River channel dredging scheme decision-making method, device and equipment based on bottom mud elution

Abstract

The invention relates to the technical field of scheme decision, and discloses a river channel clearing scheme decision method, device and equipment based on bottom mud elution, and the method comprises the steps: carrying out the data collection and standardization preprocessing of the river channel space distribution, section morphology, hydrological parameters and bottom mud pollution characteristics of a river network system, and obtaining a river channel network characteristic data set; carrying out sediment pollutant migration and diffusion analysis, and calculating a pollution propagation relationship between nodes to obtain a river sediment pollution propagation network model; carrying out river sediment pollution coupling degree calculation to obtain river sediment elution risk distribution data; and inputting the river sediment elution risk distribution data into the P center problem network optimization model, and generating a river network dredging scheme through pollution diffusion risk minimization and dredging resource utilization efficiency maximization analysis. According to the invention, a plurality of factors of the river network system are incorporated into the decision-making process, a systematized and normalized clearing scheme decision-making method is formed, and the scientificity and efficiency of river regulation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of scheme decision-making, and in particular to a method, device and equipment for making a river channel dredging scheme decision based on sediment washing. Background Art

[0002] As an important part of the water environment, the long-term enrichment and resuspension of pollutants in river sediments seriously affect the quality of the water ecological environment and the safety of river traffic. Traditional river dredging plans mainly rely on expert experience and judgment, lack of 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] As the task of river network management becomes increasingly complex, the diffusion of pollution between rivers shows a 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 schemes generally have problems such as repeated resource investment and unsatisfactory dredging effects, and lack comprehensive consideration of the integrity of the river network system and the characteristics of pollutant propagation. At present, in the decision-making process of river dredging, the assessment method of sediment elution risk is relatively simple, and it is difficult to effectively identify high-risk areas and key control sections. At the same time, the optimal allocation of dredging resources and the scientific decision-making of the plan lack quantitative evaluation standards and systematic optimization models, which makes it difficult to guarantee the implementation effect of the dredging plan. Summary of the invention

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

[0005] In a first aspect, the present invention provides a method for deciding a river channel dredging scheme based on sediment elution, the method for deciding a river channel dredging scheme based on sediment elution comprising: Data collection and standardized preprocessing of the spatial distribution, cross-sectional morphology, hydrological parameters and sediment pollution characteristics of the river network system were performed to obtain a river network characteristic data set; Based on the river network characteristic data set, the migration and diffusion analysis of sediment pollutants is performed, and the pollution propagation relationship between nodes is calculated to obtain a river sediment pollution propagation network model; Calculate the coupling degree of river sediment pollution based on the river sediment pollution propagation network model to obtain river sediment elution risk distribution data; The riverbed sediment elution risk distribution data is input into the P-center problem network optimization model, and a river network dredging plan is generated by minimizing the pollution diffusion risk and maximizing the dredging resource utilization efficiency.

[0006] In a second aspect, the present invention provides a river channel dredging scheme decision device based on sediment elution, the river channel dredging scheme decision device based on sediment elution comprising: The acquisition module is used to collect and standardize the data of the spatial distribution, cross-sectional morphology, hydrological parameters and sediment pollution characteristics of the river network system to obtain the river network characteristic data set; An analysis module is used to perform sediment pollutant migration and diffusion analysis based on the river network characteristic data set, and calculate the pollution propagation relationship between nodes to obtain a river sediment pollution propagation network model; A calculation module, used to calculate the coupling degree of river sediment pollution based on the river sediment pollution propagation network model to obtain river sediment elution risk distribution data; A generation module is used to input the riverbed sediment elution risk distribution data into the P-center problem network optimization model, and generate a river network dredging plan by minimizing the pollution diffusion risk and maximizing the dredging resource utilization efficiency.

[0007] The third aspect of the present invention provides a computer device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the computer device executes the above-mentioned river dredging plan decision-making method based on sediment elution.

[0008] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned river dredging plan decision-making method based on sediment elution.

[0009] In the technical solution provided by the present invention, the present invention constructs a riverbed sediment pollution propagation network model, combines time series analysis and causal test methods, accurately describes the migration and diffusion laws of sediment pollutants, and provides scientific data support for dredging plan decision-making. The present invention adopts a multi-dimensional evaluation index system to quantitatively calculate the coupling degree of riverbed sediment pollution, identifies risk hotspots through spatial statistical analysis methods, and effectively improves the accuracy of risk assessment. Based on the P-center problem network optimization theory, the present invention constructs an optimization model with dual objectives of minimizing the risk of pollution diffusion and maximizing the efficiency of dredging resource utilization, solving the problem of repeated resource investment in traditional dredging plans. The present invention adopts 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 constraints and optimization algorithms. In the process of optimization solution, the present invention adopts target balance analysis and constraint matrix standardization processing to establish a complete solution parameter configuration system to ensure 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, thereby improving the scientificity and efficiency of river management. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0011] Figure 1 A schematic diagram of the steps of a river dredging solution decision method based on sediment elution in an embodiment of the present invention; Figure 2 It is a schematic diagram of the structure of a river dredging solution decision-making device based on sediment elution in an embodiment of the present invention; Figure 3 It is a schematic block diagram of the structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0012] Embodiments of the present invention provide a method, device and equipment for making decisions on river dredging schemes based on sediment elution. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0013] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 In one embodiment of the present invention, a method for deciding a river channel dredging solution based on sediment washing includes: Step S1, collecting and standardizing data on the spatial distribution, cross-sectional morphology, hydrological parameters and sediment pollution characteristics of the river network system to obtain a river network characteristic data set; It is understandable that the execution subject of the present invention may be a river channel dredging solution decision device based on sediment washing, or a terminal or a server, which is not limited here. The embodiment of the present invention is described by taking the server as the execution subject as an example.

[0014] Specifically, the river network system is mapped using a geographic information system (GIS) to collect spatial coordinate data including longitude, latitude and elevation to describe the spatial distribution of the river. The original data of the spatial distribution of the river is generated through GIS technology. The original data of the spatial distribution of the river is input into the topological analysis model, and the topological structure data of the river network is generated by calculating key parameters such as node connectivity, path connectivity and network density. The topological structure data of the river network is the core of describing the complexity of the river network, reflecting the connection relationship between the rivers and the connectivity of the overall network. Based on the river network topological structure data, the morphological parameters of the river section are collected. The morphological parameters such as river width, water depth, cross-sectional area and longitudinal slope are measured. These data are obtained through various means such as field surveys, laser scanning technology or drone aerial survey technology to obtain river morphological characteristic data. Based on the river morphological characteristic data, hydrological monitoring points are set to collect key hydrological elements including cross-sectional flow velocity, flow, water level and sediment content. These hydrological characteristic data are obtained through flow meters, automatic monitoring equipment or remote sensing technology to dynamically reflect the hydrological conditions of the river. The river hydrological characteristic data is used as the sampling basis to carry out stratified sampling and pollutant detection of river sediment. The sampling process is based on the standard stratification method to ensure that sediment samples at different depths can fully reflect the distribution of pollutants. The types of pollutants, concentration distribution and spatial variability in the sediment are detected by laboratory analysis to obtain complete sediment pollutant characteristic data. In order to convert the sediment pollutant characteristic data into a spatial distribution model, a spatial distribution map of pollutants is established based on the Kriging spatial interpolation model. The model optimizes the interpolation results through variogram calculation and cross-validation, generates spatial distribution data of sediment pollutants, displays the distribution pattern of pollutants in the river network system, and helps identify pollution hotspots and pollution gradients. The river network topological structure data, river morphological characteristic data, river hydrological characteristic data and sediment pollutant spatial distribution data are normalized. Standardization methods such as Min-Max normalization or Z-score normalization are used 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 characteristic data. The standardized river channel characteristic data is input into the density-based clustering algorithm model, and the network characteristic data set of the river channel is extracted by calculating the Euclidean distance and local density of the river channel characteristic vector. In this process, the density clustering algorithm (such as DBSCAN) can effectively identify high-density areas and river sections with similar characteristics in the river network, and generate a river channel network characteristic data set containing rich information.

[0015] Step S2: Based on the river network characteristic data set, the migration and diffusion analysis of sediment pollutants is performed, and the pollution propagation relationship between nodes is calculated to obtain a river sediment pollution propagation network model; Specifically, the hydrological characteristic data are extracted from the river network characteristic data set, and the time series discretization processing is performed on them. The continuous hydrological data are discretized into the pollutant concentration change rate and water flow velocity field at multiple time points, so that the data can reflect the migration state of pollutants at different time points and obtain the basic data of pollutant migration. The pollutant migration basic data and the river morphological characteristic data are input into the autoregressive model, and the regularity of pollutant concentration changes over time is explored by calculating the autocorrelation coefficient and partial autocorrelation coefficient of pollutant concentration. The autoregressive model can capture the inherent trend in the time series, reveal the time-dependent characteristics in the process of pollutant migration, and generate a dependent characteristic sequence with time series characteristics. In order to analyze the complex behavior of pollutant migration, the time-dependent characteristic sequence is decomposed at multiple scales to extract pollutant migration pattern data at different scales. These data reveal the migration law of pollutants at micro and macro time scales and provide more comprehensive dynamic characteristic information. The pollutant migration pattern data are input into the Granger causality test model, and the causal effect of pollutant concentration is analyzed using Granger causality. By calculating the F statistic and P value of concentration change, the pollution propagation probability matrix between nodes is generated. This matrix represents the probability of pollutant transmission between any two nodes in the network, and quantifies the causal relationship of pollutant migration from a statistical perspective. The pollution transmission probability matrix is ​​fitted with a two-zone logarithmic distribution to more accurately characterize the strength characteristics of network connections. The two-zone 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 transmission network. Based on the fitting results, network connection strength data are generated to quantify the pollutant transmission capacity and interaction intensity between nodes in the network. The network connection strength data is divided into communities. Through community division, the key pollution transmission channels in the network are identified. These channels are the main paths for pollutant migration and are also the parts with 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 transmission capacity and local response capacity of the pollution transmission network. The combination of these indicators can reflect the transmission capacity index of the network. Combining the pollution transmission probability matrix between nodes, the network connection strength data and the network transmission capacity index, a river sediment pollution transmission 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 transmission bottlenecks in the network.

[0016] Step S3, 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; Specifically, the network propagation capacity index in the river sediment pollution propagation network model is graded and divided into different levels to generate pollution assessment index data. The differences in pollution levels are refined so that it can accurately reflect the propagation characteristics and strength changes of river sediment pollution, forming a clear basis for pollution assessment. Based on the pollution assessment index data, the synergistic 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 synergistic variation coefficient is achieved by constructing a pollution coupling matrix, in which each matrix unit represents the coupling intensity of water quality safety and ecological risk in different regions. Through this matrix, the comprehensive impact of pollutants on water quality and ecological environment is revealed. The release cycle and change intensity of sediment pollutants are analyzed on the pollution coupling matrix, the dynamic law of pollution diffusion is explored, and dynamic data of pollution diffusion is generated. By setting the water quality safety threshold, ecological risk threshold and governance response threshold, the dynamic data of pollution diffusion is analyzed by risk grading. Each risk level corresponds to a specific environmental impact range and governance needs, and node risk score data is generated. The local Moran index and the global Moran index are calculated for the node risk score data to reveal the spatial aggregation characteristics of risk distribution. The local Moran index is used to identify high-risk hotspots in local areas, while the global Moran index provides information on the overall spatial distribution pattern of risks. By analyzing the Moran index, the concentration or discrete state of risks in the river network system is clarified, and risk aggregation characteristic data is formed. On this basis, the kernel density analysis is applied to process the risk aggregation characteristic data to determine the spatial distribution trend and density change of high-risk areas and generate risk hotspot distribution data. Based on the risk hotspot distribution data, risk areas are identified to determine the specific scope of each high-risk area. The identification process combines geographic information systems and spatial analysis methods to display 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 region 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. By combining the above analysis process, the river sediment elution risk distribution data is generated.

[0017] Step S4, input the riverbed sediment elution risk distribution data into the P-centered problem network optimization model, and generate a river network dredging plan by minimizing the risk of pollution diffusion and maximizing the efficiency of dredging resource utilization. Specifically, the riverbed sediment elution risk distribution data is input into the data preprocessing layer of the P-centered problem network optimization model for standardization. The risk level, spatial position and dredging resource parameters are normalized 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 standardized optimization calculation basic data, the constraint condition layer of the P-centered problem network optimization model is constructed. A multidimensional constraint condition set is set, including specific restrictions such as hydrological demand constraints, river connectivity constraints and resource input constraints. The hydrological demand constraint ensures that the dredging scheme can meet the basic requirements of the river hydrological balance and avoid hydrological damage caused by dredging; the river connectivity constraint ensures 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 optimization scheme can be implemented under actual conditions. These constraints together constitute the basic rules of optimization calculation and constrain the search space of potential solutions. The dredging location selection variable is constructed for the multidimensional constraint set. A 0-1 decision variable for the dredging operation is defined for each river node, indicating whether the node is selected for dredging operation, and a selection variable for the dredging method is defined to indicate the specific dredging technology or method used 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. The decision variable matrix is ​​input into the objective function construction layer of the P-center problem network optimization model to establish a dual-objective optimization function of minimizing the risk of pollution diffusion and maximizing resource utilization efficiency. The pollution diffusion risk minimization sub-objective function aims to prioritize the dredging of high-risk nodes and reduce the possibility of further spread of pollutants; the resource utilization efficiency maximization sub-objective function optimizes the spatial and technical allocation of resources to ensure the maximum governance effect with the minimum investment. These two sub-objective functions are integrated into a dual-objective optimization problem in the optimization model, reflecting the scientific nature and practical feasibility of the dredging scheme. In order to solve the dual-objective optimization problem, the LINGO solver is configured to determine the optimization solution parameters by setting the solution accuracy, number of iterations and convergence conditions. The 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 falling into the local optimum or long-term non-convergence problem. The optimization solution parameters are input into the optimization calculation layer of the P-center problem network optimization model, and the branch and bound method is used 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 of each branch, the algorithm can quickly eliminate infeasible or non-optimal solutions and ultimately generate a globally optimal river network dredging solution.

[0018] The objective balance analysis of the dual-objective optimization function is carried out. By constructing the normalized weighted expressions of the pollution diffusion risk weight coefficient and the resource utilization efficiency weight coefficient, the weight distribution relationship between the two objectives is adjusted to ensure that the objective function can balance the pollution control priority and resource allocation efficiency, and generate the objective balance parameters. The objective balance parameters are input into the constraint matrix standardization module, and the multidimensional constraint condition set is processed by the main diagonal element normalization method. The core of this normalization method is to ensure that the numerical range of the constraint conditions is consistent, so as to avoid the adverse effect of scale differences on the optimization results. The processed standardized 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. The standardized constraint matrix is ​​initialized for linear programming. The simplex method is used to initialize the basic feasible solution, and the upper and lower bounds are set for non-basic variables. These initial solution parameters are an important basis for the optimization process, which determines the initial search space of the branch and bound method and the feasibility of the solution. By optimizing the initial solution parameters, the convergence speed of the optimization model is significantly accelerated and the accuracy of the solution is improved. The initial solution parameters are input into the branch and bound solution control module to configure the node selection rules and branch variable selection strategy. The node selection rule determines the nodes that are preferentially expanded during the optimization process, while the branch variable selection strategy is used to determine the splitting priority of the branch variables. The two together constitute the branch strategy parameters, which directly affect the efficiency and path of the optimization solution. The integer variable processing rules are constructed based on the branch strategy parameters. The solution of integer programming requires setting the gap tolerance value of the integer variable to control the accuracy of the solution, and the value range of the integer variable is dynamically adjusted by the rounding direction judgment condition. Through the reasonable setting of integer programming parameters, it is ensured that the optimization model takes into account both computational efficiency and solution accuracy when processing discrete decision variables. The convergence judgment rules are set for the integer programming parameters. By constructing the threshold conditions of the relative improvement rate of the objective function and the constraint satisfaction, the convergence of the optimization process is effectively controlled. When the relative improvement rate of the objective function is lower than the set threshold and all constraints are satisfied, the optimization process is terminated, which improves the stability and reliability of the model solution. The setting of the convergence control parameters provides a clear termination condition for the solution process to prevent unnecessary iterations from extending the solution time. The convergence control parameters are input into the solution process monitoring module, and the maximum number of iterations and the upper limit of the calculation time are set to form the termination criterion parameters. These parameters ensure that the LINGO solver can complete the optimization task within the specified time and computing resources, thereby improving the solution efficiency and avoiding the dilemma of long-term calculation. By combining integer programming parameters, convergence control parameters and termination criterion parameters, the LINGO solver is fully configured and finally the optimized solution parameters are obtained.

[0019] The standardized constraint matrix in the optimization solution parameters is relaxed, and the original 0-1 integer constraints are converted into interval constraints of continuous variables. By relaxing the restrictions on decision variables, the problem is converted into a relaxed linear programming problem, providing good computing conditions for the rapid solution of the initial solution. The relaxed linear programming problem is input into the simplex calculation module, and the initial optimal solution is obtained through iterative solution. In this process, the simplex method quickly determines the optimal solution in an efficient path search manner, and the generated node upper bound data can be used as the basis for subsequent branch variable selection. Based on the node upper bound data, a branch variable selection rule is constructed. The optimal branch variable is determined by calculating the non-integer degree and influence coefficient of each decision variable. This branch variable selection rule can give priority to variables with large non-integer degrees, thereby effectively narrowing the search space of the solution, improving the solution efficiency of the algorithm, and generating a clear branch sequence. The generated branch sequence is input into the branch and bound tree construction module, and subproblems are generated by dynamically splitting nodes and each subproblem is solved to determine its optimal solution. The solution of each subproblem 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, the priority and potential value of the node are quantified to form node evaluation data. The node evaluation data is input into the node traversal module, and the next node to be processed is selected based on the optimal depth-first search strategy. The depth-first search strategy can quickly mine high-value nodes and narrow the search range of solutions by preferentially expanding the optimal solution path of the current branch, thereby optimizing the computational efficiency. A feasibility test is performed on each node in the candidate node set, including verifying the integer characteristics and constraint satisfaction of the solution to form a feasible solution set. The integer characteristic verification 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 multidimensional constraints. After obtaining the feasible solution set, the dredging plan generation stage is entered. By mapping the decision variables to the dredging operation plan of the river node, the specific allocation of the dredging task is completed. The dredging operation plan of each node includes key information such as whether to dredge, the dredging method selection, and the amount of resource input, forming a complete river network dredging plan.

[0020] In the embodiment of the present invention, the present invention accurately describes the migration and diffusion law of sediment pollutants by constructing a river sediment pollution propagation network model, combining time series analysis and causal test methods, and provides scientific data support for dredging plan decision-making. The present invention adopts a multi-dimensional evaluation index system to quantify the coupling degree of river sediment pollution, identifies risk hotspots through spatial statistical analysis methods, and effectively improves the accuracy of risk assessment. Based on the P-center problem network optimization theory, the present invention constructs an optimization model with dual objectives of minimizing pollution diffusion risk and maximizing dredging resource utilization efficiency, solving the problem of repeated resource investment in traditional dredging plans. The present invention adopts 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 constraints and optimization algorithms. In the optimization solution process, the present invention adopts target balance analysis and constraint matrix standardization processing, establishes a complete solution parameter configuration system, and ensures 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, and improving the scientificity and efficiency of river management.

[0021] In a specific embodiment, the process of executing step S1 may specifically include the following steps: Carry out GIS mapping of the river network system, collect spatial coordinate data including longitude, latitude and elevation, and obtain the original data of river spatial distribution; The original data of river spatial distribution are input into the topological analysis model, and the river network topological structure data are obtained by calculating the node connectivity, path connectivity and network density; The morphological parameters of the river sections in the river network topological structure data are collected, and the parameters including the width, water depth, cross-sectional area and longitudinal slope are measured to obtain the morphological characteristic data of the river; Hydrological monitoring points are set based on the river morphological characteristic data to collect hydrological elements including cross-sectional flow velocity, flow, water level and sediment content to obtain river hydrological characteristic data; Taking the river hydrological characteristic data as the sampling basis, the river sediment is layered sampled and pollutant detected to obtain sediment pollutant characteristic data, which includes pollutant types, concentration distribution and spatial variability; A Kriging spatial interpolation model was established based on the characteristic data of sediment pollutants, and the spatial distribution data of sediment pollutants were obtained through variogram calculation and cross-validation. Normalize the river network topological structure data, river channel morphological characteristic data, river channel hydrological characteristic data and sediment pollutant spatial distribution data to obtain standardized river channel characteristic data; The standardized river channel characteristic data are input into the density-based clustering algorithm model, and the river network characteristic data set is obtained by calculating the Euclidean distance and local density of the river channel characteristic vectors.

[0022] Specifically, the river network system is mapped using the Geographic Information System (GIS), and data including longitude and latitude are collected. ,latitude and elevation The spatial coordinate data including the river network. The original data of river spatial distribution is obtained through high-precision satellite remote sensing, drone mapping or ground measurement. Assume that the river network contains There are key nodes, and the coordinates of each node are recorded as ,in These data constitute the geometric skeleton of the river. The original data of the river spatial distribution are input into the topological analysis model to analyze the connection relationship between nodes, path connectivity and network density. Defined as a node Number of direct connections to other nodes: ; in, is an element of the adjacency matrix. If the node and Directly connected, ,otherwise Path connectivity ( ) quantifies the degree of connectivity between any two nodes in the network, and the calculation formula is: ; Among them, ShortestPath Representation Node and The shortest path length between them. Network density It is defined as the ratio of the actual number of connections to the total number of possible connections: ; Through calculation, the topological structure data of the river network is generated to describe the overall connectivity and complexity of the river system. Based on the topological structure data, the morphological parameters of the river section are collected. By measuring the width of the river , water depth , cross-sectional area and longitudinal slope ( ), and obtain the morphological characteristics of the river channel. The cross-sectional area is expressed as: ; The longitudinal slope is defined as the change in elevation per unit distance: ; Through drone surveys or underwater detection equipment, these parameters can be collected efficiently to provide support for the analysis of the geometric characteristics of the river. Hydrological monitoring points are set according to the river morphological characteristics data, and cross-sectional flow velocity data are collected. ,flow , Water level and sand content The flow calculation formula is: ; Hydrological data can dynamically reflect the hydrodynamic behavior of the river channel and provide input basis for modeling the migration and diffusion of pollutants. After obtaining the hydrological characteristic data, it is used as a basis for conducting sediment stratification sampling and pollutant detection, and analyzing the types of pollutants in the sediment ( ), concentration distribution and spatial variability The results of stratified sampling establish the distribution function , which represents the relationship between concentration and depth. The spatial variability of pollutant concentration is analyzed by calculating the variation function Finish: ; in, Indicates the spacing is The number of sampling pairs. Based on the 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 ​​by fitting the variation function. : ; in, is the interpolation weight, which is solved by the Kriging equations. The river network topology data, river channel morphology data, river channel hydrological characteristics data and sediment pollutant spatial distribution data are normalized. The normalization formula is: ; in, is the original data, and The normalized river channel characteristic data is input into the density-based clustering algorithm model (such as DBSCAN) and the Euclidean distance of the river channel characteristic vector is calculated. and the local density Complete cluster analysis. The calculation formula of Euclidean distance is: ; in, and Respectively represent and The sample eigenvalues, is the number of feature dimensions. The local density is defined as: ; in, is the indicator function, when The value is 1 when the pollution is significant, otherwise it is 0. Through cluster analysis, the river network characteristic dataset is obtained, revealing the areas with similar pollution characteristics and potential priority areas for dredging in the river network.

[0023] In a specific embodiment, the process of executing step S2 may specifically include the following steps: The hydrological characteristic data in the river network characteristic data set are discretized into time series, and the pollutant concentration change rate and water flow velocity field at each time point are calculated to obtain the basic data of pollutant migration; The basic data of pollutant migration and river morphological characteristics are input into the autoregressive model, and the time-dependent characteristic sequence is obtained by calculating the autocorrelation coefficient and partial autocorrelation coefficient of pollutant concentration. Perform multi-scale decomposition on the time-dependent characteristic sequence to obtain pollutant migration pattern data, and then input the pollutant migration pattern data into the Granger causality test model to calculate the F statistic and P value of the pollutant concentration and obtain the probability matrix of pollution transmission between nodes; The pollution propagation probability matrix between nodes is fitted with a two-region logarithmic distribution to obtain the network connection strength data; The network connection strength data is divided into communities to obtain the key channel data of pollution propagation, and the global efficiency and local efficiency are calculated based on the key channel data of pollution propagation to obtain the network propagation capacity index; A river sediment pollution propagation network model was constructed based on the pollution propagation probability matrix between nodes, network connection strength data and network propagation capacity index.

[0024] Specifically, the hydrological characteristic data in the river network characteristic data set is discretized into time series. Assuming the time range is , at each time point Collect hydrological characteristics, including pollutant concentrations and water velocity field The pollutant concentration change rate is defined as: ; in, is the time interval; and the water velocity field It is obtained by calculating the two-dimensional or three-dimensional flow field. , or It is expressed as: ; Based on these data, basic data describing the dynamic behavior of pollutant migration are formed. The basic data on pollutant migration are combined with the river morphological characteristics data (including width ,depth ,slope etc.) into the autoregressive model (AR). The AR model uses past concentration values ​​to predict current concentrations, and the model form is: ; in, is the regression coefficient, is the lag order, is a white noise term. By calculating the autocorrelation coefficient and partial autocorrelation coefficient , quantifies the time dependence of pollutant concentrations: ; The analysis results of autocorrelation and partial autocorrelation form a time-dependent characteristic sequence, revealing the temporal law of pollutant migration. The time-dependent characteristic sequence is decomposed into multiple scales to decompose the concentration sequence into change patterns at multiple scales. Assuming the original signal is , after discrete wavelet transform, the detail component is obtained and approximation component : ; in, Indicates the number of decomposition levels. The components at each scale reflect the different dynamic characteristics of pollutant migration, forming pollutant migration pattern data. The migration pattern data is input into the Granger causality test model, and the causal relationship of pollution transmission between nodes is analyzed by calculating the F statistic and P value of pollutant concentration. Assume that two nodes and The concentration series are and , the Granger test hypothesis is " No Granger leads ”, and its test model is: ; By F test statistic: ; Among them, RSS is the residual sum of squares, is the sample size, The significance of is determined by the P value. Generate the probability matrix of contamination propagation between nodes , each element of the matrix represents a node To Node The probability of contamination spreading. Perform a two-zone logarithmic distribution fitting and calculate the network connection strength data. Assuming the transmission strength Satisfies the logarithmic distribution: ; in, and are the fitting parameters, is the error term. The fitting results reveal the statistical characteristics of the transmission intensity. The network connection strength data is used to divide the community and identify the key channels for pollution transmission. The maximization method is used to divide the community: ; in, is the adjacency matrix, and is the degree of the node, is the total number of edges, is an indicator function, only when the node and If they belong to the same community, the value is 1. Based on the community division results, the global efficiency is calculated. and local efficiency , to quantify the network propagation capacity. The global efficiency is defined as: ; in, Is a node and The shortest path length between. The local efficiency is: ; Combining the propagation probability matrix, network connection strength and propagation capacity index , , construct a river sediment pollution propagation network model.

[0025] In a specific embodiment, the process of executing step S3 may specifically include the following steps: The network propagation capacity index in the river sediment pollution propagation network model is graded to obtain pollution assessment index data; Based on the pollution assessment index data, the synergistic variation coefficient between the water quality safety index and the ecological risk index is calculated to obtain the pollution coupling matrix. The release cycle and variation intensity of sediment pollutants in the pollution coupling matrix are analyzed to obtain the dynamic data of pollution diffusion. Set water quality safety thresholds, ecological risk thresholds and governance response thresholds, and perform risk grading on pollution diffusion dynamic data to obtain node risk score data; The local Moran's index and the global Moran's index are calculated for the node risk score data to obtain the risk aggregation characteristic data, and the kernel density analysis is performed on the risk aggregation characteristic data to obtain the risk hotspot distribution data. The risk area is identified based on the risk hotspot distribution data to obtain the risk zoning results, and the pollution intensity index and spatial diffusion index of each risk area are calculated based on the risk zoning results to obtain the river sediment elution risk distribution data.

[0026] Specifically, the network propagation capacity index in the river sediment pollution propagation network model is graded. The network propagation capacity index quantifies the importance and efficiency of each node in pollution propagation. Assuming that the node propagation capacity index is ,in , using the quantile method to classify Divide into Pollution assessment index data after classification It is expressed as: ; The Rank function divides the transmission capacity into levels according to the quantiles. The result of this step assigns a corresponding evaluation level to each node to distinguish between high-risk and low-risk areas. , calculate the water quality safety index and ecological risk index The synergistic variation coefficient between them is used to construct the pollution coupling matrix. The synergistic variation coefficient reflects the comprehensive impact of pollutants on water quality and ecosystems and is defined as: ; in, and are the means of water quality safety index and ecological risk index respectively. Cooperative variation coefficient matrix It is expressed as: ; This matrix provides the correlation strength between pollution indicators. The pollution coupling matrix is ​​used to analyze the release cycle and change intensity of sediment pollutants to generate pollution diffusion dynamic data. Describe the time characteristics and intensity of pollutant release from sediment It represents the fluctuation range of pollutant concentration, which is calculated as: ; in, For Node The frequency of pollution release, is the concentration series Combine these dynamic data to capture the temporal and spatial variation of pollutants. Set water quality safety thresholds , Ecological risk threshold and governance response thresholds ), and classify the risk of pollution diffusion dynamic data. Node risk score data ( ) is calculated as: ; This score reflects the comprehensive risk level of the node in terms of water quality, ecology and governance. The node risk score data is input into the spatial statistical model to calculate the local Moran index. and the global Moran index , which is used to reveal the spatial clustering characteristics of risks. The local Moran index is defined as: ; in, is the mean of the node risk scores, is the variance, For Node and The global Moran index is: ; By calculating the local and global Moran index, it is revealed whether the high-risk area shows spatial agglomeration characteristics. Kernel density analysis is performed on the risk aggregation characteristic data to generate risk hotspot distribution data. The kernel density function is expressed as: ; in, is the kernel function, is the smoothing bandwidth, is the node location. Kernel density analysis can intuitively present the spatial distribution of high-risk areas. Based on the risk hotspot distribution data, risk areas are identified and risk zoning results are divided. Combined with the risk zoning results, the pollution intensity index of each risk area is calculated. and spatial diffusion index , and obtain the river sediment elution risk distribution data. The pollution intensity index is defined as: ; in, Represents the set of nodes in a risk partition. The spatial diffusion index is defined as: ; in, For Node and spatial distance.

[0027] In a specific embodiment, the process of executing step S4 may specifically include the following steps: The riverbed sediment washout risk distribution data is input into the data preprocessing layer of the P-center problem network optimization model, and the risk level, spatial location and dredging resource parameters are standardized to obtain the optimization calculation basic data; Input the optimization calculation basic data 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 constraints of hydrological demand constraints, river connectivity constraints and resource input constraints; The dredging location selection variables are constructed for the multi-dimensional constraint condition set, and the decision variable matrix is ​​obtained by setting the dredging operation 0-1 decision variables and dredging method selection variables for each river channel node; The decision variable matrix is ​​input into the objective function construction layer of the P-center problem network optimization model, and a dual-objective optimization function is obtained by constructing a sub-objective function for minimizing the risk of pollution diffusion and a sub-objective function for maximizing the efficiency of resource utilization. Configure the LINGO solver for the dual-objective optimization function and multi-dimensional constraint condition set, and obtain the optimization solution parameters by setting the solution accuracy, number of iterations and convergence conditions; The optimization solution parameters are input into the optimization calculation layer of the P-center problem network optimization model, and the branch and bound method is used to solve the 0-1 integer programming problem to generate a river network dredging plan.

[0028] Specifically, the riverbed sediment elution risk distribution data is input into the data preprocessing layer of the P-center problem network optimization model for standardization. The risk distribution data includes risk level , spatial location and clear resource parameters ,in Indicates a river node. To ensure the uniformity of data of different dimensions, a normalization method is used to map all variables to the range of [0,1]: ; in, They are the maximum and minimum values ​​of risk level, spatial coordinates and resource parameters respectively. After standardization, the optimization calculation basic data is obtained to ensure the numerical stability and convergence of the subsequent optimization model calculation. The optimization calculation basic data is input into the constraint condition construction layer of the P-center problem network optimization model to generate a multi-dimensional constraint condition set. The constraints include hydrological demand constraints ), river connectivity constraints ) and resource input constraints ( ). Hydrological demand constraints ensure that the dredged river meets the minimum flow requirements , the formula is: ; in, For Node The cross-sectional area of ​​the river, is the flow velocity of the node. The river connectivity constraint ensures that any two nodes and Spacing Do not exceed the maximum allowed distance To ensure the connectivity of the dredging solution: ; Resource input constraints limit the total dredging resources to not exceed the total available resources : ; These constraints together constitute the basic rules of the optimization model and the feasible range of the constraint solution. Based on the multi-dimensional constraint condition set, the dredging location selection variable matrix is ​​constructed. The dredging operation variables of each node are set. Indicates whether to clear the node , and set the dredging method selection variable Indicates that at the node Select Decision variable matrix and is defined as: ; in, Represents the total number of dredging methods. Constraint variables must satisfy: ; This means that only when the node When the pollution is being cleared, a specific clearing method can be selected. The decision variable matrix is ​​input into the objective function construction layer to generate a dual-objective optimization function of minimizing the risk of pollution diffusion and maximizing the efficiency of resource utilization. The sub-objective function of minimizing the risk of pollution diffusion is defined as: ; The sub-objective function for maximizing resource utilization efficiency is: ; The comprehensive dual objective function is expressed by weighted linear combination method as follows: ; in, and is the objective function weight, which is used to balance the importance of the two. The LINGO solver is configured for the dual-objective optimization function and the multi-dimensional constraint condition set. By setting the solution accuracy ( ), number of iterations 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 sets the upper and lower bounds of the nodes. and Gradually converge: ; The branch variable selection is based on the optimal depth-first search strategy, and the current optimal solution is iteratively updated until the convergence condition is met. or the maximum number of iterations is reached The optimization solution parameters are input into the optimization calculation layer of the P-center problem network optimization model, and the river network dredging plan is generated by the branch and bound method. The dredging plan includes the choice of whether to dredge each node and the dredging method.

[0029] In a specific embodiment, the execution step configures the LINGO solver for the dual-objective optimization function and the multi-dimensional constraint condition set, and the process of obtaining the optimized solution parameters by setting the solution accuracy, the number of iterations and the convergence condition can specifically include the following steps: The objective balance analysis of the dual-objective optimization function is carried out, and the objective balance parameters are obtained by constructing the normalized weighted expressions of the pollution diffusion risk weight coefficient and the resource utilization efficiency weight coefficient. The target balance parameters are input into the constraint matrix normalization module, and the multidimensional constraint condition set is processed based on the main diagonal element normalization method to obtain a standardized constraint matrix; Initialize the linear programming of the standardized constraint matrix, set the initial basic feasible solution and the upper and lower bounds of the non-basic variables of the simplex method, obtain the initial solution parameters, input the initial solution parameters into the branch and bound solution control module, and configure the node selection rules and branch variable selection strategy to obtain the branch strategy parameters; Based on the branch strategy parameters, integer variable processing rules are constructed, and the gap tolerance value and rounding direction judgment conditions of the integer variables are set to obtain integer programming parameters. Convergence judgment rules are set for the integer programming parameters, and threshold conditions for the relative improvement rate of the objective function and the constraint satisfaction are constructed to obtain convergence control parameters. The convergence control parameters are input into the solution process monitoring module, and the maximum number of iteration steps and the upper limit of the calculation time are set to obtain the termination criterion parameters. The LINGO solver parameters are configured based on the integer programming parameters, convergence control parameters and termination criterion parameters to obtain the optimized solution parameters.

[0030] Specifically, the objective balance analysis of the dual-objective optimization function is performed 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 , the resource utilization efficiency objective function is ,in Representation Node The pollution risk level Is a decision variable, indicating whether the node is cleared represents resource utilization selection, is the clearing resource parameter. In order to construct a balanced objective function, the two objectives are normalized to obtain the normalized weight coefficient: ; The normalized dual objective function is expressed as: ; By adjusting and , flexibly control the importance of the two objectives according to different decision preferences, and generate target balance parameters The target balance parameters are input into the constraint matrix normalization module to normalize the main diagonal elements of the multidimensional constraint condition set to improve numerical stability. Assume that the constraint matrix is , the constraint vector is , which is of the form: ; Normalization is done by Normalize the main diagonal elements of by dividing each row by the corresponding diagonal element , and obtain the standardized matrix and the normalized constraint vector : ; This process ensures that the constraints have consistent dimensions, which is convenient for subsequent operations of the solver. The linear programming is initialized for the standardized constraint matrix, and the initial basic feasible solution of the simplex method and the upper and lower bounds of the non-basic variables are set. The initial basic feasible solution is defined as the starting point that satisfies the constraints. , which is of the form: ; The upper and lower bounds of non-basic variables are defined according to the constraints: ; in and are the lower and upper bounds of the variables, respectively. By initializing these parameters, the initial solution parameters are generated, providing initial conditions for the subsequent optimization of the branch and bound method. The initial solution parameters are input into the branch and bound solution control module, the node selection rules and branch variable selection strategy are configured, and the branch strategy parameters are generated. The node selection rule adopts a depth-first search or breadth-first search strategy, and the branch variable selection is based on the priority selection of non-integer degrees. The non-integer degree is defined as: ; Selecting the variable with the largest non-integer degree as the branch variable helps to quickly approximate the integer solution. Based on the branch strategy parameters, construct integer variable processing rules. Set the gap tolerance value of the integer variable And the rounding direction judgment condition, the gap tolerance value controls the accuracy of the solution: ; The rounding direction is determined according to the objective function gradient direction to ensure that the optimization direction is towards a better solution. The convergence judgment rules are set for the integer programming parameters to construct the threshold conditions for the relative improvement rate of the objective function and the constraint satisfaction. The relative improvement rate of the objective function is defined as: ; in and are the objective function values ​​of the new solution and the old solution respectively, is the convergence threshold. The constraint satisfaction is expressed as: ; in is the constraint tolerance. Input the convergence control parameters into the solution process monitoring module and set the maximum number of iteration steps. and the upper limit of calculation time , generate the termination criteria parameters. The termination condition is: Stop if: orIteration or Time ; The integer programming parameters, convergence control parameters and termination criterion parameters are combined to configure the LINGO solver and generate the final optimization solution parameters.

[0031] In a specific embodiment, the execution step inputs the optimization solution parameters into the optimization calculation layer of the P-center problem network optimization model, and the 0-1 integer programming problem is solved by the branch and bound method. The process of generating the river network dredging plan can specifically include the following steps: The standardized constraint matrix in the optimization solution parameters is relaxed, and the 0-1 integer constraints are converted into continuous variable interval constraints to obtain the 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, obtain the node upper boundary value data, build the branch variable selection rule based on the node upper boundary value data, 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, generate subproblems and solve the optimal solutions of the subproblems, obtain the active node set, update the boundary value 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 by the optimal depth-first search strategy to obtain a candidate node set. Perform feasibility test based on the candidate node set, verify the integer characteristics of the node solution and the degree of constraint satisfaction, and obtain a feasible solution set. The dredging scheme is generated for the feasible solution set, and the decision variables are mapped to the dredging operation scheme of the river node to generate the river network dredging scheme.

[0032] Specifically, the standardized constraint matrix in the optimization solution parameters is relaxed, and the 0-1 integer constraints are converted into continuous variable interval constraints, thereby relaxing 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 variables are , whose constraint form is: ; Relaxation expands variable definitions to continuous intervals by relaxing integer constraints: ; The relaxed problem is transformed into a linear programming relaxation problem: ; in, is the coefficient vector of the objective function. The relaxed problem is input into the simplex calculation module and the simplex method is used to iteratively solve the problem to find the initial optimal solution. The simplex method starts from an initial feasible solution and searches along the direction of the constraint boundary to gradually improve the objective function value. Initial feasible solution satisfy: ; And by iteratively selecting the entry and exit of the basic variables, we finally find an optimal solution that satisfies all constraints. Through this solution process, we can get the node upper boundary data , the value of each variable represents the optimal value in the relaxed problem. Based on the node upper bound data, the branch variable selection rule is constructed and the non-integer degree of each decision variable is calculated, which is defined as: ; in, and They are The lower and upper integer values ​​of . The larger the non-integer degree, the more the variable deviates from the integer solution, and this variable should be selected for branching. Calculate the influence coefficient of each variable, defined as: ; in is the coefficient of the variable in the objective function. A branch sequence is constructed based on the non-integer degree and influence coefficient, and the variable with the greatest influence is selected for branching. The branch sequence is input into the branch and bound tree construction module, and two sub-problems are generated based on the selected branch variables: one sub-problem constraint , another subproblem constraint By solving the optimal solution of each subproblem, we get the set of active nodes, each of which represents a potential solution. We update the limit values ​​of the active node set and update the upper and lower bounds according to the optimal solution of the subproblem: ; in, is the current optimal solution, and are the minimum and maximum objective function values ​​of the sub-problems respectively. For each active node, the difference between the objective function value and the current optimal value is calculated to obtain the node evaluation data: ; in Is a node The node evaluation data is input into the node traversal module, and the next node to be processed is selected using the optimal depth-first search strategy. In the node traversal module, a feasibility test is performed based on the candidate node set to verify whether the solution satisfies the integer characteristics and constraints. Satisfy all constraints and , then the solution is included in the feasible solution set. Generate dredging solutions for the feasible solution set and map the final integer solution to the dredging operation plan of the river node. For example, suppose a river network has 5 nodes and its optimal solution is , indicating that nodes 1, 3, and 4 are cleared, while nodes 2 and 5 are not cleared. The clearing method is determined by the auxiliary variable Mapped to specific operations, such as "Mechanical dredging", "Rinse and clear".

[0033] The above describes the river channel dredging scheme decision method based on sediment elution in the embodiment of the present invention. The following describes the river channel dredging scheme decision device based on sediment elution in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, a river channel dredging solution decision device based on sediment elution includes: The acquisition module is used to collect and standardize the data of the spatial distribution, cross-sectional morphology, hydrological parameters and sediment pollution characteristics of the river network system to obtain the river network characteristic data set; The analysis module is used to analyze the migration and diffusion of sediment pollutants based on the river network characteristic data set, and calculate the pollution propagation relationship between nodes to obtain the river sediment pollution propagation network model; A calculation module is used to calculate the coupling degree of river sediment pollution based on the river sediment pollution propagation network model to obtain the river sediment elution risk distribution data; The generation module is used to input the riverbed sediment elution risk distribution data into the P-center problem network optimization model, and generate a river network dredging plan by minimizing the risk of pollution diffusion and maximizing the efficiency of dredging resource utilization. Through the collaborative cooperation of the above-mentioned components, the present invention accurately describes the migration and diffusion laws of sediment pollutants by constructing a riverbed sediment pollution propagation network model, combining time series analysis and causal test methods, and provides scientific data support for dredging plan decision-making. The present invention uses a multi-dimensional evaluation index system to quantitatively calculate the coupling degree of riverbed sediment pollution, identifies risk hotspots through spatial statistical analysis methods, and effectively improves the accuracy of risk assessment. Based on the P-center problem network optimization theory, the present invention constructs an optimization model with dual objectives of minimizing the risk of pollution diffusion and maximizing the efficiency of dredging resource utilization, which solves 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 constraints and optimization algorithms. In the optimization solution process, the present invention adopts target balance analysis and constraint matrix standardization processing, establishes a complete solution parameter configuration system, and ensures 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, which improves the scientificity and efficiency of river management.

[0034] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As 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 designed by the computer 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 method is implemented.

[0035] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0036] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and 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.

[0037] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and 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 embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0038] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0039] If 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 is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0040] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A river dredging scheme decision-making method based on sediment elution, characterized in that: The method comprises: Data collection and standardized preprocessing of the spatial distribution, cross-sectional morphology, hydrological parameters and sediment pollution characteristics of the river network system were performed to obtain a river network characteristic data set; Based on the river network characteristic data set, the migration and diffusion analysis of sediment pollutants is performed, and the pollution propagation relationship between nodes is calculated to obtain a river sediment pollution propagation network model; Calculate the coupling degree of river sediment pollution based on the river sediment pollution propagation network model to obtain river sediment elution risk distribution data; The riverbed sediment elution risk distribution data is input into the P-center problem network optimization model, and a river network dredging plan is generated by minimizing the pollution diffusion risk and maximizing the dredging resource utilization efficiency.

2. The method for river dredging decision-making based on sediment elution according to claim 1 is characterized in that: The spatial distribution, cross-sectional morphology, hydrological parameters and sediment pollution characteristics of the river network system are collected and standardized preprocessed to obtain a river network characteristic data set, including: Carry out GIS mapping of the river network system, collect spatial coordinate data including longitude, latitude and elevation, and obtain the original data of river spatial distribution; Inputting the original data of spatial distribution of the river channel into the topological analysis model, and obtaining the river network topological structure data by calculating the node connectivity, path connectivity and network density; Collecting morphological parameters of the river section in the river network topological structure data, measuring parameters including river width, water depth, cross-sectional area and longitudinal slope, and obtaining river morphological characteristic data; Based on the river morphological characteristic data, hydrological monitoring points are set to collect hydrological elements including cross-sectional flow velocity, flow, water level and sediment content to obtain river hydrological characteristic data; Using the river hydrological characteristic data as a sampling basis, stratified sampling and pollutant detection are performed on river sediment to obtain sediment pollutant characteristic data, wherein the sediment pollutant characteristic data includes pollutant types, concentration distribution and spatial variability; A Kriging spatial interpolation model is established based on the sediment pollutant characteristic data, and the sediment pollutant spatial distribution data is obtained through variance function calculation and cross-validation; Normalizing 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; The standardized river channel characteristic data is input into a density-based clustering algorithm model, and the river channel network characteristic data set is obtained by calculating the Euclidean distance and local density of the river channel characteristic vector.

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

4. The method for river dredging decision-making based on sediment elution according to claim 3 is characterized in that: The coupling degree of river sediment pollution is calculated based on the river sediment pollution propagation network model to obtain river sediment elution risk distribution data, including: Classifying the network propagation capacity index in the river sediment pollution propagation network model to obtain pollution assessment index data; Based on the pollution assessment index data, the synergistic variation coefficient between the water quality safety index and the ecological risk index is calculated to obtain a pollution coupling matrix, and the sediment pollutant release cycle and variation intensity of the pollution coupling matrix are analyzed to obtain pollution diffusion dynamic data; Setting water quality safety thresholds, ecological risk thresholds and governance response thresholds, and performing risk grading on the pollution diffusion dynamic data to obtain node risk score data; Calculating the local Moran's index and the global Moran's index on the node risk score data to obtain risk aggregation feature data, and performing kernel density analysis on the risk aggregation feature data to obtain risk hotspot distribution data; Based on the risk hotspot distribution data, risk areas are identified to obtain risk zoning results, and based on the risk zoning results, the pollution intensity index and spatial diffusion index of each risk area are calculated to obtain riverbed sediment elution risk distribution data.

5. The method for river dredging decision-making based on sediment elution according to claim 4 is characterized in that: The riverbed sediment elution risk distribution data is input into the P-center problem network optimization model, and a river network dredging plan is generated by minimizing the risk of pollution diffusion and maximizing the efficiency of dredging resource utilization, including: The riverbed sediment elution risk distribution data is input into the data preprocessing layer of the P-center problem network optimization model, and the risk level, spatial location and dredging resource parameters are standardized to obtain the optimization calculation basic data; Inputting the optimization calculation basic data into the constraint condition construction layer of the P-center problem network optimization model, and obtaining a multidimensional constraint condition set by setting the restriction conditions of hydrological demand constraint, river channel connectivity constraint and resource input constraint; Constructing dredging location selection variables for the multidimensional constraint condition set, and obtaining a decision variable matrix by setting dredging operation 0-1 decision variables and dredging method selection variables 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 pollution diffusion risk minimization sub-objective function and a resource utilization efficiency maximization sub-objective function; Performing LINGO solver configuration on the dual-objective optimization function and the multi-dimensional constraint condition set, and obtaining optimization solution parameters by setting solution accuracy, number of iterations and convergence conditions; The optimization solution parameters are input into the optimization calculation layer of the P-center problem network optimization model, and the 0-1 integer programming problem is solved by using the branch and bound method to generate a river network dredging plan.

6. The method for river dredging decision-making based on sediment elution according to claim 5 is characterized in that: The LINGO solver is configured for the dual-objective optimization function and the multi-dimensional constraint condition set, and the optimization solution parameters are obtained by setting the solution accuracy, the number of iterations and the convergence condition, including: Performing a target balance analysis on the dual-objective optimization function, and obtaining a target balance parameter by constructing a normalized weighted expression of a pollution diffusion risk weight coefficient and a resource utilization efficiency weight coefficient; Inputting the target balance parameters into a constraint matrix normalization module, processing the multidimensional constraint condition set based on a main diagonal element normalization method, and obtaining a standardized constraint matrix; Initializing the standardized constraint matrix for linear programming, setting the initial basic feasible solution and the upper and lower bounds of the non-basic variables of the simplex method, obtaining initial solution parameters, inputting the initial solution parameters into the branch and bound solution control module, and configuring the node selection rules and the branch variable selection strategy to obtain the branch strategy parameters; Based on the branch strategy parameters, integer variable processing rules are constructed, and gap tolerance values ​​and rounding direction judgment conditions of integer variables are set to obtain integer programming parameters, convergence judgment rules are set for the integer programming parameters, and threshold conditions of relative improvement rate of objective function and constraint satisfaction are constructed to obtain convergence control parameters; The convergence control parameters are input into the solution process monitoring module, and the maximum number of iteration steps and the upper limit of the calculation time are set to obtain the termination criterion parameters. The LINGO solver parameters are configured based on the integer programming parameters, the convergence control parameters and the termination criterion parameters to obtain the optimized solution parameters.

7. The method for river dredging decision-making based on sediment elution according to claim 6 is characterized in that: The optimization solution parameters are input into the optimization calculation layer of the P-center problem network optimization model, and the 0-1 integer programming problem is solved by the branch and bound method to generate a river network dredging plan, including: Performing relaxation processing on the standardized constraint matrix in the optimization solution parameters, and converting the 0-1 integer constraints into continuous variable interval constraints, to obtain a linear programming relaxation problem; Input the linear programming relaxation problem into a simplex calculation module, and iteratively solve to obtain an initial optimal solution, obtain node upper boundary value data, construct a branch variable selection rule based on the node upper boundary value data, and calculate the non-integer degree and influence coefficient of the decision variable to obtain a branch sequence; Input the branch sequence into a branch and bound tree construction module, generate subproblems and solve the optimal solutions of the subproblems, obtain an active node set, update the boundary value 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 node evaluation data; Input the node evaluation data into the node traversal module, and select the next node to be processed by 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 characteristics and constraint satisfaction degree of the node solution to obtain a feasible solution set; A dredging scheme is generated for the feasible solution set, and the decision variables are mapped to the dredging operation scheme of the river channel nodes to generate a river network dredging scheme.

8. A river dredging solution decision-making device based on sediment elution, characterized in that: Used to execute the river channel dredging scheme decision method based on sediment elution according to any one of claims 1 to 7, the river channel dredging scheme decision device based on sediment elution comprises: The acquisition module is used to collect and standardize the data of the spatial distribution, cross-sectional morphology, hydrological parameters and sediment pollution characteristics of the river network system to obtain the river network characteristic data set; An analysis module is used to perform sediment pollutant migration and diffusion analysis based on the river network characteristic data set, and calculate the pollution propagation relationship between nodes to obtain a river sediment pollution propagation network model; A calculation module, used to calculate the coupling degree of river sediment pollution based on the river sediment pollution propagation network model to obtain river sediment elution risk distribution data; A generation module is used to input the riverbed sediment elution risk distribution data into the P-center problem network optimization model, and generate a river network dredging plan by minimizing the pollution diffusion risk and maximizing the dredging resource utilization efficiency.

9. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program executable on the processor, and is characterized in that when the processor executes the computer program, the river dredging plan decision-making method based on sediment elution as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the processor is enabled to execute the river dredging plan decision-making method based on sediment elution as described in any one of claims 1 to 7.

Images