Method and device for determining water surface rate configuration scheme

Abstract

The invention relates to the technical field of water surface rate configuration, in particular to a method and device for determining a water surface rate configuration scheme. Obtaining regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network region; inputting the regional meteorological data, the hydrological data, the water quality data and the underlying surface factor data into a preset water surface rate configuration scheme model, and outputting a target water surface rate configuration scheme corresponding to the target water network region; and configuring the water surface rate corresponding to the target water network area based on the target water surface rate configuration scheme. The scientific and reasonable water surface rate configuration scheme is beneficial to balance the supply-demand relationship of water resources and guarantee the economic development and ecological stability of the target water network area. By optimizing the configuration of the water surface rate, the current requirement for water resources can be met, the water resources and the ecological environment can be protected, a foundation is laid for long-term sustainable development of a region, and coordinated development of economy, society and environment is achieved. Reasonable water surface rate configuration of the target water network area is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of water surface rate configuration, and in particular to a method and device for determining a water surface rate configuration scheme. Background Art

[0002] Water network areas are widely distributed around the world and are the core carriers for water storage, allocation and ecological service functions. In cities, water networks not only bear the task of supplying domestic and industrial water, but also play an important role in regulating urban microclimate and beautifying the environment. In rural and natural ecosystems, water networks support agricultural irrigation, maintain rich biodiversity, and are an important guarantee for ecological balance. However, with the global population growth, accelerated urbanization and the impact of climate change, water network areas are facing many severe challenges such as water shortages, flood disasters and ecological environmental deterioration.

[0003] However, traditional water surface ratio configuration technology mainly relies on empirical formulas and simple model simulations, which makes it difficult to accurately analyze and predict complex water network structures and diverse boundary conditions. The lack of support from advanced information technology and data analysis methods has led to significant limitations in data collection, processing, and solution optimization, and is unable to meet the high-precision requirements of modern water network management. To sum up, how to configure the water surface ratio in a scientific, reasonable, comprehensive and dynamically adaptable manner in the water network area has become a key issue that needs to be urgently solved in the current fields of water conservancy engineering, ecological environment and urban planning. It has extremely important practical significance for ensuring the sustainable use of water resources, maintaining ecological security and promoting sustainable social and economic development. Summary of the invention

[0004] In view of this, the present invention provides a method and device for determining a water surface ratio configuration scheme to solve the problem of how to configure the water surface ratio of a water network area in a scientific, reasonable, comprehensive, and dynamically adaptable manner.

[0005] In a first aspect, the present invention provides a method for determining a water surface rate configuration scheme, the method comprising: Acquire regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area; the underlying surface factor data includes at least one of regional topographic data, regional soil type and property data, and regional vegetation type and coverage rate data; Input regional meteorological data, hydrological data, water quality data and underlying surface factor data into the preset water surface ratio configuration scheme model, and output the target water surface ratio configuration scheme corresponding to the target water network area; The water surface rate corresponding to the target water network area is configured based on the target water surface rate configuration plan.

[0006] The method for determining the water surface rate configuration scheme provided in the embodiment of the present application obtains the regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area, provides a comprehensive and accurate basis for the water surface rate configuration scheme, makes the target water surface rate configuration scheme output by the preset water surface rate configuration scheme model more in line with the actual situation, and improves the scientificity and accuracy of the configuration scheme. Then, the regional meteorological data, hydrological data, water quality data and underlying surface factor data are input into the preset water surface rate configuration scheme model, and the target water surface rate configuration scheme corresponding to the target water network area is output, which ensures the accuracy and scientificity of the target water surface rate configuration scheme corresponding to the output target water network area. The water surface rate corresponding to the target water network area is configured based on the target water surface rate configuration scheme. A scientific and reasonable water surface rate configuration scheme helps to balance the supply and demand relationship of water resources and ensure the economic development and ecological stability of the target water network area. By optimizing the water surface rate configuration, it can not only meet the current demand for water resources, but also protect water resources and the ecological environment, lay the foundation for the long-term sustainable development of the region, and achieve the coordinated development of economy, society and environment. A reasonable water surface rate configuration is achieved for the target water network area.

[0007] In an optional implementation, the preset water surface rate configuration scheme model includes a feature extraction model and a water surface rate configuration scheme model, and the regional meteorological data, hydrological data, water quality data and underlying surface factor data are input into the preset water surface rate configuration scheme model, and the target water surface rate configuration scheme corresponding to the target water network area is output, including: Input regional meteorological data, hydrological data, water quality data and underlying surface factor data into the feature extraction model; The feature extraction model identifies the input data and determines the data types of various data in the input data; the input data are regional meteorological data, hydrological data, water quality data and underlying surface factor data; Set up multi-channel input according to the data type of various data; Based on each input channel, feature extraction is performed on various types of input data, and feature representation is output; The feature representation is input into the water surface ratio configuration scheme model, and the target water surface ratio configuration scheme corresponding to the target water network area is output based on the feature representation.

[0008] The method for determining the water surface ratio configuration scheme provided in the embodiment of the present application inputs regional meteorological data, hydrological data, water quality data and underlying surface factor data into a feature extraction model. The feature extraction model identifies the input data, determines the data types of various data in the input data, and ensures the accuracy of the data types of various data in the determined input data. Then, a multi-channel input is set according to the data types of various data. Thus, the feature extraction model can perform targeted processing according to the characteristics of different types of data. Multi-channel input allows the feature extraction model to adopt processing methods suitable for each data type respectively, avoids the loss of key information due to unified processing, ensures the effective use of data, and lays the foundation for subsequent accurate feature extraction. Then, the feature extraction model extracts features from various types of input data based on each input channel and outputs feature representation. Thus, the comprehensiveness and accuracy of the feature representation are guaranteed. These rich feature representations more comprehensively reflect the actual conditions of the target water network area, provide more detailed and accurate information for the water surface ratio configuration scheme model, and thus improve the reliability of the scheme. Then, the feature representation is input into the water surface ratio configuration scheme model, and the target water surface ratio configuration scheme corresponding to the target water network area is output based on the feature representation. The accuracy of the target water surface rate configuration plan corresponding to the output target water network area is guaranteed.

[0009] In an optional implementation, based on each input channel, feature extraction is performed on various types of input data, and a feature representation is output, including: The image data in the input data is input as image channel; For the numerical data in the input data, the numerical data is converted into a two-dimensional matrix form and channels are allocated according to the feature dimensions; Use multiple convolution kernels of different sizes to perform convolution processing on the data input from each input channel to extract target features of different scales; The target features extracted from different levels and types of data are fused according to the channel dimension to generate feature representation.

[0010] The method for determining the water surface ratio configuration scheme provided in the embodiment of the present application directly uses the image data as the image channel input, and completely retains the spatial information and visual features of the image. Converting the numerical data into a two-dimensional matrix and allocating channels according to the feature dimension can structure the abstract numerical information, which is convenient for model recognition and processing. Then, the input channel data is convoluted using multiple convolution kernels of different sizes to extract target features of different scales. Small-size convolution kernels focus on local details. When processing image data, they can accurately identify subtle features such as the edge of the water body and small ditches; large-size convolution kernels focus on global information and grasp the macro layout and structure of the entire water network. This multi-scale feature extraction method comprehensively covers various feature information of the data and enhances the depth and breadth of understanding of the data. Finally, the target features extracted from different levels and types of data are fused according to the channel dimension, and the generated feature representation combines the advantages of multi-source data. The spatial features of image data and the quantitative features of numerical data complement each other, providing the model with richer and more comprehensive information. When determining the water surface ratio configuration plan, the fused features can reflect not only the actual geographical form of the water network, but also the changes in related hydrological, meteorological and other data, making the water surface ratio configuration model decision more scientific and accurate. In addition, rich and comprehensive feature representation helps to improve the performance of the water surface ratio configuration model in subsequent tasks. The water surface ratio configuration model is analyzed based on the fused feature representation, which can more accurately capture the complex relationships and potential laws between data, improve prediction accuracy and evaluation reliability. At the same time, the introduction of multi-scale features enhances the adaptability of the model to different situations, so that it can still maintain good performance when facing complex and changing data.

[0011] In an optional implementation, the water surface rate configuration scheme model includes a plurality of sub-water surface rate configuration scheme models of different algorithms, the feature representation is input into the water surface rate configuration scheme model, and a target water surface rate configuration scheme corresponding to the target water network area is output based on the feature representation, including: Input the characteristic representation into each sub-water surface rate configuration scheme model respectively; Each sub-water surface rate configuration scheme model identifies the feature representation and outputs a plurality of first candidate water surface rate configuration schemes respectively; fusing the first candidate water surface rate configuration schemes to generate a plurality of first fused water surface rate configuration schemes; Inputting each first fused water surface rate configuration scheme into each sub-water surface rate configuration scheme model respectively; Each sub-water surface rate configuration scheme model evaluates and updates each first fused water surface rate configuration scheme, and outputs a plurality of second candidate water surface rate configuration schemes; This cycle is repeated until a preset number of times, and each sub-water surface rate configuration scheme model outputs its corresponding current optimal water surface rate configuration scheme; Identify the current optimal water surface rate configuration schemes and output the target water surface rate configuration scheme corresponding to the target water network area.

[0012] The water surface rate configuration scheme determination method provided in the embodiment of the present application, the water surface rate configuration scheme model includes a plurality of sub-water surface rate configuration scheme models of different algorithms, and each sub-water surface rate configuration scheme model processes the feature representation based on different algorithm principles and optimization strategies. Multiple algorithms are run in parallel to explore the possibility of water surface rate configuration from different angles. The output of multiple first candidate water surface rate configuration schemes has high diversity, which can cover various possible configuration situations more comprehensively, avoid the limitations of a single algorithm, and lay the foundation for the subsequent generation of better solutions. Then, multiple first candidate water surface rate configuration schemes are fused to generate a first fused water surface rate configuration scheme. The fusion process can integrate the advantages of different first candidate water surface rate configuration schemes. This fusion is not a simple superposition, but through certain rules and algorithms, it gives full play to the strengths of each scheme, makes up for the shortcomings, and improves the comprehensive performance of the scheme, which is more in line with the needs of multi-objective optimization of water surface rate configuration in practical applications. Next, the first fused water surface rate configuration scheme is input into each sub-water surface rate configuration scheme model again for evaluation and update, and the second candidate water surface rate configuration scheme is output, and this process is continuously cycled. Each cycle is an optimization and improvement of the previous solution. The sub-water surface rate configuration scheme model re-evaluates and adjusts the fusion scheme based on its own algorithm, gradually tapping the potential of the scheme and making the scheme approach the optimal solution. As the number of cycles increases, the target water surface rate configuration scheme performs better and better in meeting multiple objectives such as flood control, water resource utilization, and ecological protection, improving the accuracy and effectiveness of the water surface rate configuration scheme. During the cycle, each sub-water surface rate configuration scheme model continuously adapts to the newly input fusion scheme and evaluates and adjusts different situations. This enables the entire model system to cope with complex and changeable water network environments and data characteristics. Even in the face of data uncertainty and changes in the actual situation of the water network area, the water surface rate configuration scheme model can output a relatively stable and effective target water surface rate configuration scheme through multiple cycles and adjustments, enhancing the adaptability and robustness of the water surface rate configuration scheme model, and ensuring that a reliable target water surface rate configuration scheme can be provided in different scenarios.

[0013] In an optional implementation, each sub-water surface rate configuration scheme model identifies the feature representation and outputs a plurality of first candidate water surface rate configuration schemes, including: For one of the sub-water surface rate configuration scheme models, the sub-water surface rate configuration scheme model identifies the feature representation; Based on the target constraint conditions in the sub-water surface rate configuration scheme model, multiple initial water surface rate configuration schemes are generated; the target constraint conditions include at least one of the water surface rate range constraint, the water quality standard constraint, the water level and flood storage capacity constraint, and the land use constraint; Based on the characteristic representation, each initial water surface ratio configuration scheme is substituted into the objective function in the sub-water surface ratio configuration scheme model, and the objective function value corresponding to each initial water surface ratio configuration scheme is calculated; wherein the objective function includes at least one of the flood control and drainage sub-function, the water resource utilization sub-function, and the water environment capacity sub-function; According to the objective function values ​​corresponding to each initial water surface rate configuration scheme, determine the historical optimal water surface rate configuration scheme and the current global optimal water surface rate configuration scheme; Determine the update speed based on the historical optimal water surface rate configuration plan and the current global optimal water surface rate configuration plan; According to the update speed, each updated water surface rate configuration scheme is obtained; each updated water surface rate configuration scheme satisfies the target constraint; Calculate the objective function value corresponding to each updated water surface rate configuration scheme; Determine the updated water surface rate configuration scheme whose objective function value is greater than the preset function value as the first candidate water surface rate configuration scheme; If there is no updated water surface ratio configuration scheme whose objective function value is greater than the preset function value, the updated water surface ratio configuration scheme continues to be updated until the updated water surface ratio configuration scheme whose objective function value is greater than the preset function value is determined as the first candidate water surface ratio configuration scheme.

[0014] The method for determining the water surface rate configuration scheme provided in the embodiment of the present application generates an initial water surface rate configuration scheme based on target constraints, which fundamentally guarantees the feasibility of the scheme in practical applications. It avoids the situation that the generated scheme cannot be implemented due to violation of actual constraints. Based on the feature representation, each initial water surface rate configuration scheme is substituted into the objective function in the sub-water surface rate configuration scheme model, and the objective function value corresponding to each initial water surface rate configuration scheme is calculated, so that the advantages and disadvantages of each initial water surface rate configuration scheme can be evaluated from multiple angles. When evaluating a scheme, not only can its effect on flood control and waterlogging removal be understood, but also its impact on water resource utilization and water environment capacity can be known, thereby realizing multi-objective optimization of the water surface rate configuration scheme, so that the generated scheme is more in line with the comprehensive needs of sustainable development of the water network area. Then, the historical optimal and current global optimal water surface rate configuration schemes are determined according to the objective function value, and the update speed is determined accordingly, and the updated configuration scheme is continuously obtained. The updated scheme with an objective function value greater than the preset function value is determined as the first candidate water surface rate configuration scheme. This screening mechanism can gradually eliminate undesirable schemes and retain and optimize more advantageous schemes. If there is no solution that meets the preset conditions, it will be continuously updated to ensure that the first candidate water surface ratio configuration solution finally obtained performs well under the multi-objective comprehensive evaluation, thereby improving the quality and practicality of the first candidate water surface ratio configuration solution. In the process of continuous updating and screening, the first candidate water surface ratio configuration solution can better adapt to the complex situation of the water network area. Since each update is based on the previous optimal solution and takes into account the comprehensive impact of multi-objective functions, the generated solution can more accurately meet the actual needs. For a water network area that has both flood control needs and water shortage problems, this mechanism can find a first candidate water surface ratio configuration solution that achieves a good balance between flood control and waterlogging and water resource utilization through multiple iterative optimizations.

[0015] In an optional implementation, the target water network area includes multiple water bodies, and the objective function includes:

[0016] in, is the flood control and waterlogging removal sub-function; The water body in the target water network area i area; For water The flood storage capacity is as follows: ;in For water The water level, is the flood storage coefficient; and / or,

[0017] in, is the water resource utilization sub-function, For water The available water volume is calculated as: ; Where P is water The precipitation of E is the water body evaporation, β is the regional hydrological conversion coefficient; For water The water use efficiency is defined as: ;in, For agricultural irrigation needs; and / or,

[0018] in, is the water environment capacity subfunction; Pi is the water body The pollution load is usually related to land use and emissions; Vi is the water body The environmental capacity is calculated as: ; Among them, γ is the comprehensive degradation coefficient, which is related to the water body The dilution capacity of is the area of ​​water body i, is the water level of water body i, It is the standard concentration of water quality.

[0019] The method for determining the water surface ratio configuration scheme provided in the embodiment of the present application, the objective function includes at least one of the flood control and drainage sub-function, the water resource utilization sub-function, and the water environment capacity sub-function. Different sub-functions represent the key needs of different aspects of the water network area, covering sub-functions such as flood control and drainage, water resource utilization, and water environment capacity. The functions of the water network in flood control, water supply, ecology, etc. can be comprehensively considered to avoid the one-sided pursuit of a single goal and ignore other aspects. For example, when determining the water surface ratio, if only water resource utilization is considered, the water surface area is increased to store water resources, but the risk of flood disasters may increase during the flood period due to neglect of flood control and drainage. Comprehensively considering multiple sub-functions, it is possible to coordinate and balance and formulate a water surface ratio configuration scheme that is more in line with the overall interests of the region. In addition, each sub-function quantitatively evaluates the water surface ratio configuration scheme from different angles to provide a basis for multi-objective optimization. By calculating the values ​​of each scheme under different sub-functions, the performance of the scheme on each target can be clearly understood, and then the water surface ratio configuration can be adjusted through optimization algorithms and other means to seek a balance between multiple targets. In the flood control and drainage subfunction, the flood storage capacity and flood risk reduction degree of the region under different water surface rates can be calculated; in the water resource utilization subfunction, the efficiency of water resource utilization and the degree of meeting water demand can be evaluated. On this basis, the water surface rate configuration that can achieve multiple goals at a better level can be found to achieve the coordinated development of water resources, ecological environment and social economy. In addition, the main problems and needs faced by water networks in different regions are different. The objective function can be constructed by selecting appropriate subfunction combinations to enhance the adaptability of the scheme to specific regions. In areas where floods occur frequently, the flood control and drainage subfunction is the main one, and other subfunctions are appropriately taken into account to ensure that the configuration scheme can effectively resist floods and protect people’s lives and property. In areas where water resources are scarce, the focus is on the water resource utilization subfunction, and the water surface rate configuration is optimized to improve the efficiency of water resource utilization. This approach of adapting measures to local conditions makes the water surface rate configuration scheme more scientific and practical, and can effectively solve the actual water problems in the region.

[0020] In an optional implementation, the target water network area includes multiple water bodies, and the target constraint conditions include water surface rate range constraint, water quality standard constraint, water level and flood storage capacity constraint, and land use constraint. The water surface rate range constraint is: ; in, is the total area of ​​the target water network area; For water Water surface rate; is the upper limit of the water surface rate, is the lower limit of the water surface rate; The water quality standards are constrained by: , ; in, For water of dissolved oxygen, For water The chemical oxygen demand, For water Total nitrogen, For water of total phosphorus; Chemical oxygen demand for environmental standards, Total nitrogen for environmental standards, is the environmental standard value of total phosphorus; The water level and flood storage capacity constraints are: ; in, For water The water level, is the maximum permissible water level in the target water network area; The land use constraints are: ; in, is the total area of ​​the channel; is the total area of ​​the pond; is the total area of ​​wetlands; is the total area of ​​agricultural land, is the total area of ​​the target water network area.

[0021] The method for determining the water surface ratio configuration scheme provided in the embodiment of the present application, the target constraint conditions include at least one of the water surface ratio range constraint, the water quality standard constraint, the water level and flood storage capacity constraint and the land use constraint. These constraints are set based on the actual geographical environment, resource conditions and planning requirements to ensure that the generated water surface ratio configuration scheme is feasible in reality. The water surface ratio range constraint determines the upper and lower limits based on the regional topography and planning to avoid the water surface ratio being too high to occupy too much land resources and affect other uses, or being too low to meet the functional requirements of the water network; the land use constraint coordinates the water surface ratio configuration with other land use types such as farmland and wetlands to ensure the rational allocation and efficient use of regional land resources. Water quality standard constraints and water level and flood storage capacity constraints are crucial to protecting the ecological environment and ensuring water quality safety. Water quality standard constraints ensure that water quality meets ecological and use requirements by setting limits on pollutant indicators such as dissolved oxygen and chemical oxygen demand, and maintain the stability and health of the water ecosystem; water level and flood storage capacity constraints ensure that the water level of the water body is within a safe range to avoid damage to the ecological environment and human activities caused by floods, while ensuring the flood storage capacity of the water body, playing a role in regulating and storing when floods come, and reducing flood risks. In addition, comprehensive consideration of these constraints is conducive to achieving sustainable development in the region. Reasonable water surface ratio configuration meets the current needs of flood control, water resource utilization and ecological protection, while taking into account the needs of future development. Through water level and flood storage capacity constraints and water surface ratio range constraints, the layout of the water network is optimized, the efficiency of water resource utilization is improved, and the long-term stable development of the region is guaranteed; land use constraints coordinate the relationship between the water network and other land uses, and promote the coordinated development of the regional economy, society and environment.

[0022] In an optional implementation, each first candidate water surface rate configuration scheme is fused to generate a plurality of first fused water surface rate configuration schemes, including: Calculate the similarity between the first candidate water surface rate configuration schemes; For each first candidate water surface rate configuration scheme whose similarity is greater than a preset similarity threshold, only one is retained to obtain multiple backup water surface rate configuration schemes; The backup water surface rate configuration plans are cross-combined to generate first fused water surface rate configuration plans.

[0023] The method for determining the water surface rate configuration scheme provided in the embodiment of the present application calculates the similarity between each first candidate water surface rate configuration scheme, and only retains one of the schemes with a similarity greater than a preset threshold, which effectively avoids the redundancy of the scheme. Among the many candidate schemes, there may be some schemes that are similar in nature and have similar contributions to the final goal. If no screening is performed, the subsequent processing of these similar schemes will increase the amount of calculation, waste computing resources and time. By removing redundant schemes and retaining representative schemes, unnecessary computing burdens are reduced, the efficiency of the entire scheme optimization process is improved, and subsequent analysis and processing can be more focused on differentiated schemes. While removing similar schemes, multiple different standby water surface rate configuration schemes are retained to ensure the diversity of the schemes. These different schemes each have unique advantages and characteristics, covering different water surface rate configuration ideas and trade-offs for multiple objectives. One scheme may be outstanding in flood control and drainage, and another scheme may be more advantageous in water resource utilization. Retaining these diverse schemes provides rich materials for subsequent cross-combinations, which is conducive to generating a more comprehensive and higher-quality first fusion water surface rate configuration scheme. Then, the cross-combination of the standby water surface rate configuration schemes can integrate the advantages of different schemes and create a more comprehensive and innovative first fusion water surface rate configuration scheme. In the cross-combination process, the characteristics and advantages of different schemes are combined with each other, which may produce new configuration ideas and patterns. The two schemes that perform well in flood control and water resource utilization are cross-combined. The newly generated scheme may achieve a good balance in both aspects, thereby improving the comprehensive performance of the scheme on multiple objectives, which is more in line with the needs of multi-objective optimization of water surface rate configuration in practical applications. As a result, the generated multiple first fusion water surface rate configuration schemes have stronger adaptability and can better meet the complex and changeable actual needs of the target water network area. Due to the integration of the characteristics of different schemes, these fusion schemes can cover a wider range of situations and conditions. When facing the water network operation needs in different seasons and different climatic conditions, the fusion scheme may have better adjustment capabilities and adaptability, providing more high-quality options for the scientific planning and effective management of water network areas.

[0024] In an optional implementation, each current optimal water surface ratio configuration scheme is identified, and a target water surface ratio configuration scheme corresponding to the target water network area is output, including: Input each current optimal water surface rate configuration scheme into each sub-water surface rate configuration scheme model respectively; Each sub-water surface rate configuration scheme model evaluates the current optimal water surface rate configuration scheme, and votes on each current optimal water surface rate configuration scheme based on the evaluation results; The current optimal water surface rate configuration plan with the most votes is determined as the target water surface rate configuration plan.

[0025] The method for determining the water surface rate configuration scheme provided in the embodiment of the present application inputs each current optimal water surface rate configuration scheme into each sub-water surface rate configuration scheme model respectively. Thus, each current optimal water surface rate configuration scheme can be examined from multiple professional perspectives. Each sub-water surface rate configuration scheme model evaluates the performance of each current optimal water surface rate configuration scheme in terms of flood control and drainage, water resource utilization, water environment capacity, etc. according to its own algorithm characteristics, and votes on each current optimal water surface rate configuration scheme according to the evaluation results. The target scheme is determined by combining these evaluation results, which can fully integrate the advantages of different models, avoid the limitations of a single model, and make the final target water surface rate configuration scheme more scientific and reasonable. In addition, multiple sub-water surface rate configuration scheme models are independently evaluated and voted, which can reduce the risk of wrong decisions due to errors or limitations of individual sub-water surface rate configuration scheme models. If only one model is relied on to determine the scheme, the sub-water surface rate configuration scheme model may give unreasonable results due to factors such as data anomalies and algorithm defects. Voting on multiple sub-water surface rate configuration scheme models is similar to group decision-making, which can complement and verify each other. Even if deviations occur in the evaluation of individual models, the correct evaluation of other models can ensure the reliability of the final decision, improve the credibility of the water surface ratio configuration plan, and provide a more reliable basis for its practical application in the target water network area.

[0026] In a second aspect, the present invention provides a device for determining a water surface rate configuration scheme, the device comprising: An acquisition module is used to acquire regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area; the underlying surface factor data includes at least one of regional topographic data, regional soil type and property data, and regional vegetation type and coverage rate data; An input module is used to input regional meteorological data, hydrological data, water quality data and underlying surface factor data into a preset water surface ratio configuration scheme model, and output a target water surface ratio configuration scheme corresponding to the target water network area; The configuration module is used to configure the water surface rate corresponding to the target water network area based on the target water surface rate configuration plan.

[0027] The water surface rate configuration scheme determination device provided in the embodiment of the present application obtains the regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area, provides a comprehensive and accurate basis for the water surface rate configuration scheme, makes the target water surface rate configuration scheme output by the preset water surface rate configuration scheme model more in line with the actual situation, and improves the scientificity and accuracy of the configuration scheme. Then, the regional meteorological data, hydrological data, water quality data and underlying surface factor data are input into the preset water surface rate configuration scheme model, and the target water surface rate configuration scheme corresponding to the target water network area is output, which ensures the accuracy and scientificity of the target water surface rate configuration scheme corresponding to the output target water network area. The water surface rate corresponding to the target water network area is configured based on the target water surface rate configuration scheme. A scientific and reasonable water surface rate configuration scheme helps to balance the supply and demand relationship of water resources and ensure the economic development and ecological stability of the target water network area. By optimizing the water surface rate configuration, it can not only meet the current demand for water resources, but also protect water resources and the ecological environment, lay the foundation for the long-term sustainable development of the region, and achieve the coordinated development of economy, society and environment. A reasonable water surface rate configuration is achieved for the target water network area.

[0028] In a third aspect, the present invention provides an electronic device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method for determining a water surface rate configuration scheme according to the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0029] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for determining a water surface rate configuration scheme according to the first aspect or any corresponding embodiment thereof.

[0030] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions for causing a computer to execute the method for determining a water surface rate configuration scheme according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 is a flow chart of a method for determining a water surface rate configuration scheme according to an embodiment of the present invention; Figure 2is a flow chart of another method for determining a water surface rate configuration scheme according to an embodiment of the present invention; Figure 3 is a structural block diagram of a device for determining a water surface rate configuration scheme according to an embodiment of the present invention; Figure 4 It is a schematic diagram of the hardware structure of the electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0034] Water network areas are widely distributed around the world and are the core carriers for water storage, allocation and ecological service functions. In cities, water networks not only bear the task of supplying domestic and industrial water, but also play an important role in regulating urban microclimate and beautifying the environment. In rural and natural ecosystems, water networks support agricultural irrigation, maintain rich biodiversity, and are an important guarantee for ecological balance. However, with the global population growth, accelerated urbanization and the impact of climate change, water network areas are facing many severe challenges such as water shortages, flood disasters and ecological environmental deterioration.

[0035] However, traditional water surface ratio configuration technology mainly relies on empirical formulas and simple model simulations, which makes it difficult to accurately analyze and predict complex water network structures and diverse boundary conditions. The lack of support from advanced information technology and data analysis methods has led to significant limitations in data collection, processing, and solution optimization, and is unable to meet the high-precision requirements of modern water network management. To sum up, how to configure the water surface ratio in a scientific, reasonable, comprehensive and dynamically adaptable manner in the water network area has become a key issue that needs to be urgently solved in the current fields of water conservancy engineering, ecological environment and urban planning. It has extremely important practical significance for ensuring the sustainable use of water resources, maintaining ecological security and promoting sustainable social and economic development.

[0036] It should be noted that the method for determining the water surface rate configuration scheme provided in the embodiment of the present application, its execution subject can be a device determined by the water surface rate configuration scheme, and the device determined by the water surface rate configuration scheme can be implemented as part or all of an electronic device through software, hardware, or a combination of software and hardware, wherein the electronic device can be a server or a terminal, wherein the server in the embodiment of the present application can be a single server or a server cluster composed of multiple servers, and the terminal in the embodiment of the present application can be a smart phone, a personal computer, a tablet computer, a wearable device, and other intelligent hardware devices such as an intelligent robot. In the following method embodiments, the execution subject is an electronic device as an example for explanation.

[0037] According to an embodiment of the present invention, an embodiment of a method for determining a water surface rate configuration scheme is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0038] In this embodiment, a method for determining a water surface rate configuration scheme is provided, which can be used for the above-mentioned electronic device. Figure 1 is a flow chart of a method for determining a water surface rate configuration scheme according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps: Step S101, obtaining regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area.

[0039] The underlying surface factor data include at least one of regional topography data, regional soil type and property data, and regional vegetation type and coverage rate data.

[0040] Specifically, the electronic device can receive regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area input by the user, and can also receive regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area sent by other devices.

[0041] Among them, regional meteorological data can include precipitation, temperature, evaporation, wind speed, etc. corresponding to the target water network area. These climate factors directly affect the dynamic adjustment of the water surface rate. Hydrological data include river flow, water level changes, runoff, etc. corresponding to the target water network area. Water quality data include indicators such as dissolved oxygen, chemical oxygen demand, total nitrogen and total phosphorus corresponding to the target water network area. Regional topographic data is that topography is a key factor affecting the morphology of the water network and the movement of water flow. Different terrain types such as mountains, plains, and hills determine the direction, speed, and convergence of water flow. In mountainous areas, the terrain is undulating, the river drop is large, the water flow is fast, and concentrated runoff is easy to form, which puts forward higher requirements for flood control and drainage; while in plain areas, the terrain is flat and the water flow is relatively slow. The change of water surface rate has a more significant impact on flood regulation and water resource storage. In addition, topography also affects the distribution of precipitation. The windward slope of the mountain often has more precipitation, while the leeward slope is relatively dry. By obtaining regional topographic data, we can better understand the formation and evolution of water networks and provide a geographical basis for the rational planning of water surface rates. According to regional soil type and property data, soil type and properties have an important impact on water infiltration, evaporation and storage. Different types of soil, such as sand, loam and clay, have large differences in porosity, water permeability and water holding capacity. Sand has strong water permeability, and water is easy to infiltrate and lose, which is not conducive to the long-term preservation of surface water bodies; clay has poor water permeability, and water is easy to accumulate on the surface after rainfall, which may lead to an increase in local water surface rates. Understanding the properties of soil is helpful to analyze the movement and storage of water resources in the soil, so as to rationally plan the water surface rate and improve the efficiency of water resource utilization. In agricultural irrigation areas, rationally configuring the water surface rate according to soil type and properties can better meet the water requirements of crops and reduce water resource waste. According to regional vegetation type and coverage data, vegetation plays an important regulatory role in the ecosystem of water network areas. Vegetation affects the water cycle through transpiration. Dense vegetation can increase air humidity, intercept rainfall, reduce surface runoff, and increase soil moisture. In areas with high forest vegetation coverage, part of the rainwater is intercepted by vegetation and released slowly during rainfall, which reduces the surface runoff velocity and flow, and helps maintain a stable water surface ratio. In areas with sparse vegetation, surface runoff is large, which can easily cause soil erosion and water surface ratio fluctuations. In addition, vegetation can improve soil structure and enhance soil water retention capacity. Obtaining regional vegetation type and coverage data can evaluate the regulatory effect of vegetation on the water network and provide an ecological basis for optimizing water surface ratio configuration.

[0042] The embodiment of the present application does not specifically limit the manner in which the electronic device obtains regional meteorological data, hydrological data, water quality data, and underlying surface factor data corresponding to the target water network area.

[0043] Step S102, inputting regional meteorological data, hydrological data, water quality data and underlying surface factor data into a preset water surface ratio configuration scheme model, and outputting a target water surface ratio configuration scheme corresponding to the target water network area.

[0044] Specifically, the electronic device can integrate regional meteorological data, hydrological data, water quality data, and underlying surface factor data, and then input the integrated data into a preset water surface ratio configuration scheme model. The preset water surface ratio configuration scheme model is constructed based on certain algorithms and principles, and is intended to calculate the optimal water surface ratio configuration scheme based on the input data.

[0045] After the preset water surface ratio configuration scheme model analyzes and calculates the input data, the target water surface ratio configuration scheme corresponding to the target water network area is finally output. The scheme specifies in detail the area, layout and connectivity of different types of water bodies (such as ditches, ponds, wetlands, etc.) in the water network area to achieve multi-objective optimization. For example, in a certain target water network area, the preset water surface ratio configuration scheme model may output increasing the wetland area in a specific area to improve the storage and purification capacity of water resources, while optimizing the layout of ditches and enhancing flood control and drainage functions. The output scheme may also include an evaluation and prediction of the implementation effect of the scheme, providing a reference for subsequent decision-making and implementation. Through the implementation of the target scheme, the water resources situation in the water network area can be effectively improved, the flood control and drainage capacity can be improved, the water ecological environment can be protected, and the sustainable development of the region can be promoted.

[0046] This step will be described in detail below.

[0047] Step S103: configuring the water surface rate corresponding to the target water network area based on the target water surface rate configuration plan.

[0048] Specifically, the electronic equipment can configure the plan according to the target water surface rate to clarify the distribution of different types of water surfaces in the target water network area, such as rivers, lakes, ponds, wetlands, etc. For large rivers, considering their main functions in the region such as flood control and water delivery, determine the location and scope of their river channel widening or regulation; for lakes and ponds, reasonably plan their scale and shape according to the surrounding terrain and water demand; for wetlands, select suitable low-lying areas or river floodplains for planning to give full play to their ecological purification and storage functions.

[0049] Then, analyze the connectivity between the water bodies in the target water network area, and optimize the connectivity of the water system by digging ditches, building sluices or pumping stations, etc., according to the requirements of the configuration plan. Ensure that water can flow reasonably between different water bodies, improve the allocation capacity of water resources and the self-purification capacity of water bodies. For example, in some areas with dense river networks but poor connectivity, open up the Broken River to form a complete circulation system for the water system.

[0050] The method for determining the water surface rate configuration scheme provided in the embodiment of the present application obtains the regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area, provides a comprehensive and accurate basis for the water surface rate configuration scheme, makes the target water surface rate configuration scheme output by the preset water surface rate configuration scheme model more in line with the actual situation, and improves the scientificity and accuracy of the configuration scheme. Then, the regional meteorological data, hydrological data, water quality data and underlying surface factor data are input into the preset water surface rate configuration scheme model, and the target water surface rate configuration scheme corresponding to the target water network area is output, which ensures the accuracy and scientificity of the target water surface rate configuration scheme corresponding to the output target water network area. The water surface rate corresponding to the target water network area is configured based on the target water surface rate configuration scheme. A scientific and reasonable water surface rate configuration scheme helps to balance the supply and demand relationship of water resources and ensure the economic development and ecological stability of the target water network area. By optimizing the water surface rate configuration, it can not only meet the current demand for water resources, but also protect water resources and the ecological environment, lay the foundation for the long-term sustainable development of the region, and achieve the coordinated development of economy, society and environment. A reasonable water surface rate configuration is achieved for the target water network area.

[0051] In this embodiment, a method for determining a water surface rate configuration scheme is provided, which can be used for the above-mentioned electronic device. Figure 2 is a flow chart of a method for determining a water surface rate configuration scheme according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps: Step S201, obtaining regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area.

[0052] The underlying surface factor data include at least one of regional topography data, regional soil type and property data, and regional vegetation type and coverage rate data.

[0053] For this step, please refer to the above description of step S101, which will not be described in detail here.

[0054] Step S202, inputting regional meteorological data, hydrological data, water quality data and underlying surface factor data into a preset water surface ratio configuration scheme model, and outputting a target water surface ratio configuration scheme corresponding to the target water network area.

[0055] Specifically, the preset water surface ratio configuration scheme model includes a feature extraction model and a water surface ratio configuration scheme model. The above step S202 may include the following steps: Step S2021, inputting regional meteorological data, hydrological data, water quality data and underlying surface factor data into the feature extraction model.

[0056] Specifically, before inputting data, electronic devices can pre-process various types of data. For regional meteorological data, such as precipitation, temperature, evaporation, wind speed, etc., there may be missing data, outliers or inconsistent formats. To address these issues, interpolation is used to fill missing values, and outliers are identified and processed through statistical analysis. At the same time, the data format is unified to ensure the accuracy and completeness of the data. For river flow, water level changes, runoff, etc. included in hydrological data, data cleaning and standardization are also required to make them comparable. Water quality data such as dissolved oxygen, chemical oxygen demand, total nitrogen, total phosphorus and other indicators may vary due to different sampling time, location and method, and need to be normalized. Underlying surface factor data, such as topographic data (which can be represented by digital elevation models), soil type and property data (soil texture, porosity, etc.), vegetation type and coverage data, etc., need to be digitally converted and formatted so that they can be recognized and processed by feature extraction models.

[0057] Then, the electronic device inputs the pre-processed regional meteorological data, hydrological data, water quality data and underlying surface factor data into the feature extraction model.

[0058] Step S2022: the feature extraction model identifies the input data and determines the data types of various data in the input data.

[0059] Among them, the input data include regional meteorological data, hydrological data, water quality data and underlying surface factor data.

[0060] Specifically, the feature extraction model identifies the input data and determines the data types of various data in the input data. The data types may include image types and numerical types.

[0061] Step S2023, setting multi-channel input according to the data types of various data.

[0062] Specifically, the feature extraction model can continuously monitor the changing trends of various input data. For example, for meteorological data, it tracks the rate of change of indicators such as precipitation and temperature in the time series. In the rainy season, precipitation may rise sharply in a short period of time. The feature extraction model captures this rapid growth trend through real-time analysis of precipitation data, and determines that the weight of the impact of precipitation data on the water network system has increased significantly at this time.

[0063] Then, the feature extraction model analyzes the changing trends of various input data, determines the data fluctuations, and then determines the data feature analysis results. For example, the river flow in the hydrological data may fluctuate greatly during the flood season, but be relatively stable during the dry season. The feature extraction model evaluates the fluctuation range by calculating statistics such as the standard deviation of the flow data. When the flow fluctuation range exceeds the normal range, it indicates that the data has a more critical impact on the water network status at the current moment. Finally, the feature extraction model dynamically adjusts the computing resources of each channel based on the results of data feature analysis. When the importance of precipitation data in the rainy season is detected to increase, the feature extraction model allocates more computing units (such as neurons and computing cores in the neural network) to the precipitation data channel. These additional computing units can perform more complex operations on precipitation data, such as using higher-order convolution kernels for feature extraction, or increasing the number of neural network layers to further explore the relationship between precipitation and water network response.

[0064] In addition, the feature extraction model changes the data transmission bandwidth of the channel according to the results of data feature analysis, and sets dynamic weights for different channels, so as to set multi-channel input according to the data types of various data. For example, in the dry season, the impact of evaporation on the water resource balance of the water network increases. The feature extraction model increases the bandwidth of the evaporation data channel, allowing more data to be processed through the channel per unit time. This can be achieved by adjusting the cache size for data storage and transmission, optimizing the algorithm for data reading and writing, etc., to ensure that evaporation data can be processed by the model more quickly and accurately. The importance of each data type varies under different water network operation conditions. In the early stage of floods caused by heavy rains, precipitation and river flow data are crucial to predicting the evolution of floods. The feature extraction model will increase the weight of these two data channels. In the operation process of the model, such as in the weighted summation layer of the neural network, the weight coefficients of the two channel data are increased, so that they play a greater role in model decision-making.

[0065] Step S2024: Based on each input channel, feature extraction is performed on various types of input data, and feature representation is output.

[0066] Specifically, the above step S2024 may include: Step a1: input the image data in the input data as image channels.

[0067] Image data usually comes from a variety of channels, the most common of which are satellite remote sensing images or aerial remote sensing images obtained through remote sensing technology. These images can show the topography, water distribution, land use and other information of the target water network area from a macro perspective.

[0068] Specifically, the feature extraction model takes the image data in the input data as the image channel input, and then uses deep learning technology to achieve adaptive image enhancement. Specifically, the feature extraction model automatically determines which areas need to be enhanced and what enhancement method to use based on the content and features of the input image. For example, for water bodies with fuzzy boundaries in the water network area, the feature extraction model automatically identifies and applies edge enhancement algorithms to highlight the boundaries of the water body, so that subsequent image channel processing can more accurately capture the range and shape characteristics of the water body. This adaptive enhancement avoids the limitations of traditional fixed enhancement methods and can better adapt to image data in different scenarios.

[0069] Step a2: for the numerical data in the input data, convert the numerical data into a two-dimensional matrix form and allocate channels according to the feature dimensions.

[0070] Specifically, when numerical data has time series characteristics, such as the change of precipitation over time in meteorological data, the fluctuation of water level at different times in hydrological data, etc., the feature extraction model can construct a two-dimensional matrix based on the time series. Take time as one dimension of the matrix, and the values ​​corresponding to different time points as another dimension. For numerical data with spatial distribution characteristics, such as soil moisture and vegetation coverage in different geographical locations, the feature extraction model can construct a two-dimensional matrix based on spatial coordinates. Based on longitude and latitude or other spatial coordinate systems, the spatial position is used as the row and column index of the matrix, and the values ​​of the corresponding positions are filled into the matrix elements. For soil moisture data of multiple monitoring points in a region, it is converted into a two-dimensional matrix according to the coordinates of the monitoring points. If there are multiple spatially related numerical data, such as soil moisture, soil nutrient content, etc., they can also be integrated into a matrix, and the spatial correlation information of the data can be retained in this way. When allocating channels, channels can be divided according to spatial regions, such as allocating data from adjacent regions to the same channel, so that the model can learn spatial similarities and differences.

[0071] In addition, the feature extraction model can also mine the intrinsic relationship between numerical data and map it into a two-dimensional matrix. By calculating the correlation, causal relationship, etc. between the data, a relationship matrix is ​​constructed. When analyzing the relationship between meteorological data and hydrological data, the correlation coefficient between precipitation and river flow is calculated, and the correlation coefficient is filled into the matrix with the data type as rows and columns. For complex relationships between multiple data types, an n×n (n is the number of data types) relationship matrix is ​​formed. This matrix can highlight the interaction between data. In terms of channel allocation, data can be grouped according to the closeness of the relationship, and the dimensions corresponding to closely related data can be allocated to the same channel, which helps the model capture the coordinated change characteristics between data.

[0072] The feature extraction model can then assign channels according to feature dimensions and adjust channel weights in real time based on the importance of the data to the target task (such as water surface rate configuration plan formulation) in different scenarios.

[0073] Optionally, the feature extraction model can also create cross-fusion channels to promote the exchange and fusion of data with different feature dimensions. Select relevant dimensions from different types of numerical data and combine them into new cross-channels. Combine the precipitation dimension in meteorological data with the river velocity dimension in hydrological data into a new channel, which reflects the impact of precipitation on water flow velocity. Through this cross-fusion channel, the model can learn comprehensive features across data types, dig out information that is difficult to discover in a single data type channel, and provide a richer perspective for water network analysis.

[0074] Step a3: Use multiple convolution kernels of different sizes to perform convolution processing on the data input from each input channel to extract target features of different scales.

[0075] Specifically, the feature extraction model can dynamically select the convolution kernel size based on the features of the data input from each input channel. For example, when processing image data of a water network area, the feature extraction model first performs a preliminary analysis of the image to determine the range of the main feature scales it contains. If the image contains both large-scale features such as large lakes and small-scale features such as small ditches, the feature extraction model adaptively selects the frequency and order of using convolution kernels of different sizes based on the proportion and importance of the features. After the large-size convolution kernel obtains the global features, it passes the global feature information to the small-size convolution kernel. The small-size convolution kernel uses this global feature information to perform more refined feature mining in the local area, thereby extracting target features of different scales.

[0076] For example, after a large-size convolution kernel identifies the approximate range of a water body, a small-size convolution kernel at the boundary of the water body, combined with the global position information transmitted by the large-size convolution kernel, further extracts the subtle texture and shape change features of the boundary. This fusion method can give full play to the advantages of convolution kernels of different sizes and generate a more comprehensive and accurate feature representation.

[0077] In step a4, the target features extracted from different levels and types of data are fused according to the channel dimension to generate feature representation.

[0078] Specifically, the feature extraction model can first identify the target features extracted from different levels and different types of data, and determine the hierarchical association mechanism between each target feature. Then, according to the hierarchical association mechanism between each target feature, the weight information corresponding to each target feature is determined. For local river channel detail features), the feature extraction model can adopt a cross-scale feature fusion method. First, hierarchical processing is performed on the features of different scales, and the large-scale features are used as global features and the small-scale features are used as local features. Then, based on the weight information corresponding to each target feature, convolution operations or pooling operations are used to fuse features between different scales to generate feature representations. For example, through upsampling and downsampling operations, large-scale features and small-scale features are adjusted to the same scale, and then the channel dimension is fused. The idea of ​​jump connection can also be introduced to directly connect features of different scales, so that the fused feature representation can simultaneously contain information of different scales, and more comprehensively describe the structure and function of the water network.

[0079] Step S2025, inputting the feature representation into the water surface ratio configuration scheme model, and outputting the target water surface ratio configuration scheme corresponding to the target water network area based on the feature representation.

[0080] Specifically, the water surface ratio configuration scheme model includes a plurality of sub-water surface ratio configuration scheme models of different algorithms. The above step S2025 may include the following steps: Step b1, input the characteristic representation into each sub-water surface rate configuration scheme model respectively.

[0081] Specifically, the electronic device may input the characteristic representation into each sub-water surface ratio configuration scheme model respectively.

[0082] Step b2: each sub-water surface rate configuration scheme model identifies the feature representation and outputs a plurality of first candidate water surface rate configuration schemes respectively.

[0083] In an optional implementation manner of the present application, the above step b2 may include the following steps: Step b21, for one of the sub-water surface rate configuration scheme models, the sub-water surface rate configuration scheme model identifies the feature representation.

[0084] Specifically, for one of the sub-water surface ratio configuration scheme models, the sub-water surface ratio configuration scheme model identifies the feature representation.

[0085] Step b22, generating multiple initial water surface rate configuration schemes based on the target constraint conditions in the sub-water surface rate configuration scheme model.

[0086] Among them, the target constraints include water surface rate range constraints, water quality standard constraints, water level and flood storage capacity constraints, and land use constraints.

[0087] Among them, the water surface rate range constraint is: ; in, is the total area of ​​the target water network area; For water Water surface rate; is the upper limit of the water surface rate, is the lower limit of the water surface rate; The water quality standards are constrained by: , ; in, For water of dissolved oxygen, For water The chemical oxygen demand, For water Total nitrogen, For water of total phosphorus; Chemical oxygen demand for environmental standards, Total nitrogen for environmental standards, is the environmental standard value of total phosphorus; The water level and flood storage capacity constraints are: ; in, For water The water level, is the maximum permissible water level in the target water network area; The land use constraints are: ; in, is the total area of ​​the channel; is the total area of ​​the pond; is the total area of ​​wetlands; is the total area of ​​agricultural land, is the total area of ​​the target water network area.

[0088] Specifically, the sub-water surface ratio configuration scheme model includes multiple agents, each of which focuses on different types of target constraints. One of them, the "hydrological agent", is responsible for handling water level and flood storage capacity constraints; one, the "ecological agent", focuses on water quality standard constraints and ecological protection-related constraints; and one, the "land use agent", focuses on land use constraints and water surface ratio range constraints. These agents collaborate with each other, and when generating the initial water surface ratio configuration scheme, they comprehensively consider multiple constraints through information sharing and interaction. After the hydrological agent determines the minimum water surface area that meets the flood storage demand, it passes the information to the land use agent. On this basis, the land use agent determines the feasible water surface layout in combination with the land resource situation. The ecological agent evaluates the ecological suitability of the generated preliminary scheme and feeds it back to other agents for adjustment, and jointly generates multiple initial water surface ratio configuration schemes that meet multiple constraints.

[0089] Step b23: Based on the feature representation, each initial water surface rate configuration scheme is substituted into the objective function in the sub-water surface rate configuration scheme model to calculate the objective function value corresponding to each initial water surface rate configuration scheme.

[0090] Among them, the objective function includes at least one of the flood control and drainage sub-function, the water resource utilization sub-function, and the water environment capacity sub-function.

[0091] Specifically, the objective function includes:

[0092] in, is the flood control and waterlogging removal sub-function; The water body in the target water network area i area; For water The flood storage capacity is as follows: ;in For water The water level, is the flood storage coefficient; and / or,

[0093] in, is the water resource utilization sub-function, For water The available water volume is calculated as: ; Where P is water The precipitation of E is the water body evaporation, β is the regional hydrological conversion coefficient; For water The water use efficiency is defined as: ;in, For agricultural irrigation needs; and / or,

[0094] in, is the water environment capacity subfunction; Pi is the water body The pollution load is usually related to land use and emissions; Vi is the water body The environmental capacity is calculated as: ; Among them, γ is the comprehensive degradation coefficient, which is related to the water The dilution capacity of is the area of ​​water body i, is the water level of water body i, It is the standard concentration of water quality.

[0095] Specifically, the sub-water surface ratio configuration scheme model can dynamically adjust the weights of each sub-function in the objective function (such as flood control and drainage sub-function, water resource utilization sub-function, water environment capacity sub-function, etc.) according to the importance and relevance of each feature in the feature representation. For water network areas in flood-prone areas, in the feature representation, the features related to flood control (such as river width, terrain slope, etc.) will be more prominent. At this time, increasing the weight of the flood control and drainage sub-function in the objective function makes the evaluation of the scheme's performance in flood control more rigorous. Through this adaptive weight adjustment, the objective function can better fit the actual needs of the water network area and more accurately reflect the advantages and disadvantages of each initial scheme.

[0096] Then, use a multi-core processor or distributed computing platform to perform parallel calculations on each initial water surface ratio configuration scheme. Different schemes are assigned to different computing cores or computing nodes, and the objective function is substituted into the calculation to obtain the objective function value corresponding to each initial water surface ratio configuration scheme. This can greatly shorten the calculation time, especially when dealing with a large number of initial schemes. For example, in a large-scale water network area, hundreds or thousands of initial water surface ratio configuration schemes may be generated. Through parallel calculation, the objective function value corresponding to each scheme can be quickly obtained, which improves the evaluation efficiency and buys more time for subsequent scheme screening and optimization.

[0097] Step b24, determining the historical optimal water surface rate configuration scheme and the current global optimal water surface rate configuration scheme according to the objective function value corresponding to each initial water surface rate configuration scheme.

[0098] Specifically, the sub-water surface rate configuration scheme model can introduce multi-dimensional evaluation indicators. Multi-dimensional evaluation indicators can include factors such as the cost of scheme implementation, the long-term impact on the surrounding ecological environment, and social and economic benefits. A scheme with a high objective function value may have too high an implementation cost or have a potential negative impact on the ecological environment. By combining these multi-dimensional indicators, each initial water surface rate configuration scheme is scored. A comprehensive evaluation model is constructed to weightedly integrate the objective function value with other evaluation indicators to obtain a comprehensive score. The historical optimal and current global optimal schemes are determined based on the comprehensive score, which can avoid the limitations of focusing only on the objective function value and ignoring other important factors, making the selected scheme more valuable for practical application.

[0099] Then, the sub-water surface rate configuration scheme model can use swarm intelligence algorithms (such as ant colony algorithm, particle swarm algorithm, etc.) to assist in determining the optimal solution. Each initial water surface rate configuration scheme is regarded as an individual or particle in the swarm intelligence algorithm, and the objective function value and comprehensive evaluation score are used as the fitness value of the individual. Through the iterative search of the algorithm, the individual is allowed to continuously evolve and optimize in the solution space. In the ant colony algorithm, ants search and choose between different schemes according to the guidance of pheromones, and the strength of pheromones reflects the pros and cons of the scheme. As the iteration proceeds, the algorithm gradually converges to a better solution. Through the global search capability of the swarm intelligence algorithm, some excellent solutions that are difficult to find by traditional methods can be discovered, which improves the efficiency and quality of determining the optimal solution, thereby determining the historical optimal water surface rate configuration scheme and the current global optimal water surface rate configuration scheme.

[0100] Step b25, determining the update speed according to the historical optimal water surface rate configuration scheme and the current global optimal water surface rate configuration scheme.

[0101] Specifically, in each iteration, each initial water surface rate configuration scheme updates the speed and position according to its corresponding historical optimal water surface rate configuration scheme pbest and the current global optimal water surface rate configuration scheme gbest. The speed update formula is: , in, is the speed of the initial water surface rate configuration scheme i at time t, w is the inertia weight, which is used to balance the global and local search capabilities. As the number of iterations increases, w can decrease linearly, so that the algorithm focuses on global search in the early stage and focuses on local optimization in the later stage; c1 and c2 are learning factors, usually set to constants; r1 and r2 are random numbers between [0,1]; pbest i is the historical optimal water surface rate configuration scheme corresponding to the initial water surface rate configuration scheme i; gbest i The current global optimal water surface rate configuration scheme corresponding to the initial water surface rate configuration scheme i. is the position of the initial water surface rate configuration scheme i at time t.

[0102] Step b26, obtaining updated water surface rate configuration schemes according to the update speed.

[0103] Among them, each updated water surface rate configuration scheme meets the target constraint.

[0104] Specifically, the position update formula is:

[0105] in, Configure a scheme for each updated water surface rate.

[0106] Step b27, calculate the objective function value corresponding to each updated water surface rate configuration scheme.

[0107] Specifically, the sub-water surface ratio configuration scheme model can calculate the objective function value corresponding to each updated water surface ratio configuration scheme based on the above objective function.

[0108] Step b28, determining the updated water surface rate configuration scheme whose objective function value is greater than the preset function value as the first candidate water surface rate configuration scheme.

[0109] Specifically, the sub-water surface ratio configuration scheme model can compare the objective function value corresponding to each updated water surface ratio configuration scheme with the preset function value, and then determine the updated water surface ratio configuration scheme with the objective function value greater than the preset function value as the first candidate water surface ratio configuration scheme.

[0110] Step b29: If there is no updated water surface ratio configuration scheme whose objective function value is greater than the preset function value, the updated water surface ratio configuration scheme continues to be updated until the updated water surface ratio configuration scheme whose objective function value is greater than the preset function value is determined as the first candidate water surface ratio configuration scheme.

[0111] Specifically, if there is no updated water surface ratio configuration scheme whose objective function value is greater than the preset function value, the sub-water surface ratio configuration scheme model continues to update the updated water surface ratio configuration scheme until the updated water surface ratio configuration scheme whose objective function value is greater than the preset function value is determined as the first candidate water surface ratio configuration scheme.

[0112] In an optional implementation of the present application, other sub-water surface ratio configuration scheme models may be sub-water surface ratio configuration scheme models based on a genetic algorithm (GA), may be sub-water surface ratio configuration scheme models based on a simulated annealing algorithm (SA), or may be sub-water surface ratio configuration scheme models based on other algorithms. The embodiments of the present application do not specifically limit other sub-water surface ratio configuration scheme models.

[0113] For the sub-water surface ratio configuration scheme model based on genetic algorithm (GA), GA is an optimization algorithm based on natural selection and genetic mechanism. It simulates the biological evolution process to find the optimal solution through operations such as encoding, selection, crossover and mutation. In the water surface ratio configuration problem, the water surface ratio configuration scheme is encoded as chromosomes, and new configuration schemes are continuously generated by selecting chromosomes with high fitness for crossover and mutation, and gradually evolving towards the optimal solution.

[0114] For the sub-water surface ratio configuration scheme model based on the simulated annealing algorithm (SA), SA originates from the simulation of the solid annealing process. By controlling the temperature parameters, a worse solution is accepted with a certain probability, thereby jumping out of the local optimum. In the water surface ratio configuration optimization, starting from an initial water surface ratio configuration scheme, a new scheme is randomly searched in the solution space. If the objective function value of the new scheme is better, the scheme is accepted; otherwise, the worse scheme is accepted with a certain probability according to the current temperature and the Metropolis criterion. As the temperature gradually decreases, the algorithm converges to the global optimal solution.

[0115] Step b3: fuse the first candidate water surface rate configuration schemes to generate multiple first fused water surface rate configuration schemes.

[0116] Specifically, the above step b3 may include the following steps: Step b31, calculating the similarity between the first candidate water surface rate configuration schemes.

[0117] Specifically, the water surface ratio configuration scheme model represents each first candidate water surface ratio configuration scheme as a graph structure, where nodes represent each element in the water network system (such as rivers, lakes, pumping stations, etc.), and edges represent the relationship between elements (such as water flow direction, connectivity, etc.). These graph structures are learned and analyzed using graph neural networks (GNNs). Through the message passing mechanism of GNNs, nodes can obtain information about their neighboring nodes, thereby learning the structural characteristics of the entire graph. Then, when the water surface ratio configuration scheme model calculates the similarity between graph structures of different schemes, it can compare the feature vectors extracted by the graph neural network, or use a graph matching algorithm to measure the similarity of the graph structure. This method can capture the complex relationship between elements in the scheme, and has unique advantages in evaluating the overall layout and functional similarity of water surface ratio configuration schemes in the water network system.

[0118] Step b32, for each first candidate water surface rate configuration scheme whose similarity is greater than a preset similarity threshold, only one is retained to obtain multiple backup water surface rate configuration schemes.

[0119] Specifically, for each first candidate water surface rate configuration scheme whose similarity is greater than a preset similarity threshold, the water surface rate configuration scheme model retains only one of them, thereby obtaining a plurality of backup water surface rate configuration schemes.

[0120] Step b33, cross-combining the various standby water surface rate configuration schemes to generate various first fused water surface rate configuration schemes.

[0121] Specifically, the water surface rate configuration scheme model can disassemble each standby water surface rate configuration scheme according to different functional modules of the water network system, such as flood control module (including dam design, flood storage area planning, etc.), water resource allocation module (water source protection, water channel layout, etc.), ecological restoration module (wetland construction, aquatic plant planting, etc.) and water quality purification module (sewage treatment facilities, ecological pond setting, etc.). Then, according to the current needs and future development trends of the water network area, the importance weight of each functional module is determined, and then according to the importance weight of each functional module, each standby water surface rate configuration scheme is cross-combined to generate each first fusion water surface rate configuration scheme. For example, in periods of frequent droughts, the weight of the water resource allocation module is relatively high; while in areas with greater flood risks, the weight of the flood control module is more critical. When combining the schemes, the corresponding modules of each scheme are weighted and fused according to the weights. For the water resource allocation module, if the weight of the relevant part of plan A is 0.6 and that of plan B is 0.4, when generating the module of the fusion plan, both are considered comprehensively and combined according to the weight distribution ratio, so that the generated plan is more in line with actual needs.

[0122] Step b4, inputting each first fused water surface rate configuration scheme into each sub-water surface rate configuration scheme model respectively.

[0123] Specifically, each first fused water surface ratio configuration scheme is input into each sub-water surface ratio configuration scheme model respectively again.

[0124] Step b5: each sub-water surface rate configuration scheme model evaluates and updates each first fused water surface rate configuration scheme, and outputs a plurality of second candidate water surface rate configuration schemes.

[0125] Specifically, each sub-water surface rate configuration scheme model evaluates and updates each first fused water surface rate configuration scheme, and the process of outputting multiple second candidate water surface rate configuration schemes can be referred to the above process in which each sub-water surface rate configuration scheme model identifies the feature representation and outputs multiple first candidate water surface rate configuration schemes respectively, which will not be repeated here.

[0126] Step b6, repeat this process until a preset number of times have passed, and each sub-water surface rate configuration scheme model outputs its corresponding current optimal water surface rate configuration scheme.

[0127] Specifically, the water surface rate configuration scheme model can dynamically adjust the number of loops and loop strategies according to the optimization degree of the current optimal water surface rate configuration scheme. In the early stage of the cycle, if the water surface rate configuration scheme model finds that the optimization speed of the scheme is fast and the objective function value increases rapidly, then the number of subsequent loops can be appropriately reduced to avoid excessive optimization and waste of computing resources. On the contrary, if the scheme improves slowly, increase the number of loops or adjust the parameters and operation methods of the model in the loop. It is also possible to set an adaptive loop strategy for each sub-water surface rate configuration scheme model based on the characteristics of different sub-water surface rate configuration scheme models. For example, for a model based on deep learning, when its convergence speed slows down, adjust the learning rate or increase the amount of training data to speed up the optimization process of the scheme.

[0128] Finally, based on the adjusted number of cycles and cycle strategies, each sub-water surface rate configuration scheme model outputs its corresponding current optimal water surface rate configuration scheme.

[0129] Step b7, identifying each current optimal water surface rate configuration scheme, and outputting the target water surface rate configuration scheme corresponding to the target water network area.

[0130] Specifically, the above step b7 may include the following steps: Step b71, inputting each current optimal water surface rate configuration scheme into each sub-water surface rate configuration scheme model respectively.

[0131] Specifically, each current optimal water surface ratio configuration scheme is input into each sub-water surface ratio configuration scheme model respectively.

[0132] Step b72: each sub-water surface rate configuration scheme model evaluates the current optimal water surface rate configuration scheme, and votes on each current optimal water surface rate configuration scheme based on the evaluation result.

[0133] Specifically, each sub-water surface ratio configuration scheme model can evaluate the current optimal water surface ratio configuration scheme based on the above-mentioned objective function and the knowledge graph in the water network field to obtain the evaluation results. Among them, the knowledge graph contains rich professional knowledge, practical experience and various cases. When evaluating the current optimal water surface ratio configuration scheme, the model will refer to the relevant information in the knowledge graph for intelligent reasoning and judgment. If a plan involves the construction of a specific water conservancy facility, the knowledge graph can provide the operating results and success and failure cases of the facility under different geographical environments and climatic conditions, helping the model to more accurately evaluate the feasibility and potential risks of the plan. At the same time, the knowledge graph can also analyze the relationship between the various elements in the plan to determine whether there are potential conflicts or synergies, thereby giving a more scientific evaluation result.

[0134] Then, vote on each current optimal water surface rate configuration plan based on the evaluation results.

[0135] Step b73, determining the current optimal water surface rate configuration plan with the most votes as the target water surface rate configuration plan.

[0136] Specifically, the water surface ratio configuration scheme model determines the current optimal water surface ratio configuration scheme with the most votes as the target water surface ratio configuration scheme.

[0137] Step S203: configuring the water surface rate corresponding to the target water network area based on the target water surface rate configuration plan.

[0138] For details about this step, please refer to the above description of step S103, which will not be elaborated here.

[0139] The method for determining the water surface ratio configuration scheme provided in the embodiment of the present application inputs regional meteorological data, hydrological data, water quality data and underlying surface factor data into a feature extraction model. The feature extraction model identifies the input data, determines the data types of various data in the input data, and ensures the accuracy of the data types of various data in the determined input data. Then, a multi-channel input is set according to the data types of various data. Thus, the feature extraction model can carry out targeted processing according to the characteristics of different types of data. The time series data such as precipitation and temperature in meteorological data differ greatly from the flow and water level change data in hydrological data and water quality data in terms of characteristics. Multi-channel input allows the feature extraction model to adopt processing methods suitable for each data type respectively, avoids the loss of key information due to unified processing, ensures the effective use of data, and lays the foundation for subsequent accurate feature extraction. The image data is directly input as an image channel, and the spatial information and visual features of the image are completely retained. For example, in the remote sensing image of the water network area, the shape, location and range of the water body can be intuitively presented, providing an intuitive geographic information basis for subsequent analysis. Converting numerical data into a two-dimensional matrix and assigning channels according to the feature dimension can structure the abstract numerical information, which is convenient for model recognition and processing. For example, the flow and water level values ​​in hydrological data can be organized according to feature dimensions such as time series or spatial distribution after conversion, so that the model can better capture the relationship and change trend between values. Then, multiple convolution kernels of different sizes are used to perform convolution processing on the input channel data to extract target features of different scales. Small-sized convolution kernels focus on local details. When processing image data, they can accurately identify subtle features such as the edge of the water body and small ditches; large-sized convolution kernels focus on global information and grasp the macro layout and structure of the entire water network. When processing numerical data, small convolution kernels can capture short-term fluctuations and local changes in data, while large convolution kernels are used to analyze long-term trends and overall laws. This multi-scale feature extraction method comprehensively covers various feature information of the data and improves the depth and breadth of understanding of the data. Finally, the target features extracted from different levels and types of data are fused according to the channel dimension, and the generated feature representation combines the advantages of multi-source data. The spatial features of image data and the quantitative features of numerical data complement each other, providing the model with richer and more comprehensive information. When determining the water surface ratio configuration plan, the fused features can not only reflect the actual geographical form of the water network, but also reflect the changes in related hydrological, meteorological and other data, making the water surface ratio configuration plan model decision more scientific and accurate. In addition, rich and comprehensive feature representation helps to improve the performance of the water surface ratio configuration plan model in subsequent tasks. The water surface ratio configuration plan model is analyzed based on the fused feature representation, which can more accurately capture the complex relationships and potential laws between data, improve prediction accuracy and evaluation reliability. At the same time, the introduction of multi-scale features enhances the adaptability of the model to different situations, so that it can still maintain good performance when facing complex and changing data.

[0140] Then, the water surface ratio configuration scheme model contains multiple sub-water surface ratio configuration scheme models with different algorithms, and each sub-water surface ratio configuration scheme model processes the feature representation based on different algorithm principles and optimization strategies. For one of the sub-water surface ratio configuration scheme models, an initial water surface ratio configuration scheme is generated based on the target constraint conditions, which fundamentally guarantees the feasibility of the scheme in practical applications. It avoids the situation that the generated scheme cannot be implemented due to violation of actual constraints. Based on the feature representation, each initial water surface ratio configuration scheme is substituted into the objective function in the sub-water surface ratio configuration scheme model, and the objective function value corresponding to each initial water surface ratio configuration scheme is calculated, so that the advantages and disadvantages of each initial water surface ratio configuration scheme can be evaluated from multiple angles. When evaluating a scheme, not only can its effect on flood control and drainage be understood, but also its impact on water resource utilization and water environment capacity can be known, thereby realizing multi-objective optimization of the water surface ratio configuration scheme, so that the generated scheme is more in line with the comprehensive needs of sustainable development of the water network area. Then, the historical optimal and current global optimal water surface ratio configuration schemes are determined according to the objective function value, and the update speed is determined accordingly, and the updated configuration scheme is continuously obtained. The updated scheme with an objective function value greater than the preset function value is determined as the first candidate water surface ratio configuration scheme. This screening mechanism can gradually eliminate undesirable schemes and retain and optimize more advantageous schemes. If there is no scheme that meets the preset conditions, it will be continuously updated to ensure that the first candidate water surface ratio configuration scheme finally obtained performs well under the multi-objective comprehensive evaluation, thereby improving the quality and practicality of the first candidate water surface ratio configuration scheme. In the process of continuous updating and screening, the first candidate water surface ratio configuration scheme can better adapt to the complex situation of the water network area. Since each update is based on the previous optimal scheme and takes into account the comprehensive impact of multiple objective functions, the generated scheme can more accurately meet the actual needs. For a water network area that has both flood control needs and water resource shortage problems, this mechanism can find a first candidate water surface ratio configuration scheme that achieves a good balance between flood control and waterlogging and water resource utilization through multiple iterative optimizations.

[0141] Next, the similarity between the first candidate water surface rate configuration schemes is calculated, and only one of the schemes with a similarity greater than the preset threshold is retained, which effectively avoids the redundancy of the schemes. Among the many candidate schemes, there may be some schemes that are similar in nature and have similar contributions to the final goal. If they are not screened, the subsequent processing of these similar schemes will increase the amount of calculation, waste computing resources and time. By removing redundant schemes and retaining representative schemes, unnecessary computational burdens are reduced, the efficiency of the entire scheme optimization process is improved, and subsequent analysis and processing can be more focused on differentiated schemes. While removing similar schemes, multiple different standby water surface rate configuration schemes are retained to ensure the diversity of the schemes. These different schemes each have unique advantages and characteristics, covering different water surface rate configuration ideas and trade-offs for multiple objectives. One scheme may perform well in flood control and drainage, while another scheme may have more advantages in water resource utilization. Retaining these diverse schemes provides rich materials for subsequent cross-combinations, which is conducive to generating a more comprehensive and higher-quality first fusion water surface rate configuration scheme. Then, the cross-combination of the standby water surface rate configuration schemes can integrate the advantages of different schemes and create a more comprehensive and innovative first fusion water surface rate configuration scheme. In the cross-combination process, the characteristics and advantages of different schemes are combined with each other, which may produce new configuration ideas and patterns. The two schemes that perform well in flood control and water resource utilization are cross-combined. The newly generated scheme may achieve a good balance in both aspects, thereby improving the comprehensive performance of the scheme on multiple objectives, which is more in line with the needs of multi-objective optimization of water surface rate configuration in practical applications. As a result, the generated multiple first fusion water surface rate configuration schemes have stronger adaptability and can better meet the complex and changeable actual needs of the target water network area. Due to the integration of the characteristics of different schemes, these fusion schemes can cover a wider range of situations and conditions. When facing the water network operation needs in different seasons and different climatic conditions, the fusion scheme may have better adjustment capabilities and adaptability, providing more high-quality options for the scientific planning and effective management of water network areas.

[0142] Then, the first fused water surface rate configuration scheme is input into each sub-water surface rate configuration scheme model again for evaluation and update, and the second candidate water surface rate configuration scheme is output, and this process is repeated continuously. Each cycle is an optimization and improvement of the previous scheme. The sub-water surface rate configuration scheme model re-evaluates and adjusts the fusion scheme based on its own algorithm, gradually taps the potential of the scheme, and makes the scheme approach the optimal solution continuously. After the preset number of times, each sub-water surface rate configuration scheme model outputs its corresponding current optimal water surface rate configuration scheme. Each current optimal water surface rate configuration scheme is input into each sub-water surface rate configuration scheme model respectively. Thus, each current optimal water surface rate configuration scheme can be examined from multiple professional perspectives. Each sub-water surface rate configuration scheme model evaluates the performance of each current optimal water surface rate configuration scheme in terms of flood control and drainage, water resource utilization, water environment capacity, etc. according to its own algorithm characteristics, and votes on each current optimal water surface rate configuration scheme according to the evaluation results. The target scheme is determined by combining these evaluation results, which can fully integrate the advantages of different models, avoid the limitations of a single model, and make the final target water surface rate configuration scheme more scientific and reasonable. In addition, multiple sub-water surface ratio configuration scheme models are independently evaluated and voted, which can reduce the risk of wrong decisions due to errors or limitations of individual sub-water surface ratio configuration scheme models. If only one model is relied upon to determine the scheme, the sub-water surface ratio configuration scheme model may give unreasonable results due to data anomalies, algorithm defects and other factors. Voting on multiple sub-water surface ratio configuration scheme models is similar to group decision-making, which can complement and verify each other. Even if deviations occur in the evaluation of individual models, the correct evaluation of other models can ensure the reliability of the final decision, improve the credibility of the water surface ratio configuration scheme, and provide a more reliable basis for the actual application of the target water network area.

[0143] In this embodiment, a water surface rate configuration scheme determination device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0144] This embodiment provides a device for determining a water surface rate configuration scheme, such as Figure 3 As shown, including: The acquisition module 301 is used to acquire regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area; the underlying surface factor data includes at least one of regional topographic data, regional soil type and property data, and regional vegetation type and coverage rate data; An input module 302 is used to input regional meteorological data, hydrological data, water quality data and underlying surface factor data into a preset water surface ratio configuration scheme model, and output a target water surface ratio configuration scheme corresponding to the target water network area; The configuration module 303 is used to configure the water surface rate corresponding to the target water network area based on the target water surface rate configuration plan.

[0145] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0146] The water surface rate configuration scheme determination device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0147] The embodiment of the present invention also provides an electronic device having the above Figure 3 The water surface rate configuration scheme determining device is shown.

[0148] See also Figure 4 , Figure 4 is a schematic diagram of the structure of an electronic device provided by an optional embodiment of the present invention, such as Figure 4 As shown, the electronic device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 4 A processor 10 is taken as an example.

[0149] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0150] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0151] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0152] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0153] The electronic device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 4 The example of connecting through bus is taken in the following.

[0154] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the electronic device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator rod, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0155] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for determining a water surface rate configuration scheme, characterized in that: The method comprises: Acquire regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area; the underlying surface factor data includes at least one of regional topographic data, regional soil type and property data, and regional vegetation type and coverage rate data; Input the regional meteorological data, the hydrological data, the water quality data and the underlying surface factor data into a preset water surface ratio configuration scheme model, and output a target water surface ratio configuration scheme corresponding to the target water network area; The water surface rate corresponding to the target water network area is configured based on the target water surface rate configuration plan.

2. The method according to claim 1, characterized in that The preset water surface rate configuration scheme model includes a feature extraction model and a water surface rate configuration scheme model, and the regional meteorological data, the hydrological data, the water quality data and the underlying surface factor data are input into the preset water surface rate configuration scheme model, and the target water surface rate configuration scheme corresponding to the target water network area is output, including: Inputting the regional meteorological data, the hydrological data, the water quality data and the underlying surface factor data into the feature extraction model; The feature extraction model identifies the input data and determines the data types of various data in the input data; the input data are the regional meteorological data, the hydrological data, the water quality data and the underlying surface factor data; Set up multi-channel input according to the data type of various data; Based on each input channel, feature extraction is performed on various types of input data, and a feature representation is output; The characteristic representation is input into the water surface ratio configuration scheme model, and the target water surface ratio configuration scheme corresponding to the target water network area is output based on the characteristic representation.

3. The method according to claim 2, characterized in that The extracting features of various types of input data based on each input channel and outputting feature representations include: Taking the image data in the input data as image channel input; For the numerical data in the input data, convert the numerical data into a two-dimensional matrix form, and allocate channels according to feature dimensions; Use multiple convolution kernels of different sizes to perform convolution processing on the data input from each input channel to extract target features of different scales; The target features extracted from different levels and different types of data are fused according to the channel dimension to generate the feature representation.

4. The method according to claim 2, characterized in that: The water surface rate configuration scheme model includes a plurality of sub-water surface rate configuration scheme models of different algorithms, and the feature representation is input into the water surface rate configuration scheme model, and the target water surface rate configuration scheme corresponding to the target water network area is output based on the feature representation, including: Inputting the characteristic representations into each of the sub-water surface rate configuration scheme models respectively; Each of the sub-water surface rate configuration scheme models identifies the feature representation and outputs a plurality of first candidate water surface rate configuration schemes respectively; fusing the first candidate water surface rate configuration schemes to generate a plurality of first fused water surface rate configuration schemes; Inputting each of the first fused water surface rate configuration schemes into each of the sub-water surface rate configuration scheme models respectively; Each of the sub-water surface rate configuration scheme models evaluates and updates each of the first fused water surface rate configuration schemes, and outputs a plurality of second candidate water surface rate configuration schemes; This cycle is repeated until a preset number of times, and each of the sub-water surface rate configuration scheme models outputs the corresponding current optimal water surface rate configuration scheme; Each of the current optimal water surface ratio configuration schemes is identified, and the target water surface ratio configuration scheme corresponding to the target water network area is output.

5. The method according to claim 4, characterized in that Each of the sub-water surface rate configuration scheme models identifies the feature representation and outputs a plurality of first candidate water surface rate configuration schemes, including: For one of the sub-water surface ratio configuration scheme models, the sub-water surface ratio configuration scheme model identifies the feature representation; Generate multiple initial water surface rate configuration schemes based on the target constraint conditions in the sub-water surface rate configuration scheme model; the target constraint conditions include at least one of the water surface rate range constraint, water quality standard constraint, water level and flood storage capacity constraint, and land use constraint; Based on the characteristic representation, each of the initial water surface ratio configuration schemes is substituted into the objective function in the sub-water surface ratio configuration scheme model, and the objective function value corresponding to each of the initial water surface ratio configuration schemes is calculated; wherein the objective function includes at least one of a flood control and drainage sub-function, a water resource utilization sub-function, and a water environment capacity sub-function; Determine the historical optimal water surface rate configuration scheme and the current global optimal water surface rate configuration scheme according to the objective function value corresponding to each of the initial water surface rate configuration schemes; Determining an update speed according to the historical optimal water surface rate configuration scheme and the current global optimal water surface rate configuration scheme; According to the update speed, each updated water surface rate configuration scheme is obtained; each updated water surface rate configuration scheme satisfies the target constraint; Calculate the objective function value corresponding to each of the updated water surface rate configuration schemes; Determine the updated water surface rate configuration scheme whose objective function value is greater than the preset function value as the first candidate water surface rate configuration scheme; If there is no updated water surface ratio configuration scheme whose objective function value is greater than the preset function value, the updated water surface ratio configuration scheme continues to be updated until the updated water surface ratio configuration scheme whose objective function value is greater than the preset function value is determined as the first candidate water surface ratio configuration scheme.

6. The method according to claim 5, characterized in that The target water network area includes multiple water bodies, and the objective function includes: ， in, is the flood control and waterlogging removal sub-function; The water body in the target water network area i area; For water The flood storage capacity is as follows: ;in For water The water level, is the flood storage coefficient; and / or, ， in, is the water resource utilization sub-function, For water The available water volume is calculated as: ; Where P is water The precipitation of E is the water body evaporation, β is the regional hydrological conversion coefficient; For water The water use efficiency is defined as: ;in, For agricultural irrigation needs; and / or, ， in, is the water environment capacity subfunction; Pi is the water body The pollution load is usually related to land use and emissions; Vi is the water body The environmental capacity is calculated as: ; Among them, γ is the comprehensive degradation coefficient, which is related to the water body The dilution capacity of is the area of ​​water body i, is the water level of water body i, It is the standard concentration of water quality.

7. The method according to claim 5, characterized in that The target water network area includes multiple water bodies, the target constraint conditions include the water surface rate range constraint, the water quality standard constraint, the water level and flood storage capacity constraint and the land use constraint, and the water surface rate range constraint is: ； in, is the total area of ​​the target water network area; For water Water surface rate; is the upper limit of the water surface rate, is the lower limit of the water surface rate; The water quality standard constraints are: ， ； in, For water of dissolved oxygen, For water The chemical oxygen demand, For water Total nitrogen, For water of total phosphorus; Chemical oxygen demand for environmental standards, Total nitrogen for environmental standards, is the environmental standard value of total phosphorus; The water level and flood storage capacity constraints are: ； in, For water The water level, is the maximum allowable water level of the target water network area; The land use constraints are: ； in: is the total area of ​​the channel; is the total area of ​​the pond; is the total area of ​​wetlands; is the total area of ​​agricultural land, is the total area of ​​the target water network area.

8. The method according to claim 4, characterized in that The step of fusing the first candidate water surface rate configuration schemes to generate a plurality of first fused water surface rate configuration schemes includes: Calculating the similarity between the first candidate water surface rate configuration schemes; For each of the first candidate water surface rate configuration schemes whose similarity is greater than a preset similarity threshold, only one of them is retained to obtain multiple backup water surface rate configuration schemes; The standby water surface rate configuration schemes are cross-combined to generate the first fused water surface rate configuration schemes.

9. The method according to claim 4, characterized in that Identifying each of the current optimal water surface ratio configuration schemes and outputting the target water surface ratio configuration scheme corresponding to the target water network area includes: Inputting each of the current optimal water surface rate configuration schemes into each of the sub-water surface rate configuration scheme models respectively; Each of the sub-water surface rate configuration scheme models evaluates the current optimal water surface rate configuration scheme, and votes on each of the current optimal water surface rate configuration schemes according to the evaluation result; The current optimal water surface rate configuration scheme with the most votes is determined as the target water surface rate configuration scheme.

10. A device for determining a water surface ratio configuration scheme, characterized in that: The device comprises: An acquisition module, used to acquire regional meteorological data, hydrological data, water quality data and underlying surface factor data corresponding to the target water network area; the underlying surface factor data includes at least one of regional topographic data, regional soil type and property data, and regional vegetation type and coverage rate data; An input module, used to input the regional meteorological data, hydrological data, water quality data and underlying surface factor data into a preset water surface ratio configuration scheme model, and output a target water surface ratio configuration scheme corresponding to the target water network area; A configuration module is used to configure the water surface rate corresponding to the target water network area based on the target water surface rate configuration plan.

Images