Power transmission and transformation project water and soil loss dynamic monitoring and early warning method and platform

Abstract

The invention discloses a power transmission and transformation project water and soil loss dynamic monitoring and early warning method and platform, and relates to the field related to ecological risk alarm, and the method comprises the steps: carrying out the arrangement of monitoring point positions based on the structure distribution data of a target power transmission and transformation project, obtaining N project monitoring point positions, and deploying ground sensor networks on the monitoring point positions in sequence; collecting water and soil associated data of N point location areas, obtaining an unmanned aerial vehicle remote sensing change image according to a preset collection frequency, mapping the water and soil associated data of the N point location areas to the unmanned aerial vehicle remote sensing change image for dynamic twinborn simulation, and establishing a power transmission and transformation project water and soil loss twinborn body; performing synchronous coupling prediction on the real-time water and soil associated data of the N point locations, and outputting water and soil loss parameters of the power transmission and transformation project; and carrying out dynamic graded early warning by adopting a graded early warning mechanism. The technical problems of early warning lag and low accuracy existing in existing power transmission and transformation project water and soil loss monitoring and early warning are solved, and the technical effect of improving the early warning accuracy and timeliness is achieved.

Description

Technical Field

[0001] This application relates to the field of ecological risk alarm, and in particular to a method and platform for dynamic monitoring and early warning of soil and water loss in power transmission and transformation projects. Background Technology

[0002] Power transmission and transformation projects alter surface vegetation and soil structure during construction and operation, leading to soil erosion. This not only threatens the safe and stable operation of power grid facilities but also has long-term negative impacts on the surrounding ecological environment. Currently, the industry mainly relies on periodic manual inspections combined with static soil and water conservation facility checks, as well as post-event analysis using data collected from limited, isolated monitoring points. However, manual inspections are time-consuming and have limited coverage, making it difficult to achieve all-weather monitoring; the deployment of static monitoring points lacks scientific assessment of the risk differences in the project area, resulting in insufficient representativeness of the monitoring network.

[0003] At present, the monitoring and early warning of soil erosion in power transmission and transformation projects suffers from technical problems such as delayed early warning and low accuracy. Summary of the Invention

[0004] This application provides a method and platform for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects. It employs a ground-based sensor network with N monitoring points deployed based on the target project's structural distribution data. Sensors collect soil-water correlation data at these points, acquire UAV remote sensing images, and use dynamic twin simulation to establish a soil-water correlation twin. Based on this twin, real-time soil-water correlation data is coupled and predicted to output soil-water correlation parameters. A tiered early warning mechanism is constructed, and parameters are dynamically graded for early warning. These technical means solve the technical problems of delayed and inaccurate early warning in existing monitoring and early warning systems for soil erosion in power transmission and transformation projects, achieving the technical effect of improving the accuracy and timeliness of early warning.

[0005] This application provides a method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects, comprising: deploying monitoring points based on the structural distribution data of the target power transmission and transformation project to obtain N monitoring points; sequentially deploying a ground sensor network at the N monitoring points; collecting soil-water correlation data of the N monitoring points through the ground sensor network, and simultaneously acquiring UAV remote sensing change images at a preset acquisition frequency; mapping the soil-water correlation data of the N monitoring points to the UAV remote sensing change images to perform dynamic twin simulation and establish a soil-water erosion twin for the power transmission and transformation project; synchronously coupling and predicting the real-time soil-water correlation data of the N monitoring points based on the soil-water erosion twin for the power transmission and transformation project, and outputting soil-water erosion parameters for the power transmission and transformation project; constructing a hierarchical early warning mechanism, and using the hierarchical early warning mechanism to perform dynamic hierarchical early warning for the soil-water erosion parameters of the power transmission and transformation project.

[0006] In a possible implementation, N engineering monitoring points are obtained, and the following processing is performed: risk factors are extracted and weighted for the target power transmission and transformation project to obtain a set of risk factors related to soil erosion and a set of risk factor weight coefficients; the structural distribution data of the target power transmission and transformation project is risk-weighted according to the set of risk factors related to soil erosion and the set of risk factor weight coefficients to obtain a regional risk score set for the power transmission and transformation project; risk areas are divided into risk regions based on the regional risk score set for the power transmission and transformation project to determine a set of graded power transmission and transformation risk regions; monitoring points are deployed in the set of graded power transmission and transformation risk regions to obtain N engineering monitoring points.

[0007] In a possible implementation, monitoring points are deployed for the graded power transmission and transformation risk area set to obtain N engineering monitoring points. The following processes are then performed: monitoring point deployment rules are obtained, including representativeness of water loss risk, regional coverage, and monitoring stability; monitoring points are analyzed for the graded power transmission and transformation risk area set according to the monitoring point deployment rules to obtain an initial graded regional monitoring point set; redundant point distance thresholds are set according to the water and soil loss monitoring requirements of the power transmission and transformation project; and the initial graded regional monitoring point set is merged and optimized based on the redundant point distance thresholds to obtain the N engineering monitoring points.

[0008] In one possible implementation, a soil and water erosion twin of the power transmission and transformation project is established, and the following processing is performed: 3D modeling is performed based on the structural distribution data of the target power transmission and transformation project to generate a basic 3D model of the power transmission and transformation project; the UAV remote sensing change image is arranged temporally and soil and water change is modeled to generate a global regional soil and water change model; the global regional soil and water change model is matched and projected with the basic 3D model of the power transmission and transformation project to obtain a soil and water change model of the power transmission and transformation project; the soil and water correlation data of the N points are mapped to the soil and water change model of the power transmission and transformation project for dynamic twin simulation to establish a soil and water erosion twin of the power transmission and transformation project.

[0009] In a possible implementation, a global regional soil and water change model is generated, and the following processing is performed: geometric correction and temporal arrangement processing are performed on the UAV remote sensing change images to obtain UAV remote sensing sequence images; semantic segmentation and soil and water classification are performed on the UAV remote sensing sequence images to obtain soil and water sequence semantic images; surface disturbance analysis and erosion intensity classification are performed on the soil and water sequence semantic images to determine regional soil and water loss change data; coordinate alignment and overlay of the regional soil and water loss change data and soil and water evolution modeling are performed to generate the global regional soil and water change model.

[0010] In one possible implementation, the soil and water correlation data of the N locations are mapped to the soil and water change model of the power transmission and transformation project for dynamic twin simulation, establishing a soil and water loss twin of the power transmission and transformation project, and performing the following processing: constructing a soil and water loss prediction task list; performing prediction analysis on the soil and water correlation data of the N locations based on the soil and water loss prediction task list to obtain soil and water loss parameters of the N locations; matching and fusing the soil and water loss parameters of the N locations into the soil and water change model of the power transmission and transformation project for dynamic twin simulation, establishing a soil and water loss twin of the power transmission and transformation project.

[0011] In a possible implementation, soil erosion parameters for N location areas are obtained, and the following processing is performed: A historical soil erosion dataset of power transmission and transformation projects is collected; each prediction task in the soil erosion prediction task list is matched and identified with the historical soil erosion dataset of power transmission and transformation projects to obtain a soil erosion sample set associated with the prediction tasks; based on the soil erosion sample set associated with the prediction tasks, task prediction training, validation, optimization, and updates are performed to generate a set of soil erosion branch predictors; the set of soil erosion branch predictors is used to predict and analyze the soil erosion-related data of the N location areas to obtain the soil erosion parameters for the N location areas.

[0012] In a possible implementation, the following processing is also performed: constructing simulation scenario parameters for soil erosion in power transmission and transformation projects based on the historical soil erosion dataset; and performing simulation verification and iterative parameter optimization on the soil erosion twin of the power transmission and transformation projects based on the simulation scenario parameters.

[0013] In a possible implementation, the aforementioned graded early warning mechanism is used to dynamically grade and warn of soil erosion parameters of the power transmission and transformation project, and the following processing is performed: the aforementioned graded early warning mechanism is used to dynamically grade and evaluate the soil erosion parameters of the power transmission and transformation project to obtain the target early warning level; and the target power transmission and transformation project is given a dynamic early warning response for soil erosion based on the target early warning level.

[0014] This application also provides a dynamic monitoring and early warning platform for soil erosion in power transmission and transformation projects, comprising: a monitoring point deployment module, used to deploy monitoring points based on the structural distribution data of the target power transmission and transformation project, obtaining N monitoring points, and sequentially deploying a ground sensor network on the N monitoring points; a dynamic twin simulation module, used to collect soil-water correlation data of the N points through the ground sensor network, and simultaneously acquire UAV remote sensing change images according to a preset acquisition frequency, mapping the soil-water correlation data of the N points to the UAV remote sensing change images to perform dynamic twin simulation, and establishing a soil-water erosion twin of the power transmission and transformation project; a synchronous coupling prediction module, used to perform synchronous coupling prediction of real-time soil-water correlation data of the N points based on the soil-water erosion twin of the power transmission and transformation project, and output soil-water erosion parameters of the power transmission and transformation project; and a dynamic hierarchical early warning module, used to construct a hierarchical early warning mechanism, and use the hierarchical early warning mechanism to perform dynamic hierarchical early warning of the soil-water erosion parameters of the power transmission and transformation project.

[0015] The proposed method and platform for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects first establishes N monitoring points based on the structural distribution data of the target power transmission and transformation project. A ground-based sensor network is then deployed at each of these N monitoring points. Next, soil and water correlation data for the N monitoring points are collected using this ground-based sensor network. Simultaneously, UAV remote sensing imagery is acquired at a preset acquisition frequency. The soil and water correlation data for the N monitoring points are mapped to the UAV remote sensing imagery for dynamic twin simulation, establishing a soil and water erosion twin for the power transmission and transformation project. Then, based on this soil and water erosion twin, real-time soil and water correlation data for the N monitoring points are synchronously coupled and predicted to output soil and water erosion parameters for the power transmission and transformation project. Finally, a tiered early warning mechanism is constructed, and this mechanism is used to dynamically and hierarchically issue early warnings for the soil and water erosion parameters of the power transmission and transformation project. This achieves the technical effect of improving the accuracy and timeliness of early warnings. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a flowchart illustrating the dynamic monitoring and early warning method for soil erosion in power transmission and transformation projects provided in this application embodiment.

[0018] Figure 2This is a schematic diagram of the structure of the dynamic monitoring and early warning platform for soil and water loss in power transmission and transformation projects provided in this application embodiment.

[0019] Explanation of reference numerals in the attached diagram: 10 for monitoring point deployment module, 20 for dynamic twin simulation module, 30 for synchronous coupling prediction module, and 40 for dynamic hierarchical early warning module. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0021] This application provides a method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects, such as... Figure 1 As shown, the method includes: Step S100: Based on the structural distribution data of the target power transmission and transformation project, the monitoring points are deployed to obtain N monitoring points for the project, and a ground sensor network is deployed on the N monitoring points in sequence.

[0022] Specifically, the structural distribution data of the target power transmission and transformation project refers to the spatial location, geometric shape, and attribute information of all structures in the project, stored in CAD drawings or GIS vector data format. Based on the structural distribution data of the power transmission and transformation project, such as tower base coordinates, transmission line routes, substation boundaries, construction access roads, and earthwork areas, spatial analysis is performed using a Geographic Information System (GIS). Combined with layers such as terrain slope, soil type, and vegetation cover, a feasibility analysis of monitoring point deployment is conducted to determine N specific geographical locations for data collection. A ground sensor network is then deployed at these N geographical locations. This ground sensor network consists of various physical sensors and data acquisition and transmission equipment used to automatically collect environmental parameters, including IoT devices such as soil moisture sensors, soil compaction probes, runoff meters, and rain gauges.

[0023] In one possible implementation, N engineering monitoring points are obtained. Step S100 further includes step S110, which involves extracting and weighting risk factors for the target power transmission and transformation project to obtain a set of risk factors related to soil erosion and a set of risk factor weight coefficients. Specifically, risk factors are extracted from the engineering design and environmental assessment reports. Risk factors are environmental or engineering factors that may lead to or exacerbate soil erosion. Examples include: slope gradient, soil erodibility, vegetation cover, rainfall erosivity, and engineering disturbance intensity. The analytic hierarchy process (AHP) or entropy weight method is used to determine the proportion of each risk factor in the comprehensive risk assessment. For example, a judgment matrix is ​​constructed through expert scoring, and the weight coefficients of each factor are calculated to ultimately form a quantitative list of risk factors and their importance.

[0024] Step S120: A risk-weighted assessment is performed on the structural distribution data of the target power transmission and transformation project according to the set of soil erosion-related risk factors and the set of risk factor weight coefficients, resulting in a risk score set for the power transmission and transformation project area. Specifically, in the GIS platform, each risk factor is created as a raster layer, and a weighted overlay analysis is performed based on its weight. For example, using a raster calculator, the formula is executed: Total Risk Score = (Slope Factor Score × Slope Weight) + (Soil Erosion Factor Score × Soil Weight) + ... Finally, a risk score distribution map of the entire project area is generated, where each raster cell value represents the level of soil erosion risk at that location.

[0025] Step S130: Based on the risk score set of the power transmission and transformation project area, the structural distribution data is divided into risk areas to determine a hierarchical power transmission and transformation risk area set. Specifically, the continuous risk scores are reclassified using the natural breakpoint method or equal interval method in GIS. According to the risk score, the continuous geographic space is divided into several sub-regions with different risk levels. For example, the risk scores are divided into three levels: low-risk area, medium-risk area, and high-risk area. The total risk score corresponding to the low-risk area is 0-30 points, the total risk score corresponding to the medium-risk area is 31-70 points, and the total risk score corresponding to the high-risk area is 71-100 points. After the division, polygonal regions of different risk levels are generated, constituting the hierarchical power transmission and transformation risk area set.

[0026] Step S140 involves deploying monitoring points for the graded power transmission and transformation risk area set, resulting in N engineering monitoring points. Specifically, within the divided risk level areas, the location and number of monitoring points in each area are determined according to preset rules, generating N engineering monitoring points, which are the specific locations where sensors are installed. Higher-risk areas will have denser monitoring points, while lower-risk areas will have relatively sparser points.

[0027] In one possible implementation, monitoring points are deployed for the graded power transmission and transformation risk area set to obtain N engineering monitoring points. Step S140 further includes step S141, obtaining monitoring point deployment rules, which include representativeness of water loss risk, regional coverage, and monitoring stability. Specifically, monitoring point deployment rules are formulated and used as constraints. "Representativeness of water loss risk" means that the monitoring point should accurately reflect the overall soil and water loss characteristics of its risk area, requiring the point to be located in a typical geomorphic unit of that risk level, such as the top, middle, or bottom of a slope. "Regional coverage" means that the spatial distribution of monitoring points should effectively cover the entire target area, assessed by calculating the Thiessen polygon area served by the point to ensure no monitoring blind spots. "Monitoring stability" means that the geographical location and environment of the monitoring point can ensure the long-term stable operation of the sensor, requiring the point to have a stable foundation, be far from flood channels, and have a wireless signal strength higher than a threshold.

[0028] Step S142: Analyze the monitoring points in the graded power transmission and transformation risk area set according to the monitoring point layout rules to obtain an initial graded area monitoring point set. Specifically, spatial analysis is performed using GIS software. For example, within each risk area, representative slope locations are identified based on a digital elevation model, and then, combined with a communication base station coverage map, locations with good signal strength are selected to initially generate a set containing all candidate points.

[0029] Step S143: Based on the soil and water loss monitoring requirements of the power transmission and transformation project, set a redundancy point distance threshold. Specifically, the redundancy point distance threshold is a distance value set to avoid overly dense monitoring points. When the distance between two points is less than this value, one of the points is considered redundant. Based on monitoring accuracy requirements and cost control, set a minimum point spacing, for example, 500 meters. That is, if the distance between two initial points is less than 500 meters, they are considered redundant in terms of monitoring function.

[0030] Step S144: Based on the redundant point distance threshold, the initial hierarchical regional monitoring point set is merged and optimized to obtain the N engineering monitoring points. Specifically, a clustering algorithm or the "fusion" tool in GIS is used for processing. Based on the redundant point distance threshold, points that are too close are clustered into one class. Then, the point with the best overall conditions in this class is selected as the final monitoring point, and other redundant points are deleted, thereby obtaining the optimized N final monitoring points. By reducing the spatially dense monitoring points, the technical effects of saving costs and improving the efficiency of point deployment are achieved.

[0031] Step S200: Collect soil and water correlation data for N locations through the ground sensor network, and simultaneously acquire UAV remote sensing change images according to a preset acquisition frequency. Map the soil and water correlation data for the N locations to the UAV remote sensing change images to perform dynamic twin simulation and establish a soil and water loss twin for the power transmission and transformation project.

[0032] Specifically, ground-based sensor networks are used to collect soil-water correlation data, which is data related to the soil erosion process, such as soil moisture, rainfall, and surface runoff. Simultaneously, sensors mounted on drones are used to capture high-resolution orthophotos and multispectral images of the engineering area at regular time intervals. The soil-water correlation data is then linked to the drone remote sensing change images using timestamps and spatial coordinates. Real-time soil moisture content and other data from the soil-water correlation data are used as attribute information and linked to the digital surface model generated from the drone imagery. This creates a digital twin model in virtual space that changes synchronously with the real-world physical engineering project and reflects the soil and water conditions.

[0033] In one possible implementation, a soil erosion twin of the power transmission and transformation project is established. Step S200 further includes step S210, which involves performing three-dimensional modeling based on the structural distribution data of the target power transmission and transformation project to generate a three-dimensional model of the basic power transmission and transformation project. Specifically, using the building information model design files of the power transmission and transformation project or through oblique photogrammetry technology, a three-dimensional white model or real-scene model containing structures such as tower foundations, poles, lines, and substations is constructed to obtain a three-dimensional model of the basic power transmission and transformation project that reflects the original form and spatial location of the power transmission and transformation project structures.

[0034] Step S220 involves arranging the UAV remote sensing change images temporally and modeling soil and water changes to generate a global regional soil and water change model. Specifically, UAV images acquired at different times are arranged temporally, organizing multiple periods of remote sensing images in chronological order. Soil and water changes are modeled using image differencing and change detection algorithms to identify signs of soil erosion such as changes in surface vegetation cover, expansion of exposed soil areas, and gully development. The rate and intensity of these changes are quantified to obtain a global regional soil and water change model that reflects the spatiotemporal evolution of soil erosion throughout the entire monitoring area.

[0035] Step S230: Match and project the global regional soil and water change model with the three-dimensional model of the basic power transmission and transformation project to obtain the soil and water change model of the power transmission and transformation project. Specifically, the global regional soil and water change model reflecting surface changes is used as a texture or attribute layer and attached to the surface of the three-dimensional model of the basic power transmission and transformation project, so that the three-dimensional model can not only display the engineering structure, but also dynamically display the soil and water loss status of its surroundings and within the station.

[0036] Step S240 involves mapping the soil and water correlation data of the N locations to the soil and water change model of the power transmission and transformation project for dynamic twin simulation, thus establishing a soil and water erosion twin for the power transmission and transformation project. Specifically, real-time soil moisture and other data transmitted from ground sensors are used to generate a continuous field through spatial interpolation, which is then fused with the soil and water change model of the power transmission and transformation project. Simultaneously, a prediction model is used to extrapolate the soil and water conditions at future times, enabling the soil and water erosion twin of the power transmission and transformation project to possess real-time sensing and short-term prediction capabilities.

[0037] In one possible implementation, a global regional soil and water change model is generated. Step S220 further includes step S221, which involves geometric correction and temporal arrangement processing of the UAV remote sensing change images to obtain a UAV remote sensing sequence image. Specifically, ground control point or POS system data is used to perform geometric correction on each UAV image, eliminating geometric distortions caused by factors such as sensors and terrain in the remote sensing images. Images from all periods are then unified to the same coordinate system and sorted by acquisition time to form a time-series dataset.

[0038] Step S222: Based on the UAV remote sensing image sequence, perform semantic segmentation and soil and water classification to obtain a soil and water sequence semantic image. Specifically, semantic segmentation is an image processing technique that assigns each pixel in an image to a specific semantic category. Soil and water classification refers to classifying land surface types into categories related to soil erosion, such as vegetation and bare soil. Specifically, a pre-trained deep learning model, such as U-Net, is used to perform pixel-level classification on each image, automatically identifying and labeling each pixel in the image as a category such as vegetation, bare soil, water body, or building, generating a semantic segmentation map with category labels.

[0039] Step S223 involves performing surface disturbance analysis and erosion intensity classification on the soil and water sequence semantic images to determine regional soil erosion change data. Specifically, surface disturbance analysis is performed by comparing time-series semantic images to calculate the rate of change of exposed soil area and the trend of vegetation index changes, quantifying the changes in land cover and morphology caused by natural or human activities. Combining slope data and using methods such as pixel-based bisection models, erosion intensity is classified, dividing the severity of soil erosion into different levels, such as slight, mild, moderate, strong, and extremely strong erosion.

[0040] Step S224 involves aligning and overlaying the regional soil erosion change data using coordinates and modeling soil and water evolution to generate the global regional soil and water change model. Specifically, in GIS, the raster data of soil erosion changes at different times are aligned and overlaid with the three-dimensional model of the basic power transmission and transformation project, that is, data from different times are matched and overlaid in the same coordinate system. Using algorithms such as spatiotemporal kriging interpolation, a mathematical model is established to simulate the continuous spatial distribution and dynamic temporal evolution of soil erosion characteristics, forming a four-dimensional dynamic model, where the four dimensions include three spatial dimensions and one temporal dimension.

[0041] In one possible implementation, the soil and water correlation data of the N locations are mapped to the soil and water change model of the power transmission and transformation project for dynamic twin simulation, establishing a soil and water loss twin of the power transmission and transformation project. Step S240 further includes step S241, constructing a list of soil and water loss prediction tasks. Specifically, based on management needs, a list of specific indicators to be predicted is determined, such as: soil erosion modulus in the next 24 hours, area change rate of high-risk areas in the next week, and runoff that may be generated by a future rainfall event.

[0042] Step S242: Based on the soil and water loss prediction task list, predictive analysis is performed on the soil and water correlation data of the N location areas to obtain soil and water loss parameters for the N location areas. Specifically, for each prediction task in the soil and water loss prediction task list, the corresponding prediction model is called, such as a time series prediction model or a machine learning regression model. Current and historical soil and water correlation data of the location are input to calculate the predicted value for future time, i.e., the quantitative soil and water loss index, such as soil loss amount.

[0043] Step S243 involves matching and fusing the soil erosion parameters of the N locations into the soil and water change model of the power transmission and transformation project for dynamic twin simulation, thus establishing a soil and water erosion twin for the power transmission and transformation project. Specifically, the predicted parameters of each location are used to generate a predicted distribution map of the entire region through spatial interpolation, and this map is then updated into the soil and water change model of the power transmission and transformation project as the state at a future point in time. In this way, the soil and water erosion twin of the power transmission and transformation project can not only display the current state but also predict the future distribution of soil erosion risks.

[0044] In one possible implementation, after obtaining soil erosion parameters for N location areas, step S242 further includes step S2421: collecting historical soil erosion datasets from power transmission and transformation projects; matching and identifying each prediction task in the soil erosion prediction task list with the historical soil erosion dataset to obtain a prediction task-related soil erosion sample set. Specifically, historical data corresponding to each task in the soil erosion prediction task list is extracted from historical databases or similar project cases and labeled to form a sample set related to each prediction task. The prediction task-related soil erosion sample set is a set of historical data selected and prepared for each prediction task, containing input features and corresponding actual output values.

[0045] Step S2422 involves training, validating, optimizing, and updating the prediction model based on the soil erosion sample set associated with the prediction task, generating a set of soil erosion branch predictors. Specifically, for each prediction task, a dedicated prediction model is trained using its corresponding sample set, allowing the machine learning algorithm to learn the mapping relationship between input features and prediction targets. For example, a random forest or LSTM neural network can be used for training. Model parameters are adjusted through cross-validation, and performance is evaluated using a reserved test set. The model is then adjusted based on the evaluation results to improve its accuracy and generalization ability, ultimately generating an optimal prediction model for each task. The set of all these models constitutes the set of soil erosion branch predictors.

[0046] Step S2423: The soil erosion branch predictor set is used to predict and analyze the soil and water correlation data of the N point areas to obtain the soil erosion parameters of the N point areas. Specifically, the real-time collected soil moisture, rainfall, and other data of the N points are input into the corresponding branch predictors for analysis and prediction to obtain the soil erosion related indicators for each point area. For example, the data is input into the "soil erosion modulus predictor" to output the predicted value of the future soil erosion modulus for each point.

[0047] In one possible implementation, the method further includes step S500, constructing simulation scenario parameters for soil erosion in power transmission and transformation projects based on the historical soil erosion dataset. Specifically, the simulation scenario parameters are a set of input conditions required to reproduce or test specific soil erosion events in a simulation environment. Typical and extreme soil erosion events, such as erosion processes after severe rainstorms, are extracted from the historical soil erosion dataset to construct a series of standardized simulation test scenarios, including rainfall process curves, pre-existing soil moisture, and wind force.

[0048] Step S600: Based on the parameters of the simulated soil erosion scenario of the power transmission and transformation project, the soil erosion twin of the power transmission and transformation project is simulated and verified, and its parameters are iteratively optimized. Specifically, the constructed parameters of typical historical scenarios are input into the soil erosion twin of the power transmission and transformation project, the simulation model is run, and the simulation results are compared with historical records. If the error is large, the parameters of the twin's built-in prediction model are automatically or manually adjusted, and the process is iterated repeatedly until the simulation results match the historical observation data to the required degree, thereby improving the prediction accuracy and reliability of the twin.

[0049] Step S300: Based on the soil and water loss twin of the power transmission and transformation project, synchronously couple and predict the real-time soil and water correlation data of N points, and output the soil and water loss parameters of the power transmission and transformation project.

[0050] Specifically, based on the constructed soil and water loss twin of the power transmission and transformation project, the real-time collected soil and water data are processed, and the data from multiple points are comprehensively analyzed at the same time to predict quantitative indicators for describing the soil and water loss status of the entire project area, such as loss area and loss degree level.

[0051] Step S400: Construct a hierarchical early warning mechanism and use the hierarchical early warning mechanism to dynamically classify and issue early warnings for soil and water loss parameters of the power transmission and transformation project.

[0052] Specifically, the tiered early warning mechanism is a rule-based system that classifies risks into different levels based on preset thresholds and takes corresponding response measures. Threshold ranges for different soil erosion parameters are pre-defined. For example, an erosion modulus below a certain value triggers a blue warning, between two values ​​a yellow warning, and above a certain value a red warning. The predicted parameters output from the S300 steps are compared with these thresholds in real time, automatically triggering the corresponding level of warning.

[0053] In one possible implementation, the aforementioned graded early warning mechanism is used to dynamically grade and warn of soil erosion parameters of the power transmission and transformation project. Step S400 further includes step S410, which uses the aforementioned graded early warning mechanism to dynamically grade and evaluate the soil erosion parameters of the power transmission and transformation project to obtain the target early warning level. Specifically, the early warning engine reads the predicted soil erosion parameters of the power transmission and transformation project, such as soil erosion modulus and the proportion of high-risk area, and matches them with the early warning threshold table stored in the database. For example, if it is predicted that the erosion modulus of a certain area will exceed the "red" threshold in the next 24 hours, the target early warning level of that area will be automatically determined to be "red".

[0054] Step S420: Based on the target warning level, a dynamic early warning response for soil erosion is implemented for the target power transmission and transformation project. Specifically, the early warning system is linked with the information release platform. Once the warning level is determined, a preset response process is automatically executed. For example, a blue warning only requires a notification to be displayed on the platform interface; a yellow warning will send an SMS notification to the inspection personnel; a red warning will automatically trigger an alarm sound and push an App push message and email containing the risk location and situation to relevant personnel, and suggest activating the emergency inspection plan.

[0055] This application's embodiments employ a ground sensor network and deploy N monitoring points based on the target engineering structure distribution data. Sensors are used to collect soil and water correlation data at these points, and UAV remote sensing images are acquired. A dynamic twin simulation is used to establish a soil and water loss twin. Based on this twin, real-time soil and water correlation data is coupled and predicted to output soil and water loss parameters. A tiered early warning mechanism is constructed, and parameters are dynamically graded for early warning. These technical means solve the technical problems of delayed and inaccurate early warning in existing power transmission and transformation engineering soil and water loss monitoring and early warning systems, achieving the technical effect of improving the accuracy and timeliness of early warning.

[0056] In the above text, refer to Figure 1 This paper describes in detail a method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects according to embodiments of the present invention. Next, we will refer to... Figure 2 This invention describes a dynamic monitoring and early warning platform for soil and water loss in power transmission and transformation projects, according to an embodiment of the present invention.

[0057] The dynamic monitoring and early warning platform for soil erosion in power transmission and transformation projects according to embodiments of the present invention addresses the technical problems of delayed and inaccurate early warning in existing monitoring and early warning systems for soil erosion in power transmission and transformation projects, thereby improving the accuracy and timeliness of early warning. The dynamic monitoring and early warning platform for soil erosion in power transmission and transformation projects includes: a monitoring point deployment module 10, a dynamic twin simulation module 20, a synchronous coupling prediction module 30, and a dynamic hierarchical early warning module 40.

[0058] The monitoring point deployment module 10 is used to deploy monitoring points based on the structural distribution data of the target power transmission and transformation project, obtaining N monitoring points, and then deploying a ground sensor network on the N monitoring points in sequence. The dynamic twin simulation module 20 is used to collect soil and water correlation data of the N points through the ground sensor network, and simultaneously acquire UAV remote sensing change images according to a preset acquisition frequency. The soil and water correlation data of the N points are mapped to the UAV remote sensing change images to perform dynamic twin simulation and establish a soil and water loss twin of the power transmission and transformation project. The synchronous coupling prediction module 30 is used to perform synchronous coupling prediction of the real-time soil and water correlation data of the N points based on the soil and water loss twin of the power transmission and transformation project, and output the soil and water loss parameters of the power transmission and transformation project. The dynamic hierarchical early warning module 40 is used to construct a hierarchical early warning mechanism and use the hierarchical early warning mechanism to perform dynamic hierarchical early warning of the soil and water loss parameters of the power transmission and transformation project.

[0059] The detailed description of the specific configuration of the monitoring point layout module 10 is explained as follows: As mentioned above, N engineering monitoring points are obtained. The monitoring point layout module 10 may further include: a risk factor extraction unit for extracting risk factors and assigning weights to the target power transmission and transformation project to obtain a set of water and soil erosion-related risk factors and a set of risk factor weight coefficients; a risk weighting assessment unit for performing risk weighting assessment on the structural distribution data of the target power transmission and transformation project according to the set of water and soil erosion-related risk factors and the set of risk factor weight coefficients to obtain a set of regional risk scores for the power transmission and transformation project; a risk area division unit for dividing the structural distribution data into risk areas based on the set of regional risk scores for the power transmission and transformation project to determine a set of graded power transmission and transformation risk areas; and a monitoring point layout unit for deploying monitoring points in the set of graded power transmission and transformation risk areas to obtain N engineering monitoring points.

[0060] The process involves setting up monitoring points for the graded power transmission and transformation risk area set to obtain N engineering monitoring points. The monitoring point deployment unit may further include: a monitoring point deployment rule acquisition subunit for acquiring monitoring point deployment rules, which include representativeness of water loss risk, regional coverage, and monitoring stability; a monitoring point analysis subunit for analyzing the monitoring points in the graded power transmission and transformation risk area set according to the monitoring point deployment rules to obtain an initial graded regional monitoring point set; a redundant point distance threshold setting subunit for setting redundant point distance thresholds based on the water and soil loss monitoring requirements of the power transmission and transformation project; and a merging and optimization subunit for merging and optimizing the initial graded regional monitoring point set based on the redundant point distance thresholds to obtain the N engineering monitoring points.

[0061] The specific configuration of the dynamic twin simulation module 20 is described in detail below: As mentioned above, to establish a soil and water erosion twin for the power transmission and transformation project, the dynamic twin simulation module 20 may further include: a 3D modeling unit for performing 3D modeling based on the structural distribution data of the target power transmission and transformation project to generate a basic 3D model of the power transmission and transformation project; a soil and water change modeling unit for performing temporal arrangement and soil and water change modeling on the UAV remote sensing change image to generate a global regional soil and water change model; a matching projection unit for matching and projecting the global regional soil and water change model with the basic 3D model of the power transmission and transformation project to obtain a soil and water change model for the power transmission and transformation project; and a soil and water erosion twin establishment unit for mapping the soil and water correlation data of the N points to the soil and water change model of the power transmission and transformation project for dynamic twin simulation to establish a soil and water erosion twin for the power transmission and transformation project.

[0062] The global regional soil and water change model is generated by a soil and water change modeling unit, which may further include: a geometric correction subunit for performing geometric correction and temporal arrangement processing on the UAV remote sensing change images to obtain UAV remote sensing sequence images; a semantic segmentation subunit for performing semantic segmentation and soil and water classification based on the UAV remote sensing sequence images to obtain soil and water sequence semantic images; a surface disturbance analysis subunit for performing surface disturbance analysis and erosion intensity classification on the soil and water sequence semantic images to determine regional soil and water loss change data; and a coordinate alignment and overlay subunit for performing coordinate alignment and overlay and soil and water evolution modeling on the regional soil and water loss change data to generate the global regional soil and water change model.

[0063] Specifically, the soil and water correlation data of the N locations are mapped to the soil and water change model of the power transmission and transformation project for dynamic twin simulation to establish a soil and water erosion twin for the power transmission and transformation project. The unit for establishing the soil and water erosion twin for the power transmission and transformation project may further include: a soil and water erosion prediction task list construction subunit for constructing a soil and water erosion prediction task list; a prediction analysis subunit for performing prediction analysis on the soil and water correlation data of the N locations based on the soil and water erosion prediction task list to obtain soil and water erosion parameters of the N locations; and a parameter matching and fusion subunit for matching and fusing the soil and water erosion parameters of the N locations to the soil and water change model of the power transmission and transformation project for dynamic twin simulation to establish a soil and water erosion twin for the power transmission and transformation project.

[0064] The prediction and analysis subunit, which obtains soil and water loss parameters for N location areas, may further include: a prediction task-related soil and water loss sample set acquisition component for collecting historical power transmission and transformation project soil and water loss datasets, matching and identifying each prediction task in the soil and water loss prediction task list with the historical power transmission and transformation project soil and water loss datasets to obtain a prediction task-related soil and water loss sample set; a soil and water loss branch predictor set generation component for performing task prediction training, verification, optimization, and updating based on the prediction task-related soil and water loss sample set to generate a soil and water loss branch predictor set; and a prediction and analysis component for using the soil and water loss branch predictor set to perform prediction and analysis on the soil and water loss related data for the N location areas to obtain the soil and water loss parameters for the N location areas.

[0065] The system may further include: a parameter construction module for simulating soil erosion in power transmission and transformation projects, used to construct parameters for simulating soil erosion in power transmission and transformation projects based on the historical soil erosion dataset of power transmission and transformation projects; and a simulation verification module used to perform simulation verification and iterative parameter optimization on the soil erosion twin of the power transmission and transformation projects based on the simulation parameters for simulating soil erosion in power transmission and transformation projects.

[0066] The specific configuration of the dynamic hierarchical early warning module 40 is described in detail below: As mentioned above, the hierarchical early warning mechanism is used to dynamically classify and warn of soil erosion parameters of the power transmission and transformation project. The dynamic hierarchical early warning module 40 may further include: a dynamic hierarchical evaluation unit used to dynamically classify and evaluate the soil erosion parameters of the power transmission and transformation project using the hierarchical early warning mechanism to obtain the target early warning level; and a soil erosion dynamic early warning response unit used to perform a soil erosion dynamic early warning response for the target power transmission and transformation project based on the target early warning level.

[0067] The dynamic monitoring and early warning platform for soil and water loss in power transmission and transformation projects provided in this embodiment of the invention can execute the dynamic monitoring and early warning method for soil and water loss in power transmission and transformation projects provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0068] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0069] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects, characterized in that, The method includes: Based on the structural distribution data of the target power transmission and transformation project, monitoring points are deployed to obtain N monitoring points for the project. A ground sensor network is then deployed on the N monitoring points in sequence. The ground sensor network collects soil and water correlation data at N locations, and simultaneously acquires UAV remote sensing change images at a preset acquisition frequency. The soil and water correlation data at the N locations are mapped to the UAV remote sensing change images to perform dynamic twin simulation and establish a soil and water loss twin for the power transmission and transformation project. Based on the aforementioned soil and water loss twin of the power transmission and transformation project, synchronous coupling prediction is performed on the real-time soil and water correlation data of N points to output the soil and water loss parameters of the power transmission and transformation project. A tiered early warning mechanism is constructed, and the mechanism is used to dynamically classify and issue early warnings for soil and water loss parameters of the power transmission and transformation project.

2. The method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects as described in claim 1, characterized in that, N engineering monitoring points were obtained, including: Risk factors are extracted and weights are assigned to the target power transmission and transformation project to obtain a set of risk factors related to soil erosion and a set of risk factor weight coefficients. Based on the aforementioned set of water and soil erosion-related risk factors and the set of risk factor weight coefficients, the structural distribution data of the target power transmission and transformation project are subjected to risk weighted assessment to obtain the regional risk score set of the power transmission and transformation project. Based on the risk scoring set of the power transmission and transformation project area, the structural distribution data is divided into risk areas to determine the graded power transmission and transformation risk area set. Monitoring points are set up for the graded power transmission and transformation risk area set to obtain N engineering monitoring points.

3. The method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects as described in claim 2, characterized in that, Monitoring points were deployed for the aforementioned graded power transmission and transformation risk area set, resulting in N engineering monitoring points, including: The rules for the deployment of monitoring points are obtained, including the representativeness of loss risk, regional coverage, and monitoring stability. According to the monitoring point layout rules, the monitoring point set of the graded power transmission and transformation risk area set is analyzed to obtain the initial graded area monitoring point set. Based on the monitoring needs of soil and water loss in power transmission and transformation projects, redundant point distance thresholds are set. Based on the redundant point distance threshold, the initial hierarchical regional monitoring point set is merged and optimized to obtain the N engineering monitoring points.

4. The method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects as described in claim 1, characterized in that, Establishing a soil and water erosion twin for power transmission and transformation projects, including: Three-dimensional modeling is performed based on the structural distribution data of the target power transmission and transformation project to generate a three-dimensional model of the basic power transmission and transformation project; The UAV remote sensing change images are arranged in a time sequence and soil and water change model is generated to produce a global regional soil and water change model. The global regional soil and water change model is matched and projected with the three-dimensional model of the basic power transmission and transformation project to obtain the soil and water change model of the power transmission and transformation project. The soil and water correlation data of the N points are mapped to the soil and water change model of the power transmission and transformation project for dynamic twin simulation, and a soil and water loss twin of the power transmission and transformation project is established.

5. The method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects as described in claim 4, characterized in that, Generate a global regional soil and water change model, including: Geometric correction and temporal arrangement processing are performed on the UAV remote sensing change images to obtain UAV remote sensing sequence images; Semantic segmentation and soil and water classification are performed on the aforementioned UAV remote sensing sequence images to obtain soil and water sequence semantic images; Surface disturbance analysis and erosion intensity classification are performed on the semantic images of the soil and water sequence to determine the regional soil and water loss change data; The regional soil and water loss change data are coordinate aligned and overlaid, and soil and water evolution modeling is performed to generate the global regional soil and water change model.

6. The method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects as described in claim 4, characterized in that, The soil and water correlation data of the N points are mapped to the soil and water change model of the power transmission and transformation project for dynamic twin simulation, and a soil and water loss twin of the power transmission and transformation project is established, including: Construct a task list for predicting soil erosion; Based on the aforementioned list of soil and water loss prediction tasks, predictive analysis is performed on the soil and water correlation data of the N location areas to obtain soil and water loss parameters for the N location areas. The soil and water loss parameters of the N locations are matched and integrated into the soil and water change model of the power transmission and transformation project to perform dynamic twin simulation and establish a soil and water loss twin of the power transmission and transformation project.

7. The method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects as described in claim 6, characterized in that, We obtained soil and water loss parameters for N location areas, including: Collect a historical soil and water loss dataset of power transmission and transformation projects, and match and identify each prediction task in the soil and water loss prediction task list with the historical soil and water loss dataset of power transmission and transformation projects to obtain a soil and water loss sample set associated with the prediction task. Based on the soil erosion sample set associated with the prediction task, task prediction training, verification, optimization and updating are performed to generate a set of soil erosion branch predictors. The soil and water loss branch predictor set is used to predict and analyze the soil and water correlation data of the N points to obtain the soil and water loss parameters of the N points.

8. The method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects as described in claim 7, characterized in that, The method further includes: Based on the historical soil and water loss dataset of power transmission and transformation projects, parameters for a simulated soil and water loss scenario of power transmission and transformation projects are constructed. Based on the parameters of the simulated soil erosion scenario of the power transmission and transformation project, the soil erosion twin of the power transmission and transformation project is simulated and verified, and the parameters are iteratively optimized.

9. The method for dynamic monitoring and early warning of soil erosion in power transmission and transformation projects as described in claim 1, characterized in that, The aforementioned graded early warning mechanism is used to dynamically grade and issue early warnings for soil erosion parameters of the power transmission and transformation project, including: The aforementioned graded early warning mechanism is used to dynamically grade and evaluate the soil and water loss parameters of the power transmission and transformation project to obtain the target early warning level; Based on the target early warning level, a dynamic early warning response for soil and water loss is conducted for the target power transmission and transformation project.

10. A dynamic monitoring and early warning platform for soil erosion in power transmission and transformation projects, characterized in that, The platform is used to implement the dynamic monitoring and early warning method for soil erosion in power transmission and transformation projects as described in any one of claims 1-9, and the platform includes: The monitoring point deployment module is used to deploy monitoring points based on the structural distribution data of the target power transmission and transformation project, to obtain N project monitoring points, and then deploy a ground sensor network on the N project monitoring points in sequence; The dynamic twin simulation module is used to collect soil and water correlation data of N points in the ground sensor network, and at the same time acquire UAV remote sensing change images according to a preset acquisition frequency. The soil and water correlation data of the N points in the ground sensor network are mapped to the UAV remote sensing change images to perform dynamic twin simulation and establish a soil and water loss twin of the power transmission and transformation project. The synchronous coupling prediction module is used to perform synchronous coupling prediction on real-time soil and water correlation data of N points based on the soil and water loss twin of the power transmission and transformation project, and output the soil and water loss parameters of the power transmission and transformation project. The dynamic hierarchical early warning module is used to construct a hierarchical early warning mechanism and to use the hierarchical early warning mechanism to dynamically classify and warn of soil and water loss parameters of the power transmission and transformation project.

Images