] Specifically, electronic devices can determine an appropriate time step based on the required accuracy of the research and the temporal resolution of the data. The time step can be an hour, day, week, month, etc. Then, using the constructed base graph structure as the initial state, the state of each regional node (water-using area) in the base graph structure is adjusted according to future water resource demand. If the water resource demand of a certain water-using area increases within the current time step, the attribute of that regional node is changed accordingly, such as marking it as a high-demand state; conversely, if the demand decreases, it is marked as a low-demand state. Furthermore, the water use priorities of different water-using areas are considered; for example, domestic water use is usually prioritized over industrial and agricultural water use. This priority difference is reflected when adjusting the state of regional nodes. For example, when water supply is tight, priority is given to ensuring the needs of domestic water-using areas, and the states of other water-using areas are adjusted accordingly.

[0199] The weights of each edge (water flow channel) are adjusted based on changes in future water supply. If the water source connected to a certain water flow channel experiences an increase in supply during the current time step, enabling smoother water supply to the water-using area, the weight of that edge is increased to reflect the increased likelihood and efficiency of water flow through the channel. Conversely, if the supply decreases or the water flow is obstructed, the weight of the edge is decreased. Then, the edge weights are further adjusted by incorporating future time-specific data. For example, during the dry season, some water flow channels dependent on specific water sources may have their weights reduced due to reduced water volume at the source; while during the rainy season, some channels with normally low flow rates may have their weights increased due to rainfall replenishment. Additionally, special times such as holidays may cause changes in water demand between certain areas, thus affecting the edge weights.

[0200] After adjusting the state of regional nodes and edge weights in the basic graph structure at each time step, a new spatiotemporal graph is generated. This spatiotemporal graph reflects the state of the water network at the current time step, including the demand of each water-using area and the connectivity of water flow channels. These spatiotemporal graphs are arranged in chronological order to form multiple spatiotemporal graphs based on time series.

[0201] Step d33: Construct a multi-layer spatiotemporal graph convolutional network, and use graph convolution operations to process the spatial node features corresponding to each region node in the basic graph structure.

[0202] Specifically, a multi-layer spatiotemporal graph convolutional network typically consists of an input layer, multiple hidden layers, and an output layer. The input layer receives the basic graph structure, along with relevant region node features and edge information. The hidden layers are the core of the network, extracting and learning spatial and temporal features from the graph structure by stacking multiple graph convolutional layers. The output layer outputs corresponding results according to task requirements, such as predicted values ​​for each region node or overall evaluation metrics for the water network.

[0203] Specifically, each region node in the basic graph structure corresponds to certain spatial node features, which may include information such as the geographical location, area, population, and water use type of the water-using region. Before performing graph convolution operations, these features need to be encoded and vectorized for input into the network for processing. One-hot encoding, vector embedding, and other methods can be used to convert discrete features into continuous vectors. For numerical features, normalization can be performed to ensure they have a uniform scale.

[0204] For each region node, a weighted sum of features from its neighboring nodes is calculated through graph convolutional layers to update the feature representation of that region node. Different graph convolutional methods employ different strategies when calculating weights. For example, in graph convolutional networks, weights are pre-calculated based on the Laplacian matrix of the graph; while in graph attention convolution, weights are adaptively learned through an attention mechanism, which better reflects the importance of regions among nodes.

[0205] After aggregating neighborhood information, the features of neighboring regional nodes are fused with the features of the current regional node to update its feature representation. This updated feature not only includes information about the regional node itself but also incorporates information from its neighborhood, better reflecting the contextual relationships of the regional node within the graph structure. As the number of network layers increases, the features of regional nodes are continuously propagated and updated within the graph structure, enabling each regional node to access a wider range of graph structure information.

[0206] To enhance the expressive power of a network, non-linear activation functions, such as ReLU (Rectified Linear Unit), are typically introduced into graph convolution operations. After updating the features of region nodes, a non-linear activation function is applied to map the linearly transformed features into a non-linear space, obtaining the spatial node features corresponding to each region node, thereby enabling the learning of more complex feature relationships.

[0207] Step d34: Perform convolution operations on each spatiotemporal graph using a temporal convolutional layer to obtain the temporal node features corresponding to each region node in the basic graph structure.

[0208] Specifically, the core of a temporal convolutional layer is the convolutional kernel, which slides across the spatiotemporal graph along the time dimension. The size of the convolutional kernel determines the number of time steps processed in a single operation on the time series. For example, a convolutional kernel of size 3 will process three consecutive spatiotemporal graphs simultaneously. The depth (number of channels) of the convolutional kernel can be set according to the feature dimensions of the input spatiotemporal graph to ensure effective feature extraction. By adjusting the size and depth of the convolutional kernel, the range and precision with which the temporal convolutional layer captures time series information can be controlled.

[0209] When a temporal convolutional layer performs convolution operations on the temporal dimension of a spatiotemporal graph, the convolutional kernel slides along the time series, moving with a fixed stride (usually 1). At each location, the convolutional kernel performs element-wise multiplication with the corresponding spatiotemporal graph region and sums the results to obtain a new feature value. This process is similar to traditional image convolution operations, except that it processes graph-structured data in the temporal dimension. In this way, temporal convolutional layers can capture the local features and change patterns of the spatiotemporal graph over time.

[0210] To enhance the expressive power of a model, non-linear activation functions are typically applied to the output of temporal convolutional layers. These activation functions transform the linear output of the convolution operation into a non-linear output, enabling the model to learn more complex temporal relationships. For example, the ReLU function can set negative outputs to 0 while retaining positive outputs, thus introducing a non-linear element and helping the model better fit complex patterns in the data.

[0211] After performing convolution operations on the spatiotemporal graph in the temporal convolutional layer, each region node receives a new feature representation. This feature representation integrates information from multiple time steps for that region node. The original features of each region node in the spatiotemporal graph are updated with new features containing temporal dimension information after the convolution operation. For example, the water demand, supply, and other features of a water-using region node at different time steps are integrated into a comprehensive temporal node feature through temporal convolution, reflecting the changing trends and patterns of that region node over time.

[0212] Temporal convolutional layers may contain multiple convolutional kernels, each generating a corresponding feature map. These feature maps can be combined through aggregation operations (such as addition and concatenation) to obtain the final temporal features. The aggregated temporal features contain rich temporal information, better describing the behavior and characteristics of each region node in the base graph structure over time. These temporal features can serve as input for subsequent analysis and prediction tasks, such as predicting the future state of water networks and assessing trends in water supply and demand.

[0213] Step d35: Based on the characteristics of each spatial node, calculate the future spatial supply and demand matching degree corresponding to the target water network.

[0214] Specifically, for each regional node i, calculate the ratio Ri = Si / Di of its water supply Si and demand Di.

[0215] Based on the spatial node characteristics corresponding to each regional node, clustering algorithms (such as K-means clustering) are used to divide the regional nodes in the target water network into different categories. For example, clustering can be performed based on features such as the geographical location and water use type of regional nodes, so that regional nodes in the same category have similar spatial characteristics and functions.

[0216] For each cluster, its weight can be determined based on factors such as the number of regional nodes it contains, the total water resources, and the importance of water supply or use. For example, a cluster containing multiple large industrial water-using regional nodes, whose water demand accounts for a significant proportion of the entire water network, would have a relatively high weight. Assuming there are k clusters in total, and cluster j has a weight of c... j For region node i, if it belongs to cluster j, its weight w i =c j ×1 / n j , where n j This represents the number of region nodes in cluster j. This method considers both the overall importance of the cluster to which a region node belongs and its relative importance within the cluster, thus reflecting the node's position in the water network in greater detail.

[0217] To comprehensively consider the spatial distribution of the entire water network, a spatial weight matrix W can be introduced, which reflects the spatial correlation between nodes in each region. For example, the weights between adjacent nodes can be set higher, while the weights of nodes that are farther apart can be set lower.

[0218] Then, based on the formula: The future spatial supply and demand matching degree corresponding to the target water network is calculated, where MS is the spatial supply and demand matching degree and n is the number of nodes in the region. This formula means that the spatial weight matrix measures the degree of difference in the supply and demand ratio between different nodes; the smaller the difference, the higher the spatial supply and demand matching degree.

[0219] Step d36: Based on the characteristics of each time node, calculate the future time supply and demand matching degree of the target water network.

[0220] Specifically, the characteristics of each time point are input into a preset time series model to predict the supply St and demand D t of each regional node at each time point t.

[0221] Based on formula Calculate the future supply-demand matching degree corresponding to the target water network. Here, T represents the number of predicted time periods. The smaller the value of MT, the higher the future supply-demand matching degree.

[0222] Step d4: Based on the future spatial supply and demand matching degree and the future temporal supply and demand matching degree, determine the future water resource supply and demand matching degree corresponding to the target water network.

[0223] Specifically, the electronic device uses the future spatial supply-demand matching degree and the future temporal supply-demand matching degree as the evaluation factor set U = {u1, u2}, where u1 represents the future spatial supply-demand matching degree and u2 represents the future temporal supply-demand matching degree. Based on the actual situation, a comment set V = {v1, v2, v3, v4, v5} is established, which could be, for example, very good, excellent, average, poor, very poor. For each evaluation factor, its membership function with each level in the comment set is determined. For example, for the future spatial supply-demand matching degree u1, based on its numerical range, its membership function is determined to belong to "very good," "good," "average," "poor," and "very poor." Similarly, a corresponding membership function is determined for the future temporal supply-demand matching degree u2. Then, based on the membership function, the membership degrees of the future spatial supply-demand matching degree and the future temporal supply-demand matching degree with each level in the comment set are calculated, constructing a fuzzy relation matrix R. Each row of matrix R represents the membership vector of an evaluation factor with different comment levels.

[0224] The weight vector A = {a1, a2} is determined for the future spatial supply and demand matching degree and the future temporal supply and demand matching degree, where a1 + a2 = 1. The weights can be determined with reference to the actual situation of the water network and relevant expert opinions. Through fuzzy matrix multiplication, the weight vector A is multiplied by the fuzzy relation matrix R to obtain the comprehensive evaluation vector B = A·R. The elements in vector B represent the membership degree of the target water network's future water resource supply and demand matching degree to each level in the evaluation set. Then, based on the principle of maximum membership, the level to which the future water resource supply and demand matching degree belongs can be determined, or a comprehensive numerical value can be calculated to represent the matching degree. For example, by assigning corresponding values ​​to each level in the evaluation set, multiplying them by the membership degree, and summing the results, a specific numerical value for the future water resource supply and demand matching degree can be obtained.

[0225] Step S2033: Based on the current water resource supply and demand matching degree and the future water resource supply and demand matching degree, determine the target water resource supply and demand matching degree corresponding to the target water network.

[0226] Specifically, the current water resource supply and demand matching degree and the future water resource supply and demand matching degree are combined to determine the target water resource supply and demand matching degree corresponding to the target water network.

[0227] Step S204: Based on the target water resource supply and demand matching degree, the pattern of the target water network is planned and adjusted.

[0228] Please refer to the above description of step S104 for details on this step, which will not be repeated here.

[0229] The water network planning method provided in this application acquires future meteorological observation data and current hydrological monitoring data for each water body. Based on the water body characteristics, future meteorological observation data, and current hydrological monitoring data, node attribute information is added to the water body nodes and edge attribute information is added to the edges of each channel, constructing an initial water system network model. This facilitates the analysis of the interaction and influence mechanisms between various factors, laying the foundation for in-depth research on the dynamic changes of the water system. The water body characteristics, future meteorological observation data, and current hydrological monitoring data for each water body are input into a preset prediction model, which outputs the hydrological dynamic change data for each water body within a preset future time period. This helps to understand the possible changes in the water system in the future, providing forward-looking information for decisions on water resource management, flood control and disaster reduction, and ecological protection. The hydrological dynamic change data for each water body is fused with the initial water system network model to generate a target water system network model for the target water network. This allows the generated target water system network model to reflect the changes in the water system in real time and dynamically. The target water system network model integrates current and future water system information, providing a more comprehensive and accurate basis for decision-making in water resource management, water ecological protection, and water conservancy project planning. Through model analysis and simulation, the effects of different decision-making schemes can be evaluated, the advantages and disadvantages of various potential measures can be compared, and the optimal solution can be selected, thereby improving the scientific and rational nature of decision-making and promoting the sustainable use of water resources and the healthy development of the water system's ecological environment.

[0230] Based on the connectivity between water bodies, an initial adjacency matrix is ​​generated for each water body, providing an intuitive and concise matrix representation of the connectivity relationships within the target water network. Based on the water body characteristics, future meteorological observation data, current hydrological monitoring data, and hydrological dynamic change data corresponding to each water body in the target water network model, a dynamic data list is generated for each water body. This achieves deep integration of water network-related data, providing a rich information foundation for comprehensive analysis of water network behavior. Furthermore, the dynamic data list accurately reflects the state changes of each water body under different time and environmental conditions, providing strong support for prediction and decision-making. Each dynamic data list is associated with the initial adjacency matrix to generate a backup adjacency matrix, enriching its meaning. This allows the matrix to not only reflect the connectivity relationships between water bodies but also associate the dynamic attribute information of each water body. Based on the water body characteristics, future meteorological observation data, current hydrological monitoring data, and hydrological dynamic change data corresponding to each water body, the weight information corresponding to each edge in the backup adjacency matrix is ​​calculated to generate the target adjacency matrix. This achieves precise quantification of the connectivity relationships between water bodies. The weighting comprehensively considers various factors such as water flow resistance, water conveyance cost, and connectivity stability, enabling the matrix to more realistically reflect the actual water flow transmission situation in the water network. Furthermore, the target adjacency matrix provides core data support for scientific decision-making in water networks. In water resource planning and water network optimization, the weight information based on the matrix can accurately assess the advantages and disadvantages of different connection paths, and formulate optimal water resource allocation schemes and water conservancy project construction plans. Simultaneously, by dynamically updating the weights, changes in the water network can be reflected in real time, allowing decisions to adapt to the constantly changing water network environment and ensuring the scientific and effective management of water resources.

[0231] Based on the target adjacency matrix, the current connectivity index and the predicted connectivity index for the target water network within a preset future timeframe are calculated. This accurately quantifies the current connectivity of the target water network. Calculating the predicted connectivity index for the preset future timeframe allows for the prediction of the water network's connectivity development trend over a future period based on existing data and models. The target connectivity index is generated based on the current and predicted connectivity indices. This provides a comprehensive assessment of the target water network's connectivity from both static and dynamic dimensions. This comprehensive assessment method avoids the limitations of considering only the current state while ignoring future changes, or relying solely on predictions while ignoring current realities, providing a more accurate and comprehensive reference for the long-term planning and management of water networks. Based on the target adjacency matrix, the current shortest flow path and the predicted shortest flow path for the target water network within a preset future timeframe are calculated. Determining the current shortest flow path helps achieve efficient water resource allocation in the present. Predicting the predicted shortest flow path for the preset future timeframe takes into account the water flow direction of the water network under different future conditions. Based on the current and predicted shortest flow paths, a target shortest flow path is generated, providing a complete water flow path scheme for water network management. This scheme considers both the current actual situation of the water network and potential future changes. Based on the current and predicted shortest flow paths, the current water flow efficiency and the predicted water flow efficiency within a preset future timeframe for the target water network are calculated. Calculating the current water flow efficiency helps understand the water flow operation status of various parts of the water network, identify areas with excessively slow flow velocity or excessive energy loss, and take timely optimization measures to improve the overall operational efficiency of the water network. Calculating the predicted water flow efficiency allows for early prediction of potential efficiency problems the water network may face in the future. Based on the current and predicted water flow efficiencies, a target water flow efficiency is generated. This provides an important reference indicator for the long-term operation and management of the water network. Long-term maintenance and upgrade plans can be formulated based on the target water flow efficiency to ensure that the water network maintains high operational efficiency at different times, achieving efficient utilization and sustainable management of water resources.

[0232] Based on the current water system connectivity results, the current water resource supply and demand matching degree of the target water network is calculated. This accurately reflects the balance between water resource supply and demand of the target water network at the current moment. Future time characteristic data for a preset future duration corresponding to the target area, as well as the future water resource demand of the target area, are obtained. Based on the water body characteristics of each water body, future meteorological observation data, and current hydrological monitoring data, the future water resource supply of the target water network is calculated. Using each water-using area in the target water network as a regional node and the water flow channels between each water-using area as edges, weights are assigned to each edge according to the future water system connectivity results, constructing a basic graph structure. This can intuitively present the water flow relationships and connectivity between each water-using area in the target water network. According to the time step, based on future time characteristic data, future water resource demand, and future water resource supply, the state of each regional node and the weights of each edge in the basic graph structure are adjusted to generate multiple spatiotemporal graphs based on time series, which can fully consider the dynamic changes in water resource supply and demand over time. A multi-layer spatiotemporal graph convolutional network is constructed, and the spatial node features corresponding to each region node in the basic graph structure are analyzed through graph convolution operations. This effectively mines the spatial correlation features between different water-using areas. Temporal convolutional layers are used to perform convolution operations on each spatiotemporal graph to obtain the temporal node features corresponding to each region node in the basic graph structure. This captures the characteristics and patterns of water resource supply and demand over time. This helps to understand the evolution patterns of water resource supply and demand over time and provides important temporal information for predicting future supply and demand. By analyzing the temporal node features, the matching degree of water resource supply and demand at different time scales can be assessed, providing support for formulating reasonable time scheduling strategies. Based on the spatial node features, the future spatial supply and demand matching degree of the target water network is calculated; this allows for the assessment of the matching of water resource supply and demand among different water-using areas in the target water network from a spatial dimension, clarifying the spatial distribution of water resource surplus and deficit in different areas. Based on the temporal node features, the future temporal supply and demand matching degree of the target water network is calculated. This allows for the assessment of the matching degree of water resource supply and demand at different times from a temporal dimension, helping to formulate water resource scheduling plans adapted to different time scales. Based on the future spatial and temporal supply-demand matching degrees, the future water resource supply-demand matching degree corresponding to the target water network is determined. Based on the current and future water resource supply-demand matching degrees, the target water resource supply-demand matching degree corresponding to the target water network is determined. This achieves a comprehensive assessment of the water resource supply and demand situation of the target water network from the present to the future. This assessment method avoids the limitations of focusing only on a single current or future stage, comprehensively considering the dynamic changes in water network operation, and providing a more systematic and comprehensive perspective for water resource management. Furthermore, the target water resource supply-demand matching degree integrates the current actual situation and future development trends, providing a more scientific and reliable basis for water resource management decisions.

[0233] This embodiment also provides a water network pattern planning device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0234] This embodiment provides a water network pattern planning device, such as Figure 3 As shown, it includes:

[0235] The acquisition module 301 is used to acquire at least one water body corresponding to the target area and the water body characteristics corresponding to each water body; the water body characteristics include water body location pattern characteristics, water body morphology characteristics, water body structure characteristics, and water body flow characteristics;

[0236] The first calculation module 302 is used to calculate the target water system connectivity result of the target water network corresponding to the target area based on the water body characteristics of each water body; the target water network is based on the composition of each water body.

[0237] The second calculation module 303 is used to calculate the target water resource supply and demand matching degree corresponding to the target water network based on the target water system connectivity results.

[0238] The adjustment module 304 is used to plan and adjust the pattern of the target water network based on the target water resource supply and demand matching degree.

[0239] This invention also provides an electronic device having the above-described features. Figure 3 The water network pattern planning device is shown. Please refer to [link / reference]. Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an optional embodiment of the present invention.

[0240] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for planning a water network pattern, characterized in that, The methods include: Obtain at least one water body corresponding to the target area and the water body characteristics corresponding to each water body; Water body characteristics include water body location and pattern characteristics, water body morphological characteristics, water body structural characteristics, and water body flow characteristics; Acquire future meteorological observation data and current hydrological monitoring data for each water body; Based on the water body characteristics of each water body, future meteorological observation data and current hydrological monitoring data, node attribute information is added to the water body nodes corresponding to each water body, and edge attribute information is added to the edges corresponding to each channel to construct an initial water system network model. Input the water characteristics of each water body, future meteorological observation data and current hydrological monitoring data into the preset prediction model, and output the dynamic hydrological change data of each water body within a preset time period in the future. The hydrological dynamic change data corresponding to each water body are fused with the initial water system network model to generate the target water system network model corresponding to the target water network. The target water network is based on the composition of each water body; Based on the connection status between each water body, generate the initial adjacency matrix corresponding to each water body; Based on the water body characteristics, future meteorological observation data, current hydrological monitoring data and hydrological dynamic change data of each water body in the target water system network model, a dynamic data list corresponding to each water body is generated. Associate each dynamic data list with the initial adjacency matrix to generate a backup adjacency matrix; Based on the water characteristics of each water body, future meteorological observation data, current hydrological monitoring data, and hydrological dynamic change data, calculate the weight information corresponding to each edge in the backup adjacency matrix and generate the target adjacency matrix. Based on the target adjacency matrix, calculate the target connectivity index, target shortest flow path, and target flow efficiency corresponding to the target water network. Based on the target connectivity index, the target shortest flow path, and the target flow efficiency, the target water system connectivity result corresponding to the target water network is determined. The target water system connectivity results include the current water system connectivity results and the future water system connectivity results within a preset time period; The current water system connectivity result is obtained by fusing the current connectivity index in the target connectivity index, the current shortest flow path in the target shortest flow path, and the current flow efficiency in the target flow efficiency. The future water system connectivity result is obtained by fusing the future connectivity index in the target connectivity index, the future shortest flow path in the target shortest flow path, and the future flow efficiency in the target flow efficiency. Based on the current water system connectivity results, calculate the current water resource supply and demand matching degree corresponding to the target water network; Based on the future water system connectivity results, calculate the future water resource supply and demand matching degree corresponding to the target water network; Based on the current water resource supply and demand matching degree and the future water resource supply and demand matching degree, determine the target water resource supply and demand matching degree corresponding to the target water network; Based on the target water resource supply and demand matching degree, the pattern of the target water network is planned and adjusted.

2. The method according to claim 1, characterized in that, The step of calculating the target connectivity index, the target shortest flow path, and the target flow efficiency corresponding to the target water network based on the target adjacency matrix includes: Based on the target adjacency matrix, calculate the current connectivity index of the target water network and the predicted connectivity index within a preset time period in the future; The target connectivity index is generated based on the current connectivity index and the predicted connectivity index. Based on the target adjacency matrix, calculate the current shortest flow path corresponding to the target water network and the predicted shortest flow path within a preset time period in the future. The target shortest flow path is generated based on the current shortest flow path and the predicted shortest flow path. Based on the current shortest flow path and the predicted shortest flow path, calculate the current flow efficiency of the target water network and the predicted flow efficiency within a preset time period in the future. The target flow efficiency is generated based on the current flow efficiency and the predicted flow efficiency.

3. The method according to claim 2, characterized in that, The calculation of the future water resource supply and demand matching degree corresponding to the target water network based on the future water system connectivity results includes: Obtain future time feature data for a predetermined duration corresponding to the target area, as well as the future water resource demand corresponding to the target area; Based on the characteristics of each water body, future meteorological observation data, and current hydrological monitoring data, the future water resource supply corresponding to the target water network is calculated. Based on the future time characteristic data, the future water resource demand, the future water resource supply, and the future water system connectivity results, calculate the future spatial supply and demand matching degree and the future time supply and demand matching degree corresponding to the target water network. Based on the future spatial supply and demand matching degree and the future temporal supply and demand matching degree, the future water resource supply and demand matching degree corresponding to the target water network is determined.

4. The method according to claim 3, characterized in that, The calculation of the future spatial supply-demand matching degree and future temporal supply-demand matching degree corresponding to the target water network, based on the future time characteristic data, the future water resource demand, the future water resource supply, and the future water system connectivity results, includes: Using each water-using area in the target water network as a region node, and the water flow channels between each water-using area as edges, a basic graph structure is constructed by assigning weights to each edge according to the future water system connectivity results. According to the time step, based on the future time feature data, the future water resource demand, and the future water resource supply, the state of each regional node and the weight of each edge in the basic graph structure are adjusted to generate multiple spatiotemporal graphs based on time series. A multi-layer spatiotemporal graph convolutional network is constructed, and the spatial node features corresponding to each region node in the basic graph structure are obtained through graph convolution operations. The spatiotemporal graphs are convolved using temporal convolutional layers to obtain the temporal node features corresponding to each region node in the spatiotemporal graphs. Based on the characteristics of each spatial node, the future spatial supply and demand matching degree corresponding to the target water network is calculated; Based on the characteristics of each time node, the future time supply and demand matching degree corresponding to the target water network is calculated.

5. A water network pattern planning device, characterized in that, The device includes: The acquisition module is used to acquire at least one water body corresponding to the target area and the water body characteristics corresponding to each water body. Water body characteristics include water body location and pattern characteristics, water body morphological characteristics, water body structural characteristics, and water body flow characteristics; The first calculation module is used to obtain future meteorological observation data and current hydrological monitoring data for each water body. Based on the water body characteristics of each water body, future meteorological observation data and current hydrological monitoring data, node attribute information is added to the water body nodes corresponding to each water body, and edge attribute information is added to the edges corresponding to each channel to construct an initial water system network model. The water body characteristics, future meteorological observation data, and current hydrological monitoring data of each water body are input into the preset prediction model, and the hydrological dynamic change data of each water body within the preset time period are output; the hydrological dynamic change data of each water body is fused with the initial water system network model to generate the target water system network model corresponding to the target water network. The target water network is based on the composition of each water body; according to the connection status between each water body, an initial adjacency matrix is ​​generated for each water body. Based on the water body characteristics, future meteorological observation data, current hydrological monitoring data, and hydrological dynamic change data corresponding to each water body in the target water system network model, a dynamic data list corresponding to each water body is generated; each dynamic data list is associated with the initial adjacency matrix to generate a backup adjacency matrix; based on the water body characteristics, future meteorological observation data, current hydrological monitoring data, and hydrological dynamic change data corresponding to each water body, the weight information corresponding to each edge in the backup adjacency matrix is ​​calculated to generate the target adjacency matrix; Based on the target adjacency matrix, calculate the target connectivity index, target shortest flow path, and target flow efficiency corresponding to the target water network. Based on the target connectivity index, the target shortest flow path, and the target flow efficiency, the target water system connectivity result corresponding to the target water network is determined. The target water system connectivity results include the current water system connectivity results and the future water system connectivity results within a preset time period; The current water system connectivity result is obtained by fusing the current connectivity index in the target connectivity index, the current shortest flow path in the target shortest flow path, and the current flow efficiency in the target flow efficiency. The future water system connectivity result is obtained by fusing the future connectivity index in the target connectivity index, the future shortest flow path in the target shortest flow path, and the future flow efficiency in the target flow efficiency. The second calculation module is used to calculate the current water resource supply and demand matching degree corresponding to the target water network based on the current water system connectivity results. Based on the future water system connectivity results, calculate the future water resource supply and demand matching degree corresponding to the target water network; Based on the current water resource supply and demand matching degree and the future water resource supply and demand matching degree, determine the target water resource supply and demand matching degree corresponding to the target water network; The adjustment module is used to plan and adjust the pattern of the target water network based on the target water resource supply and demand matching degree.

6. An electronic device, characterized in that, include: A memory and a processor are interconnected, the memory stores computer instructions, and the processor executes the water network pattern planning method according to any one of claims 1 to 4 by executing the computer instructions.

Images