Insulator flashover fault monitoring method, device, equipment, system and medium

Abstract

The present invention provides a method, device, equipment, system and medium for monitoring pollution flashover faults of insulators, including: obtaining an effective noise characteristic value data set generated by an edge pickup device, the effective noise characteristic value data set is the edge pickup device collecting the surrounding original audio signal and extracting and generating it by using an edge computing algorithm, and the effective noise characteristic value data set includes the most efficient and the least efficient resolution effective noise characteristic value groups; calling a fault monitoring system to process the effective noise characteristic value data set accumulated within a preset time, generating an abnormal evaluation score, processing the most efficient resolution effective noise characteristic value group to generate a judgment result, comparing the judgment result with the value of the least efficient resolution effective noise characteristic value group to obtain an abnormal evaluation score; outputting an alarm signal according to the abnormal evaluation score. It is intended to solve the problems that the existing monitoring methods for pollution flashover of insulators are labor-intensive and financial-intensive, have poor prevention effects, and are not suitable for promotion.

Description

Technical Field

[0001] The present invention relates to the field of insulator device detection, and in particular to an insulator pollution flashover fault monitoring method, device, equipment, system and medium. Background Art

[0002] Insulators that have been operating outdoors for a long time will gradually form a dirty layer on the porcelain surface due to pollution from smoke, dust, and industrial sewage. The contaminated insulators will induce surface flashover under the action of the operating voltage. This phenomenon is referred to as "pollution flashover". Under dry environmental conditions, the pollution flashover phenomenon of insulators has little impact on power operations. However, under climatic conditions such as haze, rain, dew, and snow, the dirt layer on the surface of the insulator porcelain will cause a sharp drop in flashover voltage due to moisture, leading to serious pollution flashover events. Pollution flashover is accompanied by significant noise and flash, and it is very easy to occur even under working voltage. The characteristic of insulator pollution flashover failure is that the process lasts for a long time. Once it occurs, it is generally impossible to eliminate it by automatic closing. The fault is easy to deteriorate rapidly, causing large-scale power outages. Maintenance is difficult to implement and the recovery time is long, which seriously affects the safe and stable operation of the power system.

[0003] At present, there are four main monitoring methods for the pollution flashover of insulators. The first is the traditional manual inspection. During daily manual inspections, the pollution flashover phenomenon and severity of faulty insulators can be judged by human ears and eyes. However, due to personal sensory differences and environmental weather influences, it is difficult for the human ear to distinguish the subtle manifestations of the early stage of a fault, and it is easy to miss the best maintenance time. Regularly organize inspection personnel to clean the dirty insulators regularly or use charged water to wash them to keep the surface of the insulators clean. Since most high-voltage towers are installed in the field, the locations are numerous and scattered, so a lot of manpower is required, and the maintenance effect is poor. At the same time, there are safety hazards when personnel frequently work at high altitudes. The second is operating parameter monitoring. By monitoring the operating parameters of each node, such as current and voltage, mutual inductance current and voltage sensors are installed on the cables before and after the tower to monitor the changes in the working voltage and leakage current of the section, so as to reversely infer whether the insulator has a flashover fault. The sensors used in this method are easy to install and have mature low-power technology support. However, since the working process of power equipment is fluctuating and it is difficult to find a regular pattern, the changes in the monitored parameters may not be obvious during mild flashover, and will only change significantly during severe faults. At this time, it is already the late stage of the fault and the equipment is on the verge of shutdown, which does not play a role in fault prevention. The third is a fixed-point video monitoring device. By installing a video monitoring device around the insulator, the surface image of the insulator is taken at regular intervals and sent back, and the state of the insulator is manually judged. Since the insulator is a three-dimensional column as a whole, the image cannot cover the back and blind spots, and there are many on a single tower, up to dozens. If the insulator is to be monitored in all directions and in full quantity, the construction is extremely difficult, the project cost is extremely high, the input-output ratio is low, and it is not universal. This method can only be used for some core and important towers, and is not suitable for comprehensive promotion. The fourth method is to combine emerging visual recognition with drone technology. The use of drone programming technology and visual recognition can replace human high-altitude operations and improve inspection efficiency. However, there are still problems of high cost and difficulty in cross-regional operations, which is not conducive to promotion and use.

[0004] In addition, due to the particularity of insulators, if you choose to install mechanical or physical contact monitoring products on the insulators, there will be problems such as difficult installation, extremely high protection requirements, and dangerous operating conditions.

[0005] In view of this, this application is filed. Summary of the invention

[0006] In view of this, the purpose of the present invention is to provide an insulator flashover fault monitoring method, device, equipment, system and medium, which can effectively solve the problems of the monitoring means of insulator flashover in the prior art that consumes a lot of manpower, has poor maintenance effect, has safety hazards caused by frequent high-altitude operations of workers, has poor fault prevention effect, is costly, and is not suitable for comprehensive promotion.

[0007] The present invention discloses a method for monitoring insulator flashover faults, comprising:

[0008] Acquire an effective noise characteristic value data set generated by an edge sound pickup device, wherein the effective noise characteristic value data set is generated by extracting and generating an original audio signal of the surrounding environment collected by the edge sound pickup device using an edge computing algorithm, and the effective noise characteristic value data set includes an effective noise characteristic value group with the highest effective resolution that can retain effective information and an effective noise characteristic value group with the lowest effective resolution that can retain data change trends;

[0009] Calling the trained fault monitoring system to process the multiple groups of effective noise characteristic value data sets accumulated within a preset time to generate an abnormality assessment score, wherein the multiple groups of the most efficient resolution effective noise characteristic value groups are subjected to abnormality judgment processing to generate a judgment result, and the judgment result is compared with the values ​​of the multiple groups of the least efficient resolution effective noise characteristic value groups to obtain a corresponding abnormality assessment score;

[0010] According to the abnormality assessment score, an alarm signal is output.

[0011] Preferably, the trained fault monitoring system is called to process the multiple groups of effective noise characteristic value data sets accumulated within a preset time to generate an abnormality assessment score, specifically:

[0012] Accumulate multiple groups of the most efficient resolution effective noise eigenvalue groups within a preset time to perform autocorrelation calculations to generate an autocorrelation array that can show the periodic law of the effective eigenvalues;

[0013] Performing FFT calculation on the autocorrelation array to generate an energy value array corresponding to the autocorrelation array;

[0014] Converting the energy value array into an energy proportion array, and determining whether the value of the energy proportion array in each frequency range exceeds a preset value, to generate an abnormal data group, wherein the abnormal data group is the value of the energy proportion array that exceeds the preset value;

[0015] An anomaly score is evaluated for the abnormal data group based on the values ​​of multiple groups of the minimum effective resolution effective noise feature value groups accumulated within a preset time, and an anomaly score corresponding to the abnormal data group is generated, wherein the anomaly score is the value of the minimum effective resolution effective noise feature value group in the frequency band corresponding to the abnormal data group.

[0016] The present invention also provides an insulator flashover fault monitoring device, comprising:

[0017] An effective eigenvalue acquisition unit is used to acquire an effective noise eigenvalue data set generated by an edge sound pickup device, wherein the effective noise eigenvalue data set is generated by extracting and generating an original audio signal of the surrounding environment collected by the edge sound pickup device using an edge computing algorithm, and the effective noise eigenvalue data set includes an effective noise eigenvalue group with the highest effective resolution that can retain effective information and an effective noise eigenvalue group with the lowest effective resolution that can retain data change trends;

[0018] An abnormality analysis unit is used to call a trained fault monitoring system to process multiple groups of effective noise characteristic value data sets accumulated within a preset time to generate an abnormality evaluation score, wherein the multiple groups of the most efficient resolution effective noise characteristic value groups are subjected to abnormality judgment processing to generate a judgment result, and the judgment result is compared with the values ​​of the multiple groups of the least efficient resolution effective noise characteristic value groups to obtain a corresponding abnormality evaluation score;

[0019] The early warning unit is used to output an alarm signal according to the abnormal evaluation score.

[0020] The present invention also provides an insulator flashover fault monitoring device, comprising:

[0021] An autocorrelation calculation unit is used to perform autocorrelation calculation on a plurality of groups of effective noise eigenvalues ​​with the highest efficiency resolution accumulated within a preset time, so as to generate an autocorrelation array capable of showing the periodic law of effective eigenvalues;

[0022] A Fourier transform unit, used for performing FFT calculation on the autocorrelation array to generate an energy value array corresponding to the autocorrelation array;

[0023] An abnormality judgment unit, used for converting the energy value array into an energy proportion array, and judging whether the value of the energy proportion array in each frequency range exceeds a preset value, and generating an abnormal data group, wherein the abnormal data group is the value of the energy proportion array exceeding the preset value;

[0024] An anomaly score evaluation unit is used to evaluate the anomaly score of the abnormal data group based on the values ​​of multiple groups of the minimum effective resolution effective noise feature value groups accumulated within a preset time, and generate an anomaly score corresponding to the abnormal data group, wherein the anomaly score is the value of the minimum effective resolution effective noise feature value group in the frequency band corresponding to the abnormal data group.

[0025] The present invention also provides an insulator flashover fault monitoring device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, an insulator flashover fault monitoring method as described in any one of the above items is implemented.

[0026] The present invention also provides an insulator pollution flashover fault monitoring system, comprising an edge sound pickup device, a remote monitoring terminal and an insulator pollution flashover fault monitoring device as described in claim 5, wherein the insulator pollution flashover fault monitoring device is wirelessly connected to the edge sound pickup device, and the insulator pollution flashover fault monitoring device is wirelessly connected to the remote monitoring terminal.

[0027] Preferably, the edge pickup device is used for:

[0028] Acquire the original audio signal collected by the edge sound pickup device, wherein the original audio signal is all sounds within the effective range of the surrounding environment;

[0029] Using the edge computing algorithm to compress the original audio signal, dividing it into k first original noise feature value groups with a step size of n, and p second original noise feature value groups with a step size of h, where n, k, h, and p are preset values;

[0030] Performing FFT calculation on each of the first original noise eigenvalue groups and each of the second original noise eigenvalue groups to generate corresponding first effective noise eigenvalue groups and second effective noise eigenvalue groups;

[0031] An extraction process is performed on each of the first effective noise eigenvalue groups and each of the second effective noise eigenvalue groups to extract data of each of the first effective noise eigenvalue groups within a preset frequency band and data of each of the second effective noise eigenvalue groups, and the data are summed to generate a most efficient resolution effective noise eigenvalue group corresponding to the first effective noise eigenvalue group and a least efficient resolution effective noise eigenvalue group corresponding to the second effective noise eigenvalue group.

[0032] The present invention also provides a readable storage medium storing a computer program, wherein the computer program can be executed by a processor of a device where the storage medium is located to implement an insulator flashover fault monitoring method as described in any one of the above.

[0033] In summary, the present embodiment provides an insulator flashover fault monitoring method, device, equipment and medium, which use an edge pickup device that does not need any mechanical connection with the insulator to collect audio signals from the surrounding environment, and extract effective noise, and use the extracted effective noise as an input parameter of the fault monitoring system. After the fault monitoring system processes the effective noise characteristic value data set, it will output an abnormality evaluation score, and send an alarm signal to a remote monitoring terminal based on the abnormality evaluation score, thereby solving the problems of the monitoring means of flashover of insulators in the prior art consuming a lot of manpower, poor maintenance effect, safety hazards of frequent high-altitude operations of workers, poor fault prevention effect, high cost, and unsuitability for comprehensive promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of an insulator flashover fault monitoring method provided by an embodiment of the present invention.

[0035] Figure 2 It is a schematic diagram of an original waveform signal of 1 s with a length of 48000 provided in an embodiment of the present invention.

[0036] Figure 3 It is a schematic diagram of a continuous feature vector waveform signal of 20 seconds accumulated in the buffer provided in an embodiment of the present invention.

[0037] Figure 4 It is a schematic diagram of the periodic variation of the autocorrelation function of the high-dimensional characteristic vector when pollution flashover occurs provided by an embodiment of the present invention.

[0038] Figure 5 It is a schematic diagram of the frequency performance of the autocorrelation when no pollution flashover occurs on the insulator, provided by an embodiment of the present invention.

[0039] Figure 6-1 It is a schematic diagram of the first frequency representation of the autocorrelation when pollution flashover occurs on an insulator, provided by an embodiment of the present invention.

[0040] Figure 6-2 It is a schematic diagram of the second frequency representation of the autocorrelation when pollution flashover occurs on an insulator, provided by an embodiment of the present invention.

[0041] Figure 7 It is a healthy schematic diagram of the pollution flashover strength of insulators on the line provided by an embodiment of the present invention.

[0042] Figure 8 It is a schematic diagram of a health assessment curve of a stepped energy increasing insulator provided in an embodiment of the present invention.

[0043] Fig. 9 It is a schematic diagram of the first module of the insulator flashover fault monitoring device provided in an embodiment of the present invention.

[0044] Fig.10 It is a schematic diagram of the second module of the insulator flashover fault monitoring device provided in an embodiment of the present invention.

[0045] Fig.11 It is a structural schematic diagram of an insulator flashover fault monitoring system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but merely represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0047] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0048] See also Figure 1 The first embodiment of the present invention provides an insulator flashover fault monitoring method, which can be executed by an insulator flashover fault monitoring device (hereinafter referred to as the monitoring device), and in particular, by one or more processors in the monitoring device to implement the following steps:

[0049] S101, obtaining an effective noise feature value data set generated by an edge sound pickup device, wherein the effective noise feature value data set is generated by extracting the original audio signal of the surrounding environment collected by the edge sound pickup device and using an edge computing algorithm to extract the original audio signal, and the effective noise feature value data set includes a most efficient resolution effective noise feature value group that can retain effective information and a least efficient resolution effective noise feature value group that can retain data change trends;

[0050] Specifically, in this embodiment, a monitoring point is selected and the edge sound pickup device with anti-electromagnetic interference is arranged to collect the overall noise of the insulator transmission process in real time. The edge computing algorithm built into the edge sound pickup device is used to extract the audio information that effectively responds to the abnormal phenomenon of the insulator in real time to output a high-dimensional feature vector, and the high-dimensional feature vector is instantly transmitted to the central server for storage through the communication module built into the edge sound pickup device using 4G communication means.

[0051] In this embodiment, take the original audio signal of 1s with a length of 48000 as an example. Figure 2 As shown in the figure, the original audio signal is directly compressed into two sets of effective feature data of length 376 and 24 by the edge algorithm. The abnormal noise of the insulator is concentrated in the specified frequency domain, which is different from the frequency domain where other noises are located. Through this compression extraction, most of the interfering sounds such as insects and birds, wind noise, etc. in the noisy environment have been filtered out, and only the effective feature values ​​containing the characteristics of the insulator are extracted. Among them, the high-dimensional feature vector retains the effective information of the insulator characteristics, and covers the effective information of the original audio signal to the maximum extent with a capacity far smaller than the original sound 4-, and performs preliminary data filtering for subsequent data mining, which not only improves work efficiency, but also greatly saves the broadband cost of network transmission data and ensures the real-time nature of algorithm monitoring.

[0052] S102, calling the trained fault monitoring system to process the multiple groups of effective noise characteristic value data sets accumulated within a preset time to generate an abnormality assessment score, wherein the multiple groups of the most efficient resolution effective noise characteristic value groups are subjected to abnormality judgment processing to generate a judgment result, and the judgment result is compared with the values ​​of the multiple groups of the least efficient resolution effective noise characteristic value groups to obtain a corresponding abnormality assessment score;

[0053] Specifically, in this embodiment, multiple groups of the most efficient resolution effective noise eigenvalue groups are accumulated and obtained within a preset time to perform autocorrelation calculations to generate an autocorrelation array that can show the periodic law of the effective eigenvalues;

[0054] Performing FFT calculation on the autocorrelation array to generate an energy value array corresponding to the autocorrelation array;

[0055] Converting the energy value array into an energy proportion array, and determining whether the value of the energy proportion array in each frequency range exceeds a preset value, to generate an abnormal data group, wherein the abnormal data group is the value of the energy proportion array that exceeds the preset value;

[0056] An anomaly score is evaluated for the abnormal data group based on the values ​​of multiple groups of the minimum effective resolution effective noise feature value groups accumulated within a preset time, and an anomaly score corresponding to the abnormal data group is generated, wherein the anomaly score is the value of the minimum effective resolution effective noise feature value group in the frequency band corresponding to the abnormal data group.

[0057] Specifically, in this embodiment, adaptive clustering recognition and result calculation are performed on the effective noise characteristic value data set transmitted by the edge sound pickup device, and real-time push and automatic alarm are performed on abnormal events that meet the alarm threshold.

[0058] Specifically, in this embodiment, taking the original audio signal of 1s with a length of 48000 as an example, the effective noise feature value group with the highest effective resolution of 376 and the effective noise feature value group with the lowest effective resolution of 24 transmitted by the edge pickup device are stored every one second, and the two groups of effective feature value data are accumulated for 20s respectively, as shown in FIG. Figure 3 As shown; at the same time, an abnormality detection is performed on the most efficient resolution effective noise eigenvalue group with a length of 376*20 accumulated every 20s, that is, an autocorrelation operation is performed on the most efficient resolution effective noise eigenvalue group accumulated for 20s to obtain an autocorrelation array that can show the periodic law of the effective eigenvalue and an autocorrelation curve that represents the periodic law of the data, as shown in Figure 4 As shown; among them, when the insulator does not have an abnormality, its autocorrelation curve is the correlation curve of Gaussian noise, which is irregular and therefore has no autocorrelation; when the insulator actually has an abnormality, the autocorrelation curve extracted by the noise will show a fixed frequency change. Further, the calculated autocorrelation array is subjected to fft calculation to obtain an array of length 8000. The length of the array represents the frequency range, and the value represents the energy. At the same time, the energy value array is converted into an energy ratio array. If the energy ratio in each frequency range exceeds the set range, it means that an abnormal insulator event has occurred; the presence or absence of autocorrelation is converted into a more direct energy ratio curve. The energy ratio curve is expressed as a smooth straight line, indicating that there is no obvious periodicity, as shown in Figure 5 As shown in the figure; when the energy ratio curve fluctuates at different positions on the horizontal axis, it means that the noise characteristics have obvious periodicity in different frequency domains, indicating that different abnormalities occur in the insulator; among them, the insulator has pollution flashover abnormality, such as Figure 6-1 and Figure 6-2 As shown. When an abnormality is detected, that is, when the value of each frequency domain of the energy proportion curve exceeds the threshold for judging the abnormality, the data of the minimum effective resolution effective noise characteristic value group with a length of 24*20 is used to evaluate the degree of abnormality. The greater the energy proportion of the maximum effective resolution effective noise characteristic value group, the greater the value of the minimum effective resolution effective noise characteristic value group of the corresponding section. At this time, the higher the degree of abnormality, the lower the health assessment score of the insulator, and vice versa. Among them, according to the existing data samples, two abnormality alarm lines of 75 and 50 can be set. When the evaluation score is greater than 50 and less than or equal to 75, it is judged that a slight abnormality has occurred. When the score is less than or equal to 50, it is judged that a strong abnormality has occurred. When any alarm line is triggered, the alarm information will be pushed in real time to remind relevant personnel to perform maintenance.

[0059] S103: Outputting an alarm signal according to the abnormality assessment score.

[0060] See also Figures 7 and 8Specifically, in this embodiment, the insulator pollution flashover fault monitoring method can monitor abnormal events of insulators in real time, accurately capture abnormalities in insulator transmission, quantitatively evaluate the health status of insulators, predict the development trend of faults and send alarm information to the remote monitoring terminal. Among them, the user can monitor the abnormal evaluation score in real time through the remote control terminal, identify the analysis results, feedback the insulator status and health evaluation value curve, and fault warning information, determine whether creepage, pollution flashover, and flashover abnormalities occur, and then remind maintenance personnel to go to the site for further confirmation and maintenance work, avoid large-scale regional power outages caused by insulator creepage, greatly reduce the workload of transmission maintenance personnel, and enable transmission maintenance personnel to carry out maintenance and repair in a targeted manner.

[0061] Among them, the insulator pollution flashover fault monitoring method adopts specific feature extraction and score evaluation algorithm based on the noise characteristics of insulator pollution flashover, effectively eliminates the interference of complex outdoor environment, and can accurately identify the pollution flashover fault of insulator with high accuracy and low false alarm rate; at the same time, the fault development trend is predicted according to the health assessment curve, and early warning information is pushed in real time to remind the operation and maintenance personnel to arrange on-site maintenance in time to avoid the continuous development of the fault, causing large-scale power outages and economic losses. It has good universality and input-output ratio, suitable for large-scale deployment, ensuring the stable and efficient operation of the equipment, while reducing the work intensity of the inspection personnel and reducing safety hazards such as high-altitude operations.

[0062] See also Fig. 9 A second embodiment of the present invention provides an insulator flashover fault monitoring device, comprising:

[0063] An effective eigenvalue acquisition unit 101 is used to acquire an effective noise eigenvalue data set generated by an edge sound pickup device, wherein the effective noise eigenvalue data set is generated by extracting the original audio signal of the surrounding environment collected by the edge sound pickup device and using an edge computing algorithm to extract the original audio signal, and the effective noise eigenvalue data set includes an effective noise eigenvalue group with the highest effective resolution that can retain effective information and an effective noise eigenvalue group with the lowest effective resolution that can retain data change trends;

[0064] The abnormality analysis unit 102 is used to call the trained fault monitoring system to process the multiple groups of effective noise characteristic value data sets accumulated within a preset time to generate an abnormality evaluation score, wherein the multiple groups of the most efficient resolution effective noise characteristic value groups are subjected to abnormality judgment processing to generate a judgment result, and the judgment result is compared with the values ​​of the multiple groups of the least efficient resolution effective noise characteristic value groups to obtain a corresponding abnormality evaluation score;

[0065] The early warning unit 103 is used to output an alarm signal according to the abnormality evaluation score.

[0066] See also Fig.10 The third embodiment of the present invention provides an insulator flashover fault monitoring device, comprising:

[0067] An autocorrelation calculation unit 201 is used to perform autocorrelation calculation on a plurality of groups of effective noise eigenvalues ​​with the highest efficiency resolution accumulated within a preset time, so as to generate an autocorrelation array capable of showing the periodic law of effective eigenvalues;

[0068] A Fourier transform unit 202, configured to perform an FFT calculation on the autocorrelation array to generate an energy value array corresponding to the autocorrelation array;

[0069] The abnormality judgment unit 203 is used to convert the energy value array into an energy proportion array, and judge whether the value of the energy proportion array in each frequency range exceeds a preset value, and generate an abnormal data group, wherein the abnormal data group is the value of the energy proportion array that exceeds the preset value;

[0070] The anomaly score evaluation unit 204 is used to evaluate the anomaly score of the abnormal data group based on the values ​​of multiple groups of the minimum effective resolution effective noise feature value groups accumulated within a preset time, and generate an anomaly score corresponding to the abnormal data group, wherein the anomaly score is the value of the minimum effective resolution effective noise feature value group in the frequency band corresponding to the abnormal data group.

[0071] A fourth embodiment of the present invention provides an insulator flashover fault monitoring device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, an insulator flashover fault monitoring method as described in any one of the above items is implemented.

[0072] See also Fig.11 The fifth embodiment of the present invention provides an insulator flashover fault monitoring system, comprising an edge pickup device, a remote monitoring terminal, and an insulator flashover fault monitoring device as described in claim 5, wherein the insulator flashover fault monitoring device is wirelessly connected to the edge pickup device, and the insulator flashover fault monitoring device is wirelessly connected to the remote monitoring terminal.

[0073] Specifically, in this embodiment, the edge pickup device is used for:

[0074] Acquire the original audio signal collected by the edge sound pickup device, wherein the original audio signal is all sounds within the effective range of the surrounding environment;

[0075] Using the edge computing algorithm to compress the original audio signal, dividing it into k first original noise feature value groups with a step size of n, and p second original noise feature value groups with a step size of h, where n, k, h, and p are preset values;

[0076] Performing FFT calculation on each of the first original noise eigenvalue groups and each of the second original noise eigenvalue groups to generate corresponding first effective noise eigenvalue groups and second effective noise eigenvalue groups;

[0077] An extraction process is performed on each of the first effective noise eigenvalue groups and each of the second effective noise eigenvalue groups to extract data of each of the first effective noise eigenvalue groups within a preset frequency band and data of each of the second effective noise eigenvalue groups, and the data are summed to generate a most efficient resolution effective noise eigenvalue group corresponding to the first effective noise eigenvalue group and a least efficient resolution effective noise eigenvalue group corresponding to the second effective noise eigenvalue group.

[0078] Specifically, in this embodiment, the edge pickup device can be an omnidirectional pickup, which is usually installed at the center of the insulator area of ​​each electric tower. The distance from the pickup to each insulator is equal, so each monitored electric tower only needs to install a set of collection devices to meet the monitoring requirements, and the implementation cost and installation difficulty are extremely low; the omnidirectional pickup has the advantages of no need for any mechanical connection with the insulator, convenient installation, efficient monitoring, input-output, and low false alarm rate. At the same time, the omnidirectional pickup can effectively resist interference and is not affected by the high voltage electricity, strong magnetic field and climate environment of the electric tower. This non-contact installation method has no mechanical connection with the insulator of the monitored object, and does not require any modification of the original equipment structure, does not interfere with the normal operation of the equipment, greatly reduces the difficulty and modification cost of implementation and installation, and is not interfered and affected by high voltage electricity and strong magnetic fields, and adopts the common omnidirectional pickup on the market, without the need for high-demand sensors and complex hardware development, and the monitoring cost is low. It should be noted that in other embodiments, edge pickup devices of other types of magnetic forces can also be used, which are not specifically limited here, but these schemes are within the scope of protection of the present invention.

[0079] Specifically, in this embodiment, all sounds within the effective range are collected by the fully designated microphones; wherein the original audio signal of 1 s with a length of 48000 is taken as an example.

[0080] Specifically, in this embodiment, the original audio signal of 1s with a length of 48000 is divided into fixed k arrays with a step size of n; an FFT calculation is performed on each array to obtain an array with a length of 257, and the data representing the 10kHz-15kHz frequency band are extracted and summed to obtain the total frequency band energy value of the array; finally, the 48000 original signal data are compressed into k data k_feature that can represent the characteristics of the original signal, and the larger the k value, the higher the degree of data information retention.

[0081] In this embodiment, the edge computing algorithm is used to calculate the original audio signal twice to calculate two sets of k_feature values; one set of k is 376, which is the most efficient resolution that can retain effective information after experiments, and is used to accurately detect insulator abnormalities, named feature1, and the other set of k is 24, which is the lowest resolution that can retain data change trends after experiments, and is used to quickly convert evaluation scores, named feature2. The two sets of effective feature value data calculated by the edge algorithm are uploaded to the server through the built-in 4G communication module of the fully specified pickup to achieve real-time storage and reading.

[0082] A sixth embodiment of the present invention provides a readable storage medium storing a computer program, wherein the computer program can be executed by a processor of a device where the storage medium is located to implement an insulator flashover fault monitoring method as described in any one of the above.

[0083] Exemplarily, the computer program described in the fourth embodiment and the sixth embodiment of the present invention can be divided into one or more modules, and the one or more modules are stored in the memory and executed by the processor to complete the present invention. The one or more modules can be a series of computer program instruction segments that can complete specific functions, and the instruction segments are used to describe the execution process of the computer program in the implementation of an insulator flashover fault monitoring device. For example, the device described in the second embodiment and the third embodiment of the present invention.

[0084] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the insulator flashover fault monitoring method, and uses various interfaces and lines to connect the various parts of the insulator flashover fault monitoring method.

[0085] The memory can be used to store the computer program and / or module, and the processor realizes various functions of an insulator flashover fault monitoring method by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, a text conversion function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, text message data, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0086] Wherein, if the implemented module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0087] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0088] The above are only preferred implementations of the present invention. The protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention.

Claims

1. A method for monitoring insulator flashover faults, characterized in that: include: Acquire an effective noise characteristic value data set generated by an edge sound pickup device, wherein the effective noise characteristic value data set is generated by extracting and generating an original audio signal of the surrounding environment collected by the edge sound pickup device using an edge computing algorithm, and the effective noise characteristic value data set includes an effective noise characteristic value group with the highest effective resolution that can retain effective information and an effective noise characteristic value group with the lowest effective resolution that can retain data change trends; Calling the trained fault monitoring system to process the multiple groups of effective noise characteristic value data sets accumulated within a preset time to generate an abnormality assessment score, wherein the multiple groups of the most efficient resolution effective noise characteristic value groups are subjected to abnormality judgment processing to generate a judgment result, and the judgment result is compared with the values ​​of the multiple groups of the least efficient resolution effective noise characteristic value groups to obtain a corresponding abnormality assessment score; According to the abnormality assessment score, an alarm signal is output.

2. The insulator flashover fault monitoring method according to claim 1, characterized in that: The trained fault monitoring system is called to process the multiple sets of effective noise characteristic value data sets accumulated within a preset time to generate an abnormality evaluation score, specifically: Accumulate multiple groups of the most efficient resolution effective noise eigenvalue groups within a preset time to perform autocorrelation calculations to generate an autocorrelation array that can show the periodic law of the effective eigenvalues; Performing FFT calculation on the autocorrelation array to generate an energy value array corresponding to the autocorrelation array; Converting the energy value array into an energy proportion array, and determining whether the value of the energy proportion array in each frequency range exceeds a preset value, to generate an abnormal data group, wherein the abnormal data group is the value of the energy proportion array that exceeds the preset value; An anomaly score is evaluated for the abnormal data group based on the values ​​of multiple groups of the minimum effective resolution effective noise feature value groups accumulated within a preset time, and an anomaly score corresponding to the abnormal data group is generated, wherein the anomaly score is the value of the minimum effective resolution effective noise feature value group in the frequency band corresponding to the abnormal data group.

3. An insulator flashover fault monitoring device, characterized in that: include: An effective eigenvalue acquisition unit is used to acquire an effective noise eigenvalue data set generated by an edge sound pickup device, wherein the effective noise eigenvalue data set is generated by extracting and generating an original audio signal of the surrounding environment collected by the edge sound pickup device using an edge computing algorithm, and the effective noise eigenvalue data set includes an effective noise eigenvalue group with the highest effective resolution that can retain effective information and an effective noise eigenvalue group with the lowest effective resolution that can retain data change trends; An abnormality analysis unit is used to call a trained fault monitoring system to process multiple groups of effective noise characteristic value data sets accumulated within a preset time to generate an abnormality evaluation score, wherein the multiple groups of the most efficient resolution effective noise characteristic value groups are subjected to abnormality judgment processing to generate a judgment result, and the judgment result is compared with the values ​​of the multiple groups of the least efficient resolution effective noise characteristic value groups to obtain a corresponding abnormality evaluation score; The early warning unit is used to output an alarm signal according to the abnormal evaluation score.

4. The insulator flashover fault monitoring device according to claim 3, characterized in that: Also includes: An autocorrelation calculation unit is used to perform autocorrelation calculation on a plurality of groups of effective noise eigenvalues ​​with the highest efficiency resolution accumulated within a preset time, so as to generate an autocorrelation array capable of showing the periodic law of effective eigenvalues; A Fourier transform unit, used for performing FFT calculation on the autocorrelation array to generate an energy value array corresponding to the autocorrelation array; An abnormality judgment unit, used for converting the energy value array into an energy proportion array, and judging whether the value of the energy proportion array in each frequency range exceeds a preset value, and generating an abnormal data group, wherein the abnormal data group is the value of the energy proportion array exceeding the preset value; An anomaly score evaluation unit is used to evaluate the anomaly score of the abnormal data group based on the values ​​of multiple groups of the minimum effective resolution effective noise feature value groups accumulated within a preset time, and generate an anomaly score corresponding to the abnormal data group, wherein the anomaly score is the value of the minimum effective resolution effective noise feature value group in the frequency band corresponding to the abnormal data group.

5. An insulator flashover fault monitoring device, characterized in that: The invention comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, an insulator flashover fault monitoring method as claimed in any one of claims 1 to 2 is implemented.

6. An insulator flashover fault monitoring system, characterized in that: It comprises an edge sound pickup device, a remote monitoring terminal and an insulator flashover fault monitoring device as claimed in claim 5, wherein the insulator flashover fault monitoring device is wirelessly connected to the edge sound pickup device, and the insulator flashover fault monitoring device is wirelessly connected to the remote monitoring terminal.

7. The insulator flashover fault monitoring system according to claim 6, characterized in that: The edge pickup device is used for: Acquire the original audio signal collected by the edge sound pickup device, wherein the original audio signal is all sounds within the effective range of the surrounding environment; The original audio signal is compressed and processed by using the edge computing algorithm to be divided into k first original noise characteristic value groups with a step size of n and p second original noise characteristic value groups with a step size of h, wherein n, k, h, and p are preset values; Performing FFT calculation on each of the first original noise eigenvalue groups and each of the second original noise eigenvalue groups to generate corresponding first effective noise eigenvalue groups and second effective noise eigenvalue groups; An extraction process is performed on each of the first effective noise eigenvalue groups and each of the second effective noise eigenvalue groups to extract data of each of the first effective noise eigenvalue groups within a preset frequency band and data of each of the second effective noise eigenvalue groups, and the data are summed to generate a most efficient resolution effective noise eigenvalue group corresponding to the first effective noise eigenvalue group and a least efficient resolution effective noise eigenvalue group corresponding to the second effective noise eigenvalue group.

8. A readable storage medium, characterized in that: A computer program is stored, and the computer program can be executed by a processor of the device where the storage medium is located to implement an insulator flashover fault monitoring method as described in any one of claims 1 to 2.

Images