High-voltage lightning arrester fault detection method, device and equipment based on cross-modal fusion

Abstract

The invention provides a high-voltage lightning arrester fault detection method, device and equipment based on cross-modal fusion, and relates to the technical field of lightning arrester fault detection. The method comprises the following steps: acquiring real-time state data of the high-voltage lightning arrester based on a plurality of preset sensors, and determining first fault information of the high-voltage lightning arrester; judging whether a fault of a preset fault type exists in the first fault information or not; if yes, collecting a real-time infrared image of the high-voltage lightning arrester, determining at least one fault position of the high-voltage lightning arrester and second fault information of each fault position according to the real-time infrared image, and determining final fault information of the high-voltage lightning arrester according to the second fault information and the first fault information of each fault position; and if the first fault information does not exist, adjusting the first fault information according to the real-time environment data of the high-voltage lightning arrester to obtain final fault information. According to the invention, accurate positioning and comprehensive diagnosis of the fault of the high-voltage lightning arrester can be realized, and the fault detection accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of lightning arrester fault detection, and in particular to a high-voltage lightning arrester fault detection method, device and equipment based on cross-modal fusion. BACKGROUND

[0002] The high-voltage lightning arrester is a key component in the power system for protecting power equipment from over-voltage. Its main function is to eliminate over-voltage caused by lightning strikes or high-voltage power supply equipment switching. After discharging over-voltage, the lightning arrester should return to the original state to ensure the continuous operation of the power system. However, the lightning arrester may fail due to aging, internal moisture, surface contamination or perforation failure caused by over-current during long-term operation, resulting in a decrease in its protection function and thus a risk of damage to the power equipment. Therefore, timely detection of the fault type of the lightning arrester can effectively maintain the normal operation of the lightning arrester, which is crucial for ensuring the reliable operation of the power system.

[0003] The existing lightning arrester fault detection methods mainly rely on sensor detection and infrared imaging technology. The sensor detection method includes leakage current detection, insulation resistance measurement and direct current reference voltage test, etc. The leakage current detection judges the insulation performance of the lightning arrester by monitoring the change of its leakage current; the insulation resistance measurement is used to evaluate the insulation state of the lightning arrester; and the direct current reference voltage test is used to detect the electrical characteristics of the lightning arrester. The infrared imaging technology identifies potential faults by monitoring the temperature distribution on the surface of the lightning arrester, which has the advantages of being intuitive and fast.

[0004] However, in the existing lightning arrester fault detection methods, the sensor detection is susceptible to environmental interference, and although the infrared imaging is intuitive, it is difficult to provide quantitative data and has limited ability to identify early faults. These shortcomings result in inaccurate lightning arrester faults obtained by the existing lightning arrester fault detection methods, which are prone to judgment errors or omissions. SUMMARY

[0005] The embodiments of the present application provide a high-voltage lightning arrester fault detection method, device and equipment based on cross-modal fusion to solve the problem of inaccurate lightning arrester faults obtained by the existing lightning arrester fault detection methods, which are prone to judgment errors or omissions.

[0006] In a first aspect, the embodiments of the present application provide a high-voltage lightning arrester fault detection method based on cross-modal fusion, comprising: acquiring real-time state data of the high-voltage lightning arrester based on a plurality of pre-set sensors, and determining first fault information of the high-voltage lightning arrester according to the real-time state data; judging whether there is a fault of a preset fault type in the first fault information; if the preset fault type fault exists in the first fault information, collecting a real-time infrared image of the high-voltage surge arrester, and determining at least one fault position of the high-voltage surge arrester and second fault information of each fault position according to the real-time infrared image, and determining final fault information of the high-voltage surge arrester according to the second fault information of each fault position and the first fault information; if the preset fault type fault does not exist in the first fault information, adjusting the first fault information according to real-time environmental data of the high-voltage surge arrester to obtain the final fault information.

[0007] In a second aspect, an embodiment of the present application provides a high-voltage surge arrester fault detection device based on cross-modal fusion, comprising: a collection module configured to collect real-time state data of the high-voltage surge arrester based on a plurality of pre-set sensors, and determine first fault information of the high-voltage surge arrester according to the real-time state data; a judgment module configured to determine whether a preset fault type fault exists in the first fault information; a determination module configured to, if the preset fault type fault exists in the first fault information, collect a real-time infrared image of the high-voltage surge arrester, and determine at least one fault position of the high-voltage surge arrester and second fault information of each fault position according to the real-time infrared image, and determine final fault information of the high-voltage surge arrester according to the second fault information of each fault position and the first fault information; an adjustment module configured to, if the preset fault type fault does not exist in the first fault information, adjust the first fault information according to real-time environmental data of the high-voltage surge arrester to obtain the final fault information.

[0008] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory and a processor, the memory stores a computer program, and the processor implements the method in the first aspect or any possible implementation manner of the first aspect when executing the computer program.

[0009] In the embodiment of the present application, the real-time state data of the high-voltage arrester is collected through a plurality of sensors set in advance, and the first fault information of the high-voltage arrester is determined according to the real-time state data. When the fault of the preset fault type exists in the first fault information, the real-time infrared image of the high-voltage arrester is collected. At least one fault position of the high-voltage arrester and the second fault information of each fault position are further determined according to the real-time infrared image, and the final fault information of the high-voltage arrester is determined according to the second fault information of each fault position and the first fault information. In addition, when the fault of the preset fault type does not exist in the first fault information, the first fault information is adjusted according to the real-time and environmental data of the high-voltage arrester to obtain the final fault information. When the fault of the preset type exists, the final fault information of the high-voltage arrester is obtained by combining the real-time state data collected by the sensor and the real-time infrared image of different modal data, which can realize accurate positioning and comprehensive diagnosis of the high-voltage arrester fault and improve the fault detection accuracy. When the fault of the preset type does not exist, the final fault information is obtained by adjusting the first fault information through the environmental data, which can more accurately reflect the actual state of the high-voltage arrester and avoid fault misjudgment. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 is the application scenario of the high-voltage arrester fault detection method based on cross-modal fusion provided by the embodiment of the present application; Figure 2 is the implementation flowchart of step S130 of the high-voltage arrester fault detection method based on cross-modal fusion provided by the embodiment of the present application; Figure 3 is a structural schematic diagram of the high-voltage arrester fault detection device based on cross-modal fusion provided by the embodiment of the present application; Figure 4 is a schematic diagram of an electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0011] The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0012] Referring to Figure 1 , which shows the implementation flowchart of the high-voltage arrester fault detection method based on cross-modal fusion provided by the embodiment of the present application, and is described in detail as follows: Step S110, based on a plurality of sensors set in advance, collecting real-time state data of the high-voltage arrester, and determining first fault information of the high-voltage arrester according to the real-time state data.

[0013] In some embodiments, various types of sensors need to be deployed in advance at key positions of the high-voltage surge arrester, and their installation positions, monitoring parameters, and sampling frequencies are pre-planned according to the structure and fault characteristics of the surge arrester. For example, a current sensor is installed on the surge arrester body to monitor the change of leakage current, and a temperature sensor is deployed near the insulation component to capture abnormal temperature rise. The high-voltage surge arrester refers to an overvoltage protection device installed in a power system, mainly used to suppress the damage of lightning or operating overvoltage to electrical equipment, including valve piece resistance, ceramic insulator, and other components, which are in a high resistance state during normal operation and quickly conduct to release energy during overvoltage. Real-time state data refers to the quantitative information obtained by the sensor during the operation of the surge arrester, reflecting the current working condition of the device, including electrical parameters, physical parameters, environmental parameters, etc. The first fault information refers to the fault-related information determined by the real-time state data collected by the sensor, including fault type, fault severity level, specific performance of abnormal parameters, etc.

[0014] In one possible implementation, the specific processing of step S110 is as follows: based on the pre-set multiple sensors, the real-time state data of the high-voltage surge arrester is collected, the abnormal state data is screened from the real-time state data, and the first fault information of the high-voltage surge arrester is determined according to the abnormal state data.

[0015] In some embodiments, the abnormal state data refers to the values or signals in the real-time state data that do not conform to the normal operation parameters of the high-voltage surge arrester, which is usually manifested as parameters exceeding threshold values, abnormal fluctuation trends, or data mutations, etc. For example, the leakage current collected by the current sensor is continuously higher than the fluctuation upper limit during normal operation of the device. According to the screened abnormal data, combined with the fault knowledge base or diagnostic rules, the possible fault type and characteristics of the high-voltage surge arrester can be inferred, for example, if the abnormal state data shows that the leakage current continuously increases and is accompanied by temperature rise, the first fault information may point to "valve piece aging leading to insulation performance degradation".

[0016] Step S120, determining whether the first fault information contains a fault of a preset fault type.

[0017] In some embodiments, the preset fault types include valve piece aging, internal dampness, valve piece damage, ceramic insulator salt corrosion, and surface contamination, etc. These faults can be reflected by infrared imaging, and other fault types that can be reflected by infrared imaging can also be used as preset fault types.

[0018] Step S130, if the first fault information contains the preset fault type, collecting the real-time infrared image of the high-voltage surge arrester, determining at least one fault position of the high-voltage surge arrester and the second fault information of each fault position according to the real-time infrared image, and determining the final fault information of the high-voltage surge arrester according to the second fault information of each fault position and the first fault information.

[0019] In some embodiments, the real-time infrared image refers to the surface temperature distribution image of the high-voltage surge arrester captured in real time by the infrared thermal imaging device, and different temperature regions are visualized in different colors. The fault position refers to the specific physical part where the abnormality occurs in the high-voltage surge arrester, which can be located by the temperature abnormal region of the infrared image. If the temperature of a certain valve piece region in the infrared image is significantly higher than the temperature of the valve piece region in the normal operation, the position of the valve piece is the fault position. The second fault information is detailed feature information related to the fault position obtained by analyzing the real-time infrared image, mainly including temperature abnormality parameters, abnormal region geometric features, and deviation degree from the normal state, etc. The final fault information is a comprehensive diagnostic conclusion formed by comprehensively analyzing the first fault information and the second fault information, including fault type, specific position, severity, and multi-modal data cross-validation basis, etc.

[0020] Referring to Figure 2 The specific processing method of step S130 can include steps S1301-S1309, and the specific contents are as follows: Step S1301, if the first fault information contains the preset fault type, determining the image collection position according to the first fault information.

[0021] In some embodiments, the image collection position refers to the best installation position or shooting angle of the infrared image collection device according to the fault type and possible physical position in the first fault information. For example, if the fault type in the first fault information is valve piece damage, the image collection position can be set in front of the valve piece region of the high-voltage surge arrester body, ensuring that the infrared device can clearly capture the temperature distribution details of the region, so as to accurately locate the fault point.

[0022] Step S1302, setting the infrared image collection device at the image collection position, and collecting the real-time infrared image of the high-voltage surge arrester according to the set infrared image collection device.

[0023] In some embodiments, the infrared image acquisition device is a device for capturing the infrared radiation of the surface of the high-voltage surge arrester and converting it into a temperature distribution image, which can be an infrared thermal imager. After the infrared image acquisition device is set up, the infrared radiation data of the surface of the high-voltage surge arrester can be obtained in real time and real-time infrared images can be generated through the set-up infrared image acquisition device. The infrared signals of each part of the high-voltage surge arrester are continuously received after the device is started, and the real-time infrared images are formed after processing. The real-time infrared images are used for subsequent fault location analysis.

[0024] In step S1303, the real-time temperature distribution map of the high-voltage surge arrester is determined according to the real-time infrared image, and the real-time temperature distribution map is compared with the standard temperature distribution map of the high-voltage surge arrester in the normal operating state to determine at least one fault location of the high-voltage surge arrester.

[0025] In some embodiments, the real-time temperature distribution map is a visual temperature distribution map converted from the pixel temperature information in the real-time infrared image, usually in the form of isotherm or color block to mark the relative temperature values of each region. The standard temperature distribution map refers to the pre-established temperature distribution map of the high-voltage surge arrester in the normal operating state, which needs to be generated according to the device design parameters, historical operation data or standard data. The real-time temperature distribution map and the standard temperature distribution map are compared to obtain a comparison result, and according to the comparison result, the specific physical part of the abnormal temperature on the surface of the high-voltage surge arrester can be located. For example, if there is a significant difference between the temperature distribution of a certain valve piece region in the real-time temperature distribution map and the temperature distribution of the corresponding region in the standard temperature distribution map, it is determined that the location of the valve piece is the fault location.

[0026] In step S1304, based on the real-time infrared image and each fault location, the second fault information of each fault location of the high-voltage surge arrester is determined.

[0027] In some embodiments, according to the fault location, the temperature data of each fault location can be determined from the real-time infrared image, and according to the temperature data of each fault location, the temperature matrix of each fault location can be obtained, and according to the temperature matrix of each fault location, the corresponding second fault information of each fault location can be determined.

[0028] In step S1305, according to the location information of each fault location, the first fault data of each fault location is screened out from the first fault information.

[0029] In some embodiments, the first fault data refers to the abnormal data of electrical parameters directly related to the specific fault location in the first fault information, such as leakage current fluctuation, abnormal voltage value, specific temperature value, etc. These data are the basis for preliminary diagnosis of faults.

[0030] Step S1306, based on the second fault information of each fault position, determine the second fault data of each fault position.

[0031] In some embodiments, the second fault data is quantitative data converted from the second fault information, which is a numerical expression of the characteristics of the fault position.

[0032] Step S1307, according to the real-time environmental data of the high-voltage surge arrester, respectively adjust the first fault data and the second fault data of each fault position.

[0033] In some embodiments, the real-time environmental data refers to the environmental parameters of the high-voltage surge arrester operating site, including environmental temperature, humidity, air pressure, electromagnetic interference intensity, etc. These data can be collected in real time by environmental sensors. Environmental adaptation adjustment is a process of eliminating the interference of environmental factors on fault data through algorithms or models, so that the data more truly reflects the actual state of the equipment. For example, when the environmental temperature rises, the temperature of the surge arrester body will rise accordingly. At this time, the temperature data detected by the sensor needs to be reduced by the influence of the environmental temperature to restore the temperature of the device itself.

[0034] Step S1308, based on the first fault data and the second fault data after environmental adaptation adjustment, calculate the fault data of each fault position.

[0035] In some embodiments, the first fault data and the second fault data after environmental adaptation adjustment are fused and processed through a specific algorithm or model, which can obtain fault data that comprehensively reflects the actual fault condition of the fault position. For example, a weighted fusion method can be used to combine electrical parameters and temperature characteristics to calculate a comprehensive fault index of the fault position, so as to more comprehensively describe the severity and nature of the fault.

[0036] Step S1309, according to the fault data of each fault position, determine the final fault information of the high-voltage surge arrester.

[0037] In some embodiments, the final fault information refers to a comprehensive diagnostic conclusion formed after integrating multi-modal data, covering information such as fault type, specific physical position, severity level, etc. For example, the final fault information can be "the high-voltage surge arrester has serious aging fault in the 2nd and 4th valve pieces, causing abnormal temperature rise in the corresponding position and continuous increase of leakage current, and is judged as a II-level serious fault".

[0038] In a possible implementation, the specific processing manner of step S1309 is: for each fault position, determining the corresponding possible fault information set of the fault position based on the fault data of the fault position; calculating the matching degrees of the fault data of the remaining fault positions and each possible fault information in the possible fault information set, and screening out the possible fault information with the highest matching degree as the candidate fault information; and determining the final fault information of the high-voltage surge arrester based on the fault data of each fault position and all candidate fault information.

[0039] In some embodiments, the possible fault information set is inferred according to the fault data of a single fault position, and if the fault data of a certain position shows abnormal temperature and is accompanied by current fluctuation, the possible fault information set includes preset fault types such as valve aging, internal moisture and surface contamination that match the characteristics. The matching degree is calculated by an algorithm to determine the degree of coincidence of the fault data of the remaining fault positions with each fault type in the current possible fault information set, which can be quantified by a similarity index. For example, the fault data of fault position B has a higher similarity to fault information 1 than to fault information 2, so fault information 1 has a higher matching degree.

[0040] It should be noted that the candidate fault information is the fault type with the highest matching degree with the fault data of other positions screened from the possible fault information set of each fault position, and the candidate fault information is the alternative fault information of the position. For example, the possible fault information set of fault position A includes fault information 1, fault information 2 and fault information 3, and the remaining fault positions are fault position B and fault position C. The matching degree of the fault data of fault position B to fault information 1 is 85%, the matching degree of the fault data of fault position C to fault information 1 is 60%, the matching degree of the fault data of fault position B to fault information 2 is 82%, the matching degree of the fault data of fault position C to fault information 2 is 79%, the matching degree of the fault data of fault position B to fault information 3 is 55%, and the matching degree of the fault data of fault position C to fault information 3 is 75%. Therefore, the matching degree of fault information 2 to the remaining fault positions of fault position A is the highest, and fault information 2 is determined as the candidate fault information of fault position A.

[0041] In some embodiments, the final fault information is the fault information of the high-voltage surge arrester containing fault type, location, severity and verification basis formed by fusing sensor state data and infrared image data after environmental adjustment and matching degree calculation. After obtaining the candidate fault information of each fault location, if the similarity of the candidate fault information of each fault location exceeds the preset value, a candidate fault information is randomly selected to determine the final fault information of the high-voltage surge arrester; if the similarity of the candidate fault information of one fault location to the candidate fault information of the remaining fault locations does not exceed the preset value, a candidate fault information is randomly selected from the remaining fault locations to determine the final fault information of the high-voltage surge arrester; if the similarity of the candidate fault information of at least two fault locations to the candidate fault information of the remaining fault locations does not exceed the preset value, and the similarity of the candidate fault information of the at least two fault locations exceeds the preset value, a candidate fault information is randomly selected from the candidate fault information of the at least two fault locations, and a candidate fault information is randomly selected from the candidate fault information of the remaining fault locations, which are collectively used as the final fault information of the high-voltage surge arrester. All possible fault types are listed for each fault location, and then the matching degree of the data of other fault locations to these types is calculated to screen the candidate fault information. Finally, the final fault information is obtained by comprehensively considering the fault data and the candidate fault information of all fault locations. The correlation between different fault locations is fully considered, the single position data misjudgment is avoided, the fault type, location and severity are more accurately determined, the detection result is more comprehensive and reliable, and the missed or wrong judgment is reduced.

[0042] In step S140, if there is no fault of the preset fault type in the first fault information, the first fault information is adjusted according to the real-time environmental data of the high-voltage surge arrester to obtain the final fault information.

[0043] In some embodiments, adjustment refers to the process of correcting the deviation of the original sensor data by environmental data, and further adjusting the first fault information according to the corrected sensor data to obtain the final fault information, such as when high humidity causes the insulation resistance sensor to read low, a compensation coefficient is calculated according to the real-time humidity parameter to eliminate the disturbance of environmental factors on the data and restore the true state of the equipment.

[0044] It should be noted that after obtaining the final fault information of the high-voltage surge arrester, the fault alarm scheme of the high-voltage surge arrester can be determined according to the final fault information of the high-voltage surge arrester, and the fault alarm is performed according to the fault alarm scheme.

[0045] By combining the electrical state data collected by the sensor and the temperature distribution data of the infrared image, the fault location is accurately positioned and the fault type is comprehensively diagnosed when the preset fault is found through cross-modal data fusion, and the sensor deviation is corrected through environmental data when there is no preset fault, solving the problems of traditional methods being easily disturbed by the environment and inaccurate positioning, realizing accurate position positioning, comprehensive type diagnosis and severity evaluation of high-voltage lightning arrester faults, and effectively reducing the occurrence of fault misjudgment and omission, making the detection result more accurate and reliable, and ensuring the safe operation of the power system.

[0046] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0047] The following is a device embodiment of the present application, and for details not described in detail, reference can be made to the corresponding method embodiments described above.

[0048] Figure 3 The structure schematic diagram of the high-voltage lightning arrester fault detection device based on cross-modal fusion provided by the embodiment of the present application is shown, only the parts related to the embodiment of the present application are shown for convenience of description, and the details are described as follows: As Figure 3 shown, the high-voltage lightning arrester fault detection device 3 based on cross-modal fusion comprises: The acquisition module 31 is configured to acquire real-time state data of the high-voltage lightning arrester based on a plurality of sensors set in advance, and determine first fault information of the high-voltage lightning arrester according to the real-time state data. The judgment module 32 is configured to determine whether there is a fault of a preset fault type in the first fault information. The determination module 33 is configured to, when there is a fault of the preset fault type in the first fault information, acquire a real-time infrared image of the high-voltage lightning arrester, and determine at least one fault position of the high-voltage lightning arrester and second fault information of each fault position according to the real-time infrared image, and determine final fault information of the high-voltage lightning arrester according to the second fault information of each fault position and the first fault information. The adjustment module 34 is configured to, when there is no fault of the preset fault type in the first fault information, adjust the first fault information according to real-time environmental data of the high-voltage lightning arrester to obtain the final fault information.

[0049] In a possible implementation, the acquisition module 31 is specifically configured to: acquire real-time state data of the high-voltage lightning arrester based on a plurality of sensors set in advance, filter out abnormal state data from the real-time state data, and determine first fault information of the high-voltage lightning arrester according to the abnormal state data.

[0050] In a possible implementation, the determining module 33 is specifically configured to: if the preset fault type exists in the first fault information, determine the image acquisition position according to the first fault information; and set the infrared image acquisition device at the image acquisition position, and acquire the real-time infrared image of the high-voltage surge arrester according to the set infrared image acquisition device.

[0051] In a possible implementation, the determining module 33 is further configured to: determine the real-time temperature distribution of the high-voltage surge arrester according to the real-time infrared image, compare the real-time temperature distribution with a standard temperature distribution of the high-voltage surge arrester in a normal operation state, and determine at least one fault position of the high-voltage surge arrester; and determine second fault information of each fault position of the high-voltage surge arrester based on the real-time infrared image and each fault position.

[0052] In a possible implementation, the determining module 33 is further configured to: according to the position information of each fault position, screen the first fault data of each fault position from the first fault information; determine second fault data of each fault position based on the second fault information of each fault position; respectively perform environment-adaptive adjustment on the first fault data and the second fault data of each fault position according to real-time environment data of the high-voltage surge arrester; calculate fault data of each fault position based on the first fault data and the second fault data after the environment-adaptive adjustment; and determine final fault information of the high-voltage surge arrester according to the fault data of each fault position.

[0053] In a possible implementation, the determining module 33 is further configured to: for each fault position, determine a corresponding possible fault information set of the fault position based on the fault data of the fault position; calculate a matching degree of the corresponding fault data of the remaining fault positions and each possible fault information in the possible fault information set, and screen a possible fault information with the highest matching degree as a candidate fault information; and determine final fault information of the high-voltage surge arrester based on the fault data of each fault position and all candidate fault information.

[0054] In a possible implementation, the adjusting module 34 is specifically configured to: determine an adjustment parameter of each sensor according to real-time environment data of the high-voltage surge arrester and sensor information of each sensor; and adjust the first fault information based on the adjustment parameter of each sensor to obtain the final fault information of the high-voltage surge arrester.

[0055] In a possible implementation, the preset fault type includes valve piece aging, internal dampness, valve piece damage, ceramic insulator salt corrosion, and surface pollution.

[0056] Figure 4 is a schematic diagram of an electronic device provided by an embodiment of the present application. As shown in Figure 4As shown, the electronic device 4 of this embodiment includes a processor 40 and a memory 41. The memory 41 stores a computer program 42. The processor 40 implements the steps in the above method embodiments when executing the computer program 42. Alternatively, the processor 40 implements the functions of the modules / units in the above apparatus embodiments when executing the computer program 42.

[0057] For example, the computer program 42 can be divided into one or more modules / units, which are stored in the memory 41 and executed by the processor 40 to complete the present application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program 42 in the electronic device 4.

[0058] The electronic device 4 can include, but is not limited to, the processor 40 and the memory 41. Those skilled in the art can understand that, Figure 4 The electronic device 4 is only an example and does not constitute a limitation on the electronic device 4, which can include more or fewer components than shown, or combine certain components, or different components, for example, the electronic device 4 can also include an input / output device, a network access device, a bus, etc.

[0059] For the convenience and brevity of description, only the above-mentioned division of functional modules / units is exemplified, and in actual application, the above-mentioned functions can be completed by different functional modules / units according to needs. The above-mentioned modules / units can be realized in the form of hardware, software, or a combination of hardware and software.

[0060] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in detail in a certain embodiment can be referred to the relevant description of other embodiments. If there is no special description and no logical conflict, the terms and / or descriptions of different embodiments are consistent and can be mutually referenced, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0061] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A fault detection method for high-voltage surge arresters based on cross-modal fusion, characterized in that, include: Real-time status data of the high-voltage surge arrester is collected based on multiple pre-set sensors, and the first fault information of the high-voltage surge arrester is determined based on the real-time status data. Determine whether the first fault information contains a fault of a preset fault type; If a fault of a preset fault type exists in the first fault information, then a real-time infrared image of the high-voltage surge arrester is collected, and at least one fault location of the high-voltage surge arrester and second fault information of each fault location are determined based on the real-time infrared image. Based on the second fault information of each fault location and the first fault information, the final fault information of the high-voltage surge arrester is determined. If the first fault information does not contain a preset fault type, the first fault information is adjusted based on the real-time environmental data of the high-voltage surge arrester to obtain the final fault information.

2. The high-voltage surge arrester fault detection method based on cross-modal fusion according to claim 1, characterized in that, Determining the final fault information of the high-voltage surge arrester based on the second fault information and the first fault information for each fault location includes: Based on the location information of each fault location, the first fault data for each fault location is filtered out from the first fault information; Based on the second fault information at each fault location, determine the second fault data at each fault location; Based on the real-time environmental data of the high-voltage surge arrester, environmental adaptation adjustments are made to the first and second fault data at each fault location; Based on the first and second fault data after environmental adaptation adjustments, the fault data for each fault location is calculated. Based on the fault data at each fault location, the final fault information of the high-voltage surge arrester is determined.

3. The high-voltage surge arrester fault detection method based on cross-modal fusion according to claim 2, characterized in that, The step of determining the final fault information of the high-voltage surge arrester based on the fault data at each fault location includes: For each fault location, based on the fault data at that fault location, a set of possible fault information corresponding to that fault location is determined; the matching degree between the fault data corresponding to the remaining fault locations and each possible fault information in the set of possible fault information is calculated, and the possible fault information with the highest matching degree is selected as candidate fault information. Based on the fault data at each fault location and all candidate fault information, the final fault information of the high-voltage surge arrester is determined.

4. The high-voltage surge arrester fault detection method based on cross-modal fusion according to claim 1, characterized in that, The step of determining at least one fault location of the high-voltage surge arrester and second fault information for each fault location based on the real-time infrared image includes: Based on the real-time infrared image, a real-time temperature distribution map of the high-voltage surge arrester is determined. The real-time temperature distribution map is then compared with a standard temperature distribution map of the high-voltage surge arrester under normal operating conditions to determine at least one fault location of the high-voltage surge arrester. Based on the real-time infrared image and each fault location, second fault information is determined for each fault location of the high-voltage surge arrester.

5. The high-voltage surge arrester fault detection method based on cross-modal fusion according to claim 1, characterized in that, If the first fault information contains a fault of a preset fault type, then the real-time infrared image of the high-voltage surge arrester is collected, including: If the first fault information contains a fault of a preset fault type, then the image acquisition location is determined based on the first fault information. An infrared image acquisition device is set at the image acquisition location, and real-time infrared images of the high-voltage surge arrester are acquired according to the set infrared image acquisition device.

6. The high-voltage surge arrester fault detection method based on cross-modal fusion according to claim 1, characterized in that, Real-time status data of the high-voltage surge arrester is collected based on multiple pre-set sensors, and the first fault information of the high-voltage surge arrester is determined based on the real-time status data, including: Based on multiple pre-set sensors, real-time status data of the high-voltage surge arrester is collected. Abnormal status data is filtered out from the real-time status data, and the first fault information of the high-voltage surge arrester is determined based on the abnormal status data.

7. The high-voltage surge arrester fault detection method based on cross-modal fusion according to claim 1, characterized in that, The step of adjusting the first fault information based on the real-time environmental data of the high-voltage surge arrester to obtain the final fault information includes: Based on the real-time environmental data of the high-voltage surge arrester and the sensor information of each sensor, the adjustment parameters of each sensor are determined; Based on the adjustment parameters of each sensor, the first fault information is adjusted to obtain the final fault information of the high-voltage surge arrester.

8. The high-voltage surge arrester fault detection method based on cross-modal fusion according to claim 1, characterized in that, The preset fault types include valve plate aging, internal dampness, valve plate damage, ceramic insulator salt corrosion, and surface contamination.

9. A high-voltage surge arrester fault detection device based on cross-modal fusion, characterized in that, include: The acquisition module is used to acquire real-time status data of the high-voltage surge arrester based on multiple pre-set sensors, and to determine the first fault information of the high-voltage surge arrester based on the real-time status data. The judgment module is used to determine whether there is a preset fault type in the first fault information; The determination module is used to collect real-time infrared images of the high-voltage surge arrester when a fault of a preset fault type exists in the first fault information, and to determine at least one fault location of the high-voltage surge arrester and second fault information of each fault location based on the real-time infrared images, and to determine the final fault information of the high-voltage surge arrester based on the second fault information of each fault location and the first fault information. The adjustment module is used to adjust the first fault information according to the real-time environmental data of the high-voltage surge arrester when there is no preset fault type in the first fault information, so as to obtain the final fault information.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images