Power distribution lightning arrester aging defect identification method and device considering environment confidence, electronic equipment and storage medium

Abstract

The invention discloses a power distribution lightning arrester aging defect identification method and device considering environment confidence, electronic equipment and a storage medium, and belongs to the field of intelligent monitoring of power equipment, and the method comprises the steps: collecting the inspection image data of a power distribution lightning arrester and the time sequence synchronous environment parameters of the power distribution lightning arrester; mapping the environment parameters to defect display conditions, quantifying the adaptation degree of the environment to aging defect display, and generating an environment confidence index; visual features are extracted from the inspection image data, similarity calculation is carried out on the visual features and standard defect features, and an image similarity coefficient is obtained; and weighting the image similarity coefficient based on the environment confidence index to form a fusion defect probability value, and outputting a power distribution lightning arrester aging defect identification result according to the fusion defect probability value. Therefore, by implementing the method and the device, the problems of low identification confidence and missed detection and misjudgment caused by incapability of distinguishing'equipment defect-free 'and'environmental implicit' due to lack of quantitative evaluation on environmental suitability in the prior art can be solved.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology for power equipment, specifically to a method, device, electronic device, and storage medium for identifying aging defects in distribution surge arresters that takes into account environmental confidence levels. Background Technology

[0002] As a critical overvoltage protection device in the power distribution network, the operating status of surge arresters directly affects the safety, stability, and reliability of the power grid. Due to long-term exposure to the complex outdoor natural environment, surge arresters are highly susceptible to aging caused by factors such as moisture intrusion, dirt accumulation, and electrical stress impacts, which can lead to decreased insulation performance, thermal breakdown, or even explosion accidents.

[0003] However, most existing visual recognition technologies focus on texture analysis or feature extraction of the acquired images themselves, neglecting the decisive influence of "environmental conditions" on "defect manifestation." In fact, many aging defects in surge arresters exhibit significant "environmental dependence," meaning they only become clearly visible under specific meteorological conditions (such as specific humidity, light, or temperature differences), while remaining "latent" in unsuitable environments. Existing technologies lack a quantitative assessment mechanism for whether the current environment is suitable for defect manifestation, often relying solely on the similarity of image features for forced judgments even when environmental conditions do not support clear defect feature presentation. This single-dimensional recognition mode of "looking at images without considering the environment" cannot distinguish the essential difference between "surge arrester without defects" and "environment causing defects not to manifest," easily leading to a deviation of the confidence level of the recognition results from the true situation, resulting in missed detections or misjudgments. Summary of the Invention

[0004] This invention provides a method, device, electronic device, and storage medium for identifying aging defects in distribution surge arresters that considers environmental confidence. This solves the problem in the prior art where the lack of quantitative assessment of environmental suitability makes it impossible to distinguish between "defect-free equipment" and "environmentally hidden defects," leading to low identification confidence and missed detections or misjudgments.

[0005] An embodiment of the present invention provides a method for identifying aging defects in distribution surge arresters considering environmental confidence levels, comprising: Collect inspection image data of the power distribution surge arrester and environmental parameters that are time-synchronized with the inspection image data; The environmental parameters are compared with the preset defect display conditions to quantitatively evaluate the degree of adaptability of the environmental parameters to defect display and generate an environmental confidence index for the inspection image data; wherein, the defect display conditions characterize the conditions that need to be met for the aging defects of the distribution arrester to be displayed. Visual features are extracted from the inspection image data to generate target visual feature vectors, and the similarity between the target visual feature vectors and the preset standard defect feature vectors is calculated to obtain the image similarity coefficient. Based on the environmental confidence index, the image similarity coefficient is weighted and corrected to generate a fused defect probability value that characterizes the possibility of aging defects in the distribution surge arrester. The final aging defect identification result of the distribution surge arrester is generated based on the fusion defect probability value.

[0006] Furthermore, the defect manifestation conditions include: the intensity value and duration of the corresponding environmental factor that enables the aging defects of the distribution surge arrester to manifest. The process of comparing environmental parameters with preset defect display conditions to quantitatively evaluate the adaptability of environmental parameters to defect display and generate an environmental confidence index for inspection image data includes: The phenomenal information used to trigger the manifestation of aging defects and the accompanying information that coexist with the phenomenal information are analyzed from environmental parameters. The phenomenal information is the intensity value and duration of the positive environmental factor that enhances the manifestation of defects. The accompanying information is the intensity value and duration of the negative environmental factor that negatively affects the manifestation of defects. The intensity value of the gain environment is compared with the intensity value of the manifestation environment factor to generate an intensity consistency verification result; the duration value of the environment is compared with the manifestation duration threshold to generate a duration consistency verification result. Based on the results of the strength consistency verification and the duration consistency verification, the defect activation index, which characterizes the probability of defect manifestation, is calculated. Based on the phenomenon information, conditional phenomena are determined, and the similarity between the conditional phenomena and preset standard defect phenomena is calculated to generate a phenomenon salience index that characterizes feature clarity. Based on the associated information, calculate the degree of interference to the phenomenon information and generate the associated interference index; The defect activation index and the phenomenon significance index are positively weighted and the associated interference index is negatively weighted. An environmental confidence index is generated based on the results of the positive and negative weighted calculations.

[0007] Furthermore, the calculation of the defect activation index, which characterizes the probability of defect manifestation, based on the intensity consistency verification results and the duration consistency verification results, includes: Based on the preset intensity-probability mapping table, the intensity occurrence probability is generated according to the intensity consistency verification result; Based on the preset duration-probability mapping table, the probability of duration occurrence is generated according to the duration consistency verification result; The defect activation index is calculated based on the probability of intensity occurrence and the probability of duration occurrence.

[0008] Furthermore, the negative environmental factors include wind factors and light factors; The step of calculating the degree of interference to the phenomenon information based on the associated information and generating an associated interference index includes: The intensity value and duration of the wind factor are input into a preset wind-jitter blur conversion function to calculate the image blur coefficient, which characterizes the degree of motion blur in the image. The intensity value and duration of the illumination factor are input into a preset illumination-contrast conversion function to calculate the image overexposure coefficient, which characterizes the degree of image exposure deviation. The weighted sum of the image blur coefficient and the image overexposure coefficient is calculated to generate the interference level value, and the interference level value is determined as the accompanying interference index.

[0009] Furthermore, the step of performing a positive weighted calculation on the defect activation index and the phenomenon significance index, a negative weighted calculation on the associated interference index, and generating an environmental confidence index based on the results of the positive and negative weighted calculations includes: The corresponding positive gain weights are configured for the defect activation index and the phenomenon significance index, respectively; Based on the positive gain weight, a weighted summation is performed on the defect activation index and the phenomenon significance index to generate an environmental susceptibility value. A negative attenuation weight is configured for the associated interference index, and the associated interference index is weighted based on the negative attenuation weight to generate an environmental interference value. An environmental confidence index is generated by correcting environmental susceptibility values ​​using environmental disturbance values; wherein the environmental disturbance values ​​and the environmental confidence index are negatively correlated.

[0010] Furthermore, the step of extracting visual features from the inspection image data to generate a target visual feature vector, and calculating the similarity between the target visual feature vector and a preset standard defect feature vector to obtain an image similarity coefficient, includes: Target detection and region cropping are performed on the inspection image data to extract the image of the defect area to be tested, which contains the components of the power distribution arrester. Visual features are extracted from the image of the defective region to be tested, generating texture distribution feature data and edge shape feature data of the image of the defective region to be tested, and the texture distribution feature data and edge shape feature data are vectorized and encoded to generate target visual feature vectors. The target visual feature vector is mapped to a preset feature vector space to generate a spatially mapped feature vector; the cosine similarity value between the spatially mapped feature vector and the preset standard defect feature vector is calculated to generate an image similarity coefficient.

[0011] Furthermore, the step of generating the final aging defect identification result of the distribution arrester based on the fusion defect probability value includes: The threshold drift compensation amount is calculated based on the associated interference index, and the preset benchmark judgment threshold is summed with the threshold drift compensation amount to generate a dynamic effective threshold. The fusion defect probability value is compared with the dynamic activation threshold. If the probability value of the fusion defect is greater than the dynamic activation threshold, an identification result representing a high-confidence defect is generated; otherwise, an identification result representing suspected interference or normality is generated.

[0012] Based on the above method embodiments, the present invention provides corresponding apparatus embodiments.

[0013] One embodiment of the present invention provides a device for identifying aging defects in distribution surge arresters that considers environmental confidence, comprising: a multi-source data acquisition module, an environmental confidence assessment module, a visual feature analysis module, a probability weighted fusion module, and a defect identification and judgment module; The multi-source data acquisition module is used to acquire inspection image data of the power distribution surge arrester and environmental parameters that are time-synchronized with the inspection image data. The environmental confidence assessment module is used to compare environmental parameters with preset defect manifestation conditions, quantify the degree of adaptability of environmental parameters to defect manifestation, and generate an environmental confidence index for inspection image data; wherein, the defect manifestation conditions characterize the conditions that need to be met for the aging defects of the distribution arrester to be manifested. The visual feature analysis module is used to extract visual features from the inspection image data, generate a target visual feature vector, and calculate the similarity between the target visual feature vector and the preset standard defect feature vector to obtain the image similarity coefficient. The probability weighted fusion module is used to weight and correct the image similarity coefficient according to the environmental confidence index, and generate a fused defect probability value that characterizes the possibility of aging defects in the power distribution arrester. The defect identification and judgment module is used to generate the final aging defect identification result of the distribution surge arrester based on the fused defect probability value.

[0014] Based on the above method embodiments, the present invention provides corresponding electronic device embodiments.

[0015] An embodiment of the present invention provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method for identifying aging defects of distribution surge arresters that considers environmental confidence as described in any of the above-described method embodiments.

[0016] Based on the above method embodiments, the present invention provides corresponding storage medium embodiments.

[0017] One embodiment of the present invention provides a storage medium storing a computer program thereon, wherein, when the computer program is running, it controls the device where the storage medium is located to execute the method for identifying aging defects of distribution surge arresters considering environmental confidence level as described in any of the above-described method embodiments.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a method, apparatus, electronic device, and storage medium for identifying aging defects in distribution surge arresters, considering environmental confidence levels. The method collects inspection image data of distribution surge arresters and their time-synchronized environmental parameters; maps the environmental parameters to defect manifestation conditions, quantifies the degree of environmental adaptability to the manifestation of aging defects, and generates an environmental confidence index; extracts visual features from the inspection image data and calculates similarity with standard defect features to obtain image similarity coefficients; weights the image similarity coefficients based on the environmental confidence index to form a fused defect probability value, and outputs the aging defect identification result of the distribution surge arrester accordingly.

[0019] This invention effectively solves the problem of the lack of a quantitative assessment mechanism for the environmental suitability of defect manifestation by mapping real-time environmental parameters to preset defect manifestation conditions and generating an environmental confidence index. Furthermore, by using this environmental confidence index as a weighting factor to correct the image similarity coefficient, it overcomes the blindness of traditional methods that rely solely on image features for forced judgment without considering environmental conditions. This avoids missed detections or misjudgments caused by defects not being manifested due to unsuitable environments, and significantly improves the confidence of the recognition results. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a method for identifying aging defects in distribution surge arresters that considers environmental confidence levels, provided by an embodiment of the present invention.

[0021] Figure 2 This is a schematic diagram of the structure of a power distribution surge arrester aging defect identification device that takes into account environmental confidence level, provided in an embodiment of the present invention. Detailed Implementation

[0022] 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.

[0023] like Figure 1 As shown, to address the problem in existing technologies where the lack of quantitative assessment of environmental suitability makes it impossible to distinguish between "defect-free equipment" and "environmentally hidden defects," leading to low identification confidence and missed detections or misjudgments, an embodiment of the present invention provides a method for identifying aging defects in distribution surge arresters that considers environmental confidence, comprising at least the following steps: Step S1: Collect inspection image data of the power distribution surge arrester and environmental parameters that are time-synchronized with the inspection image data; Specifically, in one embodiment of the present invention, the method for identifying aging defects in power distribution surge arresters first performs a data acquisition step, namely, collecting inspection image data of power distribution surge arresters and environmental parameters that are time-synchronized with the inspection image data.

[0024] Regarding inspection image data for distribution surge arresters, this data can be acquired through high-definition visible light cameras mounted on drones, or through real-time capture by ground inspection robots or online monitoring cameras fixed on distribution towers. The inspection image data should clearly include the main structure of the distribution surge arrester, its skirts, and connecting hardware to ensure sufficient image information for subsequent visual feature extraction. To ensure image quality meets the requirements of algorithm analysis, the acquisition equipment automatically adjusts the focus and exposure parameters during the acquisition process to ensure that the stored inspection image data resolution meets the preset industrial inspection standards.

[0025] Regarding environmental parameters synchronized with the inspection image data, these parameters form the data foundation for subsequent assessment of the environmental confidence index. The environmental parameters must maintain a strict correspondence with the inspection image data in the time dimension. Specifically, time synchronization means that the meteorological conditions recorded by the environmental parameters must be the real-time state at the instant the inspection image data is captured. In practice, this can be achieved by assigning a unified timestamp accurate to milliseconds to each frame of acquired inspection image data and each set of environmental parameters, and then associating and storing the inspection image data and environmental parameters based on the timestamp index. If the environmental parameters are obtained from an external weather station, the environmental parameters closest to the time of the inspection image data acquisition are selected as the synchronization data.

[0026] To support the subsequent dual evaluation of "defect manifestation" and "image interference", the environmental parameters collected in this embodiment are specifically divided into two categories of key information: phenomenon information and associated information.

[0027] The first category is phenomenon information. According to the definition in this embodiment, phenomenon information specifically refers to data related to "gain environmental factors" that amplify the manifestation of aging defects, including the intensity value and duration of the gain environmental factors. Gain environmental factors are environmental factors that directly affect the distribution surge arrester and can promote the transformation of aging defects from a latent state to a manifest state. Specifically, the intensity value refers to the strength of meteorological effects under the current environment. In scenarios where damp aging is the primary detection target, the intensity value is expressed as the relative humidity or rainfall value; in scenarios where pollution aging is the primary detection target, the intensity value is expressed as the salt density or dust density value in the air. The duration of effect specifically refers to the length of time that the aforementioned gain environmental factors maintain their corresponding intensity, such as the number of minutes a high humidity environment has persisted. Collecting these two dimensions is to determine in subsequent steps whether the current environment has accumulated sufficient energy (i.e., the integral effect of intensity and time) to make the defect characteristics manifest.

[0028] The second category is associated information. According to the definition in this embodiment, associated information specifically refers to data related to "negative environmental factors" that negatively impact the manifestation of aging defects (i.e., interfere with imaging or mask features). As stated in the claim, the negative environmental factors in this embodiment specifically include "wind factors" and "lighting factors."

[0029] To comprehensively assess the interference level of negative environmental factors, the collected associated information must cover both intensity and duration dimensions. For wind factors, the collected data includes wind speed values ​​and wind duration. Wind speed values ​​are obtained using a high-precision anemometer to assess the instantaneous wind force level; wind duration is used to assess how long the wind speed is maintained, as sustained strong winds are more likely to cause continuous resonant vibrations in power distribution arresters or data acquisition equipment, resulting in motion blur that is difficult to recover from. For illumination factors, the collected data includes illumination intensity values ​​and illumination duration. Illumination intensity values ​​are obtained using a photosensor to assess the current ambient lumen value; illumination duration is used to assess the dwell time in strong or weak light environments, as continuous strong light exposure can cause thermal noise or persistent glare in the lens, leading to overexposure or abnormal contrast in the image.

[0030] By collecting multi-dimensional environmental parameters including environmental intensity, environmental duration, wind speed, and light intensity, and strictly aligning these environmental parameters with the inspection image data in a specific time sequence, this invention can provide accurate data support for subsequent quantitative assessment of the environment's adaptability to defect manifestation. This avoids the one-sidedness of relying solely on images for judgment without considering environmental conditions, and effectively ensures the integrity and logical consistency of the source data.

[0031] Step S2: Compare the environmental parameters with the preset defect display conditions, quantify and evaluate the adaptability of the environmental parameters to defect display, and generate an environmental confidence index for the inspection image data; wherein, the defect display conditions characterize the conditions that need to be met for the aging defects of the power distribution arrester to be displayed. In a preferred embodiment, the defect manifestation conditions include: the intensity value and duration of the corresponding environmental factor that enables the aging defects of the distribution surge arrester to manifest. The process of comparing environmental parameters with preset defect display conditions to quantitatively evaluate the adaptability of environmental parameters to defect display and generate an environmental confidence index for inspection image data includes: The phenomenal information used to trigger the manifestation of aging defects and the accompanying information that coexist with the phenomenal information are analyzed from environmental parameters. The phenomenal information is the intensity value and duration of the positive environmental factor that enhances the manifestation of defects. The accompanying information is the intensity value and duration of the negative environmental factor that negatively affects the manifestation of defects. The intensity value of the gain environment is compared with the intensity value of the manifestation environment factor to generate an intensity consistency verification result; the duration value of the environment is compared with the manifestation duration threshold to generate a duration consistency verification result. Based on the results of the strength consistency verification and the duration consistency verification, the defect activation index, which characterizes the probability of defect manifestation, is calculated. Based on the phenomenon information, conditional phenomena are determined, and the similarity between the conditional phenomena and preset standard defect phenomena is calculated to generate a phenomenon salience index that characterizes feature clarity. Based on the associated information, calculate the degree of interference to the phenomenon information and generate the associated interference index; The defect activation index and the phenomenon significance index are positively weighted and the associated interference index is negatively weighted. An environmental confidence index is generated based on the results of the positive and negative weighted calculations.

[0032] In a preferred embodiment, calculating the defect activation index, which characterizes the probability of defect manifestation, based on the intensity consistency verification result and the duration consistency verification result includes: Based on the preset intensity-probability mapping table, the intensity occurrence probability is generated according to the intensity consistency verification result; Based on the preset duration-probability mapping table, the probability of duration occurrence is generated according to the duration consistency verification result; The defect activation index is calculated based on the probability of intensity occurrence and the probability of duration occurrence.

[0033] In a preferred embodiment, the negative environmental factors include wind factors and light factors; The step of calculating the degree of interference to the phenomenon information based on the associated information and generating an associated interference index includes: The intensity value and duration of the wind factor are input into a preset wind-jitter blur conversion function to calculate the image blur coefficient, which characterizes the degree of motion blur in the image. The intensity value and duration of the illumination factor are input into a preset illumination-contrast conversion function to calculate the image overexposure coefficient, which characterizes the degree of image exposure deviation. The weighted sum of the image blur coefficient and the image overexposure coefficient is calculated to generate the interference level value, and the interference level value is determined as the accompanying interference index.

[0034] In a preferred embodiment, the step of positively weighting the defect activation index and the phenomenon significance index, negatively weighting the associated interference index, and generating an environmental confidence index based on the results of the positive and negative weighting calculations includes: The corresponding positive gain weights are configured for the defect activation index and the phenomenon significance index, respectively; Based on the positive gain weight, a weighted summation is performed on the defect activation index and the phenomenon significance index to generate an environmental susceptibility value. A negative attenuation weight is configured for the associated interference index, and the associated interference index is weighted based on the negative attenuation weight to generate an environmental interference value. An environmental confidence index is generated by correcting environmental susceptibility values ​​using environmental disturbance values; wherein the environmental disturbance values ​​and the environmental confidence index are negatively correlated.

[0035] Specifically, this step aims to address the technical challenge of traditional visual recognition algorithms neglecting the influence of environmental factors on the "physical manifestation" of imaging features. In its implementation, this step uses multi-dimensional environmental analysis and mathematical modeling to calculate an environmental confidence index that reflects whether the current environment is suitable for defect identification.

[0036] First, two distinct types of key information were extracted from the collected environmental parameters: “phenomenal information” that triggers the manifestation of aging defects, and “accompanying information” that accompanies the phenomenon information and has a negative effect on the manifestation of defects.

[0037] Specifically, in this embodiment, the phenomenon information corresponds to the "gain environmental factor," which is an environmental factor that enhances the manifestation of defects. The data dimensions of the phenomenon information include the intensity value of the gain environmental factor (such as the percentage of air humidity or the salt density value) and the duration of its effect (such as the number of minutes the high humidity state is maintained).

[0038] The accompanying information specifically corresponds to "negative environmental factors," which are environmental factors that interfere with image quality or mask defective features. In this embodiment, the accompanying information encompasses wind and illumination factors. To comprehensively quantify the interference, the dimensions of the collected accompanying information data include not only wind speed and illumination intensity values, but also strictly include the duration of wind and illumination. This is because sustained environmental stress (such as continuous strong winds or continuous intense light) often produces more severe image quality degradation than instantaneous fluctuations.

[0039] Next, the system calls the preset defect manifestation condition database and parses out the intensity values ​​of the manifestation environmental factors (i.e., the manifestation threshold). ) and display duration threshold ( This step compares the intensity values ​​of the acquired gain environmental factors with the intensity values ​​of the manifest environmental factors to generate an intensity consistency verification result. To quantify this comparison, this embodiment uses the "relative intensity ratio method" for calculation. The specific calculation formula is as follows: In the formula, This is the result of the strength consistency verification. These are the currently collected measured values ​​of environmental intensity; This is the preset display threshold. When When the value is less than 1, it indicates that the current environmental intensity has not reached the manifestation threshold, and the probability of defect manifestation is extremely low; when... A value ≥1 indicates that the environmental intensity has met the standard, and the larger the value, the stronger the driving force. Similarly, the environmental duration value is compared with the display duration threshold to calculate the duration consistency verification result. The specific calculation formula is as follows: In the formula, This is the result of the duration consistency check; This represents the measured duration of the current environmental condition. This is the preset display duration threshold. The "consistency verification result" here... and ) is a quantitative, dimensionless value that objectively reflects the degree to which the current environment approaches or exceeds the manifestation threshold, providing an accurate index basis for subsequent table lookups.

[0040] Subsequently, based on the above verification results, the defect activation index, representing the probability of defect manifestation, is calculated. Specifically, this embodiment uses a deterministic lookup table mechanism to replace fuzzy reasoning: based on a preset "intensity-probability mapping table," the probability of intensity occurrence is generated according to the intensity consistency verification results. It should be noted that this preset "intensity-probability mapping table" is constructed based on the statistical patterns of historical operation and maintenance big data of the distribution network. The specific construction method is as follows: statistically analyze the historically diagnosed surge arrester aging defect records in the region, trace back the environmental intensity value when each defect was discovered, statistically analyze the defect detection rate under different intensity ranges (such as humidity 40%-50%, 50%-60%, etc.), and store these discrete detection rate values ​​as a mapping table. The probability values ​​recorded in this mapping table are positively correlated with the environmental intensity. Similarly, based on a preset "duration-probability mapping table," the probability of duration occurrence is generated according to the duration consistency verification results. This "duration-probability mapping table" is set based on the physical penetration model of material aging (such as the moisture penetration kinetic model), quantifying the integral effect relationship between the duration of environmental action and the degree of defect feature manifestation. Finally, the probability of intensity occurrence and the probability of duration occurrence are fused and calculated (e.g., the weighted product of the two) to generate the defect activation index.

[0041] While calculating the defect activation index, it is also necessary to evaluate the visual clarity of the defect features, i.e., to calculate the phenomenon salience index. Based on the phenomenon information, the conditional phenomenon under the current environment (i.e., the actual meteorological state vector) is determined, and the vector space similarity between the conditional phenomenon and the preset standard defect phenomenon is calculated. The standard defect phenomenon refers to the theoretically most favorable ideal environmental state for the presentation of defect features (e.g., a specific high-humidity and rain-free condition). The higher the similarity between the two, the higher the generated phenomenon salience index, indicating that under the current environment, once the defect appears, its visual features (such as texture and color contrast) will be closer to the standard form and easier to be captured by the visual algorithm. The preset "standard defect phenomenon" is an ideal reference template built based on an expert knowledge base, i.e., a pre-selected environmental vector corresponding to several sample data with clear aging characteristics collected under typical highly suitable environments (e.g., high humidity and rain-free conditions).

[0042] Simultaneously, it is necessary to quantify the degree of interference of negative environmental factors on image quality, i.e., to calculate the associated interference index. To accurately capture the cumulative interference effect of environmental factors, this step performs the following calculations: For wind factors, wind speed values ​​and wind duration are simultaneously input into a preset "wind-flicker blur conversion function." This conversion function is obtained based on wind resistance calibration experiments of the image acquisition equipment. By testing the micro-amplitude of the equipment and the resulting image pixel displacement (motion blur kernel) under different wind speed levels and durations in a laboratory wind tunnel environment, a mapping relationship between wind parameters and image blur coefficients is constructed.

[0043] For illumination factors, both the illumination intensity value and the duration of illumination are simultaneously input into a preset "illuminance-contrast conversion function". This conversion function can evaluate the impact of continuous illumination on the sensor's photosensitive element, mapping it to an image overexposure coefficient that characterizes the degree of image exposure deviation (such as overexposure or underexposure). This conversion function is set based on the image sensor's photoelectric response characteristic curve (Gamma curve) and is used to evaluate the degree to which the illumination lumen value deviates from the sensor's linear operating range (i.e., overexposure or underexposure). Subsequently, a weighted sum of the image blur coefficient and the image overexposure coefficient is calculated to generate an interference level value, which is then defined as the associated interference index. The higher the associated interference index value, the more severe the noise interference caused by environmental factors to the image.

[0044] Finally, based on the three component indices calculated above, a weighted correction logic is used to generate the final environmental confidence index. The specific calculation process is as follows: Positive gain weights are assigned to the defect activation index and the phenomenon significance index, and a weighted summation is performed based on these weights to generate the environmental susceptibility value. Simultaneously, a negative attenuation weight is assigned to the associated interference index, and this weight is used to weight the associated interference index to generate the environmental interference value. The environmental susceptibility value is then corrected using the environmental interference value to ensure a negative correlation between the environmental interference value and the final generated environmental confidence index.

[0045] For example, the following modified formula can be used for calculation: In the formula, Represents the environmental confidence index; Represents the defect activation index; The significance index represents the phenomenon; Represents the associated interference index; , Represents positive gain weights; This represents a negative decay weight.

[0046] Through the above calculation steps, this embodiment can accurately quantify the "credibility" of the current environment for visual recognition, ensuring that the subsequent recognition process is only given high weight when the environmental confidence index meets the requirements. This effectively avoids false detections and missed detections caused by forced recognition in harsh or unsuitable environments, and significantly improves the robustness of identifying aging defects in power distribution surge arresters.

[0047] Step S3: Extract visual features from the inspection image data, generate target visual feature vectors, and calculate the similarity between the target visual feature vectors and the preset standard defect feature vectors to obtain the image similarity coefficients. In a preferred embodiment, the step of extracting visual features from the inspection image data to generate a target visual feature vector, and calculating the similarity between the target visual feature vector and a preset standard defect feature vector to obtain an image similarity coefficient, includes: Target detection and region cropping are performed on the inspection image data to extract the image of the defect area to be tested, which contains the components of the power distribution arrester. Visual features are extracted from the image of the defective region to be tested, generating texture distribution feature data and edge shape feature data of the image of the defective region to be tested, and the texture distribution feature data and edge shape feature data are vectorized and encoded to generate target visual feature vectors. The target visual feature vector is mapped to a preset feature vector space to generate a spatially mapped feature vector; the cosine similarity value between the spatially mapped feature vector and the preset standard defect feature vector is calculated to generate an image similarity coefficient.

[0048] Specifically, the core of this step is to extract key equipment features from the complex inspection background and quantify the degree of similarity between the current equipment state and the known defect state through mathematical similarity measurement.

[0049] First, refined preprocessing is performed on the inspection image data, specifically including target detection and region cropping. Since inspection image data typically contains complex background noise such as sky, towers, and trees, direct full-image analysis would waste computational resources and easily introduce false positives. Therefore, a pre-trained target detection algorithm (e.g., a detection model based on convolutional neural networks) is used to locate the position coordinates of the distribution surge arrester in the inspection image data. Based on the located position coordinates, a rectangular region containing the distribution surge arrester component is cropped from the original image to obtain the image of the defect area to be tested. The image of the defect area to be tested only contains the potential defect texture of the surge arrester body and its surface, eliminating interference from the background environment.

[0050] Next, deep visual feature extraction is performed on the image of the defective region to be tested. To comprehensively capture the morphological features of aging defects, the feature extraction process encompasses two dimensions: texture distribution and edge shape. Texture analysis algorithms (such as Gray-Level Co-occurrence Matrix, GLCM) are used to extract texture distribution feature data from the image of the defective region to be tested. This data reflects the roughness, contrast, and spatial correlation of pixel gray levels on the arrester surface, effectively characterizing the microscopic texture of dirt deposition or material degradation. Simultaneously, edge detection algorithms (such as the Canny operator or Sobel operator) are used to extract edge shape feature data from the image of the defective region to be tested. This data reflects the crack direction, damage contour, and geometric deformation of the arrester's skirt surface. Subsequently, the extracted texture distribution feature data and edge shape feature data are vectorized (e.g., serial splicing or feature fusion) to generate a high-dimensional target visual feature vector. The target visual feature vector is the numerical representation of the image of the defective region to be tested in the feature space.

[0051] Subsequently, the target visual feature vector is mapped to a predefined feature vector space to generate a spatially mapped feature vector. The predefined feature vector space is a standardized metric space constructed based on a large amount of sample data. The mapping process aims to eliminate differences in the dimensions and scales of the feature vectors, ensuring the accuracy of subsequent calculations. A pre-set database is then invoked to obtain standard defect feature vectors. These standard defect feature vectors are ideal feature vectors that characterize typical aging defects (such as typical cracks and typical ablation marks), obtained in advance through experimental or historical data accumulation.

[0052] Finally, the cosine similarity between the spatially mapped feature vector and the standard defect feature vector is calculated. Cosine similarity measures the difference by calculating the cosine of the angle between the two vectors in multidimensional space. Compared to Euclidean distance, cosine similarity focuses more on the alignment of feature vectors in direction, that is, on the similarity of defect "patterns" rather than the difference in absolute pixel values. The specific calculation formula is as follows: In the formula, Represents the image similarity coefficient; Represents the feature vector of spatial mapping; Represents the standard defect feature vector; The eigenvector representing the spatial mapping is in the th... Numerical components in each dimension; The representative standard defect feature vector is in the first... Numerical components in each dimension; This represents the total dimension of the feature vector; This represents the magnitude operation of a vector.

[0053] Through the above steps, this embodiment can accurately quantify the similarity between the device under test and the known defect morphology from two dimensions: texture and shape. The generated image similarity coefficient can objectively reflect the visual confidence that there are aging defects on the surface of the power distribution arrester, providing a reliable image basis for subsequent comprehensive judgment in combination with environmental factors.

[0054] Step S4: Based on the environmental confidence index, the image similarity coefficient is weighted and corrected to generate a fused defect probability value that characterizes the possibility of aging defects in the power distribution arrester. Specifically, this step is the logical hub connecting "environmental perception" and "visual recognition". Its core purpose is to use macroscopic environmental credibility to modulate microscopic visual similarity, thereby establishing a dynamic confidence-dependent mechanism.

[0055] Specifically, the environmental confidence index generated in step S2 is first normalized. Since the environmental confidence index is a composite value calculated based on positive and negative weighting, its original value range may not fall within the standard unit interval. Therefore, using preset maximum and minimum theoretical confidence values, a linear normalization function or a sigmoid nonlinear mapping function is constructed to map the environmental confidence index to a closed interval between zero and one, generating confidence weights. The confidence weights directly reflect the "effective contribution rate" of the current environmental conditions to visual imaging. The closer the value is to one, the more ideal the environment and the more reliable the visual features; the closer the value is to zero, the stronger the environmental interference and the lower the reliability of the visual features.

[0056] Subsequently, based on the confidence weights, a product correction operation or an exponential decay modulation operation is performed on the image similarity coefficients obtained in step S3 to generate a fused defect probability value. From a data processing perspective, the image similarity coefficients objectively characterize the degree of agreement between the target object and standard defect features in terms of texture and edge features (i.e., feature matching degree); while the confidence weights quantitatively characterize the effectiveness or reliability of the collected image data in truly reflecting the device status under the current environmental conditions. Fusing these two calculations is essentially performing a "probabilistic noise reduction based on environmental effectiveness" operation.

[0057] The specific correction calculation formula is as follows: Alternatively, a modulation formula with stronger nonlinearity can be used: In the formula, This represents the probability value of the generated fusion defect, which is a dimensionless probability score. This represents the image similarity coefficient calculated in step S3; This represents the confidence weights after normalization. This represents the preset modulation sensitivity coefficient, used to control the rate at which environmental factors affect the final result.

[0058] Through the above calculations, when the environment is harsh and the confidence weight is low, even if the image similarity coefficient calculated by the visual algorithm is high (which may be due to misjudgment caused by water stains or light spots), the final fusion defect probability value will be forcibly lowered, thereby suppressing the amplitude of the false alarm signal. Conversely, only under the dual conditions of suitable environment and matching image features will a higher fusion defect probability value be output.

[0059] By implementing this step, the present invention can achieve adaptive calibration of the identification results, ensuring that the final output of the identification results of aging defects of distribution surge arresters is strictly controlled by the interpretability of the environment. From a mathematical logic perspective, it eliminates the one-sidedness of discussing image recognition in isolation from environmental conditions and significantly reduces the false alarm rate under complex weather conditions.

[0060] Step S5: Generate the final aging defect identification result of the distribution surge arrester based on the fusion defect probability value.

[0061] In a preferred embodiment, generating the final aging defect identification result of the distribution arrester based on the fusion defect probability value includes: The threshold drift compensation amount is calculated based on the associated interference index, and the preset benchmark judgment threshold is positively superimposed with the threshold drift compensation amount to generate a dynamically effective threshold. The fusion defect probability value is compared with the dynamic activation threshold. If the probability value of the fusion defect is greater than the dynamic activation threshold, an identification result representing a high-confidence defect is generated; otherwise, an identification result representing suspected interference or normality is generated.

[0062] Specifically, this step is the final decision-making stage of the entire identification process. In order to further improve the system's anti-interference capability under complex weather conditions, this step abandons the traditional fixed threshold judgment mode and introduces a dynamic threshold drift mechanism based on the degree of environmental interference.

[0063] Specifically, firstly, the associated interference index calculated in step S2 is obtained, and a preset benchmark judgment threshold is retrieved. The benchmark judgment threshold is a judgment threshold (e.g., 0.75) determined based on an ideal experimental environment (i.e., no wind and no light interference). Subsequently, the threshold drift compensation amount is calculated based on the associated interference index. The physical meaning of the threshold drift compensation amount is: when environmental interference exists, in order to prevent noise signals from being misjudged as defect signals, an additional judgment difficulty value needs to be added. This compensation amount is calculated through linear mapping or nonlinear mapping, typically using the following calculation formula: In the formula, This represents the threshold drift compensation amount; Represents the associated interference index; This represents the preset drift sensitivity coefficient (e.g., a value of 0.2). This formula indicates that the more severe the environmental disturbance, the greater the required compensation.

[0064] Next, the preset baseline judgment threshold and the calculated threshold drift compensation amount are positively superimposed to generate the dynamically effective threshold. Positive superposition refers to numerical addition, and its calculation formula is as follows: In the formula, This represents the final dynamic threshold used for judgment; This represents the baseline judgment threshold. Through this step, the judgment threshold automatically increases as environmental interference increases. For example, in strong wind and strong light environments, the associated interference index is high, and the dynamic effective threshold may automatically drift from the baseline of 0.75 to 0.85, thus creating a more stringent "filter".

[0065] Finally, the fusion defect probability value generated in step S4 is compared with the dynamic effective threshold, and the final identification conclusion is generated based on the comparison result: If the probability value of the fusion defect is greater than the dynamic effective threshold, it indicates that even under the current level of interference, and after the weighted suppression of environmental confidence and the dynamic increase of the judgment threshold, the defect characteristics of the target still exist significantly. Therefore, an identification result representing a high-confidence defect is generated, and a corresponding alarm or maintenance work order is triggered. If the probability value of the fusion defect is less than or equal to the dynamic activation threshold, it indicates that the target has failed to pass the strict judgment criteria under the current environment to generate identification results that characterize suspected interference or normal operation. It is classified as normal equipment or suspected signal caused by environmental interference and no defect alarm is triggered.

[0066] By implementing this step, the present invention can construct an adaptive decision-making logic that is "more cautious in judgment as the environment becomes more severe," which effectively solves the problem of frequent false alarms caused by the fixed threshold being too low in severe weather conditions in traditional methods, and ensures that the final output recognition result has extremely high engineering usability and accuracy.

[0067] Based on the above method embodiments, the present invention provides corresponding apparatus embodiments.

[0068] like Figure 2As shown, an embodiment of the present invention provides a device for identifying aging defects in distribution surge arresters that considers environmental confidence, including: a multi-source data acquisition module, an environmental confidence assessment module, a visual feature analysis module, a probability weighted fusion module, and a defect identification and judgment module; The multi-source data acquisition module is used to acquire inspection image data of the power distribution surge arrester and environmental parameters that are time-synchronized with the inspection image data. The environmental confidence assessment module is used to compare environmental parameters with preset defect manifestation conditions, quantify the degree of adaptability of environmental parameters to defect manifestation, and generate an environmental confidence index for inspection image data; wherein, the defect manifestation conditions characterize the conditions that need to be met for the aging defects of the distribution arrester to be manifested. The visual feature analysis module is used to extract visual features from the inspection image data, generate a target visual feature vector, and calculate the similarity between the target visual feature vector and the preset standard defect feature vector to obtain the image similarity coefficient. The probability weighted fusion module is used to weight and correct the image similarity coefficient according to the environmental confidence index, and generate a fused defect probability value that characterizes the possibility of aging defects in the power distribution arrester. The defect identification and judgment module is used to generate the final aging defect identification result of the distribution surge arrester based on the fused defect probability value.

[0069] It should be noted that the embodiments of the device described above correspond to the embodiments of the present invention described above, and can realize the method for identifying aging defects of distribution surge arresters considering environmental confidence level as described in any one of the above embodiments of the present invention. Furthermore, the embodiments of the device described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without creative effort.

[0070] Based on the above-described method embodiments of the present invention, a corresponding embodiment of an electronic device is provided.

[0071] An embodiment of the present invention provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method for identifying aging defects of distribution surge arresters considering environmental confidence as described in any one of the present invention, or, when the processor executes the computer program, it implements the functions of each module in the above-described device embodiments.

[0072] For example, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the terminal device.

[0073] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0074] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0075] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function, etc.; the data storage area may store data created based on the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital card (SD card), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0076] Based on the above method embodiments, the present invention provides corresponding storage medium embodiments; Another embodiment of the present invention provides a storage medium including a stored computer program, wherein, when the computer program is running, it controls the device where the storage medium is located to execute any of the above-described methods for identifying aging defects in distribution surge arresters that considers environmental confidence levels.

[0077] The aforementioned storage medium is a computer-readable storage medium, and the computer program includes computer program code, which may be in the form of source code, object code, executable file, or certain intermediate forms. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0078] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0079] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for identifying aging defects in distribution surge arresters considering environmental confidence levels, characterized in that, include: Collect inspection image data of the power distribution surge arrester and environmental parameters that are time-synchronized with the inspection image data; The environmental parameters are compared with the preset defect display conditions to quantitatively evaluate the degree of adaptability of the environmental parameters to defect display and generate an environmental confidence index for the inspection image data; wherein, the defect display conditions characterize the conditions that need to be met for the aging defects of the distribution arrester to be displayed. Visual features are extracted from the inspection image data to generate target visual feature vectors, and the similarity between the target visual feature vectors and the preset standard defect feature vectors is calculated to obtain the image similarity coefficient. Based on the environmental confidence index, the image similarity coefficient is weighted and corrected to generate a fused defect probability value that characterizes the possibility of aging defects in the distribution surge arrester. The final aging defect identification result of the distribution surge arrester is generated based on the fusion defect probability value.

2. The method for identifying aging defects in distribution surge arresters considering environmental confidence level as described in claim 1, characterized in that, The defect manifestation conditions include: the intensity value and duration of the corresponding environmental factor that enables the aging defects of the distribution surge arrester to manifest. The process of comparing environmental parameters with preset defect display conditions to quantitatively evaluate the adaptability of environmental parameters to defect display and generate an environmental confidence index for inspection image data includes: The phenomenal information used to trigger the manifestation of aging defects and the accompanying information that coexist with the phenomenal information are analyzed from environmental parameters. The phenomenal information is the intensity value and duration of the positive environmental factor that enhances the manifestation of defects. The accompanying information is the intensity value and duration of the negative environmental factor that negatively affects the manifestation of defects. The intensity value of the gain environment is compared with the intensity value of the manifestation environment factor to generate an intensity consistency verification result; the duration value of the environment is compared with the manifestation duration threshold to generate a duration consistency verification result. Based on the results of the strength consistency verification and the duration consistency verification, the defect activation index, which characterizes the probability of defect manifestation, is calculated. Based on the phenomenon information, conditional phenomena are determined, and the similarity between the conditional phenomena and preset standard defect phenomena is calculated to generate a phenomenon salience index that characterizes feature clarity. Based on the associated information, calculate the degree of interference to the phenomenon information and generate the associated interference index; The defect activation index and the phenomenon significance index are positively weighted and the associated interference index is negatively weighted. An environmental confidence index is generated based on the results of the positive and negative weighted calculations.

3. The method for identifying aging defects in distribution surge arresters considering environmental confidence level as described in claim 2, characterized in that, The defect activation index, which characterizes the probability of defect manifestation, is calculated based on the strength consistency verification results and the duration consistency verification results, including: Based on the preset intensity-probability mapping table, the intensity occurrence probability is generated according to the intensity consistency verification result; Based on the preset duration-probability mapping table, the probability of duration occurrence is generated according to the duration consistency verification result; The defect activation index is calculated based on the probability of intensity occurrence and the probability of duration occurrence.

4. The method for identifying aging defects in distribution surge arresters considering environmental confidence level as described in claim 3, characterized in that, The negative environmental factors include wind factors and light factors; The step of calculating the degree of interference to the phenomenon information based on the associated information and generating an associated interference index includes: The intensity value and duration of the wind factor are input into a preset wind-jitter blur conversion function to calculate the image blur coefficient, which characterizes the degree of motion blur in the image. The intensity value and duration of the illumination factor are input into a preset illumination-contrast conversion function to calculate the image overexposure coefficient, which characterizes the degree of image exposure deviation. The weighted sum of the image blur coefficient and the image overexposure coefficient is calculated to generate the interference level value, and the interference level value is determined as the accompanying interference index.

5. The method for identifying aging defects in distribution surge arresters considering environmental confidence level as described in claim 4, characterized in that, The process involves positively weighting the defect activation index and the phenomenon significance index, negatively weighting the associated interference index, and generating an environmental confidence index based on the results of both positive and negative weighting calculations. The corresponding positive gain weights are configured for the defect activation index and the phenomenon significance index, respectively; Based on the positive gain weight, a weighted summation is performed on the defect activation index and the phenomenon significance index to generate an environmental susceptibility value. A negative attenuation weight is configured for the associated interference index, and the associated interference index is weighted based on the negative attenuation weight to generate an environmental interference value. An environmental confidence index is generated by correcting environmental susceptibility values ​​using environmental disturbance values; wherein the environmental disturbance values ​​and the environmental confidence index are negatively correlated.

6. The method for identifying aging defects in distribution surge arresters considering environmental confidence level as described in claim 5, characterized in that, The process of extracting visual features from inspection image data to generate a target visual feature vector, and calculating the similarity between the target visual feature vector and a preset standard defect feature vector to obtain an image similarity coefficient, includes: Target detection and region cropping are performed on the inspection image data to extract the image of the defect area to be tested, which contains the components of the power distribution arrester. Visual features are extracted from the image of the defective region to be tested, generating texture distribution feature data and edge shape feature data of the image of the defective region to be tested, and the texture distribution feature data and edge shape feature data are vectorized and encoded to generate target visual feature vectors. The target visual feature vector is mapped to a preset feature vector space to generate a spatially mapped feature vector; the cosine similarity value between the spatially mapped feature vector and the preset standard defect feature vector is calculated to generate an image similarity coefficient.

7. The method for identifying aging defects in distribution surge arresters considering environmental confidence level as described in claim 6, characterized in that, The step of generating the final aging defect identification result of the distribution arrester based on the fusion defect probability value includes: The threshold drift compensation amount is calculated based on the associated interference index, and the preset benchmark judgment threshold is summed with the threshold drift compensation amount to generate a dynamic effective threshold. The fusion defect probability value is compared with the dynamic activation threshold. If the probability value of the fusion defect is greater than the dynamic activation threshold, an identification result representing a high-confidence defect is generated; otherwise, an identification result representing suspected interference or normality is generated.

8. A device for identifying aging defects in distribution surge arresters considering environmental confidence levels, characterized in that, include: Multi-source data acquisition module, environmental confidence assessment module, visual feature analysis module, probability weighted fusion module, and defect identification and judgment module; The multi-source data acquisition module is used to acquire inspection image data of the power distribution surge arrester and environmental parameters that are time-synchronized with the inspection image data. The environmental confidence assessment module is used to compare environmental parameters with preset defect manifestation conditions, quantify the degree of adaptability of environmental parameters to defect manifestation, and generate an environmental confidence index for inspection image data; wherein, the defect manifestation conditions characterize the conditions that need to be met for the aging defects of the distribution arrester to be manifested. The visual feature analysis module is used to extract visual features from the inspection image data, generate a target visual feature vector, and calculate the similarity between the target visual feature vector and the preset standard defect feature vector to obtain the image similarity coefficient. The probability weighted fusion module is used to weight and correct the image similarity coefficient according to the environmental confidence index, and generate a fused defect probability value that characterizes the possibility of aging defects in the power distribution arrester. The defect identification and judgment module is used to generate the final aging defect identification result of the distribution surge arrester based on the fused defect probability value.

9. An electronic device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the method for identifying aging defects in distribution surge arresters that takes into account environmental confidence as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the storage medium is located to perform the method for identifying aging defects in distribution surge arresters that considers environmental confidence as described in any one of claims 1 to 7.

Images