Power distribution network inspection management method and system

Abstract

The invention relates to the technical field of power equipment state monitoring, in particular to a power distribution network inspection management method and system. The method comprises the following steps: obtaining tower structure ledger data of a power distribution network line, and determining an electrical component pair meeting a preset symmetry condition in a tower; the unmanned aerial vehicle obtains an ultraviolet image sequence and records spatio-temporal information; determining a symmetric analysis area in the ultraviolet image sequence according to the tower structure standing book data and the space-time information; calculating a relative discharge index of the electrical component pair in a preset ultraviolet inspection task according to the symmetric analysis area; and judging the relative discharge index, if the judgment result is abnormal, performing trend analysis on the relative discharge index sequence, and generating a maintenance decision instruction to be included in an inspection management process. According to the invention, relative discharge index calculation and trend analysis are carried out on the electrical component pair, so that the stability and comparability of ultraviolet inspection data are improved, and the reliability of defect identification and state evaluation is enhanced.

Description

Technical Field

[0001] This invention relates to the field of power equipment condition monitoring technology, specifically a distribution network inspection and management method and system. Background Technology

[0002] Using drones equipped with solar-blind ultraviolet imagers to detect discharge insulators, fittings, and other components of power distribution lines has become an important method for discovering insulation defects. The existing conventional method involves controlling the drone to photograph the target component, analyzing absolute indicators such as light spot intensity, area, or photon count in the ultraviolet image, and comparing them with preset thresholds to determine the presence of discharge defects.

[0003] However, in practical applications, the intensity of ultraviolet signals is easily affected by instantaneous meteorological conditions such as air humidity, ambient temperature, and background light. This results in significant fluctuations in the absolute values ​​of the same equipment component measured at different times and under different environments, lacking direct comparability. This makes defect identification and trend judgment based on absolute value comparisons unreliable, easily leading to false alarms or missed alarms. If inspections are strictly limited to a few similar meteorological conditions to ensure data comparability, it will severely restrict the window period and execution efficiency of inspection operations, making it difficult to meet the management needs of routine, high-frequency condition monitoring and assessment of a large power distribution network.

[0004] Therefore, how to obtain quantitative assessment results that can be compared over a long period of time and truly reflect changes in equipment status without relying on harsh environmental conditions is a contradiction that needs to be solved in the application of ultraviolet inspection technology to the lean management of power distribution networks. Summary of the Invention

[0005] The purpose of this invention is to provide a distribution network inspection management method and system to solve the problem that existing ultraviolet inspection data fluctuates greatly under different environmental conditions and is difficult to compare stably over a long period of time, resulting in insufficient reliability of condition assessment and maintenance decisions.

[0006] To achieve the above objectives, in one aspect, the present invention provides a distribution network inspection and management method, the method comprising: Step S1: Obtain the tower structure ledger data of the power distribution network line; determine the electrical component pairs in the tower that meet the preset symmetry conditions according to the tower structure ledger data, the electrical component pairs include a first component and a second component; the UAV performs a preset ultraviolet inspection task to obtain an ultraviolet image sequence containing the first component and the second component, and records the spatiotemporal information.

[0007] Step S2: Based on the tower structure ledger data and spatiotemporal information, determine the symmetrical analysis region in the ultraviolet image sequence; based on the symmetrical analysis region, calculate the relative discharge index of the electrical components in the preset ultraviolet inspection task; determine the relative discharge index, and if the determination result is abnormal, obtain the relative discharge index sequence of the abnormal electrical components in N consecutive inspections.

[0008] Step S3: Perform trend analysis on the relative discharge index sequence to obtain an abnormal trend determination result; based on the abnormal trend determination result, generate maintenance decision instructions for the abnormal electrical components and incorporate them into the inspection management process.

[0009] Furthermore, the method for determining the electrical component pairs in the tower that meet the preset symmetry conditions based on the tower structure ledger data includes: Based on the tower structure ledger data, a set of candidate electrical components located on the same tower is extracted; based on the type information and installation location information of the electrical components, candidate electrical component pairs that satisfy the symmetry relationship in structure are selected.

[0010] Obtain the ultraviolet observation data corresponding to the candidate electrical component pairs within the historical inspection cycle. The ultraviolet observation data includes the ultraviolet intensity values ​​of each electrical component within different historical inspection cycles.

[0011] Based on the ultraviolet intensity value, calculate the discrete statistical value of the intensity difference of each candidate electrical component pair within the historical inspection cycle.

[0012] Based on the discrete statistical values ​​of the strength difference, candidate electrical component pairs are selected where the strength difference variation within a preset discrete range during the historical inspection cycle.

[0013] Furthermore, the method for obtaining the ultraviolet image sequence containing the first component and the second component includes: When performing a preset ultraviolet inspection task, the imaging parameters of the ultraviolet imaging device carried by the UAV are locked. The imaging parameters include gain parameters and exposure time parameters. Within a preset acquisition time window, an initial ultraviolet image sequence is acquired according to a preset acquisition frequency. Each frame of the ultraviolet image in the initial ultraviolet image sequence contains both a first component and a second component.

[0014] From the initial ultraviolet image sequence, extract the first ultraviolet observation data of the first component in each frame of ultraviolet image and the second ultraviolet observation data of the second component in each frame of ultraviolet image; based on the first ultraviolet observation data and the second ultraviolet observation data, generate the discharge activity intensity sequence of the first component and the second component over time.

[0015] Based on the discharge activity intensity sequence of the first component and the second component, the overlapping time period in which the discharge activity intensity of both the first component and the second component exceeds the preset intensity threshold is determined.

[0016] An ultraviolet image sequence is obtained by selecting ultraviolet images that fall within the overlapping time period from the initial ultraviolet image sequence.

[0017] Furthermore, when acquiring the initial ultraviolet image sequence, the pixel scale ratio of the pixel bounding box size of the first component and the second component in the ultraviolet image is calculated.

[0018] Based on the tower structure ledger data, the spatial geometric positions of the first and second components are extracted; based on the spatial geometric positions and the UAV spatial attitude parameters recorded in the spatiotemporal information, the theoretical imaging scale ratio is calculated; when the deviation between the pixel scale ratio and the theoretical imaging scale ratio exceeds a preset deviation threshold, the ultraviolet image is subjected to perspective transformation processing.

[0019] Furthermore, the method for determining the symmetrical analysis region in the ultraviolet image sequence based on the tower structure ledger data and spatiotemporal information includes: Based on the spatial geometric location and spatiotemporal information, initial analysis regions corresponding to the first component and the second component are determined in the ultraviolet images of the ultraviolet image sequence.

[0020] Based on the spatial distribution characteristics of the ultraviolet radiation response of the ultraviolet images in the ultraviolet image sequence, the spatial range of the initial analysis area is relocated and scaled to obtain a symmetrical analysis area corresponding to the first component and the second component.

[0021] Furthermore, the method for calculating the relative discharge index of electrical components in a preset ultraviolet inspection task based on the symmetrical analysis region includes: For each frame of ultraviolet image in the ultraviolet image sequence, the intensity value sequence of the first frame within the symmetrical analysis region corresponding to the first component and the intensity value sequence of the second frame within the symmetrical analysis region corresponding to the second component are calculated; outlier frame removal is performed on the intensity value sequence of the first frame and the intensity value sequence of the second frame respectively to obtain the first effective frame sequence and the second effective frame sequence.

[0022] The first ultraviolet intensity value and the second ultraviolet intensity value are calculated based on the first effective frame sequence and the second effective frame sequence; the first discrete metric and the second discrete metric are calculated based on the first effective frame sequence and the second effective frame sequence respectively; and the stable term is determined based on the functional relationship between the first discrete metric and the second discrete metric.

[0023] The relative discharge index of the electrical components in the preset ultraviolet inspection task is calculated based on the stability term, the first ultraviolet intensity value, and the second ultraviolet intensity value.

[0024] Furthermore, the method for determining the relative discharge index includes: Obtain historical datasets of relative discharge indices for electrical component pairs within historical inspection cycles; from these historical datasets, select subsets that meet preset data quality conditions at the time of acquisition as valid historical datasets.

[0025] Based on the valid historical dataset, the statistical distribution boundary of the relative discharge index is calculated.

[0026] The relative discharge index obtained within the current preset ultraviolet inspection task is compared with the statistical distribution boundary; if the relative discharge index is outside the statistical distribution boundary, the result is determined to be abnormal; otherwise, the result is determined to be normal.

[0027] Based on the same inventive concept, this invention also provides a power distribution network inspection and management system, the system comprising: The benchmark modeling module is used to acquire the tower structure ledger data of the power distribution network line; determine the electrical component pairs in the tower that meet the preset symmetry conditions based on the tower structure ledger data, the electrical component pairs include a first component and a second component; the UAV acquires the ultraviolet image sequence containing the first component and the second component by performing a preset ultraviolet inspection task, and records the spatiotemporal information.

[0028] The differential detection module is used to determine the symmetrical analysis region in the ultraviolet image sequence based on the tower structure ledger data and spatiotemporal information; calculate the relative discharge index of electrical components in the preset ultraviolet inspection task based on the symmetrical analysis region; determine the relative discharge index, and if the determination result is abnormal, obtain the relative discharge index sequence of the abnormal electrical components in N consecutive inspections.

[0029] The trend decision module is used to perform trend analysis on the relative discharge index sequence to obtain an abnormal trend determination result; based on the abnormal trend determination result, it generates maintenance decision instructions for the abnormal electrical components and incorporates them into the inspection management process.

[0030] Compared with existing technologies, this invention identifies electrical component pairs that meet preset symmetry conditions based on tower structure ledger data, determines symmetry analysis regions in ultraviolet image sequences, and calculates the relative discharge index of electrical component pairs based on the symmetry analysis regions for judgment and trend analysis. This makes the discharge assessment results independent of a single absolute quantity index, thereby reducing the impact of environmental factors on the detection results, achieving stable comparison across time periods, and improving the reliability of defect identification and status trend judgment. Attached Figure Description

[0031] Figure 1This is a flowchart of a power distribution network inspection and management method according to Embodiment 1 of the present invention; Figure 2 This is a block diagram of a power distribution network inspection and management system according to Embodiment 2 of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] Before providing examples, it is necessary to describe the application scenarios of this invention. This embodiment is applicable to overhead power distribution line towers with structural symmetry. Typical scenarios include straight-line towers, tension towers, and branch towers. Electrical components of the same model, installation height, and operating voltage level are typically installed on both sides of the same tower, such as symmetrically arranged suspension insulator strings, jumper insulator strings, or symmetrical hardware assemblies. During long-term operation, these symmetrical components are subjected to the same electrical and climatic environmental conditions. Ultraviolet imaging detection is mainly used to identify corona discharge and partial discharge phenomena in overhead line inspections. Ultraviolet radiation intensity is affected by factors such as ambient humidity, air density, background light radiation, imaging distance, shooting posture, and equipment gain parameters. The ultraviolet photon count rate of the same component fluctuates at different inspection times. For direct comparison of absolute ultraviolet intensity values, it is difficult to eliminate the influence of environmental variables on the measurement results under cross-time conditions. In the above-mentioned tower scenarios with structural symmetry, the influence of environmental factors, imaging posture, and equipment parameters on the two components is highly consistent in the ultraviolet images acquired by the symmetrical electrical components at the same time. Based on this physical condition, differential calculations can be performed on symmetrical components within the same frame of ultraviolet image, thereby reducing the interference of external variables on the detection results.

[0034] Example 1: As Figure 1 As shown in the figure, this embodiment provides a distribution network inspection and management method, the method including: Obtain the structural data of the poles and towers of the power distribution network; determine the electrical component pairs in the poles and towers that meet the preset symmetry conditions based on the structural data of the poles and towers, wherein the electrical component pairs include a first component and a second component; the method for determining the electrical component pairs in the poles and towers that meet the preset symmetry conditions based on the structural data of the poles and towers includes: Based on the tower structure ledger data, a set of candidate electrical components located on the same tower is extracted; based on the type information and installation location information of the electrical components, candidate electrical component pairs that satisfy the symmetry relationship in structure are selected.

[0035] Obtain the ultraviolet observation data corresponding to the candidate electrical component pairs within the historical inspection cycle. The ultraviolet observation data includes the ultraviolet intensity values ​​of each electrical component within different historical inspection cycles.

[0036] Based on the ultraviolet intensity value, calculate the discrete statistical value of the intensity difference of each candidate electrical component pair within the historical inspection cycle.

[0037] Based on the discrete statistical values ​​of the intensity difference, candidate electrical component pairs are selected whose intensity difference changes are limited to a preset discrete range within the historical inspection cycle. The UAV performs a preset ultraviolet inspection task to acquire an ultraviolet image sequence containing the first and second components and records spatiotemporal information; the method for acquiring the ultraviolet image sequence containing the first and second components includes: When performing a preset ultraviolet inspection task, the imaging parameters of the ultraviolet imaging device carried by the UAV are locked, including gain parameters and exposure time parameters; within a preset acquisition time window, an initial ultraviolet image sequence is acquired according to a preset acquisition frequency, and each frame of the initial ultraviolet image sequence contains both a first component and a second component; when acquiring the initial ultraviolet image sequence, the pixel scale ratio of the pixel bounding box size of the first component and the second component in the ultraviolet image is calculated.

[0038] Based on the tower structure ledger data, the spatial geometric positions of the first and second components are extracted; based on the spatial geometric positions and the UAV spatial attitude parameters recorded in the spatiotemporal information, the theoretical imaging scale ratio is calculated; when the deviation between the pixel scale ratio and the theoretical imaging scale ratio exceeds a preset deviation threshold, the ultraviolet image is subjected to perspective transformation processing.

[0039] From the initial ultraviolet image sequence, extract the first ultraviolet observation data of the first component in each frame of ultraviolet image and the second ultraviolet observation data of the second component in each frame of ultraviolet image; based on the first ultraviolet observation data and the second ultraviolet observation data, generate the discharge activity intensity sequence of the first component and the second component over time.

[0040] Based on the discharge activity intensity sequence of the first component and the second component, the overlapping time period in which the discharge activity intensity of both the first component and the second component exceeds the preset intensity threshold is determined.

[0041] An ultraviolet image sequence is obtained by selecting ultraviolet images that fall within the overlapping time period from the initial ultraviolet image sequence.

[0042] This embodiment uses a cat-head-shaped straight tower in a 110kV overhead line for illustration. The tower structure ledger data is stored in a database, including information such as the model, phase, installation side, three-dimensional coordinates of the suspension point, and installation angle of each electrical component. A set of candidate electrical components located on the same tower is extracted from the tower structure ledger data. In this embodiment, the set of candidate electrical components includes three-phase (A, B, C) suspension insulator strings arranged on both the left and right sides, totaling six strings of insulators of the same model. Based on the type and installation location information of the electrical components, three pairs of candidate electrical components symmetrically arranged about the center face of the tower are selected: left-side A-phase and right-side A-phase insulators, left-side B-phase and right-side B-phase insulators, and left-side C-phase and right-side C-phase insulators. Ultraviolet (UV) observation data of the three pairs of candidate electrical components are retrieved during the historical inspection period. In this embodiment, the historical inspection period is the past 24 months, with a total of 12 regular UV inspections conducted. Each of the 12 scheduled ultraviolet (UV) inspections records the UV photon count rate within the corresponding area of ​​each insulator string, which is used as the UV intensity value of the electrical component, measured in photons / s. For each candidate electrical component pair, 12 intensity differences are generated across the 12 inspections. The intensity difference is the difference in UV intensity values ​​between the two insulator strings of the candidate electrical component pair within the same inspection. The standard deviation of the intensity difference sequence formed by the 12 intensity differences is used as the discrete statistical value of the intensity difference. The discrete statistical value of the intensity difference between the left and right phase A insulator strings is 45 photons / s; the discrete statistical value of the intensity difference between the left and right phase B insulator strings is 18 photons / s; and the discrete statistical value of the intensity difference between the left and right phase C insulator strings is 120 photons / s. The preset discrete range is limited to an intensity difference discrete statistical value not exceeding 50 photons / s. The left-side A-phase insulator string and the right-side A-phase insulator string, as well as the left-side B-phase insulator string and the right-side B-phase insulator string, were selected as electrical component pairs that meet the preset symmetry conditions. In this embodiment, the left-side A-phase insulator string is selected as the first component, and the right-side A-phase insulator string is selected as the second component.

[0043] A drone equipped with a UVi-260 ultraviolet imaging device performed a pre-defined ultraviolet inspection task. The drone flew to a position where it could simultaneously observe both the first and second components while maintaining stable flight attitude. The gain parameter of the ultraviolet imaging device was set to 45 dB, and the exposure time was set to 8 ms and kept constant. For approximately 30 minutes after sunset, continuous shooting was conducted at a sampling frequency of 10 Hz for 5 seconds, resulting in an initial sequence of 50 ultraviolet images. Each frame of the initial 50-frame ultraviolet image sequence simultaneously contained both the first and second components. During the acquisition process, the drone's latitude, longitude, elevation, attitude angle, and timestamp were simultaneously recorded as spatiotemporal information.

[0044] For each initial frame of the ultraviolet image, an edge detection method is used to identify the contours of the first and second components in the ultraviolet image, and the minimum bounding rectangle of each contour is calculated. The ratio of the widths of the minimum bounding rectangles of the first and second components is used as the pixel scale ratio. For example, in the first frame image, the bounding box width of the first component is 92 pixels, the bounding box width of the second component is 80 pixels, and the pixel scale ratio is 1.15.

[0045] The three-dimensional coordinates of the two insulator attachment points recorded in the tower structure ledger data are used as the spatial geometric positions of the first and second components, respectively. The latitude, longitude, elevation, and attitude angle of the UAV, recorded simultaneously during ultraviolet image acquisition, are used to determine the imaging attitude of the ultraviolet imaging device relative to the two spatial geometric positions. The ultraviolet imaging device has a focal length of 8mm, and the pixel size parameters are obtained from the device calibration data. Based on the three-dimensional coordinates of the two insulator attachment points, the UAV attitude angle, and the focal length parameters, the theoretical imaging scale ratio of the first and second components under ideal imaging conditions is calculated using a pinhole imaging model. In this embodiment, the theoretical imaging scale ratio of the first frame of the ultraviolet image is calculated to be 1.02. The relative deviation between the pixel scale ratio of 1.15 and the theoretical imaging scale ratio of 1.02 is 12.7%, exceeding the preset deviation threshold of 10%. Therefore, perspective transformation processing is performed on the first frame of the ultraviolet image. The perspective transformation matrix is ​​calculated based on the coordinates of the center point of the bounding box of the first and second components in the ultraviolet image and the corresponding three-dimensional coordinates of the attachment points. After transformation, the pixel scale ratio is recalculated, and the relative deviation between the transformed pixel scale ratio and the theoretical imaging scale ratio is less than the preset deviation threshold.

[0046] After perspective correction, the ultraviolet photon count rate of all pixels within the pixel bounding box regions of the first and second components is extracted in each frame of the ultraviolet image. The pixel values ​​are then integrated to obtain the first and second ultraviolet observation data. A discharge activity intensity sequence of length 50 for the first component and a discharge activity intensity sequence of the second component are formed from 50 frames of data. The two discharge activity intensity sequences are compared, and frames whose intensity values ​​both exceed a preset intensity threshold of 5000 photons / s are selected. Statistical results show that frames 15 to 35 meet the condition; this time interval is the overlapping period when the discharge activity of both components is significant. Twenty-one frames corresponding to frames 15 to 35 are selected from the initial ultraviolet image sequence to form an ultraviolet image sequence, and the spatiotemporal information of the corresponding frames is recorded. The preset intensity threshold is determined by taking the average background noise value after multiple measurements on a known component without discharge under locked gain and exposure parameters.

[0047] Based on the tower structure ledger data and spatiotemporal information, a symmetrical analysis region is determined in the ultraviolet image sequence; the method for determining the symmetrical analysis region based on the tower structure ledger data and spatiotemporal information in the ultraviolet image sequence includes: Based on the spatial geometric location and spatiotemporal information, initial analysis regions corresponding to the first component and the second component are determined in the ultraviolet images of the ultraviolet image sequence.

[0048] Based on the spatial distribution characteristics of the ultraviolet radiation response of the ultraviolet images in the ultraviolet image sequence, the spatial range of the initial analysis area is repositioned and scaled to obtain a symmetrical analysis area corresponding to the first and second components. Based on the symmetrical analysis area, the relative discharge index of the electrical components in the preset ultraviolet inspection task is calculated; the method for calculating the relative discharge index of the electrical components in the preset ultraviolet inspection task based on the symmetrical analysis area includes: For each frame of ultraviolet image in the ultraviolet image sequence, the intensity value sequence of the first frame within the symmetrical analysis region corresponding to the first component and the intensity value sequence of the second frame within the symmetrical analysis region corresponding to the second component are calculated; outlier frame removal is performed on the intensity value sequence of the first frame and the intensity value sequence of the second frame respectively to obtain the first effective frame sequence and the second effective frame sequence.

[0049] The first ultraviolet intensity value and the second ultraviolet intensity value are calculated based on the first effective frame sequence and the second effective frame sequence; the first discrete metric and the second discrete metric are calculated based on the first effective frame sequence and the second effective frame sequence respectively; and the stable term is determined based on the functional relationship between the first discrete metric and the second discrete metric.

[0050] The relative discharge index of the electrical component in a preset ultraviolet inspection task is calculated based on the stability term, the first ultraviolet intensity value, and the second ultraviolet intensity value. The relative discharge index is then determined; if the determination result is abnormal, the relative discharge index sequence of the abnormal electrical component in N consecutive inspections is obtained. The method for determining the relative discharge index includes: Obtain historical datasets of relative discharge indices for electrical component pairs within historical inspection cycles; from these historical datasets, select subsets that meet preset data quality conditions at the time of acquisition as valid historical datasets.

[0051] Based on the valid historical dataset, the statistical distribution boundary of the relative discharge index is calculated.

[0052] The relative discharge index obtained within the current preset ultraviolet inspection task is compared with the statistical distribution boundary; if the relative discharge index is outside the statistical distribution boundary, the result is determined to be abnormal; otherwise, the result is determined to be normal.

[0053] For example, for each frame of an ultraviolet image in a sequence, the spatial position and attitude relationship of the ultraviolet imaging device are established based on the spatial geometric positions of the first and second components recorded in the tower structure ledger data, as well as the latitude, longitude, elevation, and attitude angle of the UAV recorded during image acquisition. Combining the focal length and pixel size parameters of the ultraviolet imaging device, the spatial geometric positions of the first and second components are projected onto the image pixel coordinate system using a pinhole imaging model. The approximate projection areas of the first and second components in the ultraviolet image are then calculated. These approximate projection areas are represented by a rectangular enclosure and serve as the initial analysis areas. Taking the first frame of the ultraviolet image as an example, the pixel coordinate range of the initial analysis area for the first component is calculated to be (150, 200) to (250, 400), and the pixel coordinate range of the initial analysis area for the second component is (400, 200) to (500, 400). In each frame of ultraviolet images, the ultraviolet photon count rate of all pixels within the initial analysis region of the first component is statistically analyzed, and the pixel intensity centroid coordinates and second-order central moments are calculated. The same calculation is performed on the initial analysis region of the second component. The intensity centroids of the first component in multiple frames of ultraviolet images are concentrated around (180, 280), and the second-order moment of the pixel intensity along the insulator axis is greater than the second-order moment in the vertical direction. Based on the average coordinates of the intensity centroids, the center of the initial analysis region of the first component is moved to (180, 280). Based on the principal axis direction and length of the second-order moment, the long side of the region is rotated and appropriately extended to make the principal axis direction of the region consistent with the insulator axis direction. The intensity centroid and second-order moment of the second component are calculated using the same method, and the center of its initial analysis region is adjusted and the principal axis direction is corrected. The first and second components are arranged in a mirror-symmetric manner in the tower structure ledger data, and the two regions use the same region size parameters and principal axis determination rules during the correction process. Therefore, the two corrected regions maintain mirror symmetry under the structural correspondence. This yields the symmetrical analysis regions corresponding to the first component and the second component. The principal axes of these two symmetrical analysis regions in the image coordinate system align with the axial directions of the insulators recorded in the tower structure ledger data, and their centers coincide with their respective centroids. After processing, the symmetrical analysis regions of the first and second components are obtained for each frame of the ultraviolet image sequence.

[0054] In each frame of the ultraviolet image, the ultraviolet photon count rate of all pixels within the symmetrical analysis region corresponding to the first component is extracted, and the pixel values ​​within the region are integrated. The integration result is used as the ultraviolet observation data for the first component in that frame. In the same frame of the ultraviolet image, the ultraviolet photon count rate of all pixels within the symmetrical analysis region corresponding to the second component is extracted and integrated. The integration result is used as the ultraviolet observation data for the second component in that frame. Integration calculations are performed on 21 frames of ultraviolet images in the ultraviolet image sequence to obtain a first frame intensity value sequence and a second frame intensity value sequence of length 21. The sample mean μ1 and sample standard deviation σ1 are calculated for the first frame intensity value sequence. Frames with values ​​exceeding the range [μ1-3σ1, μ1+3σ1] are marked as outliers and removed. The remaining frames constitute the first valid frame sequence. The second valid frame sequence is calculated similarly. After removal, the first valid frame sequence contains 20 frames of data, and the second valid frame sequence contains 19 frames of data. Align the first and second valid frame sequences based on their timestamps, retaining only data frames whose timestamps appear in both sequences, resulting in an aligned set of valid frames. Calculate the arithmetic mean of the aligned first valid frame set to obtain the first ultraviolet intensity value I1 = 18500 photons / s. Calculate the arithmetic mean of the aligned second valid frame set to obtain the second ultraviolet intensity value I2 = 12000 photons / s. Calculate the sample standard deviation of the aligned first and second valid frame sets to obtain the first discrete metric σ1 = 850 and the second discrete metric σ2 = 600. The stability term ε is calculated using the function ε = 50 + 0.1 × (σ1 + σ2). Substituting σ1 = 850 and σ2 = 600, we obtain ε = 195. Substituting into the formula RDI = (I1 - I2) / (I1 + I2 + ε), we calculate the relative discharge index RDI ≈ 0.212. The relative discharge index ranges from -1 to 1, and the relative discharge index calculated in this embodiment is 0.212.

[0055] The relative discharge index of the electrical component, calculated from each inspection over the past 24 months, is obtained to form a historical dataset of relative discharge index. In this embodiment, 12 historical inspection records are stored. Data quality is screened from these 12 historical inspection records. The data quality conditions are: relative humidity below 80% during UV image acquisition, UV imaging equipment gain parameters between 40 dB and 50 dB, and exposure time maintained at 8 ms. Each historical inspection record is checked against the acquisition log, and records that simultaneously meet the above conditions are selected to obtain a valid historical dataset, consisting of 8 historical relative discharge index data points. The sample mean μ is calculated from these 8 valid historical data points. h and sample standard deviation σ h The calculated μ h =0.05, σ h =0.08. The statistical distribution boundary is based on the interval [μh -3σ h ,μ h +3σ h The calculation yields a boundary of [-0.19, 0.29]. The relative discharge index of 0.212 calculated in this inspection is compared with the statistical distribution boundary. 0.212 is within the interval [-0.19, 0.29], and the result is considered normal. Assume the relative discharge index calculated in this inspection is 0.35. When the relative discharge index is greater than the upper limit of the statistical distribution boundary, the first component is determined to be an abnormal component; when the relative discharge index is less than the lower limit of the statistical distribution boundary, the second component is determined to be an abnormal component. Comparing 0.35 with the statistical distribution boundary [-0.19, 0.29], 0.35 is greater than 0.29, therefore the first component is determined to be an abnormal component. The relative discharge indices calculated for the first component in the most recent five consecutive inspections are read from the database, forming a relative discharge index sequence. The five inspections are arranged in chronological order, with an interval of approximately one month between adjacent inspections, and the sequence values ​​are [0.05, 0.08, 0.15, 0.25, 0.35].

[0056] Trend analysis is performed on the relative discharge index sequence to obtain an abnormal trend determination result; based on the abnormal trend determination result, maintenance decision instructions for the abnormal electrical components are generated and incorporated into the inspection management process.

[0057] In this embodiment, a linear fit is performed on the relative discharge index sequence. Using inspection sequence numbers 1 to 5 as independent variables and the corresponding relative discharge index as the dependent variable, the least squares method is used to calculate the fitted straight line. The calculated slope of the fitted straight line is 0.073 per month. The trend judgment threshold is set to 0.02 per month. The trend judgment threshold is determined based on the statistical results of the relative discharge index sequence of historical normal components. In historical normal samples, the absolute value of the slope is distributed within 95% of the range not exceeding 0.02 per month. If the fitted slope of 0.073 per month is greater than 0.02 per month, the abnormal trend judgment result is deterioration. Based on the abnormal trend judgment result, a maintenance decision instruction is generated. The maintenance decision instruction includes the tower number, phase A, component identification (left insulator string), anomaly type (abnormal relative discharge index with an upward trend), suggested handling method (to climb the tower to inspect the insulator sheath and hardware connections), and handling priority (high). The maintenance decision instruction is written into the work order module of the production management system in structured data form, generating a corresponding defect elimination work order record, and is associated with the inspection management file of that tower.

[0058] Example 2: Based on the same inventive concept, such as Figure 2 As shown in the figure, this embodiment also provides a power distribution network inspection and management system, the system including: The benchmark modeling module is used to acquire the tower structure ledger data of the power distribution network line; determine the electrical component pairs in the tower that meet the preset symmetry conditions based on the tower structure ledger data, the electrical component pairs include a first component and a second component; the UAV acquires the ultraviolet image sequence containing the first component and the second component by performing a preset ultraviolet inspection task, and records the spatiotemporal information.

[0059] The differential detection module is used to determine the symmetrical analysis region in the ultraviolet image sequence based on the tower structure ledger data and spatiotemporal information; calculate the relative discharge index of electrical components in the preset ultraviolet inspection task based on the symmetrical analysis region; determine the relative discharge index, and if the determination result is abnormal, obtain the relative discharge index sequence of the abnormal electrical components in N consecutive inspections.

[0060] The trend decision module is used to perform trend analysis on the relative discharge index sequence to obtain an abnormal trend determination result; based on the abnormal trend determination result, it generates maintenance decision instructions for the abnormal electrical components and incorporates them into the inspection management process.

[0061] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0062] Finally, it should be noted that although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for managing power distribution network inspections, characterized in that, The method includes: Obtain the tower structure ledger data of the power distribution network line; determine the electrical component pairs in the tower that meet the preset symmetry conditions based on the tower structure ledger data, the electrical component pairs include a first component and a second component; the UAV performs a preset ultraviolet inspection task to obtain an ultraviolet image sequence containing the first component and the second component, and records the spatiotemporal information; Based on the tower structure ledger data and spatiotemporal information, a symmetrical analysis region is determined in the ultraviolet image sequence; based on the symmetrical analysis region, the relative discharge index of electrical components in a preset ultraviolet inspection task is calculated; the relative discharge index is judged, and if the judgment result is abnormal, the relative discharge index sequence of the abnormal electrical components in N consecutive inspections is obtained. Trend analysis is performed on the relative discharge index sequence to obtain an abnormal trend determination result; based on the abnormal trend determination result, maintenance decision instructions for the abnormal electrical components are generated and incorporated into the inspection management process.

2. The distribution network inspection and management method according to claim 1, characterized in that, The method for determining pairs of electrical components in a tower that meet preset symmetry conditions based on the tower structure ledger data includes: Based on the tower structure ledger data, a set of candidate electrical components located on the same tower is extracted; based on the type information and installation location information of the electrical components, candidate electrical component pairs that satisfy the symmetry relationship in structure are selected. Obtain the ultraviolet observation data corresponding to the candidate electrical component pairs within the historical inspection cycle, wherein the ultraviolet observation data includes the ultraviolet intensity values ​​of each electrical component within different historical inspection cycles; Based on the ultraviolet intensity value, calculate the discrete statistical value of the intensity difference of each candidate electrical component pair within the historical inspection cycle; Based on the discrete statistical values ​​of the strength difference, candidate electrical component pairs are selected where the strength difference variation within a preset discrete range during the historical inspection cycle.

3. The distribution network inspection and management method according to claim 2, characterized in that, The method for obtaining an ultraviolet image sequence containing a first component and a second component includes: When performing a preset ultraviolet inspection task, the imaging parameters of the ultraviolet imaging device carried by the UAV are locked. The imaging parameters include gain parameters and exposure time parameters. Within a preset acquisition time window, an initial ultraviolet image sequence is acquired according to a preset acquisition frequency. Each frame of the ultraviolet image in the initial ultraviolet image sequence contains both a first component and a second component. From the initial ultraviolet image sequence, extract the first ultraviolet observation data of the first component in each frame of ultraviolet image and the second ultraviolet observation data of the second component in each frame of ultraviolet image; based on the first ultraviolet observation data and the second ultraviolet observation data, generate the discharge activity intensity sequences of the first component and the second component over time, respectively. Based on the discharge activity intensity sequence of the first component and the second component, the overlapping time period in which the discharge activity intensity of both the first component and the second component exceeds the preset intensity threshold is determined; An ultraviolet image sequence is obtained by selecting ultraviolet images that fall within the overlapping time period from the initial ultraviolet image sequence.

4. The distribution network inspection and management method according to claim 3, characterized in that, When acquiring the initial ultraviolet image sequence, calculate the pixel scale ratio of the pixel bounding box size of the first component and the second component in the ultraviolet image; Based on the tower structure ledger data, extract the spatial geometric positions of the first and second components; Based on the spatial geometric location and the UAV spatial attitude parameters recorded in the spatiotemporal information, the theoretical imaging scale ratio is calculated; when the deviation between the pixel scale ratio and the theoretical imaging scale ratio exceeds a preset deviation threshold, the ultraviolet image is subjected to perspective transformation processing.

5. The distribution network inspection and management method according to claim 4, characterized in that, The method for determining the symmetrical analysis region in the ultraviolet image sequence based on the tower structure ledger data and spatiotemporal information includes: Based on the spatial geometric location and spatiotemporal information, the initial analysis regions corresponding to the first component and the second component are determined respectively in the ultraviolet images of the ultraviolet image sequence. Based on the spatial distribution characteristics of the ultraviolet radiation response of the ultraviolet images in the ultraviolet image sequence, the spatial range of the initial analysis region is relocated and scaled to obtain a symmetrical analysis region corresponding to the first and second components.

6. The distribution network inspection and management method according to claim 5, characterized in that, The method for calculating the relative discharge index of electrical components in a preset ultraviolet inspection task based on the symmetrical analysis region includes: For each frame of ultraviolet image in the ultraviolet image sequence, the intensity value sequence of the first frame within the symmetrical analysis region corresponding to the first component and the intensity value sequence of the second frame within the symmetrical analysis region corresponding to the second component are calculated; outlier frame removal is performed on the first frame intensity value sequence and the second frame intensity value sequence respectively to obtain the first effective frame sequence and the second effective frame sequence; The first ultraviolet intensity value and the second ultraviolet intensity value are calculated based on the first effective frame sequence and the second effective frame sequence; the first discrete metric and the second discrete metric are calculated based on the first effective frame sequence and the second effective frame sequence respectively; the stable term is determined based on the functional relationship between the first discrete metric and the second discrete metric. The relative discharge index of the electrical components in the preset ultraviolet inspection task is calculated based on the stability term, the first ultraviolet intensity value, and the second ultraviolet intensity value.

7. The distribution network inspection and management method according to claim 6, characterized in that, The method for determining the relative discharge index includes: Obtain historical datasets of relative discharge indices for electrical component pairs within historical inspection cycles; from these historical datasets, select subsets that meet preset data quality conditions at the time of acquisition as valid historical datasets; Based on the valid historical dataset, calculate the statistical distribution boundary of the relative discharge index; The relative discharge index obtained within the current preset ultraviolet inspection task is compared with the statistical distribution boundary; if the relative discharge index is outside the statistical distribution boundary, the result is determined to be abnormal; otherwise, the result is determined to be normal.

8. A power distribution network inspection and management system, characterized in that, The system includes: The baseline modeling module is used to acquire the tower structure ledger data of the power distribution network; determine the electrical component pairs in the tower that meet the preset symmetry conditions based on the tower structure ledger data, the electrical component pairs including a first component and a second component; the UAV acquires the ultraviolet image sequence containing the first component and the second component by performing a preset ultraviolet inspection task, and records the spatiotemporal information; The differential detection module is used to determine the symmetrical analysis region in the ultraviolet image sequence based on the tower structure ledger data and spatiotemporal information; calculate the relative discharge index of electrical components in the preset ultraviolet inspection task based on the symmetrical analysis region; determine the relative discharge index; if the determination result is abnormal, obtain the relative discharge index sequence of the abnormal electrical components in N consecutive inspections; The trend decision module is used to perform trend analysis on the relative discharge index sequence to obtain an abnormal trend determination result; based on the abnormal trend determination result, it generates maintenance decision instructions for the abnormal electrical components and incorporates them into the inspection management process.

Images