Cooperative enhanced electrical equipment insulation state monitoring method and system

Abstract

The invention relates to the field of equipment monitoring, in particular to a collaborative enhanced electrical equipment insulation state monitoring method and system. The method comprises the following steps: performing image capture on the surface of the electrical equipment to obtain image data; adopting a preset recognition algorithm to recognize and analyze the discharge trace and the corrosion condition to obtain an analysis result; if the analysis result shows that no obvious trace exists, a covering type monitoring step is adopted: comprehensive electrical parameter detection is carried out on the electrical equipment, comprehensive electrical parameters and physical state information of the equipment are obtained, and a comprehensive monitoring result is obtained; and if the analysis result shows that an obvious trace exists, entering a key monitoring step: based on the recorded trace position and according to the analysis result, carrying out targeted electrical parameter detection on a related area to obtain a key monitoring result, and carrying out deep fusion on the analysis result, the comprehensive monitoring result and the key monitoring result to obtain comprehensive evaluation. According to the invention, the monitoring efficiency is improved, problems can be found quickly, and targeted detection is realized.

Description

Technical Field

[0001] The present invention relates to the field of equipment monitoring, and in particular to a collaboratively enhanced electrical equipment insulation status monitoring method and system. Background Art

[0002] In the power system, the insulation state of electrical equipment is directly related to the operational safety and reliability of the equipment. With the continuous expansion of the power system and the development of technology, the traditional insulation monitoring method of electrical equipment can no longer meet the current needs. The traditional method mainly relies on regular power outage detection, which is not only inefficient, but also unable to achieve real-time monitoring of the insulation state of the equipment, resulting in the inability to timely detect and prevent potential insulation failures. For example, the real-time insulation state detection system described in patent document CN118050606A, although a real-time monitoring system is proposed, the system mainly focuses on real-time acquisition of insulation state parameters corresponding to electrical equipment through a data acquisition device, and uses an insulation recognition model for analysis, lacking a detailed description of direct image analysis and targeted monitoring of the equipment surface.

[0003] In the existing technology, insulation monitoring of electrical equipment mainly relies on a single detection method, such as electrical parameter testing or manual visual inspection. These methods have obvious limitations: electrical parameter testing can usually only be performed in a power-off state, and online monitoring cannot be achieved; while manual visual inspection is limited by the subjective judgment and experience of the inspectors, and accuracy and consistency are difficult to guarantee. In addition, with the expansion of equipment scale and the complexity of the operating environment, a single monitoring method has been difficult to fully cover all parts of the equipment, especially in high-voltage and complex electrical equipment, monitoring blind spots and data islands are becoming increasingly prominent. Patent document CN118362844A provides a method for detecting the insulation status of a ring network box. The method deploys a variety of sensors, including numerical type sensors and image type sensors, to obtain numerical data and terahertz spectral images directly related to the insulation status, but the method may face challenges in data integration and real-time response in practical applications, and there is no focus on detection and targeted detection.

[0004] In order to improve the accuracy and real-time performance of insulation status monitoring of electrical equipment, researchers and engineers have begun to explore collaborative monitoring methods that combine multiple monitoring technologies. These methods aim to achieve comprehensive and real-time monitoring of the insulation status of equipment by integrating multiple sensors and monitoring technologies. However, existing collaborative monitoring methods still face challenges in data processing and information fusion. How to effectively integrate data from different monitoring methods, conduct rapid and targeted detection, and provide accurate comprehensive evaluation is a key issue in the current technological development. Summary of the invention

[0005] The present invention provides a collaborative enhanced electrical equipment insulation status monitoring method and system, aiming to solve the problem of how to achieve comprehensive, rapid and targeted detection of the insulation status of electrical equipment and provide accurate comprehensive evaluation.

[0006] To achieve the above purpose, the following technical solution is adopted.

[0007] A method for monitoring insulation status of a collaboratively enhanced electrical device comprises the following steps:

[0008] Using a high-resolution camera to capture images of the surface of the electrical equipment to obtain image data;

[0009] According to the image data, the preset recognition algorithm is used to identify and analyze the discharge traces and corrosion conditions to obtain the analysis results;

[0010] If the analysis results show that there are no obvious traces, then a covering monitoring step is taken: a comprehensive electrical parameter test is conducted on the electrical equipment to obtain the comprehensive electrical parameters and physical status information of the equipment and obtain a comprehensive monitoring result;

[0011] If the analysis results show that there are obvious traces, the key monitoring step is entered: based on the recorded trace positions and according to the analysis results, targeted electrical parameter detection is performed on the relevant areas to obtain key monitoring results;

[0012] The analysis results, comprehensive monitoring results, and key monitoring results are deeply integrated to obtain a comprehensive assessment of the insulation status of electrical equipment.

[0013] Optionally, the electrical parameter detection includes but is not limited to ultrasonic detection and partial discharge detection.

[0014] Optionally, it also includes a correlation extension detection step: after the key detection step is completed, an extended detection is performed on the equipment area related to the trace position to evaluate potential insulation state changes and obtain a correlation detection result.

[0015] Optionally, based on the image data, a preset recognition algorithm is used to identify and analyze the discharge traces and corrosion conditions to obtain the analysis results, specifically including:

[0016] The image data captured using a high-resolution camera is input into a preset recognition algorithm;

[0017] Perform noise removal and contrast enhancement on the input image data to obtain a preprocessed image;

[0018] The convolutional neural network of the YOLOv5 algorithm is used to extract features from the preprocessed images to obtain a series of feature maps;

[0019] Through the feature pyramid network structure, the feature maps of different levels in the YOLOv5 algorithm are fused to obtain a fused feature map;

[0020] Apply sparse coding technology to the obtained fusion feature map to extract key features;

[0021] Use recursive neural networks to recursively process the extracted key features in time series;

[0022] The recursively processed features are input into the classifier for final identification of discharge traces and corrosion conditions;

[0023] The final identification results are output to form analysis results of discharge traces and corrosion conditions, which serve as a basis for further detection of the insulation status of electrical equipment.

[0024] Optionally, the covering monitoring step specifically includes:

[0025] Based on the layout of electrical equipment and historical fault data, the equipment is divided into multiple monitoring areas, each of which corresponds to specific monitoring parameters and methods;

[0026] Develop a monitoring plan for each monitoring area, including monitoring frequency, detection technology to be used, and expected monitoring results;

[0027] According to the monitoring plan, ultrasonic detection and partial discharge detection technology are used to collect electrical parameters in each monitoring area to obtain original electrical parameter data;

[0028] Filtering, denoising and normalizing the obtained raw electrical parameter data to obtain pre-processed data;

[0029] Analyze the preprocessed data using a preset machine learning algorithm to identify abnormal patterns or trends and obtain abnormal detection results. The machine learning algorithm is trained based on historical monitoring data and is used to identify parameter changes that deviate from normal operating conditions;

[0030] Combined with the abnormal detection results, a comprehensive status assessment is conducted on the electrical equipment in each monitoring area. The assessment results include the health index of the equipment and the potential risk prediction;

[0031] The evaluation results are synthesized to form a comprehensive monitoring report of the equipment, recording the status of each monitoring area and providing maintenance recommendations and risk warnings.

[0032] Optionally, the key monitoring steps specifically include:

[0033] After the discharge traces are identified in the image analysis step, the positions and features of the traces are confirmed, and the confirmed information is used as input data for key monitoring;

[0034] According to the confirmed trace location, select the monitoring point on the electrical equipment corresponding to the trace;

[0035] Based on the characteristics and historical data of the monitoring points, develop customized testing plans for each monitoring point, including testing parameters, testing methods and testing cycles;

[0036] Perform synchronous multi-parameter detection on each monitoring point, including but not limited to partial discharge, insulation resistance, temperature and humidity, to obtain comprehensive electrical parameter data;

[0037] Synchronize and fuse the comprehensive electrical parameter data obtained to obtain fused data;

[0038] Perform trend analysis on the fused data to identify potential insulation degradation trends and trigger an early warning mechanism when the trend analysis results exceed the preset threshold;

[0039] Based on the trend analysis results, targeted maintenance decision support is provided, including suggestions on maintenance timing, maintenance scope and maintenance methods.

[0040] Optionally, the analysis results, comprehensive monitoring results, and key monitoring results are deeply integrated to obtain a comprehensive evaluation of the insulation status of the electrical equipment, specifically including:

[0041] Integrate the image analysis results, data from comprehensive monitoring reports and results from key monitoring reports to form a unified data set;

[0042] Verify the consistency of the integrated data set to ensure that the temporal and spatial correspondence of data from different sources is accurate and obtain a verified data set;

[0043] Using the verified data set, the features of different monitoring results are mapped into a unified reference framework to obtain a mapped feature set;

[0044] According to the importance and reliability of each monitoring result, weights are assigned to the mapped feature set to obtain a weighted feature set;

[0045] Based on the weighted feature set, the insulation status of the electrical equipment is calculated to reflect the comprehensive score of the insulation health level of the equipment; based on the weighted feature set, the change trend of the insulation status of the electrical equipment over time is analyzed to obtain a trend analysis report;

[0046] Based on the trend analysis report and comprehensive score, combined with maintenance priorities and resource allocation, specific maintenance decisions and action plans are made to obtain a maintenance decision report.

[0047] Optionally, the step of selecting a monitoring point on the electrical equipment corresponding to the trace according to the confirmed trace position specifically includes:

[0048] Convert the identified discharge trace locations into digital coordinates of the electrical device;

[0049] Map the physical layout of electrical equipment to a digital coordinate system;

[0050] According to the digital coordinates of the trace, the monitoring points on the electrical equipment directly related to the trace are screened out, and the monitoring points cover the trace and the surrounding areas that may be affected;

[0051] Based on the characteristics of the traces and historical monitoring data, the selected monitoring points are optimized, redundant monitoring points are eliminated, and the layout of monitoring points that fully covers the trace area is determined;

[0052] Conduct on-site calibration of selected monitoring points;

[0053] Attach data tags to each monitoring point, including the number, location, associated trace features and historical monitoring records of the monitoring point.

[0054] Optionally, a trend analysis is performed on the fused data to identify potential insulation degradation trends, and a warning mechanism is triggered when the trend analysis results exceed a preset threshold, specifically including:

[0055] Based on the fused data, determine the statistical features reflecting the insulation status change for trend analysis, including but not limited to the mean value, standard deviation, skewness, and peak frequency;

[0056] Compare current statistical characteristics with those of the same monitoring points in history to identify long-term trends in insulation status;

[0057] Pre-train trend models based on historical data to predict future trends in insulation status;

[0058] Apply pre-trained trend models to analyze real-time monitoring data, monitor real-time changes in insulation status, and generate trend reports;

[0059] Compare the statistical features in the generated trend report with the warning threshold to determine whether there is potential insulation degradation. The warning threshold is set based on historical data and expert experience;

[0060] If the trend analysis results exceed the preset threshold, an early warning signal is automatically generated to indicate possible insulation degradation problems;

[0061] Based on the warning signals, formulate corresponding warning response measures, including but not limited to immediate maintenance recommendations, further monitoring requirements or urgent safety measures.

[0062] A collaborative enhanced electrical equipment insulation status monitoring system, comprising:

[0063] An image capturing unit is used to capture an image of the surface of the electrical device using a high-resolution camera to obtain image data.

[0064] The image analysis unit is used to identify and analyze discharge traces and corrosion conditions based on image data using a preset recognition algorithm to obtain analysis results.

[0065] The monitoring decision unit is used to determine whether to perform a coverage monitoring step or a key monitoring step according to the analysis result of the image analysis unit.

[0066] The comprehensive monitoring execution unit is used to execute the covering monitoring steps under the instruction of the monitoring decision unit, conduct comprehensive electrical parameter detection on the electrical equipment, obtain the comprehensive electrical parameters and physical status information of the equipment, and obtain comprehensive monitoring results.

[0067] The key monitoring execution unit is used to execute the key monitoring steps under the instruction of the monitoring decision unit, and to conduct targeted electrical parameter detection on the relevant areas based on the recorded trace positions and the analysis results to obtain the key monitoring results.

[0068] The data fusion unit is used to deeply integrate the analysis results, comprehensive monitoring results, and key monitoring results to obtain a comprehensive assessment of the insulation status of electrical equipment.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] The collaboratively enhanced electrical equipment insulation status monitoring method provided by the present application realizes real-time monitoring of discharge traces and corrosion conditions on the surface of electrical equipment by combining high-resolution image capture and preset recognition algorithms. This method overcomes the limitations of traditional monitoring methods that cannot monitor and accurately identify insulation problems in real time, and provides a more accurate and comprehensive monitoring method. The present invention improves the efficiency of electrical equipment insulation status monitoring through key detection steps, quickly discovers problems, and conducts targeted detection.

[0071] By deeply integrating the analysis results, comprehensive monitoring results, and key monitoring results, the method of this application can provide a comprehensive assessment of the insulation status of electrical equipment. This deep fusion technology not only improves the accuracy and detection efficiency of the monitoring results, but also enhances the ability to predict the trend of changes in the insulation status of the equipment, thereby providing a more scientific basis for equipment maintenance and fault prevention.

[0072] The method also includes triggering an early warning mechanism when the trend analysis results exceed a preset threshold, which can promptly indicate possible insulation degradation problems and provide an important time window for timely maintenance and fault prevention of equipment. Through this early warning mechanism, the risk of unexpected power outages can be effectively reduced and the reliability and safety of the power system can be improved.

[0073] In summary, the synergistically enhanced electrical equipment insulation status monitoring method provided in this application solves the limitations of the existing technology through technological innovation, realizes comprehensive and real-time monitoring of the insulation status of electrical equipment, and provides accurate comprehensive evaluation and timely early warning, which has important practical value and broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 It is a flow chart of an embodiment of a collaborative enhanced electrical equipment insulation status monitoring method of the present invention.

[0075] Figure 2 It is a module schematic diagram of an embodiment of a collaboratively enhanced electrical equipment insulation status monitoring system of the present invention. DETAILED DESCRIPTION

[0076] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0077] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present application belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention.

[0078] Example 1

[0079] like Figure 1 As shown, a collaboratively enhanced method for monitoring the insulation status of electrical equipment is provided. The method realizes comprehensive monitoring and evaluation of the insulation status of electrical equipment by comprehensively using image processing technology, pattern recognition algorithm and electrical parameter detection technology.

[0080] First, a high-resolution camera is used to capture images of the surface of the electrical equipment to obtain clear image data of the equipment surface. These image data will serve as the basis for subsequent analysis to identify and analyze discharge marks and corrosion on the equipment surface.

[0081] Next, the captured image data is input into a preset recognition algorithm. The recognition algorithm includes a series of image processing steps, such as noise removal and contrast enhancement, to improve the recognizability of features in the image. Then, a deep learning model, especially the convolutional neural network of the YOLOv5 algorithm, is used to extract features from the preprocessed image to obtain a series of feature maps. Through the feature pyramid network structure, feature maps at different levels are fused to obtain a richer fused feature map. Furthermore, sparse coding technology is applied to the fused feature map to extract key features, and these key features are recursively processed in time series using a recursive neural network to identify discharge marks and corrosion conditions. Finally, the recursively processed features are input into the classifier for final recognition of discharge marks and corrosion conditions, and the analysis results are output.

[0082] If the analysis results show that there are no obvious traces on the surface of the electrical equipment, the covering monitoring step is performed. This step involves comprehensive electrical parameter testing of the electrical equipment, including but not limited to ultrasonic testing and partial discharge testing, to obtain comprehensive electrical parameters and physical status information of the equipment. These data will be used to generate comprehensive monitoring results.

[0083] If the analysis results show that there are obvious traces, the key monitoring step is entered. Based on the recorded trace locations, the monitoring points corresponding to the traces are selected, and a customized detection plan is formulated based on the characteristics and historical data of the monitoring points. Synchronous multi-parameter detection is performed on each monitoring point, including parameters such as partial discharge, insulation resistance, temperature and humidity, to obtain comprehensive electrical parameter data. The acquired data is synchronized and fused to obtain the fused data, and trend analysis is performed on it to identify potential insulation degradation trends.

[0084] Finally, the image analysis results, comprehensive monitoring results, and key monitoring results are deeply integrated to obtain a comprehensive assessment of the insulation status of electrical equipment. This step involves integrating data from different sources and types to provide a comprehensive view of the equipment status through data fusion technology. The comprehensive assessment results can be used to guide equipment maintenance and fault prevention, ensuring the safe and stable operation of the power system.

[0085] Through the above implementation, the method can realize real-time and comprehensive monitoring of the insulation status of electrical equipment, timely discover potential insulation problems, reduce the risk of unexpected power outages, and provide a scientific basis for equipment maintenance and fault prevention.

[0086] As a specific example of the above example, in performing the coverage monitoring step or the key monitoring step, the method includes but is not limited to the following electrical parameter detection technologies:

[0087] Ultrasonic testing: Use ultrasonic sensors to test electrical equipment to identify potential defects inside the equipment, such as cracks, corrosion, or internal discharge. The reflection and propagation characteristics of ultrasonic signals can reveal abnormal conditions inside the equipment. Specifically, ultrasonic waves of a specific frequency are emitted, and echoes reflected by the internal structure of the equipment are received. The time, amplitude, and frequency changes of these echoes are analyzed to identify and locate internal defects.

[0088] Partial discharge detection: Use partial discharge detection equipment to monitor discharge activity inside electrical equipment. Partial discharge refers to the localized electrical breakdown that occurs in the insulation material of electrical equipment, which is usually an early sign of insulation damage. By capturing and analyzing weak signals associated with partial discharge events, such as electromagnetic waves, optical radiation, or sound waves, the insulation status of the equipment can be evaluated.

[0089] When implementing ultrasonic testing and partial discharge testing, the method further comprises the following steps:

[0090] Calibration of testing equipment: Before testing, the ultrasonic sensor and partial discharge testing equipment are calibrated to ensure the accuracy and reliability of the test results.

[0091] Data acquisition: During the detection process, the data generated by the ultrasonic and partial discharge detection equipment is collected in real time, including the signal amplitude, frequency, time and other related parameters.

[0092] Signal processing: Preprocess the collected raw data, including filtering, denoising and normalization, to improve the quality of the signal and facilitate subsequent analysis.

[0093] Feature extraction: Extract features related to the insulation status of the equipment from the preprocessed data, such as the attenuation rate of the ultrasonic signal, the frequency and energy distribution of partial discharge.

[0094] Anomaly Identification: Using machine learning algorithms or preset thresholds, the extracted features are analyzed to identify and classify anomalies, such as partial discharge events or unusual changes in ultrasonic signals.

[0095] Result Recording and Reporting: Record the test results and generate detailed test reports including test data, analysis results and maintenance recommendations.

[0096] This method can comprehensively and accurately monitor the insulation status of electrical equipment, timely discover potential insulation problems, and provide a scientific basis for equipment maintenance and fault prevention. The implementation of this method not only improves the efficiency and accuracy of monitoring, but also enhances the ability to predict the trend of equipment insulation status changes, thereby providing a strong guarantee for the safe and stable operation of the power system.

[0097] As a specific example, the identification step specifically includes:

[0098] Use high-resolution cameras to capture detailed images of the surface of electrical equipment to ensure that the acquired image data can accurately reflect the discharge marks and corrosion conditions of the equipment.

[0099] The captured raw image data is input into the preset image processing algorithm. First, noise removal is performed to eliminate random interference in the image and improve image quality. Then contrast enhancement is performed to make the features in the image more obvious, which is convenient for subsequent feature extraction.

[0100] The convolutional neural network (CNN) of the YOLOv5 algorithm is used to extract features from the preprocessed images. YOLOv5 is an efficient target detection algorithm, and its convolutional neural network can extract key visual features from images to form a series of feature maps.

[0101] The feature pyramid network (FPN) structure is used to fuse the feature maps of different levels in the YOLOv5 algorithm. FPN can integrate feature information of different scales, enhance the expressiveness of features, and improve the accuracy of detection.

[0102] Sparse coding technology is applied to the fused feature map to extract key features. Sparse coding technology can extract the most representative features from a large amount of data, providing strong support for the identification of discharge traces and corrosion conditions.

[0103] The extracted key features are recursively processed on the time series using a recursive neural network (RNN). RNN is particularly suitable for processing sequence data and can capture the changing trend of features over time, which is particularly important for analyzing the dynamic changes of discharge traces and corrosion conditions.

[0104] The recursively processed features are input into a classifier for final identification of discharge traces and corrosion conditions. The classifier can be a deep learning-based model, such as a convolutional neural network, which can classify discharge traces and corrosion conditions based on the learned features.

[0105] The final identification results are output to form a detailed analysis report of discharge traces and corrosion conditions. These results will serve as a basis for further testing of the insulation status of electrical equipment and provide a scientific basis for equipment maintenance and fault prevention.

[0106] This method can accurately monitor the insulation status of electrical equipment, timely discover potential insulation problems, reduce the risk of unexpected power outages, and provide a scientific basis for equipment maintenance and fault prevention. The implementation of this method not only improves the accuracy and reliability of monitoring, but also enhances the ability to predict the trend of equipment insulation status changes, thereby providing a strong guarantee for the safe and stable operation of the power system.

[0107] As a specific example, the coverage monitoring steps specifically include:

[0108] First, the equipment is divided into multiple monitoring areas based on the layout of the electrical equipment and historical failure data. This step involves a detailed survey of the equipment to determine the characteristics and historical failure modes of each area, so that specific monitoring parameters and methods can be customized for each area.

[0109] A detailed monitoring plan is developed for each assigned monitoring area. The monitoring plan includes determining the monitoring frequency, selecting appropriate detection technologies such as ultrasonic testing and partial discharge detection, and the expected monitoring results. The monitoring frequency may vary from area to area, depending on the historical failure rate of the area and the importance of the equipment.

[0110] According to the monitoring plan, electrical parameters are collected for each monitoring area using ultrasonic testing and partial discharge testing technology. Ultrasonic testing is used to detect potential defects inside the equipment, while partial discharge testing is used to identify partial discharge activity in the insulation material. This step involves setting up the testing equipment, performing the test, and recording the raw electrical parameter data.

[0111] The collected raw electrical parameter data is filtered, denoised and normalized. Filtering is used to eliminate high-frequency noise in the data, denoising further reduces the impact of random noise, and normalization ensures that the data is on the same scale for easy analysis.

[0112] The pre-processed data is analyzed using a preset machine learning algorithm to identify abnormal patterns or trends. The machine learning algorithm is trained based on historical monitoring data and can identify parameter changes that deviate from normal operating conditions. This step involves inputting the pre-processed data into the trained model, which outputs anomaly detection results.

[0113] Combined with the abnormal detection results, a comprehensive status assessment is conducted on the electrical equipment in each monitoring area. The assessment results include the health index of the equipment and the potential risk prediction, providing a quantitative measure of the operating status of the equipment.

[0114] The evaluation results are synthesized to form a comprehensive monitoring report for the equipment. The report records the status of each monitoring area, provides maintenance suggestions and risk warnings, and provides decision support for equipment maintenance and fault prevention.

[0115] As a specific example, the key monitoring steps include:

[0116] First, a high-resolution camera is used to capture an image of the surface of the electrical equipment, and the discharge traces are identified through an image analysis step. In this step, the image data is processed using a preset recognition algorithm to determine the location and characteristics of the traces.

[0117] According to the position of the discharge traces identified in the image analysis step, the monitoring points corresponding to the traces on the electrical equipment are selected. These monitoring points are selected as key monitoring areas for more in-depth electrical parameter detection.

[0118] Based on the characteristics of the monitoring point and historical monitoring data, a customized detection plan is developed for each monitoring point. The detection plan includes determining the detection parameters (such as partial discharge, insulation resistance, temperature and humidity, etc.), detection methods (such as ultrasonic detection and partial discharge detection technology) and detection cycle.

[0119] Perform synchronous multi-parameter detection on each monitoring point to obtain comprehensive electrical parameter data, including but not limited to partial discharge level, insulation resistance, temperature and humidity, which provide a basis for subsequent data analysis.

[0120] The electrical parameter data obtained from different monitoring points are synchronized and integrated to form a unified data set. This step ensures the consistency and comparability of the data and provides an accurate data basis for trend analysis.

[0121] Perform trend analysis on the fused data to identify potential insulation degradation trends. In this step, the data is analyzed using a preset machine learning algorithm to identify abnormal patterns or trends. When the trend analysis results exceed the preset threshold, an early warning mechanism is triggered to indicate possible insulation degradation issues.

[0122] Based on the trend analysis results, targeted maintenance decision support is provided, including suggestions on maintenance timing, maintenance scope and maintenance methods to achieve preventive maintenance and failure prevention.

[0123] This method can accurately monitor the insulation status of electrical equipment, timely discover potential insulation problems, reduce the risk of unexpected power outages, and provide a scientific basis for equipment maintenance and fault prevention. The implementation of this method not only improves the accuracy and reliability of monitoring, but also enhances the ability to predict the trend of equipment insulation status changes, thereby providing a strong guarantee for the safe and stable operation of the power system.

[0124] As a specific example of the above example, the step of selecting the monitoring point on the electrical device corresponding to the trace according to the confirmed trace position specifically includes, after the discharge traces are identified in the image analysis step, first converting the position of these traces from the image coordinates to the digital coordinates of the electrical device. This step involves using image processing software or algorithms to convert the position information of the trace in the image into the actual spatial coordinates of the device so as to accurately locate it on the physical device.

[0125] Map the physical layout of electrical equipment to a digital coordinate system. This is usually achieved by creating a digital twin model of the equipment, which accurately reflects the physical structure and component locations of the equipment. In this way, it can be ensured that the digital coordinates correspond to the actual location of the equipment.

[0126] Based on the digital coordinates of the trace, select the monitoring points on the electrical equipment directly related to the trace. The selection of these monitoring points covers the trace and the surrounding areas that may be affected, ensuring the comprehensiveness of the monitoring. The screening process may involve geographic information system (GIS) technology or similar spatial analysis tools.

[0127] Based on the characteristics of the trace and historical monitoring data, the selected monitoring points are optimized. This step includes eliminating redundant monitoring points and determining the monitoring point layout that fully covers the trace area. The optimization process may involve statistical analysis and machine learning techniques to predict which monitoring points are most likely to provide useful information about insulation status changes.

[0128] On-site calibration of selected monitoring points ensures accurate docking of the monitoring equipment with the monitoring points. This step is necessary to eliminate any possible sources of error and ensure the accuracy of the monitoring data. The calibration process may include adjusting the position, angle or other parameters of the monitoring equipment.

[0129] Attach data tags to each monitoring point, including the number, location, associated trace characteristics, and historical monitoring records of the monitoring point. These data tags are used for data management and analysis, so that the data of each monitoring point can be quickly retrieved and analyzed. The creation and maintenance of data tags can be achieved through a database management system.

[0130] As a specific example of the above example, the steps of performing trend analysis on the fused data, identifying potential insulation degradation trends, and triggering an early warning mechanism when the trend analysis results exceed a preset threshold value specifically include:

[0131] From the fused data, a set of statistical features are selected that can quantify key parameters of insulation condition. Features include but are not limited to:

[0132] Average: Calculates the average value of an electrical parameter over a specific period of time, such as the average intensity of partial discharge activity.

[0133] Standard deviation: measures the degree of dispersion of electrical parameter values ​​and reflects the stability of insulation status.

[0134] Skewness: Describes the asymmetry of the electrical parameter distribution and identifies potential abnormal patterns.

[0135] Peak frequency: records the frequency at which electrical parameters reach their peak values, indicating the frequency of partial discharge activity.

[0136] Compare and analyze the current statistical characteristics with the historical statistical characteristics of the same monitoring point. This step compares the changes in current data with historical data to identify the long-term trend of insulation status and thus assess the possibility of insulation degradation.

[0137] Based on historical data, a trend prediction model is trained. The model uses machine learning techniques such as random forest, support vector machine or neural network to train on historical data and learn the changing patterns of insulation status.

[0138] Apply pre-trained trend models to analyze real-time monitoring data to monitor real-time changes in insulation status. This step involves inputting real-time data into the model, and the model output includes the predicted results of insulation status changes.

[0139] Generate a trend report that describes the changing trend of the insulation status in detail. The report contains the real-time value, historical comparison value and predicted value of the statistical characteristics, providing a basis for subsequent early warning judgment.

[0140] Compare the statistical features in the generated trend report with the warning thresholds. The warning thresholds are set based on historical data and expert experience to determine if there are abnormal changes in the insulation state. These thresholds can be static or dynamically adjusted based on the model's predictions.

[0141] If the trend analysis results exceed the preset threshold, the system automatically generates a warning signal to indicate possible insulation degradation problems. The generation of warning signals can be automatic or semi-automatic, depending on the system configuration and user needs.

[0142] Based on the early warning signals, formulate corresponding early warning response measures. These measures include but are not limited to:

[0143] Immediate maintenance recommendations: such as cleaning insulation surfaces and checking the integrity of insulation materials.

[0144] Further monitoring needs: Increase the monitoring frequency in specific areas or introduce more accurate monitoring equipment.

[0145] Emergency safety measures: such as taking power-off measures to prevent equipment failure when early warning signals indicate potential major insulation problems.

[0146] Through the above implementation, the method can realize real-time monitoring and early warning of the insulation status of electrical equipment, timely discover potential insulation degradation problems, and take corresponding maintenance measures, thereby improving the reliability and safety of the power system.

[0147] As a specific example, it also includes:

[0148] First, the image analysis results, data from the comprehensive monitoring report, and the results of the key monitoring report were integrated to form a unified data set. This step involves converting data from different sources and formats into a unified data format and ensuring the consistency of the data in time series and spatial location for subsequent analysis.

[0149] The consistency of the integrated data set is verified to ensure that the correspondence between data from different sources in time and space is accurate. This step includes checking the data's timestamp, spatial location mark, and data quality to ensure the accuracy and reliability of the data set and obtain a verified data set.

[0150] Using the validated dataset, the features of different monitoring results are mapped into a unified reference framework. This step involves converting the features of various monitoring data into a comparable format so that they can be analyzed under the same framework to obtain a mapped feature set.

[0151] According to the importance and reliability of each monitoring result, weights are assigned to the mapped feature set. This step considers the contribution of different monitoring data to insulation status assessment, and the weights are determined through expert experience and historical data analysis to obtain a weighted feature set.

[0152] Based on the weighted feature set, a comprehensive score of the insulation status of the electrical equipment is calculated. This step applies a mathematical model, such as weighted average or other statistical methods, to comprehensively consider all monitoring results to reflect the insulation health level of the equipment.

[0153] Based on the weighted feature set, analyze the change trend of the insulation status of electrical equipment over time. This step uses time series analysis techniques, such as moving average or trend line fitting, to identify the change pattern of insulation status and potential degradation trends, and obtain a trend analysis report.

[0154] Based on the trend analysis report and comprehensive score, combined with maintenance priorities and resource allocation, specific maintenance decisions and action plans are made. This step involves evaluating the cost-effectiveness of different maintenance strategies, determining the best time and method for maintenance, and the rational allocation of resources to obtain a maintenance decision report.

[0155] Example 2

[0156] A collaborative enhanced electrical equipment insulation status monitoring method also includes a correlation extension detection step: after the key detection step is completed, an extended detection is performed on the equipment area related to the trace position to evaluate potential insulation status changes and obtain correlation detection results.

[0157] Analysis of key test results: After completing the key monitoring steps, the key monitoring results obtained are analyzed in detail. These results include data collected from the monitoring points on parameters such as partial discharge, insulation resistance, temperature and humidity. The purpose of the analysis is to determine whether the insulation condition at the trace location and its surrounding areas shows signs of degradation.

[0158] Determine extended inspection areas: Based on the focused inspection results, determine the equipment areas that require extended inspection. These areas may include parts of the electrical equipment near the trace location, as well as adjacent areas that may be affected by insulation degradation. The purpose of the extended inspection is to evaluate changes in the insulation condition in these areas so that preventive measures can be taken.

[0159] Customized testing plan: Customized testing plan is formulated for the determined extended testing area. The plan includes testing parameters, testing methods and testing cycles. Testing parameters may include but are not limited to partial discharge level, insulation resistance, temperature and humidity, etc. Testing methods may involve ultrasonic testing, partial discharge testing and other technologies.

[0160] Perform extended testing: Perform synchronized multi-parameter testing of the extended test area according to the customized test plan. This step involves using appropriate test equipment and techniques to collect electrical parameter data of the extended test area.

[0161] Data fusion and trend analysis: The data collected by the extended detection is integrated with the key monitoring results. Then, the integrated data is analyzed using trend analysis techniques, such as time series analysis, to identify the long-term trend of insulation status.

[0162] Assess potential changes: Based on the trend analysis results, assess the potential changes in insulation status within the extended inspection area. This assessment may involve qualitative and quantitative analysis of insulation degradation trends, as well as prediction of potential risks.

[0163] Generate Correlation Test Results: The evaluation results are collated into a Correlation Test Report that details changes in insulation condition over the extended test area, as well as any potential insulation degradation issues and recommended maintenance actions.

[0164] Through the above implementation, the method can conduct a more in-depth and comprehensive monitoring of the insulation status of electrical equipment, especially after the key monitoring step is completed, the equipment area related to the trace position can be extended to detect, so as to more effectively evaluate and prevent potential insulation problems. This method not only improves the accuracy and reliability of monitoring, but also provides a more scientific basis for equipment maintenance and fault prevention.

[0165] Example 3

[0166] like Figure 2As shown, a collaborative enhanced electrical equipment insulation status monitoring system comprises:

[0167] 1. Image capture unit:

[0168] High-resolution cameras are equipped with multiple high-resolution cameras with autofocus and optical zoom functions to meet the monitoring needs of different distances and angles.

[0169] Image stabilization system. To ensure image quality, the camera integrates an image stabilization system to reduce image blur caused by device vibration or wind.

[0170] Environmental adaptability: the camera is designed to be waterproof, dustproof, and resistant to high and low temperatures to adapt to various outdoor environments.

[0171] Data transmission module: image data is transmitted to the image analysis unit in real time via wireless or wired means.

[0172] 2. Image analysis unit:

[0173] Preprocessing software is used to perform image preprocessing operations such as noise removal, contrast enhancement, and edge detection.

[0174] The deep learning platform integrates deep learning models such as YOLOv5 for feature extraction and identification of discharge traces and corrosion conditions.

[0175] Analysis results storage,The analysis results are stored in the database for subsequent data fusion and historical trend analysis.

[0176] 3. Monitoring decision-making unit:

[0177] The decision-making algorithm, based on the image analysis results, uses logical judgment and machine learning algorithms to automatically determine the monitoring strategy.

[0178] User interaction interface, provides a user interface that allows operators to manually adjust the monitoring strategy or confirm the automatically generated monitoring plan.

[0179] The instruction output module outputs the monitoring decision results in the form of instructions to the comprehensive monitoring execution unit and the key monitoring execution unit.

[0180] 4. Comprehensive monitoring of execution units:

[0181] Electrical parameter detection equipment, including ultrasonic detectors, partial discharge detectors, insulation resistance testers, etc., are used to collect comprehensive electrical parameters of the equipment.

[0182] The data acquisition system is responsible for digitizing the data collected by the detection equipment and transmitting it to the data fusion unit in real time.

[0183] Mobile monitoring platform,In order to achieve comprehensive coverage of large equipment, a mobile monitoring platform, such as a drone or robot, may be required to carry detection equipment for on-site inspection.

[0184] 5. Key monitoring execution units:

[0185] The high-precision positioning system ensures the precise positioning of the monitoring points for targeted detection.

[0186] Customized detection equipment: Depending on the characteristics of the monitoring point, specific detection equipment may be required, such as a highly sensitive partial discharge sensor.

[0187] On-site calibration tools are used to calibrate testing equipment to ensure data accuracy.

[0188] 6. Data fusion unit:

[0189] The data integration platform integrates data from different monitoring units, including image data, electrical parameter data and environmental data.

[0190] Feature mapping and weighting, maps features from different sources into a unified reference frame and assigns weights according to their importance and reliability.

[0191] The comprehensive evaluation algorithm calculates the insulation status score of the equipment based on the weighted feature set and analyzes the changing trend of the insulation status.

[0192] The report generation system automatically generates monitoring reports, including equipment health index, potential risk prediction and maintenance recommendations.

[0193] As a preferred example, it also includes:

[0194] 7. Trend analysis unit:

[0195] The statistical feature extraction module is responsible for extracting key statistical features for trend analysis from the monitoring data, such as mean, standard deviation, skewness, peak frequency, etc.

[0196] The historical data comparison module compares the current statistical characteristics with the statistical characteristics of the same monitoring point in history to identify the long-term change trend of the insulation status.

[0197] Trend prediction model, a model trained based on historical data, is used to predict future changes in insulation status.

[0198] The real-time change monitoring module applies trend models to analyze real-time monitoring data, monitors real-time changes in insulation status, and generates trend reports.

[0199] 8. Early warning mechanism unit:

[0200] The threshold setting module sets the warning threshold based on historical data and expert experience to determine whether there is an abnormal change in the insulation status.

[0201] The early warning signal generation module automatically generates an early warning signal when the trend analysis result exceeds the preset threshold, indicating possible insulation degradation problems.

[0202] The response measures planning module formulates corresponding early warning response measures based on the early warning signals, including immediate maintenance suggestions, further monitoring requirements or urgent safety measures.

[0203] 9. Data Management and Reporting Unit:

[0204] The data storage module is responsible for storing all monitoring data, analysis results and trend reports in a secure database.

[0205] The report preparation module automatically prepares detailed monitoring reports based on monitoring results and trend analysis, providing equipment health index, potential risk prediction and maintenance recommendations.

[0206] User Access Interface, which provides a user interface that allows operators to access monitoring data, analyze results and reports, and manually enter maintenance and response actions.

[0207] 10. Communication and integration unit:

[0208] The data communication module is responsible for data transmission between units within the system and data exchange with external systems (such as the central monitoring system).

[0209] System integration module ensures that the monitoring system can be integrated with other power system management tools and platforms, such as SCADA system, ERP system, etc.

[0210] 11. Security and Authentication Unit:

[0211] User authentication module ensures that only authorized personnel can access system data and control monitoring processes.

[0212] The data encryption module encrypts the transmitted and stored data to protect the data from unauthorized access.

[0213] The system security monitoring module monitors the operation of the system to ensure the security and stability of the system.

[0214] It is known from common technical knowledge that the present invention can be implemented by other embodiments that do not deviate from its spirit or essential features. Therefore, the above disclosed embodiments are only illustrative in all respects and are not exclusive. All changes within the scope of the present invention or within the scope equivalent to the present invention are included in the present invention.

Claims

1. A collaborative enhanced electrical equipment insulation status monitoring method, characterized in that: The following steps are included: Using a high-resolution camera to capture images of the surface of the electrical equipment to obtain image data; According to the image data, the preset recognition algorithm is used to identify and analyze the discharge traces and corrosion conditions to obtain the analysis results; If the analysis results show that there are no obvious traces, then a covering monitoring step is taken: a comprehensive electrical parameter test is conducted on the electrical equipment to obtain the comprehensive electrical parameters and physical status information of the equipment and obtain a comprehensive monitoring result; If the analysis results show that there are obvious traces, the key monitoring step is entered: based on the recorded trace positions and according to the analysis results, targeted electrical parameter detection is performed on the relevant areas to obtain key monitoring results; The analysis results, comprehensive monitoring results, and key monitoring results are deeply integrated to obtain a comprehensive assessment of the insulation status of electrical equipment.

2. A method for monitoring insulation status of synergistically enhanced electrical equipment according to claim 1, characterized in that: The electrical parameter detection includes but is not limited to ultrasonic detection and partial discharge detection.

3. A method for monitoring insulation status of synergistically enhanced electrical equipment according to claim 1, characterized in that: It also includes a correlation extension detection step: after the key detection step is completed, an extended detection is performed on the equipment area related to the trace position to evaluate potential insulation state changes and obtain correlation detection results.

4. A method for monitoring insulation status of synergistically enhanced electrical equipment according to claim 1, characterized in that: According to the image data, a preset recognition algorithm is used to identify and analyze the discharge traces and corrosion conditions, and the steps of obtaining the analysis results include: The image data captured using a high-resolution camera is input into a preset recognition algorithm; Perform noise removal and contrast enhancement on the input image data to obtain a preprocessed image; The convolutional neural network of the YOLOv5 algorithm is used to extract features from the preprocessed images to obtain a series of feature maps; Through the feature pyramid network structure, the feature maps of different levels in the YOLOv5 algorithm are fused to obtain a fused feature map; Apply sparse coding technology to the obtained fusion feature map to extract key features; Use recursive neural networks to recursively process the extracted key features in time series; The recursively processed features are input into the classifier for final identification of discharge traces and corrosion conditions; The final identification results are output to form analysis results of discharge traces and corrosion conditions, which serve as a basis for further detection of the insulation status of electrical equipment.

5. The method for monitoring insulation status of a synergistically enhanced electrical device according to claim 1, characterized in that: The covering monitoring step specifically includes: Based on the layout of electrical equipment and historical fault data, the equipment is divided into multiple monitoring areas, each of which corresponds to specific monitoring parameters and methods; Develop a monitoring plan for each monitoring area, including monitoring frequency, detection technology to be used, and expected monitoring results; According to the monitoring plan, ultrasonic detection and partial discharge detection technology are used to collect electrical parameters in each monitoring area to obtain original electrical parameter data; Filtering, denoising and normalizing the obtained raw electrical parameter data to obtain pre-processed data; Analyze the preprocessed data using a preset machine learning algorithm to identify abnormal patterns or trends and obtain abnormal detection results. The machine learning algorithm is trained based on historical monitoring data and is used to identify parameter changes that deviate from normal operating conditions; Combined with the abnormal detection results, a comprehensive status assessment is conducted on the electrical equipment in each monitoring area. The assessment results include the health index of the equipment and the potential risk prediction; The evaluation results are synthesized to form a comprehensive monitoring report of the equipment, recording the status of each monitoring area and providing maintenance recommendations and risk warnings.

6. A method for monitoring insulation status of synergistically enhanced electrical equipment according to claim 1, characterized in that: The key monitoring steps specifically include: After the discharge traces are identified in the image analysis step, the positions and features of the traces are confirmed, and the confirmed information is used as input data for key monitoring; According to the confirmed trace location, select the monitoring point on the electrical equipment corresponding to the trace; Based on the characteristics and historical data of the monitoring points, develop customized testing plans for each monitoring point, including testing parameters, testing methods and testing cycles; Perform synchronous multi-parameter detection on each monitoring point, including but not limited to partial discharge, insulation resistance, temperature and humidity, to obtain comprehensive electrical parameter data; Synchronize and fuse the comprehensive electrical parameter data obtained to obtain fused data; Perform trend analysis on the fused data to identify potential insulation degradation trends and trigger an early warning mechanism when the trend analysis results exceed the preset threshold; Based on the trend analysis results, targeted maintenance decision support is provided, including suggestions on maintenance timing, maintenance scope and maintenance methods.

7. A method for monitoring insulation status of synergistically enhanced electrical equipment according to claim 1, characterized in that: The steps of deeply integrating the analysis results, comprehensive monitoring results, and key monitoring results to obtain a comprehensive assessment of the insulation status of electrical equipment include: Integrate the image analysis results, data from comprehensive monitoring reports, and results from key monitoring reports to form a unified data set; Verify the consistency of the integrated data set to ensure that the temporal and spatial correspondence of data from different sources is accurate and obtain a verified data set; Using the verified data set, the features of different monitoring results are mapped into a unified reference framework to obtain a mapped feature set; According to the importance and reliability of each monitoring result, weights are assigned to the mapped feature set to obtain a weighted feature set; Based on the weighted feature set, the insulation status of the electrical equipment is calculated to provide a comprehensive score reflecting the insulation health level of the equipment; Based on the weighted feature set, analyze the changing trend of the insulation status of electrical equipment over time and obtain a trend analysis report; Based on the trend analysis report and comprehensive score, combined with maintenance priorities and resource allocation, specific maintenance decisions and action plans are made to obtain a maintenance decision report.

8. A method for monitoring insulation status of synergistically enhanced electrical equipment according to claim 6, characterized in that: According to the confirmed trace location, the steps of selecting the monitoring point on the electrical equipment corresponding to the trace include: Convert the identified discharge trace locations into digital coordinates of the electrical device; Map the physical layout of electrical equipment to a digital coordinate system; According to the digital coordinates of the trace, the monitoring points on the electrical equipment directly related to the trace are screened out, and the monitoring points cover the trace and the surrounding areas that may be affected; Based on the characteristics of the traces and historical monitoring data, the selected monitoring points are optimized, redundant monitoring points are eliminated, and the layout of monitoring points that fully covers the trace area is determined; Conduct on-site calibration of selected monitoring points; Attach data tags to each monitoring point, including the number, location, associated trace features and historical monitoring records of the monitoring point.

9. A method for monitoring insulation status of synergistically enhanced electrical equipment according to claim 6, characterized in that: Perform trend analysis on the fused data to identify potential insulation degradation trends and trigger an early warning mechanism when the trend analysis results exceed the preset threshold, including: Based on the fused data, determine the statistical features reflecting the insulation status change for trend analysis, including but not limited to the mean value, standard deviation, skewness, and peak frequency; Compare current statistical characteristics with those of the same monitoring points in history to identify long-term trends in insulation status; Pre-train trend models based on historical data to predict future trends in insulation status; Apply pre-trained trend models to analyze real-time monitoring data, monitor real-time changes in insulation status, and generate trend reports; Compare the statistical features in the generated trend report with the warning threshold to determine whether there is potential insulation degradation. The warning threshold is set based on historical data and expert experience; If the trend analysis results exceed the preset threshold, an early warning signal is automatically generated to indicate insulation degradation problems; Based on the warning signals, formulate corresponding warning response measures, including but not limited to immediate maintenance recommendations, further monitoring requirements or urgent safety measures.

10. A synergistically enhanced electrical equipment insulation status monitoring system, based on a synergistically enhanced electrical equipment insulation status monitoring method according to any one of claims 1 to 9, characterized in that: include, An image capturing unit is used to capture an image of the surface of the electrical device using a high-resolution camera to obtain image data. The image analysis unit is used to identify and analyze discharge traces and corrosion conditions based on image data using a preset recognition algorithm to obtain analysis results. The monitoring decision unit is used to determine whether to perform a coverage monitoring step or a key monitoring step according to the analysis result of the image analysis unit. The comprehensive monitoring execution unit is used to execute the covering monitoring steps under the instruction of the monitoring decision unit, conduct comprehensive electrical parameter detection on the electrical equipment, obtain the comprehensive electrical parameters and physical status information of the equipment, and obtain comprehensive monitoring results. The key monitoring execution unit is used to execute the key monitoring steps under the instruction of the monitoring decision unit, and to conduct targeted electrical parameter detection on the relevant areas based on the recorded trace positions and the analysis results to obtain the key monitoring results. The data fusion unit is used to deeply integrate the analysis results, comprehensive monitoring results, and key monitoring results to obtain a comprehensive assessment of the insulation status of electrical equipment.

Images