Ring main unit state monitoring and early warning method based on edge calculation
By constructing a transient pulse reference library and dynamically adjusting the frequency of edge devices, the problem of insufficient capture of partial discharge signal characteristics in ring main units was solved, frequency coordination was achieved, and the accuracy and reliability of monitoring and early warning were improved.
Patent Information
- Application Number
- CN202511748706.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-26
- Publication Date
- 2026-02-27
AI Technical Summary
In existing technologies, it is difficult to fully capture the various signal characteristics of partial discharge in ring main units using a single frequency, and there are conflicts when dynamically adjusting the main and auxiliary acquisition frequencies, which leads to a decrease in the accuracy of monitoring and early warning.
By constructing a transient pulse reference library, the main and auxiliary acquisition frequencies of edge devices are dynamically adjusted to ensure frequency coordination. The maximum tolerable frequency difference is determined by combining historical data analysis, and frequency adjustment is performed to adapt to different types of partial discharge.
It achieves comprehensive capture of partial discharge signals from ring main units, improves monitoring accuracy and reliability, avoids judgment errors caused by frequency imbalance, and enhances the accuracy and reliability of early warning.
Smart Images
Figure CN121578060A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of ring main unit monitoring, in particular to a ring main unit state monitoring and early warning method based on edge computing. BACKGROUND
[0002] In the ring main unit partial discharge monitoring and early warning, the "sensor + edge computing" architecture has become the mainstream, but the fixed main sampling frequency design often falls into the "monitoring blind area" dilemma - the transient pulse length of different types of partial discharge is significantly different, the corona discharge pulse can reach nanoseconds, and the surface discharge is mostly microseconds, and a single frequency cannot completely capture the characteristics of various signals.
[0003] In the prior art, the problem of single frequency not being able to completely capture the characteristics of various signals, the lack of dynamic adjustment of data acquisition frequency, and the new conflict caused by dynamic adjustment of data acquisition frequency, that is, the conflict between dynamic adjustment of main sampling frequency and main auxiliary sampling linkage, specifically, the system usually cooperates "main sampling + auxiliary sampling" - the main sampling captures high-frequency core signals such as very high frequency, and the auxiliary sampling synchronously collects environmental and operating parameters such as temperature and humidity, switch temperature, etc., to improve the monitoring and early warning accuracy through multi-dimensional joint analysis, and the cooperation of the two relies on strict time synchronization, but when the main sampling frequency changes dynamically, the auxiliary sampling fixed frequency cannot be adapted: if the auxiliary sampling is fixed, the main sampling high-frequency data and the auxiliary sampling parameters "mismatch", resulting in a decrease in the accuracy of monitoring and early warning.
[0004] Therefore, the application provides a ring main unit state monitoring and early warning method based on edge computing. SUMMARY
[0005] In order to make up for the deficiencies of the prior art and solve at least one technical problem proposed in the background art.
[0006] The technical scheme adopted by the application to solve the technical problem is: a ring main unit state monitoring and early warning method based on edge computing, comprising the following steps: Step S10: According to the historical partial discharge data of the ring main unit, the transient pulse reference length under different types of partial discharge is analyzed and determined, and is integrated into a pulse length reference library; Step S20: Obtain the main sampling frequency of each edge device when monitoring different types of partial discharge of the ring main unit, and respectively combine and compare with the pulse length reference library to determine the type of partial discharge that needs to adjust the main sampling frequency of the edge device; Step S30: For the type of partial discharge that needs to adjust the main sampling frequency of the edge device, according to the historical main and auxiliary sampling frequency difference data under different types of partial discharge and the historical discharge monitoring results of the ring main unit, the maximum tolerance frequency difference under different types of partial discharge monitoring is analyzed and determined; Step S40: According to the maximum tolerable frequency difference under different types of partial discharge monitoring, it is judged whether the edge device main acquisition frequency adjustment of the partial discharge meets the requirement of the main and auxiliary acquisition frequency difference under the partial discharge monitoring. If not, when the edge device main acquisition frequency adjustment is performed, the edge device auxiliary acquisition frequency is adjusted at the same time according to the maximum tolerable frequency difference under the partial discharge monitoring. Step S50: After the edge device main acquisition frequency and auxiliary acquisition frequency are adjusted, the ring network cabinet state monitoring and early warning are performed.
[0007] As a further scheme of the application: the historical partial discharge data of the ring network cabinet includes the type of each partial discharge occurring in the historical operation cycle of the ring network cabinet and the pulse signal duration generated when the partial discharge occurs.
[0008] As a further scheme of the application: the process of determining the transient pulse reference duration under different types of partial discharge is: Based on any type of partial discharge, the pulse signal duration generated at each partial discharge is extracted from the historical partial discharge data and integrated into a pulse signal duration set. The coefficient of variation of the pulse signal duration set is calculated to obtain a pulse signal duration stability value, and the stability of the pulse signal duration generated under the partial discharge is judged. If the stable signal is high, the pulse signal duration set is subjected to mean value operation to obtain the transient pulse reference time under the partial discharge. If the stable signal is low, the minimum pulse signal duration in the pulse signal duration set is selected as the transient pulse reference duration under the partial discharge.
[0009] As a further scheme of the application: the process of determining the partial discharge type that needs to be adjusted in the edge device main acquisition frequency is: Based on the transient pulse reference duration under different partial discharges in the pulse duration reference library, the target main acquisition frequency under different partial discharges is calculated. Based on any type of partial discharge, the main acquisition frequency of the current edge device for ring network cabinet partial discharge monitoring is obtained, which is recorded as the actual main acquisition frequency. If the actual main acquisition frequency is lower than the target main acquisition frequency, it indicates that the edge device main acquisition frequency adjustment is needed.
[0010] As a further scheme of the application: the historical main and auxiliary acquisition frequency difference data includes the main and auxiliary acquisition frequency difference when different types of partial discharge monitoring, i.e. the absolute frequency deviation between the main acquisition frequency of the main acquisition sensor and the auxiliary acquisition frequency of the auxiliary acquisition sensor. The historical discharge monitoring result includes whether the edge device can accurately monitor the partial discharge after the partial discharge occurs under different types of partial discharge monitoring.
[0011] As a further scheme of the present application: the process of analyzing and determining the maximum tolerable frequency difference under different types of partial discharge monitoring is: Based on any type of partial discharge, the difference between the main and auxiliary collection frequencies of multiple partial discharges is extracted, and based on any main and auxiliary collection frequency difference, the number of discharge monitoring compliance times under the main and auxiliary collection frequency difference is obtained. The main and auxiliary collection frequency differences are integrated to obtain a main and auxiliary collection frequency difference sequence, and the number of discharge monitoring compliance times under each main and auxiliary collection frequency difference is integrated to obtain a discharge monitoring compliance sequence. The Pearson correlation coefficient between the main and auxiliary collection frequency difference sequence and the discharge monitoring compliance sequence is calculated, and absolute value processing is performed to obtain a collection-monitoring correlation coefficient, and the correlation type between the main and auxiliary collection frequency difference and the partial discharge compliance is determined. If there is a linear correlation, the least squares method is used to fit the main and auxiliary collection frequency difference sequence and the discharge monitoring compliance sequence to obtain a collection-monitoring correlation model. If there is a nonlinear correlation, the main and auxiliary collection frequency difference sequence and the discharge monitoring compliance sequence are nonlinearly fitted, and the fitting model with the highest fitting goodness is selected as the collection-monitoring correlation model after fitting. The discharge monitoring compliance threshold is substituted into the collection-monitoring correlation model to obtain the maximum tolerable frequency difference under partial discharge monitoring.
[0012] As a further scheme of the present application: the acquisition method of the number of discharge monitoring compliance times is: Based on any main and auxiliary collection frequency difference, the number of times that the edge device can accurately monitor and judge the partial discharge under the main and auxiliary collection frequency difference is counted, and the number of times is proportionally calculated with the total number of partial discharge monitoring to obtain the number of discharge monitoring compliance times.
[0013] As a further scheme of the present application: the process of determining whether the main collection frequency adjustment of the edge device for partial discharge meets the requirement of the main and auxiliary collection frequency difference under partial discharge monitoring is: Based on any type of partial discharge, the auxiliary collection frequency under partial discharge monitoring is obtained, and the absolute difference between the auxiliary collection frequency and the target main collection frequency under partial discharge monitoring is calculated to obtain the main and auxiliary collection frequency difference after the main collection frequency adjustment. If the main and auxiliary collection frequency difference after the main collection frequency adjustment is greater than the maximum tolerable frequency difference under partial discharge monitoring, it indicates that the main collection frequency adjustment of the edge device for partial discharge does not meet the requirement of the main and auxiliary collection frequency difference under partial discharge monitoring.
[0014] As a further scheme of the present application: the process of simultaneously adjusting the auxiliary collection frequency of the edge device according to the maximum tolerable frequency difference under partial discharge monitoring is: The absolute difference between the main and auxiliary collection frequency difference after the main collection frequency adjustment and the maximum tolerable frequency difference under partial discharge monitoring is calculated to obtain the main and auxiliary collection frequency adjustment value. Sum the main auxiliary acquisition frequency adjustment value and the auxiliary acquisition frequency under partial discharge monitoring to obtain an auxiliary acquisition frequency target value, and adjust the edge device auxiliary acquisition frequency according to the auxiliary acquisition frequency target value.
[0015] As a further scheme of the present application: the process of ring net cabinet state monitoring and early warning is: The captured pulse signals collected according to the main acquisition frequency and the auxiliary acquisition frequency are respectively subjected to feature extraction, and the main acquisition feature group and the auxiliary acquisition feature group are obtained by integration; The main acquisition feature group and the auxiliary acquisition feature group are respectively subjected to corresponding Euclidean distance calculation with the historical main acquisition feature group and the historical auxiliary acquisition feature group in the existing pulse signal feature library, to obtain the main acquisition feature proximity value and the auxiliary acquisition feature proximity value; If the main acquisition feature proximity value and the auxiliary acquisition feature proximity value are both greater than or equal to a preset threshold value, it indicates that the partial discharge risk is high, and a high-level early warning signal is generated; If the main acquisition feature proximity value or the auxiliary acquisition feature proximity value is greater than or equal to a preset threshold value, it indicates that the partial discharge risk is low, and a low-level early warning signal is generated.
[0016] The beneficial effects of the present application are as follows: by constructing a transient pulse reference library based on historical partial discharge data, the target main acquisition frequency is calculated in combination with the pulse duration characteristics, to ensure that the main acquisition frequency after adjustment can completely capture different types of partial discharge signals; by analyzing historical data to determine the maximum tolerance difference of the main and auxiliary acquisition frequencies, when the main acquisition frequency is adjusted due to not meeting the monitoring requirements, it is simultaneously checked whether the difference between the adjusted main and auxiliary frequencies exceeds the threshold value, and if the threshold value is exceeded, the auxiliary acquisition frequency is adaptively adjusted, forming a main and auxiliary frequency cooperative linkage mechanism. This scheme completely solves the pain points of the traditional monitoring method, i.e., the single adjustment of the main acquisition frequency easily leads to a large difference between the main and auxiliary frequencies and the disorder of data timing, avoids the judgment errors caused by frequency imbalance during joint analysis of partial discharge data, and significantly improves the accuracy and reliability of partial discharge monitoring on the basis of ensuring comprehensive capture of pulse signals. BRIEF DESCRIPTION OF DRAWINGS
[0017] The present application will be further described below with reference to the accompanying drawings.
[0018] Fig. 1 is a step flow chart of a ring net cabinet state monitoring and early warning method based on edge computing according to an embodiment of the present application; Fig. 2 is a logic judgment chart of a ring net cabinet state monitoring and early warning method based on edge computing according to an embodiment of the present application. DETAILED DESCRIPTION
[0019] In order to make the technical means, creative features, purposes and effects achieved by the present application easy to understand, the present application will be further described below with reference to the specific embodiments.
[0020] Example: Please refer to Figs. 1-2 As shown in the figure, the ring main unit status monitoring and early warning method based on edge computing described in this embodiment of the invention specifically includes the following steps: Step S10: Based on the historical partial discharge data of the ring main unit, analyze and determine the transient pulse reference duration under different types of partial discharge, and integrate it into a pulse duration reference library; In step S10, the historical partial discharge data of the ring main unit includes the type of partial discharge that occurs each time the ring main unit is in the historical operating cycle and the duration of the pulse signal generated when the partial discharge occurs. The historical partial discharge data can be obtained through the historical operation monitoring records of the ring main unit. The historical operating cycle represents the period between when the ring main unit is put into use and the current time. In step S10, the different types of partial discharge include, but are not limited to, corona discharge, internal discharge, and surface discharge; In step S10, the process of determining the transient pulse reference duration under different types of partial discharge is as follows: Based on any type of partial discharge; Extract the duration of the pulse signal generated during each partial discharge from historical partial discharge data and integrate them into a pulse signal duration set; Calculate the coefficient of variation of the pulse signal duration set to obtain the pulse signal duration stable value; The stability of the duration of the pulse signal generated under partial discharge is determined by comparing the sustained stable value of the pulse signal with a preset threshold. If the sustained stable value of the pulse signal is greater than the preset threshold, it indicates that the duration stability of the pulse signal generated by partial discharge is low. If the sustained stable value of the pulse signal is less than or equal to the preset threshold, it indicates that the duration of the pulse signal generated by partial discharge is highly stable. Based on the high stability of the duration of the pulse signal generated by partial discharge, the pulse signal duration set is averaged to obtain the transient pulse reference duration under partial discharge. Since the pulse signal duration generated by partial discharge has low stability, the minimum pulse signal duration within the pulse signal duration set is selected as the transient pulse reference duration under partial discharge. A traversal operation is performed on each type of partial discharge to obtain the transient pulse reference duration under different types of partial discharge, and the data is summarized and integrated to obtain a pulse duration reference library. In this implementation scheme, it is understood that the purpose of selecting the minimum pulse signal duration as the transient pulse reference duration under partial discharge is to ensure that the subsequent data acquisition frequency adjustment can meet the monitoring and capture of the pulse signal as much as possible. In this implementation scheme, step S10 serves the following purpose: By mining historical partial discharge data from the ring main unit, step S10 analyzes the stability of pulse signal duration using the coefficient of variation for different types of partial discharge, such as corona discharge and internal discharge. It then determines the transient pulse reference duration and constructs a reference library by combining averaging or minimum duration selection strategies. This step ensures the scientific validity of the reference duration—using the average value when stability is high to align with actual patterns, and selecting the minimum value when stability is low to avoid pulse omissions—and provides core data support for the accurate calculation of the subsequent main sampling frequency. This solves the problem of blind frequency setting caused by the lack of targeted pulse duration references in traditional monitoring, laying the foundation for the complete capture of partial discharge signals.
[0021] Step S20: Obtain the main sampling frequency of each edge device when monitoring different types of partial discharge in the ring network cabinet, and compare and analyze it with the pulse duration reference library to determine the type of partial discharge that requires adjustment of the main sampling frequency of the edge device; In step S20, each edge device refers to an embedded terminal deployed locally in the ring network cabinet, which has the ability to acquire and control data, perform local calculations, and make frequency adjustment decisions. It can manage and control the frequency acquisition and adjustment of sensors. The sensors include, but are not limited to, UHF, TEV, and ultrasonic sensors. In step S20, the main sampling frequency refers to the core sampling frequency set to fully capture the pulse signal when monitoring different types of partial discharge in a ring network cabinet. For example, for the main sampling UHF sensor and the auxiliary sampling TEV sensor for corona discharge, the main sampling frequency is the sampling frequency corresponding to the main sampling UHF sensor. In step S20, the process of determining the type of partial discharge that requires edge device main sampling frequency adjustment is as follows: Based on the transient pulse reference duration under different partial discharges in the pulse duration reference library, the target main sampling frequency under different partial discharges is calculated. The target main sampling frequency is calculated as: k / transient pulse reference duration, where k is the sampling point coefficient and the value of k is 20 (general standard for industrial monitoring). Based on any type of partial discharge, obtain the main sampling frequency when the current edge device performs partial discharge monitoring of the ring network cabinet, and record it as the actual main sampling frequency; Compare the actual primary sampling frequency with the target primary sampling frequency; If the actual main sampling frequency is higher than or equal to the target main sampling frequency, it means that the main sampling frequency of the current edge device when performing partial discharge monitoring of the ring network cabinet meets the requirements, and no operation is performed. If the actual main sampling frequency is lower than the target main sampling frequency, it means that the main sampling frequency of the current edge device when performing partial discharge monitoring of the ring network cabinet does not meet the requirements, and it means that the main sampling frequency of the edge device needs to be adjusted for this type of partial discharge. In this implementation scheme, it is understood that if the actual main sampling frequency is lower than the target main sampling frequency, it means that in this type of partial discharge monitoring, if the edge device collects pulse signals according to the actual main sampling frequency, it may not be able to capture the pulse signals generated by the partial discharge. Therefore, the main sampling frequency of the edge device needs to be adjusted under this type of partial discharge monitoring. In this implementation scheme, the significance of step S20 is as follows: Step S20 calculates the target master sampling frequency based on the pulse duration reference library, and accurately identifies the type of partial discharge that needs adjustment by comparing the current master sampling frequency of each edge device with the target value. The edge devices, relying on their local computing and frequency adjustment capabilities, combined with the characteristics of sensors such as UHF and TEV, ensure the real-time and targeted nature of the frequency comparison. This step solves the problem of fixed master sampling frequencies in traditional monitoring, which cannot adapt to the pulse characteristics of different types of partial discharges. By dynamically judging the adjustment needs, it provides a clear direction for subsequent frequency optimization, ensuring the integrity and effectiveness of partial discharge signal acquisition.
[0022] Step S30: For partial discharge types that require adjustment of the main acquisition frequency of edge devices, based on the historical main and auxiliary acquisition frequency difference data under different types of partial discharge and the historical discharge monitoring results of the ring network cabinet, analyze and determine the maximum tolerable frequency difference under different types of partial discharge monitoring. In step S30, the historical primary and secondary acquisition frequency difference data includes the primary and secondary acquisition frequency difference during different types of partial discharge monitoring, that is, the absolute frequency deviation between the primary acquisition frequency of the primary acquisition sensor and the secondary acquisition frequency of the secondary acquisition sensor. In step S30, the historical discharge monitoring results of the ring network cabinet include whether the edge device can accurately detect the partial discharge after a partial discharge occurs during monitoring of different types of partial discharge. In step S30, the process of analyzing and determining the maximum tolerance frequency difference under different types of partial discharge monitoring is as follows: Based on any type of partial discharge; Extract the different primary and secondary acquisition frequency differences during multiple partial discharges, and obtain the proportion of discharge monitoring compliance times under any primary and secondary acquisition frequency difference; The method for obtaining the proportion of discharge monitoring compliance is as follows: based on any difference between the main and auxiliary acquisition frequencies, the number of times the edge device can accurately monitor and judge partial discharge under the difference between the main and auxiliary acquisition frequencies is counted, and the proportion is calculated with the total number of partial discharge monitoring to obtain the proportion of discharge monitoring compliance. It should be noted that the total number of partial discharge monitoring counts represents the actual number of partial discharges that occurred in the ring main unit under the main and auxiliary acquisition frequency difference. The frequency differences between the main and auxiliary acquisitions are integrated to obtain the main and auxiliary acquisition frequency difference sequence. The proportion of discharge monitoring compliance times under each main and auxiliary acquisition frequency difference is integrated to obtain the discharge monitoring compliance sequence. The Pearson correlation coefficient between the primary and secondary acquisition frequency difference sequence and the discharge monitoring compliance sequence was calculated and the absolute value was processed to obtain the acquisition-monitoring correlation coefficient. The correlation coefficient between acquisition and monitoring is compared with the threshold of the acquisition-monitoring correlation coefficient to determine the type of association between the difference in primary and secondary acquisition frequencies and partial discharge compliance. If the acquisition-monitoring correlation coefficient is greater than or equal to the acquisition-monitoring correlation coefficient threshold, it indicates that there is a linear correlation between the difference in primary and secondary acquisition frequencies and the achievement of partial discharge standards. If the acquisition-monitoring correlation coefficient is less than the acquisition-monitoring correlation coefficient threshold, it indicates that there is a non-linear correlation between the difference in primary and secondary acquisition frequencies and the achievement of partial discharge standards. If there is a linear correlation, the least squares method is used to fit the main and auxiliary acquisition frequency difference sequence and the discharge monitoring compliance sequence to obtain the acquisition-monitoring correlation model. If there is a nonlinear correlation, the main and auxiliary acquisition frequency difference sequence and the discharge monitoring compliance sequence are nonlinearly fitted, and the fitting model with the highest goodness of fit is selected as the acquisition-monitoring correlation model after fitting. Nonlinear fitting methods include, but are not limited to, exponential function fitting and quadratic function fitting. Substituting the discharge monitoring compliance threshold into the acquisition-monitoring correlation model, the maximum tolerable frequency difference under partial discharge monitoring is obtained; The discharge monitoring compliance threshold is based on the technical guidelines for partial discharge monitoring of power equipment. The accurate identification rate of partial discharge monitoring for 10kV and above ring main units must be ≥90%, meaning the discharge monitoring compliance threshold must not be lower than 90%. It should be noted that there is usually a positive relationship between the difference in the main and auxiliary acquisition frequencies and the achievement of partial discharge standards. As the difference in the main and auxiliary acquisition frequencies increases, the achievement of partial discharge standards decreases. Therefore, by substituting the discharge monitoring achievement threshold into the acquisition-monitoring correlation model, the maximum tolerable frequency difference under partial discharge monitoring can be obtained. The maximum tolerable frequency difference ensures the accuracy of partial discharge monitoring. In this implementation plan, the significance of step S30 is as follows: Step S30 combines historical primary and secondary acquisition frequency difference data with discharge monitoring results, determines the correlation type using the Pearson correlation coefficient, constructs an acquisition-monitoring correlation model using linear or nonlinear fitting, and determines the maximum tolerable frequency difference based on a 90% monitoring compliance threshold. This step overcomes the limitation of traditional monitoring where the primary and secondary frequency difference lacks clear constraints. By quantifying the maximum tolerable range, it ensures the accuracy of partial discharge monitoring and provides a scientific basis for subsequent coordinated adjustment of primary and secondary frequencies, avoiding monitoring failures caused by excessive primary and secondary frequency differences, and further improving the reliability and standardization of partial discharge monitoring.
[0023] Step S40: Based on the maximum tolerable frequency difference under different types of partial discharge monitoring, determine whether the adjustment of the main sampling frequency of the edge device for partial discharge meets the requirement of the main and auxiliary sampling frequency difference under partial discharge monitoring. If it does not meet the requirement, when adjusting the main sampling frequency of the edge device, adjust the auxiliary sampling frequency of the edge device simultaneously according to the maximum tolerable frequency difference under partial discharge monitoring. In step S40, the process of determining whether the adjustment of the main sampling frequency of the edge device for partial discharge meets the requirement of the main and auxiliary sampling frequency difference under partial discharge monitoring is as follows: Based on any type of partial discharge; The auxiliary sampling frequency under partial discharge monitoring is obtained, and the absolute difference between it and the target main sampling frequency under partial discharge monitoring is calculated to obtain the main and auxiliary sampling frequency difference after the main sampling frequency is adjusted. In this implementation scheme, it can be understood that the adjustment of the main sampling frequency of the edge device for partial discharge is due to the fact that the actual main sampling frequency has not reached the target main sampling frequency. Therefore, the adjustment of the main sampling frequency is to adjust the actual main sampling frequency to the target main sampling frequency. Compare the difference between the main and auxiliary acquisition frequencies after the main acquisition frequency is adjusted with the maximum tolerable frequency difference under partial discharge monitoring. If the difference between the main and auxiliary acquisition frequencies after the main acquisition frequency adjustment is greater than the maximum tolerable frequency difference under partial discharge monitoring, it indicates that the main and auxiliary acquisition frequencies after the main acquisition frequency adjustment exceed the standard, and that the adjustment of the main acquisition frequency of the edge device of partial discharge does not meet the requirement of the main and auxiliary acquisition frequency difference under partial discharge monitoring. If the difference between the main and auxiliary acquisition frequencies after the main acquisition frequency adjustment is less than or equal to the maximum tolerable frequency difference under partial discharge monitoring, it means that the main and auxiliary acquisition frequencies after the main acquisition frequency adjustment have not exceeded the standard, and it means that the adjustment of the main acquisition frequency of the edge device of partial discharge meets the requirement of the difference between the main and auxiliary acquisition frequencies under partial discharge monitoring. Then, proceed directly to step S50. In step S40, the process of simultaneously adjusting the auxiliary sampling frequency of the edge device according to the maximum tolerable frequency difference under partial discharge monitoring is as follows: The absolute difference between the main and auxiliary acquisition frequency difference after the main acquisition frequency adjustment and the maximum tolerable frequency difference under partial discharge monitoring is calculated to obtain the main and auxiliary acquisition frequency adjustment value. The main and auxiliary acquisition frequency adjustment values are summed with the auxiliary acquisition frequency under partial discharge monitoring to obtain the auxiliary acquisition frequency target value. The auxiliary acquisition frequency of the edge device is adjusted according to the auxiliary acquisition frequency target value. It should be noted that adjusting the main and auxiliary sampling frequencies of edge devices is actually adjusting the sampling frequencies of the main and auxiliary sampling sensors in the edge device management. In this implementation scheme, the significance of step S40 is as follows: Based on the adjustment of the main acquisition frequency, step S40 optimizes the auxiliary acquisition frequency synchronously by verifying the matching between the adjusted main and auxiliary acquisition frequency difference and the maximum tolerable frequency difference. When the difference exceeds the limit, the auxiliary acquisition frequency is corrected by calculating the adjustment value to ensure the coordinated adaptation of the main and auxiliary sensor acquisition frequencies. This step solves the problem of main and auxiliary frequency imbalance that may be caused by adjusting the main acquisition frequency alone, realizes the dynamic coordination of the main and auxiliary acquisition frequencies, ensures the integrity of partial discharge signal capture, avoids monitoring loopholes caused by frequency adjustment, and improves the coordination and stability of the edge device acquisition system.
[0024] Step S50: After adjusting the main sampling frequency and auxiliary sampling frequency of the edge device, perform ring network cabinet status monitoring and early warning. In step S50, the process of performing status monitoring and early warning for the ring main unit is as follows: The pulse signals captured by acquisition according to the main acquisition frequency and the auxiliary acquisition frequency are subjected to feature extraction respectively, and integrated to obtain the main acquisition feature group and the auxiliary acquisition feature group; The main mining feature group and the auxiliary mining feature group are respectively compared with the historical main mining feature group and the historical auxiliary mining feature group in the existing pulse signal feature library to calculate the corresponding Euclidean distance, and the main mining feature proximity value and the auxiliary mining feature proximity value are obtained. In this implementation scheme, the existing pulse signal feature library contains historical main sampling feature groups and historical auxiliary sampling feature groups of pulse signals generated under different partial discharge types. These are all obtained by summarizing historical partial discharge monitoring. The historical main sampling feature groups and historical auxiliary sampling feature groups both contain signal features such as the amplitude, pulse width, frequency, phase, and duration of the pulse signal. Similarly, the main sampling feature groups and auxiliary sampling feature groups also contain these features. It should be noted that when the main mining feature group and the auxiliary mining feature group are compared with the historical main mining feature group and the historical auxiliary mining feature group, the corresponding partial discharge types are consistent. The proximity values of the primary and secondary sampling features are compared with preset thresholds respectively; If both the primary sampling feature and the secondary sampling feature are greater than or equal to the preset threshold, it indicates a high risk of partial discharge and generates an advanced early warning signal. If the main sampling feature or the auxiliary sampling feature is greater than or equal to the preset threshold, it indicates that the risk of partial discharge is low and a low-level early warning signal is generated. If both the primary sampling feature and the secondary sampling feature are less than the preset threshold, no operation will be performed, and partial discharge monitoring will continue. In this implementation scheme, the significance of step S50 is as follows: Step S50 extracts multi-dimensional features of the pulse signals collected by the main and auxiliary sensors after adjustment, compares them with the historical feature database using Euclidean distance, and generates warning signals in a graded manner based on the dual feature group verification results. This step relies on the dual verification mechanism of main and auxiliary features, overcoming the limitations of single feature monitoring and improving the accuracy and comprehensiveness of partial discharge risk identification. By distinguishing between high-level warnings, low-level warnings, and no warning scenarios, precise control of partial discharge risks is achieved, enabling timely warning of potential fault hazards, providing scientific support for ring main unit operation and maintenance decisions, and effectively ensuring the safe and stable operation of the ring main unit.
[0025] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.
Claims
1. An edge-computing-based ring main unit state monitoring and early warning method, characterized in that: The method comprises the following steps: Step S10: According to the historical partial discharge data of the ring main unit, the transient pulse reference duration under different types of partial discharge is analyzed and determined, and is integrated into a pulse duration reference library; Step S20: The main sampling frequency of the current edge device for monitoring different types of partial discharge of the ring main unit is obtained, and is combined and compared with the pulse duration reference library respectively to determine the type of partial discharge that needs to be adjusted in the main sampling frequency of the edge device; Step S30: For the type of partial discharge that needs to be adjusted in the main sampling frequency of the edge device, according to the historical main and auxiliary collection frequency difference data under different types of partial discharge and the historical discharge monitoring result of the ring main unit, the maximum tolerant frequency difference under different types of partial discharge monitoring is analyzed and determined; Step S40: According to the maximum tolerant frequency difference under different types of partial discharge monitoring, it is judged whether the main sampling frequency adjustment of the edge device for partial discharge monitoring meets the requirement of the main and auxiliary collection frequency difference, if not, when the main sampling frequency of the edge device is adjusted, the auxiliary sampling frequency of the edge device is also adjusted according to the maximum tolerant frequency difference under partial discharge monitoring; Step S50: After the main and auxiliary sampling frequencies of the edge device are adjusted, the state monitoring and early warning of the ring main unit are carried out.
2. The ring main unit condition monitoring and early warning method based on edge computing according to claim 1, characterized in that: The historical partial discharge data of the ring main unit includes the type of each partial discharge occurring in the historical operation cycle of the ring main unit and the duration of the pulse signal generated when the partial discharge occurs.
3. The method according to claim 2, wherein the method comprises the following steps: The process of determining the transient pulse reference duration under different types of partial discharge is as follows: Based on any type of partial discharge, the duration of the pulse signal generated at each partial discharge is extracted from the historical partial discharge data and integrated into a pulse signal duration set; The coefficient of variation of the pulse signal duration set is calculated to obtain a stable pulse signal duration value, and the stability of the duration of the pulse signal generated under partial discharge is judged; If the stable signal is high, the mean value operation is performed on the pulse signal duration set to obtain the transient pulse reference duration under partial discharge, and if the stable signal is low, the minimum pulse signal duration in the pulse signal duration set is selected as the transient pulse reference duration under partial discharge.
4. The method according to claim 3, characterized in that: The process of determining the type of partial discharge that needs to be adjusted in the main sampling frequency of the edge device is as follows: Based on the transient pulse reference duration under different types of partial discharge in the pulse duration reference library, the target main sampling frequency under different types of partial discharge is calculated; Based on any type of partial discharge, the main sampling frequency of the current edge device for monitoring the partial discharge of the ring main unit is obtained, which is recorded as the actual main sampling frequency; If the actual main sampling frequency is lower than the target main sampling frequency, it indicates that the main sampling frequency adjustment of the edge device is needed.
5. The method for monitoring and early warning of the ring main unit state based on edge computing according to claim 1, characterized in that: The historical main and auxiliary collection frequency difference data includes the main and auxiliary collection frequency difference under different types of partial discharge monitoring, that is, the absolute frequency deviation between the main sampling frequency of the main sampling sensor and the auxiliary sampling frequency of the auxiliary sampling sensor; The historical discharge monitoring result includes whether the edge device can accurately monitor the partial discharge after the partial discharge occurs under different types of partial discharge monitoring.
6. The method for monitoring and early warning of ring main unit status based on edge computing according to claim 5, characterized in that: The process of analyzing and determining the maximum tolerant frequency difference under different types of partial discharge monitoring is as follows: Based on any type of partial discharge, the difference between the main and auxiliary collection frequencies of multiple partial discharges is extracted, and based on any main and auxiliary collection frequency difference, the discharge monitoring compliance frequency ratio under the main and auxiliary collection frequency difference is obtained; The main and auxiliary collection frequency differences are integrated to obtain a main and auxiliary collection frequency difference sequence, and the discharge monitoring compliance frequency ratios under the main and auxiliary collection frequency differences are integrated to obtain a discharge monitoring compliance sequence; The Pearson correlation coefficient between the main and auxiliary collection frequency difference sequence and the discharge monitoring compliance sequence is calculated, and absolute value processing is performed to obtain a collection-monitoring correlation coefficient, and the correlation type between the main and auxiliary collection frequency difference and the partial discharge compliance is determined, if there is a linear correlation, then the least squares method is used to fit the main and auxiliary collection frequency difference sequence and the discharge monitoring compliance sequence to obtain a collection-monitoring correlation model, if there is a nonlinear correlation, then the main and auxiliary collection frequency difference sequence and the discharge monitoring compliance sequence are nonlinearly fitted, and after fitting, the fitting model with the highest goodness of fit is selected as the collection-monitoring correlation model; The discharge monitoring compliance threshold is substituted into the collection-monitoring correlation model to obtain the maximum tolerable frequency difference under partial discharge monitoring.
7. The method according to claim 6, characterized in that: The discharge monitoring compliance frequency ratio is obtained in the following way: Based on any main and auxiliary collection frequency difference, the number of times that the edge device can accurately monitor and judge partial discharge under the main and auxiliary collection frequency difference is counted, and the ratio with the total number of partial discharge monitoring is calculated to obtain the discharge monitoring compliance frequency ratio.
8. The edge computing-based ring main unit state monitoring and early warning method according to claim 1, characterized in that: The process of determining whether the edge device main collection frequency adjustment of partial discharge meets the requirement of the main and auxiliary collection frequency difference under partial discharge monitoring is as follows: Based on any type of partial discharge, the auxiliary collection frequency under partial discharge monitoring is obtained, and the absolute difference with the target main collection frequency under partial discharge monitoring is calculated to obtain the main and auxiliary collection frequency difference after the main collection frequency adjustment. If the main and auxiliary collection frequency difference after the main collection frequency adjustment is greater than the maximum tolerable frequency difference under partial discharge monitoring, it indicates that the edge device main collection frequency adjustment of partial discharge does not meet the requirement of the main and auxiliary collection frequency difference under partial discharge monitoring.
9. The method according to claim 8, characterized in that: The process of adjusting the edge device auxiliary collection frequency according to the maximum tolerable frequency difference under partial discharge monitoring is as follows: The absolute difference between the main and auxiliary collection frequency difference after the main collection frequency adjustment and the maximum tolerable frequency difference under partial discharge monitoring is calculated to obtain the main and auxiliary collection frequency adjustment value; The main and auxiliary collection frequency adjustment value and the auxiliary collection frequency under partial discharge monitoring are summed to obtain the auxiliary collection frequency target value, and the edge device auxiliary collection frequency is adjusted according to the auxiliary collection frequency target value.
10. The method according to claim 1, characterized in that: The process of ring network cabinet state monitoring and early warning is as follows: The pulse signals captured by collection according to the main and auxiliary collection frequencies are respectively subjected to feature extraction, and the main and auxiliary collection feature groups are obtained by integration; The main and auxiliary collection feature groups are respectively subjected to Euclidean distance calculation corresponding to the historical main and auxiliary collection feature groups in the existing pulse signal feature library to obtain the main and auxiliary collection feature proximity values; If the main and auxiliary collection feature proximity values are both greater than or equal to the preset threshold, it indicates that the risk of partial discharge is high, and a high-level early warning signal is generated. If the main sampling characteristic value or the auxiliary sampling characteristic value is greater than or equal to the preset threshold value, it is indicated that the partial discharge risk is low, and a low-level early warning signal is generated.