Composite insulator contamination hydrophobicity imaging diagnosis system and method

Abstract

This invention discloses a system and method for imaging and diagnosing the hydrophobicity of composite insulators, relating to the field of power equipment testing technology. The method includes: receiving operating environment data, geometric structure data, infrared time-series matrix, and multispectral feature matrix of a target insulator; generating a temperature response baseline for the target insulator; extracting the target region on the insulator surface based on the operating environment data and geometric structure data, dividing it into multiple pixel sub-regions; extracting temperature change data for each pixel sub-region; combining the temperature response baseline to screen a first candidate region set that may contain microcracks; and, based on the multispectral feature matrix, selecting a second candidate region set from the first candidate region set; performing spatial connectivity analysis and temporal stability analysis on the second candidate region set to determine the hydrophobic failure region of the target insulator. Its beneficial effect is that it can improve the accuracy of detecting small hydrophobic failure regions on the surface of composite insulators in high-altitude areas.

Description

Technical Field

[0001] This invention relates to the field of power equipment testing technology, and in particular to an imaging diagnostic system and method for the hydrophobicity of composite insulators. Background Technology

[0002] Composite insulators, as an important component of transmission lines, are mainly used to support conductors and isolate parts with different potentials. The surface of the insulator is usually covered with a layer of organic material with hydrophobic properties. Hydrophobicity refers to the ability of the insulator surface to repel water. Good hydrophobicity can effectively inhibit the formation of a continuous conductive water film by dirt in a humid environment, reduce the conductivity of the insulator surface, and thus improve the operational safety and reliability of the transmission line. Therefore, accurately diagnosing the hydrophobicity status of the insulator surface is of great significance for predicting the insulator's service life, formulating scientific maintenance plans, and preventing insulation accidents.

[0003] Currently, the diagnosis of hydrophobicity in composite insulators mainly relies on contact angle measurement, spraying, optical imaging, multispectral detection, and infrared thermal imaging analysis. These methods can effectively detect large-scale degradation of hydrophobicity on the insulator surface under stable environmental conditions. However, in high-altitude areas, due to large diurnal temperature differences and frequent temperature fluctuations, the surface material of the insulator is susceptible to thermal expansion and contraction, resulting in microcracks and localized hydrophobic degradation. This degradation area is small in scale and scattered, making it difficult for existing diagnostic methods to detect sensitively. Furthermore, the low temperature and low air pressure environment at high altitudes often leads to condensation or frost covering the insulator surface. These wet conditions not only alter the optical and thermal properties of the insulator surface but also mask the imaging signals of local hydrophobicity differences, further reducing detection accuracy. If a missed detection occurs, the local failure area of ​​the hydrophobic layer will form a continuous conductive water film in a humid environment, causing a surge in leakage current and electric field distortion, which may ultimately lead to a flashover trip accident, seriously threatening the safe operation of transmission lines.

[0004] Therefore, a composite insulator pollution hydrophobicity imaging diagnostic system and method are proposed. Summary of the Invention

[0005] In view of the above-mentioned prior art, this application is hereby filed. Embodiments of this application provide a composite insulator pollution hydrophobicity imaging diagnostic system and method, which can distinguish between microcrack thermal anomalies caused by hydrophobicity degradation and thermal anomalies caused by wet pollution / frost, improving the detection accuracy of small hydrophobic failure areas on the surface of composite insulators in high-altitude areas.

[0006] According to one aspect of this application, an imaging diagnostic method for the hydrophobicity of composite insulators is provided, comprising: receiving operating environment data, geometric structure data, and a registered and normalized infrared time-series matrix and multispectral feature matrix of a target insulator; generating a temperature response baseline of the target insulator based on the operating environment data and a pre-constructed physical thermal network model; extracting a target region on the surface of the target insulator that is more prone to hydrophobic degradation under thermal expansion and contraction and wet pollution conditions based on the operating environment data and the geometric structure data; dividing the target region into multiple pixel sub-regions; and extracting each pixel sub-region based on the infrared time-series matrix. Temperature change data of the element region; based on the temperature change data and the temperature response baseline, pixel sub-regions whose temperature change characteristics deviate from a preset threshold are selected to obtain a first candidate region set that may have microcracks; pixel sub-regions in the first candidate region set that have thermal anomalies due to wet contamination or frost are removed based on the spectral feature matrix to obtain a second candidate region set; spatial connectivity analysis and temporal stability analysis are performed on the second candidate region set, and connected regions whose temperature change characteristics deviate from the temperature response baseline by more than a preset deviation threshold in multiple time windows are identified as hydrophobic failure regions of the target insulator.

[0007] According to another aspect of this application, a composite insulator pollution hydrophobicity imaging diagnostic system is provided, comprising: a receiving module for receiving operating environment data, geometric structure data, and a registered and normalized infrared time-series matrix and multispectral feature matrix of a target insulator; a baseline generation module for generating a temperature response baseline of the target insulator based on the operating environment data and a pre-built physical thermal network model; a region segmentation module for extracting target regions on the surface of the target insulator that are more prone to hydrophobic degradation under thermal expansion and contraction and wet pollution conditions based on the operating environment data and the geometric structure data; a sub-region segmentation module for dividing the target region into multiple pixel sub-regions; and a temperature change extraction module for extracting data based on the infrared... A time-series matrix is ​​used to extract temperature change data for each pixel sub-region; a first filtering module is used to filter pixel sub-regions whose temperature change characteristics deviate from a preset threshold based on the temperature change data and the temperature response baseline, to obtain a first candidate region set that may contain microcracks; a second filtering module is used to remove pixel sub-regions in the first candidate region set that have thermal anomalies due to wet contamination or frost based on the spectral feature matrix, to obtain a second candidate region set; a decision module is used to perform spatial connectivity analysis and temporal stability analysis on the second candidate region set, and to identify connected regions in multiple time-series windows whose temperature change characteristics deviate from the temperature response baseline by more than a preset deviation threshold as hydrophobic failure regions of the target insulator.

[0008] According to another aspect of this application, an electronic device is provided, including a memory and a processor, the memory being used to store computer-executable instructions, and the processor being used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the method described above.

[0009] According to another aspect of this application, a computer storage medium is provided that stores computer-executable instructions thereon, which, when executed by a processor, implement the steps of the method described above.

[0010] Compared with the prior art, the composite insulator pollution hydrophobicity imaging diagnostic system and method according to the embodiments of this application can generate a temperature response baseline by fusing infrared time-series matrix and multispectral feature matrix, combined with physical thermal network model, and accurately locate hydrophobic failure area based on spatial connectivity and time stability analysis. It can effectively distinguish between microcracks and thermal anomalies caused by wet pollution / frost, solve the problem of low detection accuracy of the prior art in high-altitude environments, and has the advantage of improving the detection accuracy of small hydrophobic failure areas. Attached Figure Description

[0011] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0012] Figure 1 This is a flowchart of the composite insulator pollution hydrophobicity imaging diagnostic method of the present invention.

[0013] Figure 2 This is a block diagram of the composite insulator pollution hydrophobicity imaging diagnostic system of the present invention.

[0014] Figure 3 This is a block diagram of an electronic device according to the present invention. Detailed Implementation

[0015] Hereinafter, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0016] Application Overview

[0017] In traditional methods for diagnosing the hydrophobicity of composite insulators, localized hydrophobic degradation caused by microcracks on the insulator surface is difficult to detect effectively in high-altitude environments. At the same time, optical and thermal signal interference caused by wet pollution or frost coverage significantly reduces diagnostic accuracy. Existing infrared thermal imaging and multispectral imaging detection methods, which rely on drastic temperature fluctuations and surface icing and frost, cannot accurately distinguish between thermal anomalies caused by material degradation and temperature changes caused by environmental factors, which can easily lead to misjudgments of the hydrophobicity degradation of insulators.

[0018] If the above problems are not solved, the hydrophobic degradation of the microcrack region cannot be identified in time. Under continuous thermal stress, the cracks will expand to form through defects, accelerating the aging of organic materials. In humid environments, the defect region will form a continuous conductive channel, causing the leakage current density to locally increase to 8-10 times that of the normal region, resulting in electric field distribution distortion. This phenomenon will significantly reduce the surface flashover voltage of the insulator, and induce irreversible insulation breakdown accidents under operating overvoltage or lightning strike conditions.

[0019] Faced with the aforementioned problems, this application first considers how to accurately distinguish between material degradation and external environmental interference in complex high-altitude environments. Traditional single-sensor data is easily affected by sudden temperature changes and surface icing, resulting in a high false positive rate. To address this, this application attempts to integrate multi-dimensional data for cross-validation: a temperature response baseline is established by constructing a physical thermal network model to eliminate baseline drift caused by environmental temperature fluctuations. At the same time, a multispectral feature matrix is ​​introduced for secondary screening. False anomaly signals are eliminated by utilizing the difference in reflectivity between the material degradation area and the frost layer in specific bands. Furthermore, it is found that hydrophobic degradation caused by microcracks has spatial continuity and temporal persistence, while anomalies caused by environmental factors are mostly isolated points or instantaneous phenomena. Therefore, a strategy combining spatial connectivity analysis and multi-temporal window stability verification is proposed to effectively improve detection reliability.

[0020] Exemplary methods

[0021] Figure 1The illustration shows a method for imaging and diagnosing the hydrophobicity of composite insulators according to an embodiment of this application, comprising: receiving operating environment data, geometric structure data, and a registered and normalized infrared time series matrix and multispectral feature matrix of the target insulator; generating a temperature response baseline of the target insulator based on the operating environment data and a pre-constructed physical thermal network model; extracting target regions on the surface of the target insulator that are more prone to hydrophobic degradation under thermal expansion and contraction and wet pollution conditions based on the operating environment data and geometric structure data; dividing the target region into multiple pixel sub-regions; extracting temperature change data of each pixel sub-region based on the infrared time series matrix; filtering pixel sub-regions whose temperature change characteristics deviate from a preset threshold based on the temperature change data and the temperature response baseline to obtain a first candidate region set that may contain microcracks; removing pixel sub-regions in the first candidate region set that have thermal anomalies due to wet pollution or frost based on the spectral feature matrix to obtain a second candidate region set; performing spatial connectivity analysis and temporal stability analysis on the second candidate region set, and identifying connected regions in multiple time series windows whose temperature change characteristics deviate from the temperature response baseline by more than a preset deviation threshold as hydrophobic failure regions of the target insulator.

[0022] Among them, the operating environment data refers to the temperature, humidity, wind speed, and solar radiation intensity parameters of the environment in which the target insulator is located. These parameters can be obtained through meteorological sensors or historical meteorological databases and are used as input boundary conditions when establishing a physical thermal network model.

[0023] Among them, geometric structure data refers to the insulator's skirt diameter, shed spacing, and shed disc tilt angle parameters, which can be obtained through three-dimensional laser scanning or engineering drawing analysis, and are used to calculate the surface thermal stress distribution and wet contamination accumulation areas.

[0024] The infrared time series matrix refers to a collection of infrared thermal image data from multiple time periods that has undergone time registration and spatial normalization. Specifically, it can be periodically acquired and registered using an infrared thermal imager to extract temperature change trend features.

[0025] Among them, the multispectral feature matrix refers to the set of visible light and near-infrared reflectance data after spatial registration. Specifically, it can be acquired synchronously using a multispectral imager to distinguish between material degradation and wet pollution interference.

[0026] Among them, the physical thermal network model refers to the simulation model of the insulator temperature field based on the heat conduction equation. Specifically, it can be implemented using the finite element method or the lumped parameter method to generate the temperature response baseline under defect-free conditions.

[0027] Among them, the temperature response baseline refers to the theoretical temperature change curve of the insulator when there is no hydrophobic degradation, which is used to identify abnormal areas where the actual temperature deviates from the norm.

[0028] Among them, the pixel sub-region refers to the smallest analysis unit that divides the target area according to the resolution of the infrared thermal imager. Specifically, it can be implemented by a grid division algorithm and is used to locate local thermal anomalies caused by microcracks.

[0029] Temperature change data refers to the temperature value and rate of change of each pixel sub-region at different time points. Specifically, it can be calculated through temporal matrix interpolation and is used to detect abnormal fluctuations that deviate from the baseline.

[0030] Spatial connectivity analysis refers to detecting the topological connectivity between adjacent pixel sub-regions in a candidate region. Specifically, it can be achieved using region growing algorithms or morphological processing to eliminate pseudo-defects formed by isolated noise points.

[0031] Among them, time stability analysis refers to the statistical analysis of the anomaly persistence of candidate regions within a continuous monitoring period. Specifically, the sliding window method can be used to calculate the anomaly frequency, which is used to distinguish between transient interference and persistent defects.

[0032] The core innovation of this application lies in fusing infrared time-series temperature fields and multispectral features, combining them with a physical model to generate a dynamic baseline, and employing a multi-stage screening mechanism to effectively distinguish between hydrophobic degradation caused by microcracks and thermal anomalies caused by wet contamination and frost. First, vulnerable areas are accurately located based on environmental and structural data. Microscopic defects are captured through pixel-level temperature change analysis. Then, spectral reflectance characteristics are used to eliminate interference from wet contamination. Finally, spatiotemporal joint analysis confirms persistent failure areas. This technical solution solves the technical challenges of significant interference from thermal expansion and contraction and the masking of defects by wet contamination signals in microcrack detection at high altitudes.

[0033] As a preferred embodiment, the solution of this application is specifically implemented as follows:

[0034] First, environmental data of the target insulator, including ambient temperature, humidity, and wind speed, is collected through a sensor network. Then, geometric data of the insulator is obtained through a 3D scanner. Finally, infrared thermal imagers and multispectral cameras are used to collect the infrared time-series matrix and multispectral feature matrix of the insulator surface, and image registration and normalization are performed.

[0035] Secondly, using a pre-established physical thermal network model and current operating environment data, the temperature response baseline of the target insulator under normal conditions is calculated. Based on the operating environment data and geometric structure data, areas of concentrated thermal stress and areas prone to wet contamination on the insulator surface are identified, and the intersection of these areas is determined as the key target area of ​​focus.

[0036] Furthermore, the target area is divided into several pixel-level sub-regions. For each sub-region, its temperature change data is extracted from the infrared time-series matrix and compared with the temperature response baseline. If the temperature change characteristics of a sub-region deviate from the baseline by more than a preset threshold, it is included in the first candidate region set.

[0037] Subsequently, the first candidate region set was further screened using the multispectral feature matrix. By analyzing the spectral reflectance characteristics of each sub-region in a specific band, thermal anomaly regions caused by wetness or frost were identified and eliminated, resulting in a more accurate second candidate region set.

[0038] Finally, spatial connectivity analysis is performed on the sub-regions in the second candidate region set, and adjacent abnormal sub-regions are clustered into connected components. Stability analysis is performed on each connected component within multiple time windows. If its temperature change characteristics continuously deviate from the baseline and exceed a preset deviation threshold, the connected component is identified as a hydrophobic failure region of the insulator.

[0039] This application effectively solves the problem of difficult detection of localized hydrophobic degradation caused by microcracks on the surface of composite insulators in complex high-altitude environments. By fusing multi-dimensional data and employing a multi-stage screening strategy, the accuracy and reliability of detection are significantly improved. This method can effectively distinguish between thermal anomalies caused by material degradation and temperature changes caused by environmental factors, greatly reducing the false positive rate. Simultaneously, spatial connectivity analysis and temporal stability verification further enhance the ability to identify micro-scale hydrophobic degradation regions. This improvement enables accurate diagnosis of the hydrophobic state of insulators even under complex conditions such as drastic temperature fluctuations and surface icing, providing a reliable basis for timely detection of potential faults and the development of scientific maintenance plans, thereby effectively improving the operational safety and reliability of transmission lines.

[0040] In some of the above-mentioned solutions in this application, the extraction of the target area specifically includes: calculating the thermal stress amplitude of each region of the target insulator based on operating environment data and geometric structure data, and classifying the regions with thermal stress amplitudes higher than a preset threshold as thermal anomaly sensitive areas; identifying the water film and frost-prone areas of the target insulator based on operating environment data and geometric structure data, and classifying the water film and frost-prone areas as wet pollution / frost-prone high-incidence areas; and marking the intersection of the thermal anomaly sensitive areas and the wet pollution / frost-prone high-incidence areas as the target area.

[0041] Among them, the thermal stress amplitude is calculated by the coefficient of thermal expansion and the temperature change gradient, and the preset threshold is set to 30%-50% of the material yield strength; the high-incidence areas of wetness / frost are identified by the surface curvature and contact angle distribution model, and areas with a contact angle of less than 60 degrees are identified as areas prone to accumulation; the intersection area is generated by the spatial superposition algorithm, and only the areas that meet both conditions are retained.

[0042] Specifically, the division of thermally sensitive areas is based on material fatigue characteristics. When the thermal stress amplitude exceeds a threshold, it indicates that the area is prone to microcracks under alternating hot and cold conditions. Areas prone to dampness / frost are identified through geometric indentation features, such as the edge of an umbrella skirt and the threaded connection; these areas are prone to moisture retention due to their structural characteristics. By taking the intersection of these two types of areas, misjudgments caused by a single factor can be eliminated, such as areas subjected only to thermal stress without dampness, or areas subjected only to dampness without thermal stress. For example, when the thermal stress amplitude in a certain area reaches 0.8 MPa and the contact angle is 55 degrees, this area is simultaneously marked as both a thermally sensitive area and a dampness-prone area, and is ultimately included in the target area. This method limits the target area to regions where thermomechanical stress and dampness act together, significantly improving the localization accuracy of subsequent microcrack detection.

[0043] Through the above technical solution, this application can accurately identify the key areas on the insulator surface most prone to hydrophobic degradation. By combining thermal stress analysis and wet contamination / frost prediction, the detection range can be narrowed down in a targeted manner, improving the efficiency and accuracy of subsequent hydrophobicity diagnosis. Furthermore, this method considers the structural characteristics of the insulator and operating environment factors, and can adapt to the detection needs of different types of insulators in various complex environments, demonstrating strong versatility and practicality.

[0044] In some of the above-mentioned schemes of this application, filtering the first candidate region set based on the spectral feature matrix specifically includes: extracting reflectance data of each pixel sub-region in the first candidate region set in N preset bands based on the spectral feature matrix to form a spectral vector, where N is a positive integer greater than one; calculating the spectral deviation between the spectral vector of each pixel sub-region and the reference spectral vector of the preset hydrophobicity degradation characterization; and removing pixel sub-regions with spectral deviation less than a preset deviation threshold from the first candidate set.

[0045] The preset wavelength selection needs to cover the region of difference in characteristic absorption peaks between hydrophobic degraded materials and wet / frosted coverings, for example, a combination of the 450-500nm visible light band, the 1450-1550nm near-infrared band, and the 2100-2200nm short-wave infrared band. The spectral vector is constructed using normalized reflectance data to eliminate ambient light interference, and the reference spectral vector is established using reflectance data from laboratory simulations of hydrophobic degraded samples. Deviation calculation can use Euclidean distance or cosine similarity algorithms, and the deviation threshold is determined by statistically analyzing the spectral differences between wet and microcracked samples.

[0046] Specifically, when a pixel sub-region is covered with ice crystals due to surface frost, its reflectivity in the 1450-1550nm wavelength range is significantly higher than that of the hydrophobic degradation region, while it exhibits a reflectivity trough in the 2100-2200nm wavelength range due to the absorption characteristics of ice crystals. By calculating the deviation between the spectral vector of the candidate region and the reference vector, the reflectivity changes caused by material aging and the reflectivity anomalies caused by surface coverings can be effectively distinguished. For example, when the deviation value is lower than a set threshold, it indicates that the reflectivity characteristics of the region are closer to wet stains than material degradation, and thus it is excluded from the candidate set. This process, through multi-dimensional spectral feature comparison, accurately eliminates misjudgments caused by changes in surface state and improves the specificity of microcrack detection.

[0047] Through the above technical solution, this application can effectively eliminate pixel sub-regions in the first candidate region set that suffer from thermal anomalies due to wet contamination or frost, thus improving the accuracy of hydrophobic failure region identification. Furthermore, by comprehensively analyzing multi-band spectral data, the ability to capture hydrophobic degradation characteristics is enhanced, reducing the interference of environmental factors on the diagnostic results. Therefore, the solution of this application can more accurately locate hydrophobic failure regions on the surface of composite insulators, providing a reliable basis for subsequent maintenance and replacement decisions.

[0048] In some of the above-mentioned schemes of this application, the spatial connectivity analysis and temporal stability analysis of the second candidate region set specifically include: dividing the second candidate region set into multiple connected domains, each connected domain consisting of spatially adjacent pixel sub-regions in the second candidate region set; for each connected domain, calculating the sum of the deviations of the temperature change data of all pixel sub-regions within it relative to the temperature response baseline, as the cumulative deviation of the connected domain; calculating the average deviation of each connected domain based on the number of pixel sub-regions contained in each connected domain and the corresponding cumulative deviation; within M consecutive time windows, statistically analyzing the proportion of windows where the average deviation of each connected domain reaches a preset deviation threshold, where M is a positive integer greater than one; and marking the connected domains where the window proportion exceeds the preset proportion as hydrophobic failure regions of the target insulator.

[0049] The partitioning of connected components is achieved through image morphological closing operations, ensuring that spatially adjacent pixel sub-regions with similar thermal anomaly patterns are aggregated into the same connected component. The cumulative deviation is calculated using a weighted summation method, with the weight coefficients dynamically adjusted based on the position of the pixel sub-region on the insulator surface. The calculation of the average deviation incorporates an area normalization factor to eliminate comparison biases between connected components of different sizes. The number of time windows, M, is set to 3-5 cycles, with a preset ratio set to the 70%-80% range to balance detection sensitivity and anti-interference capability.

[0050] Specifically, spatial connectivity analysis clusters discrete pixel sub-regions into continuous regions, effectively eliminating interference from isolated noise points. The calculation of cumulative deviation integrates the overall thermal anomaly intensity of the region, avoiding random errors from single-pixel detection. The average deviation index combines region area and anomaly intensity to establish a comparability evaluation benchmark. Temporal stability analysis requires candidate regions to consistently exhibit thermal anomaly characteristics across multiple detection cycles; for example, if a region reaches the deviation threshold in four out of five consecutive detection cycles, it can be identified as a persistent failure region. The preset deviation threshold is determined through laboratory calibration, and its specific value is dynamically adjusted based on the thermal expansion coefficient of the insulator material and the range of ambient temperature fluctuations. This technical solution effectively distinguishes between sporadic thermal anomalies and persistent hydrophobic failures through a dual screening mechanism of spatial clustering and time-series verification, thereby improving detection accuracy.

[0051] In some of the solutions described above in this application, when calculating the cumulative deviation of connected regions, the possibility that small connected regions may be formed by instantaneous noise or isolated interference points is not considered. Although such regions may deviate within a single time window, they lack spatial continuity. If they are directly involved in subsequent analysis, it will lead to an increased misjudgment rate and affect the accuracy of diagnosis.

[0052] This application further proposes that, before calculating the cumulative deviation, the following steps are included: counting the number of pixel sub-regions contained in each connected component; and filtering out connected components with a number of pixel sub-regions less than a preset threshold.

[0053] The preset threshold number is determined based on the minimum spatial scale of microcracks on the insulator surface in actual application scenarios. For example, it can be obtained through experimental calibration or historical data statistics, with a typical range of 5 to 10 pixel sub-regions. The screening operation eliminates pseudo-anomaly regions formed by isolated noise points through spatial continuity verification, retaining candidate connected regions with spatial extensibility. This step, together with the subsequent temporal stability analysis, forms a collaborative verification mechanism. Spatial continuity verification filters out transient interference, and temporal stability verification filters out occasional anomalies. This dual verification reduces the probability of false detection.

[0054] Specifically, in the spatial connectivity analysis stage, the connected regions formed by all spatially adjacent pixel sub-regions in the second candidate region set are first identified. For each connected region, the number of pixel sub-regions it contains is counted and compared with a preset threshold. When the number of pixel sub-regions is lower than the threshold, the connected region is determined to be formed by isolated noise or transient interference and is directly screened out, no longer participating in the subsequent cumulative deviation calculation. For example, if the preset threshold is set to 8 pixel sub-regions, all connected regions containing fewer than 8 pixels are excluded. The retained connected regions enter the subsequent temporal stability analysis process, and the persistence of their deviation behavior is verified through multiple time windows. This screening mechanism effectively eliminates false anomaly signals caused by single-point thermal noise or local transient wet contamination, ensuring that subsequent analysis is only carried out on real hydrophobic failure areas with spatial expansion and temporal persistence, significantly improving the reliability of diagnostic results.

[0055] Through the above technical solution, this application can effectively filter out small-area thermal anomaly regions caused by random noise or local hotspots, improving the reliability of hydrophobic failure region identification. Since small-area connected regions may be caused by sensor noise, local reflection, or other non-hydrophobic degradation factors, excluding them can reduce the false alarm rate. Simultaneously, by setting a reasonable preset threshold, potential hydrophobic failure regions of a certain scale can be retained, ensuring that important degradation regions are not missed. This screening mechanism can balance detection sensitivity and accuracy, improving the reliability of diagnostic results.

[0056] In some of the solutions described above in this application, hydrophobic failure regions are identified by statistically analyzing the average deviation of connected components within multiple time windows and setting a window ratio threshold. However, this method may overlook some potential failure regions that have small deviations within a single window but significant long-term cumulative effects, resulting in insufficient detection sensitivity.

[0057] This application further proposes: for each connected component, the sum of its cumulative deviations within M time windows is calculated; connected components whose sum of cumulative deviations exceeds a preset cumulative threshold are marked as hydrophobic failure regions of the target insulator.

[0058] The calculation of the cumulative deviation sum is based on the sum of the deviations of all pixel sub-regions within the connected domain relative to the temperature response baseline in different time windows. The preset cumulative threshold is determined based on historical data or experimental calibration. The statistical process covers multiple consecutive time windows to ensure continuous monitoring of abnormal temperature changes. The preset cumulative threshold is used to filter regions with high long-term cumulative deviations, avoiding misjudgments caused by accidental thermal interference within a single window.

[0059] Specifically, within M consecutive time windows, the sum of cumulative deviations for each connected region is obtained by summing the deviation values ​​of all pixel sub-regions within each window. This accumulation process can capture the cumulative effect of temperature anomalies over time, and is particularly suitable for scenarios in high-altitude areas where large diurnal temperature differences and frequent hot-cold cycles lead to the gradual propagation of microcracks. When the sum of cumulative deviations exceeds a preset cumulative threshold, it indicates that the connected region has a significant temperature anomaly that persists for a long time, which is highly correlated with the thermal properties of hydrophobic failure areas. By introducing a judgment condition for the sum of cumulative deviations, interference signals caused by instantaneous environmental fluctuations or temporary wet cover are further eliminated, improving the accuracy of identifying long-term stability failure areas. For example, the preset cumulative threshold can be set to 1.5 times the average cumulative deviation of historical failure areas to ensure the reliability of the screening results.

[0060] Through the above technical solution, this application can effectively identify areas with persistent temperature anomalies and eliminate interference from short-term temperature fluctuations. This improves the accuracy and reliability of diagnosing hydrophobic failure areas and avoids misjudgments and missed diagnoses. Furthermore, this method can detect minute but persistent temperature anomalies, which helps to identify hydrophobic degradation problems on the insulator surface early and provides a basis for preventive maintenance.

[0061] In some of the solutions described above in this application, the pixel sub-regions of the second candidate region set may contain abnormal signals caused by environmental noise or instantaneous temperature fluctuations. These low signal-to-noise ratio regions may lead to misjudgments in subsequent spatial connectivity and temporal stability analyses, reducing the accuracy of hydrophobic failure region detection.

[0062] This application further proposes that before performing spatial connectivity analysis and temporal stability analysis on the second candidate region set, the following steps are also included: calculating the temperature change signal-to-noise ratio of each pixel sub-region in the second candidate region set based on the infrared time series matrix and operating environment data; and removing pixel sub-regions whose temperature change signal-to-noise ratio is lower than a preset signal-to-noise ratio threshold from the second candidate region set.

[0063] The temperature change signal-to-noise ratio (SNR) is calculated by jointly analyzing the temperature data sequence at continuous time points in the infrared time series matrix and the temperature fluctuation parameters in the environmental data. For example, the ratio of the standard deviation of temperature change to the average temperature change amplitude is used as the SNR indicator. Wind speed and ambient temperature change rate parameters from the operating environment data are used to correct the temperature fluctuation model and distinguish between actual thermal anomalies and environmental noise. The preset SNR threshold is set based on the minimum SNR statistical value of the effective thermal anomaly area in historical data, for example, within the range of 2.5 to 3.0.

[0064] Specifically, the calculation process for the temperature change signal-to-noise ratio (SNR) is as follows: The temperature change curve of each pixel sub-region within M consecutive time windows is extracted from the infrared time series matrix. Combined with the ambient temperature change rate and wind speed data from the operating environment data, a noise baseline model of the temperature change is constructed. The SNR quantification value is obtained by calculating the deviation between the actual temperature change curve and the noise baseline model. When the SNR is below a threshold, it indicates that the temperature change in that region is mainly caused by random noise or transient environmental interference, rather than a stable thermal anomaly caused by the degradation of the material's hydrophobicity. By filtering out such regions, only pixel sub-regions with significant thermal anomaly characteristics are retained in the second candidate region set, allowing subsequent connected component analysis to focus on the actual failure region. For example, if the standard deviation of the temperature change in a pixel sub-region within 5 consecutive time windows is 0.8℃, while the theoretical noise standard deviation caused by ambient temperature fluctuations is 0.5℃, its SNR is 1.6, which is below the preset threshold of 2.5. Therefore, it is judged as an invalid signal and discarded.

[0065] Through the above technical solutions, this application can effectively improve the accuracy and reliability of imaging diagnosis of hydrophobicity in composite insulators. By introducing temperature change signal-to-noise ratio analysis, false thermal anomaly areas caused by environmental noise can be filtered out, reducing misjudgments. This method is particularly suitable for areas with complex environmental conditions such as high altitudes, and can more accurately identify thermal anomaly areas truly caused by hydrophobic degradation, providing a more reliable data foundation for subsequent spatial connectivity analysis and temporal stability analysis. Therefore, the solution of this application can more accurately locate hydrophobic failure areas on the surface of composite insulators, providing strong support for the timely detection and handling of potential insulation hazards, and contributing to improving the operational safety and reliability of transmission lines.

[0066] Exemplary System

[0067] Figure 2The figure illustrates a composite insulator pollution hydrophobicity imaging diagnostic system according to an embodiment of this application, comprising: a receiving module for receiving operating environment data, geometric structure data, and a registered and normalized infrared time-series matrix and multispectral feature matrix of the target insulator; a baseline generation module for generating a temperature response baseline of the target insulator based on the operating environment data and a pre-built physical thermal network model; a region segmentation module for extracting target regions on the surface of the target insulator that are more prone to hydrophobic degradation under thermal expansion and contraction and wet pollution conditions based on the operating environment data and geometric structure data; and a sub-region segmentation module for dividing the target region into multiple pixel sub-regions; and a temperature change imaging module. The module extracts temperature change data for each pixel sub-region based on the infrared time series matrix. The first filtering module filters pixel sub-regions whose temperature change characteristics deviate from a preset threshold based on the temperature change data and the temperature response baseline, obtaining a first candidate region set that may contain microcracks. The second filtering module removes pixel sub-regions from the first candidate region set that exhibit thermal anomalies due to dampness or frost based on the spectral feature matrix, obtaining a second candidate region set. The decision module performs spatial connectivity and temporal stability analysis on the second candidate region set, identifying connected regions whose temperature change characteristics deviate from the temperature response baseline by more than a preset deviation threshold across multiple time series windows. The system includes: a receiver module for the hydrophobic failure region of the target insulator, used to receive the target insulator's operating environment data, geometric structure data, and registered and normalized infrared time-series matrix and multispectral feature matrix; a baseline generation module, used to generate the temperature response baseline of the target insulator based on the operating environment data and a pre-built physical thermal network model; a region segmentation module, used to extract the target regions on the surface of the target insulator that are more prone to hydrophobic degradation under thermal expansion and contraction and wet pollution conditions, based on the operating environment data and geometric structure data; a sub-region segmentation module, used to divide the target region into multiple pixel sub-regions; and a temperature change extraction module, used to extract the temperature change based on the infrared time-series matrix. The system analyzes the temperature change data of each pixel sub-region. A first filtering module filters pixel sub-regions whose temperature change characteristics deviate from a preset threshold based on the temperature change data and the temperature response baseline, obtaining a first candidate region set that may contain microcracks. A second filtering module removes pixel sub-regions from the first candidate region set that exhibit thermal anomalies due to wetness or frost based on the spectral feature matrix, obtaining a second candidate region set. A decision module performs spatial connectivity analysis and temporal stability analysis on the second candidate region set, identifying connected regions whose temperature change characteristics deviate from the temperature response baseline by more than a preset deviation threshold within multiple time windows as hydrophobic failure regions of the target insulator.

[0068] In one example, the region segmentation module extracts the target region by: calculating the thermal stress amplitude of each region of the target insulator based on operating environment data and geometric structure data, and classifying regions with thermal stress amplitudes higher than a preset threshold as thermal anomaly sensitive regions; identifying areas of water film and frost accumulation on the target insulator based on operating environment data and geometric structure data, and classifying these areas as wet pollution / frost high-incidence areas; and marking the intersection of the thermal anomaly sensitive regions and the wet pollution / frost high-incidence areas as the target region.

[0069] In one example, the second screening module obtains the second candidate region by: extracting reflectance data of each pixel sub-region in the first candidate region set in N preset bands based on the spectral feature matrix to form a spectral vector, where N is a positive integer greater than one; calculating the spectral deviation between the spectral vector of each pixel sub-region and the reference spectral vector of the preset hydrophobicity degradation characterization; and removing pixel sub-regions with spectral deviation less than a preset deviation threshold from the first candidate set.

[0070] In one example, the decision module performs spatial connectivity and temporal stability analysis on the second candidate region set, including: dividing the second candidate region set into multiple connected domains, each of which consists of spatially adjacent pixel sub-regions in the second candidate region set; for each connected domain, calculating the sum of the deviations of the temperature change data of all pixel sub-regions within it from the temperature response baseline, as the cumulative deviation of the connected domain; calculating the average deviation of each connected domain based on the number of pixel sub-regions contained in each connected domain and the corresponding cumulative deviation; within M consecutive time windows, calculating the proportion of windows where the average deviation of each connected domain reaches a preset deviation threshold, where M is a positive integer greater than one; and marking connected domains where the window proportion exceeds the preset proportion as hydrophobic failure regions of the target insulator.

[0071] In one example, before calculating the cumulative deviation, the decision module also includes: counting the number of pixel sub-regions contained in each connected component; and filtering out connected components with a number of pixel sub-regions less than a preset threshold.

[0072] In one example, the decision module further includes: for each connected component, calculating the sum of its cumulative deviations over M time windows; and marking connected components whose cumulative deviations exceed a preset cumulative threshold as hydrophobic failure regions of the target insulator.

[0073] In one example, before performing spatial connectivity analysis and temporal stability analysis on the second candidate region set, the decision module also includes: calculating the temperature change signal-to-noise ratio of each pixel sub-region in the second candidate region set based on the infrared time series matrix and operating environment data; and removing pixel sub-regions with a temperature change signal-to-noise ratio lower than a preset signal-to-noise ratio threshold from the second candidate region set.

[0074] Exemplary electronic devices

[0075] Figure 3 An electronic device according to an embodiment of this application is illustrated. The electronic device may be the mobile device itself, or a standalone device independent of it, which may communicate with the mobile device to receive collected input signals from it and send selected target driving behaviors to it.

[0076] Figure 3 A block diagram of an electronic device according to an embodiment of this application is illustrated.

[0077] like Figure 3 As shown, the electronic device includes one or more processors and memory.

[0078] A processor can be a central processing unit (CPU) or other form of processing unit with data processing and / or instruction execution capabilities, and can control other components in an electronic device to perform desired functions.

[0079] The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and a processor may execute the program instructions to implement the driving behavior decision-making methods of the various embodiments of this application described above, and / or other desired functions.

[0080] In one example, the electronic device may also include input devices and output devices, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0081] Of course, for the sake of simplicity, Figure 3 Only some of the components of the electronic device relevant to this application are shown in this illustration; components such as buses, input / output interfaces, etc., are omitted. In addition, the electronic device may include any other suitable components depending on the specific application.

[0082] Exemplary computer-readable media

[0083] Embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the driving behavior decision-making methods according to various embodiments of this application described in the "Exemplary Methods" section above.

[0084] Computer-readable storage media may take the form of any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0085] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0086] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0087] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0088] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0089] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A method for imaging diagnosis of the hydrophobicity of composite insulators, characterized in that, include: Receive the target insulator's operating environment data, geometric structure data, and the registered and normalized infrared time series matrix and multispectral feature matrix; The temperature response baseline of the target insulator is generated based on the operating environment data and the pre-constructed physical thermal network model; Based on the operating environment data and the geometric structure data, target areas on the surface of the target insulator that are more prone to hydrophobic degradation under thermal expansion and contraction and wet pollution conditions are extracted. The target region is divided into multiple pixel sub-regions; Temperature change data of each pixel sub-region is extracted based on the infrared time series matrix; Based on the temperature change data and the temperature response baseline, pixel sub-regions whose temperature change characteristics deviate from a preset threshold are selected to obtain a first candidate region set that may contain microcracks. Based on the spectral feature matrix, pixel sub-regions in the first candidate region set that are thermally abnormal due to wetness or frost are removed from the first candidate region set to obtain the second candidate region set. Spatial connectivity analysis and temporal stability analysis are performed on the second candidate region set. Connected regions whose temperature change characteristics deviate from the temperature response baseline by more than a preset deviation threshold in multiple time windows are identified as hydrophobic failure regions of the target insulator.

2. The imaging diagnostic method for the hydrophobicity of composite insulators according to claim 1, characterized in that, The target region to be extracted includes: The thermal stress amplitude of each region of the target insulator is calculated based on the operating environment data and the geometric structure data, and the regions with thermal stress amplitudes higher than a preset threshold are classified as thermal anomaly sensitive areas. Based on the operating environment data and the geometric structure data, the target insulator's water film and areas prone to frost accumulation are identified, and these areas are classified as high-incidence wet / frost zones. The intersection area of ​​the thermally sensitive area and the area prone to wetness / frost is marked as the target area.

3. The imaging diagnostic method for the hydrophobicity of composite insulators according to claim 1, characterized in that, The second candidate region is obtained as follows: The reflectance data of each pixel sub-region in the first candidate region set in N preset bands are extracted based on the spectral feature matrix to form a spectral vector, where N is a positive integer greater than one. Calculate the spectral deviation between the spectral vector of each pixel sub-region and the reference spectral vector of the preset hydrophobicity degradation characterization; and Pixel sub-regions with spectral deviations less than a preset deviation threshold are removed from the first candidate region set.

4. The imaging diagnostic method for the hydrophobicity of composite insulators according to claim 1, characterized in that, The spatial connectivity analysis and temporal stability analysis of the second candidate region set include: The second candidate region set is divided into multiple connected regions, and each connected region is composed of spatially adjacent pixel sub-regions in the second candidate region set; For each connected component, the sum of the deviations of the temperature change data of all pixel sub-regions within it relative to the temperature response baseline is calculated as the cumulative deviation of the connected component. The average deviation of each connected component is calculated based on the number of pixel sub-regions contained in each connected component and the corresponding cumulative deviation. Within M consecutive time windows, the proportion of windows in which the average deviation of each connected component reaches a preset deviation threshold is calculated, where M is a positive integer greater than one. Connected regions whose window ratio exceeds a preset ratio are marked as hydrophobic failure regions of the target insulator.

5. The imaging diagnostic method for the hydrophobicity of composite insulators according to claim 4, characterized in that, The calculation of cumulative deviation is preceded by: Count the number of pixel sub-regions contained in each of the connected components; Connected regions whose number of pixel sub-regions is less than a preset threshold are filtered out.

6. The imaging diagnostic method for the hydrophobicity of composite insulators according to claim 4, characterized in that, Also includes: For each connected component, calculate the sum of its cumulative deviations within the M time windows; Connected regions whose sum of cumulative deviations exceeds a preset cumulative threshold are marked as hydrophobic failure regions of the target insulator.

7. The imaging diagnostic method for the hydrophobicity of composite insulators according to claim 1, characterized in that, Before performing spatial connectivity analysis and temporal stability analysis on the second candidate region set, the following steps are also included: Calculate the temperature change signal-to-noise ratio of each pixel sub-region in the second candidate region set based on the infrared time series matrix and the operating environment data; Pixel sub-regions whose temperature change signal-to-noise ratio is lower than a preset signal-to-noise ratio threshold are removed from the second candidate region set.

8. A composite insulator pollution hydrophobicity imaging diagnostic system, characterized in that, include: The receiving module is used to receive the target insulator's operating environment data, geometric structure data, and the registered and normalized infrared timing matrix and multispectral feature matrix. The baseline generation module is used to generate the temperature response baseline of the target insulator based on the operating environment data and the pre-built physical thermal network model. The region segmentation module is used to extract target regions on the surface of the target insulator that are more prone to hydrophobic degradation under thermal expansion and contraction and wet pollution conditions, based on the operating environment data and the geometric structure data. The sub-region division module is used to divide the target region into multiple pixel sub-regions; The temperature change extraction module is used to extract temperature change data of each pixel sub-region based on the infrared time series matrix. The first screening module is used to screen out pixel sub-regions whose temperature change characteristics deviate from a preset threshold based on the temperature change data and the temperature response baseline, so as to obtain a first candidate region set that may have microcracks. The second screening module is used to remove pixel sub-regions in the first candidate region set that have thermal anomalies due to wetness or frost based on the spectral feature matrix, thereby obtaining a second candidate region set. The decision module is used to perform spatial connectivity analysis and temporal stability analysis on the second candidate region set. It identifies the connected regions in which the temperature change characteristics deviate from the temperature response baseline by more than a preset deviation threshold in multiple time windows as the hydrophobic failure regions of the target insulator.

9. An electronic device comprising a memory and a processor, characterized in that, The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the method as described in any one of claims 1 to 7.

10. A computer storage medium storing computer-executable instructions thereon, characterized in that, When the computer-executable instructions are executed by a processor, they implement the steps of the method as described in any one of claims 1 to 7.

Images