Image monitoring analysis method and system for laser cleaning of disconnector contacts

Abstract

The application provides an image monitoring analysis method and system for laser cleaning of isolator contacts, and belongs to the technical field of monitoring analysis, and specifically comprises: determining attention image features in image features according to abnormal changes in operating temperatures of the isolator contacts in different associated load intervals; determining the image monitoring analysis method in the cleaning process of the isolator contacts in the associated load interval according to the update processing result of the attention image features and in combination with the running time of the isolator contacts corresponding to different attention image features; determining the stop control mode of the isolator contacts in the cleaning process based on the image monitoring analysis method; and determining the verification control target of the attention image features based on the stop processing data in different isolator contacts, thereby improving the reliability of the cleaning process.

Description

Technical Field

[0001] This invention belongs to the field of monitoring and analysis technology, and in particular relates to an image monitoring and analysis method and system for laser cleaning of disconnector switch contacts. Background Technology

[0002] During long-term operation, the contact of a disconnector switch will form a dirt layer on its surface due to oxidation, corrosion, arc erosion, dust accumulation, etc., which will increase the contact resistance and aggravate the heat generation. In severe cases, it may cause equipment failure or power outage.

[0003] To address the aforementioned issues, laser cleaning, as a non-contact, high-precision, and environmentally friendly cleaning technology, has been increasingly applied to the cleaning of contacts in power equipment in recent years. However, it still faces the following technical challenges: When using laser cleaning equipment for cleaning, there may be abnormalities after cleaning, which may lead to temperature anomalies after a short period of operation. Therefore, how to extract and verify the image features that may cause temperature anomalies based on the temperature anomalies during operation, so as to ensure the reliability of the cleaning process, has become an urgent technical problem to be solved.

[0004] Specifically, this application provides an image monitoring and analysis method and system for laser cleaning of disconnector switch contacts. Summary of the Invention

[0005] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides an image monitoring and analysis method for laser cleaning of disconnector switch contacts, which includes: S1 uses the operating data of the disconnector contact after cleaning to determine the abnormal changes in the operating temperature of the disconnector contact, and combines the abnormal correlation time range of the operating temperature of different disconnector contacts to determine the image feature extraction and processing strategy of the disconnector contact during the laser cleaning process. S2 determines the load range for extracting and processing the image features of the disconnector switch contacts based on the extraction and processing strategy, takes the load range for extracting and processing the image features of the disconnector switch contacts as the associated load range, and determines the image features of interest in the image features based on the abnormal changes in the operating temperature of the disconnector switch contacts in each associated load range. S3 determines the image monitoring and analysis method for the cleaning process of the disconnector contacts within the associated load range based on the updated processing results of the image features of interest and the running time of the disconnector contacts corresponding to different image features of interest. S4 determines the stop control method for the disconnector contacts during the cleaning process based on the image monitoring and analysis method, and determines the verification and control target of the image features of interest based on the stop processing data of different disconnector contacts.

[0006] The beneficial effects of this invention are as follows: By analyzing the abnormal temperature fluctuations of the disconnector switch contacts, the reliability of the cleaning process based on the laser cleaning device is determined. Based on the reliability of the cleaning process based on the laser cleaning device, a strategy for extracting and processing image features of the disconnector switch contacts during the laser cleaning process is determined, thus laying the foundation for reliable identification and processing of image features when the cleaning process is terminated.

[0007] By using the interruption processing data in different disconnector switch contacts, the impact of the verification processing under the current monitoring and processing method is determined. Based on the impact of the verification processing under the current monitoring and processing method, the verification control targets of the image features of concern are determined. This ensures the reliability of the verification processing of different image features of concern while also reducing the impact on different disconnector switch contacts during the cleaning process.

[0008] Furthermore, the operating data of the disconnecting switch contacts after cleaning includes the operating temperature of the disconnecting switch contacts during the operation process after cleaning.

[0009] Furthermore, the abnormal fluctuations in the operating temperature of the disconnector contact include the abnormal time periods of the operating temperature of the disconnector contact and the cleaning duration range during which the abnormal time periods occur.

[0010] Furthermore, the abnormal operating temperature correlation duration range of the disconnecting switch contact is the duration range of the first abnormal operating temperature of the disconnecting switch contact.

[0011] Furthermore, the method for determining the image feature extraction and processing strategy of the disconnecting switch contacts during the laser cleaning process is as follows: S11 uses the abnormal fluctuations in the operating temperature of the disconnecting switch contacts to identify the disconnecting switch contacts that have abnormal operating temperatures during certain periods, and identifies them as abnormal switch contacts. S12 determines, based on the abnormal association duration interval of the abnormal switch contact, the abnormal switch contact that belongs to the abnormal association duration interval in different duration intervals, and regards it as the associated abnormal contact; S13 utilizes the associated abnormal contacts within different time intervals to determine the image feature extraction and processing strategy for the disconnecting switch contacts during the laser cleaning process.

[0012] Furthermore, the method for determining the image monitoring and analysis method during the cleaning process of the disconnector contacts within the associated load range is as follows: S31 determines the number of features of interest based on the update processing result of the features of interest in the image; S32 takes the isolation switch contact corresponding to the image feature of interest as the associated contact, and takes the associated contact whose running length is greater than the maximum value of the endpoint of the filtering time interval as the filtering associated contact of the image feature of interest. S33 uses the number of image features of interest, the distribution data of associated contacts with different image features of interest, and the abnormal time periods of screening associated contacts to determine the image monitoring and analysis method for the cleaning process of disconnector contacts within the associated load range.

[0013] In a second aspect, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described image monitoring and analysis method for laser cleaning of disconnector contacts when running the computer program.

[0014] Other features and advantages will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0015] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0016] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart of an image monitoring and analysis method for laser cleaning of disconnector switch contacts; Figure 2 This is a flowchart illustrating the method for determining the image feature extraction and processing strategy for disconnector switch contacts during laser cleaning. Figure 3 This is a flowchart of a method for determining image features of interest in image features. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that the invention will be thorough and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore their detailed description will be omitted.

[0019] The terms “a,” “one,” “the,” and “the” are used to indicate the existence of one or more elements / components / etc.; the terms “including” and “having” are used to indicate an open-ended meaning of inclusion and that other elements / components / etc. may exist in addition to the listed elements / components / etc.

[0020] Example 1 To solve the above problems, according to one aspect of the present invention, such as Figure 1 As shown, an image monitoring and analysis method for laser cleaning of disconnector switch contacts is provided, specifically including: S1 uses the operating data of the disconnector contact after cleaning to determine the abnormal changes in the operating temperature of the disconnector contact, and combines the abnormal correlation time range of the operating temperature of different disconnector contacts to determine the image feature extraction and processing strategy of the disconnector contact during the laser cleaning process. S2 determines the load range for extracting and processing the image features of the disconnector switch contacts based on the extraction and processing strategy, takes the load range for extracting and processing the image features of the disconnector switch contacts as the associated load range, and determines the image features of interest in the image features based on the abnormal changes in the operating temperature of the disconnector switch contacts in each associated load range. S3 determines the image monitoring and analysis method for the cleaning process of the disconnector contacts within the associated load range based on the updated processing results of the image features of interest and the running time of the disconnector contacts corresponding to different image features of interest. S4 determines the stop control method for the disconnector contacts during the cleaning process based on the image monitoring and analysis method, and determines the verification and control target of the image features of interest based on the stop processing data of different disconnector contacts.

[0021] Furthermore, the operating data of the disconnecting switch contacts after cleaning includes the operating temperature of the disconnecting switch contacts during the operation process after cleaning.

[0022] Specifically, the abnormal fluctuations in the operating temperature of the disconnector switch contacts include the abnormal periods in the operating temperature of the disconnector switch contacts and the cleaning duration range during which the abnormal periods occur.

[0023] It should be noted that the abnormal operating temperature correlation interval of the disconnecting switch contact is the interval between the first occurrence of an abnormal operating temperature of the disconnecting switch contact.

[0024] Specifically, such as Figure 2 As shown, the method for determining the image feature extraction and processing strategy of the disconnecting switch contacts during laser cleaning is as follows: In this embodiment, the reliability of the cleaning process based on the abnormal temperature fluctuation of the disconnector contact is determined by the laser cleaning device. Based on the reliability of the cleaning process based on the laser cleaning device, the strategy for extracting and processing the image features of the disconnector contact during the laser cleaning process is determined, thereby laying the foundation for the reliable identification and processing of image features of the cleaning process that has been terminated.

[0025] S11 uses the abnormal fluctuations in the operating temperature of the disconnecting switch contacts to identify the disconnecting switch contacts that have abnormal operating temperatures during certain periods, and identifies them as abnormal switch contacts. The abnormal switch contact refers to the disconnect switch contact that, during operation after cleaning, experienced an abnormal period in which its operating temperature exceeded the normal temperature range; that is, the contact where abnormal temperature fluctuations existed.

[0026] Suppose a batch of disconnector contacts are put into operation after cleaning. Check the temperature curve of each contact during the operation period. If the operating temperature of a contact exceeds the normal fluctuation range of the normal operating temperature during a certain period, then the contact is identified as an abnormal switch contact.

[0027] This step involves screening abnormal switch contacts during periods of abnormal temperature. Its significance lies in quickly identifying a set of contacts that may have cleaning quality issues. This provides a streamlined set of analysis objects for subsequent classification of associated abnormal contacts and determination of image feature extraction strategies, thereby improving analysis efficiency.

[0028] S12 determines, based on the abnormal association duration interval of the abnormal switch contact, the abnormal switch contact that belongs to the abnormal association duration interval in different duration intervals, and regards it as the associated abnormal contact; The abnormal association duration interval refers to the cleaning duration interval when the abnormal switch contact first shows an abnormal operating temperature; the associated abnormal contact refers to the abnormal switch contact after being classified according to the abnormal association duration interval within each duration interval.

[0029] Assuming the cleaning time of the abnormal switch contacts is divided into multiple time intervals, the abnormal switches are classified according to the cleaning time interval in which the operating temperature of each abnormal switch contact first becomes abnormal, and the abnormal contacts in the same time interval are grouped into a group of related abnormal contacts.

[0030] This step involves classifying and grouping the abnormal switch contacts according to the abnormal associated duration interval. Its significance lies in transforming the time-series information of abnormal operating temperature into structured classification features in the duration dimension, providing data support for subsequent differentiated processing based on the time interval to determine image feature extraction strategies.

[0031] S13 utilizes the associated abnormal contacts within different time intervals to determine the image feature extraction and processing strategy for the disconnecting switch contacts during the laser cleaning process.

[0032] Assuming that the classification of associated abnormal contacts for each time interval has been completed, based on the distribution of the number of associated abnormal contacts in different time intervals, determine which load intervals' disconnector contacts will be processed for image feature extraction, as well as the scope and conditions of the extraction process.

[0033] This step utilizes the correlation of abnormal contact distribution over a time interval to determine the extraction and processing strategy. Its significance lies in transforming the temporal distribution information of abnormal operating temperature into a basis for selecting image feature extraction strategies, thereby enabling the adoption of differentiated extraction strategies under different cleaning reliability scenarios and achieving reasonable allocation of monitoring resources.

[0034] Specifically, by utilizing the associated abnormal contacts within different time intervals, a strategy for extracting and processing image features of the disconnector switch contacts during laser cleaning is determined, including: By identifying the associated abnormal contacts within different time intervals, the time intervals that do not meet the requirements are determined and used as the filtering time intervals; It should be noted that the duration intervals that do not meet the requirements are duration intervals with short intervals, which can be determined by using a threshold method.

[0035] The screening time interval refers to a time interval with a short interval duration, that is, a time interval where the interval between the endpoints of the time interval does not meet the preset interval duration requirement. It is used to screen out the set of contacts that have abnormal operating temperature within the short cleaning time interval. The interval duration refers to the difference between the upper and lower limits of the time interval. The shorter the interval duration, the more concentrated the cleaning time corresponding to the interval is. When the abnormal operating temperature occurs within the short time interval, it indicates that the degree of abnormality in the cleaning is higher.

[0036] Assuming the cleaning time is divided into multiple time intervals with different interval durations, if the interval duration of a certain interval is shorter than a preset interval duration threshold, then that interval is determined as the screening time interval, and the associated abnormal contacts within that interval are entered into subsequent analysis.

[0037] This step filters out short cleaning intervals as the focus of analysis by using time intervals. The significance of this step is that abnormal operating temperatures that occur within short intervals are often closely related to the degree of abnormality in cleaning parameter settings. This allows for a more accurate identification of the correlation between cleaning parameters and cleaning quality, thereby improving the targeting of image feature extraction.

[0038] It should be noted that the requirement that the number of associated abnormal contacts within the screening time interval does not meet the requirements means that the number exceeds the preset number threshold, i.e., when the number of associated abnormal contacts is large and the degree of abnormality of laser cleaning is relatively serious.

[0039] Case 1: If the number of associated abnormal contacts within the screening time interval does not meet the requirements, and the number of associated abnormal contacts within the screening time interval is large, the degree of abnormality in laser cleaning is more serious. In this case, the image feature extraction and processing strategy for the isolating switch contacts during the laser cleaning process is determined to be to extract and process the image features of all isolating switch contacts within the load interval. Case 1 refers to the situation where the number of associated abnormal contacts is large and the degree of abnormality in laser cleaning is severe when determining the extraction and processing strategy. In this case, a full coverage strategy is adopted to extract and process the image features of the disconnector contacts in all load ranges.

[0040] If the number of associated abnormal contacts within the screening time interval exceeds the preset threshold, it indicates that the abnormal operating temperature is relatively concentrated and severe within the short cleaning interval, and the cleaning reliability of the laser cleaning device within this interval is low. Therefore, image feature extraction processing is performed on the disconnector contacts in all load intervals to fully cover any possible cleaning quality problems.

[0041] This step employs a full-coverage extraction strategy when the laser cleaning anomalies are severe. The significance of this strategy is that when there are a large number of associated abnormal contacts within the screening time interval, it indicates that the overall reliability of the cleaning process is low. Extracting only a portion of the interval may lead to the omission of cleaning quality issues. Therefore, the most comprehensive extraction strategy is adopted to ensure that all potential problems are identified, thereby avoiding cleaning quality hazards caused by insufficient extraction range.

[0042] Additionally, it should be noted that in Case 2: if the number of associated abnormal contacts within the screening time interval meets the requirements, the cleaning abnormality coefficient of the disconnector contacts within the load interval is determined based on the proportion of disconnector contacts within the load interval to the number within the screening time interval. It is then determined whether there are load intervals where the cleaning abnormality coefficient does not meet the requirements. If so, the image feature extraction and processing strategy for the disconnector contacts during the laser cleaning process is to perform image feature extraction and processing on all disconnector contacts within the load interval where associated abnormal contacts exist within the screening time interval. If not, then all disconnector contacts within the load interval where the cleaning abnormality coefficient is above the preset abnormality coefficient threshold are subjected to image feature extraction and processing.

[0043] Case 2 refers to the situation where the number of associated abnormal contacts within the screening time interval meets the requirements (i.e., the number does not exceed the threshold) and the degree of laser cleaning abnormality is not serious. In this case, it is necessary to first calculate the cleaning abnormality coefficient of each load interval, and then determine the extraction strategy according to whether there are load intervals where the cleaning abnormality coefficient does not meet the requirements.

[0044] The cleaning anomaly coefficient refers to the proportion of the number of associated abnormal contacts of the disconnector switch contacts within the load range during the screening time interval to the total number of contacts within that load range. It is used to quantify the degree of concentration of cleaning anomalies within that load range.

[0045] Assuming the number of associated abnormal contacts within the screening time interval does not exceed the threshold, the ratio of the number of associated abnormal contacts to the total number of contacts in each load interval is calculated as the cleaning abnormality coefficient. If there are load intervals where the cleaning abnormality coefficient does not meet the requirements, all contacts in the load intervals containing associated abnormal contacts within the screening time interval are extracted. If there are no such load intervals, contacts in the load intervals where the cleaning abnormality coefficient is above the preset abnormality coefficient threshold are extracted.

[0046] This step further refines the extraction strategy by using the cleaning anomaly coefficient when the degree of laser cleaning anomaly is not severe. Its significance lies in identifying the load range where local anomalies are relatively concentrated by analyzing the anomaly coefficient at the load range dimension when the overall degree of anomaly is not severe. This ensures anomaly coverage while avoiding over-extraction, thereby achieving a refined extraction and processing strategy and saving resources.

[0047] The cleaning anomaly coefficient not meeting the requirements means that the cleaning anomaly coefficient is lower than the preset anomaly coefficient threshold, indicating that the degree of cleaning anomaly in this load range is relatively mild.

[0048] Specifically, such as Figure 3 As shown, the method for determining the image features of interest in the image features is as follows: In this embodiment, based on the number of associated load intervals and the abnormal temperature fluctuations of the disconnector contacts within the associated load intervals, the reliability of identifying potential cleaning anomalies in the image features of interest and the identification processing requirements are determined. Based on the reliability of identifying potential cleaning anomalies in the image features of interest and the identification processing requirements, the image features of interest are identified, thus laying the foundation for stopping the disconnector contacts during the cleaning process based on the image features of interest and realizing the identification processing of image features of potential cleaning anomalies.

[0049] The image features of interest refer to visual image features that are associated with cleaning anomalies during the cleaning process of disconnector switch contacts and that need to be focused on in subsequent monitoring and analysis, such as the distribution pattern of laser action marks on the contact surface and the texture features of residual oxide layer areas; the recognition reliability refers to the credibility of recognizing image features of interest based on associated load interval data, which is related to the number of associated load intervals and the abnormal changes.

[0050] If the number of associated load intervals is small, it means that the identification of image features of interest needs to be performed in a small number of load intervals. The identification reliability is high and the processing requirements are low, so a more lenient determination condition is adopted. If the number of associated load intervals is large, the identification reliability is relatively low and the processing requirements are high, so a more stringent determination condition is adopted.

[0051] This step determines the features of the image of interest by associating the number of load intervals and abnormal changes. Its significance lies in transforming the extraction and processing strategy into a specific set of features of the image of interest, providing a feature-level basis for determining subsequent image monitoring and analysis methods.

[0052] S21 uses the associated load interval data to determine the number of associated load intervals; The number of associated load intervals refers to the number of load intervals that need to be processed for image feature extraction, as determined by the processing strategy in S1.

[0053] Assuming that the extraction and processing strategy of S1 is determined to be to perform image feature extraction processing on the disconnector contacts in the load intervals L1 and L2, then the number of associated load intervals is 2.

[0054] This step, by counting the number of associated load intervals, is significant in that it provides a basis for determining the number of intervals in the subsequent determination of features of the image of interest, thereby dynamically adjusting the strictness of the determination of features of the image of interest based on the number of associated load intervals.

[0055] In the above steps, if the number of associated load intervals is less than the preset threshold for the number of associated load intervals, the impact of cleaning and limiting processing based on the image features of interest is relatively small. Therefore, the image features of interest are determined as follows: if the number of isolating switch contacts with the image features of interest is above the first threshold after cleaning is completed in all the load intervals of interest, then the image features of interest are determined to be image features of interest.

[0056] The preset threshold for the number of associated load intervals refers to a critical value for determining whether the number of associated load intervals is too small; the first threshold refers to a critical value for determining whether a certain image feature needs to reach the minimum occurrence frequency requirement in all associated load intervals.

[0057] If the number of associated load intervals is less than the preset threshold for the number of associated load intervals, it indicates that there are fewer load intervals that need to be focused on. In this case, a more lenient determination condition is adopted: as long as the number of disconnector contacts present after a certain image feature has been cleaned in all associated load intervals reaches more than the first threshold, the image feature is determined to be an image feature of interest.

[0058] This step uses lenient determination conditions when the number of associated load intervals is small. Its significance is that when the scope of concern is small, a lower standard can be used to screen out image features related to cleaning anomalies, avoiding the omission of potential risk features due to excessively high standards.

[0059] It is also understood that if the number of associated load intervals is not less than the preset threshold for the number of associated load intervals, then proceed to step S22; If the number of associated load intervals is not less than the preset threshold for the number of associated load intervals, it indicates that there are many load intervals that need attention and the processing requirements are high. Then, proceed to S22 to calculate the cleaning anomaly coefficient and perform further analysis.

[0060] This step proceeds to further analysis when there are a large number of associated load intervals. Its significance lies in the fact that when the scope of concern is large, it is necessary to further refine the degree of anomaly in each interval by cleaning the anomaly coefficient, thereby achieving a refined determination of the features of the image of concern.

[0061] S22 determines the cleaning anomaly coefficient for different associated load ranges based on the abnormal temperature fluctuations of the disconnector contacts in different associated load ranges. The cleaning anomaly coefficient refers to the proportion of disconnector contacts with abnormal operating temperatures within a certain associated load range to the total number of contacts within that range, and is used to quantify the degree of concentration of cleaning anomalies within that load range.

[0062] Suppose there are multiple disconnector contacts within a certain associated load range, and some of these contacts exhibit abnormal temperature fluctuations after cleaning. The ratio of the number of abnormal contacts to the total number of contacts within the range is the cleaning abnormality coefficient for that associated load range.

[0063] This step calculates the cleaning anomaly coefficient for each associated load interval. Its significance lies in quantifying the degree of cleaning quality anomaly in each interval into a comparable coefficient value, providing a quantitative basis for determining the degree of interval anomaly in subsequent image features of interest.

[0064] The above steps include the following: Using the cleaning anomaly coefficients of different associated load intervals, determine whether there are associated load intervals where the cleaning anomaly coefficient does not meet the requirements. If so, proceed to step S23. If not, determine the image features of interest in the image features. If the number of isolating switch contacts with the image features after cleaning is completed in all the load intervals of interest is above the second threshold (greater than the first threshold), then determine that the image features belong to the image features of interest. The cleaning anomaly coefficient not meeting the requirements means that the cleaning anomaly coefficient is greater than the preset anomaly coefficient threshold, indicating that the cleaning anomaly degree of the associated load interval is high; the second threshold is a critical value for judging whether a certain image feature needs to reach a higher occurrence frequency requirement in all associated load intervals, and the second threshold is greater than the first threshold.

[0065] Assuming that the cleaning anomaly coefficients for all associated load intervals have been calculated, if there are associated load intervals whose cleaning anomaly coefficients do not meet the requirements, i.e., associated load intervals with a high degree of anomaly, then proceed to S23 for more detailed analysis; if the cleaning anomaly coefficients for all associated load intervals meet the requirements, then the degree of cleaning anomaly is low, and a strict standard is adopted: only when the number of disconnector contacts present after a certain image feature has been cleaned in all associated load intervals reaches the second threshold or above, is the image feature determined to be an image feature of interest.

[0066] This step determines the features of interest based on whether there are associated load intervals where the cleaning anomaly coefficient does not meet the requirements. The significance of this step is that when the degree of cleaning anomaly in all associated load intervals is low, a more stringent standard is adopted, thereby ensuring that the features in the set of interest are highly correlated with the cleaning anomaly.

[0067] S23 uses the number of associated load intervals and the cleaning anomaly coefficients of different associated load intervals to determine the image features of interest in the image features.

[0068] Assuming that the cleaning anomaly coefficient and the number of associated load intervals have been determined, different strategies for determining the features of interest are adopted based on the judgment results of S231 and S232.

[0069] This step combines the number of associated load intervals and the cleaning anomaly coefficient to determine the features of the image of interest. Its significance lies in enabling adaptive adjustment of the strategy for determining the features of the image of interest.

[0070] Furthermore, by utilizing the number of associated load intervals and the cleaning anomaly coefficients of different associated load intervals, the image features of interest in the image features are determined, specifically including: S231, determine whether the average value of the cleaning anomaly coefficient of different associated load intervals is greater than the preset value of the anomaly coefficient. If so, determine the image feature of interest in the image features. If the number of isolating switch contacts of the image feature is above the first threshold after cleaning is completed in all the load intervals of interest, then determine that the image feature belongs to the image feature of interest. If not, proceed to step S232. The preset value of the abnormality coefficient refers to the critical value for judging whether the average value of the cleaning abnormality coefficient of each associated load interval is high.

[0071] If the average value of the cleaning anomaly coefficient in each associated load interval is greater than the preset value of the anomaly coefficient, it indicates that the overall cleaning anomaly level is high. In this case, the determination condition corresponding to the first threshold is adopted.

[0072] This step uses the determination conditions corresponding to the first threshold when the overall cleaning anomaly level is high. The significance of this step is that when the overall cleaning anomaly level is high, even if a relatively lenient standard is used, the determined features of the image of interest are still highly correlated with the cleaning anomaly.

[0073] S232 determines the control impact value based on the cleaning anomaly coefficient of different associated load intervals and the number of associated load intervals, and determines whether the control impact value is greater than a preset impact threshold. If so, the image feature of interest in the image features is determined to be a feature of interest. If the number of disconnector contacts of the image feature after cleaning is completed in all the load intervals of interest is greater than the second threshold (greater than the first threshold), the image feature is determined to be a feature of interest. If not, the image feature of interest in the image features is determined to be a feature of interest if the number of disconnector contacts of the image feature after cleaning is completed in all the load intervals of interest is greater than the second threshold, or if the number of disconnector contacts of the image feature after cleaning is completed in the load intervals of the preset number of load intervals of interest is greater than the first threshold.

[0074] Furthermore, the lower the cleaning anomaly coefficient of different associated load intervals and the greater the number of associated load intervals, the greater the control impact value.

[0075] The control impact value refers to a quantitative indicator that reflects the degree of influence of the features of the image of concern on the control of cleaning, calculated by comprehensively considering the number of associated load intervals and the cleaning anomaly coefficient.

[0076] The calculation method for the control impact value is as follows: Let N be the number of associated load intervals, and let Cavg be the average cleaning anomaly coefficient for different associated load intervals. Then: Control impact value = N × (1 - Cavg).

[0077] If the number of associated load intervals is large and the average value of the cleaning anomaly coefficient is low, the control impact value is large, indicating that the determination of the image features of concern has a high degree of influence on cleaning control, and a more stringent second threshold condition needs to be adopted; if the control impact value is small, the union condition is adopted, that is, the first threshold can be reached above the second threshold or within the number of concerned load intervals of the preset interval.

[0078] This step further refines the strategy for determining the features of the image under control by controlling the impact value. Its significance lies in the fact that when the average value of the cleaning anomaly coefficient is not greater than the preset value of the anomaly coefficient, different determination conditions are adopted according to the magnitude of the control impact value, thereby achieving a refined classification of the strategy.

[0079] This embodiment, through steps S21 to S23, realizes a complete process from counting the number of associated load intervals to calculating the cleaning anomaly coefficient and then determining the features of the image of interest. Its core value is reflected in three aspects: First, it dynamically adjusts the strictness of determining the features of the image of interest by the number of associated load intervals, and adopts lenient conditions to ensure the integrity of feature coverage when the number of intervals is small; second, it achieves refined classification of the strategy by using differentiated determination conditions when the overall anomaly degree is different through the graded judgment of the cleaning anomaly coefficient and the control impact value; and third, it achieves an adaptive balance between the accuracy and coverage of the feature recognition of the image of interest by flexibly applying the first threshold, the second threshold and the union condition.

[0080] Continuing with the calculation results of S1, the associated load intervals are L1 and L2, a total of 2 load intervals, each with 40 disconnector contacts. The cleaning anomaly coefficient for interval L1 is 0.050, and the cleaning anomaly coefficient for interval L2 is 0.025.

[0081] Step S21: The number of associated load intervals is 2.

[0082] Determine whether the number of associated load intervals 2 is less than the preset threshold number of associated load intervals 3: 2 is less than 3, which satisfies the "less than" condition. Therefore, the first threshold is used to determine the condition: if the number of isolating switch contacts with the image features is above the first threshold (3) after cleaning in all associated load intervals, then the image features are determined to be image features of interest. Therefore, the image features of interest are F1 and F3, a total of 2.

[0083] Specifically, the method for determining the image monitoring and analysis method during the cleaning process of the disconnector contacts within the associated load range is as follows: In this embodiment, the degree of influence of the image features on the cleaning process of the disconnector contacts is determined based on the number of image features of interest. Combined with the running time of the disconnector contacts corresponding to different image features of interest, the verification processing requirements for different image features of interest are determined. Based on the verification processing requirements for different image features of interest and the degree of influence of the image features on the cleaning process of the disconnector contacts, the image monitoring and analysis method for the cleaning process of the disconnector contacts within the associated load range is determined, i.e., how to stop the cleaning process to achieve verification processing of the image features of interest, thereby ensuring the reliability and efficiency of the verification and identification processing of the image features of interest.

[0084] S31 determines the number of features of interest based on the update processing result of the features of interest in the image; The number of image features of interest refers to the total number of currently determined image features of interest.

[0085] Assuming that S2 determines the image features of interest to be F1 and F3, and no changes occur after the update process, then the number of image features of interest is 2.

[0086] This step involves counting the number of image features of interest, which serves as a basis for determining whether to adopt a complex termination decision rule.

[0087] It should be noted that if the number of the image features of interest meets the requirements, the cleaning process will stop when the similarity between the disconnector contact and the image features of interest meets the requirements during the cleaning process.

[0088] Specifically, if the number of image features of interest is small, even if the cleaning process is stopped in time, it will not affect the contact of the disconnector switch. Therefore, the number of image features of interest is determined to meet the requirements, which is specifically determined by a threshold.

[0089] The number of image features of interest meets the requirement that the number is relatively small and does not exceed the preset threshold.

[0090] Assuming the number of image features of interest is small, a simple termination strategy is adopted: as long as the real-time image of the disconnector contact is detected to be similar to any image feature of interest during the cleaning process, the cleaning process is stopped immediately.

[0091] This step employs a simple termination strategy when the number of image features is small. The significance of this strategy is that when the number of features is small, any match of any feature has high risk warning value, and timely cessation of cleaning can prevent the further development of cleaning anomalies.

[0092] It should be noted that when the similarity coefficient between the disconnector contact and the feature of the image of interest is greater than a preset similarity coefficient threshold, the similarity between the disconnector contact and the feature of the image of interest is determined to meet the requirements.

[0093] The preset similarity coefficient threshold is determined as follows: its core determining factor is the similarity coefficient value corresponding to the endpoint of the sharply rising segment of the relationship curve between image feature similarity coefficient and the accuracy of cleaning anomaly detection. When the similarity coefficient exceeds this value, the improvement in the accuracy of cleaning anomaly detection tends to level off, and this value is the critical point for distinguishing between "low-confidence matching" and "high-confidence matching". Auxiliary reference factors include the type discrimination of image features (the higher the discrimination of feature types, the greater the difference in similarity coefficient between normal and abnormal features, and the threshold can be appropriately increased) and the signal-to-noise ratio of image acquisition during laser cleaning (the lower the signal-to-noise ratio, the greater the estimation error of the similarity coefficient, and the threshold needs to be appropriately reduced). If the threshold is too high, the similarity coefficient will be difficult to reach the threshold, and the abort decision will be frequently not triggered, which may lead to the cleaning anomaly problem not being detected in time; if the threshold is too low, the normal cleaning process will be misjudged as needing to be stopped, resulting in unnecessary cleaning interruption. Therefore, the similarity coefficient value corresponding to the endpoint of the sharply rising segment of the relationship curve is taken as the preset similarity coefficient threshold.

[0094] It is also understood that if the number of image features of interest does not meet the requirements, the process proceeds to step S32.

[0095] If the number of image features of interest is too large, exceeding the preset threshold, a complex termination decision rule is adopted, and the process proceeds to S32 for further analysis.

[0096] This step transitions to complex decision rules when focusing on a large number of image features. The significance of this is that when the number of features is large, the strategy of stopping the process as soon as any feature is matched may lead to frequent and unnecessary stoppages. Therefore, it is necessary to use more complex rules to integrate the matching of multiple features.

[0097] S32 takes the isolation switch contact corresponding to the image feature of interest as the associated contact, and takes the associated contact whose running length is greater than the maximum value of the endpoint of the filtering time interval as the filtering associated contact of the image feature of interest. The associated contact refers to the disconnector contact corresponding to the feature of the image of interest, that is, the contact is found to have the feature of the image of interest in the analysis of S2; the screened associated contact refers to the contact among the associated contacts whose running length is greater than the maximum value of the endpoint of the screening time interval, that is, the contact with a sufficiently long running length and more reliable information on the correlation between abnormal running temperature and cleaning quality.

[0098] Assuming the maximum value of the endpoint of the filtering duration interval is 180 days (the upper limit of the T1 interval), then contacts with a running duration exceeding 180 days among the associated contacts will be filtered as associated contacts.

[0099] This step filters out highly reliable associated contacts based on runtime. The significance of this step is that the correlation between abnormal operating temperature and cleaning quality is more stable and reliable for contacts with longer runtime. Using the data of such contacts as a basis to determine subsequent stop decision rules can improve the reliability of the decision rules.

[0100] It is understood that, based on the associated contact data of the image features of interest, the number of associated contacts of the image features of interest is determined, and it is determined whether the number of associated contacts of the image features of interest is greater than a preset threshold for the number of associated contacts. If yes, then proceed to step S33; if no, then when it is determined that the similarity between the isolating switch contact and the image features of interest meets the requirements during the cleaning process, the cleaning process is stopped.

[0101] The preset threshold for the number of associated contacts refers to the critical value used to determine whether the number of associated contacts is sufficient to support complex termination decision rules.

[0102] If the number of associated contacts for a certain image feature of interest is large, exceeding the preset threshold for the number of associated contacts, then proceed to S33 for more refined analysis; if the number is small, then a simple termination strategy is adopted.

[0103] This step determines whether to enter a complex decision based on the number of associated contacts. Its significance lies in using refined termination decision rules to improve decision accuracy when the number of associated contacts is sufficient, and using simple strategies to ensure decision reliability when the number is insufficient.

[0104] S33 uses the number of image features of interest, the distribution data of associated contacts with different image features of interest, and the abnormal time periods of screening associated contacts to determine the image monitoring and analysis method for the cleaning process of disconnector contacts within the associated load range.

[0105] Assuming that the abnormal time period distribution data of the associated contacts and the screening associated contacts for each image feature of interest have been determined, the termination decision rules corresponding to each image feature of interest are determined according to the judgment results of S331 and S332.

[0106] This step comprehensively considers the number of image features and the distribution data of abnormal time periods to determine the image monitoring and analysis method. Its significance lies in realizing the adaptive adjustment of the termination decision rules.

[0107] It is understandable that the above steps include the following: S331, when the similarity between the disconnector contact and the image feature of interest meets the requirements during the cleaning process, the image feature of interest for which the cleaning process is stopped is taken as the image feature of the cleaning process. It is determined whether the proportion of the image feature of the cleaning process in all the image features of interest is above the target proportion threshold. If so, it is determined that the cleaning process will not be stopped even if the similarity between the disconnector contact and the image feature of interest meets the requirements during the cleaning process. If not, proceed to step S332. The image features to be cleaned refer to the image features of interest that adopt a simple termination strategy (i.e., stop cleaning when the similarity requirement is met); the target proportion threshold refers to the critical value for judging whether the proportion of image features to be cleaned is high enough.

[0108] If the proportion of cleaned image features among all features of interest reaches the target proportion threshold, it indicates that most features of interest have adopted a simple termination strategy. Even if the remaining few features match, the cleaning will not stop, so as to avoid excessive termination and affect the cleaning efficiency. If the proportion does not reach the target proportion threshold, then proceed to S332 for more refined analysis.

[0109] This step determines whether to adopt a non-stop strategy for the remaining features by measuring the proportion of image features processed by the cleaning process. Its significance lies in the fact that when most features have already adopted a simple stop strategy, adopting a non-stop strategy for the remaining few features can reduce unnecessary stoppages while ensuring overall risk coverage.

[0110] S332, based on the number of associated contacts of the image features of interest and the proportion of associated contacts with abnormal time periods in the filtered associated contacts, a verification requirement value is determined. It is then determined whether the verification requirement value is greater than a preset requirement threshold. If not, it is determined that even if the similarity between the disconnector contact and the image features of interest meets the requirements during the cleaning process, the cleaning process will not be stopped for the remaining image features of interest. If yes, it is determined that the similarity between the disconnector contact and the image features of interest meets the requirements during the cleaning process, and the number of times the cleaning process has been stopped during the cleaning process in history is less than a preset number threshold, then the cleaning process is stopped.

[0111] It should be noted that the fewer the number of associated contacts with the image features being monitored, and the lower the proportion of associated contacts with abnormal time periods among the filtered associated contacts, the higher the verification requirement value.

[0112] The verification requirement value refers to an indicator that quantifies the urgency of further verification processing of the image features being monitored. The fewer the number of associated contacts and the lower the proportion of abnormal time periods among the screened associated contacts, the less sufficient the evidence of the association between the feature and the cleaning anomaly is. Therefore, a higher verification requirement value is needed to trigger more stringent verification processing.

[0113] The verification requirement value is calculated as follows: Let Nc be the number of associated contacts for the image features of interest, and Rab be the proportion of associated contacts with abnormal time periods among the filtered associated contacts. Then: the verification requirement value = 1 / ((1+Nc / N1) × (1+Rab / R1)), where N1 and R1 are the preset values ​​of unit normalization processing, and their values ​​are 1.

[0114] If the verification requirement value of a certain image feature of interest is greater than the preset requirement threshold, it indicates that the evidence of association between the feature and the cleaning anomaly is not sufficient, and the cleaning process needs to be stopped only when the historical number of stops is added during the matching process; if the verification requirement value is not greater than the preset requirement threshold, the cleaning will not be stopped when the feature is matched.

[0115] This step further refines the termination decision rules by verifying the required values. Its significance lies in determining the termination conditions based on the verification required values ​​of each image feature when the proportion of cleaned image features is insufficient, thereby achieving a fine balance between verification reliability and cleaning efficiency.

[0116] This embodiment, through steps S31 to S33, realizes a complete process from counting the number of image features to screening associated contacts and determining the image monitoring and analysis method. Its core value is reflected in three aspects: First, by dynamically adjusting the complexity of the termination decision rule based on the number of image features, a simple strategy is adopted to ensure efficiency when the number is small; second, by determining whether to adopt a non-termination strategy for the remaining features based on the proportion of cleaned image features, unnecessary terminations are reduced while ensuring overall risk coverage; and third, by using a hierarchical judgment of verification requirements, a fine balance is achieved between verification reliability and cleaning efficiency.

[0117] Continuing with the calculation results of S2, we focus on two image features: F1 and F3.

[0118] The preset quantity threshold is set to 1, the preset similarity coefficient threshold is set to 0.85, the preset associated contact number threshold is set to 5, the target proportion threshold is set to 0.6, the preset demand threshold is set to 0.3, and the preset number of times threshold is set to 3.

[0119] Step S31: Focus on 2 image features.

[0120] Determine whether the number of image features of interest, 2, meets the requirement (i.e., the number is small and does not exceed the preset number threshold 1): 2 is greater than 1, does not meet the requirement (the number is not small), and proceed to S32.

[0121] Step S32: Select the disconnector contacts corresponding to the image features F1 and F3 as associated contacts, and select the associated contacts whose running length is greater than the maximum value of the endpoint of the filtering time interval as the filtering associated contacts. There are 5 associated contacts for F1 and 3 filtering associated contacts. There are 7 associated contacts for F3 and 5 filtering associated contacts.

[0122] Determine whether the number of associated contacts for each feature of interest in the image exceeds a preset threshold of 5. F1: The number of associated contacts is 5, which is not greater than 5. Since the condition of "greater than" is not met, the cleaning process is stopped when the similarity between the isolating switch contact and F1 meets the requirements. F1 is the image feature of the cleaning process.

[0123] It is understood that the method for determining the verification and control target focusing on image features is as follows: In this embodiment, the impact of the verification process under the current monitoring and processing method is determined by using the interruption processing data in different disconnector switch contacts. Based on the impact of the verification process under the current monitoring and processing method, the verification control target of the image features of concern is determined. This ensures the reliability of the verification process of different image features of concern while also reducing the impact on different disconnector switch contacts during the cleaning process.

[0124] S41 uses the interruption processing data in different disconnector switch contacts to determine the interruption-affected contact in the disconnector switch contacts; It should be noted that the stop-affect contact mentioned is the isolating switch contact that is stopped during the cleaning process.

[0125] The term "stop-affect contact" refers to an isolating switch contact that stops the cleaning process during laser cleaning because it detects that the similarity to the features of the image of interest meets the requirements.

[0126] Suppose that during the cleaning process of a certain batch of contacts, some contacts are stopped from cleaning because they are detected to match the features of the image of interest. These contacts are the contacts affected by the stoppage.

[0127] This step identifies the trigger point of the suspension by suspending processing data. Its significance lies in providing basic data for the subsequent verification and determination of the scope of the suspension impact.

[0128] Specifically, if the number of interrupted contacts in the disconnector switch contacts is greater than a preset threshold for the number of affected contacts, then all the image features of interest are determined to belong to the verification and control target, that is, the same image feature of interest will not be verified in the same disconnector switch contact.

[0129] If the number of contacts affected by the suspension exceeds the preset threshold, indicating a large scope of the suspension, then all image features of interest will be identified as verification and control targets, and the same image feature of interest will not be repeatedly verified in all disconnector contacts.

[0130] This step involves incorporating all features into the verification and control targets when the number of contacts affected by the shutdown is high. Its significance lies in the fact that when the scope of the shutdown impact is large, incorporating all features into the control can comprehensively avoid duplicate verification, thereby improving the standardization of the verification process without significantly affecting the cleaning efficiency.

[0131] It should also be noted that if the number of interruption-affecting contacts in the disconnecting switch contacts is not greater than the preset threshold for the number of affected contacts, proceed to step S42. If the number of affected contacts does not exceed the preset threshold for the number of affected contacts, it indicates that the scope of the impact of the suspension is small, and further analysis is needed to determine the verification control target.

[0132] This step proceeds to further analysis when the number of contacts affected by the suspension is small. Its significance lies in the fact that when the scope of the suspension impact is small, it is necessary to refine the management and verification of the control range by calculating the suspension impact weight coefficient and the suspension impact value, thereby avoiding the impact of over-control on cleaning efficiency.

[0133] S42, based on the termination processing data of different termination-affected contacts, determine the number of termination processing times in different termination-affected contacts; It is understandable that the product of the number of interruption processes in the interruption contact and the preset factor is used to determine the interruption impact weight coefficient of the interruption contact. It is then determined whether the average value of the interruption impact weight coefficients of different interruption contactes is greater than the preset impact weight coefficient threshold. If so, proceed to step S43. If not, it is determined that all the image features of interest do not belong to the verification control target. The interruption impact weight coefficient refers to the weight value calculated based on the number of interruption processes and preset factors, which quantifies the degree of impact of the interruption impact contact on the cleaning process.

[0134] The calculation method for the weighting coefficient of the suspension impact is as follows: Let Tk be the number of interruption processes for the k-th interruption-affected contact, and let F be the preset factor, then: The weighting coefficient for the termination effect of the kth termination contact is Wk = Tk × F.

[0135] Assuming the number of interruption processes for each interruption-affected contact has been determined, calculate the interruption impact weight coefficient for each contact. If the average value is greater than the preset impact weight coefficient threshold, proceed to S43 for further analysis; if the average value is not greater than the threshold, then all image features of interest do not belong to the verification control target, i.e., no verification control restrictions are imposed.

[0136] This step determines whether further management and control scope is needed by using the impact weighting coefficient of the suspension. Its significance is that when the impact of the suspension is small, no control is implemented to reduce intervention in the cleaning process, and when the impact of the suspension is large, it is transitioned to refined management.

[0137] The preset factor is determined as follows: its core determining factor is the normalized coefficient of the impact of a single interruption on the contact cleaning quality—there is a physical benchmark value for the degree of influence of a single interruption on the heat-affected zone and oxide layer peeling uniformity of the contact surface. This benchmark value, after normalization, becomes the preset factor, ensuring that the interruption impact weight coefficient can reflect the cumulative impact of the number of interruptions on the cleaning quality. Auxiliary reference factors include the power recovery characteristics of the laser cleaning device (the slower the power recovery, the greater the impact of a single interruption, and the more the preset factor needs to be increased) and the thermal sensitivity of the contact material (the higher the thermal sensitivity, the greater the thermal impact of a single interruption, and the more the preset factor needs to be increased). Therefore, the normalized coefficient of the impact of a single interruption on the contact cleaning quality is taken as the preset factor.

[0138] S43, using the interruption-affecting contacts in the disconnecting switch contacts and the number of interruption processes in different interruption-affecting contacts, determine the verification and control target of the image features of interest.

[0139] Specifically, in the above steps, the stopping influence value is determined by the proportion of stopping influence contacts in the disconnector switch contacts and the stopping influence weight coefficient of different stopping influence contacts. It is then determined whether the stopping influence value meets the requirements. If not, the verification and control target of the image features of interest is determined to be the image features of interest whose verification requirement value is less than the preset requirement threshold. If so, all image features of interest are determined to belong to the verification and control target.

[0140] The suspension impact value refers to a quantitative indicator that reflects the overall impact of the verification process under the current monitoring and processing method, calculated by combining the suspension impact contact ratio and the suspension impact weight coefficient.

[0141] The calculation method for the impact value of the suspension is as follows: Let Nh be the number of contacts affected by the interruption, and Ntotal be the total number of contacts of the disconnecting switch. The proportion of contacts affected by the interruption is P = Nh / Ntotal. Let the average weighting coefficient of the termination effect of different termination-affecting contacts be Wavg, then: The impact of the suspension = P × Wavg.

[0142] If the calculated impact value of the suspension meets the requirements, it indicates that the impact of the verification process under the current monitoring and processing method is controllable. In this case, all image features of interest are identified as verification control targets. If the impact value of the suspension does not meet the requirements, it indicates that the impact is significant. In this case, only image features of interest whose verification requirement value is less than the preset requirement threshold are identified as verification control targets to narrow the control scope.

[0143] This step determines the scope of the verification control target by stopping the impact value. Its significance is that when the impact is controllable, all features are fully controlled, while when the impact is large, only features with lower verification requirement values ​​are controlled to narrow the control scope, thereby achieving a balance between verification processing reliability and cleaning efficiency.

[0144] It should be noted that if the image features of interest belong to the verification and control target, then no repeated verification processing will be performed on any of the disconnector contacts.

[0145] If a certain image feature of interest is identified as the target for verification and control, then in the subsequent cleaning process of all disconnector switch contacts, once the feature has been verified, the same feature will not be verified again in other contacts, so as to reduce the impact on the cleaning process.

[0146] This step links the verification control objectives with the limitations of repeated verification processing. Its significance lies in reducing the cumulative impact of verification processing on the cleaning process by avoiding repeated verification of the same image features of interest in multiple contacts, thereby improving cleaning efficiency while ensuring verification reliability.

[0147] This embodiment, through steps S41 to S43, realizes a complete process from determining the termination impact contact to calculating the termination impact weight coefficient and then determining the verification control target. Its core value is reflected in three aspects: First, by judging the size of the termination impact range through the number of termination impact contacts, all features are fully controlled when the impact range is small; second, by judging the grading of the termination impact weight coefficient and termination impact value, the verification control range is dynamically adjusted when the degree of impact is different; third, by further screening the verification requirement value, only low verification requirement features are controlled when the degree of impact is large, so as to narrow the control range, thereby achieving an adaptive balance between verification processing reliability and cleaning efficiency.

[0148] Continuing with the calculation results of S3, we focus on image features F1 and F3. During the cleaning process of a batch of 120 disconnector switch contacts, the cleaning process was stopped for 15 contacts because the similarity to F1 met the requirements. These contacts are the ones that caused the stoppage.

[0149] Step S41: The number of affected contacts is 15.

[0150] Determine whether the number of affected contacts 15 is greater than the preset threshold of the number of affected contacts 10: If 15 is greater than 10, proceed to S42.

[0151] Step S42: Determine the number of stop processing operations for each stop-affected contact. The number of stop processing operations for the 15 stop-affected contacts are as follows: 1 time for each of the 5 contacts, 2 times for each of the 7 contacts, and 3 times for each of the 3 contacts.

[0152] Calculate the termination impact weighting coefficient for each termination-affected contact (preset factor is 0.5): 5 contacts: Stoppage impact weighting coefficient = 1 × 0.5 = 0.5; 7 contacts: Stoppage impact weighting coefficient = 2 × 0.5 = 1.0; Three contacts: Stoppage impact weighting coefficient = 3 × 0.5 = 1.5.

[0153] The average value of the impact weight coefficient for the termination is calculated as (5 × 0.5 + 7 × 1.0 + 3 × 1.5) / 15 = (2.5 + 7.0 + 4.5) / 15 = 14.0 / 15 ≈ 0.933. The average value of 0.933 is then checked against the preset impact weight coefficient threshold of 0.8. If 0.933 is greater than 0.8, proceed to step S43.

[0154] Step S43: Calculate the impact value of termination.

[0155] The impact of the suspension on the contact ratio P = 15 / 120 = 0.125; The average weighted coefficient of the impact of the suspension is Wavg = 0.933; Impact of suspension = P × Wavg = 0.125 × 0.933 ≈ 0.117; Determine if the impact value of the termination is 0.117 (not greater than 0.10): 0.117 is greater than 0.10, therefore it does not meet the requirement.

[0156] Based on the path that does not meet the requirements, the verification control target is the image features of interest whose verification requirement value is less than the preset requirement threshold (0.3).

[0157] Example 2 In a second aspect, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described image monitoring and analysis method for laser cleaning of disconnector contacts when running the computer program.

[0158] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0159] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0160] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. An image monitoring and analysis method for laser cleaning of disconnector switch contacts, characterized in that, Specifically, it includes: Based on the operating data of the disconnector contacts after cleaning, the abnormal temperature fluctuations of the disconnector contacts are determined, and combined with the abnormal correlation time range of the operating temperatures of different disconnector contacts, the image feature extraction and processing strategy of the disconnector contacts during the laser cleaning process is determined. Based on the extraction and processing strategy, the load range for extracting and processing the image features of the disconnector switch contacts is determined. The load range for extracting and processing the image features of the disconnector switch contacts is taken as the associated load range. Based on the abnormal changes in the operating temperature of the disconnector switch contacts in each associated load range, the image features of interest in the image features are determined. Based on the update processing results of the image features of interest, and combined with the running time of the disconnector contacts corresponding to different image features of interest, the image monitoring and analysis method for the cleaning process of the disconnector contacts within the associated load range is determined. Based on the image monitoring and analysis method, the interruption control mode of the disconnector switch contacts during the cleaning process is determined, and the verification and control target of the image features of interest is determined based on the interruption processing data of different disconnector switch contacts.

2. The image monitoring and analysis method for laser cleaning of disconnector switch contacts as described in claim 1, characterized in that, The operating data of the disconnecting switch contacts after cleaning includes the operating temperature of the disconnecting switch contacts during the operation process after cleaning.

3. The image monitoring and analysis method for laser cleaning of disconnector switch contacts as described in claim 1, characterized in that, Abnormal fluctuations in the operating temperature of the disconnector contact include abnormal periods of operating temperature and the cleaning duration range during which the abnormal periods occur.

4. The image monitoring and analysis method for laser cleaning of disconnector switch contacts as described in claim 1, characterized in that, The abnormal operating temperature correlation interval of the disconnecting switch contact is the interval during which the operating temperature of the disconnecting switch contact first becomes abnormal.

5. The image monitoring and analysis method for laser cleaning of disconnector switch contacts as described in claim 1, characterized in that, The method for determining the image feature extraction and processing strategy of the disconnecting switch contacts during the laser cleaning process is as follows: Based on the abnormal fluctuations in the operating temperature of the disconnecting switch contacts, identify the disconnecting switch contacts in which the operating temperature is abnormal during certain periods, and designate them as abnormal switch contacts. Based on the abnormal association duration interval of the abnormal switch contact, determine the abnormal switch contacts that belong to the abnormal association duration interval in different duration intervals, and regard them as associated abnormal contacts; By utilizing the associated abnormal contacts within different time intervals, an image feature extraction and processing strategy for the isolating switch contacts during the laser cleaning process is determined.

6. The image monitoring and analysis method for laser cleaning of disconnector switch contacts as described in claim 5, characterized in that, By utilizing the associated abnormal contacts within different time intervals, an image feature extraction and processing strategy for the disconnecting switch contacts during laser cleaning is determined, specifically including: Based on the associated abnormal contacts in different time intervals, the time intervals that do not meet the requirements are identified and used as the screening time intervals. If the number of associated abnormal contacts in the screening time interval does not meet the requirements, then the number of associated abnormal contacts in the screening time interval is relatively large, and the degree of abnormality in laser cleaning is relatively serious. Therefore, the image feature extraction and processing strategy for the disconnecting switch contacts in the laser cleaning process is determined to be to perform image feature extraction and processing on all disconnecting switch contacts in all load intervals.

7. The image monitoring and analysis method for laser cleaning of disconnector switch contacts as described in claim 1, characterized in that, The method for determining the image features of interest in the image features is as follows: The number of associated load intervals is determined using associated load interval data; Based on the abnormal temperature fluctuations of the disconnector contacts in different associated load ranges, the cleaning abnormality coefficients for different associated load ranges are determined. By utilizing the number of associated load intervals and the cleaning anomaly coefficients of different associated load intervals, the image features of interest in the image features are determined.

8. The image monitoring and analysis method for laser cleaning of disconnector switch contacts as described in claim 7, characterized in that, Using the number of associated load intervals and the cleaning anomaly coefficients of different associated load intervals, the image features of interest in the image features are determined, specifically including: If the average value of the cleaning anomaly coefficient in different associated load intervals is greater than the preset value of the anomaly coefficient, then the image feature of interest in the image features is determined to be one of the image features of interest. If, after cleaning, the number of isolating switch contacts with the image feature is above a first threshold in all the load intervals of interest, then the image feature is determined to belong to the image feature of interest.

9. The image monitoring and analysis method for laser cleaning of disconnector switch contacts as described in claim 1, characterized in that, The method for determining the verification and control target of the image feature being monitored is as follows: Based on the interruption processing data in different disconnector switch contacts, determine the interruption-affected contacts in the disconnector switch contacts; Based on the termination processing data of different termination-affected contacts, determine the number of termination processing times for different termination-affected contacts; By utilizing the interruption-affecting contacts in the disconnecting switch contacts and the number of interruption processes in different interruption-affecting contacts, the verification and control target of the image features of interest is determined.

10. A computer system, comprising: A memory and processor connected in communication, and a computer program stored in the memory and capable of running on the processor, characterized in that, when the processor runs the computer program, it executes an image monitoring and analysis method for laser cleaning of disconnector contacts as described in any one of claims 1-9.

Images