Power transmission line insulator use state monitoring system and method

Abstract

The embodiment of the invention discloses a power transmission line insulator use state monitoring system and method. According to the technical scheme provided by the embodiment of the invention, the vibration signal and the audio signal of the insulator are synchronously collected, fused and analyzed under the condition that the environment excitation condition is met, so that the system can firstly judge the state of a mechanical structure through vibration analysis and then recognize the discharge activity through audio waveform comparison; and finally, fault root cause diagnosis is realized according to the association relationship between the mechanical state and the discharge activity. Compared with an existing mode depending on single vibration detection and manual contact type re-checking, the method achieves non-contact type accurate diagnosis of fault root causes, improves the inspection efficiency, the operation safety and the maintenance decision accuracy, and meets the operation and maintenance requirements of a power grid for large-scale and preventive state monitoring of the power transmission line.

Description

Technical Field

[0001] This application relates to the field of power transmission line technology, and in particular to a system and method for monitoring the service status of power transmission line insulators. Background Technology

[0002] Currently, during the operation and maintenance of power transmission lines, it is frequently necessary to monitor the condition of insulators to ensure their normal operation. Since insulators are typically installed on high transmission towers, non-contact monitoring using drones equipped with sensors is commonly employed. A common method involves using drones equipped with high-definition visible light or infrared thermal imaging equipment to capture images of the insulators and analyze their appearance and temperature distribution to identify anomalies such as surface contamination, damage, or localized overheating. Another method uses vibration detection, which collects and analyzes the vibration signal waveforms, frequencies, and modal characteristics of the insulators to assess the integrity of their mechanical structure and determine the presence of defects such as cracks or loosening.

[0003] However, the aforementioned detection methods based on images or vibration signals primarily target the external defects and mechanical condition of insulators. They cannot directly and accurately detect and assess the insulation performance or determine the presence of abnormal discharge. When potential structural abnormalities are detected in the insulator, manual climbing of the tower or the use of specialized tools for contact-based partial discharge detection is still required to ultimately confirm the presence of an abnormal discharge. This detection method is not only inefficient and risky, but also difficult to implement for routine monitoring of large-scale lines. Furthermore, even if a discharge phenomenon is confirmed, it is impossible to effectively distinguish whether the discharge is caused by surface contamination or internal mechanical damage, thus affecting the accuracy of fault root cause assessment and maintenance decisions. Summary of the Invention

[0004] This application provides a system and method for monitoring the service status of insulators in transmission lines. It first determines the mechanical structure status through vibration analysis, then identifies discharge activity through audio waveform comparison, and finally diagnoses the root cause of faults based on the correlation between the mechanical status and discharge activity. This solves the technical problems of existing image-based or single-vibration detection methods, which cannot effectively identify abnormal discharges, let alone distinguish whether the discharge is caused by internal mechanical damage or surface contamination.

[0005] In a first aspect, embodiments of this application provide a power transmission line insulator usage status monitoring system, comprising: The acquisition module is used to acquire the vibration signal of the target insulator under environmental excitation when the current operating environment of the target insulator meets the environmental excitation conditions, and simultaneously acquire the audio signal of the target insulator. The identification module is used to determine the mechanical structure state of the target insulator based on the vibration signal, and to compare the audio waveform of the audio signal with the historical audio of the target insulator. If the audio waveform comparison results indicate that there is a set signal difference between the audio signal and the historical audio, the module determines the difference features between the audio signal and the historical audio, inputs the difference features into the pre-trained discharge identification model, and outputs the discharge anomaly identification result of the target insulator based on the discharge identification model. The output module is used to output the root cause of the discharge anomaly of the target insulator based on the mechanical structure status when the discharge anomaly identification result determines that the target insulator has a discharge anomaly. Specifically, if the mechanical structure status indicates that the target insulator has a structural anomaly, the root cause of the discharge anomaly is determined to be mechanical structural damage to the target insulator; if the mechanical structure status indicates that the target insulator does not have a structural anomaly, the root cause of the discharge anomaly is determined to be surface contamination of the target insulator.

[0006] In a second aspect, embodiments of this application provide a method for monitoring the service status of transmission line insulators, including: Under the condition that the current operating environment of the target insulator meets the environmental excitation conditions, the vibration signal of the target insulator under environmental excitation is collected, and the audio signal of the target insulator is collected simultaneously. The mechanical structure state of the target insulator is determined based on the vibration signal, and the audio waveform is compared with the historical audio of the target insulator based on the audio signal. If the audio waveform comparison results indicate that there is a set signal difference between the audio signal and the historical audio, the difference features between the audio signal and the historical audio are determined. The difference features are input into the pre-trained discharge recognition model, and the discharge anomaly recognition result of the target insulator is output based on the discharge recognition model. If the discharge anomaly is determined to exist in the target insulator based on the discharge anomaly identification results, the root cause of the discharge anomaly is determined according to the mechanical structure status. If the mechanical structure status indicates that the target insulator has a structural anomaly, the root cause of the discharge anomaly is determined to be mechanical structural damage to the target insulator. If the mechanical structure status indicates that the target insulator does not have a structural anomaly, the root cause of the discharge anomaly is determined to be surface contamination of the target insulator.

[0007] Furthermore, before acquiring the vibration signal of the target insulator under environmental excitation, the following steps are also included: Based on the ambient wind speed collected by the wind speed sensor, if the ambient wind speed is within the preset effective wind speed range, it is determined that the current operating environment of the target insulator meets the environmental excitation conditions.

[0008] Furthermore, the mechanical structural state of the target insulator is determined based on the vibration signal, including: Extract the real-time vibration characteristics of the vibration signal, which include the time-domain and / or frequency-domain characteristics of the vibration signal; The real-time vibration characteristics are compared with the historical vibration characteristics of the target insulator to obtain the first comparison result, and the real-time vibration characteristics are compared with the vibration anomaly database to obtain the second comparison result. The mechanical structure state of the target insulator is determined based on the first comparison result and the second comparison result.

[0009] Furthermore, audio waveform comparison is performed based on the audio signal and the historical audio of the target insulator, including: Obtain the waveform characteristics benchmark of the historical audio data of the target insulator; Calculate the feature difference between the waveform features of the audio signal and the waveform feature reference; If the feature difference reaches a set difference threshold, it is determined that there is a set signal difference between the audio signal and the historical audio.

[0010] Furthermore, the differences between the audio signal and historical audio are determined, and these differences are input into a pre-trained discharge recognition model. Based on the discharge recognition model, the discharge anomaly identification result of the target insulator is output, including: Based on the feature difference degree, the difference features between the audio signal and the historical audio are constructed, and the difference features are input into the pre-trained discharge recognition model. The discharge recognition model is pre-trained based on the audio sample data of insulators with known states. The discharge identification model processes and analyzes the differential features of the input, and outputs the corresponding discharge anomaly identification results, which include the classification type and its corresponding confidence level.

[0011] Furthermore, the classification types include partial discharge anomalies; after outputting the root cause of the discharge anomaly of the target insulator based on the mechanical structure condition, it also includes: If the classification type is partial discharge anomaly and the mechanical structure condition indicates that the target insulator has a structural anomaly, then the root cause of the discharge anomaly is verified to be mechanical structural damage to the target insulator.

[0012] Furthermore, the classification types include corona discharge anomalies; after outputting the root cause of the discharge anomaly of the target insulator based on the mechanical structure condition, it also includes: If the classification type is corona discharge anomaly, and the mechanical structure condition indicates that the target insulator does not have structural abnormalities, then the root cause of the discharge anomaly is verified to be surface contamination of the target insulator.

[0013] In a third aspect, embodiments of this application provide an electronic device, including: Memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the transmission line insulator usage status monitoring method as described in the first aspect.

[0014] In a fourth aspect, embodiments of this application provide a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the transmission line insulator usage status monitoring method as described in the first aspect.

[0015] This embodiment of the application collects the vibration signal of the target insulator under environmental excitation when the current operating environment of the target insulator meets the environmental excitation conditions, and simultaneously collects the audio signal of the target insulator. Based on the vibration signal, the mechanical structure state of the target insulator is determined, and the audio waveform is compared with the historical audio of the target insulator based on the audio waveform comparison result. If the audio signal and the historical audio are found to have a set signal difference based on the audio waveform comparison result, the difference feature between the audio signal and the historical audio is determined, and the difference feature is input into a pre-trained discharge recognition model. Based on the discharge recognition model, the discharge anomaly recognition result of the target insulator is output. If the discharge anomaly recognition result determines that the target insulator has a discharge anomaly, the root cause of the discharge anomaly of the target insulator is determined according to the mechanical structure state. Specifically, if the mechanical structure state indicates that the target insulator has a structural anomaly, the root cause of the discharge anomaly is determined to be mechanical structural damage of the target insulator; if the mechanical structure state indicates that the target insulator does not have a structural anomaly, the root cause of the discharge anomaly is determined to be surface contamination of the target insulator. By employing the aforementioned technical means, and through the simultaneous acquisition and fusion analysis of vibration and audio signals from insulators under environmental excitation conditions, the system can first determine the mechanical structure status through vibration analysis, then identify discharge activity through audio waveform comparison, and finally achieve root cause diagnosis of faults based on the correlation between mechanical status and discharge activity. Compared with existing methods that rely on single vibration detection and manual contact verification, this application achieves non-contact and accurate diagnosis of fault roots, improving inspection efficiency, operational safety, and the accuracy of maintenance decisions, thus meeting the power grid's operational and maintenance needs for large-scale, preventative condition monitoring of transmission lines. Attached Figure Description

[0016] Figure 1 This is a flowchart of a method for monitoring the service status of transmission line insulators provided in Embodiment 1 of this application; Figure 2 This is a schematic diagram of the insulator status monitoring in Embodiment 1 of this application; Figure 3 This is a flowchart illustrating the determination of the mechanical structure state in Embodiment 1 of this application; Figure 4This is a flowchart illustrating the signal difference determination process in Embodiment 1 of this application; Figure 5 This is a flowchart of the discharge anomaly identification process in Embodiment 1 of this application; Figure 6 This is a schematic diagram of the structure of a power transmission line insulator usage status monitoring system provided in Embodiment 2 of this application; Figure 7 This is a schematic diagram of the structure of an electronic device provided in Embodiment 3 of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, specific embodiments of this application will be described in further detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely for explaining this application and not for limiting it. It should also be noted that, for ease of description, only the parts relevant to this application are shown in the drawings, not all of them. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe operations (or steps) as sequential processes, many of these operations can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but may also have additional steps not included in the drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0018] Example 1: Figure 1 A flowchart of a method for monitoring the service status of transmission line insulators according to Embodiment 1 of this application is provided. This method can be executed by a transmission line insulator service status monitoring device, which can be implemented through software and / or hardware. The device can consist of two or more physical entities, or it can be a single physical entity. Generally, this device can be an inspection device such as a drone.

[0019] The following description uses inspection equipment as the main body for implementing the method of monitoring the service status of transmission line insulators. (Refer to...) Figure 1 The specific methods for monitoring the service status of insulators on transmission lines include: S110. Under the condition that the current operating environment of the target insulator meets the environmental excitation conditions, the vibration signal of the target insulator under environmental excitation is collected, and the audio signal of the target insulator is collected simultaneously.

[0020] The inspection equipment of this application embodiment, during the inspection of power transmission lines, monitors the usage status of insulators on the power transmission lines by incorporating visual sensors, vibration sensors, audio sensors, and corresponding data processing units. (Refer to...) Figure 2 Taking drone A as an inspection device, the drone travels along the power transmission line and reaches the corresponding position of insulator B that needs to be monitored for its usage status. It maintains stable flight and performs usage status monitoring of insulator B based on the corresponding sensors carried by drone A.

[0021] Specifically, the insulator currently being monitored for its operational status is defined as the target insulator. During operational status detection, the inspection equipment determines whether the operating environment of the target insulator meets preset environmental excitation conditions. If the environmental excitation conditions are confirmed to be met, a synchronous data acquisition process is initiated. The inspection equipment controls its onboard vibration sensors (such as laser vibrometers) to collect vibration signals of the target insulator under environmental excitation (such as ambient wind) in a non-contact manner, while simultaneously controlling audio sensors (such as ultrasonic microphones) to synchronously collect sound signals generated by the target insulator during operation. The vibration signals and audio signals are consistent in the time dimension, thus providing a data foundation for analyzing the potential causal relationship between the mechanical and electrical states of the insulator.

[0022] Optionally, before acquiring the vibration signal of the target insulator under environmental excitation, the method further includes: Based on the ambient wind speed collected by the wind speed sensor, if the ambient wind speed is within the preset effective wind speed range, it is determined that the current operating environment of the target insulator meets the environmental excitation conditions.

[0023] The inspection equipment uses an integrated wind speed sensor to collect real-time ambient wind speed data at the location of the target insulator, determining whether the ambient wind speed falls within the corresponding effective wind speed range. This effective wind speed range is a pre-defined numerical interval based on engineering experiments and simulations. Its lower limit must ensure that the excitation energy of the natural wind on the insulator structure is sufficient to produce a vibration response that can be accurately detected by the sensor, while the upper limit must prevent excessive wind speed from causing severe, nonlinear random vibrations of the insulator or instability of the UAV platform, thereby affecting its characteristic frequency signal or making data acquisition impossible. It also avoids the introduction of significant vibration noise. By comparing the real-time wind speed with the preset range, only when the wind speed remains within this range for a certain duration (e.g., 10 seconds) is the environmental excitation condition deemed met, triggering subsequent commands to synchronously collect vibration and audio signals. This ensures the quality and consistency of the collected vibration signals, making the mechanical condition assessment based on vibration signal analysis reliable. It should be noted that since the insulators are installed on high-altitude towers and are exposed to natural wind fields for a long time, the ambient wind is transformed into a standardized excitation source that does not require additional excitation devices. This enables efficient and low-cost vibration signal acquisition, improves the accuracy and reliability of subsequent analysis, avoids inefficient or invalid data acquisition under invalid wind speeds, and optimizes the energy consumption and operational efficiency of the inspection equipment.

[0024] Furthermore, the inspection equipment can also actively apply a standardized excitation to collect vibration signals. For example, a drone can be equipped with a dedicated acoustic exciter (such as a sound wave emitting device of a specific frequency) or a micro-impact device. After arriving at the detection location, it can actively apply an excitation of known intensity and spectrum to the insulator and collect its vibration response. This application does not impose fixed restrictions on the specific vibration signal acquisition method, and will not elaborate further here.

[0025] S120. Determine the mechanical structure state of the target insulator based on the vibration signal, and compare the audio waveform with the historical audio of the target insulator based on the audio signal. If the audio waveform comparison results indicate that there is a set signal difference between the audio signal and the historical audio, determine the difference features between the audio signal and the historical audio, input the difference features into the pre-trained discharge recognition model, and output the discharge anomaly recognition result of the target insulator based on the discharge recognition model.

[0026] After acquiring vibration and audio signals, the inspection equipment first analyzes the vibration signals to determine the mechanical structural state of the target insulator. By preprocessing the vibration signals, characteristic parameters that characterize the structural dynamics are extracted. These parameters may include time-domain features (such as root mean square value and peak factor) and frequency-domain features (such as natural frequency, damping ratio, and mode shape). The mechanical structural state is determined by analyzing the extracted features in real time. For example, this feature is compared to a historical vibration characteristic benchmark, which represents the average historical vibration characteristics of the target insulator under normal conditions. If the real-time vibration characteristics of the current signal do not conform to this historical vibration characteristic benchmark, it indicates a mechanical structural abnormality; otherwise, it is considered normal.

[0027] Furthermore, for the acquired audio signal, the current acquired audio signal is compared with the historical normal audio of the target insulator, and the signal differences in the time and frequency domains are calculated. This signal difference serves as a preliminary screening; if the signal difference does not exceed a set threshold, the current audio state is considered normal, saving resources for subsequent deep computation; if it exceeds the threshold, it indicates a significant signal difference. At this point, the inspection equipment further extracts the specific difference features constituting this difference from the audio signal and inputs them into a pre-trained discharge identification model. This discharge identification model is pre-trained based on a large amount of known insulator audio sample data (normal, abnormal discharge, etc.), enabling it to analyze the input difference features and output a classification and identification result for the discharge abnormality along with its corresponding confidence level, thus obtaining the discharge abnormality identification result.

[0028] Optionally, refer to Figure 3 Determining the mechanical structural state of the target insulator based on vibration signals includes: S1201. Extract the real-time vibration characteristics of the vibration signal, including the time-domain and / or frequency-domain characteristics of the vibration signal. S1202. Compare the real-time vibration characteristics with the historical vibration characteristics of the target insulator to obtain the first comparison result, and compare the real-time vibration characteristics with the vibration anomaly database to obtain the second comparison result. S1203. Determine the mechanical structure state of the target insulator based on the first comparison result and the second comparison result.

[0029] The inspection equipment extracts features from the raw vibration signal to obtain real-time vibration characteristics including time-domain and / or frequency-domain features. These real-time vibration characteristics are then compared with historical vibration characteristic benchmarks for the target insulator. These benchmarks are statistically generated from past detection data under various healthy conditions (e.g., feature mean). By comparing these historical vibration characteristic benchmarks, abnormal trends deviating from the target insulator's historical normal levels can be detected, thus providing a direct visual indication of the target insulator's gradual deterioration.

[0030] Based on this, the inspection equipment also matches real-time vibration characteristics with a pre-established vibration anomaly database covering various typical fault modes. This database stores vibration characteristic information for various known defects in insulators. The deviation of real-time characteristics from historical benchmarks and the similarity to various anomaly patterns in the database are calculated as the first and second comparison results characterizing the mechanical structural state of the target insulator. A comprehensive judgment is then made by combining the first and second comparison results. For example, if the first comparison result shows that the deviation of real-time vibration characteristics from historical vibration characteristic benchmarks reaches a deviation threshold, and the second comparison result shows that the similarity to a certain anomaly pattern in the vibration anomaly database reaches a similarity threshold, then the target insulator is determined to have a mechanical structural anomaly. Conversely, if the first comparison result shows that the deviation of real-time vibration characteristics from historical vibration characteristic benchmarks is below the deviation threshold, and the second comparison result shows that the similarity to various anomaly patterns in the vibration anomaly database is below the similarity threshold, then the mechanical structural state of the target insulator is determined to be normal. Furthermore, if a single comparison result reaches the threshold index, the weighted threshold index can be compared based on the weighted fusion value of deviation and similarity. Thus, when the weighted fusion value exceeds the weighted threshold index, it can be determined that the target insulator has a mechanical structural abnormality.

[0031] By combining the historical trend comparison of target insulators with the comparison of known vibration anomaly databases, the comprehensiveness and reliability of mechanical condition detection are significantly improved, enabling timely detection of early hidden dangers that are developing slowly, while accurately identifying sudden typical structural faults.

[0032] Furthermore, referring to Figure 4 Audio waveform comparison is performed based on the audio signal and the historical audio of the target insulator, including: S1204. Obtain the waveform characteristic reference of the historical audio of the target insulator; S1205. Calculate the feature difference between the waveform features of the audio signal and the waveform feature reference. S1206. When the feature difference reaches the set difference threshold, determine that there is a set signal difference between the audio signal and the historical audio.

[0033] While determining the mechanical structure status, the inspection equipment simultaneously performs audio signal comparison and analysis. First, it acquires a historical audio waveform feature benchmark for the target insulator. This benchmark is obtained by repeatedly collecting audio signals from the insulator under known healthy conditions during normal operation and statistically analyzing their characteristic mean values. For example, a set of audio waveforms characterizing its acoustic steady-state features is extracted from historical normal audio data, and the mean value of its audio waveform features is calculated. The waveform features of the audio waveform may include the statistical average of pulse density in the time domain, the distribution parameters of waveform amplitude, and the energy proportion of specific sub-bands and the centroid of the spectrum in the frequency domain. By calculating the mean of these parameters, a multi-dimensional benchmark feature vector is constructed. After acquiring an audio signal, the inspection equipment performs the same feature extraction process to obtain the waveform feature vector of the current audio signal.

[0034] Subsequently, the feature difference between the current feature vector and the historical baseline feature vector is quantified by calculating Euclidean distance and cosine similarity. The feature difference reflects the degree of deviation of the current audio pattern from its own health baseline. This calculated feature difference is then compared to a pre-set difference threshold. The difference threshold is pre-determined through statistical distribution analysis of the difference in a large number of normal samples (e.g., taking the mean plus the corresponding standard deviation) or by combining engineering experience. Based on the comparison results, if the feature difference does not reach this threshold, the current audio state is determined to be normal and has no significant difference from the historical baseline, and the process can be terminated to save computing power; if the feature difference reaches the set threshold, a set signal difference is confirmed, thereby triggering the subsequent discharge identification and analysis process.

[0035] By quantitatively comparing real-time audio features with the historical health baseline of the target insulator, an efficient preliminary screening of changes in the acoustic state of the insulator is achieved. This effectively filters out normal situations with no obvious changes, saving computational resources and time for subsequent complex discharge pattern recognition models.

[0036] Then, refer to Figure 5 The system identifies the differences between the audio signal and historical audio data, inputs these differences into a pre-trained discharge identification model, and outputs the discharge anomaly identification results for the target insulator based on the model. These results include: S1207. Construct the difference features between the audio signal and historical audio based on the feature difference degree, and input the difference features into the pre-trained discharge recognition model. The discharge recognition model is pre-trained based on the audio sample data of insulators with known states. S1208. Based on the discharge identification model, the differential features of the input are processed and analyzed, and the corresponding discharge anomaly identification results are output. The discharge anomaly identification results include the classification type and its corresponding confidence level.

[0037] Once the audio signal is confirmed to be significantly different from the historical baseline through the aforementioned comparison, a difference feature vector is constructed based on this feature difference degree, and a pre-trained discharge recognition model is used to identify discharge anomalies.

[0038] To improve recognition efficiency, this application directly constructs a structured difference feature vector based on the calculated feature difference degree. The difference feature vector aims to characterize the deviation of the current signal from a healthy baseline from multiple dimensions. This difference feature vector may include: the time-domain pulse density change rate, i.e., the relative increase or decrease in the number of discharge pulses per unit time, reflecting the activity level of the discharge; the time-domain average amplitude offset, characterizing the overall trend of discharge pulse intensity changes; the spectral centroid offset, used to describe the overall movement of signal energy in the frequency distribution, indicating possible changes in the physical process of discharge; and the change in the energy proportion of a specific frequency band, which determines the spectral characteristics of the discharge by monitoring the relative enhancement or weakening of energy in specific frequency bands sensitive to discharge type (such as high-frequency bands characterizing partial discharge). By fusing the above multi-dimensional feature differences, a multi-dimensional difference feature vector is constructed.

[0039] Subsequently, the differential feature vector is input into a pre-trained discharge identification model. This model is a classifier built on a machine learning algorithm, trained offline using massive amounts of precisely labeled insulator audio sample data. The training samples cover known states such as normal and abnormal discharge. Furthermore, for abnormal discharge samples, it can be determined whether the abnormality is caused by corona discharge or partial discharge. Through model training, the model learns the distinguishing boundaries between different discharge types in the differential feature space. Later, in the model inference stage, the input differential feature vector is processed and analyzed to understand its internal nonlinear mapping information, thereby calculating the probability that the vector belongs to each discharge category. Finally, the model outputs the abnormal discharge identification result, which includes a clear classification type (e.g., normal or abnormal discharge) and a corresponding confidence level (a probability value between 0% and 100%), thus providing a reliable metric for subsequent decision-making.

[0040] S130. If the discharge anomaly is determined to exist in the target insulator based on the discharge anomaly identification result, the root cause of the discharge anomaly of the target insulator shall be determined according to the mechanical structure status. If the mechanical structure status indicates that the target insulator has a structural anomaly, the root cause of the discharge anomaly shall be determined to be mechanical structural damage to the target insulator. If the mechanical structure status indicates that the target insulator does not have a structural anomaly, the root cause of the discharge anomaly shall be determined to be surface contamination of the target insulator.

[0041] After obtaining the assessment results of the mechanical structure status and the identification results of the discharge anomaly, the inspection equipment uses these results to perform fusion diagnosis and root cause analysis of the discharge anomaly. It is understandable that, from the perspective of the insulator fault mechanism, the essence of the discharge anomaly is the failure of insulation performance. The main causes of insulation performance failure can be classified into two categories: one is damage to the internal mechanical structure of the insulator body (such as cracks or brittle fracture of the core rod), which causes internal electric field distortion and air gaps, directly triggering partial discharge; the other is surface contamination and moisture, forming a conductive layer that leads to uneven distribution of the electric field along the surface, triggering surface discharge or corona discharge. Therefore, this application can accurately distinguish between these two types of root causes of discharge anomalies based on the insulator's mechanical structure status combined with the discharge anomaly identification mechanism.

[0042] If the discharge identification model indicates abnormal discharge activity in the target insulator, a root cause analysis is performed. If the mechanical structure analysis also indicates structural abnormalities in the target insulator, the current discharge anomaly is determined to be caused by internal mechanical damage. Conversely, if the discharge identification model confirms an anomaly, but the mechanical structure analysis shows the target insulator is structurally intact and without abnormal features (i.e., vibration characteristics are consistent with healthy baselines), the discharge anomaly is determined to be surface discharge caused by contaminants deposited on the insulator surface forming conductive channels in a humid environment. If the discharge identification model indicates no abnormal discharge activity in the target insulator, the mechanical structure detection results are directly output without further analysis.

[0043] Ultimately, based on the above determinations, the inspection equipment can generate a structured diagnostic report containing insulator identification, mechanical condition, discharge type, root cause analysis, and confidence level. This report is then transmitted wirelessly to the back-end monitoring center for auxiliary decision-making in transmission line operation and maintenance. By introducing vibration and audio signal sharing, simultaneous detection and correlation analysis of insulator mechanical and insulation defects are achieved, improving the efficiency of transmission line condition monitoring and the accuracy of maintenance decisions.

[0044] Optionally, the classification type includes partial discharge anomalies; after outputting the root cause of the discharge anomaly of the target insulator based on the mechanical structure condition, it also includes: If the classification type is partial discharge anomaly and the mechanical structure condition indicates that the target insulator has a structural anomaly, then the root cause of the discharge anomaly is verified to be mechanical structural damage to the target insulator.

[0045] When the discharge identification model outputs a classification of partial discharge anomaly, and the mechanical structure analysis also indicates a structural anomaly in the target insulator, further verification logic can be executed. Partial discharge typically originates from microscopic air gaps, cracks, or other defects within the insulating medium (such as the core rod of a composite insulator or the interior of a porcelain insulator) or at the interface, leading to repetitive breakdown under a high electric field. The structural anomalies detected by vibration analysis (such as a decrease in natural frequency or an abnormal damping ratio) are direct manifestations of such internal mechanical damage (cracks, delamination) or loose connections in structural dynamics. Therefore, the simultaneous occurrence of partial discharge anomalies and mechanical structure anomalies mutually corroborates each other. This consistency check confirms that the root cause judgment of the discharge being caused by mechanical structure damage has high physical rationality and credibility, thus outputting a verified final diagnostic conclusion with higher confidence. Depending on actual needs, the verification result can also directly increase the confidence of the root cause of the discharge anomaly by adding a set percentage of confidence value to the original root cause analysis result, thus visually reflecting the final root cause analysis result of the discharge anomaly.

[0046] Optionally, the classification type includes corona discharge anomaly; after outputting the root cause of the discharge anomaly of the target insulator based on the mechanical structure condition, it also includes: If the classification type is corona discharge anomaly, and the mechanical structure condition indicates that the target insulator does not have structural abnormalities, then the root cause of the discharge anomaly is verified to be surface contamination of the target insulator.

[0047] When the discharge identification model outputs a classification of corona discharge anomaly, and the mechanical structure analysis indicates that the target insulator has no structural abnormalities, the root cause verification of the discharge anomaly is also performed. It is understandable that corona discharge is mainly caused by the concentration of electric field on the surface of electrodes or conductors. For example, under humid and polluted conditions, isolated water droplets or regions formed by conductive substances in the pollutant layer can lead to a surge in local field strength, causing air ionization. Since this process mainly occurs on the outer surface of the insulator, it is usually not accompanied by structural damage inside the insulator body, so the vibration characteristics may remain normal. Based on this characteristic, the signals of corona discharge anomalies and the vibration analysis results of normal structures can jointly point to non-mechanical factors outside the insulator, namely, surface contamination and moisture. This verifies the determination that the discharge is caused by surface contamination.

[0048] By introducing a cross-validation mechanism between discharge type and mechanical state, the diagnostic conclusions no longer rely on a single data source, thereby improving the accuracy and reliability of state judgment.

[0049] In one embodiment, when monitoring the operational status of insulators, the inspection equipment adaptively adjusts the analysis rules for vibration and audio signals by collecting real-time environmental parameters such as wind speed, wind direction, temperature, humidity, and insulator type, thereby further improving the accuracy of the analysis. For example, under conditions of low wind speed but high humidity, it can be determined that the current environmental wind excitation is insufficient, but high humidity may enhance the discharge audio signal of polluted insulators. Therefore, the weight of audio analysis can be increased, and a more sensitive vibration signal extraction algorithm can be adopted.

[0050] Optionally, this application can also perform synchronous time-frequency transformation on the synchronously acquired vibration and audio signals to obtain their time-frequency spectra. Considering that early mechanical loosening or microcracks may be weakly represented in vibration signals during insulator operation, but may be correlated in audio signals of specific frequency bands (such as friction sounds from loose parts), this application can train a deep learning network based on this principle. One input to the network is the vibration time-frequency spectrum, and the other is the audio time-frequency spectrum. The network learns the coupling correlation pattern of the two signals in the time-frequency domain. For example, it can identify the joint feature of a slight disturbance in the 100Hz vibration mode accompanied by an acoustic emission energy envelope with a duration of 8kHz, which corresponds to the abnormal condition of early insulator loosening. By learning this correlation, the model can accurately identify potential abnormalities in the insulator based on the two input time-frequency spectra, achieving more forward-looking monitoring of the insulator's operating status.

[0051] The above describes a process where, under conditions where the current operating environment of the target insulator meets environmental excitation requirements, vibration signals of the target insulator under environmental excitation are collected, along with audio signals. Based on the vibration signals, the mechanical structure state of the target insulator is determined. Furthermore, the audio waveforms of the audio signals are compared with historical audio data of the target insulator. If a predetermined signal difference is found between the audio signals and historical audio data based on the comparison results, the difference characteristics are identified and input into a pre-trained discharge identification model. Based on this model, the discharge anomaly identification result for the target insulator is output. If the discharge anomaly identification result indicates a discharge anomaly in the target insulator, the root cause of the discharge anomaly is determined based on the mechanical structure state. Specifically, if the mechanical structure state indicates a structural anomaly, the root cause is determined to be mechanical structural damage; if the mechanical structure state indicates no structural anomaly, the root cause is determined to be surface contamination of the target insulator. By employing the aforementioned technical means, and through the simultaneous acquisition and fusion analysis of vibration and audio signals from insulators under environmental excitation conditions, the system can first determine the mechanical structure status through vibration analysis, then identify discharge activity through audio waveform comparison, and finally achieve root cause diagnosis of faults based on the correlation between mechanical status and discharge activity. Compared with existing methods that rely on single vibration detection and manual contact verification, this application achieves non-contact and accurate diagnosis of fault roots, improving inspection efficiency, operational safety, and the accuracy of maintenance decisions, thus meeting the power grid's operational and maintenance needs for large-scale, preventative condition monitoring of transmission lines.

[0052] Example 2: Based on the above embodiments, Figure 6 This is a schematic diagram of a power transmission line insulator usage status monitoring system provided in Embodiment 2 of this application. (Reference) Figure 6 The transmission line insulator usage status monitoring system provided in this embodiment specifically includes: The acquisition module 21 is used to acquire the vibration signal of the target insulator under environmental excitation when the current operating environment of the target insulator meets the environmental excitation conditions, and to simultaneously acquire the audio signal of the target insulator. The identification module 22 is used to determine the mechanical structure state of the target insulator based on the vibration signal, and to compare the audio waveform with the historical audio of the target insulator based on the audio signal. If it is determined that there is a set signal difference between the audio signal and the historical audio based on the audio waveform comparison result, the difference features between the audio signal and the historical audio are determined, and the difference features are input into the pre-trained discharge identification model. Based on the discharge identification model, the discharge anomaly identification result of the target insulator is output. Output module 23 is used to output the root cause of the discharge anomaly of the target insulator according to the mechanical structure status when the discharge anomaly identification result determines that the target insulator has a discharge anomaly; wherein, if the mechanical structure status indicates that the target insulator has a structural anomaly, the root cause of the discharge anomaly is determined to be mechanical structural damage to the target insulator; if the mechanical structure status indicates that the target insulator does not have a structural anomaly, the root cause of the discharge anomaly is determined to be surface contamination of the target insulator.

[0053] Specifically, before collecting the vibration signal of the target insulator under environmental excitation, the following steps are also included: Based on the ambient wind speed collected by the wind speed sensor, if the ambient wind speed is within the preset effective wind speed range, it is determined that the current operating environment of the target insulator meets the environmental excitation conditions.

[0054] Specifically, determining the mechanical structural state of the target insulator based on vibration signals includes: Extract the real-time vibration characteristics of the vibration signal, which include the time-domain and / or frequency-domain characteristics of the vibration signal; The real-time vibration characteristics are compared with the historical vibration characteristics of the target insulator to obtain the first comparison result, and the real-time vibration characteristics are compared with the vibration anomaly database to obtain the second comparison result. The mechanical structure state of the target insulator is determined based on the first comparison result and the second comparison result.

[0055] Specifically, the audio waveform is compared with the historical audio of the target insulator based on the audio signal, including: Obtain the waveform characteristics benchmark of the historical audio data of the target insulator; Calculate the feature difference between the waveform features of the audio signal and the waveform feature reference; If the feature difference reaches a set difference threshold, it is determined that there is a set signal difference between the audio signal and the historical audio.

[0056] Specifically, the differences between the audio signal and historical audio are determined, and these differences are input into a pre-trained discharge identification model. Based on the discharge identification model, the discharge anomaly identification result of the target insulator is output, including: Based on the feature difference degree, the difference features between the audio signal and the historical audio are constructed, and the difference features are input into the pre-trained discharge recognition model. The discharge recognition model is pre-trained based on the audio sample data of insulators with known states. The discharge identification model processes and analyzes the differential features of the input, and outputs the corresponding discharge anomaly identification results, which include the classification type and its corresponding confidence level.

[0057] Specifically, the classification types include partial discharge anomalies; after outputting the root cause of the discharge anomaly of the target insulator based on the mechanical structure condition, it also includes: If the classification type is partial discharge anomaly and the mechanical structure condition indicates that the target insulator has a structural anomaly, then the root cause of the discharge anomaly is verified to be mechanical structural damage to the target insulator.

[0058] Specifically, the classification types include corona discharge anomalies; after outputting the root cause of the discharge anomaly of the target insulator based on the mechanical structure condition, it also includes: If the classification type is corona discharge anomaly, and the mechanical structure condition indicates that the target insulator does not have structural abnormalities, then the root cause of the discharge anomaly is verified to be surface contamination of the target insulator.

[0059] The above describes a process where, under conditions where the current operating environment of the target insulator meets environmental excitation requirements, vibration signals of the target insulator under environmental excitation are collected, along with audio signals. Based on the vibration signals, the mechanical structure state of the target insulator is determined. Furthermore, the audio waveforms of the audio signals are compared with historical audio data of the target insulator. If a predetermined signal difference is found between the audio signals and historical audio data based on the comparison results, the difference characteristics are identified and input into a pre-trained discharge identification model. Based on this model, the discharge anomaly identification result for the target insulator is output. If the discharge anomaly identification result indicates a discharge anomaly in the target insulator, the root cause of the discharge anomaly is determined based on the mechanical structure state. Specifically, if the mechanical structure state indicates a structural anomaly, the root cause is determined to be mechanical structural damage; if the mechanical structure state indicates no structural anomaly, the root cause is determined to be surface contamination of the target insulator. By employing the aforementioned technical means, and through the simultaneous acquisition and fusion analysis of vibration and audio signals from insulators under environmental excitation conditions, the system can first determine the mechanical structure status through vibration analysis, then identify discharge activity through audio waveform comparison, and finally achieve root cause diagnosis of faults based on the correlation between mechanical status and discharge activity. Compared with existing methods that rely on single vibration detection and manual contact verification, this application achieves non-contact and accurate diagnosis of fault roots, improving inspection efficiency, operational safety, and the accuracy of maintenance decisions, thus meeting the power grid's operational and maintenance needs for large-scale, preventative condition monitoring of transmission lines.

[0060] The transmission line insulator usage status monitoring system provided in Embodiment 2 of this application can be used to execute the transmission line insulator usage status monitoring method provided in Embodiment 1 above, and has corresponding functions and beneficial effects.

[0061] Example 3: This application provides an electronic device in embodiment three, referring to... Figure 7 The electronic device includes a processor 31, a memory 32, a communication module 33, an input device 34, and an output device 35. The electronic device may have one or more processors and one or more memories. The processor, memory, communication module, input device, and output device of the electronic device can be connected via a bus or other means.

[0062] Memory, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the transmission line insulator service status monitoring method described in any embodiment of this application (e.g., the acquisition module, identification module, and output module in the transmission line insulator service status monitoring system). Memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function; the data storage area may store data created based on the use of the device, etc. Furthermore, memory may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some instances, memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the device via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0063] The communication module is used for data transmission.

[0064] The processor executes various functional applications and data processing of the device by running software programs, instructions, and modules stored in memory, thereby realizing the above-mentioned method for monitoring the usage status of transmission line insulators.

[0065] Input devices can be used to receive input numerical or character information, and to generate key signal inputs related to user settings and function control of the device. Output devices may include display devices such as displays.

[0066] The electronic equipment provided above can be used to execute the transmission line insulator usage status monitoring method provided in Embodiment 1 above, and has corresponding functions and beneficial effects.

[0067] Example 4: This application embodiment also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to execute a method for monitoring the service status of transmission line insulators. This method includes: when the current operating environment of a target insulator meets environmental excitation conditions, collecting vibration signals of the target insulator under environmental excitation, and simultaneously collecting audio signals of the target insulator; determining the mechanical structure status of the target insulator based on the vibration signals, and comparing the audio waveforms of the audio signals with historical audio data of the target insulator; if the audio waveform comparison results indicate a set signal difference between the audio signals and historical audio data, determining the difference features between the audio signals and historical audio data, inputting the difference features into a pre-trained discharge identification model, and outputting a discharge anomaly identification result for the target insulator based on the discharge anomaly identification result; if the discharge anomaly identification result indicates a discharge anomaly in the target insulator, determining the root cause of the discharge anomaly based on the mechanical structure status; wherein, if the mechanical structure status indicates a structural anomaly in the target insulator, the root cause of the discharge anomaly is determined to be mechanical structural damage to the target insulator; if the mechanical structure status indicates no structural anomaly in the target insulator, the root cause of the discharge anomaly is determined to be surface contamination of the target insulator.

[0068] Storage medium – any type of memory device or storage device. The term “storage medium” is intended to include: mounting media, such as CD-ROM, floppy disk, or magnetic tape devices; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (e.g., hard disk or optical storage); registers or other similar types of memory elements, etc. Storage medium may also include other types of memory or combinations thereof. Furthermore, storage medium may reside in a first computer system in which the program is executed, or it may reside in a different second computer system connected to the first computer system via a network (such as the Internet). The second computer system can provide program instructions to the first computer for execution. The term “storage medium” can include two or more storage media residing in different locations (e.g., in different computer systems connected via a network). Storage medium may store program instructions (e.g., specifically implemented as a computer program) executable by one or more processors.

[0069] Of course, the computer-executable instructions provided in the embodiments of this application are not limited to the transmission line insulator usage status monitoring method described above, but can also execute related operations in the transmission line insulator usage status monitoring method provided in any embodiment of this application.

[0070] The transmission line insulator usage status monitoring system, storage medium, and electronic equipment provided in the above embodiments can execute the transmission line insulator usage status monitoring method provided in any embodiment of this application. For technical details not described in detail in the above embodiments, please refer to the transmission line insulator usage status monitoring method provided in any embodiment of this application.

[0071] The above description is merely a preferred embodiment and the technical principles employed in this application. This application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions that can be made by those skilled in the art will not depart from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of this application, the scope of which is determined by the scope of the claims.

Claims

1. A monitoring system for the service status of insulators in transmission lines, characterized in that, include: The acquisition module is used to acquire the vibration signal of the target insulator under environmental excitation when the current operating environment of the target insulator meets the environmental excitation conditions, and to simultaneously acquire the audio signal of the target insulator; The identification module is used to determine the mechanical structure state of the target insulator based on the vibration signal, and to compare the audio waveform of the audio signal with the historical audio of the target insulator. If the audio waveform comparison result determines that there is a set signal difference between the audio signal and the historical audio, the module determines the difference feature between the audio signal and the historical audio, inputs the difference feature into a pre-trained discharge identification model, and outputs the discharge anomaly identification result of the target insulator based on the discharge identification model. The output module is used to output the root cause of the discharge anomaly of the target insulator according to the mechanical structure state when the discharge anomaly identification result determines that the target insulator has a discharge anomaly; wherein, if the mechanical structure state indicates that the target insulator has a structural anomaly, the root cause of the discharge anomaly is determined to be mechanical structural damage to the target insulator; if the mechanical structure state indicates that the target insulator does not have a structural anomaly, the root cause of the discharge anomaly is determined to be surface contamination of the target insulator.

2. A method for monitoring the service status of insulators in transmission lines, characterized in that, include: Under the condition that the current operating environment of the target insulator meets the environmental excitation conditions, the vibration signal of the target insulator under environmental excitation is collected, and the audio signal of the target insulator is collected simultaneously. The mechanical structure state of the target insulator is determined based on the vibration signal, and the audio waveform is compared with the historical audio of the target insulator based on the audio signal. If the audio waveform comparison result indicates that there is a set signal difference between the audio signal and the historical audio, the difference feature between the audio signal and the historical audio is determined. The difference feature is input into the pre-trained discharge recognition model, and the discharge anomaly recognition result of the target insulator is output based on the discharge recognition model. If the discharge anomaly identification result determines that the target insulator has a discharge anomaly, the root cause of the discharge anomaly of the target insulator is determined according to the mechanical structure status; wherein, if the mechanical structure status indicates that the target insulator has a structural anomaly, the root cause of the discharge anomaly is determined to be mechanical structural damage to the target insulator; if the mechanical structure status indicates that the target insulator does not have a structural anomaly, the root cause of the discharge anomaly is determined to be surface contamination of the target insulator.

3. The method for monitoring the service status of transmission line insulators according to claim 2, characterized in that, Before acquiring the vibration signal of the target insulator under environmental excitation, the method further includes: Based on the ambient wind speed collected by the wind speed sensor, if the ambient wind speed is within the preset effective wind speed range, it is determined that the current operating environment of the target insulator meets the environmental excitation conditions.

4. The method for monitoring the service status of transmission line insulators according to claim 2, characterized in that, Determining the mechanical structural state of the target insulator based on the vibration signal includes: Extract the real-time vibration characteristics of the vibration signal, wherein the real-time vibration characteristics include the time-domain characteristics and / or frequency-domain characteristics of the vibration signal; The real-time vibration characteristics are compared with the historical vibration characteristics of the target insulator to obtain a first comparison result, and the real-time vibration characteristics are compared with the vibration anomaly database to obtain a second comparison result. The mechanical structure state of the target insulator is determined based on the first comparison result and the second comparison result.

5. The method for monitoring the service status of transmission line insulators according to claim 2, characterized in that, The audio waveform comparison based on the audio signal and the historical audio of the target insulator includes: Obtain the waveform feature reference of the historical audio of the target insulator; Calculate the feature difference degree between the waveform features of the audio signal and the waveform feature reference; If the feature difference reaches a set difference threshold, it is determined that the audio signal has a set signal difference from the historical audio.

6. The method for monitoring the service status of transmission line insulators according to claim 5, characterized in that, The step of determining the difference features between the audio signal and the historical audio, inputting the difference features into a pre-trained discharge recognition model, and outputting the discharge anomaly recognition result of the target insulator based on the discharge recognition model includes: Based on the feature difference degree, the difference features between the audio signal and the historical audio are constructed, and the difference features are input into the pre-trained discharge recognition model. The discharge recognition model is pre-trained based on insulator audio sample data with known states. Based on the discharge identification model, the input differential features are processed and analyzed, and the corresponding discharge anomaly identification results are output. The discharge anomaly identification results include the classification type and its corresponding confidence level.

7. The method for monitoring the service status of transmission line insulators according to claim 6, characterized in that, The classification types include partial discharge anomalies; After outputting the root cause of the discharge anomaly of the target insulator based on the mechanical structure state, the method further includes: If the classification type is partial discharge anomaly and the mechanical structure condition indicates that the target insulator has a structural anomaly, then the root cause of the discharge anomaly is verified to be mechanical structural damage to the target insulator.

8. The method for monitoring the service status of transmission line insulators according to claim 6, characterized in that, The classification types include corona discharge anomalies; After outputting the root cause of the discharge anomaly of the target insulator based on the mechanical structure state, the method further includes: If the classification type is corona discharge abnormality, and the mechanical structure status indicates that the target insulator does not have structural abnormalities, then the root cause of the discharge abnormality is verified to be surface contamination of the target insulator.

9. An electronic device, characterized in that, include: Memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the transmission line insulator usage status monitoring method as described in any one of claims 2-8.

10. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform the method for monitoring the service status of transmission line insulators as described in any one of claims 2-8.

Images