Decision-making Method, Device and Equipment for River Channel Dredging Scheme Based on Sediment Elution

Abstract

The present invention relates to the technical field of solution decision-making, and discloses a method, device and equipment for decision-making of river channel dredging solutions based on sediment elution. The method includes: collecting and performing standardized preprocessing on the river channel spatial distribution, cross-section morphology, hydrological parameters and sediment pollution characteristics of the river network system to obtain a river channel network feature data set; performing analysis on the migration and diffusion of sediment pollutants and calculating the pollution propagation relationship between nodes to obtain a river channel sediment pollution propagation network model; calculating the coupling degree of river channel sediment pollution to obtain river channel sediment elution risk distribution data; inputting the river channel sediment elution risk distribution data into a P-center problem network optimization model, and generating a river network dredging solution through analysis of minimizing pollution diffusion risk and maximizing the utilization efficiency of dredging resources. The present invention incorporates multiple factors of the river network system into the decision-making process, forms a systematic and standardized decision-making method for dredging solutions, and improves the scientificity and efficiency of river channel treatment.

Description

Technical Field

[0001] The present invention relates to the technical field of solution decision-making, and particularly relates to a method, device and equipment for decision-making of river channel dredging solutions based on sediment elution. Background Art

[0002] As an important part of the water environment, the pollutants in the river bottom sediment with long-term enrichment and resuspension phenomena seriously affect the water ecological environment quality and the navigation safety of the river channel. Traditional river channel dredging solutions mainly rely on expert experience judgment, lacking systematic network analysis methods and scientific decision-making support tools, and it is difficult to accurately grasp the migration and diffusion laws and spatial distribution characteristics of sediment pollutants.

[0003] With the increasing complexity of the river network system governance tasks, the pollution diffusion among river channels shows strong coupling correlation, and the propagation mechanism of sediment pollutants in the river network and the risk quantification calculation method have not been deeply studied. Existing dredging solutions generally have problems such as repeated resource investment and unsatisfactory dredging effects, lacking comprehensive consideration of the integrity of the river network system and the pollutant propagation characteristics. In the current river channel dredging decision-making process, the evaluation method for sediment elution risk is relatively single, and it is difficult to effectively identify high-risk areas and key control sections. At the same time, the optimization allocation of dredging resources and the scientific decision-making of solutions lack quantitative evaluation criteria and systematic optimization models, resulting in the implementation effect of the dredging solution being difficult to be guaranteed. Summary of the Invention

[0004] The present invention provides a method, device and equipment for decision-making of river channel dredging solutions based on sediment elution. The present invention incorporates multiple factors of the river network system into the decision-making process, forming a systematic and standardized decision-making method for dredging solutions, and improving the scientificity and efficiency of river channel treatment.

[0005] In the first aspect, the present invention provides a method for decision-making of river channel dredging solutions based on sediment elution, and the method for decision-making of river channel dredging solutions based on sediment elution includes:

[0006] Collect and perform standardized preprocessing on the river channel spatial distribution, cross-section morphology, hydrological parameters and sediment pollution characteristics of the river network system to obtain a river channel network feature data set;

[0007] Based on the river channel network feature data set, perform analysis on the migration and diffusion of sediment pollutants and calculate the pollution propagation relationship between nodes to obtain a river channel sediment pollution propagation network model;

[0008] Based on the river channel sediment pollution propagation network model, calculate the coupling degree of river channel sediment pollution to obtain river channel sediment elution risk distribution data;

[0009] Input the elution risk distribution data of the river channel sediment into the P - center problem network optimization model, and generate a river network dredging plan through the analysis of minimizing pollution diffusion risk and maximizing the utilization efficiency of dredging resources.

[0010] In a second aspect, the present invention provides a decision - making device for a river channel dredging plan based on sediment elution. The decision - making device for a river channel dredging plan based on sediment elution includes:

[0011] A collection module, configured to collect and perform standardized pre - processing on data such as the river channel spatial distribution, cross - section morphology, hydrological parameters, and sediment pollution characteristics of the river network system to obtain a river channel network feature data set;

[0012] An analysis module, configured to perform an analysis of the migration and diffusion of sediment pollutants based on the river channel network feature data set, and calculate the pollution propagation relationship between nodes to obtain a river channel sediment pollution propagation network model;

[0013] A calculation module, configured to calculate the coupling degree of the river channel sediment pollution based on the river channel sediment pollution propagation network model to obtain the elution risk distribution data of the river channel sediment;

[0014] A generation module, configured to input the elution risk distribution data of the river channel sediment into the P - center problem network optimization model, and generate a river network dredging plan through the analysis of minimizing pollution diffusion risk and maximizing the utilization efficiency of dredging resources.

[0015] In a third aspect of the present invention, a computer device is provided, including: a memory and at least one processor, wherein instructions are stored in the memory; the at least one processor invokes the instructions in the memory to cause the computer device to execute the above - mentioned decision - making method for a river channel dredging plan based on sediment elution.

[0016] In a fourth aspect of the present invention, a computer - readable storage medium is provided, in which instructions are stored. When the instructions run on a computer, the computer is caused to execute the above - mentioned decision - making method for a river channel dredging plan based on sediment elution.

[0017] In the technical solution provided by the present invention, by constructing a river bottom sediment pollution propagation network model and combining time series analysis and causal test methods, the present invention accurately describes the migration and diffusion law of sediment pollutants, providing scientific data support for the decision-making of dredging plans. The present invention uses a multi-dimensional evaluation index system to quantitatively calculate the coupling degree of river bottom sediment pollution, and identifies risk hotspots through spatial statistical analysis methods, effectively improving the accuracy of risk assessment. Based on the P-center problem network optimization theory, the present invention constructs an optimization model with the dual objectives of minimizing the pollution diffusion risk and maximizing the utilization efficiency of dredging resources, solving the problem of repeated resource investment in traditional dredging plans. The present invention uses the branch and bound method to solve the 0-1 integer programming problem, and through strict constraint conditions and optimization algorithms, ensures the feasibility and optimality of the dredging plan. In the process of optimization solution, the present invention uses objective balance analysis and constraint matrix standardization processing to establish a complete solution parameter configuration system, ensuring the convergence and stability of the optimization results. The present invention incorporates multiple factors such as the integrity of the river network system, the characteristics of pollutant propagation, and the utilization efficiency of resources into the decision-making process, forming a systematic and standardized decision-making method for dredging plans, improving the scientificity and efficiency of river regulation. Brief Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a schematic diagram of the steps of the decision-making method for the river dredging plan based on sediment elution in the embodiments of the present invention;

[0020] Figure 2 It is a schematic diagram of the structure of the decision-making device for the river dredging plan based on sediment elution in the embodiments of the present invention;

[0021] Figure 3 It is a schematic block diagram of the structure of the computer device in the embodiments of the present invention. Detailed Embodiments

[0022] The embodiments of the present invention provide a decision-making method, device and equipment for river channel dredging solutions based on sediment elution. The terms "first", "second", "third", "fourth", etc. (if any) in the description, claims and above-mentioned drawings of the present invention are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the term "comprising" or "having" and any variation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or equipment comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or equipment.

[0023] For ease of understanding, the specific process of the embodiments of the present invention will be described below. Please refer to Figure 1 , an embodiment of the decision-making method for river channel dredging solutions based on sediment elution in the embodiments of the present invention includes:

[0024] Step S1: Collect and perform standardized preprocessing on the river channel spatial distribution, cross-section morphology, hydrological parameters and sediment pollution characteristics of the river network system to obtain a river channel network feature dataset;

[0025] It can be understood that the execution subject of the present invention can be a decision-making device for river channel dredging solutions based on sediment elution, or a terminal or a server. Specifically, it is not limited here. The embodiments of the present invention will be described by taking the server as the execution subject as an example.

[0026] Specifically, conduct a Geographic Information System (GIS) survey of the river network system to collect spatial coordinate data including longitude, latitude, and elevation to describe the spatial distribution of river channels. Through GIS technology, generate the original spatial distribution data of the river channels. Input the original spatial distribution data of the river channels into a topological analysis model, and generate the topological structure data of the river network by calculating key parameters such as node connectivity, path connectivity, and network density. The topological structure data of the river network is the core for describing the complexity of the river channel network, reflecting the connection relationships between river channels and the connectivity of the overall network. Based on the topological structure data of the river network, collect the morphological parameters of the river channel cross-sections. Measure morphological parameters such as river channel width, water depth, cross-sectional area, and longitudinal slope. These data are obtained through various means such as on-site surveys, laser scanning technology, or unmanned aerial vehicle (UAV) aerial survey technology to obtain the river channel morphological feature data. Based on the river channel morphological feature data, set up hydrological monitoring points to collect key hydrological elements including cross-sectional flow velocity, flow rate, water level, and sediment concentration. These hydrological feature data are obtained through means such as flow meters, automatic monitoring equipment, or remote sensing technology to dynamically reflect the hydrological conditions of the river channel. Use the river channel hydrological feature data as the sampling basis to conduct stratified sampling and pollutant detection of the river channel sediment. The sampling process is based on standard stratified methods to ensure that sediment samples at different depths can comprehensively reflect the distribution of pollutants. Through laboratory analysis, detect the types, concentration distributions, and spatial variability of pollutants in the sediment to obtain complete sediment pollutant feature data. In order to convert the sediment pollutant feature data into a spatial distribution model, establish a spatial distribution map of pollutants based on the Kriging spatial interpolation model. This model optimizes the interpolation results through variogram calculation and cross-validation to generate the spatial distribution data of sediment pollutants, showing the distribution pattern of pollutants in the river network system and helping to identify pollution hotspots and pollution gradients. Normalize the topological structure data of the river network, the river channel morphological feature data, the river channel hydrological feature data, and the spatial distribution data of sediment pollutants. Adopt standardization methods such as Min-Max normalization or Z-score normalization to convert data of different scales and units into a unified dimensionless form to improve the consistency and comparability of the data and obtain standardized river channel feature data. Input the standardized river channel feature data into a density-based clustering algorithm model, and extract the river channel network feature dataset by calculating the Euclidean distance and local density of the river channel feature vectors. Density clustering algorithms (such as DBSCAN) can effectively identify high-density regions and river reaches with similar features in the river network during this process, generating a river channel network feature dataset containing rich information.

[0027] Step S2: Conduct an analysis of the migration and diffusion of sediment pollutants based on the river channel network feature dataset, and calculate the pollution propagation relationship between nodes to obtain a river channel sediment pollution propagation network model;

[0028] Specifically, hydrological feature data is extracted from the river network feature dataset and discretized for time series processing. The continuous hydrological data is discretized into the pollutant concentration change rate and water flow velocity field at multiple time points, enabling the data to reflect the migration state of pollutants at different time points and obtaining the basic data for pollutant migration. The basic data for pollutant migration and the river channel morphological feature data are input into an autoregressive model. By calculating the autocorrelation coefficient and partial autocorrelation coefficient of the pollutant concentration, the regularity of the pollutant concentration change over time is mined. The autoregressive model can capture the inherent trend in the time series, reveal the time-dependent characteristics existing in the pollutant migration process, and generate a dependent feature sequence with time series characteristics. To analyze the complex behavior of pollutant migration, the time-dependent feature sequence is decomposed at multiple scales to extract the pollutant migration pattern data at different scales. These data reveal the migration laws of pollutants at microscopic and macroscopic time scales and provide more comprehensive dynamic feature information. The pollutant migration pattern data is input into a Granger causality test model, and the Granger causality is used to analyze the causal effect of the pollutant concentration. By calculating the F statistic and P value of the concentration change, a pollution propagation probability matrix between nodes is generated. This matrix represents the probability of pollutant propagation between any two nodes in the network, quantifying the causal relationship of pollutant migration from a statistical perspective. A two-region logarithmic distribution fitting is performed on the pollution propagation probability matrix to more precisely characterize the strength characteristics of network connections. The two-region logarithmic distribution fitting can reveal the distribution law of the connection strength between nodes, providing an important basis for the structural analysis of the pollution propagation network. Based on the fitting results, network connection strength data is generated to quantify the pollutant propagation ability and interaction strength between each node in the network. Community detection is performed on the network connection strength data. Through community detection, the key channels for pollution propagation in the network are identified. These channels are the main paths for pollutant migration and are also parts of high importance in the network structure. After identifying the key channels, the global efficiency and local efficiency of the network are calculated to quantify the overall propagation ability and local response ability of the pollution propagation network. The combination of these indicators can reflect the propagation ability index of the network. Combining the pollution propagation probability matrix between nodes, the network connection strength data, and the network propagation ability index, a river sediment pollution propagation network model is constructed. This model can describe the dynamic diffusion process of pollutants in the river network system and reveal the key nodes, paths, and propagation bottlenecks in the network.

[0029] Step S3: Based on the river sediment pollution propagation network model, calculate the coupling degree of river sediment pollution to obtain the river sediment elution risk distribution data;

[0030] Specifically, classification is carried out based on the network propagation ability index in the river sediment pollution propagation network model, and the network propagation ability index is divided into different levels to generate pollution assessment index data. The differences in pollution levels are refined to accurately reflect the propagation characteristics and the changes in the strength of river sediment pollution, forming a clear basis for pollution assessment. Based on the pollution assessment index data, the co-variation coefficient between the water quality safety index and the ecological risk index is calculated to quantify the dynamic correlation between the two. The calculation of the co-variation coefficient is achieved by constructing a pollution coupling degree matrix, where each matrix element represents the coupling strength of water quality safety and ecological risk in different regions. Through this matrix, the comprehensive impact degree of pollutants on water quality and the ecological environment is revealed. The sediment pollutant release cycle and change intensity of the pollution coupling degree matrix are analyzed to explore the dynamic law of pollution diffusion and generate pollution diffusion dynamic data. By setting water quality safety thresholds, ecological risk thresholds, and treatment response thresholds, risk classification analysis is carried out on the pollution diffusion dynamic data. Each risk level corresponds to a specific environmental impact range and treatment requirement, generating node risk score data. The local Moran's I and global Moran's I of the node risk score data are calculated to reveal the spatial aggregation characteristics of the risk distribution. The local Moran's I is used to identify high-risk hotspots in local areas, while the global Moran's I provides information on the overall spatial distribution pattern of risks. Through the analysis of Moran's I, the concentrated or dispersed state of risks in the river network system is clarified, forming risk aggregation characteristic data. On this basis, kernel density analysis is applied to process the risk aggregation characteristic data, thereby determining the spatial distribution trend and density change of high-risk areas and generating risk hotspot distribution data. Based on the risk hotspot distribution data, risk area identification is carried out to determine the specific scope of each high-risk area. The identification process combines geographic information systems and spatial analysis methods to show the key areas and main paths of pollution diffusion. According to the identified risk zoning results, the pollution intensity index and spatial diffusion index of each risk area are calculated. The pollution intensity index can reflect the total amount of pollution in the area and its threat to the environment, while the spatial diffusion index measures the ability and trend of pollution to spread from high-concentration areas to the periphery. These indicators together constitute a quantitative assessment of the risk of river sediment elution. Through the above comprehensive analysis process, river sediment elution risk distribution data is generated.

[0031] Step S4: Input the riverbed sediment elution risk distribution data into the P - center problem network optimization model. Through the analysis of minimizing pollution diffusion risk and maximizing the utilization efficiency of dredging resources, a river network dredging plan is generated. Specifically, input the riverbed sediment elution risk distribution data into the data pre - processing layer of the P - center problem network optimization model for standardization processing. Uniformly normalize the risk level, spatial location, and dredging resource parameters to eliminate the differences in data scale and dimension, ensure the accuracy of subsequent optimization calculations, and obtain the basic data for optimization calculations, so that each node in the river network system and its corresponding risk level, spatial distribution, and allocable resources can be clearly quantified. Based on the optimized calculation basic data after standardization processing, construct the constraint condition layer of the P - center problem network optimization model. Set a multi - dimensional constraint condition set, which includes specific limiting conditions such as hydrological demand constraints, river channel connectivity constraints, and resource input constraints. The hydrological demand constraint ensures that the dredging plan can meet the basic requirements of river channel hydrological balance and avoid hydrological damage caused by dredging; the river channel connectivity constraint guarantees that the overall function of the river network system is not affected, especially the dredging of key nodes must be coordinated with the network connectivity characteristics; the resource input constraint reasonably restricts the allocation of dredging resources to ensure that the optimized plan can be implemented under actual conditions. These constraint conditions together constitute the basic rules for optimization calculations and restrict the search space of potential solutions. Construct the dredging location selection variables for the multi - dimensional constraint condition set. Define the 0 - 1 decision variables for the dredging operation for each river channel node, indicating whether the node is selected for dredging operations, and at the same time define the selection variables for the dredging method to show the specific dredging techniques or methods adopted at different nodes. These decision variables are represented in matrix form and constitute the core input of the optimization model, which can not only reflect the dredging priority of each node but also guide the effective allocation of dredging resources. Input the decision variable matrix into the objective function construction layer of the P - center problem network optimization model to establish a bi - objective optimization function of minimizing pollution diffusion risk and maximizing resource utilization efficiency. The sub - objective function of minimizing pollution diffusion risk aims to prioritize the dredging of high - risk nodes to reduce the possibility of further pollutant diffusion; the sub - objective function of maximizing resource utilization efficiency optimizes the spatial and technical allocation of resources to ensure the maximum treatment effect with the minimum input. These two sub - objective functions are integrated into a bi - objective optimization problem in the optimization model, reflecting the scientific nature and practical feasibility of the dredging plan. To solve the bi - objective optimization problem, configure the LINGO solver, and determine the optimization solution parameters by setting the solution accuracy, iteration times, and convergence conditions. The high efficiency and flexibility of the LINGO solver make it a tool for solving large - scale optimization problems. The core of the solver configuration is to ensure that the optimization process can find the optimal solution that meets the accuracy requirements within a reasonable time, while avoiding getting stuck in local optima or not converging for a long time. Input the optimization solution parameters into the optimization calculation layer of the P - center problem network optimization model, and use the branch - and - bound method to solve the 0 - 1 integer programming problem.The branch and bound method is an optimization algorithm that can effectively handle discrete optimization problems with complex constraints. By gradually dividing the solution space and estimating the bounds for each branch, the algorithm can quickly eliminate infeasible or non-optimal solutions and finally generate a globally optimal river network dredging plan.

[0032] Perform objective balance analysis on the bi-objective optimization function. By constructing a normalized weighted expression for the pollution diffusion risk weight coefficient and the resource utilization efficiency weight coefficient, adjust the weight distribution relationship between the two objectives to ensure that the objective function can balance the priority of pollution control and the efficiency of resource allocation, and generate the objective balance parameter. Input the objective balance parameter into the constraint matrix normalization module, and process the multi-dimensional constraint condition set by the main diagonal element normalization method. The core of this normalization method lies in ensuring that the numerical ranges of the constraint conditions are consistent, thus avoiding adverse effects on the optimization results due to scale differences. The processed normalized constraint matrix not only improves the solution efficiency of the model, but also enhances the numerical stability of the constraint conditions during the optimization process. Perform the initialization settings for the linear programming of the normalized constraint matrix. Use the simplex method to initialize the basic feasible solution, and at the same time set the upper and lower bounds for the non-basic variables. These initial solution parameters are an important basis for the optimization process and determine the initial search space and the feasibility of the solution of the branch and bound method. By optimizing the initial solution parameters, significantly accelerate the convergence speed of the optimization model and improve the solution accuracy. Input the initial solution parameters into the branch and bound solution control module, and configure the node selection rule and the branch variable selection strategy. The node selection rule determines the nodes to be expanded preferentially during the optimization process, while the branch variable selection strategy is used to judge the splitting priority of the branch variables. These two together constitute the branch strategy parameters, which directly affect the efficiency and path of the optimization solution. Construct the integer variable processing rule based on the branch strategy parameters. The solution of integer programming requires setting the gap tolerance value of the integer variable to control the solution accuracy, and at the same time dynamically adjust the value range of the integer variable through the rounding direction judgment condition. Through reasonable setting of the integer programming parameters, ensure that the optimization model takes into account both the computational efficiency and the solution accuracy when dealing with discrete decision variables. Set the convergence judgment rule for the integer programming parameters. By constructing the threshold conditions for the relative improvement rate of the objective function and the constraint satisfaction degree, effectively control the convergence of the optimization process. When the relative improvement rate of the objective function is lower than the set threshold and all constraint conditions are satisfied, the optimization process terminates, improving the stability and reliability of the model solution. The setting of the convergence control parameters provides a clear termination condition for the solution process, preventing unnecessary iterations from prolonging the solution time. Input the convergence control parameters into the solution process monitoring module, and set the maximum number of iterations and the upper limit of the calculation time to form the termination criterion parameters. These parameters ensure that the LINGO solver can complete the optimization task within the specified time and computational resources, thereby improving the solution efficiency and avoiding being trapped in a long-time calculation dilemma. By combining the integer programming parameters, the convergence control parameters and the termination criterion parameters, comprehensively configure the parameters of the LINGO solver, and finally obtain the optimized solution parameters.

[0033] Relax the standardized constraint matrix in the optimization solution parameters, converting the original 0-1 integer constraint into an interval constraint of continuous variables. By relaxing the restrictions on decision variables, the problem is transformed into a relaxed problem of linear programming, providing good computational conditions for the rapid solution of the initial solution. Input the linear programming relaxation problem into the simplex calculation module, and obtain the initial optimal solution through iterative solution. In this process, the simplex method quickly determines the optimal solution with an efficient path search method, and the upper bound value data of the generated nodes can be used as the basis for the selection of subsequent branch variables. Based on the upper bound value data of the nodes, construct a branch variable selection rule. By calculating the non-integer degree and influence coefficient of each decision variable, determine the optimal branch variable. This branch variable selection rule can preferentially process variables with a larger non-integer degree, thus effectively reducing the search space of the solution, improving the solution efficiency of the algorithm, and generating a clear branch sequence. Input the generated branch sequence into the branch and bound tree construction module, generate sub-problems by dynamically splitting nodes, and solve each sub-problem to determine its optimal solution. The solution of each sub-problem is used to update the active node set, and each node in the active node set represents a potential feasible solution. By calculating the difference between the objective function value of each active node and the current optimal value, quantify the priority and potential value of the node to form node evaluation data. Input the node evaluation data into the node traversal module, and select the next node to be processed based on the optimal depth-first search strategy. The depth-first search strategy can quickly mine high-value nodes and reduce the search scope of the solution by preferentially expanding the optimal solution path of the current branch, thereby optimizing the computational efficiency. Conduct a feasibility test on each node in the candidate node set, including verifying the integer characteristics of the solution and the degree of constraint satisfaction, to form a feasible solution set. The verification of integer characteristics ensures that the optimized solution meets the basic requirements of 0-1 integer programming, while the constraint satisfaction test confirms that the solution can meet all multi-dimensional constraint conditions. After obtaining the feasible solution set, enter the stage of generating the dredging plan. Complete the specific allocation of dredging tasks by mapping decision variables to the dredging operation plan of river network nodes. The dredging operation plan for each node includes key information such as whether to conduct dredging, the selection of dredging methods, and the amount of resource input, forming a complete river network dredging plan.

[0034] In the embodiments of the present invention, by constructing a river bottom sediment pollution propagation network model and combining time series analysis and causal test methods, the migration and diffusion laws of bottom sediment pollutants are accurately described, providing scientific data support for the decision-making of dredging plans. The present invention uses a multi-dimensional evaluation index system to quantitatively calculate the coupling degree of river bottom sediment pollution, and identifies risk hot spots through spatial statistical analysis methods, effectively improving the accuracy of risk assessment. Based on the P-center problem network optimization theory, the present invention constructs an optimization model with the dual objectives of minimizing pollution diffusion risk and maximizing the utilization efficiency of dredging resources, solving the problem of repeated resource investment in traditional dredging plans. The present invention uses the branch and bound method to solve the 0-1 integer programming problem, and through strict constraint conditions and optimization algorithms, ensures the feasibility and optimality of the dredging plan. In the process of optimization solution, the present invention uses objective balance analysis and constraint matrix standardization processing to establish a complete solution parameter configuration system, ensuring the convergence and stability of the optimization results. The present invention incorporates multiple factors such as the integrity of the river network system, pollutant propagation characteristics, and resource utilization efficiency into the decision-making process, forming a systematic and standardized dredging plan decision-making method, improving the scientificity and efficiency of river regulation.

[0035] In a specific embodiment, the process of executing step S1 may specifically include the following steps:

[0036] Perform geographic information system mapping on the river network system, collect spatial coordinate data including longitude, latitude, and elevation, and obtain the original data of the river channel spatial distribution;

[0037] Input the original data of the river channel spatial distribution into the topological analysis model, and obtain the river network topological structure data through the calculation of node connectivity, path connectivity, and network density;

[0038] Collect the morphological parameter of the river channel section in the river network topological structure data, measure the parameters including river channel width, water depth, cross-sectional area, and longitudinal slope, and obtain the river channel morphological characteristic data;

[0039] Based on the river channel morphological characteristic data, set up hydrological monitoring points, collect hydrological elements including cross-sectional velocity, flow rate, water level, and sediment concentration, and obtain the river channel hydrological characteristic data;

[0040] Take the river channel hydrological characteristic data as the sampling basis, conduct stratified sampling and pollutant detection on the river bottom sediment, and obtain the bottom sediment pollutant characteristic data, which includes pollutant types, concentration distribution, and spatial variability;

[0041] According to the bottom sediment pollutant characteristic data, establish a Kriging spatial interpolation model, and obtain the bottom sediment pollutant spatial distribution data through variogram calculation and cross-validation;

[0042] Normalize the river network topological structure data, river channel morphological feature data, river channel hydrological feature data, and spatial distribution data of sediment pollutants to obtain standardized river channel feature data;

[0043] Input the standardized river channel feature data into the density-based clustering algorithm model, and obtain the river channel network feature dataset by calculating the Euclidean distance and local density of the river channel feature vectors.

[0044] Specifically, conduct Geographic Information System (GIS) mapping on the river network system, and collect spatial coordinate data including longitude , latitude , and elevation . Through high-precision satellite remote sensing, unmanned aerial vehicle mapping, or ground survey, obtain the original data of the river channel spatial distribution. Assume that the river network contains key nodes, and the coordinates of each node are recorded as , where . These data constitute the geometric skeleton of the river channel. Input the original data of the river channel spatial distribution into the topological analysis model to analyze the connection relationship, path connectivity, and network density between nodes. The node connectivity is defined as the number of nodes directly connected to other nodes: ;

[0045] where is an element of the adjacency matrix. If nodes and are directly connected, then , otherwise . The path connectivity( ) quantifies the connectivity degree between any two nodes in the network, and the calculation formula is: ;

[0046] where, ShortestPath represents the shortest path length between nodes and . The network density is defined as the ratio of the actual number of connections to the number of all possible connections: ;

[0047] Through calculation, generate the topological structure data of the river network, which is used to describe the overall connectivity and complexity of the river channel system. Based on the topological structure data, collect the morphological parameters of the river channel cross-section. By measuring the river channel width , water depth , cross-sectional area , and longitudinal slope( ), obtain the morphological feature data of the river channel. The cross-sectional area is expressed as: ;

[0048] The longitudinal slope is defined as the elevation change within a unit distance: ;

[0049] These parameters can be efficiently collected through UAV aerial surveys or underwater detection equipment, providing support for the analysis of the geometric characteristics of the river channel. Hydrological monitoring points are set according to the river channel morphological characteristic data, and hydrological element data including cross-sectional velocity , discharge , water level and sediment concentration are collected. The calculation formula for discharge is: ;

[0050] Hydrological data can dynamically reflect the hydrodynamic behavior of the river channel and provide input basis for the modeling of pollutant migration and diffusion. After obtaining the hydrological characteristic data, based on it, bottom sediment stratified sampling and pollutant detection are carried out to analyze the types of pollutants in the bottom sediment( ), concentration distribution and spatial variability . The results of stratified sampling establish a distribution function , representing the variation relationship of concentration with depth. The analysis of the spatial variability of pollutant concentration is completed by calculating the variogram : ;

[0051] Among them, represents the number of sampling pairs with a spacing of . Based on the bottom sediment pollutant characteristic data, the Kriging spatial interpolation model is used to generate pollutant spatial distribution data. The Kriging model generates spatial interpolation prediction values through variogram fitting: ;

[0052] Among them, is the interpolation weight, which is solved by the Kriging equation system. The river network topological structure data, river channel morphological characteristic data, river channel hydrological characteristic data and bottom sediment pollutant spatial distribution data are normalized. The normalization formula is: ;

[0053] Among them, is the original data, and are its minimum and maximum values respectively. The normalized river channel characteristic data is input into a density-based clustering algorithm model (such as DBSCAN), and the clustering analysis is completed by calculating the Euclidean distance and local density of the river channel feature vectors. The calculation formula for the Euclidean distance is: ;

[0054] Among them, and respectively represent the th and th eigenvalue of the th sample, is the number of feature dimensions. The local density is defined as: ;

[0055] Among them, is an indicator function, which takes the value of 1 when , otherwise 0. Through cluster analysis, a river network feature dataset is obtained, revealing areas with similar pollution characteristics and potential priority areas for dredging in the river network.

[0056] In a specific embodiment, the process of executing step S2 may specifically include the following steps:

[0057] Perform time series discretization processing on the hydrological feature data in the river network feature dataset, and calculate the pollutant concentration change rate and water flow velocity field at each time point to obtain the basic data for pollutant migration;

[0058] Input the basic data for pollutant migration and the river channel morphology feature data into an autoregressive model, and calculate the autocorrelation coefficient and partial autocorrelation coefficient of the pollutant concentration to obtain a time-dependent feature sequence;

[0059] Perform multi-scale decomposition on the time-dependent feature sequence to obtain pollutant migration pattern data, and input the pollutant migration pattern data into a Granger causality test model to calculate the F statistic and P value of the pollutant concentration to obtain a pollution propagation probability matrix between nodes;

[0060] Perform a two-region logarithmic distribution fitting on the pollution propagation probability matrix between nodes to obtain network connection strength data;

[0061] Perform community division on the network connection strength data to obtain pollution propagation key channel data, and calculate the global efficiency and local efficiency based on the pollution propagation key channel data to obtain a network propagation ability index;

[0062] Construct a river sediment pollution propagation network model based on the pollution propagation probability matrix between nodes, network connection strength data, and network propagation ability index.

[0063] Specifically, perform time series discretization processing on the hydrological feature data in the river network feature dataset. Assume that the time range is , and collect hydrological features at each time point , including pollutant concentration and water flow velocity field . The pollutant concentration change rate is defined as: ;

[0064] Among them, is the time interval; and the water flow velocity field is obtained by two-dimensional or three-dimensional flow field calculation, and each of its components , or is expressed as: ;

[0065] Based on these data, the basic data describing the dynamic behavior of pollutant migration is formed. The basic data of pollutant migration and the river channel morphological characteristic data (including width , depth , slope , etc.) are input into the autoregressive model (AR). The AR model uses past concentration values to predict the current concentration, and the model form is: ;

[0066] Among them, is the regression coefficient, is the lag order, is the white noise term. By calculating the autocorrelation coefficient and the partial autocorrelation coefficient , the time dependence of pollutant concentration is quantified: ;

[0067] The analysis results of autocorrelation and partial autocorrelation form a time-dependent feature sequence, revealing the temporal law of pollutant migration. The time-dependent feature sequence is decomposed at multiple scales, and the concentration sequence is decomposed into change patterns at multiple scales. Assuming the original signal is , after discrete wavelet transform, the detail component and the approximation component are obtained: ;

[0068] Among them, represents the decomposition level. The components at each scale reflect different dynamic characteristics of pollutant migration, forming pollutant migration pattern data. The migration pattern data is input into the Granger causality test model, and by calculating the F statistic and P value of pollutant concentration, the causal relationship of pollution propagation between nodes is analyzed. Assuming the concentration sequences of two nodes and are and respectively, the hypothesis of the Granger test is " does not Granger cause ", and its test model is: ;

[0069] Through the F - test statistic: ;

[0070] where RSS is the residual sum of squares, is the number of samples, The significance of is determined by the P - value. Generate the pollution propagation probability matrix between nodes to . Each element of the matrix represents the pollution propagation probability from node Perform a two - zone logarithmic distribution fitting on the propagation probability matrix and calculate the network connection strength data. Assume that the propagation intensity ;

[0071] where and are fitting parameters, is the error term. The fitting results reveal the statistical characteristics of the propagation intensity. Use the network connection strength data for community division to identify the key channels of pollution propagation. Perform community division by maximizing the modularity : ;

[0072] where is the adjacency matrix, and are the degrees of the nodes, is the total number of edges, is the indicator function, which takes 1 only when nodes and belong to the same community. Based on the community division results, calculate the global efficiency and the local efficiency to quantify the network propagation ability. The global efficiency is defined as: ;

[0073] where is the shortest path length between nodes and . The local efficiency is: ;

[0074] Combine the propagation probability matrix, network connection strength and propagation ability index , to construct a river sediment pollution propagation network model.

[0075] In a specific embodiment, the process of executing step S3 may specifically include the following steps:

[0076] Classify the network propagation ability index in the river sediment pollution propagation network model to obtain pollution assessment index data;

[0077] Based on the pollution assessment index data, calculate the co-variation coefficient between the water quality safety index and the ecological risk index to obtain a pollution coupling degree matrix, and analyze the sediment pollutant release period and change intensity of the pollution coupling degree matrix to obtain pollution diffusion dynamic data;

[0078] Set the water quality safety threshold, ecological risk threshold and governance response threshold, and conduct risk classification on the pollution diffusion dynamic data to obtain node risk score data;

[0079] Calculate the local Moran index and global Moran index for the node risk score data to obtain risk aggregation characteristic data, and conduct kernel density analysis on the risk aggregation characteristic data to obtain risk hot spot distribution data; Identify risk regions based on the risk hot spot distribution data to obtain risk zoning results, and according to the risk zoning results, calculate the pollution intensity index and spatial diffusion index of each risk region to obtain the river sediment elution risk distribution data.

[0080] Specifically, classify the network propagation ability index in the river sediment pollution propagation network model. The network propagation ability index quantifies the importance and efficiency of each node in pollution propagation. Assume the node propagation ability index is , where , use the quantile method for classification, and divide into levels. The classified pollution assessment index data is expressed as: ;

[0081] Among them, the Rank function divides the propagation ability levels according to quantiles. The result of this step assigns a corresponding assessment level to each node to distinguish high-risk and low-risk regions. Based on the pollution assessment index data , calculate the water quality safety index and the co-variation coefficient between the ecological risk index to construct a pollution coupling degree matrix. The co-variation coefficient reflects the comprehensive impact of pollutants on water quality and the ecosystem, and is defined as: ;

[0082] Among them, and are the means of the water quality safety index and the ecological risk index respectively. The co-variation coefficient matrix is expressed as: ;

[0083] This matrix provides the correlation strength between pollution indicators. Analyze the sediment pollutant release cycle and change intensity of the pollution coupling degree matrix to generate pollution diffusion dynamic data. The release cycle describes the time characteristics of pollutant release from the sediment, and the change intensity represents the fluctuation range of pollutant concentration, which are calculated respectively as: ;

[0084] where is the pollution release frequency of node , is the standard deviation of the concentration sequence . Combine these dynamic data to capture the variation laws of pollutants in time and space. Set the water quality safety threshold , the ecological risk threshold and the governance response threshold ), and conduct risk grading on the pollution diffusion dynamic data. The node risk score data ( ) is calculated as: ; ;

[0085] This score reflects the comprehensive risk level of the node at the water quality, ecological and governance levels. Input the node risk score data into the spatial statistical model to calculate the local Moran's I and the global Moran's I to reveal the spatial clustering characteristics of risks. The local Moran's I is defined as: ;

[0086] where is the mean of the node risk scores, is the variance, is the node and 's spatial weight. The global Moran's I is: ;

[0087] By calculating the local and global Moran's I, reveal whether high-risk areas show spatial clustering characteristics. Conduct kernel density analysis on the risk clustering characteristic data to generate risk hotspot distribution data. The kernel density function is expressed as: ;

[0088] where is the kernel function, is the smoothing bandwidth, is the node location. Kernel density analysis can visually present the spatial distribution of high-risk areas. Based on the risk hotspot distribution data, conduct risk area identification and divide the risk zoning results. Combine the risk zoning results to calculate the pollution intensity index and the spatial diffusion index , the risk distribution data of riverbed sediment elution is obtained. The pollution intensity index is defined as: ;

[0089] Among them, represents the set of nodes in a certain risk zone. The spatial diffusion index is defined as: ;

[0090] Among them, is the node and 's spatial distance.

[0091] In a specific embodiment, the process of executing step S4 may specifically include the following steps:

[0092] Input the risk distribution data of riverbed sediment elution into the data preprocessing layer of the P-center problem network optimization model, and perform standardization processing on the risk level, spatial location, and dredging resource parameters to obtain the basic data for optimized calculation;

[0093] Input the basic data for optimized calculation into the constraint condition construction layer of the P-center problem network optimization model, and obtain a multi-dimensional constraint condition set by setting the limiting conditions of hydrological demand constraints, river connectivity constraints, and resource input constraints;

[0094] Construct the dredging location selection variables for the multi-dimensional constraint condition set, and obtain the decision variable matrix by setting the 0-1 decision variables for dredging operations and the selection variables for dredging methods at each river node;

[0095] Input the decision variable matrix into the objective function construction layer of the P-center problem network optimization model, and obtain a double-objective optimization function by constructing a sub-objective function for minimizing pollution diffusion risk and a sub-objective function for maximizing resource utilization efficiency;

[0096] Configure the LINGO solver for the double-objective optimization function and the multi-dimensional constraint condition set, and obtain the optimized solution parameters by setting the solution accuracy, iteration times, and convergence conditions;

[0097] Input the optimized solution parameters into the optimized calculation layer of the P-center problem network optimization model, and use the branch and bound method to solve the 0-1 integer programming problem to generate a river network dredging plan.

[0098] Specifically, input the risk distribution data of riverbed sediment elution into the data preprocessing layer of the P-center problem network optimization model for standardization processing. The risk distribution data includes the risk level , spatial location and dredging resource parameters , where represents the river node. To ensure the unity of data in different dimensions, the normalization method is used to map all variables to the range of [0,1]: ;

[0099] Among them, are the maximum and minimum values of the risk level, spatial coordinates, and resource parameters respectively. After standardization, the basic data for optimized calculation are obtained to ensure the numerical stability and convergence of subsequent optimized model calculations. The basic data for optimized calculation are input into the constraint condition construction layer of the P-center problem network optimization model to generate a multi-dimensional constraint condition set. The constraint conditions include hydrological demand constraints ), river channel connectivity constraints ), and resource input constraints ( ). The hydrological demand constraints ensure that the river channel after dredging meets the minimum flow requirement , and the formula is: ;

[0100] Among them, is the cross-sectional area of the river channel at node , and is the flow velocity at the node. The river channel connectivity constraints ensure that the distance and between any two nodes does not exceed the maximum allowable distance to ensure the connectivity of the dredging plan: ;

[0101] The resource input constraints limit that the total dredging resources shall not exceed the total available resources : ;

[0102] These constraints together constitute the basic rules of the optimization model and restrict the feasible range of the solution. Based on the multi-dimensional constraint condition set, a dredging location selection variable matrix is constructed. The dredging operation variable for each node is set to indicate whether to dredge node , and at the same time, the dredging method selection variable is set to indicate to select the th dredging method at node . The definition of the decision variable matrices and is: ;

[0103] Among them, represents the total number of dredging methods. The constraint variables need to satisfy: ;

[0104] This indicates that only when node Only when dredging is to be carried out can a specific dredging method be selected. The decision variable matrix is input into the objective function construction layer to generate a bi-objective optimization function that minimizes the pollution diffusion risk and maximizes the resource utilization efficiency. The sub-objective function for minimizing the pollution diffusion risk is defined as: ;

[0105] The sub-objective function for maximizing the resource utilization efficiency is: ;

[0106] The comprehensive bi-objective function is expressed using the weighted linear combination method as: ;

[0107] Among them, and are the weights of the objective functions, used to balance the importance of the two. Configure the LINGO solver for the bi-objective optimization function and the multi-dimensional constraint condition set, and obtain the optimization solution parameters by setting the solution accuracy ( ), the number of iterations and the convergence condition. The branch and bound method is used in the solution process to optimize the 0-1 integer programming problem. The branch and bound method decomposes the problem into several sub-problems and gradually converges by setting the upper and lower bounds and of the nodes: ;

[0108] The selection of the branching variable is based on the optimal depth-first search strategy, and the current optimal solution is iteratively updated until the convergence condition is satisfied or the maximum number of iterations is reached. Input the optimization solution parameters into the optimization calculation layer of the P-center problem network optimization model, and generate a river network dredging plan through the branch and bound method. The dredging plan includes whether each node is to be dredged and the selection of the dredging method.

[0109] In a specific embodiment, the process of configuring the LINGO solver for the bi-objective optimization function and the multi-dimensional constraint condition set and obtaining the optimization solution parameters by setting the solution accuracy, the number of iterations, and the convergence condition may specifically include the following steps:

[0110] Conduct an objective balance analysis on the bi-objective optimization function, and obtain the objective balance parameter by constructing a normalized weighted expression of the pollution diffusion risk weight coefficient and the resource utilization efficiency weight coefficient;

[0111] Input the objective balance parameter into the constraint matrix normalization module, and process the multi-dimensional constraint condition set based on the main diagonal element normalization method to obtain the normalized constraint matrix;

[0112] Initialize the standardized constraint matrix by linear programming, set the initial basic feasible solution and the upper and lower bounds of non-basic variables of the simplex method to obtain the initial solution parameters, input the initial solution parameters into the branch and bound solving control module, and configure the node selection rule and the branch variable selection strategy to obtain the branch strategy parameters;

[0113] Construct the integer variable processing rule based on the branch strategy parameters, set the gap tolerance value and the rounding direction judgment condition of the integer variable to obtain the integer programming parameters, set the convergence judgment rule for the integer programming parameters, and construct the threshold conditions of the relative improvement rate of the objective function and the constraint satisfaction degree to obtain the convergence control parameters;

[0114] Input the convergence control parameters into the solving process monitoring module, set the maximum number of iteration steps and the upper limit of the calculation time to obtain the termination criterion parameters, and configure the LINGO solver parameters based on the integer programming parameters, the convergence control parameters and the termination criterion parameters to obtain the optimization solving parameters.

[0115] Specifically, perform objective balance analysis on the bi-objective optimization function to ensure the coordination of the two objectives of minimizing the pollution diffusion risk and maximizing the resource utilization efficiency. Assume that the pollution diffusion risk objective function is , and the resource utilization efficiency objective function is , where represents the pollution risk level of node , is a decision variable indicating whether the node is dredged represents the resource utilization selection, is the dredging resource parameter. In order to construct a balanced objective function, normalize the two objectives to obtain the normalized weight coefficients: ;

[0116] The normalized bi-objective function is expressed as: ;

[0117] By adjusting and , flexibly control the importance of the two objectives according to different decision preferences to generate the objective balance parameter . Input the objective balance parameter into the constraint matrix standardization module to perform the normalization processing of the main diagonal elements of the multi-dimensional constraint condition set to improve the numerical stability. Assume that the constraint matrix is , and the constraint vector is , and its form is: ;

[0118] The standardization is to normalize the main diagonal elements of the matrix , divide each row by the corresponding diagonal element to obtain the standardized matrix and the standardized constraint vector : ;

[0119] This processing ensures that the constraint conditions have consistent dimensions, facilitating subsequent operations of the solver. Initialize the linear programming of the standardized constraint matrix, and set the initial basic feasible solution and the upper and lower bounds of the non-basic variables of the simplex method. The initial basic feasible solution is defined as the starting point that satisfies the constraint conditions , and its form is: ;

[0120] The upper and lower bounds of the non-basic variables are defined according to the constraint conditions as: ;

[0121] where and are the lower and upper bounds of the variable respectively. By initializing these parameters, initial solution parameters are generated, providing initial conditions for the optimization of the subsequent branch and bound method. Input the initial solution parameters into the branch and bound solution control module, configure the node selection rule and the branch variable selection strategy, and generate the branch strategy parameters. The node selection rule adopts the depth-first search or breadth-first search strategy, and the branch variable selection is based on the priority selection of the non-integer degree. The non-integer degree is defined as: ;

[0122] Selecting the variable with the largest non-integer degree as the branch variable helps to quickly approach the integer solution. Based on the branch strategy parameters, construct the integer variable processing rule. Set the gap tolerance value and the rounding direction judgment condition. The gap tolerance value controls the accuracy of the solution: ;

[0123] The rounding direction is judged according to the gradient direction of the objective function to ensure that the optimization direction is towards a better solution. Set the convergence judgment rule for the integer programming parameters, and construct the threshold conditions for the relative improvement rate of the objective function and the constraint satisfaction degree. The relative improvement rate of the objective function is defined as: ;

[0124] where and are the objective function values of the new solution and the old solution respectively, is the convergence threshold. The constraint satisfaction degree is expressed as: ;

[0125] where is the constraint tolerance. Input the convergence control parameters into the solution process monitoring module, set the maximum number of iteration steps and the upper limit of the calculation time , and generate the termination criterion parameters. The termination condition is: Stop if: or Iteration or Time ;

[0126] Combine the integer programming parameters, convergence control parameters, and termination criterion parameters, configure the LINGO solver, and generate the final optimized solution parameters.

[0127] In a specific embodiment, the process of inputting the optimized solution parameters into the optimization calculation layer of the P - center problem network optimization model and using the branch - and - bound method to solve the 0 - 1 integer programming problem to generate the river network dredging plan may specifically include the following steps:

[0128] Perform relaxation processing on the standardized constraint matrix in the optimized solution parameters, and convert the 0 - 1 integer constraint into a continuous variable interval constraint to obtain a linear programming relaxation problem;

[0129] Input the linear programming relaxation problem into the simplex calculation module, iteratively solve to obtain the initial optimal solution, obtain the node upper - bound value data, construct a branch variable selection rule based on the node upper - bound value data, and calculate the non - integer degree and influence coefficient of the decision variable to obtain a branch sequence;

[0130] Input the branch sequence into the branch - and - bound tree construction module, generate sub - problems and solve the optimal solutions of the sub - problems to obtain an active node set, update the bound values for the active node set, and calculate the difference between the objective function value of each active node and the current optimal value to obtain node evaluation data;

[0131] Input the node evaluation data into the node traversal module, select the next node to be processed using the optimal depth - first search strategy to obtain a candidate node set, perform a feasibility test based on the candidate node set, and verify the integer property and constraint satisfaction degree of the node solution to obtain a feasible solution set;

[0132] Generate a dredging plan for the feasible solution set, map the decision variables to the dredging operation plan of the river channel nodes, and generate a river network dredging plan.

[0133] Specifically, perform relaxation processing on the standardized constraint matrix in the optimized solution parameters, convert the 0 - 1 integer constraint into a continuous variable interval constraint, so as to relax the original integer programming problem into a linear programming problem for initial solution. Assume that the original constraint matrix is , the constraint vector is , the decision variable is , and its constraint form is: ;

[0134] The relaxation processing relaxes the integer constraint and expands the variable definition to a continuous interval: ;

[0135] The relaxed problem is transformed into a linear programming relaxation problem: ;

[0136] where, is the objective function coefficient vector. The relaxed problem is input into the simplex calculation module, and the simplex method is used for iterative solution to find the initial optimal solution. The simplex method starts from an initial basic feasible solution, searches along the direction of the constraint boundary, and gradually improves the objective function value. The initial feasible solution satisfies: ;

[0137] and through iteration, the entering basic variable and the leaving basic variable are selected, and finally an optimal solution that satisfies all constraints is found . Through this solution process, the node upper bound value data is obtained, and the value of each variable represents the optimal value in the relaxed problem. Based on the node upper bound value data, a branching variable selection rule is constructed, and the non-integer degree of each decision variable is calculated, defined as: ;

[0138] where, and are the floor value and the ceiling value of respectively. The larger the non-integer degree, the higher the degree of deviation of the variable from the integer solution, and this variable should be preferentially selected for branching. The influence coefficient of each variable is calculated, defined as: ;

[0139] where is the coefficient of the variable in the objective function. According to the non-integer degree and the influence coefficient, a branching sequence is constructed, and the variable with the greatest influence is preferentially selected for branching. The branching sequence is input into the branch and bound tree construction module, and based on the selected branching variable, two sub-problems are generated: one sub-problem constraint , and the other sub-problem constraint . By solving the optimal solutions of each sub-problem, an active node set is obtained, and each active node represents a potential solution. The bound value of the active node set is updated, and the upper and lower bounds are updated according to the optimal solution of the sub-problem: ;

[0140] where, is the current optimal solution, and are the minimum and maximum objective function values of the sub-problem respectively. The difference between the objective function value and the current optimal value is calculated for each active node to obtain the node evaluation data: ;

[0141] where is a node of the objective function value. According to the node evaluation data, input it into the node traversal module, and use the optimal depth-first search strategy to select the next node to be processed. In the node traversal module, based on the candidate node set, perform a feasibility test to verify whether the solution satisfies the integer property and constraint conditions. If the solution of a certain node satisfies all constraints and , then this solution is classified into the feasible solution set. Generate a dredging plan for the feasible solution set, and map the final integer solution to the dredging operation plan of the river channel nodes. For example, assume that a river network has 5 nodes, and its optimal solution is , indicating that nodes 1, 3, and 4 are dredged, while nodes 2 and 5 are not dredged. The dredging method is mapped to specific operations through the auxiliary variable , such as "mechanical dredging", "flushing dredging".

[0142] The above describes the decision-making method for the river channel dredging plan based on sediment elution in the embodiments of the present invention. Next, the decision-making device for the river channel dredging plan based on sediment elution in the embodiments of the present invention will be described. Please refer to Figure 2 , an embodiment of the decision-making device for the river channel dredging plan based on sediment elution in the embodiments of the present invention includes:

[0143] An acquisition module, configured to collect and perform standardized preprocessing on the river channel spatial distribution, cross-section morphology, hydrological parameters, and sediment pollution characteristics of the river network system to obtain a river channel network feature data set;

[0144] An analysis module, configured to perform sediment pollutant migration and diffusion analysis based on the river channel network feature data set, and calculate the pollution propagation relationship between nodes to obtain a river channel sediment pollution propagation network model;

[0145] A calculation module, configured to calculate the coupling degree of river channel sediment pollution based on the river channel sediment pollution propagation network model to obtain river channel sediment elution risk distribution data;

[0146] A generation module is used to input the elution risk distribution data of riverbed sediment into the P - center problem network optimization model. Through the analysis of minimizing pollution diffusion risk and maximizing the utilization efficiency of dredging resources, a river network dredging plan is generated. Through the collaborative cooperation of the above - mentioned various components, the present invention accurately describes the migration and diffusion law of sediment pollutants by constructing a riverbed sediment pollution propagation network model and combining time - series analysis and causal test methods, providing scientific data support for the decision - making of dredging plans. The present invention uses a multi - dimensional evaluation index system to quantitatively calculate the coupling degree of riverbed sediment pollution, and identifies risk hot - spot areas through spatial statistical analysis methods, effectively improving the accuracy of risk assessment. Based on the P - center problem network optimization theory, the present invention constructs an optimization model with the dual objectives of minimizing pollution diffusion risk and maximizing the utilization efficiency of dredging resources, solving the problem of repeated resource investment in traditional dredging plans. The present invention uses the branch - and - bound method to solve the 0 - 1 integer programming problem, and ensures the feasibility and optimality of the dredging plan through strict constraint conditions and optimization algorithms. During the optimization solution process, the present invention adopts objective balance analysis and constraint matrix standardization processing to establish a complete solution parameter configuration system, ensuring the convergence and stability of the optimization results. The present invention incorporates multiple factors such as the integrity of the river network system, the characteristics of pollutant propagation, and the utilization efficiency of resources into the decision - making process, forming a systematic and standardized decision - making method for dredging plans, improving the scientificity and efficiency of river regulation.

[0147] Referring to Figure 3 , an embodiment of the present invention also provides a computer device. This computer device can be a server, and its internal structure can be as Figure 3 shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non - volatile storage medium and an internal memory. The non - volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non - volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above - mentioned method is implemented.

[0148] Those skilled in the art can understand that Figure 3 the structure shown in

[0149] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0150] Those of ordinary skill in the art can understand that all or part of the processes in the above method embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above method embodiments. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0151] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, system, and unit can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0152] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0153] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A decision-making method for river dredging scheme based on sediment elution, characterized in that, The method includes: Collecting and preprocessing data on the channel spatial distribution, cross-section morphology, hydrological parameters, and sediment pollution characteristics of the river network system to obtain a river network feature dataset; Performing sediment pollutant migration and diffusion analysis based on the river network feature dataset, and calculating the pollution propagation relationship between nodes to obtain a river sediment pollution propagation network model; Calculating the coupling degree of river sediment pollution based on the river sediment pollution propagation network model to obtain river sediment elution risk distribution data; Inputting the river sediment elution risk distribution data into the P-center problem network optimization model, and generating a river network dredging plan through the analysis of minimizing pollution diffusion risk and maximizing the utilization efficiency of dredging resources; specifically including: inputting the river sediment elution risk distribution data into the data preprocessing layer of the P-center problem network optimization model, performing standardization processing on the risk level, spatial location, and dredging resource parameters to obtain basic data for optimization calculation; inputting the basic data for optimization calculation into the constraint condition construction layer of the P-center problem network optimization model, and obtaining a multi-dimensional constraint condition set by setting the limiting conditions of hydrological demand constraints, river channel connectivity constraints, and resource input constraints; constructing dredging location selection variables for the multi-dimensional constraint condition set, and obtaining a decision variable matrix by setting the 0-1 decision variable for the dredging operation and the dredging method selection variable for each river channel node; inputting the decision variable matrix into the objective function construction layer of the P-center problem network optimization model, and obtaining a dual-objective optimization function by constructing a sub-objective function for minimizing pollution diffusion risk and a sub-objective function for maximizing resource utilization efficiency; configuring the LINGO solver for the dual-objective optimization function and the multi-dimensional constraint condition set, and obtaining optimization solution parameters by setting the solution accuracy, iteration times, and convergence conditions; inputting the optimization solution parameters into the optimization calculation layer of the P-center problem network optimization model, and using the branch and bound method to solve the 0-1 integer programming problem to generate a river network dredging plan.

2. The decision-making method for river dredging plan based on sediment elution according to claim 1, wherein The collecting and preprocessing data on the channel spatial distribution, cross-section morphology, hydrological parameters, and sediment pollution characteristics of the river network system to obtain a river network feature dataset includes: Performing geographical information system mapping on the river network system, collecting spatial coordinate data including longitude, latitude, and elevation to obtain original data on river channel spatial distribution; Inputting the original data on river channel spatial distribution into a topological analysis model, and obtaining river network topological structure data through calculations of node connectivity, path connectivity, and network density; Collecting morphological parameter data for the river channel cross-sections in the river network topological structure data, and measuring parameters including river channel width, water depth, cross-sectional area, and longitudinal slope to obtain river channel morphological feature data; Setting hydrological monitoring points based on the river channel morphological feature data, and collecting hydrological elements including cross-sectional velocity, flow rate, water level, and sediment concentration to obtain river channel hydrological feature data; Taking the river channel hydrological characteristic data as the sampling basis, stratified sampling and pollutant detection are carried out on the river channel sediment to obtain sediment pollutant characteristic data, and the sediment pollutant characteristic data includes pollutant types, concentration distribution and spatial variability; According to the sediment pollutant characteristic data, a Kriging spatial interpolation model is established, and through variogram calculation and cross-validation, sediment pollutant spatial distribution data is obtained; Normalize the river network topological structure data, the river channel morphological characteristic data, the river channel hydrological characteristic data and the sediment pollutant spatial distribution data to obtain standardized river channel characteristic data; Input the standardized river channel characteristic data into a density-based clustering algorithm model, and by calculating the Euclidean distance and local density of the river channel characteristic vectors, a river channel network characteristic data set is obtained.

3. The decision-making method for river dredging plan based on sediment elution according to claim 2, characterized in that, Based on the river channel network characteristic data set, sediment pollutant migration and diffusion analysis is carried out, and the pollution propagation relationship between nodes is calculated to obtain a river channel sediment pollution propagation network model, including: Perform time series discretization processing on the hydrological characteristic data in the river channel network characteristic data set, and calculate the pollutant concentration change rate and water flow velocity field at each time point to obtain basic pollutant migration data; Input the basic pollutant migration data and the river channel morphological characteristic data into an autoregressive model, and by calculating the autocorrelation coefficient and partial autocorrelation coefficient of the pollutant concentration, a time-dependent characteristic sequence is obtained; Perform multi-scale decomposition on the time-dependent characteristic sequence to obtain pollutant migration pattern data, and input the pollutant migration pattern data into a Granger causality test model to calculate the F statistic and P value of the pollutant concentration to obtain a pollution propagation probability matrix between nodes; Perform double-region logarithmic distribution fitting on the pollution propagation probability matrix between nodes to obtain network connection strength data; Perform community division on the network connection strength data to obtain pollution propagation key channel data, and calculate the global efficiency and local efficiency based on the pollution propagation key channel data to obtain a network propagation ability index; Construct a river channel sediment pollution propagation network model based on the pollution propagation probability matrix between nodes, the network connection strength data and the network propagation ability index.

4. The decision-making method for the river dredging plan based on sediment elution according to claim 3, wherein Based on the river channel sediment pollution propagation network model, calculate the coupling degree of river channel sediment pollution to obtain river channel sediment elution risk distribution data, including: Classify the network propagation ability index in the river channel sediment pollution propagation network model to obtain pollution assessment index data; Based on the pollution assessment index data, calculate the co-variation coefficient between the water quality safety index and the ecological risk index to obtain a pollution coupling degree matrix, and analyze the sediment pollutant release period and change intensity of the pollution coupling degree matrix to obtain pollution diffusion dynamic data; Set water quality safety thresholds, ecological risk thresholds and governance response thresholds, and perform risk classification on the pollution diffusion dynamic data to obtain node risk score data; Calculate the local Moran index and global Moran index on the node risk score data to obtain risk aggregation characteristic data, and perform kernel density analysis on the risk aggregation characteristic data to obtain risk hotspot distribution data; Based on the risk hot spot distribution data, risk areas are identified to obtain the risk zoning results. According to the risk zoning results, the pollution intensity index and spatial diffusion index of each risk area are calculated to obtain the risk distribution data of river bottom sediment elution.

5. The decision-making method for river dredging plan based on sediment elution according to claim 1, characterized in that, Configure the LINGO solver for the dual-objective optimization function and the multi-dimensional constraint set. By setting the solution accuracy, number of iterations, and convergence conditions, the optimized solution parameters are obtained, including: Conduct target balance analysis on the dual-objective optimization function. By constructing a normalized weighted expression of the pollution diffusion risk weight coefficient and the resource utilization efficiency weight coefficient, the target balance parameter is obtained; Input the target balance parameter into the constraint matrix normalization module, and process the multi-dimensional constraint set based on the main diagonal element normalization method to obtain the normalized constraint matrix; Initialize the linear programming of the normalized constraint matrix, and set the initial basic feasible solution of the simplex method and the upper and lower bounds of the non-basic variables to obtain the initial solution parameters. Input the initial solution parameters into the branch and bound solution control module, and configure the node selection rule and the branch variable selection strategy to obtain the branch strategy parameters; Construct an integer variable processing rule based on the branch strategy parameters, and set the gap tolerance value and the rounding direction judgment condition of the integer variable to obtain the integer programming parameters. Set the convergence judgment rule for the integer programming parameters, and construct the threshold conditions of the relative improvement rate of the objective function and the constraint satisfaction degree to obtain the convergence control parameters; Input the convergence control parameters into the solution process monitoring module, and set the maximum number of iteration steps and the upper limit of the calculation time to obtain the termination criterion parameters. Configure the LINGO solver parameters based on the integer programming parameters, the convergence control parameters, and the termination criterion parameters to obtain the optimized solution parameters.

6. The decision-making method for river dredging plan based on sediment elution according to claim 5, characterized in that Input the optimized solution parameters into the optimization calculation layer of the P-center problem network optimization model, and use the branch and bound method to solve the 0-1 integer programming problem to generate a river network dredging plan, including: Relax the normalized constraint matrix in the optimized solution parameters, and convert the 0-1 integer constraint into a continuous variable interval constraint to obtain a linear programming relaxation problem; Input the linear programming relaxation problem into the simplex calculation module, and iteratively solve it to obtain the initial optimal solution to obtain the node upper bound value data. Based on the node upper bound value data, construct a branch variable selection rule, and calculate the non-integer degree and influence coefficient of the decision variable to obtain the branch sequence; Input the branch sequence into the branch and bound tree construction module, and generate sub-problems and solve the optimal solutions of the sub-problems to obtain the active node set. Update the bound values of the active node set, and calculate the difference between the objective function value of each active node and the current optimal value to obtain the node evaluation data; Input the node evaluation data into the node traversal module, and select the next node to be processed according to the optimal depth-first search strategy to obtain the candidate node set. Conduct a feasibility test based on the candidate node set, and verify the integer characteristics and constraint satisfaction degree of the node solution to obtain the feasible solution set; Generate a dredging plan for the feasible solution set, and map the decision variables to the dredging operation plan of the river channel nodes to generate a river network dredging plan.

7. A decision-making device for river dredging plan based on sediment elution, characterized in that, For implementing the decision-making method for the river channel dredging plan based on sediment elution as described in any one of claims 1-6, the decision-making device for the river channel dredging plan based on sediment elution includes: A collection module, configured to collect and perform standardized preprocessing on the river channel spatial distribution, cross-section morphology, hydrological parameters, and sediment pollution characteristics of the river network system to obtain a river channel network feature data set; An analysis module, configured to perform sediment pollutant migration and diffusion analysis based on the river channel network feature data set, and calculate the pollution propagation relationship between nodes to obtain a river channel sediment pollution propagation network model; A calculation module, configured to calculate the coupling degree of river channel sediment pollution based on the river channel sediment pollution propagation network model to obtain river channel sediment elution risk distribution data; A generation module, configured to input the river channel sediment elution risk distribution data into a P-center problem network optimization model, and generate a river network dredging plan through the analysis of minimizing pollution diffusion risk and maximizing the utilization efficiency of dredging resources.

8. A computer device, characterized in that, It includes a memory and a processor, and the memory stores a computer program that can run on the processor. It is characterized in that when the processor executes the computer program, it implements the decision-making method for the river channel dredging plan based on sediment elution as described in any one of claims 1 to 6.

9. A computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, the processor is caused to execute the decision-making method for the river channel dredging plan based on sediment elution as described in any one of claims 1 to 6.

Images