Power distribution network inspection processing method and system based on unmanned aerial vehicle

Abstract

The invention provides a power distribution network inspection processing method and system based on an unmanned aerial vehicle, and belongs to the technical field of inspection management, and the method specifically comprises the steps: obtaining the inspection fault types of a coincidence risk machine nest under different weather types, and combining the similar conditions of the inspection fault types and the inspection fault types under different weather types, according to the method, the inspection control machine nest in the overlapped risk machine nest is determined, the inspection processing method of the machine nest under different weather types is determined according to the inspection control machine nest data and the matching data of the overlapped risk machine nest of the machine nest under different weather types, and the accuracy of fault identification processing is improved.

Description

Technical Field

[0001] This invention belongs to the field of inspection management technology, and in particular relates to a method and system for power distribution network inspection based on unmanned aerial vehicles (UAVs). Background Technology

[0002] To improve the efficiency of power distribution network inspection, existing technologies often utilize drones for inspection. Specifically, patent application CN202511100172.0, "A Power Distribution Network Line Health Assessment System and Method Based on Drones," describes a method that divides the power distribution network into regions based on line and environmental information, uses drones to collect images, and combines image processing techniques to extract abnormal features, thereby calculating a network line health score. This improves inspection efficiency and accuracy, thus contributing to the safe and stable operation of the power grid. However, the aforementioned technical solution has the following technical problems: When multiple drone nests exist within the same power distribution line, differences in external lighting conditions due to weather type may lead to higher identification reliability for certain fault types under specific weather conditions. Therefore, if drones in different nests are simultaneously conducting inspections, if multiple drones are found to be faulty, the cloud platform may need to process multiple drones at the same time, making it difficult to meet the identification reliability requirements. Thus, it is necessary to determine the identification reliability of drones under different weather types and avoid the technical problem of all drones focusing too much on a certain weather type for fault identification, which would result in poor fault identification reliability.

[0003] To address the aforementioned technical issues, this application provides a method and system for power distribution network inspection based on unmanned aerial vehicles (UAVs). Summary of the Invention

[0004] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides a method for power distribution network inspection based on unmanned aerial vehicles (UAVs), which includes: S1 uses the data of UAV nests in the power distribution line as a basis to determine the inspection matching area of ​​UAVs in different nests. Based on the overlap of historical fault inspection data of different nests corresponding to the inspection matching area under different weather conditions, the nests with overlapping risks are identified. When it is determined that UAV inspection control processing is required, the process proceeds to the next step. S2 obtains the inspection failure types of the overlapping risk nest under different weather types, and determines the inspection control nest in the overlapping risk nest by combining the similarity between the inspection failure types and the inspection failure types under different weather types. S3 determines the inspection and handling method for the nest under different weather conditions based on the inspection and control nest data and the matching data of nests with overlapping risk under different weather conditions.

[0005] The beneficial effects of this invention are as follows: Based on historical fault inspection data of overlapping risk nests under different weather types, it is determined whether UAV inspection and control processing is required. This achieves the assessment of the inspection and control processing requirements of UAVs in cases where the overlap of weather types with high fault identification reliability for overlapping risk nests is high. By optimizing the weather types used for inspection processing of some overlapping risk nests when the overlap of weather types with high fault identification reliability is high, the technical problem of poor fault identification reliability caused by inspecting too many nests under the same weather type is avoided.

[0006] Based on the inspection control unit data and the matching data of overlapping risk units under different weather types, the inspection and processing methods for units under different weather types are determined. This takes into account the number of inspection control units that need to be re-inspected under different weather types, the impact of not being able to troubleshoot in a timely manner on the operational reliability of the transmission line, and the number of overlapping risk units under different weather types. Therefore, in weather types with fewer overlapping risk units, the weather types used for inspection and verification are determined by comprehensively considering operational reliability. While ensuring the operational reliability of the distribution line, the optimization of the weather types for inspection also reduces the occurrence of technical problems such as poor inspection reliability caused by excessive overlapping risk.

[0007] Furthermore, the inspection matching area of ​​the UAV is determined based on the area of ​​the power distribution line inspected by the UAV.

[0008] Furthermore, the weather types are classified based on sunshine duration and wind speed, specifically, dates with sunshine duration and wind speed within the same range are classified into the same weather type.

[0009] Furthermore, the overlap of the fault inspection data is determined based on the fault type detected by the UAV in the inspection matching area of ​​the nest under the weather type.

[0010] Furthermore, the method for determining the overlapping risk nests in the nest is as follows: Based on the fault inspection data of the inspection matching area of ​​the nest under different weather types, determine the fault type obtained by the inspection matching area of ​​the nest under the weather type in history, and use it as the inspection fault type. Based on the number of times the nest identifies different inspection fault types under the weather type, the matching inspection fault type of the nest under the weather type is determined. Based on the deviation between the number of times the nest identifies the matching inspection fault type under the weather type and the number of times it identifies other weather types, the weather type of concern in the weather type is determined. By using the matched inspection fault type data of the nest under the weather type of concern, as well as the matched inspection fault type data of other nests under the weather type of concern, it is determined whether the nest is a nest with overlapping risk.

[0011] Furthermore, it was determined that drone inspection and control procedures were necessary, specifically including: Based on the overlapping risk nest data, determine the weather type of the overlapping risk nest in the power distribution line; Based on the historical fault inspection data of the overlapping risk nests under different weather types, determine the matching inspection fault type of the overlapping risk nests under the weather types of concern. Based on the weather type of the overlapping risk nest in the power distribution line and the matching inspection fault type of the overlapping risk nest under the weather type, it is determined whether inspection control processing of the UAV is required.

[0012] Secondly, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described method for power distribution network inspection based on unmanned aerial vehicles when running the computer program.

[0013] Other features and advantages will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0014] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0015] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0016] Figure 1 This is a flowchart of a power distribution network inspection and processing method based on drones; Figure 2 This is a flowchart illustrating the method for determining overlapping risk nests within a nest; Figure 3This is a flowchart for determining the need for drone inspection and control processes. Detailed Implementation

[0017] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that the invention will be thorough and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore their detailed description will be omitted.

[0018] The terms “a,” “one,” “the,” and “the” are used to indicate the existence of one or more elements / components / etc.; the terms “including” and “having” are used to indicate an open-ended meaning of inclusion and that other elements / components / etc. may exist in addition to the listed elements / components / etc.

[0019] Example 1 To solve the above problems, according to one aspect of the present invention, such as Figure 1 As shown, a method for power distribution network inspection based on unmanned aerial vehicles (UAVs) is provided, specifically including: S1 uses the data of UAV nests in the power distribution line as a basis to determine the inspection matching area of ​​UAVs in different nests. Based on the overlap of historical fault inspection data of different nests corresponding to the inspection matching area under different weather conditions, the nests with overlapping risks are identified. When it is determined that UAV inspection control processing is required, the process proceeds to the next step. Furthermore, the inspection matching area of ​​the UAV is determined based on the area of ​​the power distribution line inspected by the UAV.

[0020] Furthermore, the weather types are classified based on sunshine duration and wind speed, specifically, dates with sunshine duration and wind speed within the same range are classified into the same weather type.

[0021] Furthermore, the overlap of the fault inspection data is determined based on the fault type detected by the UAV in the inspection matching area of ​​the nest under the weather type.

[0022] In a provincial power transmission and distribution network with complex geographical and climatic conditions, the operation and maintenance department has deployed multiple intelligent drone nests to automate the inspection of thousands of kilometers of lines. Historical data shows that specific weather conditions can significantly amplify the visibility or probability of certain types of defects. For example, mechanical faults such as conductor galloping and loose fittings are more easily observed in windy weather; visual defects such as insulator contamination and cracks have higher imaging contrast under strong sunlight. Under these specific weather conditions, drones not only identify more faults, but also with higher confidence levels, resulting in more reliable data.

[0023] The core decision-making objective of this embodiment is to use inspection data generated under specific high-reliability weather conditions to intelligently filter out "overlapping risk nests" with highly similar failure modes from the entire network of nests, providing accurate early warnings for preventing regional and systemic risks.

[0024] Inspection matching area: The physical line segment covered by the drone operation range of a drone nest.

[0025] Logic: By clearly defining the geographical boundaries of the data being analyzed, it ensures that subsequent fault statistics and comparisons are conducted within physical units that are relevant to operation and maintenance management, thus avoiding mixing up data from unrelated lines.

[0026] Weather type: A classification based on historical meteorological data, which divides sunshine and wind speed into several combined intervals.

[0027] Logical reasoning: Fault visibility is strongly correlated with weather conditions. Strong light facilitates the identification of external defects (such as rust and breakage), while strong winds easily trigger and expose mechanical faults (such as swaying and loosening). This classification essentially groups data according to the "triggering conditions for fault manifestation," allowing subsequent analysis to focus on fault datasets that are more comparable and generated under the same triggering conditions.

[0028] Matching inspection fault type: For a given nest and weather type, the fault type that has been identified the most times in historical inspections.

[0029] Logical: It represents the "most prominent feature" or "most vulnerable link" of the nest in this specific environment. Using "most" instead of "average" or "sum" is to highlight the dominance and typicality of its failure mode.

[0030] Pay attention to weather type: For a given nest, its fault characteristics exhibit a high degree of "specificity" or "diversity" in weather type.

[0031] Logic: Specificity Path (Obstacle Criterion): If a certain fault is detected frequently only (or mainly) under a certain type of weather, and rarely under other weather conditions, it indicates that the fault is extremely sensitive to that type of weather condition, the data has strong indicative significance, and the reliability is high. Multivariate Path (Quantitative Criterion): If, under a certain type of weather, the number of identifications of multiple faults is significantly higher than other faults, it indicates that this type of weather is a "stress test" scenario that triggers multiple hidden dangers in the area, and it also deserves close attention.

[0032] Overlapping risk nests: Nests whose dominant failure modes significantly overlap with other nests under the "Focus on Weather Type" setting.

[0033] Logic: The failure mode of a single cell may be accidental. However, when multiple cells exhibit the same dominant failure mode (matching the inspection failure type) under the same weather conditions that are highly likely to induce failure (focus on weather type), this strongly suggests the existence of a regional common cause that extends beyond a single line segment.

[0034] Specifically, such as Figure 2 As shown, the method for determining the overlapping risk nests in the nest is as follows: S11 uses the fault inspection data of the inspection matching area of ​​the nest under different weather types to determine the fault type obtained by the inspection matching area of ​​the nest in the history under the weather type, and uses it as the inspection fault type. Step S11: Data Extraction and Fault Type Statistics. Execution Process: The system reads all inspection work orders from the past year for Cell A. Based on the weather records at the time of each work order execution, it categorizes them into preset weather types (e.g., Type I - Strong Light and Light Breeze, Type II - Moderate Wind and Rain, Type III - Strong Wind and Darkness). Then, it accumulates the total number of times different fault codes (e.g., F01: Insulator Damage, F02: Foreign Object in Wire, F03: Tree Obstruction, F04: Hardware Corrosion) are identified under each weather type.

[0035] Detailed Explanation: This is the data preprocessing stage. The core process involves cleaning and classifying the raw data according to weather conditions, constructing a structured data cube (nest × weather type × fault type × frequency) for subsequent analysis. This ensures that comparisons are made under the same environmental conditions, improving the accuracy of the analysis.

[0036] S12 determines the matching inspection fault type of the nest under the weather type based on the number of times the nest is identified under different inspection fault types under the weather type, and determines the weather type of concern in the weather type based on the deviation between the number of identifications of the matching inspection fault type under the weather type and the number of identifications under other weather types. Determine the "weather type of concern" for Nest A. Execution process: View the statistics table generated in step S11.

[0037] Under type I: F01 appears 28 times (maximum), and F02 appears 5 times. Calculate the difference between the number of times F01 appears in type I (28) and in type II (10) and type III (6): 28-10=18>12, 28-6=22>12. The deviation criterion is met.

[0038] Under Type II: F03 appeared 15 times (maximum), and F04 appeared 14 times. F03 appeared 8 times and 4 times under Type I and Type III respectively, with differences of 7 (<12) and 11 (<12) respectively, which do not meet the "both greater than" requirement. Although the matched fault types are F03 and F04 (the number is 2, equal to the threshold), it does not exceed the threshold, so the quantity criterion is not valid.

[0039] In type III: F02 appears 12 times (the most), but it also appears 5 times in type I. The difference is 7 < 12, which does not meet the requirement.

[0040] Logical Explanation: This step aims to identify the weather scenarios that are most diagnostically valuable for the nest.

[0041] For Type I: The number of times fault F01 was identified peaked under Type I weather conditions, and was significantly higher than in other weather conditions. This indicates that Type I (strong light and light wind) is the best observation window to expose the F01 (insulator damage) problem on the line under the jurisdiction of machine nest A, as the data signal-to-noise ratio is the highest and the conclusions are the most reliable under this condition.

[0042] For Type II: Although both types of faults are relatively prominent, their absolute frequency and differences from other weather conditions are not significant enough, indicating that this type of weather may just be a common high-incidence scenario rather than a specific scenario.

[0043] Conclusion: The weather type of concern for Nest A is determined to be Type I.

[0044] Furthermore, the weather type of interest refers to the weather type in which the deviation between the number of identifications under a certain matching inspection fault type and the number of identifications under other weather types is greater than a preset deviation threshold, or the weather type in which the number of matching inspection fault types is greater than a preset threshold for the number of matching fault types.

[0045] S13 uses the matching inspection fault type data of the nest under the weather type of concern, as well as the matching inspection fault type data of other nests under the weather type of concern, to determine whether the nest is a nest with overlapping risk.

[0046] It should be noted that the matching inspection fault type under the aforementioned weather type is the inspection fault type that is identified most frequently under that weather type.

[0047] Specifically, the other cellars are the cellars in the power distribution lines other than the aforementioned cellars.

[0048] Specifically, by using the matched inspection fault type data of the nest under the weather type of concern, and the matched inspection fault type data of other nests under the weather type of concern, it is determined whether the nest is a nest with overlapping risk, which specifically includes: Determine if other nests have matching inspection fault types under the weather type of concern. If yes, proceed to the next step; otherwise, determine that the nest does not belong to the overlapping risk nest. Determine whether the number of matched inspection fault types of the nest under the weather type of concern is greater than the preset threshold for the number of matched types. If yes, determine that the nest belongs to the overlapping risk nest. If no, proceed to the next step. Other nests that match the inspection fault type under the weather type of concern are selected as overlapping nests. It is determined whether the number of overlapping nests is greater than the preset threshold for the number of overlapping nests. If yes, the nest is determined to be a nest with overlapping risk. If no, proceed to the next step. The number of matching inspection fault types of the nest under the weather type of concern and the number of matching inspection fault types of overlapping nests under the weather type of concern are used as the total number of matching types. Based on the total number of matching types, it is determined whether the nest is an overlapping risk nest.

[0049] It should be noted that when the total number of matching types under the weather type of the nest is greater than the preset threshold, the nest is determined to be a nest with overlapping risk.

[0050] Determine whether nest A is a nest with overlapping risk under "Weather Type I of Concern": Step S13.1: Determine whether other machine nests have a matching inspection fault type under Type I? Execution process: Query the fault statistics of other nests (B, C, D, E) under weather type I. Findings: Nest B had matching faults of F01 (25 times), and nest C had matching faults of F03 (20 times). Nests D and E had no data or very low numbers of faults under type I.

[0051] Detailed Explanation: This is the first step in finding "potential overlapping objects". If other nests do not form a clear dominant fault under Type I (i.e., no matching fault type), it means that nest A has a low probability of overlapping inspections even if it is inspected under Type I, and therefore does not belong to the nests with overlapping risks.

[0052] Conclusion: Nests B and C exist. Proceed to the next step.

[0053] Step S13.2: Determine whether the number of matched inspection fault types for cell A under type I is greater than 1. Execution process: Nest A has a set of matching fault types {F01} under type I, with a quantity of 1.

[0054] Logical Explanation: This step assesses the complexity of its own characteristics. If a nest exhibits multiple (>1) equally dominant fault types under specific weather conditions, it indicates that its risk profile is highly complex. Regardless of whether it overlaps with others, it should be considered a high-risk object and directly marked.

[0055] Conclusion: Quantity 1 is not greater than threshold 1. Proceed to the next step.

[0056] Step S13.3: Determine if the number of "screening overlapping nests" is greater than 1? Execution process: "Filter overlapping nests" = Other nests with matching fault types under type I = {nest B, nest C}. The quantity is 2.

[0057] Detailed Explanation: This step determines the breadth of the overlapping area. If multiple other nests (>1) have a clear fault mode under the same weather condition of concern, then even if nest A's own mode is simple, it is in a "common fault atmosphere." This indicates that the fault (F01) may not be unique to nest A, but rather part of a broader regional phenomenon. Triggering this condition will trigger an alert.

[0058] Conclusion: Quantity 2 > Threshold 1. The condition is met, and nest A is determined to be a nest with overlapping risk.

[0059] Step S13.4: Determine based on "Total Number of Matching Types" (Alternative Path): Detailed Explanation of the Logic: This step is a fine-grained quantitative judgment used when the overlap range is not wide (step 3 is not triggered). It calculates the total number of unique fault types for all relevant nests (their own nests + the selected overlapping nests) under the weather conditions under the focus.

[0060] Hypothetical scenario: If only nest B overlaps with nest A (both have matching fault F01), then the number of "screening overlapping nests" is 1, which is not greater than the threshold, and proceed to this step.

[0061] Execution process: Total number of matching types = {F01 of Nest A} ∪ {F01 of Nest B} = {F01}, the total number is 1.

[0062] Logical Explanation: A small total number (≤3) indicates that although nest B and A have the same failure mode, the overall failure type involved is simple and may only be a local common problem of the lines under the jurisdiction of these two nests. Therefore, the risk of synchronous identification on the cloud platform is low.

[0063] Conclusion: 1 ≤ preset type quantity threshold 3. Therefore, under this hypothetical scenario, nest A does not belong to the overlapping risk nest.

[0064] This embodiment is a follow-up decision-making step to the preceding "overlapping risk nest identification method". In the preceding analysis, the system has identified several "overlapping risk nests" from the entire network. These nests exhibit similar failure modes to other nests under specific "weather types of concern".

[0065] The current decision-making objective is to determine whether to initiate global inspection control measures (e.g., limiting concurrent inspection tasks for certain drone nests under specific weather conditions). The core purpose is to balance cloud-based AI image recognition resources. When a large number of drones simultaneously transmit massive amounts of images under the same weather conditions (especially for high-risk faults of the same type), the pressure on cloud-based concurrent processing increases dramatically, potentially leading to recognition delays, decreased efficiency, or even system overload. Therefore, a decision needs to be made regarding whether to implement proactive control measures.

[0066] Focus on weather type: The weather type with the most significant fault characteristics determined from the perspective of each overlapping risk nest individual (such as "Type I - strong light and light wind" in the previous example).

[0067] Matching inspection fault type: Under the weather type of concern, this is the fault type that is identified most frequently by the overlapping risk nest.

[0068] Inspection control and processing: refers to the global scheduling and intervention of drone inspection tasks. For example, when a specific weather is predicted, the inspection tasks of drone nests belonging to the same "weather of concern" are proactively reduced or staggered to alleviate the instantaneous processing pressure on the cloud.

[0069] Disruptive Risk Nests: In the final screening, those individuals that, while belonging to overlapping risk nests, did not exceed the threshold in terms of the total number of nests under their respective weather types are considered potential "disruptors" and require further evaluation of the concentration of their risk patterns.

[0070] Identification Matching Factor: Used to quantify the concentration of fault modes under a specific "weather type of concern". The calculation formula is: (sum of the number of matching fault types for interference risk nests) / (number of all possible inspection fault types under this weather type). The higher the ratio, the more concentrated the risk is on a few types of faults under this weather condition.

[0071] Specifically, such as Figure 3 As shown, it has been determined that inspection and control processing of the drone is required, specifically including: S21 determines the weather type of concern for the overlapping risk nests in the power distribution line based on the overlapping risk nest data; Execution process: The system reads data from all overlapping risk nests and extracts the "weather type of concern" attribute for each nest.

[0072] Logical Explanation: This is the data preparation step, which summarizes the individual conclusions from the previous steps into a global list, serving as input for subsequent population analysis.

[0073] Output: List of weather types to watch = {Type I, Type I, Type I, Type II, Type II, Type III}.

[0074] S22 determines the matching inspection fault type of the overlapping risk nest under the weather type of concern based on the historical fault inspection data of the overlapping risk nest under different weather types. Determine the type of fault to be matched during inspection Execution process: For each overlapping risk nest, associate it with the "matching inspection fault type" determined in the previous analysis under its respective "weather type of concern".

[0075] Logical Explanation: This step binds the nest to its most typical risk characteristics (failure type), so that subsequent analysis can not only see under what weather conditions the risk is concentrated, but also what the risk is.

[0076] Table 1 Matching Fault Types

[0077] S23 determines whether UAV inspection control processing is required based on the weather type of the overlapping risk nest in the power distribution line and the matching inspection fault type of the overlapping risk nest under the weather type.

[0078] It should be noted that this includes the following specific situations: Scenario 1: If the number of overlapping risk drone nests is greater than the preset threshold for the number of overlapping risk drone nests, it is determined that drone inspection control processing is required, thereby reducing the number of drone nests to be inspected under certain weather types of concern, and ensuring the efficiency and reliability of image recognition processing of multiple drone nests on the cloud platform at the same time.

[0079] Scenario 1: Is the overall quantity too large? Execution process: Count the total number of overlapping risk nests. Currently, there are 6. Compare this with the preset threshold for overlapping risk nests, K_total=5. 6>5.

[0080] Detailed Explanation: This is the top-level, global assessment. If the total number of overlapping risk nests is too high (more than 5), it indicates a widespread, cross-regional overlap of fault modes across the entire network. In this situation, regardless of the specific weather conditions these nests are concentrated in, predicting these "weather types of concern" could trigger a large number of nests to simultaneously perform high-value inspections and transmit data, placing significant concurrent pressure on the cloud. To ensure the stability of core services (cloud identification efficiency and reliability), global inspection control must be initiated in advance to stagger or limit task activity.

[0081] Decision output: Inspection and control of the drone is required. Process complete.

[0082] Scenario 2: If the number of overlapping risk drone nests is not greater than the preset threshold for the number of overlapping risk drone nests, deduplication is performed based on the weather type of the overlapping risk drone nests to obtain the number of weather types of interest. If the number of weather types of interest is greater than the preset threshold for the number of weather types of interest, it is determined that drone inspection control processing is required, thereby reducing the number of drone nests to be inspected under some weather types of interest, and ensuring the efficiency and reliability of image recognition processing of drones in multiple nests on the cloud platform at the same time.

[0083] Scenario 2 Assessment: Are the types of high-risk weather too dispersed? Execution process: Based on the list of weather types of concern with overlapping risk nests, deduplication is performed. This results in a unique set of weather types of concern: {Type I, Type II, Type III}. The number of these types is calculated to be 3. This is compared to the preset weather type quantity threshold W_th = 2. 3 > 2.

[0084] Detailed Explanation: When the total number of risky data nests is not large, it's necessary to check if they are concentrated under a few weather types. If there are too many risky weather types (more than two), it means that for a long period of time (covering multiple weather conditions), data nests may be in their "high-risk period," causing the cloud to frequently handle high-value data streams from different weather scenarios, resulting in persistently high load. To reduce the pressure on the system from being in a high-concurrency preparedness state for a long time, control measures and optimized task scheduling are needed.

[0085] Decision output: Inspection and control of the drone is required. Process complete.

[0086] Scenario 3: If the number of weather types of concern is not greater than the preset threshold for the number of weather types, then the number of overlapping risk drone nests belonging to different weather types is determined. When there is a weather type where the number of overlapping risk drone nests belonging to different weather types of concern is greater than the preset threshold for the number of overlapping drone nests, then it is determined that drone inspection control processing is required, thereby reducing the number of drone nests to be inspected under some weather types of concern, and ensuring the efficiency and reliability of image recognition processing of multiple drone nests on the cloud platform at the same time.

[0087] If the preset threshold for the number of weather types is 3, then case 3 is judged as follows: Is there a high concentration of risk nests under a single weather type? Execution process: For each weather type, count the number of overlapping risk nests belonging to that "weather type of concern".

[0088] Type I: There are 3 types: RC-01, RC-02, and RC-03.

[0089] Type II: There are RC-04 and RC-05, a total of 2.

[0090] Type III: There is RC-06, 1 in total.

[0091] Judgment: Check if the number of overlapping nests under any weather type is greater than the preset threshold C_th=3. The number (3) under type I is equal to (not greater than) the threshold 3, so the condition is not met.

[0092] Detailed Explanation: When there are few types of risky weather, further investigation is needed to determine if any particular weather condition could trigger a "cluster" risk. If a large number (>3) of risky cloud clusters are concentrated under a certain type of weather (such as Type I), then once this type of weather is forecasted, these clusters may simultaneously initiate inspections, instantly placing peak pressure on the cloud. To avoid such instantaneous overload under specific weather conditions, control measures are required.

[0093] Decision output: Case 3 not triggered. Case 4 is entered.

[0094] Scenario 4: When there is no weather type with a number of overlapping risk drone nests that are greater than the preset threshold for the number of overlapping drone nests, the overlapping risk drone nests that are overlapping risk drone nests under the weather type will be regarded as interference risk drone nests. Based on the number of matching inspection fault types of the interference risk drone nests under the weather type, it will be determined whether inspection control processing of the drone is required.

[0095] It should be noted that, based on the number of matched inspection fault types for the drone nest under the aforementioned weather type, it is determined whether drone inspection and control processing is required. Specifically, this includes: The identification matching factor for interference risk nests under the weather type is determined by the ratio of the number of matching inspection fault types to the number of inspection fault types. When the sum of the identification matching factors of the weather type with interference risk exceeds the preset matching factor threshold, it is determined that inspection and control processing of the drone is required.

[0096] Assess the concentration of risk patterns (identify matching factors): Execution process: Identify interference risk nests: Since the number of overlapping risk nests does not exceed C_th(3) under types I, II, and III, these three nests RC-01, RC-02, RC-03, RC-04, RC-05, and RC-06 are all considered "interference risk nests" under the current judgment level.

[0097] Grouping and calculating matching factors by weather type: Weather type I: Interference risk nest = {RC-01, RC-02, RC-03}. Their matched fault types are {F01, F01, F03}.

[0098] Assume that the total number of inspection fault types defined by the system is 10 (F01 to F10).

[0099] The identification matching factor (type I) for interference risk nests is 1 / 10 = 0.1.

[0100] Weather type II: Interference risk nest = {RC-04, RC-05}. Matching fault type is {F02, F02}, with a duplicate count of 1.

[0101] The identification matching factor (Type II) for interference risk nests = 1 / 10 = 0.1.

[0102] Weather type III: Interference risk nest = {RC-06}. Matching fault type is {F04}, with a duplicate count of 1.

[0103] Identification matching factor (Type III) = 1 / 10 = 0.1.

[0104] Judgment: Calculate the sum of the identification matching factors of "interference risk nests" under each weather type. Here, "sum" refers to the sum of factor contributions of all interference risk nests under the same weather type.

[0105] Check: Type I factor 3 multiplied by 0.1 = 0.3 < 0.7, Type II factor 0.2 < 0.7, Type III factor 021 < 0.7.

[0106] Detailed Explanation: Scenario 4 is an in-depth assessment for risks that are neither "widespread" (Scenario 1), "frequent" (Scenario 2), nor "clustered" (Scenario 3). It focuses on the singularity or concentration of risk patterns among different interference risk nests under specific weather conditions. A low identification matching factor (e.g., 0.2) means that the risk under this weather condition is concentrated on a few faults. Even if concurrent anomalies exist, their impact on the overall reliability of fault identification is low, therefore no inspection and control measures are required.

[0107] Decision output: The matching factor for all weather types is no greater than 0.7. Therefore, it is determined that no UAV inspection control processing is required. The system can maintain the current inspection concurrency strategy.

[0108] S2 obtains the inspection failure types of the overlapping risk nest under different weather types, and determines the inspection control nest in the overlapping risk nest by combining the similarity between the inspection failure types and the inspection failure types under different weather types. 1. Implementation scenario: In the same regional power grid described in embodiment S22, the system has identified six overlapping risk cells (RC-01, RC-02, RC-03, RC-04, RC-05, and RC-06) through preliminary analysis, and clarified their respective weather types of concern. Now, the operation and maintenance decision-making enters the next stage: it is necessary to identify which of these six cells require the application of more refined control strategies.

[0109] 2. Decision-making objectives: The core decision-making objective of this embodiment is to perform a secondary screening based on the identified overlapping risk nests (RC-01 to RC-06) to locate the inspection control nest.

[0110] Target identification: Establish rules to automatically identify nests whose "signature faults" may be lurking under other weather conditions but are not adequately identified.

[0111] Control Objective: Implement a "trigger-verification" mechanism for the marked "inspection control nests". When an alarm is triggered under the weather of interest, the nest will be automatically scheduled to be verified under other weather conditions to confirm the cross-weather identifiability of the fault. This will ultimately provide data to dynamically optimize the inspection plan for the nest and reduce over-reliance on a single high-risk weather window.

[0112] Specifically, the method for determining the inspection control cell in the aforementioned cell is as follows: Matching inspection fault types: The fault type that has been identified most frequently in the history of overlapping risk nests under their "weather type of concern".

[0113] Other identification counts: refers to the total number of historical inspection identifications for this matched inspection fault type under other weather types besides 'weather of concern'.

[0114] Inspection control nests: These refer to overlapping risk nests that need to be "controlled" and have their cross-weather identification reliability verified. Their characteristics are: while the matched fault type appears in other weather conditions (indicating the fault exists), the frequency of occurrence may be unstable or relatively low, suggesting that the current identification system's reliability in detecting this fault under other weather conditions is questionable.

[0115] Reliable identification: This refers to the identification of "matching inspection fault types" at a similar confidence level as under "weather types of concern" conditions, even when the weather type is not of concern. Verification typically requires manual review or enhanced AI analysis.

[0116] Entity definition (completely consistent with S22): Nests: RC-01, RC-02, RC-03, RC-04, RC-05, RC-06.

[0117] Weather type: Type I (Strong Light, Light Breeze): Light intensity > 80,000 Lux, wind speed < level 3.

[0118] Type II (moderate wind with rain): Sunshine intensity 40,000-60,000 Lux, wind speed 3-4, accompanied by precipitation.

[0119] Type III (Windy and Dry): Sunlight > 70,000 Lux, wind speed > level 5.

[0120] Fault types (used and refined): F1 (Insulator spontaneous explosion / crack): Damage to the porcelain or glass insulator caused by aging or stress.

[0121] F2 (Broken strand in conductor or ground wire): A broken metal strand appears on the outer or inner layer of the conductor or ground wire.

[0122] F3 (New structures within the corridor): Newly built illegal houses, tower cranes, etc. have appeared within the line protection zone.

[0123] S31 uses the inspection fault type data of the overlapping risk nest under the weather type of concern to determine the matching inspection fault type of the overlapping risk nest under the weather type of concern. Determine the fault type of RC-01 during the matching inspection: Execution process: Statistically analyze the historical fault data of RC-01 under Type I (strong light and light wind).

[0124] Example data: F1 (Insulator Self-Destruction): 36 times identified F2 (Conductor Wire Breakage): Identified 18 times F3 (Add Structure): Recognized 7 times Detailed Explanation: This step is a risk profile. Under the optimal imaging conditions of "Type I," insulator problems (F1) on the lines under RC-01 are most frequently detected. This indicates that insulator defects are the primary risk characteristic of this line under high visibility conditions.

[0125] Output: Matching inspection fault type of RC-01 = F1.

[0126] S32 determines the number of identifications under different weather types based on the matching inspection fault types under the weather types of concern, the number of historical inspections under weather types that do not belong to the weather types of concern, and uses these as other identifications. Determine the "other identification counts" for matching fault F1. Execution process: Count the number of times fault F1 is identified under type II (moderate wind with rain) and type III (strong wind and dry).

[0127] Example data: In Type II, F1 was recognized 5 times.

[0128] In Type III, F1 was identified once.

[0129] Detailed Explanation: This step involves cross-weather exploration. Data shows that even under poor imaging conditions (Type II, with rain), insulator problems (F1) were still recorded 5 times; while under windy and dry conditions (Type III), almost no records were found. This leads to a crucial conclusion: insulator defects are persistent (as they were also found under Type II), but their identification probability may be highly dependent on weather conditions (most frequent under Type I, least frequent under Type III). This dependence needs to be verified and quantified.

[0130] Output: Other recognition counts for F1: 5 for Type II, 1 for Type III.

[0131] S33 determines whether the overlapping risk nest is an inspection control nest based on the matching inspection fault type of the overlapping risk nest under the weather type of concern and the number of other identifications of different matching inspection fault types.

[0132] It should be noted that the above steps specifically include: S331 determines the matching inspection fault type of the overlapping risk nest under the weather type of concern, and whether the proportion of the number of inspection fault types of the overlapping risk nest is greater than the preset proportion threshold. If yes, the overlapping risk nest is determined to be an inspection control nest; otherwise, proceed to the next step. S332 uses the number of other identifications for different matching inspection fault types to determine whether there are other matching inspection fault types with a number of identifications that meet the requirements (i.e., greater than the preset number threshold). If yes, proceed to the next step; otherwise, determine that the overlapping risk nest does not belong to the inspection control nest. S333 determines the total number of times the matching inspection fault types are identified under different weather conditions based on other matching inspection fault types that meet the identification requirements. Based on the total number of times the matching inspection fault types are identified under different weather conditions, it determines whether the overlapping risk nest is an inspection control nest.

[0133] S331: Determine if the overall proportion of fault F1 is too high (>60%)? Execution and Judgment: Assume that the total number of failures of RC-01 is 95, and the total number of failures of F1 is 36+5+1=42, accounting for 44.2% < 60%.

[0134] Detailed Explanation of the Logic: This step is a quick screening process. Since F1 is not the absolute dominant defect in this aircraft cluster (accounting for less than 60%), it cannot be used as the sole basis for deciding to implement cross-weather-specific control; a more refined rule analysis is necessary.

[0135] Decision: Proceed to S332.

[0136] S332: Determine if there is a matching fault where “other recognition times” > C1 (4 times)? Execution and Judgment: Check F1's "Other Recognition Counts": 5 times > 4 times under Type II (satisfied), 1 time < 4 times under Type III (not satisfied).

[0137] Logical Explanation: This step establishes a validity threshold. Reaching 5 times under Type II, exceeding the threshold of 4 times, proves that the phenomenon of "insulator defects being effectively identified to a certain extent even under non-ideal weather conditions (Type II)" is not accidental and is worthy of further analysis.

[0138] Decision: There is a weather type that meets the requirements (Type II), proceed to S333.

[0139] S333: Final determination based on weather type that meets the requirements: Execution process: The set of weather types that meet the requirements is: {Type II}.

[0140] Get the "total number" of F1 in other weather types: 5 + 1 = 6 times.

[0141] Determine if the number of times is greater than C2 (2 times): 6 > 4, so it is true.

[0142] Detailed Explanation of the Logic: This step is for stability verification. It requires that the matched fault not only occur a valid number of times (S332) under at least one other weather condition, but also have a certain frequency (>4 times) to confirm its stability. The F1 fault of RC-01 occurred 6 times under Type II, which is more than the threshold for stability.

[0143] Final decision: All conditions are met, RC-01 is determined to be the "inspection control unit".

[0144] IV. Synergistic Value Achieving "Dual Weather Dimension Collaboration" in Risk Monitoring: Traditional monitoring strategies may only intensify inspections during high-incidence weather (focusing on weather types). This method, by identifying "inspection control nests" and performing cross-weather verification, collaborates with "high-incidence weather" and "high-reliability identification weather." For example, for RC-01, after an alarm in "Type I" (high incidence, high identification rate), the system will proactively re-verify in "Type I" or "Type II," thereby cross-validating the same fault under different meteorological conditions, significantly improving the accuracy and robustness of defect diagnosis.

[0145] The system drives the collaborative evolution of inspection strategies from "static planning" to "dynamic intelligent strategies": Verification results are directly used to optimize strategies. If verification shows that F1 can be reliably identified even under "Type II" weather conditions, the system can generate a collaborative strategy: "For insulator defects (F1) of RC-01, when resources are scarce in "Type I" weather, some of its inspection tasks can be adjusted to be performed in "Type II" weather windows, thereby balancing the inspection load across the entire network." This makes the inspection plan no longer a fixed schedule, but an intelligent system that can dynamically adjust based on fault characteristics, identification reliability, and system resources.

[0146] Achieving a closed-loop collaboration between "data acquisition" and "algorithm optimization": Fault image data collected across weather conditions, showing the same location but different weather, is a valuable asset for optimizing AI recognition algorithms. By comparing the image features of the same insulator in sunny and rainy weather, the algorithm can be trained specifically to improve its recognition capabilities under complex weather conditions. This forms an enhanced closed loop of "algorithm discovering potential problems -> triggering verification and data acquisition -> data feedback to optimize the algorithm."

[0147] Promoting O&M collaboration between "point-based response" and "systemic defense": The value of verifying a single "inspection control unit" can be extended to similar units. If the verification conclusion of RC-01 (such as "the confidence level of F1 fault identification decreases by 20% in rainy and foggy weather") is generalized, the O&M department can collaboratively develop targeted defense measures, such as conducting a special inspection of insulators on all similar lines before the rainy and foggy season, thus elevating fault handling from passive response to systemic prevention.

[0148] It should be noted that, based on the total number of times matching inspection fault types were identified under different weather conditions, the determination of whether the overlapping risk nest is an inspection control nest specifically includes: When there are matching inspection fault types with a total number of occurrences exceeding a preset threshold under different weather conditions, the overlapping risk nest is determined to be an inspection control nest.

[0149] It should be noted that if the overlapping risk nest is an inspection control nest, then when a matching inspection fault type occurs under the weather type of concern, inspection processing will be carried out under different weather types to determine whether the matching inspection fault type under the weather type of concern can be reliably identified under other weather types.

[0150] S3 determines the inspection and handling method for the nest under different weather conditions based on the inspection and control nest data and the matching data of nests with overlapping risk under different weather conditions.

[0151] Scenario: Within the same regional power grid, we possess the following "risk profile" data derived from prior analysis: List of all hives: RC-01, RC-02, RC-03, RC-04, RC-05, RC-06 (6 in total).

[0152] Overlapping risk nest set: {RC-01, RC-02, RC-03, RC-04, RC-05, RC-06} (all 6 are overlapping risk nests).

[0153] Inspection control nest set: Assuming that, based on the previous embodiment, {RC-01, RC-04} are marked as inspection control nests.

[0154] Weather types of concern for each satellite constellation (from previous steps such as S22): The weather type of concern for RC-01, RC-02, and RC-03 is Type I.

[0155] The weather type of concern for RC-04 and RC-05 is Type II.

[0156] The weather type of concern for RC-06 is Type III.

[0157] Objective: Based on the above profile, develop refined inspection and handling methods for each nest under different weather conditions (e.g., key inspection, routine inspection, suspended inspection, trigger verification, etc.).

[0158] Specifically, the method for determining the inspection and handling methods for the aircraft nest under different weather types is as follows: S41 uses the inspection control unit data to determine the proportion of inspection control units in the power distribution line's units and uses it as the control unit proportion. Calculate the control nest ratio. Execution process: Number of control nests = 2 (RC-01, RC-04). Total number of nests = 6.

[0159] Calculation: Control nest ratio = 2 / 6 ≈ 33.3%.

[0160] Detailed Explanation: This ratio reflects the percentage of nests across the entire network that require the "trigger-verify" high-level control strategy. A higher ratio indicates a more complex overall network risk due to untimely fault handling, necessitating a more cautious global strategy.

[0161] S42 determines the overlapping risk nests belonging to the weather type of concern in different weather types based on the matching data of the nests in different weather types. Define "focused nests" for each weather type. The process involves categorizing data by weather type based on input data. The three watch nests under weather type I (i.e., overlapping risk nests under weather type I) are RC-01, RC-02, and RC-03.

[0162] Two aircraft nests of concern under Weather Type II: RC-04 and RC-05.

[0163] Attention nest under Weather Type III: RC-06, 1 in total.

[0164] Logical Explanation: This step strongly correlates nest risk with specific weather conditions, forming a combat map that shows "which nests are at the highest risk under what weather conditions".

[0165] S43 uses the controlled nest ratio and the overlapping risk nests belonging to the weather type of concern in different weather types to determine the inspection and handling method of the nests under different weather types.

[0166] Furthermore, if the nest is a nest with overlapping risk, then inspection processing is only required in weather types where there is a matching inspection fault type, and when a fault occurs, there is no need to perform verification and identification processing under different weather types.

[0167] For overlapping risk nests (this rule applies to all 6 nests): Rule: "Inspection processing should only be performed in weather types where there is a matching inspection fault type, and when a fault occurs, there is no need to perform verification and identification processing under different weather types." Detailed Explanation: Overlapping risk nests have already identified clear and high-frequency failure modes under specific weather conditions. Therefore, resources should be concentrated on their "high-incidence windows" (i.e., their weather types of concern). For example, for RC-02 with weather type I of concern, inspections should be prioritized under type I weather to most efficiently capture its typical failure (F1). Once a failure is detected, since the failure has been sufficiently verified as high-incidence under this weather condition, it can directly enter the handling process without needing to initiate cross-weather verification as with "inspection control nests," thereby simplifying the process and improving efficiency.

[0168] Application example: For RC-02 (Concern Type I): Primary inspection resources are allocated to weather type I. Routine or low-frequency inspections can be performed under types II and III.

[0169] For RC-05 (Concern Type II): The main inspection resources are arranged for weather type II.

[0170] Additionally, it should be noted that if the nest is not a nest with overlapping risk, then the control nest ratio is determined, and it is judged whether the control nest ratio is greater than the preset control nest ratio threshold. If so, inspection processing is only required in weather types with matching inspection fault types, and when a fault occurs, there is no need to perform verification and identification processing under different weather types, thereby ensuring the operational reliability of the power distribution line. If not, proceed to the next step. For non-overlapping risk nests (which do not exist in this example, we will extrapolate), we will conduct an analysis. Rule deduction: If a nest X does not belong to a nest with overlapping risk, then the decision is made according to the following sub-steps: Is the control nest ratio greater than 30%? The current ratio is 33.3%, which is greater than 30%, so the condition is met.

[0171] Decision: For non-risk nest X, the strategy is to "focus on inspections only during (hypothetically existing) matching fault weather, and no verification is required after the fault occurs".

[0172] Detailed Explanation: When more than 30% of the network's data centers are identified as requiring complex control (inspection and control of data centers), it indicates a severe overall risk environment. To concentrate limited "verification resources" on these high-uncertainty data centers, a simplified, conservative, yet efficient strategy should be adopted for low-risk, non-overlapping risk data centers. This means "focusing on the main issues" and ensuring inspections are conducted during their primary risk weather conditions, avoiding the diversion of operational efforts due to overly complex verification processes.

[0173] Using the overlapping risk nest data of different weather types belonging to the weather type of concern, determine the number of overlapping risk nests belonging to the weather type of concern in the weather type. Determine whether the number of overlapping risk nests belonging to the weather type of concern in the weather type is less than a preset nest number threshold. If yes, proceed to the next step; otherwise, determine that no inspection will be performed in the weather type.

[0174] Based on the overlap between the overlapping risk nests belonging to the weather type of concern in the aforementioned weather type and the overlapping risk nests belonging to the weather type of concern in other weather types, the number of overlapping risk nests belonging to the weather type of concern in the aforementioned weather type and other weather types is determined. Based on the maximum value of the number of overlapping risk nests under other weather types, an interference value is determined. Based on the interference value, the inspection and handling method of the nest under different weather types is determined.

[0175] Specifically, when the interference value is less than the preset interference threshold, it is determined that the nest can only be inspected and re-examined under the specified weather type.

[0176] (Déjà vu) If the control nest ratio is ≤ 30%, proceed to the next step and analyze the number of nests of concern for each weather type.

[0177] Develop a global strategy for a specific weather type (taking weather type III as an example). When it is necessary to assess whether to conduct inspections under a certain weather type, the system will perform the following analysis (this analysis serves the overall strategy formulation, such as deciding whether to suspend all inspections under Type III weather): Determine if the number of watch nests under this weather type is less than 2. For weather type III, the number of nests to watch is 1 (RC-06).

[0178] 1 < 2, the condition is met. Proceed to the next step.

[0179] Calculate the "interference value": Concept: The interference value of weather type A = the maximum number of other weather types that share the same attention nest with weather type A.

[0180] Calculation process (for weather type III): The weather type III watchdog is {RC-06}.

[0181] Check other weather types: For Type I, the aircraft nests of concern are {RC-01, RC-02, RC-03}, which do not overlap with Type III. For Type II, the aircraft nests of concern are {RC-04, RC-05}, which also do not overlap with Type III.

[0182] Therefore, the overlap between weather type III and other weather types is 0. Its interference value = 0.

[0183] Determine if the interference value is <1? If 0 < 1, the condition is met.

[0184] Decision: It is determined that the nest can be inspected and processed under weather type III.

[0185] Detailed Explanation: Weather type III has only one "high-risk customer" (RC-06), and this customer is exclusively its responsibility (interference value = 0). This means that conducting inspections under weather type III has a very clear service target (RC-06) and will not conflict with inspection targets in other weather types. Therefore, the system can schedule inspections and perform follow-up checks when faults occur, specifically serving the high-risk monitoring of RC-06 under its weather of interest. Conversely, if the interference value is high (e.g., = 2), it indicates that the nest is also a high-risk target in other weather types, resulting in ineffective inspections under multiple weather types. In this case, reducing or canceling independent inspections under that weather type should be considered.

[0186] Example 2 Secondly, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described method for power distribution network inspection based on unmanned aerial vehicles when running the computer program.

[0187] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0188] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0189] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for power distribution network inspection and processing based on unmanned aerial vehicles (UAVs), characterized in that, Specifically, it includes: Based on the data of UAV nests in power distribution lines, the inspection matching areas of UAVs in different nests are determined. Based on the overlap of historical fault inspection data of different nests corresponding to the inspection matching areas under different weather conditions, the nests with overlapping risks are identified. Based on the historical fault inspection data of nests with overlapping risks under different weather conditions, when it is determined that UAV inspection control processing is required, the next step is initiated. Obtain the inspection failure types of the overlapping risk nests under different weather conditions, and determine the inspection control nests in the overlapping risk nests by combining the similarity between the inspection failure types and the inspection failure types under different weather conditions. Based on the inspection control nest data and the matching data of overlapping risk nests under different weather types, the inspection and handling methods for the nests under different weather types are determined.

2. The method for power distribution network inspection and processing based on unmanned aerial vehicles as described in claim 1, characterized in that, The inspection matching area of ​​the drone is determined based on the area of ​​the power distribution line inspected by the drone.

3. The method for power distribution network inspection and processing based on unmanned aerial vehicles as described in claim 1, characterized in that, The weather types are categorized based on sunlight intensity and wind speed.

4. The method for power distribution network inspection and processing based on unmanned aerial vehicles as described in claim 1, characterized in that, The overlap of the fault inspection data is determined based on the fault type detected by the UAV in the inspection matching area of ​​the nest under the weather type.

5. The method for power distribution network inspection and processing based on unmanned aerial vehicles as described in claim 1, characterized in that, The method for determining the overlapping risk nests in the nest is as follows: Based on the fault inspection data of the inspection matching area of ​​the nest under different weather types, determine the fault type obtained by the inspection matching area of ​​the nest under the weather type in history, and use it as the inspection fault type. Based on the number of times the nest identifies different inspection fault types under the weather type, the matching inspection fault type of the nest under the weather type is determined. Based on the deviation between the number of times the nest identifies the matching inspection fault type under the weather type and the number of times it identifies other weather types, the weather type of concern in the weather type is determined. By using the matched inspection fault type data of the nest under the weather type of concern, as well as the matched inspection fault type data of other nests under the weather type of concern, it is determined whether the nest is a nest with overlapping risk.

6. The method for power distribution network inspection and processing based on unmanned aerial vehicles as described in claim 5, characterized in that, The weather type of interest is a weather type in which the deviation between the number of identifications under a certain matching inspection fault type and the number of identifications under other weather types is greater than a preset deviation threshold, or a weather type in which the number of matching inspection fault types is greater than a preset threshold for the number of matching fault types.

7. The method for power distribution network inspection and processing based on unmanned aerial vehicles as described in claim 5, characterized in that, The matching inspection fault type under the weather type is the inspection fault type that is identified most frequently under the weather type.

8. The method for power distribution network inspection and processing based on unmanned aerial vehicles as described in claim 1, characterized in that, The method for determining the inspection control cell in the aforementioned cell is as follows: Based on the inspection fault type data of the overlapping risk nests under the weather type of concern, determine the matching inspection fault type of the overlapping risk nests under the weather type of concern; Based on the matching inspection fault type under the weather type of concern, the number of historical inspections under weather types that do not belong to the weather type of concern is used to determine the number of identifications under different weather types and these are used as other identifications. The overlapping risk nest is determined as an inspection control nest based on the matching inspection fault type under the weather type of concern and the number of other identifications for different matching inspection fault types.

9. The method for power distribution network inspection and processing based on unmanned aerial vehicles as described in claim 1, characterized in that, The method for determining the inspection and handling methods for the aircraft nest under different weather types is as follows: Based on the inspection control unit data, determine the proportion of inspection control units in the units of the power distribution line, and use it as the control unit proportion; Based on the matching data of overlapping risk nests in different weather types, overlapping risk nests belonging to the weather type of concern are identified in different weather types. By using the controlled nest ratio and the overlapping risk nests belonging to the weather type of concern in different weather types, the inspection and handling methods of the nests under different weather types are determined.

10. A computer system, comprising: A memory and processor connected by communication, and a computer program stored in the memory and capable of running on the processor, characterized in that, when the processor runs the computer program, it executes a UAV-based power distribution network inspection processing method according to any one of claims 1-9.

Images