Power transmission project non-standard runoff plot intelligent identification modeling method and system

Abstract

The application provides a power transmission and transformation engineering non-standard runoff plot intelligent identification modeling method and system, relates to the technical field of data processing, and the method comprises the following steps: twin modeling is carried out according to real-time monitoring data sets, non-standard runoff plot identification and division are carried out, water and soil loss coupling characteristics are optimized according to a first runoff plot distribution model, direct and indirect water and soil loss accidents under multiple nodes are deduced for a second runoff plot distribution model according to a power transmission and transformation engineering construction scheme and future meteorological data sets, global disturbance optimization under multiple nodes is carried out for the second runoff plot distribution model according to a direct and indirect loss deduction atlas, a third runoff plot distribution model is obtained, and the technical problems that the existing technology has unreasonable non-standard runoff plot division of power transmission and transformation engineering, resulting in insufficient water and soil loss prevention and control and engineering safety are solved. The technical effects of accurately identifying and dividing the non-standard runoff plot of the power transmission and transformation engineering and improving the water and soil loss prevention and control and the engineering safety are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to an intelligent identification modeling method and system for non-standard runoff areas of power transmission projects. BACKGROUND

[0002] Power transmission projects have the characteristics of long lines, wide areas, complex terrain, strong construction disturbance, etc. They usually cross complex terrain areas such as mountains, hills, and valleys. Construction and construction involve a large number of foundation excavation, tower foundation construction, road opening, and wire erection operation links. Frequent disturbance of the earth's surface changes the topographic structure and runoff path, which can easily lead to changes in the characteristics of surface runoff collection, thereby forming non-standard runoff phenomena, i.e., the runoff formation and confluence process does not follow the traditional hydrological zoning rules, leading to local confluence anomalies, concentrated soil erosion, and increased risk of slope instability.

[0003] In the prior art, the runoff area division method relies on fixed terrain zoning or expert experience, mainly using static indicators such as slope, flow direction, and catchment area, and lacks comprehensive consideration of dynamic construction disturbance and time-varying meteorological factors, which leads to the inability to accurately identify the distribution characteristics and change rules of non-standard runoff areas, resulting in problems such as unreasonable soil and water conservation design and prevention and control measures, and further affecting project safety and ecological stability.

[0004] The prior art has the technical problem of unreasonable division of non-standard runoff areas of power transmission projects, leading to insufficient soil and water loss prevention and control and project safety. SUMMARY

[0005] The purpose of the present application is to provide an intelligent identification modeling method and system for non-standard runoff areas of power transmission projects, which solves the technical problem of unreasonable division of non-standard runoff areas of power transmission projects in the prior art, leading to insufficient soil and water loss prevention and control and project safety.

[0006] In view of the above problems, the present application provides an intelligent identification modeling method and system for non-standard runoff areas of power transmission projects.

[0007] In a first aspect, the application provides a method for intelligent identification and modeling of non-standard runoff cells in power transmission and transformation projects, which comprises: performing twin modeling based on real-time monitoring data sets of a power transmission and transformation project area to construct an engineering area model; identifying and dividing non-standard runoff cells based on the engineering area model to obtain a first runoff cell distribution model; optimizing water and soil loss coupling characteristics based on the first runoff cell distribution model to obtain a second runoff cell distribution model; performing direct water and soil loss accident deduction under multiple nodes on the second runoff cell distribution model based on a power transmission and transformation project construction scheme and future meteorological data sets of the power transmission and transformation project area to obtain a first water and soil loss deduction atlas; performing indirect water and soil loss accident deduction under multiple nodes on the second runoff cell distribution model based on the power transmission and transformation project construction scheme and the future meteorological data sets to obtain a second water and soil loss deduction atlas; and performing global disturbance optimization under multiple nodes on the second runoff cell distribution model based on the first water and soil loss deduction atlas and the second water and soil loss deduction atlas to obtain a third runoff cell distribution model.

[0008] In a second aspect, the application provides an intelligent identification and modeling system for non-standard runoff cells in power transmission and transformation projects, which comprises: an engineering area model construction module for performing twin modeling based on real-time monitoring data sets of a power transmission and transformation project area to construct an engineering area model; a first runoff cell distribution model obtaining module for identifying and dividing non-standard runoff cells based on the engineering area model to obtain a first runoff cell distribution model; a second runoff cell distribution model obtaining module for optimizing water and soil loss coupling characteristics based on the first runoff cell distribution model to obtain a second runoff cell distribution model; a first water and soil loss deduction atlas obtaining module for performing direct water and soil loss accident deduction under multiple nodes on the second runoff cell distribution model based on a power transmission and transformation project construction scheme and future meteorological data sets of the power transmission and transformation project area to obtain a first water and soil loss deduction atlas; a second water and soil loss deduction atlas obtaining module for performing indirect water and soil loss accident deduction under multiple nodes on the second runoff cell distribution model based on the power transmission and transformation project construction scheme and the future meteorological data sets to obtain a second water and soil loss deduction atlas; and a third runoff cell distribution model obtaining module for performing global disturbance optimization under multiple nodes on the second runoff cell distribution model based on the first water and soil loss deduction atlas and the second water and soil loss deduction atlas to obtain a third runoff cell distribution model.

[0009] The one or more technical solutions provided in the application have at least the following technical effects or advantages:

[0010] The method provided by the embodiment of the application comprises the following steps: performing twin modeling according to real-time monitoring data sets of a power transmission and transformation project area, and constructing an engineering area model; performing non-standard runoff plot identification and division according to the engineering area model, and obtaining a first runoff plot distribution model; performing water and soil loss coupling feature optimization according to the first runoff plot distribution model, and obtaining a second runoff plot distribution model; performing direct water and soil loss accident deduction under multiple nodes on the second runoff plot distribution model according to a power transmission and transformation project construction scheme of the power transmission and transformation project area and a future meteorological data set, and obtaining a first water and soil loss deduction atlas; performing indirect water and soil loss accident deduction under multiple nodes on the second runoff plot distribution model according to the power transmission and transformation project construction scheme and the future meteorological data set, and obtaining a second water and soil loss deduction atlas; and performing global disturbance optimization under multiple nodes on the second runoff plot distribution model according to the first water and soil loss deduction atlas and the second water and soil loss deduction atlas, and obtaining a third runoff plot distribution model, so that the technical effect of accurately identifying and dividing the non-standard runoff plot of the power transmission and transformation project and improving water and soil loss prevention and control and engineering safety is achieved.

[0011] The above description is only a summary of the technical solutions of the application. In order to enable one skilled in the art to better understand the technical means of the application, the application can be implemented according to the content of the specification, and in order to enable the above and other purposes, features and advantages of the application to be more obvious and easy to understand, the following specific embodiments of the application are described. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the application, nor is it intended to limit the scope of the application. Other features of the application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only exemplary, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of the provided drawings.

[0013] Figure 1 The flowchart of the intelligent identification and modeling method of the non-standard runoff plot of the power transmission and transformation project provided by the application.

[0014] Figure 2 The structural schematic diagram of the intelligent identification and modeling system of the non-standard runoff plot of the power transmission and transformation project provided by the application.

[0015] Explanation of reference signs: engineering area model construction module 11, first runoff plot distribution model obtaining module 12, second runoff plot distribution model obtaining module 13, first soil erosion deduction atlas obtaining module 14, second soil erosion deduction atlas obtaining module 15, third runoff plot distribution model obtaining module 16. DETAILED DESCRIPTION

[0016] The application provides a power transmission and transformation engineering non-standard runoff plot intelligent identification modeling method and system, which is used for solving the technical problems that the existing power transmission and transformation engineering non-standard runoff plot division is unreasonable, resulting in insufficient soil erosion prevention and control and engineering safety.

[0017] Hereinafter, the technical solutions in the application will be described clearly and completely with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the application, rather than all the embodiments of the application. It should be understood that the application is not limited by the example embodiments described herein. Based on the embodiments of the application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the application. In addition, it should be noted that, for the convenience of description, only the parts related to the application are shown in the drawings, not all.

[0018] Embodiment one, as shown in the application provides a power transmission and transformation engineering non-standard runoff plot intelligent identification modeling method, which comprises: Figure 1 According to the real-time monitoring data set of the engineering area, a twin modeling is performed to construct an engineering area model.

[0019]

[0020] ​Specifically, the power transmission and transformation project construction and operation area is comprehensively and real-time monitored according to a plurality of monitoring devices deployed in the power transmission and transformation project area, wherein the plurality of monitoring devices are used to monitor a plurality of dimension data such as terrain, surface and hydrological data of the power transmission and transformation project area in real time. For example, the terrain data of the power transmission and transformation project area is monitored according to devices such as laser radar and total station, including elevation, slope and slope direction; the soil moisture of the ground is monitored according to the soil moisture sensor, and the ground roughness is evaluated through the three-dimensional laser scanner, the high-resolution camera carried by the unmanned aerial vehicle is photographed, the vegetation coverage information is analyzed, and the ground information including soil moisture, ground roughness and vegetation coverage is obtained; the flow velocity and flow of the river or drainage channel are measured in real time according to the flow meter, and the rainfall is measured according to the rain gauge, and the hydrological parameters such as runoff and rainfall are obtained. Through methods such as geographic coordinate unification, noise filtering and interpolation correction, the collected multi-dimensional data is cleaned to obtain a real-time monitoring data set, and the real-time monitoring data set includes terrain data, surface data and hydrological data.

[0021] According to the real-time monitoring data set, a twin modeling is performed, the twin modeling refers to a method of using digital twin technology to map in a virtual space using the real-time monitoring data set to digitally model the characteristics and data of the power transmission and transformation project area, the core is to construct a virtual model that is real-time synchronized with the entity, to form accurate mapping by integrating real-time data and physical mechanism model, to support simulation analysis and decision optimization. Specifically, the terrain data and surface data in the real-time monitoring data set are used, according to the data elevation model and the data surface model, a three-dimensional terrain and surface model of the project area is generated by using three-dimensional geographic information system (GIS) and BIM technology, real-time hydrological monitoring data is used, a hydrology-geomorphology coupling model is used to dynamically simulate surface runoff, soil permeability and precipitation to form a three-dimensional hydrological model, time and space alignment is realized through the time stamp and geographic coordinates of the collected data, the integration of the three-dimensional terrain and surface model and the three-dimensional hydrological model is realized, and finally an engineering area model containing terrain, surface and hydrological parameters is formed on a three-dimensional visualization platform.

[0022] According to the engineering area model, a non-standard runoff plot is identified and divided to obtain a first runoff plot distribution model.

[0023] Further, according to the engineering area model, non-standard runoff plot identification and division are performed to obtain a first runoff plot distribution model, including: according to the engineering area model, multi-scale water and soil loss area feature identification is performed to obtain a first scale area feature sequence, a second scale area feature sequence, and a third scale area feature sequence; according to the first scale area feature sequence, non-standard runoff plot division is performed on the engineering area model to obtain a runoff plot division first model; according to the second scale area feature sequence, non-standard runoff plot division is performed on the engineering area model to obtain a runoff plot division second model; according to the third scale area feature sequence, non-standard runoff plot division is performed on the engineering area model to obtain a runoff plot division third model; according to the runoff plot division first model, the runoff plot division second model, and the runoff plot division third model, multi-scale comparative analysis is performed to obtain a multi-scale division comparative analysis result; and according to the multi-scale division comparative analysis result, global collaborative reconstruction is performed on the runoff plot division first model, the runoff plot division second model, and the runoff plot division third model to generate the first runoff plot distribution model.

[0024] Specifically, a runoff plot is a test facility for quantitative research on the law of water and soil loss on a slope and the law of water and soil loss in a small watershed, and generally consists of a side ridge, a plot enclosed by the side ridge, a catchment tank, a runoff and sediment storage device, a protection belt, and a drainage system. A non-standard runoff plot refers to a plurality of runoff division units dynamically generated according to multi-dimensional features such as actual topography, ground surface, and hydrological parameters, with the boundary and size dynamically adjusted and changed based on environmental features to be closer to the actual law of water and soil loss, without relying on traditional division rules or fixed grid division methods. Non-standard runoff plot identification and division refer to dividing a power transmission and transformation engineering area into a plurality of areas with different water and soil loss characteristics and runoff responses for fine control.

[0025] The specific process of identifying and dividing non-standard runoff plots according to the engineering area model first identifies the characteristics of water and soil loss in the engineering area model at multiple scales. The identification of water and soil loss characteristics at multiple scales is performed at different spatial resolutions, including a first scale, a second scale, and a third scale. The first scale is used to analyze the overall topography, water system, and soil type of the entire engineering area. The second scale is used to analyze local slope, channel, and vegetation coverage area for medium-range water and soil loss characteristics. The third scale is used to analyze micro-topographic changes and small-range runoff paths. Using GIS spatial analysis tools, the slope, water system, soil type, and vegetation coverage area of the engineering area model are analyzed to obtain a first scale region feature sequence, a second scale region feature sequence, and a third scale region feature sequence. The first scale region feature sequence identifies water and soil loss characteristics at a larger macro spatial scale based on the engineering area model. It analyzes the overall topographic slope distribution, water system orientation, and overall soil type of the engineering area. The second scale region feature sequence identifies water and soil loss characteristics at a medium meso scale based on the engineering area model. For example, it analyzes the influence of meso features such as local slope, soil moisture, secondary channels, and vegetation coverage on runoff formation. The third scale region feature sequence identifies water and soil loss characteristics at a smaller micro spatial scale. For example, it analyzes micro-topographic changes and local runoff paths. Each of the first scale region feature sequence, the second scale region feature sequence, and the third scale region feature sequence contains information such as spatial location, slope, hydrological properties, soil type, and vegetation coverage.

[0026] Based on the first scale region feature sequence, the second scale region feature sequence, and the third scale region feature sequence, non-standard runoff plot division is performed on the engineering area model using GIS spatial segmentation algorithms such as the watershed algorithm based on slope and basin, clustering algorithms such as K-means, and division conditions such as dividing high slope and soil erosion sensitive areas into plots. The plot boundaries are formed based on the intersection of primary and secondary channels, and the plots are further subdivided based on small-scale features. This results in runoff plot division first model, runoff plot division second model, and runoff plot division third model. The runoff plot division first model is a large-scale plot generated based on the first scale feature sequence, with overall connectivity and basin structure characteristics. The runoff plot division second model is a medium-range plot divided based on the second scale feature sequence, with local slope, secondary channels, and hydrological differences. The runoff plot division third model is a small-range plot further divided based on the third scale sequence.

[0027] The first runoff plot division model, the second runoff plot division model and the third runoff plot division model are cross-validated through multi-scale comparative analysis, the boundaries, areas and slopes of the plots are compared, the boundary conflict areas are marked, the plots with inconsistent characteristics are identified, the boundary positions and the soil and water loss sensitivity differences of the three models are analyzed, and the multi-scale division comparative analysis results are obtained. Based on the multi-scale division comparative analysis results, the first runoff plot division model, the second runoff plot division model and the third runoff plot division model are globally reconstructed in collaboration, the plot boundaries are adjusted, and a first runoff plot distribution model is generated, the first runoff plot distribution model comprising a plurality of runoff plots. For example, a multi-objective optimization method is used to construct multi-objective optimization indexes, including macro-connectivity, local feature guarantee and small-range sensitivity, etc. The macro-connectivity is used to ensure the hydrological basin connectivity between the plots, and the local feature guarantee is used to retain the actual slope, soil type, channel and vegetation coverage differences. Constraint conditions are set according to the multi-objective optimization indexes, for example, the plot boundaries are continuous and do not intersect, the high-sensitivity areas need to be separately divided, and the total area of the multiple plots and the actual engineering area are consistent. In the optimization process, the boundaries of the three scales are superimposed as the initial solution, the plot boundary point positions are fine-tuned through mutation operation, the plot shape and area are adjusted, the boundary characteristics of the three scales are fused through crossover operation, candidate schemes are generated, the multi-objective fitness of each candidate scheme is calculated, the Pareto optimal solution is screened and continuously iterated and optimized until the boundaries converge or the preset iteration number is reached, and the first runoff plot division model, the second runoff plot division model and the third runoff plot division model are globally reconstructed in collaboration.

[0028] Through multi-scale soil and water loss area feature identification and non-standard runoff plot division, the accuracy of the division is improved, and then the first runoff plot distribution model is generated through global collaborative reconstruction, thereby improving the pertinence and reliability of soil and water loss prediction and construction safety evaluation.

[0029] Further, a second runoff plot distribution model is obtained through soil and water loss coupling feature optimization based on the first runoff plot distribution model.

[0030] Further, a second runoff plot distribution model is obtained through soil and water loss coupling feature optimization based on the first runoff plot distribution model, including: soil and water loss feature identification is performed on each runoff plot in the first runoff plot distribution model to obtain a plurality of plot soil and water loss features; pairwise coupling evaluation is performed on the plurality of plot soil and water loss features to obtain a soil and water loss coupling evaluation sequence; based on a soil and water loss coupling evaluation threshold, the plurality of plot soil and water loss features are super-coupling associated based on the soil and water loss coupling evaluation sequence to obtain a soil and water loss super-coupling relationship network; and boundary correction is performed on the first runoff plot distribution model based on the soil and water loss super-coupling relationship network to generate the second runoff plot distribution model.

[0031] Specifically, the water and soil loss characteristics of each runoff plot in the first runoff plot distribution model are extracted to obtain a plurality of plot water and soil loss characteristics, each plot water and soil loss characteristic including but not limited to slope, aspect, soil moisture, runoff, rainfall, and vegetation coverage, etc. According to the obtained plurality of plot water and soil loss characteristics, a two-by-two coupling evaluation is performed, which refers to the correlation between different plots in the process of water and soil loss, such as the erosion enhancement caused by the runoff of the upstream into the downstream, or the synergistic relationship between adjacent plots in vegetation restoration and soil stabilization. For example, the Pearson correlation coefficient, cosine similarity, and other correlation analysis methods are used to evaluate the coupling relationship of water and soil loss characteristics of different runoff plots. According to the plurality of coupling evaluation results, a water and soil loss coupling evaluation sequence is formed, which contains the correlation strength between each runoff plot and other plots.

[0032] According to actual needs and expert experience, a water and soil loss coupling evaluation threshold is set to screen a plurality of plot water and soil loss characteristics with high correlation coupling relationship. According to the water and soil loss coupling evaluation threshold, the water and soil loss coupling evaluation sequence is super-coupling correlation combed, the coupling relationship higher than the water and soil loss coupling evaluation threshold is screened out, and a water and soil loss super-coupling relationship network is constructed according to the screened super-coupling relationship, which can be represented using graph theory, wherein the nodes represent runoff plots, and the strength of the edges represents the strength of the coupling relationship. According to the water and soil loss super-coupling relationship network, the boundary of the first runoff plot distribution model is corrected through spatial analysis tools in GIS, such as buffer analysis, spatial overlay analysis, etc., the boundaries of a plurality of runoff plots are adjusted, the boundaries of high-coupling plots are adjusted or reconstructed, and the boundaries of weak-coupling plots are smoothed, so that the division of the plots is more in line with the actual distribution of the actual water and soil loss characteristics. By introducing the water and soil loss coupling mechanism to correct the boundary of the first runoff plot distribution model, a second runoff plot distribution model is formed, so that the second runoff plot distribution model formed after correction not only considers the geometric characteristics of the terrain and topography, but also more accurately reflects the actual changes of the water and soil loss characteristics, improving the accuracy and reliability of the runoff plot division.

[0033] According to the power transmission project construction scheme and future meteorological data set of the power transmission project area, a direct water and soil loss accident deduction under a plurality of nodes is performed on the second runoff plot distribution model to obtain a first water and soil loss deduction atlas.

[0034] Further, the second runoff cell distribution model is subjected to direct water and soil loss accident deduction under multiple nodes according to the power transmission and transformation engineering construction scheme of the power transmission and transformation engineering area and a future meteorological data set, to obtain a first water and soil loss deduction atlas, including: multiple node engineering construction schemes are obtained according to the power transmission and transformation engineering construction scheme; multiple node construction scene feature carding is performed according to the multiple node engineering construction schemes and the future meteorological data set, to establish multiple node construction scene vectors; the Sth runoff cell model is extracted according to the second runoff cell distribution model, S being a positive integer; the Sth runoff cell model is subjected to simulated construction respectively based on the multiple node construction scene vectors, to obtain multiple node cell simulation data; a direct water and soil loss deduction channel is established through distillation loss iterative learning under balanced confrontation according to a direct water and soil loss event set; the multiple node cell simulation data are input into the direct water and soil loss deduction channel, to obtain multiple direct water and soil loss deduction paths; the multiple direct water and soil loss deduction paths are sorted, to obtain the Sth cell direct water and soil loss accident distribution, and the Sth cell direct water and soil loss accident distribution is added to the first water and soil loss deduction atlas.

[0035] Specifically, the power transmission and transformation engineering construction scheme is divided into multiple construction nodes according to the power transmission and transformation engineering construction drawings and the construction progress plan, each node corresponding to a specific construction stage or construction area, for example, foundation pit excavation, tower foundation construction, road hardening and backfilling and remediation stages, to form multiple node engineering construction schemes. Meanwhile, the future meteorological data of the area are obtained from a meteorological service platform to form a future meteorological data set, which includes rainfall, rainfall intensity, wind speed, etc. The multiple node engineering construction schemes and the future meteorological data are subjected to multiple node construction scene feature carding, that is, the multiple node engineering construction schemes and the corresponding future meteorological data are subjected to spatiotemporal superposition to form multiple node construction scene vectors, wherein each node construction scene vector represents a construction scene of a construction node under the corresponding meteorological condition.

[0036] From the second runoff cell distribution model containing multiple runoff cell models, the Sth runoff cell model is extracted, S being a positive integer. The extracted Sth runoff cell model is subjected to simulated construction respectively in the digital twin environment according to the multiple node construction scene vectors as input, to simulate dynamic changes such as ground disturbance, runoff path change, soil structure loosening and slope exposure, and to record water and soil loss conditions during the simulated construction. Through physical modeling and data driving, cell response data under each node are obtained, to generate multiple node cell simulation data.

[0037] The direct soil and water loss event data is collected, including a collection of soil and water loss events directly caused by construction disturbance or local meteorological factors under similar topography and construction conditions in the past. For example, erosion, increased runoff, sediment output, and other phenomena caused by excavation of tower foundations, earthwork piling, construction road opening, or direct action of local heavy rainfall on bare ground during the construction process of power transmission and transformation projects. The direct soil and water loss event set records the occurrence time, intensity, location and range of soil and water loss under direct disturbance conditions. Through sample balancing adversarial learning, the direct soil and water loss event set is balanced, ensuring the balance of the training data. Through iterative learning of the distillation loss function, the performance of the deduction model is optimized, and a direct soil and water loss deduction channel is established, which can accurately and quickly output the small area soil and water loss response path. Multiple node small area simulation data are input into the direct soil and water loss deduction channel to generate multiple direct soil and water loss deduction paths, each of which represents the spatial evolution process of runoff convergence, erosion intensification and sediment migration in the small area under specific construction nodes and corresponding meteorological conditions. The multiple direct soil and water loss deduction paths are spatio-temporally arranged, for example, the construction stages of different nodes may have a continuous relationship, and the multiple direct soil and water loss deduction paths are aligned according to the construction time sequence to form a Sth small area direct soil and water loss accident distribution. The Sth small area direct soil and water loss accident distribution is superimposed on the time sequence map of the overall region to generate a first soil and water loss deduction map. The second soil and water loss deduction map not only shows the prediction path and range of direct soil and water loss, but also reflects the intensity of soil and water loss through color depth, for example, dark areas represent very serious soil and water loss, and light areas represent relatively light soil and water loss, that is, the first soil and water loss deduction map can directly show the potential soil and water loss risk area, impact range and intensity of each small area under different construction nodes.

[0038] By simulating the construction process combined with future meteorological conditions, the possible occurrence time of direct soil and water loss is predicted, the first soil and water loss deduction map is generated, and the accuracy, comprehensiveness and intuitiveness of soil and water loss risk prediction are improved. Moreover, by predicting soil and water loss accidents in advance, effective preventive measures can be taken to reduce the impact of soil and water loss on the environment and ensure the smooth progress of power transmission and transformation projects.

[0039] Further, according to the direct water and soil loss event set, the distillation loss iterative learning under balanced confrontation is established, and the direct water and soil loss deduction channel is established, including: according to the direct water and soil loss event set, the fault tree is traced back, and the first water and soil loss tracing distribution is obtained; according to the first water and soil loss tracing distribution, the first direct water and soil loss deduction model is trained; according to the first water and soil loss tracing distribution, the sample balance evaluation is carried out, and the first sample balance coefficient is obtained; based on the first sample balance coefficient, according to the sample balance threshold, the first water and soil loss tracing distribution is injected into the balanced confrontation sample, and the second water and soil loss tracing distribution is obtained; according to the second water and soil loss tracing distribution, the second direct water and soil loss deduction model is trained; according to the first direct water and soil loss deduction model and the second direct water and soil loss deduction model, the distillation loss iterative learning is carried out, and the direct water and soil loss deduction channel is generated.

[0040] Specifically, the fault tree analysis method is used to analyze the direct water and soil loss event set. Fault tree analysis is a top-down event logical reasoning method, which traces back the causes and processes of water and soil loss events from the results of water and soil loss events. Specifically, the direct water and soil loss event is taken as the top event, which is decomposed into several bottom causes, such as heavy rain, construction disturbance, vegetation damage, excessive surface slope, etc., and the fault tree structure is constructed through logical relationship. For each historical direct water and soil loss event sample, based on meteorological data, construction progress records, terrain and surface data, etc., the satisfaction and contribution probability of each water and soil loss event within a period of time before the accident is calculated, and the causes and occurrence path of each direct water and soil loss event are obtained by matching the fault tree model, and the first water and soil loss tracing distribution is generated.

[0041] According to the first water and soil loss tracing distribution, a suitable machine learning model is selected, for example, a long short-term memory network is used to learn the time series data. The input features in the first water and soil loss tracing distribution are arranged in time sequence as input sample sequence, including meteorological data, construction scene and terrain and surface data, etc., and the long short-term memory network model is trained. The long short-term memory network model analyzes the causes and paths of the first water and soil loss tracing distribution, and inputs the water and soil loss prediction path result. In the training stage, the mean square error is used as the loss function for optimization, and the trained long short-term memory network model is verified by the validation set, and finally the first direct water and soil loss deduction model is obtained, which can output the corresponding water and soil loss response trend according to the new input construction data and meteorological data.

[0042] The samples in the first water and soil loss trace distribution are classified and counted, for example, classified according to water and soil loss intensity, including mild, moderate and severe, or grouped according to event type, including rainfall-induced and construction disturbance, etc., the number of samples in each category is counted, the proportion and variance of each category are calculated, the sample balance average is formed, and the first sample balance coefficient is used to reflect the difference degree of the number of samples of different categories in the direct water and soil loss event set. When the balance coefficient is large, it indicates that the sample distribution deviation is large. A sample balance threshold is set to determine whether the sample type in the water and soil loss event set reaches a balanced state. When the first sample balance coefficient exceeds the sample balance threshold, the sample enhancement process is triggered to supplement the minority samples. For example, new samples are generated through a generative adversarial network or a time series data enhancement method. Specifically, the target data of the minority class samples is calculated using the first sample balance coefficient, a generative adversarial network is generated based on the original time series samples, and time series data consistent with the real distribution but with diverse sample attributes are generated through adversarial learning. Through sample consistency verification, the rationality of the newly added minority class samples is analyzed, and the newly added minority class samples that meet the rationality are injected into the first water and soil loss trace distribution to generate a second water and soil loss trace distribution. Similarly, according to the second water and soil loss trace distribution, a second direct water and soil loss deduction model is trained to make the second direct water and soil loss deduction model better learn the characteristics of the minority class events, and further improve the prediction accuracy of the direct water and soil loss deduction model.

[0043] Finally, according to the first direct water and soil loss deduction model and the second direct water and soil loss deduction model, distillation loss iterative learning is performed, the second direct water and soil loss deduction model is used as a teacher model, and the first direct water and soil loss deduction model is used as a student model. The first direct water and soil loss deduction model is trained using a knowledge distillation strategy. The second direct water and soil loss deduction model provides soft labels of high-precision events, i.e., probability distribution of events, and the first direct water and soil loss deduction model is optimized by minimizing the weighted combination of hard label loss and soft label distillation loss. Through multiple rounds of iterative training, the first direct water and soil loss deduction model gradually absorbs the high-order feature expression of the second direct water and soil loss deduction model while maintaining its own efficiency, achieving a balance between model performance and direct water and soil loss deduction efficiency, and generating the direct water and soil loss deduction channel. By generating the direct water and soil loss deduction channel, the occurrence path and impact range of the direct water and soil loss accident can be accurately and quickly predicted according to different node construction scenes and future meteorological conditions, providing accurate and reliable data support for dynamic assessment of water and soil loss risk in power transmission and transformation engineering construction.

[0044] According to the power transmission and transformation engineering construction scheme and the future meteorological data set, the second runoff cell distribution model is used to deduce indirect water and soil loss accidents under multiple nodes to obtain a second water and soil loss deduction graph.

[0045] Further, the second runoff plot distribution model is subjected to indirect soil and water loss accident deduction under multiple nodes according to the power transmission and transformation project construction scheme and the future meteorological data set, to obtain a second soil and water loss deduction atlas, including: performing fault tree backtracking according to the indirect soil and water loss event set to obtain a third soil and water loss backtracking distribution; performing distillation loss iterative learning under balanced confrontation according to the third soil and water loss backtracking distribution to establish an indirect soil and water loss deduction channel; inputting the multiple node plot simulation data into the indirect soil and water loss deduction channel to obtain multiple indirect soil and water loss deduction paths; arranging the multiple indirect soil and water loss deduction paths to obtain an Sth plot indirect soil and water loss accident distribution, and adding the Sth plot indirect soil and water loss accident distribution to the second soil and water loss deduction atlas.

[0046] Specifically, historical indirect soil and water loss event data is collected to form an indirect soil and water loss event set, which refers to a collection of secondary or chain soil and water loss events caused by direct disturbance events through hydrological, topographical or soil conduction. For example, upstream construction causes runoff to collect or increase sediment output, and downstream plots are affected to produce erosion or slope instability, etc. indirect soil and water loss events. Using fault tree analysis method, the causes and processes of indirect soil and water loss events are traced from the results of indirect soil and water loss events to generate a third soil and water loss backtracking distribution, including the causal relationship and occurrence path of indirect soil and water loss events. Similarly, according to the third soil and water loss backtracking distribution, a first indirect soil and water loss deduction model is trained, and sample balanced evaluation is performed according to the third soil and water loss backtracking distribution, and according to the sample balanced evaluation result and the sample balanced threshold, the third soil and water loss backtracking distribution is subjected to balanced confrontation sample injection to increase a small number of sample data to form a fourth soil and water loss backtracking distribution. And train a second indirect soil and water loss deduction model based on the fourth soil and water loss backtracking distribution, and perform iterative learning through distillation loss to build an indirect soil and water loss deduction channel.

[0047] The multiple node cell simulation data are input into the indirect soil erosion deduction channel, and the indirect soil erosion deduction channel analyzes the input multiple node cell simulation data and outputs multiple indirect soil erosion deduction paths. The multiple indirect soil erosion deduction paths are sorted to obtain the Sth cell indirect soil erosion event distribution, including the range, intensity and possible influence area of soil erosion. The Sth cell indirect soil erosion event distribution is added to the second soil erosion deduction atlas, and the second soil erosion deduction atlas not only shows the path range of indirect soil erosion, but also reflects the intensity of soil erosion through the color depth, for example, the dark color area represents very serious soil erosion, and the light color area represents relatively light soil erosion. Through the analysis of the indirect soil erosion event set, the second soil erosion deduction atlas is constructed, which further improves the comprehensiveness and reliability of the soil erosion risk prediction, optimizes the construction plan and protection measures of the power transmission and transformation project, reduces the influence of soil erosion on the environment, and ensures the safety and ecological stability of the power transmission and transformation project.

[0048] According to the first soil erosion deduction atlas and the second soil erosion deduction atlas, the second runoff cell distribution model is globally disturbed and optimized under multiple nodes to obtain a third runoff cell distribution model.

[0049] Further, according to the first soil erosion deduction atlas and the second soil erosion deduction atlas, the second runoff cell distribution model is globally disturbed and optimized under multiple nodes to obtain a third runoff cell distribution model, including: according to the first soil erosion deduction atlas, the multiple node disturbance influence of each runoff cell model in the second runoff cell distribution model is analyzed to obtain a first cell disturbance analysis atlas; according to the second soil erosion deduction atlas, the multiple node disturbance influence of each runoff cell model is analyzed to obtain a second cell disturbance analysis atlas; according to the first cell disturbance analysis atlas and the second cell disturbance analysis atlas, the multiple node disturbance influence coupling analysis of each runoff cell model is performed to obtain a disturbance influence coupling distribution; based on the disturbance influence coupling distribution, the multiple node cooperative optimization of the first cell disturbance analysis atlas and the second cell disturbance analysis atlas is performed to obtain a third cell disturbance analysis atlas; according to the third cell disturbance analysis atlas, the multiple node global optimization of each runoff cell model is performed to generate the third runoff cell distribution model.

[0050] Specifically, using a geographic information system overlay analysis calculation, the first soil erosion deduction atlas is used to spatially register the spatial data of the plurality of runoff plot models in the second runoff plot distribution model, and the plurality of runoff plot models are subjected to multi-node disturbance influence analysis. Using a digital twin model, through simulation, the sensitivity and response intensity of external disturbances, including excavation, filling, drainage changes, etc. to the soil erosion state in each runoff plot under different construction nodes and meteorological conditions are obtained, including erosion intensity, erosion range, response time and cumulative effect, a plot disturbance analysis first atlas is generated, reflecting the disturbance distribution under direct soil erosion. Similarly, according to the second soil erosion deduction atlas, the multi-node disturbance influence analysis of each runoff plot model in the second runoff plot distribution model is performed, and a plot disturbance analysis second atlas is generated, reflecting the disturbance distribution under indirect soil erosion.

[0051] According to the plot disturbance analysis first atlas and the plot disturbance analysis second atlas, the multi-node disturbance influence coupling analysis of each runoff plot model is performed, for example, by using an additive method, the weights are set according to the intensity, range, duration, etc. of the disturbance, and the interaction of direct and indirect disturbances in each plot in time and space is identified through weighted calculation, and a disturbance influence coupling distribution is obtained, which is used to reflect which plots form synergistic enhancement or mutual weakening under multi-node conditions. Based on the disturbance influence coupling distribution, through an optimization algorithm, such as a genetic algorithm or a particle swarm optimization algorithm, the plot disturbance analysis first atlas and the plot disturbance analysis second atlas are subjected to multi-node synergistic optimization, and an optimized plot disturbance analysis third atlas is generated. For example, the optimization target is set, such as minimizing the intensity of inter-plot disturbance risk, balancing the disturbance conduction effect of upstream and downstream plots, improving the stability and connectivity of plot division, and taking the boundary position, area ratio and interference response of the plot as adjustable variables, through the optimization algorithm to search and iteratively update under multi-node disturbance conditions, through the particle swarm optimization algorithm to simulate the position movement of the group in the multi-dimensional space, to perform global search according to the disturbance risk distribution gradient, and after multiple iterations, the solution with the minimum overall disturbance risk and stable spatial structure is output, forming the plot disturbance analysis third atlas.

[0052] Finally, according to the third graph of the cell disturbance analysis, a multi-node optimization is performed on each runoff cell model by a global optimization algorithm such as the simulated annealing algorithm. The third graph of the cell disturbance analysis is taken as the initial state, and a new candidate partition scheme is generated by randomly disturbing the position or form of the cell boundary. The overall disturbance intensity change after adjustment is calculated, and the candidate partition scheme is selected to be retained or discarded according to the acceptance criterion of the simulated annealing. Through multiple iterations, the global optimal solution is gradually approached, and the third runoff cell distribution model is generated. The third runoff cell distribution model takes into account the direct and indirect effects of soil erosion, and can more accurately reflect the soil erosion risk distribution in the power transmission and transformation project area. It provides comprehensive and reliable data support for the construction of power transmission and transformation stations, improves the effectiveness and rationality of risk prevention and control in the construction area of power transmission and transformation stations, reduces soil erosion, and ensures the safety of the project and the stability of the ecology.

[0053] Further, according to the real-time monitoring data set of the power transmission and transformation project area, a twin modeling is performed, including: performing real-time monitoring on the power transmission and transformation project area to obtain a regional data set; performing data cleaning according to the regional data set to obtain the real-time monitoring data set.

[0054] Specifically, the power transmission and transformation project is monitored in real time by using various monitoring devices deployed in the project area to obtain multi-dimensional data including topography, ground surface and hydrological parameters, forming a regional data set. Then the collected regional data is cleaned, and through coordinate conversion and projection, the regional data geographic coordinates are unified, and the data of different monitoring devices are unified into a unified geographic coordinate system to ensure the accuracy and consistency of multi-dimensional data in space. Through statistical analysis, noise and outliers in the regional data set are identified and removed, and missing values in the regional data set are corrected using interpolation methods such as linear interpolation and Kriging interpolation to fill in data gaps, thereby obtaining the real-time monitoring data set. Through data cleaning, the data accuracy and completeness of the real-time monitoring data set are improved, thereby ensuring the accuracy and reliability of the constructed project area model.

[0055] Further, the project area model is constructed, including: performing bias detection on the project area model according to the real-time monitoring data set to obtain a modeling bias detection result; and performing feedback optimization on the project area model according to the modeling bias detection result.

[0056] Specifically, after the engineering area model is constructed, it is run, and a deviation detection mechanism is used to monitor deviations in the model based on real-time monitoring datasets. The real-time monitoring datasets and the running data of the engineering area model are compared and analyzed to calculate the deviations between multiple real-time monitoring data and their corresponding running data. Based on these deviations, a modeling deviation detection result is generated. When deviations are found, the engineering area model is optimized based on the modeling deviation detection result, adjusting its parameters. Deviation detection is then performed again after optimization until the engineering area model accurately reflects the actual data information of the power transmission and transformation engineering area. By detecting and providing feedback on the engineering area model, its accuracy and reliability are improved, thus enabling it to accurately reflect the actual situation of the power transmission and transformation engineering area.

[0057] Example 2, based on the same inventive concept as the intelligent identification and modeling method for non-standard runoff zones in power transmission and transformation projects in the previous examples, such as... Figure 2 As shown, this application provides an intelligent identification and modeling system for non-standard runoff communities in power transmission and transformation projects. The system includes:

[0058] The engineering area model construction module 11 is used to construct an engineering area model by performing twin modeling based on the real-time monitoring dataset of the power transmission and transformation engineering area; the first runoff cell distribution model acquisition module 12 is used to identify and classify non-standard runoff cells based on the engineering area model to obtain a first runoff cell distribution model; the second runoff cell distribution model acquisition module 13 is used to optimize the water and soil erosion coupling characteristics based on the first runoff cell distribution model to obtain a second runoff cell distribution model; the first water and soil erosion projection map acquisition module 14 is used to obtain a first water and soil erosion projection map based on the power transmission and transformation engineering construction plan and future meteorological dataset of the power transmission and transformation engineering area. The second runoff plot distribution model is used to perform direct soil erosion accident simulation under multiple nodes to obtain a first soil erosion simulation map; the second soil erosion simulation map acquisition module 15 is used to perform indirect soil erosion accident simulation under multiple nodes on the second runoff plot distribution model based on the power transmission and transformation project construction plan and the future meteorological dataset to obtain a second soil erosion simulation map; the third runoff plot distribution model acquisition module 16 is used to perform global disturbance optimization under multiple nodes on the second runoff plot distribution model based on the first soil erosion simulation map and the second soil erosion simulation map to obtain a third runoff plot distribution model.

[0059] Further, the first runoff cell distribution model obtaining module 12 is further used for: performing multi-scale soil and water loss region feature identification according to the engineering region model, to obtain a first scale region feature sequence, a second scale region feature sequence and a third scale region feature sequence; performing non-standard runoff cell division on the engineering region model according to the first scale region feature sequence, to obtain a runoff cell division first model; performing non-standard runoff cell division on the engineering region model according to the second scale region feature sequence, to obtain a runoff cell division second model; performing non-standard runoff cell division on the engineering region model according to the third scale region feature sequence, to obtain a runoff cell division third model; performing multi-scale comparison analysis according to the runoff cell division first model, the runoff cell division second model and the runoff cell division third model, to obtain a multi-scale division comparison analysis result; and performing global collaborative reconstruction on the runoff cell division first model, the runoff cell division second model and the runoff cell division third model according to the multi-scale division comparison analysis result, to generate the first runoff cell distribution model.

[0060] Further, the first runoff cell distribution model obtaining module 12 is further used for: performing soil and water loss feature identification according to each runoff cell in the first runoff cell distribution model, to obtain a plurality of cell soil and water loss features; performing pairwise coupling evaluation according to the plurality of cell soil and water loss features, to obtain a soil and water loss coupling evaluation sequence; performing super-coupling correlation combing on the plurality of cell soil and water loss features according to the soil and water loss coupling evaluation sequence based on a soil and water loss coupling evaluation threshold, to obtain a soil and water loss super-coupling relationship network; and performing boundary correction on the first runoff cell distribution model according to the soil and water loss super-coupling relationship network, to generate the second runoff cell distribution model.

[0061] Further, the first soil and water loss deduction atlas obtaining module 14 is further used for: performing multi-node division according to the power transmission and transformation engineering construction scheme, to obtain a plurality of node engineering construction schemes; performing multi-node construction scene feature combing according to the plurality of node engineering construction schemes and the future meteorological data set, to establish a plurality of node construction scene vectors; extracting an Sth runoff cell model according to the second runoff cell distribution model, S being a positive integer; performing simulated construction on the Sth runoff cell model based on the plurality of node construction scene vectors, to obtain a plurality of node cell simulation data; establishing a direct soil and water loss deduction channel through distillation loss iterative learning under balanced confrontation according to a direct soil and water loss event set; inputting the plurality of node cell simulation data into the direct soil and water loss deduction channel, to obtain a plurality of direct soil and water loss deduction paths; arranging the plurality of direct soil and water loss deduction paths, to obtain an Sth cell direct soil and water loss accident distribution, and adding the Sth cell direct soil and water loss accident distribution to the first soil and water loss deduction atlas.

[0062] Further, the first water and soil erosion deduction graph obtaining module 14 is further used for: performing fault tree tracing according to the direct water and soil erosion event set to obtain a first water and soil erosion tracing distribution; training a first direct water and soil erosion deduction model according to the first water and soil erosion tracing distribution; performing sample balance evaluation according to the first water and soil erosion tracing distribution to obtain a first sample balance coefficient; based on the first sample balance coefficient, performing balance counter sample injection on the first water and soil erosion tracing distribution according to a sample balance threshold to obtain a second water and soil erosion tracing distribution; training a second direct water and soil erosion deduction model according to the second water and soil erosion tracing distribution; and performing distillation loss iterative learning according to the first direct water and soil erosion deduction model and the second direct water and soil erosion deduction model to generate the direct water and soil erosion deduction channel.

[0063] Further, the second water and soil erosion deduction graph obtaining module 15 is further used for: performing fault tree tracing according to an indirect water and soil erosion event set to obtain a third water and soil erosion tracing distribution; performing distillation loss iterative learning under balance counter to establish an indirect water and soil erosion deduction channel; inputting the multiple node cell simulation data into the indirect water and soil erosion deduction channel to obtain multiple indirect water and soil erosion deduction paths; collating the multiple indirect water and soil erosion deduction paths to obtain an Sth cell indirect water and soil erosion accident distribution, and adding the Sth cell indirect water and soil erosion accident distribution to the second water and soil erosion deduction graph.

[0064] Further, the third runoff cell distribution model obtaining module 16 is further used for: performing multi-node disturbance influence analysis on each runoff cell model in the second runoff cell distribution model according to the first water and soil erosion deduction graph to obtain a first cell disturbance analysis graph; performing multi-node disturbance influence analysis on the each runoff cell model according to the second water and soil erosion deduction graph to obtain a second cell disturbance analysis graph; performing multi-node disturbance influence coupling analysis on the each runoff cell model according to the first cell disturbance analysis graph and the second cell disturbance analysis graph to obtain a disturbance influence coupling distribution; performing multi-node collaborative optimization on the first cell disturbance analysis graph and the second cell disturbance analysis graph based on the disturbance influence coupling distribution to obtain a third cell disturbance analysis graph; and performing multi-node global optimization on the each runoff cell model according to the third cell disturbance analysis graph to generate the third runoff cell distribution model.

[0065] Further, the engineering region model construction module 11 is further used for: performing real-time monitoring on the power transmission and transformation engineering region to obtain a region data set; and performing data cleaning according to the region data set to obtain the real-time monitoring data set.

[0066] Further, the engineering area model construction module 11 is further configured to perform bias detection on the engineering area model according to the real-time monitoring data set, and obtain a modeling bias detection result; and perform feedback optimization on the engineering area model according to the modeling bias detection result.

[0067] The above description of disclosed embodiments enables one of ordinary skill in the art to make or use the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0068] Obviously, for those skilled in the art, some improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the present application.

Claims

1. A method for intelligent identification and modeling of non-standard runoff plots in power transmission and transformation projects, characterized in that, The method comprises the following steps: According to the real-time monitoring data set of the power transmission and transformation project area, a twin modeling is performed to construct an engineering area model; According to the engineering area model, a non-standard runoff cell identification and division is performed to obtain a first runoff cell distribution model; According to the first runoff cell distribution model, a water and soil loss coupling feature optimization is performed to obtain a second runoff cell distribution model; According to the power transmission and transformation project construction scheme of the power transmission and transformation project area and the future meteorological data set, a direct water and soil loss accident deduction under multiple nodes is performed on the second runoff cell distribution model to obtain a first water and soil loss deduction atlas; According to the power transmission and transformation project construction scheme and the future meteorological data set, an indirect water and soil loss accident deduction under multiple nodes is performed on the second runoff cell distribution model to obtain a second water and soil loss deduction atlas; According to the first water and soil loss deduction atlas and the second water and soil loss deduction atlas, a global disturbance optimization under multiple nodes is performed on the second runoff cell distribution model to obtain a third runoff cell distribution model; According to the first water and soil loss deduction atlas and the second water and soil loss deduction atlas, a global disturbance optimization under multiple nodes is performed on the second runoff cell distribution model to obtain a third runoff cell distribution model, comprising: According to the first water and soil loss deduction atlas, a multi-node disturbance influence analysis is performed on each runoff cell model in the second runoff cell distribution model to obtain a first cell disturbance analysis atlas; According to the second water and soil loss deduction atlas, a multi-node disturbance influence analysis is performed on each runoff cell model to obtain a second cell disturbance analysis atlas; According to the first cell disturbance analysis atlas and the second cell disturbance analysis atlas, a multi-node disturbance influence coupling analysis is performed on each runoff cell model to obtain a disturbance influence coupling distribution; Based on the disturbance influence coupling distribution, a multi-node collaborative optimization is performed on the first cell disturbance analysis atlas and the second cell disturbance analysis atlas to obtain a third cell disturbance analysis atlas; According to the third cell disturbance analysis atlas, a multi-node global optimization is performed on each runoff cell model to generate the third runoff cell distribution model.

2. The power transmission project non-standard runoff plot intelligent identification modeling method of claim 1, wherein, According to the engineering area model, a non-standard runoff cell identification and division is performed to obtain a first runoff cell distribution model, comprising: According to the engineering area model, a multi-scale water and soil loss area feature identification is performed to obtain a first scale area feature sequence, a second scale area feature sequence and a third scale area feature sequence; According to the first scale area feature sequence, a non-standard runoff cell division is performed on the engineering area model to obtain a first runoff cell division model; According to the second scale area feature sequence, a non-standard runoff cell division is performed on the engineering area model to obtain a second runoff cell division model; According to the third scale area feature sequence, a non-standard runoff cell division is performed on the engineering area model to obtain a third runoff cell division model; According to the first runoff cell division model, the second runoff cell division model and the third runoff cell division model, a multi-scale comparative analysis is performed to obtain a multi-scale division comparative analysis result; According to the multi-scale division contrast analysis result, the first runoff plot division model, the second runoff plot division model and the third runoff plot division model are globally cooperatively reconstructed to generate the first runoff plot distribution model.

3. The power transmission project non-standard runoff plot intelligent identification modeling method of claim 1, wherein, According to the first runoff plot distribution model, water and soil loss coupling characteristics are optimized to obtain a second runoff plot distribution model, including: According to the first runoff plot distribution model, water and soil loss characteristics of each runoff plot are identified to obtain a plurality of plot water and soil loss characteristics; According to the plurality of plot water and soil loss characteristics, two-by-two coupling evaluation is performed to obtain a water and soil loss coupling evaluation sequence; Based on a water and soil loss coupling evaluation threshold, the water and soil loss coupling evaluation sequence is used to perform super-coupling correlation combing on the plurality of plot water and soil loss characteristics to obtain a water and soil loss super-coupling relationship network; According to the water and soil loss super-coupling relationship network, the first runoff plot distribution model is boundary corrected to generate the second runoff plot distribution model.

4. The power transmission project non-standard runoff plot intelligent identification modeling method of claim 1, wherein, According to the power transmission and transformation engineering construction scheme and the future meteorological data set of the power transmission and transformation engineering area, direct water and soil loss accident deduction under multiple nodes is performed on the second runoff plot distribution model to obtain a first water and soil loss deduction atlas, including: According to the power transmission and transformation engineering construction scheme, multiple node engineering construction schemes are obtained; According to the plurality of node engineering construction schemes and the future meteorological data set, a plurality of node construction scene feature vectors are established; According to the second runoff plot distribution model, an Sth runoff plot model is extracted, S being a positive integer; Based on the plurality of node construction scene feature vectors, the Sth runoff plot model is simulated for construction respectively to obtain a plurality of node plot simulation data; According to the direct water and soil loss event set, distillation loss iterative learning under balanced confrontation is performed to establish a direct water and soil loss deduction channel; The plurality of node plot simulation data is input into the direct water and soil loss deduction channel to obtain a plurality of direct water and soil loss deduction paths; The plurality of direct water and soil loss deduction paths are arranged to obtain an Sth plot direct water and soil loss accident distribution, and the Sth plot direct water and soil loss accident distribution is added to the first water and soil loss deduction atlas.

5. The power transmission project non-standard runoff plot intelligent identification modeling method of claim 4, wherein, According to the direct water and soil loss event set, distillation loss iterative learning under balanced confrontation is performed to establish a direct water and soil loss deduction channel, including: According to the direct water and soil loss event set, an accident tree is traced back to obtain a first water and soil loss tracing distribution; According to the first water and soil loss tracing distribution, a first direct water and soil loss deduction model is trained; According to the first water and soil loss tracing distribution, sample balanced evaluation is performed to obtain a first sample balanced coefficient; Based on the first sample balanced coefficient, sample balanced threshold is used to perform balanced confrontation sample injection on the first water and soil loss tracing distribution to obtain a second water and soil loss tracing distribution; According to the second water and soil loss tracing distribution, a second direct water and soil loss deduction model is trained; According to the first direct water and soil loss deduction model and the second direct water and soil loss deduction model, distillation loss iterative learning is performed to generate the direct water and soil loss deduction channel.

6. The power transmission project non-standard runoff plot intelligent identification modeling method of claim 4, wherein, According to the power transmission and transformation project construction scheme and the future meteorological data set, indirect water and soil loss accident deduction under multiple nodes is performed on the second runoff small area distribution model to obtain a second water and soil loss deduction atlas, including: According to the indirect water and soil loss event set, fault tree tracing is performed to obtain a third water and soil loss tracing distribution; According to the third water and soil loss tracing distribution, distillation loss iterative learning under balance confrontation is performed to establish an indirect water and soil loss deduction channel; The multiple node small area simulation data are input into the indirect water and soil loss deduction channel to obtain multiple indirect water and soil loss deduction paths; The multiple indirect water and soil loss deduction paths are sorted to obtain an Sth small area indirect water and soil loss accident distribution, and the Sth small area indirect water and soil loss accident distribution is added to the second water and soil loss deduction atlas.

7. The power transmission project non-standard runoff plot intelligent identification modeling method of claim 1, wherein, According to the real-time monitoring data set of the power transmission and transformation project area, twin modeling is performed, including: The power transmission and transformation project area is monitored in real time to obtain a regional data set; According to the regional data set, data cleaning is performed to obtain the real-time monitoring data set.

8. The power transmission project non-standard runoff plot intelligent identification modeling method of claim 1, wherein, An engineering area model is constructed, including: According to the real-time monitoring data set, bias detection is performed on the engineering area model to obtain a modeling bias detection result; According to the modeling bias detection result, feedback optimization is performed on the engineering area model.

9. A power transmission project non-standard runoff plot intelligent identification modeling system, characterized in that, Steps for implementing the power transmission and transformation project non-standard runoff small area intelligent identification modeling method in any one of claims 1 to 8, including: An engineering area model construction module is configured to perform twin modeling according to the real-time monitoring data set of the power transmission and transformation project area to construct an engineering area model; A first runoff small area distribution model obtaining module is configured to perform non-standard runoff small area identification and division according to the engineering area model to obtain a first runoff small area distribution model; A second runoff small area distribution model obtaining module is configured to perform water and soil loss coupling feature optimization according to the first runoff small area distribution model to obtain a second runoff small area distribution model; A first water and soil loss deduction atlas obtaining module is configured to perform direct water and soil loss accident deduction under multiple nodes on the second runoff small area distribution model according to the power transmission and transformation project construction scheme and the future meteorological data set of the power transmission and transformation project area to obtain a first water and soil loss deduction atlas; A second water and soil loss deduction atlas obtaining module is configured to perform indirect water and soil loss accident deduction under multiple nodes on the second runoff small area distribution model according to the power transmission and transformation project construction scheme and the future meteorological data set to obtain a second water and soil loss deduction atlas; A third runoff small area distribution model obtaining module is configured to perform global disturbance optimization under multiple nodes on the second runoff small area distribution model according to the first water and soil loss deduction atlas and the second water and soil loss deduction atlas to obtain a third runoff small area distribution model.

Images