Power transmission line fault hidden danger integrated monitoring method and device

Abstract

The invention provides a power transmission line fault hidden danger integrated monitoring method and device, and relates to the technical field of power transmission line monitoring, and the method comprises the steps: collecting traveling wave related parameters and monitoring a traveling wave signal when a first monitoring period is satisfied; carrying out statistics on baseline traveling wave signals of the fault-free line samples meeting the traveling wave related parameters, and calculating a deviation vector; the fault triggering probability of the power transmission line meeting the deviation vector is counted; and when the fault triggering probability is greater than or equal to a fault probability threshold, retrieving a high-frequency fault detection duration of the power transmission line meeting the deviation vector, shortening the high-frequency fault detection duration by at least 0.5 times to obtain a second monitoring period, and replacing the first monitoring period with the second monitoring period. According to the invention, the problem that the sampling frequency is difficult to equalize and configure in the prior art can be solved, the objective of equalize configuration of the sampling frequency is achieved, and the technical effect of improving the power transmission line monitoring efficiency is achieved.

Description

Technical Field

[0001] This application relates to the technical field of transmission line monitoring, and particularly to an integrated monitoring method and device for potential faults in transmission lines. Background Art

[0002] In the field of transmission line fault monitoring, traditional monitoring methods mainly rely on the traveling wave theory. The traveling wave theory realizes rapid early warning and precise positioning of faults by detecting and analyzing the propagation characteristics of traveling waves in transmission lines. In practical applications, the setting of the sampling frequency has a direct impact on the accuracy of fault detection and the amount of calculation. The higher the sampling frequency, the more details of the traveling wave can be captured, thereby improving the accuracy of fault detection. However, a high sampling frequency also means a larger amount of calculation, which poses higher requirements for both the hardware and software of the monitoring system. On the other hand, if the sampling frequency is too low, some fault situations may be missed, thus affecting the reliability of the entire monitoring system. Therefore, how to reasonably configure the sampling frequency to reduce the amount of calculation and avoid resource waste while ensuring the accuracy of fault detection has become an urgent problem to be solved in the current transmission line fault monitoring technology.

[0003] Currently, the existing traditional transmission line potential fault monitoring technologies have deficiencies in sampling frequency configuration. A high sampling frequency leads to an excessive amount of calculation, while a low sampling frequency may miss fault situations, and there is a technical problem that it is difficult to achieve an equilibrium configuration of the sampling frequency. Summary of the Invention

[0004] The purpose of this application is to provide an integrated monitoring method for potential faults in transmission lines to solve the technical problems existing in the existing traditional transmission line potential fault monitoring technologies in sampling frequency configuration, where a high sampling frequency leads to an excessive amount of calculation, while a low sampling frequency may miss fault situations, and it is difficult to achieve an equilibrium configuration of the sampling frequency.

[0005] This application provides an integrated monitoring method for potential faults in transmission lines. Among them, the integrated monitoring method for potential faults in transmission lines includes: when the first monitoring period is satisfied, collecting traveling wave-related parameters and monitoring traveling wave signals; statistically analyzing the baseline traveling wave signals of fault-free line samples that meet the traveling wave-related parameters, and calculating the deviation vector between the baseline traveling wave signals and the monitored traveling wave signals; statistically analyzing the fault trigger probability of the transmission lines that meet the deviation vector; when the fault trigger probability is greater than or equal to the fault probability threshold, retrieving the high-frequency fault detection duration of the transmission lines that meet the deviation vector, obtaining a second monitoring period after shortening the high-frequency fault detection duration by at least 0.5 times, and using the second monitoring period to replace the first monitoring period.

[0006] The present application also provides an integrated monitoring device for hidden dangers of transmission line faults. Among them, the integrated monitoring device for hidden dangers of transmission line faults includes: a collection and monitoring module, which is used to collect traveling wave related parameters and monitor traveling wave signals when the first monitoring period is satisfied; a deviation vector calculation module, which is used to count the baseline traveling wave signals of fault-free line samples that meet the traveling wave related parameters and calculate the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal; a statistics module, which is used to count the fault trigger probability of the transmission line that meets the deviation vector; an update module, which is used to retrieve the high-frequency fault detection duration of the transmission line that meets the deviation vector when the fault trigger probability is greater than or equal to the fault probability threshold, obtain a second monitoring period after shortening the high-frequency fault detection duration by at least 0.5 times, and use the second monitoring period to replace the first monitoring period.

[0007] One or more technical solutions provided in the present application have at least the following technical effects or advantages: By collecting traveling wave related parameters and monitoring traveling wave signals when the first monitoring period is satisfied; counting the baseline traveling wave signals of fault-free line samples that meet the traveling wave related parameters and calculating the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal; counting the fault trigger probability of the transmission line that meets the deviation vector; when the fault trigger probability is greater than or equal to the fault probability threshold, retrieving the high-frequency fault detection duration of the transmission line that meets the deviation vector, obtaining a second monitoring period after shortening the high-frequency fault detection duration by at least 0.5 times, and using the second monitoring period to replace the first monitoring period. That is to say, when the first monitoring period is satisfied, traveling wave related parameters and traveling wave signals are collected, the baseline traveling wave signals of fault-free line samples are counted and the deviation vector is calculated, and then the fault trigger probability is counted. When the fault trigger probability reaches the threshold, the first monitoring period is updated according to the second monitoring period obtained by shortening the high-frequency fault detection duration. By dynamically adjusting the monitoring period according to the actual operating state of the transmission line, the balanced configuration of the sampling frequency is realized, and while ensuring the accuracy of fault detection, the calculation amount is effectively reduced, resource waste is avoided, and the technical effect of improving the monitoring efficiency of hidden dangers of transmission line faults, timely discovering potential faults, and ensuring the stable operation of the power system is achieved.

[0008] The above description is only an overview of the technical solutions of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specific embodiments of the present application are specifically given. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easily understood through the following description. Description of the Drawings

[0009] To more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only exemplary. For those of ordinary skill in the art, other drawings can be obtained based on the provided drawings without creative efforts.

[0010] Figure 1 It is a schematic flowchart of an integrated monitoring method for hidden dangers of transmission line faults in the present application; Figure 2 It is a schematic flowchart of collecting traveling wave related parameters and monitoring traveling wave signals in an integrated monitoring method for hidden dangers of transmission line faults in the present application. Detailed implementation manners

[0011] By providing an integrated monitoring method for hidden dangers of transmission line faults, the present application solves the technical problem that the existing traditional monitoring technology for hidden dangers of transmission line faults has deficiencies in sampling frequency configuration. High sampling frequency leads to excessive calculation amount, while low sampling frequency may miss fault situations, making it difficult to achieve balanced configuration of sampling frequency. By dynamically adjusting the monitoring period according to the actual operating state of the transmission line, the balanced configuration of sampling frequency is achieved, and while ensuring the accuracy of fault detection, the calculation amount is effectively reduced, avoiding waste of resources, and achieving the technical effect of improving the monitoring efficiency of hidden dangers of transmission line faults, timely discovering potential faults, and ensuring the stable operation of the power system.

[0012] Next, the technical solutions in the present application will be clearly and completely described with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described here. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application. Additionally, it should be noted that for the sake of description, only the parts related to the present application are shown in the drawings rather than all of them.

[0013] Embodiment 1. Please refer to the attached Figure 1 , the present application provides an integrated monitoring method for hidden dangers of transmission line faults. Among them, the integrated monitoring method for hidden dangers of transmission line faults specifically includes the following steps: Step P10: When the first monitoring period is satisfied, collect traveling wave related parameters and monitor traveling wave signals; Specifically, the sampling frequency configuration is optimized by dynamically adjusting the monitoring period. When the first monitoring period is satisfied, the traveling wave related parameters are collected and the traveling wave signal is monitored. The comprehensive consideration of these multi-dimensional parameters makes the fault monitoring more comprehensive and accurate. That is to say, first, when the first monitoring period is satisfied, the traveling wave related parameters are collected and the traveling wave signal is monitored. Among them, the traveling wave related parameters include parameters in multiple dimensions such as wire temperature, vibration, current, altitude, visible light image, infrared image, ambient temperature and humidity. These parameters reflect the operating state of the transmission line from different angles and provide comprehensive data support for subsequent fault monitoring. Then, the baseline traveling wave signal of the fault-free line samples that meet these traveling wave related parameters is statistically analyzed, and the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal is calculated. Next, the fault trigger probability of the transmission line that meets the deviation vector is statistically analyzed. If the fault trigger probability is greater than or equal to the fault probability threshold, the high-frequency fault detection duration of the transmission line that meets the deviation vector is retrieved and shortened by at least 0.5 times, so as to obtain the second monitoring period to update the first monitoring period. In a specific example, assume that the first monitoring period is 10 minutes, and the system collects the traveling wave related parameters and monitors the traveling wave signal within this period. Through statistical analysis, it is found that the fault trigger probability of some transmission lines under specific traveling wave parameters is relatively high, reaching or exceeding the set fault probability threshold. At this time, the high-frequency fault detection duration of these lines is retrieved, assumed to be 5 minutes, and shortened by at least 0.5 times, that is, adjusted to 2.5 minutes, so as to obtain a new second monitoring period to update the original 10-minute monitoring period. This can dynamically adjust the monitoring period according to the actual operating state and achieve a reasonable configuration of the sampling frequency.

[0014] To sum up, by dynamically adjusting the monitoring period, the optimization of the transmission line fault monitoring is realized. It can reduce the calculation amount and resource waste while ensuring the accuracy of fault detection, improve the monitoring efficiency, discover potential faults in time, and ensure the stable operation of the power system.

[0015] Step P20: Statistically analyze the baseline traveling wave signal of the fault-free line samples that meet the traveling wave related parameters, and calculate the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal; Specifically, by statistically analyzing the baseline traveling wave signal and calculating the deviation vector from the monitored traveling wave signal, it provides a basis for subsequent statistical analysis of fault triggering probability and monitoring period update. The core lies in using the baseline traveling wave signal of the fault-free line sample as a reference to accurately identify the abnormal changes in the currently monitored traveling wave signal, thereby realizing early warning and effective monitoring of potential faults in the transmission line, and improving the accuracy and reliability of fault monitoring. Specifically, first, the baseline traveling wave signals of the fault-free line samples that meet the traveling wave-related parameters are statistically analyzed. This process involves collecting and analyzing a large amount of fault-free line sample data to ensure that the baseline traveling wave signal can accurately reflect the traveling wave characteristics under normal operating conditions. Then, the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal is calculated. The calculation of the deviation vector is achieved by comparing the differences between the baseline traveling wave signal and the monitored traveling wave signal in various dimensions, which may include key characteristics such as signal amplitude, frequency, and phase. The magnitude and direction of the deviation vector can intuitively reflect the degree of deviation of the monitored traveling wave signal from the normal state, thus providing a quantitative basis for subsequent fault diagnosis.

[0016] In a specific example, assume that the average amplitude of the baseline traveling wave signal in a certain dimension is 100 units, while the amplitude of the monitored traveling wave signal in this dimension is 120 units. Then the deviation value in this dimension is 20 units. By calculating and integrating the deviations in all relevant dimensions, the overall deviation vector is obtained. If the deviation vector exceeds the preset threshold, it indicates that there is a significant difference between the monitored traveling wave signal and the baseline traveling wave signal, and there may be potential faults, which requires further analysis and processing.

[0017] In summary, by statistically analyzing the baseline traveling wave signal and calculating the deviation vector, accurate monitoring and abnormal identification of the traveling wave signal of the transmission line are achieved.

[0018] Step P30: Statistically analyze the fault triggering probability of the transmission line that meets the deviation vector; Specifically, by statistically analyzing the fault triggering probability of the transmission line that meets the deviation vector, it provides data support for subsequent monitoring period update. First, the fault triggering probability of the transmission line that meets the deviation vector is statistically analyzed. This process involves collecting, sorting, and analyzing a large amount of historical fault data and real-time monitoring data. Through data mining and statistical modeling methods, a mapping relationship between the deviation vector and the fault triggering probability is established. Then, based on the statistically obtained fault triggering probability, it is judged whether the monitoring period needs to be updated. If the fault triggering probability is greater than or equal to the fault probability threshold, the high-frequency fault detection duration of the transmission line that meets the deviation vector will be retrieved and shortened by at least 0.5 times to obtain a new monitoring period.

[0019] In a specific example, assume that through statistical analysis, it is found that when the value of the deviation vector in a certain dimension exceeds 20 units, the fault triggering probability of the transmission line significantly increases to 15%. If the set fault probability threshold is 10%, then at this time, it is considered that the current monitoring period is not sufficient to cope with the potential fault risk and needs to be updated. Retrieve the high-frequency fault detection duration of these transmission lines, assume it is 5 minutes, and shorten it by at least 0.5 times, that is, adjust it to 2.5 minutes, so as to obtain a new monitoring period to improve the response speed and monitoring efficiency for faults.

[0020] Generally speaking, by statistically analyzing the fault triggering probability of the transmission line, early warning and accurate monitoring of potential transmission line faults are realized.

[0021] Step P40: When the fault triggering probability is greater than or equal to the fault probability threshold, retrieve the high-frequency fault detection duration of the transmission line that meets the deviation vector, shorten the high-frequency fault detection duration by at least 0.5 times to obtain a second monitoring period, and use the second monitoring period to replace the first monitoring period.

[0022] Specifically, when the fault triggering probability reaches or exceeds the preset threshold, the high-frequency fault detection duration is retrieved and shortened to obtain a new monitoring period to update the original period. First, judge whether the fault triggering probability is greater than or equal to the fault probability threshold. If this condition is met, it means that the current transmission line is in a high fault risk state and the monitoring intensity needs to be strengthened. Then retrieve the high-frequency fault detection duration of the transmission line that meets the deviation vector. The high-frequency fault detection duration refers to the time required to quickly detect a fault at a high sampling frequency. Then shorten the retrieved high-frequency fault detection duration by at least 0.5 times to obtain a new second monitoring period, thereby updating the first monitoring period. In a specific example, assume that the first monitoring period is 10 minutes. During the monitoring process, it is found that the fault triggering probability reaches 15%, and the set fault probability threshold is 10%. At this time, the system will retrieve the high-frequency fault detection duration of the transmission line that meets the deviation vector, assume it is 5 minutes. According to the requirements of the scheme, shorten it by at least 0.5 times, that is, adjust it to 2.5 minutes, so as to obtain a new second monitoring period. This can monitor the transmission line more frequently, capture potential fault information in time, and improve the timeliness and accuracy of fault detection.

[0023] In summary, the monitoring frequency is dynamically adjusted according to the fault risk, so as to ensure the timeliness of fault detection while optimizing the utilization efficiency of monitoring resources, avoiding unnecessary resource waste, and enhancing the operation efficiency of the entire monitoring system. That is to say, when the fault risk is high, the monitoring cycle is automatically shortened and the monitoring intensity is enhanced to ensure that potential fault hazards are detected in a timely manner; while when the fault risk is low, the monitoring cycle is maintained or appropriately extended to avoid resource waste caused by over-monitoring. This flexible monitoring cycle adjustment mechanism not only improves the efficiency and reliability of fault monitoring, but also effectively reduces the overall monitoring cost and ensures the stable operation of the power system.

[0024] Furthermore, as shown in the appendix Figure 2 When the first monitoring cycle is satisfied, relevant parameters of traveling waves are collected and traveling wave signals are monitored, including: Obtain the monitoring attribute set and the traveling wave signal attribute set; Perform correlation analysis between the first traveling wave signal attribute of the traveling wave signal attribute set and the monitoring attribute set to obtain the first monitoring attribute correlation set; Until the Nth traveling wave signal attribute of the traveling wave signal attribute set is used to perform correlation analysis with the monitoring attribute set to obtain the Nth monitoring attribute correlation set; Extract the monitoring attributes whose first monitoring attribute correlation set to the Nth monitoring attribute correlation set is greater than or equal to the correlation threshold, and add them to the relevant parameters of the traveling wave.

[0025] Specifically, through the correlation analysis of the monitoring attribute set and the traveling wave signal attribute set, the monitoring attributes highly correlated with the traveling wave signal attributes are screened out and added to the relevant parameters of the traveling wave. The core lies in using the correlation analysis method to quantitatively evaluate the correlation between different attributes, so as to accurately construct the relevant parameters of the traveling wave and improve the accuracy and reliability of fault monitoring.

[0026] First, obtain the monitoring attribute set and the traveling wave signal attribute set. The monitoring attribute set covers various attributes that may affect the fault monitoring of transmission lines, such as wire temperature, vibration, current, altitude, visible light image, infrared image, environmental temperature and humidity, etc.; the traveling wave signal attribute set contains the characteristic attributes of traveling wave signals in different dimensions, such as the amplitude, frequency, phase, etc. of the signal. Then, according to the first traveling wave signal attribute in the traveling wave signal attribute set, perform a correlation analysis with the monitoring attribute set to obtain the first monitoring attribute correlation set. This process quantifies and evaluates the degree of association between them by calculating indicators such as the correlation coefficient or similarity between the two attributes. Then repeat this process until the correlation analysis is also completed for the Nth traveling wave signal attribute in the traveling wave signal attribute set to obtain the Nth monitoring attribute correlation set. During the correlation analysis process, filter out the monitoring attributes that are greater than or equal to the correlation threshold and add them to the traveling wave related parameters. The setting of the correlation threshold is to ensure that only the monitoring attributes highly correlated with the traveling wave signal attributes can be selected, thus ensuring the quality and effectiveness of the traveling wave related parameters.

[0027] In a specific example, assume that the monitoring attribute set contains 10 attributes and the traveling wave signal attribute set contains 5 attributes. The system first performs a correlation analysis on the first traveling wave signal attribute and the monitoring attribute set, and calculates to obtain the first monitoring attribute correlation set, in which the correlation degrees of 3 monitoring attributes may be greater than or equal to the set threshold of 0.6. Then, perform the same analysis on the second traveling wave signal attribute, and 2 new monitoring attributes may meet the conditions. And so on until the analysis of all 5 traveling wave signal attributes is completed. Finally, the system adds these monitoring attributes that meet the conditions to the traveling wave related parameters, so that the traveling wave related parameters can more comprehensively and accurately reflect the operating state of the transmission line.

[0028] In summary, through the correlation analysis of the monitoring attribute set and the traveling wave signal attribute set, the optimization construction of the traveling wave related parameters is realized. It can effectively improve the accuracy and reliability of fault monitoring, ensure that the traveling wave related parameters can accurately capture the fault characteristics of the transmission line, provide high-quality data support for subsequent fault detection and diagnosis, thus ensuring the stable operation of the power system and reducing the probability and loss of faults.

[0029] Further, according to the first traveling wave signal attribute of the traveling wave signal attribute set, perform a correlation analysis with the monitoring attribute set to obtain the first monitoring attribute correlation set, including: Collect a number of first traveling wave signal attribute anomaly logs where the first traveling wave signal attribute is abnormal. Among them, any one of the number of first traveling wave signal attribute anomaly logs includes an abnormal monitoring attribute set; Traverse the set of monitoring attributes, and count the proportion of trigger frequencies in the several first traveling wave signal attribute anomaly logs, which is set as the frequency correlation degree set; Extract the frequency correlation attribute set in which the frequency correlation degree is greater than or equal to the frequency correlation degree threshold from the set of monitoring attributes; Taking the fluctuation value of the first traveling wave signal attribute as the reference data sequence and the fluctuation value of the frequency correlation attribute set as the comparison data sequence, configure a grey correlation degree matrix for correlation degree analysis to obtain the first monitoring attribute correlation degree set.

[0030] Specifically, through the collection and analysis of the first traveling wave signal attribute anomaly logs, the monitoring attributes highly related to fault triggering are screened out and constructed into the first monitoring attribute correlation degree set. First, collect several logs of the first traveling wave signal attribute anomaly, and each log contains the set of anomaly monitoring attributes. Then traverse the set of monitoring attributes, and count the proportion of trigger frequencies of these attributes in the anomaly logs to form the frequency correlation degree set. Next, extract the attributes in the set of monitoring attributes whose frequency correlation degree is greater than or equal to the frequency correlation degree threshold to form the frequency correlation attribute set. Finally, taking the fluctuation value of the first traveling wave signal attribute as the reference data sequence and the fluctuation value of the frequency correlation attribute set as the comparison data sequence, configure a grey correlation degree matrix for correlation degree analysis to obtain the first monitoring attribute correlation degree set.

[0031] In a specific example, assume that 100 logs of the first traveling wave signal attribute anomaly are collected, and each log contains 10 monitoring attributes. Traverse the set of monitoring attributes, and count the proportion of trigger frequencies of each attribute in the anomaly logs. It is found that the trigger frequencies of 5 of these attributes exceed 10%, forming the frequency correlation attribute set. Then, taking the fluctuation value of the first traveling wave signal attribute as the reference and the fluctuation values of these 5 attributes as the comparison sequence, construct a grey correlation degree matrix, and calculate to obtain the first monitoring attribute correlation degree set. The attributes with higher correlation degree values will be selected as key monitoring parameters.

[0032] Generally speaking, through anomaly log analysis and correlation degree evaluation using methods such as frequency correlation degree and grey correlation degree matrix, the correlation degree between the monitoring attributes and the traveling wave signal attributes is quantitatively evaluated, so as to realize the accurate construction of traveling wave related parameters. Further, it can effectively screen out the monitoring attributes highly related to fault triggering, improve the accuracy and reliability of fault monitoring, reduce false alarms and missed alarms, and provide a solid guarantee for the stable operation of the power system.

[0033] Further, count the baseline traveling wave signals of the fault-free line samples that meet the traveling wave related parameters, and calculate the deviation vector between the baseline traveling wave signals and the monitored traveling wave signals, including: Retrieve multiple recorded traveling wave signals of multiple fault-free line samples where the topology of the first recorded line is the same as that of the transmission line and the parameters related to the first recorded traveling wave are consistent with the traveling wave related parameters; Perform median sorting of the same attributes on the multiple recorded traveling wave signals to obtain the concentration intervals of each attribute, which are set as the baseline traveling wave signals; Calculate the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal.

[0034] Specifically, by retrieving and performing median sorting of the same attributes on the recorded traveling wave signals of fault-free line samples, a baseline traveling wave signal that can reflect the normal operating state is constructed and compared with the monitored traveling wave signal for analysis, accurately identifying the abnormal state of the transmission line and providing a scientific basis for subsequent fault diagnosis and monitoring cycle adjustment.

[0035] First, retrieve multiple recorded traveling wave signals of multiple fault-free line samples where the topology of the first recorded line is the same as that of the transmission line and the parameters related to the first recorded traveling wave are consistent with the traveling wave related parameters. Then, perform median sorting of the same attributes on the multiple recorded traveling wave signals to obtain the concentration intervals of each attribute, which are set as the baseline traveling wave signals. Next, calculate the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal. In a specific example, assume that 100 recorded traveling wave signals of qualified fault-free line samples are retrieved, and each signal contains 5 key attributes. Perform median sorting of the same attributes on each attribute. For example, for the amplitude attribute, it is found that the amplitudes of most samples are concentrated in the range of 100±5 units, and this range is set as the amplitude concentration interval of the baseline traveling wave signal. After completing the median sorting of all attributes, a complete baseline traveling wave signal is obtained. Then, compare each attribute value of the monitored traveling wave signal with the corresponding concentration interval of the baseline traveling wave signal and calculate the deviation vector. For example, if the amplitude of the monitored traveling wave signal is 120 units, compared with the baseline concentration interval of 100±5 units, the deviation value is 15 units, exceeding the allowable range, indicating a significant deviation in the amplitude attribute and possibly predicting an abnormal state of the transmission line.

[0036] In summary, by generating the baseline traveling wave signal and calculating the deviation vector, accurate monitoring of the operating state of the transmission line is achieved. It can effectively identify abnormal changes in the transmission line, provide strong support for early warning and diagnosis of faults, improve the accuracy and reliability of fault monitoring, reduce the probability and loss of faults, and ensure the stable operation of the power system.

[0037] Furthermore, calculating the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal includes: Extract the first - type traveling - wave signal attributes and the second - type traveling - wave signal attributes of the traveling - wave signal. Among them, the first - type traveling - wave signal attributes represent the set of attributes whose correlation coefficient with the fault is greater than or equal to the correlation - coefficient threshold, and the second - type traveling - wave signal attributes represent the set of attributes whose correlation coefficient with the fault is less than the correlation - coefficient threshold; Perform the same - attribute deviation calculation on the first - type traveling - wave signal attributes to obtain the first - type traveling - wave signal deviation vector of the baseline traveling - wave signal and the monitored traveling - wave signal; Count the number of attributes of the second - type traveling - wave signal attributes that deviate from the baseline to obtain the second - type traveling - wave signal deviation vector of the baseline traveling - wave signal and the monitored traveling - wave signal; Add the first - type traveling - wave signal deviation vector and the second - type traveling - wave signal deviation vector to the deviation vector.

[0038] Specifically, through the classification and targeted analysis of the traveling - wave signal attributes, accurately capture the abnormal changes of the traveling - wave signal, providing a scientific basis for subsequent fault diagnosis and monitoring - cycle adjustment. First, extract the first - type traveling - wave signal attributes and the second - type traveling - wave signal attributes of the traveling - wave signal. The first - type traveling - wave signal attributes represent the set of attributes whose correlation coefficient with the fault is greater than or equal to the correlation - coefficient threshold. These attributes have a strong association with the occurrence of the fault and can directly reflect the fault characteristics; the second - type traveling - wave signal attributes represent the set of attributes whose correlation coefficient with the fault is less than the correlation - coefficient threshold. These attributes have a weak association with the fault, but still have certain reference value in the comprehensive evaluation. Then, perform the same - attribute deviation calculation on the first - type traveling - wave signal attributes to obtain the first - type traveling - wave signal deviation vector of the baseline traveling - wave signal and the monitored traveling - wave signal. Next, count the number of attributes of the second - type traveling - wave signal attributes that deviate from the baseline to obtain the second - type traveling - wave signal deviation vector of the baseline traveling - wave signal and the monitored traveling - wave signal. Finally, integrate the first - type traveling - wave signal deviation vector and the second - type traveling - wave signal deviation vector and add them to the total deviation vector.

[0039] In a specific example, assume that the correlation - coefficient threshold is set to 0.6, and 5 first - type traveling - wave signal attributes and 10 second - type traveling - wave signal attributes are extracted. For the first - type traveling - wave signal attributes, the attribute values of the baseline traveling - wave signal are [100, 200, 150, 50, 80] respectively, and the attribute values of the monitored traveling - wave signal are [120, 210, 160, 45, 90] respectively. Through the same - attribute deviation calculation, the first - type traveling - wave signal deviation vector is [20, 10, 10, 5, 10]. For the second - type traveling - wave signal attributes, it is statistically found that 4 attributes deviate from the baseline, so the second - type traveling - wave signal deviation vector is 4. Finally, integrate these two parts of the deviation vectors and add them to the total deviation vector to form a complete deviation - evaluation system.

[0040] In summary, through the deviation calculation method of classification and integration, the comprehensive monitoring of the operation status of transmission lines is realized. It can effectively capture the attribute changes highly related to faults, while taking into account the abnormal conditions of other attributes, improve the accuracy and reliability of fault monitoring, and reduce false alarms and missed alarms.

[0041] Furthermore, the fault triggering probability of the transmission line satisfying the deviation vector is statistically calculated, including: Obtain a fault triggering probability prediction network, where the fault triggering probability prediction network is generated based on machine learning training through multiple groups of data, and any one of the multiple groups of data includes: deviation vector record data of historical transmission lines with the same topology as the transmission line and labels indicating the true values of the fault triggering probability; Obtain the set of training deviation vectors in the preset time zone of the fault triggering probability prediction network; Calculate the Euclidean distance mean of the deviation vector and the set of training deviation vectors. According to the rule that the larger the Euclidean distance mean, the more the number of integrated models, perform output mean integration of multiple fault triggering probability prediction networks to obtain a temporary fault triggering probability prediction model. Process the deviation vector, output the fault triggering probability, and delete the temporary fault triggering probability prediction model at the same time.

[0042] Specifically, a fault triggering probability prediction network is constructed by machine learning methods, and the Euclidean distance mean and the integrated model rules are used to improve the accuracy and reliability of the prediction. First, obtain a fault triggering probability prediction network generated based on machine learning training through multiple groups of data, which includes deviation vector record data of historical transmission lines with the same topology as the transmission line and labels indicating the true values of the fault triggering probability. Then, obtain the set of training deviation vectors in the preset time zone of the prediction network. Next, calculate the Euclidean distance mean between the current deviation vector and the set of training deviation vectors. According to the rule that the larger the Euclidean distance mean, the more the number of integrated models, perform output mean integration of multiple fault triggering probability prediction networks to obtain a temporary fault triggering probability prediction model. Finally, use the temporary prediction model to process the deviation vector, output the fault triggering probability, and delete the temporary prediction model after the prediction is completed.

[0043] In a specific example, it is assumed that the fault trigger probability prediction network is trained based on historical data from the past 5 years, including thousands of different deviation vector record data and corresponding fault trigger probability labels. The set of training deviation vectors for the past month of this network is obtained, which contains 1000 deviation vector samples. The average Euclidean distance between the current deviation vector and these 1000 samples is calculated, and it is found that the average value is relatively large. According to the rule, a relatively large number of models need to be integrated. Thus, 5 different fault trigger probability prediction networks are integrated, and a temporary prediction model is generated by outputting the average value. The current deviation vector is processed using this model, and the fault trigger probability is obtained as 15%. Based on this result, it is decided whether to adjust the monitoring period. After the prediction is completed, the temporary prediction model is automatically deleted to save computing resources and avoid model overfitting.

[0044] In summary, by combining historical data to train and generate a prediction model, and dynamically adjusting and optimizing the model structure, accurate prediction of the fault trigger probability of transmission lines is achieved, providing a scientific basis for subsequent monitoring period adjustment and fault warning. At the same time, by dynamically adjusting the number of integrated models and timely deleting temporary models, the utilization of computing resources is optimized, and the operating efficiency of the system is improved.

[0045] Furthermore, according to the rule that the larger the average Euclidean distance, the more the number of integrated models, multiple fault trigger probability prediction networks are integrated by outputting the average value to obtain a temporary fault trigger probability prediction model, including: Calculate the square value of the average Euclidean distance, and set it as the number of integrated prediction networks; Integrate multiple fault trigger probability prediction networks according to the number of integrated prediction networks to obtain the temporary fault trigger probability prediction model.

[0046] Specifically, the number of integrated prediction networks is determined by calculating the square value of the average Euclidean distance, and multiple fault trigger probability prediction networks are integrated accordingly to obtain a temporary fault trigger probability prediction model. First, calculate the square value of the average Euclidean distance and set it as the number of integrated prediction networks. By quantitatively evaluating the difference between the current deviation vector and the set of training deviation vectors, the number of prediction networks to be integrated is determined. Then, according to the number of integrated prediction networks, multiple fault trigger probability prediction networks are integrated to obtain the temporary fault trigger probability prediction model. During the integration process, multiple prediction networks with a relatively high similarity in waveform signals to the current deviation vector are selected, and through a certain fusion strategy, such as outputting the average value, weighted average, etc., a comprehensive temporary prediction model is constructed. Finally, the current deviation vector is processed using this temporary prediction model, the fault trigger probability is output, and the temporary prediction model is deleted after the prediction task is completed to release computing resources and avoid model overfitting.

[0047] In a specific example, assume that the mean Euclidean distance between the current deviation vector and the set of training deviation vectors is 3, and its squared value is 9, that is, the number of prediction network integrations is 9. From multiple pre-trained fault trigger probability prediction networks, 9 networks with a relatively high similarity to the current deviation vector are selected for integration. These networks may be trained based on different historical data segments, different algorithms, or different parameters, each having unique prediction characteristics. The prediction results of these 9 networks are fused by taking the mean value of the outputs to obtain a temporary prediction model for the final fault trigger probability. Using this model to process the current deviation vector, a fault trigger probability of 18% is obtained, and based on this result, it is decided whether to adjust the monitoring period or take other warning measures. After the prediction is completed, the temporary prediction model is automatically deleted so that the most suitable prediction model can be reconstructed according to the new deviation vector during the next prediction. In summary, by dynamically determining the number of prediction network integrations and constructing a temporary prediction model, accurate prediction of the fault trigger probability of the transmission line is achieved.

[0048] Further, retrieving the high-frequency fault detection duration of the transmission line that satisfies the deviation vector includes: Retrieving the multiple fault detection durations of multiple fault line samples where the second recorded line topology is the same as the transmission line topology and the recorded deviation vector is the same as the deviation vector, where the fault detection duration represents the interval duration of the occurrence of a fault after the recorded deviation vector appears; Performing a central tendency analysis on the multiple fault detection durations to obtain the minimum value of the central fault detection duration, which is set as the high-frequency fault detection duration.

[0049] Specifically, by retrieving historical fault data that is consistent with the current transmission line topology and deviation vector and performing a central tendency analysis, the high-frequency fault detection duration is accurately determined. Using the statistical analysis of historical data provides a scientific basis for subsequent monitoring period adjustment, thereby optimizing the utilization efficiency of monitoring resources while ensuring the timeliness of fault detection and enhancing the operating efficiency of the entire monitoring system.

[0050] First, retrieve the fault detection durations of multiple fault line samples where the second recorded line topology is the same as the transmission line topology and the recorded deviation vector is consistent with the said deviation vector. The fault detection duration refers to the interval duration of the occurrence of a fault after the recorded deviation vector appears. This data can reflect the speed and possibility of fault development under specific operating conditions. Then, perform a central tendency analysis on the multiple fault detection durations to obtain the minimum value of the centralized fault detection duration, which is set as the high-frequency fault detection duration. Central tendency analysis usually includes calculating statistical indicators such as the mean, median, and mode. Through these indicators, the central tendency of the fault detection duration in most cases can be determined. Selecting the minimum value as the high-frequency fault detection duration is to adopt a more conservative and timely strategy in the monitoring period adjustment to ensure that potential faults can be detected as early as possible.

[0051] In a specific example, assume that the system retrieves 50 fault line samples that meet the conditions, and the fault detection durations are 2 minutes, 3 minutes, 3 minutes, 4 minutes, 5 minutes, etc. different durations. By performing a central tendency analysis on these data, the calculated mean is 3.2 minutes, the median is 3 minutes, and the mode is 3 minutes. According to the requirements of the scheme, select the minimum value of 2 minutes as the high-frequency fault detection duration. In this way, in the subsequent monitoring period adjustment, the new monitoring period will be determined based on this high-frequency fault detection duration, combined with other factors, to improve the response speed and monitoring efficiency for faults.

[0052] In summary, through the retrieval of historical fault data and central tendency analysis, the accurate determination of the high-frequency fault detection duration is achieved. It can effectively improve the timeliness and accuracy of fault monitoring, provide a scientific basis for the subsequent monitoring period adjustment, thus ensuring the stable operation of the power system, reducing the probability and losses of faults. At the same time, through the data-driven analysis method, the utilization of monitoring resources is optimized, the operation efficiency and adaptability of the system are improved, enabling it to better respond to the changes in the operating state of the transmission line.

[0053] In summary, the integrated monitoring method for hidden dangers of transmission line faults provided by this application has the following technical effects: When the first monitoring period is satisfied, collect traveling wave related parameters and monitor traveling wave signals; count the baseline traveling wave signals of the fault-free line samples that meet the traveling wave related parameters, and calculate the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal; count the fault trigger probability of the transmission line that meets the deviation vector; when the fault trigger probability is greater than or equal to the fault probability threshold, retrieve the high-frequency fault detection duration of the transmission line that meets the deviation vector, obtain the second monitoring period after shortening the high-frequency fault detection duration by at least 0.5 times, and use the second monitoring period to replace the first monitoring period. That is to say, when the first monitoring period is satisfied, collect traveling wave related parameters and monitor traveling wave signals, count the baseline traveling wave signals of the fault-free line samples and calculate the deviation vector, and then count the fault trigger probability. When the fault trigger probability reaches the threshold, update the first monitoring period according to the second monitoring period obtained by shortening the high-frequency fault detection duration. By dynamically adjusting the monitoring period according to the actual operating state of the transmission line, the balanced configuration of the sampling frequency is realized, and while ensuring the accuracy of fault detection, the calculation amount is effectively reduced, resource waste is avoided, and the technical effect of improving the monitoring efficiency of potential faults in the transmission line, timely discovering potential faults, and ensuring the stable operation of the power system is achieved.

[0054] Embodiment 2. Based on the same inventive concept as the integrated monitoring method for potential faults in a transmission line in the foregoing embodiment, the present application further provides an integrated monitoring device for potential faults in a transmission line. The integrated monitoring device for potential faults in a transmission line includes: An acquisition and monitoring module 11, which is used to collect traveling wave related parameters and monitor traveling wave signals when the first monitoring period is satisfied; A deviation vector calculation module 12, which is used to count the baseline traveling wave signals of the fault-free line samples that meet the traveling wave related parameters, and calculate the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal; A statistics module 13, which is used to count the fault trigger probability of the transmission line that meets the deviation vector; An update module 14, which is used to retrieve the high-frequency fault detection duration of the transmission line that meets the deviation vector when the fault trigger probability is greater than or equal to the fault probability threshold, obtain the second monitoring period after shortening the high-frequency fault detection duration by at least 0.5 times, and use the second monitoring period to replace the first monitoring period.

[0055] Further, the acquisition and monitoring module 11 is further used for: Obtain a monitoring attribute set and a traveling wave signal attribute set; Perform correlation analysis on the first traveling wave signal attribute of the traveling wave signal attribute set and the monitoring attribute set to obtain a first monitoring attribute correlation set; Until the Nth traveling wave signal attribute in the set of traveling wave signal attributes is subjected to a correlation analysis with the set of monitoring attributes to obtain the Nth monitoring attribute correlation set; Extract the monitoring attributes in the first monitoring attribute correlation set until the Nth monitoring attribute correlation set is greater than or equal to the correlation threshold, and add them to the traveling wave related parameters.

[0056] Furthermore, the acquisition and monitoring module 11 is further configured to: Acquire a number of first traveling wave signal attribute anomaly logs with abnormal first traveling wave signal attributes, where any one of the number of first traveling wave signal attribute anomaly logs includes an abnormal monitoring attribute set; Traverse the set of monitoring attributes, and count the proportion of trigger frequencies in the number of first traveling wave signal attribute anomaly logs, which is set as the frequency correlation set; Extract the frequency correlation attribute set in the set of monitoring attributes where the frequency correlation set is greater than or equal to the frequency correlation threshold; Using the fluctuation value of the first traveling wave signal attribute as the reference data sequence and the fluctuation value of the frequency correlation attribute set as the comparison data sequence, configure a grey correlation matrix for correlation analysis to obtain the first monitoring attribute correlation set.

[0057] Furthermore, the deviation vector calculation module 12 is further configured to: Retrieve multiple recorded traveling wave signals of multiple fault-free line samples where the first recorded line topology is the same as the transmission line topology and the first recorded traveling wave related parameters are the same as the traveling wave related parameters; Perform median sorting of the same attributes on the multiple recorded traveling wave signals to obtain the concentration interval of each attribute, which is set as the baseline traveling wave signal; Calculate the deviation vector between the baseline traveling wave signal and the monitored traveling wave signal.

[0058] Furthermore, the deviation vector calculation module 12 is further configured to: Extract the first type of traveling wave signal attributes and the second type of traveling wave signal attributes of the traveling wave signal, where the first type of traveling wave signal attributes represent the set of attributes with a correlation coefficient with the fault greater than or equal to the correlation coefficient threshold, and the second type of traveling wave signal attributes represent the set of attributes with a correlation coefficient with the fault less than the correlation coefficient threshold; Perform deviation calculation of the same attributes on the first type of traveling wave signal attributes to obtain the first type of traveling wave signal deviation vector between the baseline traveling wave signal and the monitored traveling wave signal; Count the number of attributes deviating from the baseline for the second type of traveling wave signal attributes to obtain the second type of traveling wave signal deviation vector between the baseline traveling wave signal and the monitored traveling wave signal; Add the first type of traveling wave signal deviation vector and the second type of traveling wave signal deviation vector into the deviation vector.

[0059] Further, the statistical module 13 is further configured to: Obtain a fault trigger probability prediction network, where the fault trigger probability prediction network is generated based on machine learning training through multiple groups of data, and any one of the multiple groups of data includes: deviation vector record data of a historical transmission line with the same topology as the transmission line and a label identifying the true value of the fault trigger probability; Obtain a set of training deviation vectors in a preset time zone of the fault trigger probability prediction network; Calculate the average Euclidean distance between the deviation vector and the set of training deviation vectors, and according to the rule that the larger the average Euclidean distance, the more the number of integrated models, output and average to integrate multiple fault trigger probability prediction networks to obtain a temporary fault trigger probability prediction model, process the deviation vector, output the fault trigger probability, and at the same time, delete the temporary fault trigger probability prediction model.

[0060] Further, the statistical module 13 is further configured to: Calculate the square value of the average Euclidean distance, and set it as the number of integrated prediction networks; Integrate multiple fault trigger probability prediction networks according to the number of integrated prediction networks to obtain the temporary fault trigger probability prediction model.

[0061] Further, the update module 14 is further configured to: Retrieve multiple fault detection durations of multiple fault line samples with the same topology as the transmission line and the same recorded deviation vector as the deviation vector, where the fault detection duration represents the interval duration of a fault occurring after the recorded deviation vector appears; Perform a central tendency analysis on the multiple fault detection durations to obtain the minimum value of the centralized fault detection duration, and set it as the high-frequency fault detection duration.

[0062] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0063] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of this application and its equivalent technologies, this application is also intended to cover these changes and modifications.

Claims

1. An integrated monitoring method for potential faults in transmission lines, characterized in that, An integrated monitoring terminal applied to a target transmission line, comprising: When the first monitoring period is satisfied, collect traveling wave related parameters and monitor traveling wave signals; Statistically analyze the baseline traveling wave signals of the fault-free line samples that meet the traveling wave related parameters, and calculate the deviation vector between the baseline traveling wave signals and the monitored traveling wave signals; Statistically analyze the fault trigger probability of the transmission lines that meet the deviation vector; When the fault trigger probability is greater than or equal to the fault probability threshold, retrieve the high-frequency fault detection duration of the transmission lines that meet the deviation vector, shorten the high-frequency fault detection duration by at least 0.5 times to obtain the second monitoring period, and use the second monitoring period to replace the first monitoring period.

2. The method according to claim 1, characterized in that, When the first monitoring period is satisfied, collect traveling wave related parameters and monitor traveling wave signals, including: Obtain a monitoring attribute set and a traveling wave signal attribute set; Perform correlation analysis on the first traveling wave signal attribute of the traveling wave signal attribute set and the monitoring attribute set to obtain a first monitoring attribute correlation set; Until the Nth traveling wave signal attribute of the traveling wave signal attribute set is used to perform correlation analysis on the monitoring attribute set to obtain the Nth monitoring attribute correlation set; Extract the monitoring attributes of the first monitoring attribute correlation set to the Nth monitoring attribute correlation set that are greater than or equal to the correlation threshold, and add them to the traveling wave related parameters.

3. The method according to claim 2, wherein Perform correlation analysis on the first traveling wave signal attribute of the traveling wave signal attribute set and the monitoring attribute set to obtain a first monitoring attribute correlation set, including: Collect a number of first traveling wave signal attribute anomaly logs where the first traveling wave signal attribute is abnormal. Any one of the number of first traveling wave signal attribute anomaly logs includes an abnormal monitoring attribute set; Traverse the monitoring attribute set, and statistically analyze the trigger frequency ratio in the number of first traveling wave signal attribute anomaly logs, and set it as the frequency correlation set; Extract the frequency correlation attribute set that is greater than or equal to the frequency correlation threshold from the monitoring attribute set in the frequency correlation set; Use the fluctuation value of the first traveling wave signal attribute as the reference data sequence, use the fluctuation value of the frequency correlation attribute set as the comparison data sequence, configure a grey correlation matrix for correlation analysis, and obtain the first monitoring attribute correlation set.

4. The method according to claim 1, characterized in that Statistically analyze the baseline traveling wave signals of the fault-free line samples that meet the traveling wave related parameters, and calculate the deviation vector between the baseline traveling wave signals and the monitored traveling wave signals, including: Retrieve multiple recorded traveling wave signals of multiple fault-free line samples where the first recorded line topology is the same as the transmission line topology and the first recorded traveling wave related parameters are consistent with the traveling wave related parameters; Perform median sorting of the same attributes on the multiple recorded traveling wave signals to obtain the concentration interval of each attribute, and set it as the baseline traveling wave signal; Calculate the deviation vector between the baseline traveling wave signals and the monitored traveling wave signals.

5. The method according to claim 4, wherein Calculate the deviation vector between the baseline traveling wave signals and the monitored traveling wave signals, including: Extract the first-class traveling wave signal attributes and the second-class traveling wave signal attributes of the traveling wave signal, where the first-class traveling wave signal attributes represent the set of attributes with a correlation coefficient with the fault greater than or equal to the correlation coefficient threshold, and the second-class traveling wave signal attributes represent the set of attributes with a correlation coefficient with the fault less than the correlation coefficient threshold; Perform the same-attribute deviation calculation on the first-class traveling wave signal attributes to obtain the first-class traveling wave signal deviation vector of the baseline traveling wave signal and the monitored traveling wave signal; Count the number of attributes deviating from the baseline for the second-class traveling wave signal attributes to obtain the second-class traveling wave signal deviation vector of the baseline traveling wave signal and the monitored traveling wave signal; Add the first-class traveling wave signal deviation vector and the second-class traveling wave signal deviation vector to the deviation vector.

6. The method according to claim 1, characterized in that, Count the fault trigger probability of the transmission line that satisfies the deviation vector, including: Obtain a fault trigger probability prediction network, where the fault trigger probability prediction network is generated based on machine learning training with multiple groups of data, and any one of the multiple groups of data includes: the deviation vector record data of the historical transmission line with the same topology as the transmission line and the label identifying the true value of the fault trigger probability; Obtain the training deviation vector set in the preset time zone of the fault trigger probability prediction network; Calculate the Euclidean distance mean of the deviation vector and the training deviation vector set, and according to the rule that the larger the Euclidean distance mean, the more the number of integrated models, output the mean to integrate multiple fault trigger probability prediction networks to obtain a temporary fault trigger probability prediction model, process the deviation vector, output the fault trigger probability, and delete the temporary fault trigger probability prediction model at the same time.

7. The method according to claim 6, characterized in that According to the rule that the larger the Euclidean distance mean, the more the number of integrated models, output the mean to integrate multiple fault trigger probability prediction networks to obtain a temporary fault trigger probability prediction model, including: Calculate the square value of the Euclidean distance mean and set it as the number of integrated prediction networks; Integrate multiple fault trigger probability prediction networks according to the number of integrated prediction networks to obtain the temporary fault trigger probability prediction model.

8. The method according to claim 1, characterized in that, Retrieve the high-frequency fault detection duration of the transmission line that satisfies the deviation vector, including: Retrieve the multiple fault detection durations of multiple fault line samples with the same topology as the transmission line in the second recorded line and the recorded deviation vector consistent with the deviation vector, where the fault detection duration represents the interval duration of the fault occurring after the recorded deviation vector appears; Perform a central tendency analysis on the multiple fault detection durations to obtain the minimum value of the centralized fault detection duration and set it as the high-frequency fault detection duration.

9. An integrated monitoring device for potential faults in transmission lines, characterized in that, The device is communicatively connected to the integrated monitoring terminal, and the device is used to execute the method according to any one of claims 1-8. The device includes: An acquisition monitoring module, which is used to acquire traveling wave related parameters and monitored traveling wave signals when the first monitoring period is satisfied; A deviation vector calculation module, which is used to count the baseline traveling wave signals of the fault-free line samples that satisfy the traveling wave related parameters and calculate the deviation vector of the baseline traveling wave signal and the monitored traveling wave signal; A statistical module, which is used to statistically calculate the fault triggering probability of a transmission line that meets the deviation vector; An update module, which is used to, when the fault triggering probability is greater than or equal to a fault probability threshold, retrieve the high-frequency fault detection duration of a transmission line that meets the deviation vector, obtain a second monitoring period after shortening the high-frequency fault detection duration by at least 0.5 times, and use the second monitoring period to replace the first monitoring period.

Images