An intelligent planning method for water and soil conservation measures of slope farmland

Abstract

The present application relates to the technical field of agricultural management, in particular to a kind of slope farmland water and soil conservation measure intelligent planning method, comprising the following steps, obtain elevation, soil and rainstorm data to generate drainage network, in combination with flood peak flow and soil erosion resistance to assess erosion risk, and layout drain and access point, later based on risk and cost optimization network, to generate access scheme with Hungarian algorithm, again identify key node to determine operating threshold, finally through the turbidity and flow data and threshold comparison, judge soil erosion state.In the present application, by fusing elevation, soil and rainstorm intensity and other multi-source data, the global drainage network topology can be accurately calculated, the erosion risk can be quantitatively evaluated, and the global optimization can be performed by weighing the reinforcement and reconstruction cost, then the optimal connection scheme is solved by algorithm, and then the key monitoring node is automatically identified, the turbidity and flow data are analyzed in real time, the soil erosion dynamic monitoring and immediate warning are realized, and the passive management is changed into active prevention and control.

Description

Technical Field

[0001] This invention relates to the field of agricultural management technology, and in particular to an intelligent planning method for soil and water conservation measures on sloping farmland. Background Technology

[0002] The field of intelligent planning methods for soil and water conservation measures on sloping farmland is an interdisciplinary application field integrating computer science, geographic information systems, and soil and water conservation engineering. Its core aspects include acquiring multi-source heterogeneous data on the topography, soil properties, and rainfall erosion capacity of sloping farmland, and constructing soil erosion prediction models and measure benefit evaluation models to systematically provide quantitative and intelligent planning solutions for regional soil erosion control. Traditional agricultural management, in contrast, focuses on determining the type of prevention and control measures and engineering layout for specific plots during soil and water conservation planning. This typically involves technicians carrying surveying tools to conduct on-site surveys, obtaining information such as topographic slope and soil thickness, and manually laying out engineering measures such as terraces and drainage ditches based on general soil and water conservation design specifications or personal experience.

[0003] Traditional agricultural management methods heavily rely on on-site surveys by technical personnel. This point-sampling method is difficult to fully obtain information on the complex topography of sloping farmland, especially in large areas or areas with varied terrain. It is easy to misjudge the risk of soil erosion due to blind spots in the survey. At the same time, the planning and deployment of prevention and control measures are highly dependent on individual work experience or general standards. There is a lack of quantitative evaluation and optimization process for the economic and effectiveness of the project layout, which leads to highly subjective solutions and difficulty in achieving optimal results. For example, a drainage system based on experience may cause local erosion because it fails to accurately identify key confluence paths. Finally, due to the lack of dynamic monitoring and feedback mechanisms after construction, the potential erosion risk cannot be warned in advance when encountering unexpected rainfall, and often only passive remedial measures can be taken after the disaster occurs. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and propose an intelligent planning method for soil and water conservation measures on sloping farmland.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an intelligent planning method for soil and water conservation measures on sloping farmland, comprising the following steps:

[0006] S1: Obtain elevation model data, soil distribution data and rainstorm intensity, calculate the tree-like drainage network topology structure from the elevation model data, calculate the peak flow, compare the scour resistance of the soil distribution data, generate a scour risk quantification value, obtain terrace contour data, and set up discrete points to generate a set of drainage outlets and a set of access points.

[0007] S2: Based on the scour risk quantification value, compare the reinforcement and reconstruction costs, optimize the tree-like drainage network topology to generate a drainage network topology, call the set of outlets and the set of access points, use the Hungarian algorithm to solve the connection cost, and generate an access scheme.

[0008] S3: Identify key monitoring nodes using the drainage network topology, plan the deployment path of monitoring equipment according to the access scheme, calculate the peak flow and sediment carrying capacity of the key monitoring nodes, and determine the operating benchmark threshold.

[0009] S4: Collect turbidity data and flow data at the key monitoring nodes and compare them with the operating benchmark threshold. If the turbidity data exceeds the operating benchmark threshold, soil erosion is determined and an erosion status indicator is generated.

[0010] As a further aspect of the present invention, the scour risk quantification value includes a unique identifier for the ditch section, a risk level, and a difference value; the set of spillways includes the geographical coordinates of the spillways and the terrace number to which they belong; the set of access points includes the geographical coordinates of the access points and the ditch number to which they belong; the drainage network topology specifically includes ditch connection relationships, node types, and ditch protection levels; the access scheme specifically refers to the matching pair between spillways and access points and the engineering cost of connecting channels; the operational benchmark threshold includes the design flood peak flow threshold and the theoretical sediment carrying capacity threshold; and the erosion status identifier includes the location of the warning node, the extent of turbidity exceeding the standard, and the warning timestamp.

[0011] As a further aspect of the present invention, the steps for obtaining the set of drainage outlets and the set of access points are specifically as follows:

[0012] S101: Obtain elevation model data, soil distribution data and rainstorm intensity; calculate and generate a tree-like drainage network topology based on the terrain information of the elevation model data; calculate the peak flow by combining the rainstorm intensity and the tree-like drainage network topology; extract multiple soil type erosion resistance parameters from the soil distribution data and compare them numerically with the peak flow to generate a scour risk quantification value.

[0013] S102: Obtain terrace contour data, set a scour risk judgment threshold for the scour risk quantification value, search for spatial regions where the scour risk quantification value exceeds the scour risk judgment threshold, and perform overlay analysis on the vector boundary of the spatial region and the terrace contour data to obtain a discrete set of layout points.

[0014] S103: Call the tree-like drainage network topology and the discrete layout point set, calculate the spatial distance from multiple points in the discrete layout point set to the main flow line of the tree-like drainage network topology, classify the multiple points according to the preset distance judgment value and point elevation information, and establish a set of drainage outlets and a set of access points.

[0015] As a further aspect of the present invention, the step of obtaining the access scheme specifically includes:

[0016] S201: Based on the scour risk quantification value, obtain the unit cost of reinforcement of drainage network nodes and the unit cost of flow path reconstruction. Traverse multiple flow path units in the tree-like drainage network topology, calculate the reinforcement cost by combining the scour risk quantification value and the unit cost of reinforcement of drainage network nodes, and retrieve alternative flow paths by combining the unit cost of flow path reconstruction to calculate the reconstruction cost. Based on the numerical comparison results of the reinforcement cost and the reconstruction cost, establish a node optimization judgment set.

[0017] S202: Call the node optimization decision set, filter the flow path units whose decision result is reconstruction, replace them with the alternative flow path in the tree-like drainage network topology, and update the connection relationship and weight parameters between nodes to generate the drainage network topology.

[0018] S203: Based on the drainage network topology, and by calling the set of drainage outlets and the set of access points, calculate the connection path cost between multiple drainage outlets and multiple access points, construct a connection cost matrix, solve the connection cost matrix, obtain the connection pairing with the minimum total cost, and generate an access scheme.

[0019] As a further aspect of the present invention, the step of obtaining the operating benchmark threshold specifically includes:

[0020] S301: Calculate the network centrality parameter and the upstream confluence contribution parameter using multiple network nodes in the drainage network topology, perform a weighted summation on the network centrality parameter and the upstream confluence contribution parameter to generate a node criticality score, set a node screening score threshold, retrieve the node criticality score and extract network nodes that exceed the node screening score threshold, and generate a set of key monitoring nodes.

[0021] S302: According to the access scheme, call the set of key monitoring nodes to obtain the spatial location coordinates of multiple key monitoring nodes and the terrain connectivity data between nodes. Based on the spatial location coordinates and the terrain connectivity data, calculate the total path cost of feasible deployment sequences that traverse all key monitoring nodes, select the deployment sequence with the minimum total path cost, and generate the deployment path of the monitoring equipment.

[0022] S303: For each key monitoring node in the set of key monitoring nodes, acquire elevation model data and soil distribution data of the control watershed area, calculate its peak flow rate in combination with the rainfall intensity, calculate the sediment carrying capacity, set multi-level response values ​​based on the calculation results of the peak flow rate and the sediment carrying capacity, and establish an operating benchmark threshold.

[0023] As a further aspect of the present invention, the step of obtaining the erosion state identifier specifically includes:

[0024] S401: Based on the key monitoring nodes, deploy monitoring equipment and set data acquisition frequency parameters and data upload cycle. The monitoring equipment synchronously collects turbidity data and flow data according to the data acquisition frequency parameters, encodes the collected turbidity data and flow data into independent data frames respectively, aggregates multiple sets of data frames at the end of the data upload cycle, and adds node identifier and timestamp information to each data frame to obtain the real-time data sequence of the monitoring nodes.

[0025] S402: Receive the real-time data sequence of the monitoring node, parse the data frame to extract the turbidity data, the flow data, the node identifier and the timestamp information, call the multi-level turbidity response threshold set for the differential flow data value range in the running benchmark threshold, perform sequential iterative comparison between the current turbidity data value and the multi-level turbidity response threshold corresponding to the current flow data range, and obtain the erosion event judgment result;

[0026] S403: Based on the erosion event determination result, if the determination result indicates that the turbidity data value exceeds any level of turbidity response threshold, then extract the node identifier that triggered the determination, the timestamp information, and the threshold level parameter that was exceeded, and connect the three in a structured manner, wherein the turbidity exceedance is characterized by the threshold level parameter that was exceeded, and establish an erosion state identifier.

[0027] As a further aspect of the present invention, the step of generating the scour risk quantification value specifically includes:

[0028] Obtain the elevation model data, the soil distribution data, and the upstream catchment area of ​​multiple channel segments in the tree-like drainage network topology, and call the peak flow rate;

[0029] Based on the soil distribution data, soil initiation flow parameters and erosion resistance parameters corresponding to multiple soil types are extracted;

[0030] Based on the scour risk quantification value of each channel segment in the tree-like drainage network topology, Through the formula:

[0031] calculate;

[0032] in, The quantified value representing the scour risk of the channel section. This represents the channel morphology correction factor that combines channel tortuosity and roughness. This represents the design peak flow rate of the current channel section. The soil initiation flow parameter represents the current location of the channel segment. Represents the upstream catchment area of ​​the channel section. Soil erosion resistance parameters representing the location of the gully section;

[0033] Based on the calculated scour risk quantification value Based on the preset risk level classification standard, each channel segment is assigned a unique identifier, risk level, and the difference value indicating exceeding the critical value, thereby generating the scour risk quantification value.

[0034] As a further aspect of the present invention, the steps for establishing the node optimization decision set are as follows:

[0035] The system calls the scour risk quantification value, the unit cost of the drainage network node reinforcement, and the unit cost of the flow path reconstruction, and traverses all flow path units in the tree-like drainage network topology.

[0036] For any flow path unit, the required ditch protection standard is determined based on the risk level in the corresponding scour risk quantification value. The reinforcement cost of the flow path unit is calculated by combining the length of the flow path unit and the terrain parameters, and by calling the unit cost of the drainage network node reinforcement.

[0037] Based on the tree-like drainage network topology, an alternative flow path with the same start and end points as the current flow path unit and within a preset spatial search radius is searched. The length and expected workload of the alternative flow path are calculated, and the unit cost of the flow path reconstruction is called to calculate the reconstruction cost.

[0038] Compare the reinforcement cost with the reconstruction cost. If the reinforcement cost is greater than the reconstruction cost, the optimization strategy is determined to be reconstruction; otherwise, it is reinforcement.

[0039] The unique identifier of each flow path unit and its corresponding optimization strategy judgment result are structured and paired to generate a node optimization judgment set.

[0040] As a further aspect of the present invention, the step of generating the node criticality score specifically includes:

[0041] Invoke the drainage network topology and traverse all network nodes within it;

[0042] For each network node, calculate its betweenness centrality in the drainage network topology to obtain the original betweenness centrality parameter, and calculate the sum of the catchment areas of all channel segments upstream of the node to obtain the original upstream catchment contribution parameter.

[0043] Normalization is performed on the original betweenness centrality parameter and the original upstream confluence contribution parameter of all network nodes to obtain the network centrality parameter and the upstream confluence contribution parameter;

[0044] The node key score Through the formula:

[0045] calculate;

[0046] in, The node criticality score represents the node's performance in the network. The weighting coefficients represent the network centrality used to adjust for the influence of network structure importance. The weighting coefficient representing the upstream confluence contribution used to adjust for the importance of the hydrological response. The parameter representing the original betweenness centrality of the current node. and These represent the maximum and minimum original betweenness centrality parameters in the entire drainage network topology, respectively. The parameter representing the original upstream confluence contribution of the current node. and These represent the maximum and minimum original upstream confluence contribution parameters in the entire drainage network topology, respectively.

[0047] The calculated node criticality score for each network node Associate it with its unique node identifier to generate a node criticality score.

[0048] As a further aspect of the present invention, the step of obtaining the erosion event determination result specifically includes:

[0049] The system invokes the operating baseline threshold and receives the real-time data sequence from the monitoring node, parsing it to obtain the current turbidity data and current flow data at the target timestamp;

[0050] Based on the value of the current traffic data, match and search for the traffic data value range to which it belongs in the operating benchmark threshold;

[0051] Obtain the multi-level turbidity response threshold corresponding to the numerical range of the flow data, wherein the multi-level turbidity response threshold includes multiple incremental thresholds from level one to multiple levels;

[0052] The current turbidity data is compared with the first-level turbidity response threshold in the multi-level turbidity response threshold. If the current turbidity data is greater than the first-level turbidity response threshold, it is determined that an erosion event has occurred, and the threshold level parameter that has been exceeded is recorded as level one.

[0053] If it is less than or equal to, then continue to compare the current turbidity data with the secondary turbidity response threshold, iterate until the highest threshold or the judgment event occurs, and record the threshold level parameter that is exceeded.

[0054] If the current turbidity data does not exceed the turbidity response threshold for all levels, it is determined that no erosion event has occurred, and an erosion event determination result is generated.

[0055] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0056] In this invention, by integrating multi-source data such as elevation models, soil distribution, and rainfall intensity, the topology of the tree-like drainage network across the entire region can be accurately calculated, and the erosion risk can be quantitatively assessed, providing a quantitative basis for soil and water conservation planning. Based on this, the network topology is globally optimized by balancing the costs of reinforcement and reconstruction according to the quantified erosion risk value. Then, the optimal connection scheme of each drainage outlet and access point is solved by an algorithm to ensure the economy and rationality of the engineering layout. Furthermore, key monitoring nodes in the drainage network are automatically identified, and operating benchmark thresholds are preset. By collecting and analyzing turbidity and flow data in real time, dynamic monitoring and immediate early warning of soil erosion status can be achieved, transforming passive treatment into proactive prevention and control. Attached Figure Description

[0057] Figure 1 This is a flowchart of an intelligent planning method for soil and water conservation measures on sloping farmland according to the present invention.

[0058] Figure 2 This is a flowchart illustrating the risk quantification and site selection process for the present invention.

[0059] Figure 3 This is a flowchart illustrating the generation process of the drainage network topology optimization and access scheme of the present invention.

[0060] Figure 4 This is a flowchart illustrating the key monitoring node identification and deployment path planning process of this invention.

[0061] Figure 5 This is a flowchart of the soil erosion status monitoring and early warning system of the present invention. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of this invention clearer, the software-based technical solution is described in detail below with reference to system architecture diagrams and embodiments. It should be understood that the specific embodiments described herein are only for explaining the technical solutions of this invention and do not constitute a limitation on the scope of protection.

[0063] In the description of this invention, the system architecture relationships or data processing flows indicated by terms such as "layer," "module," "interface," "data flow," "client," and "server" are all defined based on the architecture diagram or flowchart corresponding to the embodiments. This way of describing is only used to clearly illustrate the logical relationships between the elements in the technical solution, and not to limit the physical deployment form. The term "multiple" includes two or more technical units, including but not limited to multiple data nodes, processing threads, service instances, or functional components and other scalable elements. The specific number is determined according to the actual business scenario and needs to be specifically specified.

[0064] Please see Figure 1 and Figure 2 This invention provides a technical solution: an intelligent planning method for soil and water conservation measures on sloping farmland, comprising the following steps:

[0065] S1: Acquire elevation model data, soil distribution data and rainstorm intensity, calculate elevation model data to generate tree-like drainage network topology, calculate peak flow, compare soil distribution data to assess scour resistance, generate scour risk quantification value, acquire terrace contour data, and set up discrete points to generate a set of drainage outlets and access points.

[0066] The scour risk quantification includes the unique identifier of the ditch section, the risk level, and the difference value.

[0067] The set of drainage outlets includes the geographical coordinates of the drainage outlets and the terrace number to which they belong;

[0068] The set of access points includes the geographical coordinates of the access points and the channel number to which they belong;

[0069] The specific steps for obtaining the set of drainage outlets and the set of access points are as follows:

[0070] S101: Obtain elevation model data, soil distribution data and rainfall intensity; calculate and generate a tree-like drainage network topology based on the terrain information of the elevation model data; calculate the peak flow by combining the rainfall intensity and the tree-like drainage network topology; extract the erosion resistance parameters of multiple soil types from the soil distribution data and compare them numerically with the peak flow to generate a scour risk quantification value.

[0071] S102: Obtain terrace contour data, set a scour risk judgment threshold for the scour risk quantification value, search for spatial areas where the scour risk quantification value exceeds the scour risk judgment threshold, and overlay the spatial area vector boundary with the terrace contour data to obtain a discrete set of layout points.

[0072] S103: Call the tree-like drainage network topology and discrete layout point set, calculate the spatial distance from multiple points in the discrete layout point set to the main flow line of the tree-like drainage network topology, classify the multiple points according to the preset distance judgment value and point elevation information, and establish the drainage outlet set and the access point set.

[0073] The specific steps for generating the scour risk quantification value are as follows: obtain elevation model data, soil distribution data, and the upstream catchment area of ​​multiple channel segments in the tree-like drainage network topology, and call the peak flow rate;

[0074] Based on soil distribution data, soil initiation flow parameters and erosion resistance parameters corresponding to multiple soil types were extracted;

[0075] Based on the scour risk quantification value of each channel segment in the tree-like drainage network topology. Through the formula:

[0076] calculate;

[0077] in, The quantified value representing the scour risk of the channel section. This represents the channel morphology correction factor that combines channel tortuosity and roughness. This represents the design peak flow rate of the current channel section. The soil initiation flow parameter represents the current location of the channel segment. Represents the upstream catchment area of ​​the channel section. Soil erosion resistance parameters representing the location of the gully section;

[0078] Based on the calculated scour risk quantification value Based on the preset risk level classification standards, each channel segment is assigned a unique identifier, risk level, and a value representing the difference between exceeding the critical value, thereby generating a scour risk quantification value.

[0079] A 1.5 square kilometer sloping farmland area located in the southern Loess Plateau was selected as the implementation target. First, a 5m × 5m digital elevation model (DEM) of the area was acquired. Simultaneously, soil distribution data for the area was collected; the results are shown in Table 1. A rainfall intensity was set, referencing the "Outdoor Drainage Design Standard" (GB50014-2021), and a rainfall intensity formula with a 20-year return period was used for calculation, yielding a design rainfall intensity of 265.8 mm / hour.

[0080] The above calculation using the storm intensity formula with a return period of 20 years yields a design storm intensity of 265.8 mm / h, which is based on the storm intensity formula provided in the "Outdoor Drainage Design Standard" (GB50014-2021). Perform calculations; among which, Represents the intensity of heavy rainfall (liters per second per hectare). Represents the return period of rainfall (in years). Represents the duration of rainfall (in minutes). , , , The parameters are calculated based on local rainfall data; for the area in this embodiment, the parameters are determined by consulting local rainfall statistics. , , , Set the rainfall recurrence period Year, duration of rainfall Based on the confluence time of the study area being calculated as 30 minutes, the design rainfall intensity was obtained by substituting the values ​​into the formula. Liters per second per hectare, when converted to millimeters per hour, is approximately 265.8 millimeters per hour;

[0081] Table 1 Soil types and erosion resistance parameters of the target area

[0082] ;

[0083] Table 1 lists the starting flow rate and scour resistance parameters of the main soil types in the study area.

[0084] Based on the 5-meter resolution DEM data, depression filling was first performed, followed by calculating the flow direction of each grid cell to generate a cumulative runoff network. A runoff accumulation threshold of 500 was set; that is, when the number of upstream runoff grid cells of a grid cell exceeds 500, that grid cell is defined as part of the channel network. This method generates a tree-like drainage network topology covering the entire area, consisting of 25 main channels and 112 branch channels. Each channel segment is assigned a unique identifier; for example, main channel 1 is denoted as M01, and its branch channels are denoted as S01-1, S01-2, etc.

[0085] Next, the peak flood discharge is calculated by combining the design rainfall intensity and the drainage network topology. For channel section S05-3, the upstream catchment area, obtained from DEM data, is 0.08 square kilometers. The design peak flood discharge is calculated using a deductive formula method, with a runoff coefficient of 0.7 and a design rainfall intensity of 265.8 mm / h. The calculated design peak flood discharge for this channel section is then obtained. The flow rate is 4.13 cubic meters per second.

[0086] The design peak flow was calculated using the inference formula method described above, with a runoff coefficient of 0.7 and a design rainfall intensity of 265.8 mm / hour. The calculated design peak flow for this channel section was then obtained. The figure of 4.13 cubic meters per second is calculated using the Rational Method, a commonly used method in hydrology. The formula is as follows: ;in, Represents the design peak flow rate (cubic meters per second). Represents the runoff coefficient (dimensionless). Represents the design rainfall intensity (mm / hour). Represents the upstream catchment area (square kilometers); in this embodiment, , mm / hour square kilometers, substituting into the formula to calculate... cubic meters per second;

[0087] Soil distribution data were extracted, and the soil type at the location of ditch section S05-3 is loess. Referring to Table 1, the soil initiation flow parameters are as follows: The soil erosion resistance parameter is 0.015 cubic meters per second. It is 0.85.

[0088] Quantitative value of scouring risk The calculation formula is:

[0089] ;

[0090] in, The quantified value representing the scour risk of the channel section. This represents the channel morphology correction factor that combines channel tortuosity and roughness. This represents the design peak flow rate of the current channel section. The soil initiation flow parameter represents the current location of the channel segment. Represents the upstream catchment area of ​​the channel section. Soil erosion resistance parameters representing the location of the gully section;

[0091] Assign values ​​and perform calculations to the parameters in the formula:

[0092] The design references the actual morphology of channel section S05-3. Through DEM data analysis, the tortuosity of this channel section is 1.2 (the ratio of channel length to the straight-line distance between the two endpoints), and the roughness, based on on-site surveys, is determined to be 0.03. To determine... A series of scouring simulation experiments were conducted. In a controlled water tank, different combinations of tortuosity and roughness were set, a constant water flow was applied, and the scouring rate was measured. By performing binary linear regression analysis on the experimental data, the fitted relationship was obtained: Substituting the curvature of the channel section S05-3 (1.2) and roughness (0.03) into the equation, the calculation yields... .

[0093] The above is for confirmation A series of scouring simulation experiments were conducted. In a controlled water tank, different combinations of tortuosity and roughness were set, a constant water flow was applied, and the scouring rate was measured. By performing binary linear regression analysis on the experimental data, the fitted relationship was obtained: This refers to the provision of specific experimental data recording and analysis processes. The control tank used in the experiment was 5 meters long, 0.5 meters wide, and 0.5 meters high, with an adjustable slope. In the experiment, 20 working conditions were set up for the loess soil samples as shown in Table 2. Under each working condition, a constant water flow of 0.05 cubic meters per second was applied for 10 minutes, and the scouring depth was measured using a high-precision laser scanner, converted into the scouring rate per unit time. The scouring rate was normalized and used as the observed value. A binary linear regression was performed with the tortuosity and roughness to obtain the above-mentioned fitted relationship, with its coefficient of determination... It is 0.92.

[0094] Table 2 Excerpt of experimental data for coefficient fitting

[0095] ;

[0096] The calculated flow rate is 4.13 cubic meters per second.

[0097] According to Table 1, the starting flow rate parameter for loess is 0.015 cubic meters per second.

[0098] The area has been estimated at 0.08 square kilometers.

[0099] According to Table 1, the erosion resistance parameter of loess is 0.85.

[0100] Substitute the above parameter values ​​into the formula:

[0101] ;

[0102] The result indicates that the scour risk quantification value of channel section S05-3 is 63.54;

[0103] A threshold for assessing scour risk was set, based on historical scour data from 100 different gully segments within the study area. Statistical analysis of the calculated values. Statistics show that when... When the value is below 15, more than 95% of the channel sections do not experience significant scouring; when When the value is between 15 and 30, varying degrees of scouring occur in the channel section; when When the value is higher than 30, 98% of the channel sections experience severe scouring. Therefore, the scouring risk assessment threshold is set at 15. The risk level classification criteria are as follows: Low risk; Medium risk; High risk. Section S05-3 of the ditch. The value is 63.54, which falls under the high-risk category. The difference between this value and the critical value is [value missing]. Ultimately, the scour risk quantification value of ditch section S05-3 was recorded as {Ditch section unique identifier: S05-3, Risk level: High risk, Difference value: 48.54}.

[0104] The threshold mentioned above was set based on historical scour data from 100 different gully segments within the study area and... Statistical analysis of the calculated values; therefore, setting the scour risk assessment threshold to 15 refers to providing a specific analytical process. Specifically, remote sensing images and ground survey data of 100 gully segments within the study area over the past five years (2020-2024) were collected, and the actual scour situation was divided into two categories: "no obvious scour" and "obvious scour" as the true labels. The scour risk threshold for these 100 gully segments was calculated. The values ​​were analyzed using receiver operating characteristic (ROC) curve analysis. The values ​​are predictor variables, with whether or not scouring occurs as the outcome variable. The analysis results show that when... When the value is 15, the area under the ROC curve (AUC) reaches its maximum value of 0.91, and the Youden's index also reaches its maximum value, indicating that this point is the optimal critical point for distinguishing between scour and non-scour. Therefore, 15 is selected as the threshold for judging scour risk.

[0105] Acquire the terrace contour data (vector format) of the region, and perform GIS overlay analysis on the vector boundaries of all spatial areas with medium and high risk levels and the terrace contour data to identify the terraces located within these risk areas. At the lowest points or historical confluence points of these terraces, discrete points are established to initially generate a discrete point set containing 350 points.

[0106] The generated tree-like drainage network topology and the discrete set of 350 points are used. For each point, the shortest spatial distance to all main flow lines in the tree-like drainage network topology is calculated. The distance threshold is set to 10 meters. If the spatial distance from a point to a main flow line is less than 10 meters and its elevation is lower than the average elevation of the terrace, the point is classified as an access point; conversely, if the spatial distance from a point to a main flow line is greater than or equal to 10 meters and its elevation is at the lowest point of the terrace, the point is classified as a drainage outlet. After this classification process, a drainage outlet set is finally established, containing 210 drainage outlets, their geographical coordinates, and the terrace number to which they belong; at the same time, an access point set is established, containing 140 access points, their geographical coordinates, and the ditch number to which they belong.

[0107] Please see Figure 1 and Figure 3 S2: Based on the scour risk quantification value, compare the reinforcement and reconstruction costs, optimize the tree-like drainage network topology to generate the drainage network topology, call the set of outlets and the set of access points, use the Hungarian algorithm to solve the connection cost, and generate the access scheme;

[0108] The drainage network topology specifically includes the channel connection relationships, node types, and channel protection levels.

[0109] The access scheme specifically refers to the matching of the drainage outlet and the access point, as well as the engineering cost of the connecting channel;

[0110] The specific steps to obtain the access solution are as follows:

[0111] S201: Based on the scour risk quantification value, obtain the unit cost of reinforcement of drainage network nodes and the unit cost of flow path reconstruction. Traverse multiple flow path units in the tree-like drainage network topology, calculate the reinforcement cost by combining the scour risk quantification value and the unit cost of reinforcement of drainage network nodes, and retrieve alternative flow paths to calculate the reconstruction cost by combining the unit cost of flow path reconstruction. Based on the numerical comparison results of reinforcement cost and reconstruction cost, establish a node optimization judgment set.

[0112] S202: Call the node optimization decision set, filter the flow path units whose decision result is reconstruction, replace them with alternative flow paths in the tree-like drainage network topology, update the connection relationship and weight parameters between nodes, and generate the drainage network topology.

[0113] S203: Based on the drainage network topology and by calling the set of drainage outlets and the set of access points, calculate the connection path cost between multiple drainage outlets and multiple access points, construct a connection cost matrix, solve the connection cost matrix, obtain the connection pairing with the minimum total cost, and generate an access scheme.

[0114] The steps for establishing the node optimization judgment set are as follows: call the scour risk quantification value, the unit cost of strengthening the drainage network node and the unit cost of flow path reconstruction, and traverse all flow path units in the tree-like drainage network topology.

[0115] For any flow path unit, the required ditch protection standard is determined based on the risk level in its corresponding scour risk quantification value. The reinforcement cost of the flow path unit is calculated by combining the length of the flow path unit and the terrain parameters, and by calling the unit cost of reinforcement of the drainage network node.

[0116] Based on the tree-like drainage network topology, alternative flow paths with the same start and end points as the current flow path unit and within a preset spatial search radius are searched. The length and expected workload of the alternative flow paths are calculated, the unit cost of flow path reconstruction is called, and the reconstruction cost is calculated.

[0117] Compare the reinforcement cost with the reconstruction cost. If the reinforcement cost is greater than the reconstruction cost, the optimization strategy is determined to be reconstruction; otherwise, the optimization strategy is reinforcement.

[0118] The unique identifier of each flow path unit and its corresponding optimization strategy judgment result are structured and paired to generate a node optimization judgment set.

[0119] Based on the scour risk quantification results generated by S1, the unit cost of drainage network node reinforcement and the unit cost of flow path reconfiguration were obtained. These two cost figures were determined after investigating local water and soil conservation engineering material and labor market prices. The unit cost of drainage network node reinforcement was set according to the protection level; for example, for medium-risk ditches, the unit cost of masonry slope protection was 150 yuan / meter, while for high-risk ditches, the unit cost of concrete lining was 300 yuan / meter. The unit cost of flow path reconfiguration, involving both excavation of new ditches and backfilling of old ditches, was set at a comprehensive unit cost of 250 yuan / meter.

[0120] The aforementioned cost data were determined after investigating the local market prices for materials and labor in soil and water conservation projects, providing a detailed investigation and calculation process. For example, in October 2025, price quotes were obtained from three different building materials markets (Building Materials City A, Hardware Market B, and Sand and Gravel Plant C) and two construction teams (Li's Engineering Team and Wang's Construction Team) in the project area. The comprehensive material and labor costs for masonry slope protection were 135 yuan / meter, 155 yuan / meter, 148 yuan / meter, 158 yuan / meter, and 152 yuan / meter, respectively. The arithmetic average of these five sets of quotes was then calculated. The unit cost was determined to be 150 yuan / meter after rounding. Other cost data were also obtained through similar market research and calculation methods.

[0121] Traverse all 137 flow path units in the tree-like drainage network topology. Taking the channel segment S05-3 calculated in S1 as an example, its risk level is high and it requires reinforcement. The measured length of this flow path unit is 80 meters, and the terrain parameters (such as average slope) are 8%. The required channel protection standard is concrete lining. Using the drainage network node reinforcement unit cost of 300 yuan / meter, the reinforcement cost of this flow path unit is calculated as follows: Yuan;

[0122] Next, within the tree-like drainage network topology, an alternative flow path with the same start and end points as S05-3 (starting point at the end of S05-2, ending point at the main channel of M05) is searched. The spatial search radius is set to 100 meters. Within this search radius, based on DEM data, a topographically feasible alternative flow path, S05-3_alt, is calculated. This alternative flow path is 95 meters long, and due to the need to cross a small rocky area, the expected workload is increased compared to the original path. The unit cost of flow path reconstruction is 250 yuan / meter, and the calculated reconstruction cost is: Yuan.

[0123] Comparing reinforcement costs with reconstruction costs: Since the cost of reinforcement is greater than the cost of reconstruction, the optimization strategy for S05-3 is determined to be reconstruction. The unique identifier "S05-3" of this flow path unit and its corresponding optimization strategy determination result "reconstruction" are then structurally paired. This process is repeated for all 137 flow path units to finally generate a node optimization determination set. Specific examples are shown in Table 3.

[0124] Table 3 Example of Node Optimization Decision Set

[0125] ;

[0126] As shown in Table 3, this table records the cost comparison of some flow path units and the final optimization decision.

[0127] The node optimization decision set is invoked to filter out all flow path units with a decision result of "reconstruction", such as S08-1. In the original tree-like drainage network topology, the original flow path S08-1 is deleted and replaced with its substitute flow path S08-1_alt. Simultaneously, the connection relationships between nodes related to this flow path are updated, such as updating the orientation of upstream and downstream nodes, and updating its weight parameters in the network model based on parameters such as the length and slope of the new flow path. After completing all reconstruction operations, the final drainage network topology is generated. This topology clarifies the connection relationships of all channels, node types (ordinary nodes, confluence nodes, and outlet nodes), and the protection level of each channel segment determined based on its final risk status.

[0128] Based on the optimized drainage network topology, and using the set of 210 outlets and 140 access points generated in S1, the connection path cost between each outlet and each access point is calculated. The connection path cost is determined by the channel length, terrain slope, and land use type. For example, the shortest planned path from outlet W_021 to access point A_058 is 55 meters, with 30 meters crossing woodland and 25 meters crossing grassland. The unit construction cost for woodland is 80 yuan / meter, and for grassland it is 60 yuan / meter. Therefore, the connection path cost is... Yuan. In this way, the connection cost between all 210 drainage outlets and 140 access points is calculated, and a 210×140 connection cost matrix is ​​constructed.

[0129] To solve the connection cost matrix and find the connection pairing scheme with the minimum total cost, the Hungarian algorithm is executed. This process first transforms the cost matrix into a square matrix (by adding virtual nodes and high-cost values), then performs row and column reduction on the matrix to find augmenting paths and adjust the matching until an optimal match is found. This optimal match is the unique pairing of 140 outlets with 140 access points, minimizing the total cost of the connection channel engineering. The remaining 70 outlets are not connected to channels in this round of planning. The final access scheme is generated, which clarifies the matching pairs of the 140 outlets and access points, and the total channel engineering cost required to connect these pairs is 856,300 yuan.

[0130] Please see Figure 1 and Figure 4 S3: Identify key monitoring nodes using the drainage network topology, plan the deployment path of monitoring equipment according to the access scheme, calculate the peak flow and sediment carrying capacity of key monitoring nodes, and determine the operating benchmark threshold.

[0131] The operational benchmark thresholds include the design peak flow threshold and the theoretical sediment carrying capacity threshold;

[0132] The specific steps for obtaining the benchmark threshold are as follows:

[0133] S301: Calculate the network centrality parameter and upstream confluence contribution parameter using multiple network nodes in the drainage network topology, perform a weighted summation on the network centrality parameter and upstream confluence contribution parameter to generate a node criticality score, set a node screening score threshold, retrieve the node criticality score and extract network nodes that exceed the node screening score threshold, and generate a set of key monitoring nodes.

[0134] S302: Based on the access scheme, call the set of key monitoring nodes, obtain the spatial coordinates of multiple key monitoring nodes and the terrain connectivity data between nodes, calculate the total path cost of feasible deployment sequences that traverse all key monitoring nodes based on the spatial coordinates and terrain connectivity data, select the deployment sequence with the minimum total path cost, and generate the deployment path of the monitoring equipment.

[0135] S303: For each key monitoring node in the set of key monitoring nodes, obtain elevation model data and soil distribution data of the control watershed area, calculate its peak flow rate in combination with the rainfall intensity, calculate the sediment carrying capacity, set multi-level response values ​​based on the calculation results of peak flow rate and sediment carrying capacity, and establish an operational benchmark threshold.

[0136] The specific steps for generating node criticality scores are as follows: call the drainage network topology and traverse all network nodes within it;

[0137] For each network node, calculate its betweenness centrality in the drainage network topology to obtain the original betweenness centrality parameter, and calculate the sum of the catchment areas of all channel segments upstream of the node to obtain the original upstream catchment contribution parameter.

[0138] Normalize the original betweenness centrality parameter and the original upstream confluence contribution parameter of all network nodes to obtain the network centrality parameter and the upstream confluence contribution parameter.

[0139] Node criticality score Through the formula:

[0140] calculate;

[0141] in, The node criticality score represents the node's performance in the network. The weighting coefficients represent the network centrality used to adjust for the influence of network structure importance. The weighting coefficient representing the upstream confluence contribution used to adjust for the importance of the hydrological response. The parameter representing the original betweenness centrality of the current node. and These represent the maximum and minimum original betweenness centrality parameters in the entire drainage network topology, respectively. The parameter representing the original upstream confluence contribution of the current node. and These represent the maximum and minimum original upstream confluence contribution parameters in the entire drainage network topology, respectively.

[0142] The calculated node criticality score for each network node Associate it with its unique node identifier to generate a node criticality score.

[0143] The final drainage network topology generated using S2 consists of 358 nodes. To identify key monitoring nodes, the network centrality parameter and upstream runoff contribution parameter for each node need to be calculated. Betweenness centrality is used as the network centrality parameter, which measures the degree to which a node acts as a "bridge" in the network. The upstream runoff contribution parameter directly uses the upstream runoff area of ​​the node.

[0144] For all 358 network nodes, calculate their original betweenness centrality parameters. For example, node N128 is located at the confluence of two main tributaries, and many shortest paths from upstream to downstream exits in the network pass through it; therefore, its original betweenness centrality parameter is calculated. The value is 2540.5. Simultaneously, the total catchment area of ​​all upstream channel segments is calculated to obtain the original upstream catchment contribution parameter. The area is 0.45 square kilometers. After traversing all nodes, the parameter of the maximum original betweenness centrality in the entire network topology is obtained. The minimum is 3100.0. =0; Maximum original upstream confluence contribution parameter It is 1.5 square kilometers, the smallest It is 0.0025 square kilometers (the area of ​​a single grid cell).

[0145] Node criticality score The calculation formula is:

[0146] ;

[0147] in, The node criticality score represents the node's performance in the network. The weighting coefficients represent the network centrality used to adjust for the influence of network structure importance. The weighting coefficient representing the upstream confluence contribution used to adjust for the importance of the hydrological response is... , The parameter representing the original betweenness centrality of the current node. and These represent the maximum and minimum original betweenness centrality parameters in the entire drainage network topology, respectively. The parameter representing the original upstream confluence contribution of the current node. and These represent the maximum and minimum original upstream confluence contribution parameters in the entire drainage network topology, respectively.

[0148] Assign values ​​and perform calculations to the parameters in the formula:

[0149] Weighting coefficient and The design is based on an analysis of historical soil erosion events in the region. The analysis revealed that scour events triggered by key points in the network structure (such as confluences) accounted for 60% of the total, while runoff scour events triggered by large catchment areas accounted for 40%. To reflect this historical pattern, the design... , .

[0150] The above weighting coefficients and The setup is based on the analysis of historical soil erosion events in the region, and is designed to reflect this historical pattern. , This refers to the provision of a specific analytical process. Specifically, it analyzed archival data from 2020 to 2024 recorded by soil and water conservation stations in the project area, detailing 10 severe scour events that caused significant economic losses. Through on-site investigations and causal analysis reports of the event locations, it was determined that the primary cause of 6 events was concentrated scour at the confluence of main gullies (corresponding to high betweenness centrality nodes), while the primary cause of the other 4 events was strong runoff formed by the confluence of runoff on broad slopes (corresponding to large upstream confluence area nodes). Based on this 6:4 attribution ratio, the weights were set as follows: , This is to enable the scoring model to reflect the main causal characteristics of local scour disasters.

[0151] For node N128, its , square kilometers. , . square kilometers, square kilometers.

[0152] Substitute the above parameter values ​​into the formula:

[0153] ;

[0154] ;

[0155] ;

[0156] ;

[0157] This result indicates that node N128 has a node criticality score of 0.6112.

[0158] The node screening score threshold was set to 0.5. This threshold was determined by evaluating all 358 nodes. The scores were sorted, and the top 10% of nodes were selected. The 36th node had a score of 0.508, so the threshold was set to 0.5. The critical scores of all nodes were retrieved, and 38 network nodes with scores exceeding 0.5 were extracted to generate a set of key monitoring nodes.

[0159] Based on the access scheme generated by S2 and the set of 38 key monitoring nodes, their spatial coordinates and terrain connectivity data between the nodes are obtained. To plan the deployment path of the monitoring equipment, this problem is transformed into a Traveling Salesman Problem (TSP). The total path cost of feasible deployment sequences traversing all 38 key monitoring nodes is calculated, taking into account both distance and elevation gain. The deployment sequence with the minimum total path cost, with a total length of 5.8 kilometers, is selected as the final deployment path for the monitoring equipment using simulated annealing.

[0160] The above-mentioned method of using simulated annealing to select the deployment sequence with the minimum total path cost refers to the following parameters when using simulated annealing: the initial temperature is set to 1000, the temperature decay coefficient (cooling rate) is set to 0.99, the number of internal loop iterations (Markov chain length) at each temperature is set to 100, and the algorithm's termination condition is set to the point where the value of the optimal solution does not improve in 50 consecutive external loop iterations.

[0161] For each node in the key monitoring node set, such as N128, elevation model data and soil distribution data of the 0.45 square kilometer watershed it controls were acquired. Combined with the rainfall intensity (120 mm / h) over a 20-year return period, the design peak flow was recalculated, yielding a value of 3.5 cubic meters per second. Simultaneously, based on the soil type and slope information of the watershed, its theoretical sediment-carrying capacity was calculated, yielding a value of 8.8 kilograms per second.

[0162] The theoretical sediment-carrying capacity calculated based on the soil type and slope information of the watershed, resulting in a value of 8.8 kg / s, refers to the use of an empirical formula applicable to the local area. Perform the calculations. Among them, Represents sand-carrying capacity (unit: kg / s). Represents peak flow rate (unit: cubic meters per second). Represents the average slope of the ditch (dimensionless). This represents the watershed sediment yield coefficient related to soil type. For the watershed controlled by node N128, which is mainly composed of loess and loess, it was calibrated based on historical measured data. The value is set to 15. This represents the peak flow rate of this node. cubic meters per second, average slope of the ditch Substituting into the formula, we can calculate... kilograms per second.

[0163] Based on the calculation results, multi-level response values ​​are set to establish operational baseline thresholds. Specific examples are shown in Table 4.

[0164] Table 4. Operating Baseline Thresholds for Key Monitoring Node N128

[0165] ;

[0166] As shown in Table 4, this table establishes graded turbidity warning thresholds for node N128 under different flow levels, which constitute its operating benchmark threshold system.

[0167] Please see Figure 1 and Figure 5 S4: Collect turbidity and flow data at key monitoring nodes and compare them with the operating baseline threshold. If the turbidity data exceeds the operating baseline threshold, soil erosion is judged and an erosion status indicator is generated.

[0168] The erosion status indicators include the location of the warning node, the extent of turbidity exceeding the standard, and the warning timestamp;

[0169] The specific steps for obtaining the erosion status indicator are as follows:

[0170] S401: Based on key monitoring nodes, deploy monitoring equipment and set data acquisition frequency parameters and data upload cycle. The monitoring equipment synchronously collects turbidity data and flow data according to the data acquisition frequency parameters, encodes the collected turbidity data and flow data into independent data frames, aggregates multiple sets of data frames at the end of the data upload cycle, and adds node identifier and timestamp information to each data frame to obtain the real-time data sequence of the monitoring node.

[0171] S402: Receives real-time data sequences from monitoring nodes, parses data frames to extract turbidity data, flow data, node identifiers and timestamp information, calls the multi-level turbidity response threshold set for the differential flow data value range in the running benchmark threshold, sequentially iterates and compares the current turbidity data value with the multi-level turbidity response threshold corresponding to the current flow data range, and obtains the erosion event judgment result.

[0172] S403: Based on the erosion event judgment result, if the judgment result indicates that the turbidity data value exceeds any level of turbidity response threshold, then extract the node identifier that triggered the judgment, the timestamp information and the threshold level parameter that was exceeded, and connect the three in a structured manner. The turbidity exceedance is characterized by the threshold level parameter that was exceeded, and an erosion status identifier is established.

[0173] The specific steps for obtaining the erosion event determination result are as follows: call the running baseline threshold, receive the real-time data sequence of the monitoring node, and parse to obtain the current turbidity data and current flow data under the target timestamp;

[0174] Based on the current traffic data value, match and search for the traffic data value range to which it belongs in the operating baseline threshold;

[0175] Obtain the multi-level turbidity response threshold corresponding to the numerical range of the flow data. The multi-level turbidity response threshold includes multiple incremental thresholds from level one to multiple levels.

[0176] The current turbidity data is compared with the first-level turbidity response threshold in the multi-level turbidity response threshold. If the current turbidity data is greater than the first-level turbidity response threshold, it is determined that an erosion event has occurred, and the threshold level parameter that has been exceeded is recorded as the first level.

[0177] If it is less than or equal to, continue to compare the current turbidity data with the secondary turbidity response threshold, iterate until the highest threshold or the judgment event occurs, and record the threshold level parameter that is exceeded.

[0178] If the current turbidity data does not exceed the turbidity response threshold for all levels, it is determined that no erosion event has occurred, and an erosion event determination result is generated.

[0179] Based on the key monitoring node set and deployment path generated by S3, monitoring devices integrating flow meters and turbidity meters were deployed at 38 key monitoring nodes, including N128. The data acquisition frequency was set to once every 5 minutes, and the data upload cycle was once every 30 minutes. Following the set frequency, at 3:00:00 PM on October 11, 2025, the monitoring devices simultaneously collected turbidity data of 1350 NTU and flow rate data of 2.8 cubic meters per second. The devices encoded these two sets of data into independent data frames, attaching the node identifier "N128" and the timestamp "2025-10-11 15:00:00". At 3:30:00 PM, at the end of the data upload cycle, the six sets of data frames collected every 5 minutes were aggregated and uploaded to the data center via a wireless network, forming a real-time data sequence for the monitoring nodes.

[0180] The data center receives a real-time data sequence from node N128 and parses the data frame with the timestamp "15:00:00". It extracts the turbidity data as 1350 NTU, the flow rate data as 2.8 cubic meters per second, the node identifier as "N128", and the timestamp information as "2025-10-11 15:00:00".

[0181] The operating baseline threshold established for node N128 in S3 is invoked (see Table 4). Based on the current flow rate of 2.8 cubic meters per second, the flow rate range corresponding to this value is searched within the operating baseline threshold. Since 2.8 is greater than 2.5, a range ">2.5" is matched. The corresponding multi-level turbidity response thresholds are obtained: Level 1: 1200 NTU, Level 2: 2000 NTU, and Level 3: 3500 NTU.

[0182] The current turbidity data value of 1350 NTU is compared sequentially with the multi-level turbidity response thresholds for this range. First, 1350 NTU is compared with the first-level turbidity response threshold of 1200 NTU. Since 1350 > 1200, an erosion event is determined to have occurred, and the exceeded threshold level parameter is recorded as level one. Because the level one threshold has been triggered, the comparison stops, and comparisons with level two and level three thresholds are no longer performed. Finally, the erosion event determination result is: a level one erosion event has occurred.

[0183] Based on the erosion event determination result of "Level 1 erosion event occurred", the node identifier "N128" that triggered the determination, the timestamp information "2025-10-11 15:00:00", and the threshold level parameter "Level 1" that was exceeded are extracted. These three are then concatenated in a structured manner to generate an erosion status identifier. The extent of turbidity exceeding the standard is characterized by the threshold level parameter "Level 1". The final erosion status identifier is {Warning node location: N128, Turbidity exceedance range: Level 1, Warning timestamp: 2025-10-11 15:00:00}.

[0184] The above embodiments illustrate preferred embodiments of the present invention. Any equivalent adjustments to the technical solution based on software engineering methods are within the scope of protection, including but not limited to: implementing algorithm logic using different programming languages, refactoring functional modules into services, adjusting data interaction protocols, and optimizing resource scheduling strategies. Any implementation scheme derived from reasonable modifications to the data processing flow, service call chain, or system architecture layer without departing from the core technology of the present invention should be considered within the scope of protection defined by the claims of the present invention.

Claims

1. A method for intelligent planning of water and soil conservation measures for sloping farmland, characterized in that, The method comprises the following steps: S1: obtaining elevation model data, soil distribution data and storm intensity, calculating the elevation model data to generate a tree-shaped drainage network topology, calculating flood peak flow, comparing the soil distribution data's anti-impact ability to generate a quantitative value of erosion risk, obtaining terrace contour data, and arranging discrete points to generate a set of drainage outlets and a set of access points; S2: comparing reinforcement and reconstruction costs based on the quantitative value of erosion risk, optimizing the tree-shaped drainage network topology to generate a drainage network topology, calling the set of drainage outlets and the set of access points, and using the Hungarian algorithm to solve the connection cost to generate an access scheme; S3: identifying key monitoring nodes using the drainage network topology, planning a deployment path of monitoring equipment according to the access scheme, calculating the flood peak flow and sediment carrying capacity of the key monitoring nodes, and determining an operating reference threshold; S4: collecting turbidity data and flow data at the key monitoring nodes, and comparing them with the operating reference threshold, if the turbidity data exceeds the operating reference threshold, judging soil erosion and generating an erosion state identifier; The obtaining step of the set of drainage outlets and the set of access points specifically comprises: S101: obtaining elevation model data, soil distribution data and storm intensity, calculating the tree-shaped drainage network topology based on the topographic information of the elevation model data, combining the storm intensity and the tree-shaped drainage network topology to calculate the flood peak flow, extracting the anti-impact ability parameters of multiple soil types in the soil distribution data and comparing them with the flood peak flow to generate a quantitative value of erosion risk; S102: obtaining terrace contour data, setting a risk of erosion threshold for the quantitative value of erosion risk, searching for a spatial region whose quantitative value of erosion risk exceeds the risk of erosion threshold, and performing overlay analysis on the spatial region vector boundary and the terrace contour data to obtain a set of discrete distribution points; S103: calling the tree-shaped drainage network topology and the set of discrete distribution points, calculating the spatial distance from multiple points in the set of discrete distribution points to the main stream line of the tree-shaped drainage network topology, classifying the multiple points according to a preset distance judgment value and point elevation information, and establishing a set of drainage outlets and a set of access points; The obtaining step of the operating reference threshold specifically comprises: S301: calculating network centrality parameters and upstream convergence contribution parameters of multiple network nodes in the drainage network topology, performing weighted summation on the network centrality parameters and the upstream convergence contribution parameters to generate node criticality scores, setting a node screening score threshold, searching for network nodes whose node criticality scores exceed the node screening score threshold, and generating a set of key monitoring nodes; S302: according to the access scheme, calling the set of key monitoring nodes, obtaining spatial position coordinates of multiple key monitoring nodes and node-to-node topographic connectivity data, based on the spatial position coordinates and the topographic connectivity data, calculating the path total cost of all possible deployment sequences of the key monitoring nodes, screening a deployment sequence with the smallest path total cost, and generating a monitoring equipment deployment path; S303: For each key monitoring node in the set of key monitoring nodes, obtain the elevation model data and soil distribution data in the control flow domain range, calculate the flood peak flow by combining the rainstorm intensity, and calculate the sediment carrying capacity. According to the calculation results of the flood peak flow and the sediment carrying capacity, a multi-level response value is set, and a running benchmark threshold is established; The generation step of the erosion risk quantitative value is specifically: Obtain the elevation model data, the soil distribution data, and the upstream catchment area of multiple channel segments in the tree-shaped drainage network topology structure, and call the flood peak flow; For the soil distribution data, extract the soil starting flow parameters and erosion resistance parameters corresponding to multiple soil types; a quantitative value of a risk of erosion for each channel segment in the tree-like drainage network topology by the formula: computing; wherein, a scour risk quantification value representative of the channel segment, a channel morphology correction factor representative of the combination of the channel sinuosity and the roughness, a design peak flow representative of the current channel segment, a soil initiation flow parameter representative of the location where the current channel segment is located, an upstream catchment area representative of the channel segment, a soil type erosion resistance parameter representative of the location where the channel segment is located. According to the calculated erosion risk quantification value According to the calculated erosion risk quantification value According to the calculated erosion risk quantification value According to the calculated erosion risk quantification value According to the calculated erosion risk quantification value According to the calculated erosion risk quantification value According to the calculated erosion risk quantification value According to the calculated erosion risk quantification value According to the calculated erosion risk quantification value According to the calculated erosion 2. The method of claim 1, wherein, The erosion risk quantitative value includes a unique identifier of a channel segment, a risk level, and a difference value. The set of drainage outlets includes the geographic coordinates of the drainage outlets and the corresponding terrace numbers. The set of access points includes the geographic coordinates of the access points and the corresponding channel numbers. The drainage network topology includes channel connection relationships, node types, and channel protection levels. The access scheme specifically refers to the matching pairs and connection channel engineering costs of drainage outlets and access points. The running benchmark threshold includes a design flood peak flow threshold and a theoretical sediment carrying capacity threshold. The erosion state identifier includes the location of the warning node, the turbidity exceeding standard amplitude, and the warning timestamp.

3. The method of claim 1, wherein, The access scheme acquisition step is specifically: S201: Based on the erosion risk quantitative value, obtain the drainage network node reinforcement unit cost and the flow path reconstruction unit cost. Traverse multiple flow path units in the tree-shaped drainage network topology structure. Calculate the reinforcement cost by combining the erosion risk quantitative value and the drainage network node reinforcement unit cost. Retrieve the alternative flow path and calculate the reconstruction cost by combining the flow path reconstruction unit cost. According to the numerical comparison results of the reinforcement cost and the reconstruction cost, a node optimization judgment set is established; S202: Call the node optimization judgment set to filter the flow path units with reconstruction judgment results. Replace the alternative flow path in the tree-shaped drainage network topology structure, update the connection relationship and weight parameter between nodes, and generate a drainage network topology; S203: According to the drainage network topology, call the set of drainage outlets and the set of access points, calculate the connection path cost between multiple drainage outlets and multiple access points, construct a connection cost matrix, solve the connection cost matrix, and obtain the connection pairing with the minimum total cost to generate an access scheme.

4. The method of claim 1, wherein, The erosion state identifier acquisition step is specifically: S401: Based on the key monitoring node, deploy a monitoring device and set data acquisition frequency parameters and data upload periods. The monitoring device synchronously acquires turbidity data and flow data according to the data acquisition frequency parameters, encodes the acquired turbidity data and flow data into independent data frames, aggregates multiple data frames at the end of the data upload period, adds a node identifier and a timestamp to each data frame, and acquires a monitoring node real-time data sequence; S402: receiving the monitoring node real-time data sequence, parsing the data frame to extract the turbidity data, the flow data, the node identifier and the timestamp information, calling the multi-level turbidity response threshold value set in the running reference threshold value for the differentiated flow data value interval, sequentially iterating the current turbidity data value and the multi-level turbidity response threshold value of the current flow data corresponding interval, and obtaining an erosion event determination result; S403: according to the erosion event determination result, if the determination result indicates that the turbidity data value exceeds any level of turbidity response threshold value, the node identifier, the timestamp information and the threshold value grade parameter of the exceeding are extracted, which are structured in series, wherein the turbidity exceeding amplitude is represented by the threshold value grade parameter of the exceeding, and an erosion state identifier is established. 5.The intelligent planning method for slope farmland soil and water conservation measure according to claim 3, characterized in that, The establishment step of the node optimization determination set is specifically: calling the scour risk quantization value, the drainage network node reinforcement unit cost and the flow path reconstruction unit cost, and traversing all flow path units in the tree-shaped drainage network topology structure; for any flow path unit, according to the risk level in the corresponding scour risk quantization value, the required channel protection standard is determined, combined with the length and terrain parameters of the flow path unit, and the drainage network node reinforcement unit cost is called to calculate the reinforcement cost of the flow path unit; based on the tree-shaped drainage network topology structure, searching for an alternative flow path with the same starting and ending points as the current flow path unit within a preset spatial search radius, calculating the length and expected engineering quantity of the alternative flow path, and calling the flow path reconstruction unit cost to calculate the reconstruction cost; comparing the reinforcement cost and the reconstruction cost, if the reinforcement cost is greater than the reconstruction cost, the optimization strategy is determined as reconstruction, otherwise as reinforcement; structurally pairing the unique identifier of each flow path unit and its corresponding optimization strategy determination result to generate a node optimization determination set. 6.The intelligent planning method of slope farmland soil and water conservation measure according to claim 1, characterized in that, The generation step of the node criticality score is specifically: calling the drainage network topology and traversing all network nodes in it; for each network node, calculating its betweenness centrality in the drainage network topology to obtain an original betweenness centrality parameter, and calculating the total convergence area of all channel segments upstream of the node to obtain an original upstream convergence contribution degree parameter; performing normalization processing on the original betweenness centrality parameter and the original upstream convergence contribution degree parameter of all network nodes respectively to obtain the network centrality parameter and the upstream convergence contribution degree parameter; The node criticality score By the formula: computing; wherein, a node criticality score representing a network node, a weight coefficient representing a network centrality for adjusting the importance influence of the network structure, a weight coefficient representing an upstream catchment contribution for adjusting the importance influence of the hydrological response, a raw betweenness centrality parameter of a current node, a maximum and a minimum raw betweenness centrality parameter, respectively, a maximum and a minimum raw upstream catchment contribution parameter, respectively, a raw upstream catchment contribution parameter of a current node, a maximum and a minimum raw upstream catchment contribution parameter, respectively, a maximum and a minimum raw upstream catchment contribution parameter, respectively, The node criticality score of each network node is calculated The node criticality score is generated in association with its node unique identification. 7.The intelligent planning method of slope farmland soil and water conservation measure according to claim 6, characterized in that, The erosion event determination result obtaining step is specifically: calling the running reference threshold value and receiving the monitoring node real-time data sequence to parse the current turbidity data and the current flow data under the target timestamp; according to the value of the current flow data, matching and searching its belonging flow data value interval in the running reference threshold value; obtaining the multi-level turbidity response threshold value corresponding to the flow data value interval, the multi-level turbidity response threshold value including multiple increasing threshold values from level one to level many; comparing the current turbidity data with a first turbidity response threshold value in the multi-level turbidity response threshold value, if the current turbidity data is greater than the first turbidity response threshold value, determining that an erosion event occurs, and recording the threshold level parameter that is exceeded as the first level; if less than or equal to, continuing to compare the current turbidity data with a second turbidity response threshold value, and iteratively comparing in turn until the highest level threshold value or determining that an event occurs, and recording the threshold level parameter that is exceeded; if the current turbidity data does not exceed the turbidity response threshold value of all levels, determining that an erosion event does not occur, and generating an erosion event determination result.

Images