Cable accessory operation state online monitoring method and system based on edge computing

Abstract

The application relates to the field of data processing, in particular to a cable accessory operation state online monitoring method and system based on edge computing, wherein the method comprises the following steps: acquiring current sequences of an accessory to be monitored at multiple preset monitoring points, performing sliding window division on a current time period to obtain multiple time intervals and corresponding current subsequences; calculating energy density and energy entropy based on the current subsequences, and acquiring local aggregation degree in combination with time weight; performing differential processing on the energy density to obtain a decay degree, and acquiring a local decay degree in combination with a monitoring weight; and calculating a state abnormality degree according to the local aggregation degree and the local decay degree, so as to realize online monitoring and abnormal early warning of local discharge and insulation aging state of the cable accessory. The application can improve the local discharge abnormality recognition precision and enhance the dynamic perception ability of the insulation degradation trend.

Description

Technical Field

[0001] This application relates to the field of data processing. In particular, it relates to a method and system for online monitoring of the operating status of cable accessories based on edge computing. Background Technology

[0002] With the rapid development of urban power grids, high-load transmission networks, and underground utility tunnels, medium- and high-voltage power cables are widely used in long-distance, high-capacity power transmission scenarios. As a critical connection point in the power transmission link, the fusion splice in cable accessories directly affects the safety of the entire cable system. Because fusion splices typically contain multiple heterogeneous structures such as conductor connection layers, insulation recovery layers, and shielding recovery layers, they are prone to defects such as microcracks, air gaps, water trees, and localized aging under the combined effects of long-term thermal, electrical, and mechanical stresses, which can induce partial discharge phenomena.

[0003] Existing technologies typically monitor the operating status of fusion joints by methods such as partial discharge amplitude detection, partial discharge frequency statistics, or temperature threshold analysis. These methods rely on instantaneous abnormal signals from fixed monitoring locations, making it difficult to reflect the local collapse of the electric field structure caused by the spatial accumulation trend of partial discharge energy. This results in insufficient ability to identify abnormal operating states caused by early gradual insulation defects, and is prone to false alarms or missed alarms in complex electromagnetic environments. Summary of the Invention

[0004] This application provides an online monitoring method and system for the operating status of cable accessories based on edge computing. It is used to solve the problem that the existing technology only monitors the operating status of fusion splices by partial discharge amplitude detection, partial discharge frequency statistics or temperature thresholds. This method relies on instantaneous abnormal signals from fixed monitoring locations and ignores the local collapse of the electric field structure caused by the spatial accumulation trend of partial discharge energy under early gradual insulation defects, which leads to false alarms or missed alarms in the operating status of fusion splices. The method aims to improve the online identification capability of abnormal operating status of fusion splices.

[0005] In the first aspect, the online monitoring method for the operating status of cable accessories based on edge computing includes: Obtain the current sequence of the accessory to be monitored at each preset monitoring point within the current time period; The current time period is divided into multiple time intervals, and the current subsequence of each preset monitoring point in each time interval is obtained based on the time intervals and the current sequence. The energy density of each preset monitoring point in each time interval is obtained according to the current subsequence, and the energy entropy of the monitoring accessory in each time interval is obtained according to the energy density. Obtain the timeliness weight of each time interval, and perform weighted fusion of the energy entropy according to the timeliness weight to obtain the static aggregation degree of the monitored attachment; The energy entropy sequence of the monitored attachment is constructed based on the energy entropy, the energy entropy sequence is differentially processed to obtain the dynamic clustering degree of the monitored attachment, and the static clustering degree and the dynamic clustering degree are fused to obtain the local clustering degree of the monitored attachment. Based on the energy density, the monitoring weight of each preset monitoring point is obtained, the energy density is differentially processed to obtain the attenuation degree of each preset monitoring point, and the attenuation degree is fused based on the monitoring weight to obtain the local attenuation degree of the accessory to be monitored. The degree of abnormality of the monitored attachment is obtained based on the local aggregation degree and the local decay degree.

[0006] Preferably, dividing the current time period into multiple time intervals, and obtaining the current sub-sequence of each preset monitoring point in each time interval based on the time intervals and the current sequence includes: Get the set window length, and divide the current time period into multiple time intervals by sliding according to the set window length; Based on the current sequence of each preset monitoring point and the current data in each time interval, a current subsequence of each preset monitoring point in each time interval is obtained.

[0007] Preferably, the energy density of each preset monitoring point in each time interval is obtained according to the current subsequence, and the energy entropy of the monitored accessory in each time interval is obtained according to the energy density, including: Calculate the energy value of each current subsequence to obtain the energy value of each preset monitoring point in each time interval; The energy density is calculated as the ratio between the energy value and a scalar of a set window length, and the energy density of each preset monitoring point in each time interval is obtained. The information entropy is calculated based on the energy density of each preset monitoring point within the same time interval to obtain the energy entropy of the monitoring attachment in each time interval.

[0008] Preferably, obtaining the timeliness weight for each of the time intervals includes: Calculate the time interval between the start time of each time interval and the end time of the current time interval, and perform negative exponential mapping and normalization on the time interval to obtain the time-sensitivity weight of each time interval.

[0009] Preferably, the step of weighting and fusing the energy entropy according to the time-sensitivity weight to obtain the static aggregation degree of the monitored attachment includes: Based on the timeliness weight of each time interval, the energy entropy of the monitoring attachment in each time interval is fused to obtain the concentration score of the monitoring attachment in each time interval, and the sum of the concentration scores is calculated to obtain the static aggregation degree of the monitoring attachment. Preferably, the step of constructing the energy entropy sequence of the attachment to be monitored based on the energy entropy, performing differential processing on the energy entropy sequence to obtain the dynamic clustering degree of the attachment to be monitored, and fusing the static clustering degree and the dynamic clustering degree to obtain the local clustering degree of the attachment to be monitored includes: The energy entropy sequence of the monitored attachment is constructed based on the energy entropy of the attachment in each time interval. The energy entropy sequence is differentially processed to obtain the entropy difference sequence. The dynamic aggregation degree of the monitored attachment is calculated based on the positive difference value and all difference values ​​in the entropy difference sequence. The local clustering degree of the attachment to be monitored is obtained by fusing the static clustering degree and the dynamic clustering degree.

[0010] Preferably, obtaining the monitoring weight of each preset monitoring point based on energy density includes: Calculate the sum of energy densities of each preset monitoring point within the same time interval, calculate the density ratio between the energy density of each preset monitoring point within the time interval and the sum of densities within the time interval, and normalize the density ratio to obtain the monitoring weight of each preset monitoring point.

[0011] Preferably, the step of differentially processing the energy density to obtain the attenuation level of each of the preset monitoring points, and fusing the attenuation levels based on the monitoring weights to obtain the local attenuation level of the accessory to be monitored includes: A density sequence is constructed based on the energy density of each preset monitoring point in each time interval, and the density sequence is differentially processed to obtain a density difference sequence for each preset monitoring point. The attenuation level is calculated based on all negative difference values ​​in the density difference sequence to obtain the attenuation level of each preset monitoring point.

[0012] The attenuation degree is weighted and fused based on the monitoring weight of each preset monitoring point to obtain the local attenuation degree of the attachment to be monitored.

[0013] Preferably, the degree of abnormality in the state of the monitored attachment is obtained based on local aggregation and local attenuation, including: The degree of abnormality of the monitored attachment is obtained by fusing the local aggregation degree and the local decay degree.

[0014] Secondly, an online monitoring system for the operating status of cable accessories based on edge computing includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the online monitoring method for the operating status of cable accessories based on edge computing described in any one of the claims is implemented.

[0015] This application has the following effects: 1. This application divides time intervals and calculates energy entropy based on the energy density differences of each preset monitoring point within different time intervals. It analyzes the aggregation phenomenon of partial discharge energy space, quantifies the static aggregation degree of partial discharge, and realizes frequent discharge monitoring of insulation aging locations. At the same time, it analyzes the temporal entropy increase trend of energy entropy through the differential sequence analysis of energy entropy, analyzes the temporal aggregation phenomenon of partial discharge energy space, quantifies the dynamic aggregation degree of partial discharge, realizes two-dimensional monitoring of local collapse of electric field structure, improves the monitoring sensitivity of abnormal operating status of cable accessories, and avoids untimely monitoring of abnormal operating status.

[0016] 2. This application quantifies the monitoring weight of the energy retention trend of each preset monitoring point to the monitoring of the operating status by measuring the energy density change trend of each preset monitoring point. Based on the monitoring weight, the energy retention trend of each preset monitoring point is fused to realize the energy retention intensity analysis after partial discharge occurs at the location of insulation layer defect, accurately monitor the abnormal operating status caused by early insulation layer defects, and accurately identify the abnormal operating status.

[0017] The above and other objects, advantages and features of this application will become more apparent to those skilled in the art from the following detailed description of specific embodiments in conjunction with the accompanying drawings. Attached Figure Description

[0018] The following sections will describe some specific embodiments of this application in detail by way of example and not limitation, with reference to the accompanying drawings. The same reference numerals in the drawings denote the same or similar parts or components. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 This is a schematic flowchart of an online monitoring method for the operating status of cable accessories based on edge computing, according to an embodiment of this application. Figure 2 This is a schematic structural diagram of an online monitoring system for the operating status of cable accessories based on edge computing, according to an embodiment of this application. Detailed Implementation

[0019] The following reference Figures 1 to 2This application describes an edge computing-based online monitoring method and system for the operational status of cable accessories. In this description, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature, that is, include one or more of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified. When a feature "includes or contains" one or more of the features it encompasses, unless otherwise specifically described, this indicates that other features are not excluded and may be further included.

[0020] In the description of this embodiment, the terms "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0021] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0022] Please see Figure 1 The schematic flowchart of the online monitoring method for the operating status of cable accessories based on edge computing includes steps S1-S8, as follows: S1: Obtain the current sequence of the accessory to be monitored at each preset monitoring point within the current time period.

[0023] During long-term operation, the insulation restoration layer of fusion joints is prone to developing microcracks, air gaps, water trees, and localized aging areas under the combined effects of thermal stress, electrical stress, and humid environments. These defects lead to distortion of the local electric field and the formation of electric field concentration in localized areas. When the local electric field strength exceeds the local dielectric breakdown strength, partial discharge occurs. Partial discharge is a transient micro-breakdown process on the nanosecond to microsecond scale, with extremely rapid current changes, thus generating a large number of MHz-level high-frequency components and forming high-frequency partial discharge pulse currents. Since high-frequency partial discharge pulse currents can directly reflect changes in the local electric field, partial discharge energy accumulation, and insulation degradation process within the fusion joint, they are more sensitive to early insulation defects than traditional parameters such as temperature, current, or leakage current. Therefore, collecting high-frequency partial discharge pulse currents near fusion joints is a core data point for online monitoring of operational status.

[0024] Specifically, the preset monitoring points are located on the two cables connected by the fusion splice of the cable. The locations are determined by those skilled in the art based on known locations of partial discharges around historically defective fusion splices. Multiple monitoring points are continuously and evenly placed within the distribution range by varying the distances between these partial discharge locations, thus enabling comprehensive monitoring of abnormal partial discharge states caused by defects such as aging or damage to the fusion splice. In this application, the preset monitoring points include three points: the left grounding lead, the middle grounding loop, and the right grounding lead of the fusion splice.

[0025] The fusion splice in the circuit network to be monitored is taken as the accessory to be monitored. A high-frequency current transformer is set at each of the preset monitoring points to collect high-frequency partial discharge pulse current data in the current time period. The high-frequency partial discharge pulse current data is Z-score normalized to obtain the current sequence of the accessory to be monitored at each preset monitoring point.

[0026] Specifically, the Z-score standardization process is implemented by calculating the standard deviation of the sequence to which the value to be standardized belongs as the denominator, calculating the difference between the value to be standardized and the mean of the sequence as the numerator, and using the ratio of the two as the standard value to achieve standardization.

[0027] S2: Divide the current time period into multiple time intervals, and obtain the current subsequence of each preset monitoring point in each time interval based on the time intervals and the current sequence.

[0028] Specifically, the set window length is the set time interval length. Its size is calculated by those skilled in the art based on the high-frequency partial discharge pulse currents of several known defective historical fusion joints during partial discharge. Different window lengths are used to divide and calculate the variance of the information entropy, and the window length with the smallest variance is selected as the set window length. Within this set window, the ability to suppress random fluctuations of partial discharge pulses and the dynamic response capability of partial discharge can be taken into account simultaneously.

[0029] Set a window length, and then slide to divide the time range from the start time to the end time of the current time period according to the set window length to obtain multiple time intervals.

[0030] Taking one of the preset monitoring points as an example, this preset monitoring point is set as the target monitoring point. Based on the current sequence of the target monitoring point in each time interval, the current subsequence of the target monitoring point in each time interval is obtained. According to the above current subsequence acquisition steps, the current subsequence of each preset monitoring point in each time interval is obtained.

[0031] S3: Obtain the energy density of each preset monitoring point in each time interval according to the current subsequence, and obtain the energy entropy of the monitoring accessory in each time interval according to the energy density.

[0032] Calculate the energy value of each current subsequence to obtain the energy value of each preset monitoring point in each time interval.

[0033] Specifically, the energy value is realized by calculating the sum of the squares of each element in the current subsequence as the energy value of the current subsequence.

[0034] The energy density is calculated as the ratio between the energy value and a scalar of the set window length, thus obtaining the energy density of each preset monitoring point within each time interval. Since energy density is a scalar characteristic calculated based on a dimensionless energy value and a scalar of the window length, where the energy value is related to the window length, the energy density obtained by calculating the ratio between the energy value and the scalar of the set window length characterizes the energy level of the monitoring point per unit time. A higher energy density indicates a greater high-frequency partial discharge pulse current intensity at that monitoring point.

[0035] Taking one time interval as an example, this time interval is set as the target interval. The information entropy value is calculated based on the energy density of each preset monitoring point in the target interval, which is then used as the energy entropy of the accessory to be monitored in the target interval. The energy entropy of the accessory to be monitored in each time interval is obtained according to the above energy entropy calculation steps.

[0036] Specifically, the information entropy value satisfies the following relationship: In the formula, This represents the information entropy value of the attachment to be monitored within the target interval. Indicates the number of preset monitoring points. Indicates the first The ratio between the energy density of each preset monitoring point and the sum of the energy densities of all preset monitoring points. When When it is 0, it is defined It is 0.

[0037] S4: Obtain the timeliness weight of each time interval, and perform weighted fusion of the energy entropy according to the timeliness weight to obtain the static aggregation degree of the attachment to be monitored.

[0038] The time interval between the start time of the target interval and the end time of the current time interval is calculated. A negative exponential mapping is applied to this time interval to obtain the timeliness score of the target interval. The timeliness scores of the target interval are then summed and normalized to obtain the timeliness weight of the target interval. Since the timeliness weight is obtained based on the negative exponential mapping of the time interval, the larger the time interval, the smaller the timeliness weight. Because the size of the time interval is unstable, a very small time interval can easily cause weight explosion. The negative exponential mapping achieves exponential decay to avoid local time intervals dominating the timeliness weight.

[0039] Based on the above steps for obtaining the timeliness weight, the timeliness weight of each time interval is obtained.

[0040] Specifically, the negative exponential mapping is implemented by using... As a negative exponential function, The value to be mapped.

[0041] Specifically, the summation and normalization is achieved by using the sum of the time-efficiency scores of all time intervals as the denominator, the time-efficiency score of the target interval as the numerator, and calculating the ratio as the time-efficiency weight of the target interval to achieve summation and normalization.

[0042] The product of the energy entropy and the time-related weight in the target interval is calculated as the concentration score of the monitored attachment within the target interval. The sum of the concentration scores of the monitored attachment in each time interval is calculated as the static clustering degree of the monitored attachment. Since the static clustering degree is calculated based on energy entropy and time-related weight, the weight of data in the time interval closer to the current time is greater, and the greater the energy entropy, the greater the static clustering degree. This indicates that the probability of localized energy concentration and discharge around the monitored attachment is higher, and the insulation layer degradation is more severe. A higher energy entropy indicates that the energy value at a certain monitoring location is greater than that at other locations, and the phenomenon of energy concentration is more pronounced.

[0043] Specifically, the static aggregation degree satisfies the following relationship: In the formula, This indicates the static aggregation degree of the attachments to be monitored. Indicates the number of time intervals. Indicates the first Timeliness weight for each time interval The attached document to be monitored is in the [number]th [section]. Energy entropy within a time interval The attached document to be monitored is in the [number]th [section]. Concentrated fractions within a time interval.

[0044] S5: Construct the energy entropy sequence of the attachment to be monitored based on the energy entropy, perform differential processing on the energy entropy sequence to obtain the dynamic clustering degree of the attachment to be monitored, and fuse the static clustering degree and the dynamic clustering degree to obtain the local clustering degree of the attachment to be monitored.

[0045] An energy entropy sequence is constructed based on the energy entropy of the monitored accessory within each time interval. This energy entropy sequence is then differentially processed to obtain an entropy difference sequence. The average of all positive differences in the entropy difference sequence is calculated as the numerator, and the absolute average of all differences in the entropy difference sequence is calculated as the denominator. The ratio between the numerator and denominator is used to obtain the dynamic concentration degree of the monitored accessory. Since the dynamic concentration degree is calculated based on the energy entropy difference, a larger positive difference value relative to the average of the overall difference values ​​indicates a greater intensity of entropy increase, a more concentrated partial discharge current energy at a certain location, more severe insulation aging, and a more abnormal operating state. The difference value measures the temporal evolution of the partial discharge energy concentration phenomenon, characterizing the energy accumulation trend in the partial discharge space for analysis of insulation aging phenomena.

[0046] The local clustering degree of the attachment to be monitored is obtained by calculating the sum of the static clustering degree and the dynamic clustering degree.

[0047] S6: Obtain the monitoring weight of each preset monitoring point based on the energy density.

[0048] Calculate the sum of energy densities of each preset monitoring point within the target interval, and calculate the ratio between the energy density of the target monitoring point and the sum of densities within the target interval as the density ratio of the target monitoring point within the target interval.

[0049] The average density of the target monitoring points is calculated as the average density of the target monitoring points within each time interval. The average density is then summed and normalized to obtain the monitoring weight of each target monitoring point. Following the steps for obtaining the monitoring weight described above, the monitoring weight of each preset monitoring point is obtained. Since the monitoring weight is obtained based on the density ratio, a larger density ratio indicates a larger monitoring weight, and the attenuation level of that monitoring point is more important for condition monitoring. The density ratio measures the magnitude of the partial discharge energy at the monitoring point's location relative to other locations. A larger ratio indicates more severe insulation aging at that location, a more pronounced partial discharge phenomenon, and a better reflection of the energy retention phenomenon at that location, thus preventing it from being smoothed out by other locations.

[0050] Specifically, the summation and normalization is achieved by using the sum of the average densities of all preset monitoring points as the denominator, the average density of the target monitoring point as the numerator, and calculating the ratio as the monitoring weight of the target monitoring point to achieve summation and normalization.

[0051] S7: Perform differential processing on the energy density to obtain the attenuation degree of each preset monitoring point, and fuse the attenuation degree based on the monitoring weight to obtain the local attenuation degree of the accessory to be monitored.

[0052] A density sequence is constructed based on the energy density of each preset monitoring point in each time interval, and the density sequence is differentially processed to obtain a density difference sequence for each preset monitoring point.

[0053] The absolute average value of all negative difference values ​​in the density difference sequence of the target monitoring point is calculated as the attenuation degree of the target monitoring point. The attenuation degree of each preset monitoring point is obtained according to the attenuation degree acquisition steps. Since the attenuation degree is calculated based on the negative difference values ​​in the density difference sequence, the larger the negative difference value, the greater the attenuation intensity when the energy decays. When the insulation layer of the fusion joint ages, partial discharge occurs nearby, and there is a certain degree of energy retention as the aging state progresses. The more severe the aging, the slower the attenuation. Therefore, calculating the attenuation intensity and analyzing the aging state is used for monitoring the operating status of the fusion joint.

[0054] The local attenuation degree of the accessory to be monitored is obtained by summing the products between the monitoring weight of each preset monitoring point and the attenuation degree.

[0055] S8: Obtain the degree of abnormality of the monitored accessory based on the local aggregation degree and the local attenuation degree, input the degree of abnormality of the monitored accessory into the central system, determine whether there is an abnormality, and complete the online monitoring of the operating status of the cable accessory.

[0056] The local attenuation degree is mapped to a negative exponential value to obtain the attenuation degree mapping value. The sum of the local clustering degree and the attenuation degree mapping value is calculated to obtain the degree of abnormality of the state of the monitored attachment.

[0057] The abnormal status of the accessory to be monitored is input into the central system. When the abnormal status exceeds the preset threshold, an early warning of abnormal operation status of the fusion splice is issued, and relevant personnel are arranged to replace the fusion splice.

[0058] Specifically, the preset threshold is set by monitoring personnel in the field who perform a traversal test on the current data of each monitoring point during the operation of multiple known abnormal and normal fusion splices to obtain the degree of abnormality of the known fusion splices. The median between the average baseline of the degree of abnormality of the known abnormal fusion splices and the average baseline of the degree of abnormality of the known normal fusion splices is used as the preset threshold.

[0059] The flowchart provided in this embodiment is not intended to indicate that the operations of the method will be performed in any particular order, or that all operations of the method are included in every case. Furthermore, the method may include additional operations. Within the scope of the technical concept provided by the method in this embodiment, additional variations can be made to the above method.

[0060] It should be understood that in some embodiments, the components may be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods may be implemented using software or firmware stored in memory and executed by a suitable instruction execution system.

[0061] This application also provides an online monitoring system for the operating status of cable accessories based on edge computing. For example... Figure 2 As shown, the system includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the online monitoring method for the operating status of cable accessories based on edge computing according to the first aspect of this application. The system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. Their configuration and functions are known in the art and will not be described in detail here.

[0062] Therefore, those skilled in the art should recognize that although many exemplary embodiments of this application have been shown and described in detail herein, many other variations or modifications conforming to the principles of this application can be directly determined or derived from the disclosure of this application without departing from the spirit and scope of this application. Thus, the scope of this application should be understood and construed as covering all such other variations or modifications.

Claims

1. A method for online monitoring of the operating status of cable accessories based on edge computing, characterized in that, include: Obtain the current sequence of the accessory to be monitored at each preset monitoring point within the current time period; The current time period is divided into multiple time intervals, and the current subsequence of each preset monitoring point in each time interval is obtained based on the time intervals and the current sequence. The energy density of each preset monitoring point in each time interval is obtained according to the current subsequence, and the energy entropy of the monitoring accessory in each time interval is obtained according to the energy density. Obtain the timeliness weight of each time interval, and perform weighted fusion of the energy entropy according to the timeliness weight to obtain the static aggregation degree of the monitored attachment; The energy entropy sequence of the attachment to be monitored is constructed based on the energy entropy. The energy entropy sequence is differentially processed to obtain the dynamic clustering degree of the attachment to be monitored. The static clustering degree and the dynamic clustering degree are fused to obtain the local clustering degree of the attachment to be monitored. The monitoring weight of each preset monitoring point is obtained based on the energy density, the energy density is differentially processed to obtain the attenuation degree of each preset monitoring point, and the attenuation degree is fused based on the monitoring weight to obtain the local attenuation degree of the accessory to be monitored. The degree of abnormality of the monitored attachment is obtained based on the local aggregation degree and the local decay degree.

2. The method for online monitoring of the operating status of cable accessories based on edge computing according to claim 1, characterized in that, The step of dividing the current time period into multiple time intervals and obtaining the current sub-sequence of each preset monitoring point in each time interval based on the time intervals and the current sequence includes: Get the set window length, and divide the current time period into multiple time intervals by sliding according to the set window length; Based on the current sequence of each preset monitoring point and the current data in each time interval, a current subsequence of each preset monitoring point in each time interval is obtained.

3. The online monitoring method for the operating status of cable accessories based on edge computing according to claim 1, characterized in that, The step of obtaining the energy density of each preset monitoring point in each time interval based on the current subsequence, and obtaining the energy entropy of the monitored accessory in each time interval based on the energy density, includes: Calculate the energy value of each current subsequence to obtain the energy value of each preset monitoring point in each time interval; The energy density is calculated as the ratio between the energy value and a scalar of a set window length, and the energy density of each preset monitoring point in each time interval is obtained. The information entropy is calculated based on the energy density of each preset monitoring point within the same time interval to obtain the energy entropy of the monitoring attachment in each time interval.

4. The online monitoring method for the operating status of cable accessories based on edge computing according to claim 1, characterized in that, The step of obtaining the timeliness weight for each time interval includes: Calculate the time interval between the start time of each time interval and the end time of the current time interval, and perform negative exponential mapping and normalization on the time interval to obtain the time-sensitivity weight of each time interval.

5. The method for online monitoring of the operating status of cable accessories based on edge computing according to claim 1, characterized in that, The step of weighting and fusing the energy entropy according to the time-sensitivity weight to obtain the static aggregation degree of the monitored attachments includes: Based on the timeliness weight of each time interval, the energy entropy of the monitoring attachment in each time interval is fused to obtain the concentration score of the monitoring attachment in each time interval, and the sum of the concentration scores is calculated to obtain the static aggregation degree of the monitoring attachment.

6. The method for online monitoring of the operating status of cable accessories based on edge computing according to claim 1, characterized in that, The step of constructing an energy entropy sequence of the monitored attachments based on the energy entropy, performing differential processing on the energy entropy sequence to obtain the dynamic clustering degree of the monitored attachments, and fusing the static clustering degree and the dynamic clustering degree to obtain the local clustering degree of the monitored attachments includes: The energy entropy sequence of the monitored attachment is constructed based on the energy entropy of the attachment in each time interval. The energy entropy sequence is differentially processed to obtain the entropy difference sequence. The dynamic aggregation degree of the monitored attachment is calculated based on the positive difference value and all difference values ​​in the entropy difference sequence. The local clustering degree of the attachment to be monitored is obtained by fusing the static clustering degree and the dynamic clustering degree.

7. The method for online monitoring of the operating status of cable accessories based on edge computing according to claim 1, characterized in that, The method of obtaining the monitoring weight of each preset monitoring point based on energy density includes: Calculate the sum of energy densities of each preset monitoring point within the same time interval, calculate the density ratio between the energy density of each preset monitoring point within the time interval and the sum of densities within the time interval, and normalize the density ratio to obtain the monitoring weight of each preset monitoring point.

8. The method for online monitoring of the operating status of cable accessories based on edge computing according to claim 1, characterized in that, The step of differentially processing the energy density to obtain the attenuation level of each preset monitoring point, and fusing the attenuation levels based on the monitoring weights to obtain the local attenuation level of the accessory to be monitored, includes: A density sequence is constructed based on the energy density of each preset monitoring point in each time interval, and the density sequence is differentially processed to obtain a density difference sequence for each preset monitoring point. The attenuation level is calculated based on all negative difference values ​​in the density difference sequence to obtain the attenuation level of each preset monitoring point; The attenuation degree is weighted and fused based on the monitoring weight of each preset monitoring point to obtain the local attenuation degree of the attachment to be monitored.

9. The online monitoring method for the operating status of cable accessories based on edge computing according to claim 1, characterized in that, The process of obtaining the degree of abnormality in the state of the monitored attachments based on local aggregation and local decay includes: The degree of abnormality of the monitored attachment is obtained by fusing the local aggregation degree and the local decay degree.

10. An online monitoring system for the operating status of cable accessories based on edge computing, characterized in that, include: A processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement the steps of the online monitoring method for the operating status of cable accessories based on edge computing according to any one of claims 1-9.

Images